face presentation on ultrasound

Face and Brow Presentation

• Author: Teresa Marino, MD; Chief Editor: Carl V Smith, MD  more...
• Sections Face and Brow Presentation
• Mechanism of Labor
• Labor Management

At the onset of labor, assessment of the fetal presentation with respect to the maternal birth canal is critical to the route of delivery. At term, the vast majority of fetuses present in the vertex presentation, where the fetal head is flexed so that the chin is in contact with the fetal thorax. The fetal spine typically lies along the longitudinal axis of the uterus. Nonvertex presentations (including breech, transverse lie, face, brow, and compound presentations) occur in less than 4% of fetuses at term. Malpresentation of the vertex presentation occurs if there is deflexion or extension of the fetal head leading to brow or face presentation, respectively.

In a face presentation, the fetal head and neck are hyperextended, causing the occiput to come in contact with the upper back of the fetus while lying in a longitudinal axis. The presenting portion of the fetus is the fetal face between the orbital ridges and the chin. The fetal chin (mentum) is the point designated for reference during an internal examination through the cervix. The occiput of a vertex is usually hard and has a smooth contour, while the face and brow tend to be more irregular and soft. Like the occiput, the mentum can present in any position relative to the maternal pelvis. For example, if the mentum presents in the left anterior quadrant of the maternal pelvis, it is designated as left mentum anterior (LMA).

In a brow presentation, the fetal head is midway between full flexion (vertex) and hyperextension (face) along a longitudinal axis. The presenting portion of the fetal head is between the orbital ridge and the anterior fontanel. The face and chin are not included. The frontal bones are the point of designation and can present (as with the occiput during a vertex delivery) in any position relative to the maternal pelvis. When the sagittal suture is transverse to the pelvic axis and the anterior fontanel is on the right maternal side, the fetus would be in the right frontotransverse position (RFT).

Face presentation occurs in 1 of every 600-800 live births, averaging about 0.2% of live births. Causative factors associated with a face presentation are similar to those leading to general malpresentation and those that prevent head flexion or favor extension. Possible etiology includes multiple gestations, grand multiparity, fetal malformations, prematurity, and cephalopelvic disproportion. At least one etiological factor may be identified in up to 90% of cases with face presentation.

Fetal anomalies such as hydrocephalus, anencephaly, and neck masses are common risk factors and may account for as many as 60% of cases of face presentation. For example, anencephaly is found in more than 30% of cases of face presentation. Fetal thyromegaly and neck masses also lead to extension of the fetal head.

A contracted pelvis or cephalopelvic disproportion, from either a small pelvis or a large fetus, occurs in 10-40% of cases. Multiparity or a large abdomen can cause decreased uterine tone, leading to natural extension of the fetal head.

Face presentation is diagnosed late in the first or second stage of labor by examination of a dilated cervix. On digital examination, the distinctive facial features of the nose, mouth, and chin, the malar bones, and particularly the orbital ridges can be palpated. This presentation can be confused with a breech presentation because the mouth may be confused with the anus and the malar bones or orbital ridges may be confused with the ischial tuberosities. The facial presentation has a triangular configuration of the mouth to the orbital ridges compared to the breech presentation of the anus and fetal genitalia. During Leopold maneuvers, diagnosis is very unlikely. Diagnosis can be confirmed by ultrasound evaluation, which reveals a hyperextended fetal neck. [ 1 , 2 ]

Brow presentation is the least common of all fetal presentations and the incidence varies from 1 in 500 deliveries to 1 in 1400 deliveries. Brow presentation may be encountered early in labor but is usually a transitional state and converts to a vertex presentation after the fetal neck flexes. Occasionally, further extension may occur resulting in a face presentation.

The causes of a persistent brow presentation are generally similar to those causing a face presentation and include cephalopelvic disproportion or pelvic contracture, increasing parity and prematurity. These are implicated in more than 60% of cases of persistent brow presentation. Premature rupture of membranes may precede brow presentation in as many as 27% of cases.

Diagnosis of a brow presentation can occasionally be made with abdominal palpation by Leopold maneuvers. A prominent occipital prominence is encountered along the fetal back, and the fetal chin is also palpable; however, the diagnosis of a brow presentation is usually confirmed by examination of a dilated cervix. The orbital ridge, eyes, nose, forehead, and anterior fontanelle are palpated. The mouth and chin are not palpable, thus excluding face presentation. Fetal ultrasound evaluation again notes a hyperextended neck.

As with face presentation, diagnosis is often made late in labor with half of cases occurring in the second stage of labor. The most common position is the mentum anterior, which occurs about twice as often as either transverse or posterior positions. A higher cesarean delivery rate occurs with a mentum transverse or posterior [ 3 ] position than with a mentum anterior position.

The mechanism of labor consists of the cardinal movements of engagement, descent, flexion, internal rotation, and the accessory movements of extension and external rotation. Intuitively, the cardinal movements of labor for a face presentation are not completely identical to those of a vertex presentation.

While descending into the pelvis, the natural contractile forces combined with the maternal pelvic architecture allow the fetal head to either flex or extend. In the vertex presentation, the vertex is flexed such that the chin rests on the fetal chest, allowing the suboccipitobregmatic diameter of approximately 9.5 cm to be the widest diameter through the maternal pelvis. This is the smallest of the diameters to negotiate the maternal pelvis. Following engagement in the face presentation, descent is made. The widest diameter of the fetal head negotiating the pelvis is the trachelobregmatic or submentobregmatic diameter, which is 10.2 cm (0.7 cm larger than the suboccipitobregmatic diameter). Because of this increased diameter, engagement does not occur until the face is at +2 station.

Fetuses with face presentation may initially begin labor in the brow position. Using x-ray pelvimetry in a series of 7 patients, Borrell and Ferstrom demonstrated that internal rotation occurs between the ischial spines and the ischial tuberosities, making the chin the presenting part, lower than in the vertex presentation. [ 4 , 5 ] Following internal rotation, the mentum is below the maternal symphysis, and delivery occurs by flexion of the fetal neck. As the face descends onto the perineum, the anterior fetal chin passes under the symphysis and flexion of the head occurs, making delivery possible with maternal expulsive forces.

The above mechanisms of labor in the term infant can occur only if the mentum is anterior and at term, only the mentum anterior face presentation is likely to deliver vaginally. If the mentum is posterior or transverse, the fetal neck is too short to span the length of the maternal sacrum and is already at the point of maximal extension. The head cannot deliver as it cannot extend any further through the symphysis and cesarean delivery is the safest route of delivery.

Fortunately, the mentum is anterior in over 60% of cases of face presentation, transverse in 10-12% of cases, and posterior only 20-25% of the time. Fetuses with the mentum transverse position usually rotate to the mentum anterior position, and 25-33% of fetuses with mentum posterior position rotate to a mentum anterior position. When the mentum is posterior, the neck, head and shoulders must enter the pelvis simultaneously, resulting in a diameter too large for the maternal pelvis to accommodate unless in the very preterm or small infant.

Three labor courses are possible when the fetal head engages in a brow presentation. The brow may convert to a vertex presentation, to a face presentation, or remain as a persistent brow presentation. More than 50% of brow presentations will convert to vertex or face presentation and labor courses are managed accordingly when spontaneous conversion occurs.

In the brow presentation, the occipitomental diameter, which is the largest diameter of the fetal head, is the presenting portion. Descent and internal rotation occur only with an adequate pelvis and if the face can fit under the pubic arch. While the head descends, it becomes wedged into the hollow of the sacrum. Downward pressure from uterine contractions and maternal expulsive forces may cause the mentum to extend anteriorly and low to present at the perineum as a mentum anterior face presentation.

If internal rotation does not occur, the occipitomental diameter, which measures 1.5 cm wider than the suboccipitobregmatic diameter and is thus the largest diameter of the fetal head, presents at the pelvic inlet. The head may engage but can descend only with significant molding. This molding and subsequent caput succedaneum over the forehead can become so extensive that identification of the brow by palpation is impossible late in labor. This may result in a missed diagnosis in a patient who presents later in active labor.

If the mentum is anterior and the forces of labor are directed toward the fetal occiput, flexing the head and pivoting the face under the pubic arch, there is conversion to a vertex occiput posterior position. If the occiput lies against the sacrum and the forces of labor are directed against the fetal mentum, the neck may extend further, leading to a face presentation.

The persistent brow presentation with subsequent delivery only occurs in cases of a large pelvis and/or a small infant. Women with gynecoid pelvis or multiparity may be given the option to labor; however, dysfunctional labor and cephalopelvic disproportion are more likely if this presentation persists.

Labor management of face and brow presentation requires close observation of labor progression because cephalopelvic disproportion, dysfunctional labor, and prolonged labor are much more common. As mentioned above, the trachelobregmatic or submentobregmatic diameters are larger than the suboccipitobregmatic diameter. Duration of labor with a face presentation is generally the same as duration of labor with a vertex presentation, although a prolonged labor may occur. As long as maternal or fetal compromise is not evident, labor with a face presentation may continue. [ 6 ] A persistent mentum posterior presentation is an indication for delivery by cesarean section.

Continuous electronic fetal heart rate monitoring is considered mandatory by many authors because of the increased incidence of abnormal fetal heart rate patterns and/or nonreassuring fetal heart rate patterns. [ 7 ] An internal fetal scalp electrode may be used, but very careful application of the electrode must be ensured. The mentum is the recommended site of application. Facial edema is common and can obscure the fetal facial anatomy and improper placement can lead to facial and ophthalmic injuries. Oxytocin can be used to augment labor using the same precautions as in a vertex presentation and the same criteria of assessment of uterine activity, adequacy of the pelvis, and reassuring fetal heart tracing.

Fetuses with face presentation can be delivered vaginally with overall success rates of 60-70%, while more than 20% of fetuses with face presentation require cesarean delivery. Cesarean delivery is performed for the usual obstetrical indications, including arrest of labor and nonreassuring fetal heart rate pattern.

Attempts to manually convert the face to vertex (Thom maneuver) or to rotate a posterior position to a more favorable anterior mentum position are rarely successful and are associated with high fetal morbidity and mortality and maternal morbidity, including cord prolapse, uterine rupture, and fetal cervical spine injury with neurological impairment. Given the availability and safety of cesarean delivery, internal rotation maneuvers are no longer justified unless cesarean section cannot be readily performed.

Internal podalic version and breech extraction are also no longer recommended in the modern management of the face presentation. [ 8 ]

Operative delivery with forceps must be approached with caution. Since engagement occurs when the face is at +2 position, forceps should only be applied to the face that has caused the perineum to bulge. Increased complications to both mother and fetus can occur [ 9 ] and operative delivery must be approached with caution or reserved when cesarean section is not readily available. Forceps may be used if the mentum is anterior. Although the landmarks are different, the application of any forceps is made as if the fetus were presenting directly in the occiput anterior position. The mouth substitutes for the posterior fontanelle, and the mentum substitutes for the occiput. Traction should be downward to maintain extension until the mentum passes under the symphysis, and then gradually elevated to allow the head to deliver by flexion. During delivery, hyperextension of the fetal head should be avoided.

As previously mentioned, the persistent brow presentation has a poor prognosis for vaginal delivery unless the fetus is small, premature, or the maternal pelvis is large. Expectant management is reasonable if labor is progressing well and the fetal well-being is assessed, as there can be spontaneous conversion to face or vertex presentation. The earlier in labor that brow presentation is diagnosed, the higher the likelihood of conversion. Minimal intervention during labor is recommended and some feel the use of oxytocin in the brow presentation is contraindicated.

The use of operative vaginal delivery or manual conversion of a brow to a more favorable presentation is contraindicated as the risks of perinatal morbidity and mortality are unacceptably high. Prolonged, dysfunctional, and arrest of labor are common, necessitating cesarean section delivery.

The incidence of perinatal morbidity and mortality and maternal morbidity has decreased due to the increased incidence of cesarean section delivery for malpresentation, including face and brow presentation.

Neonates delivered in the face presentation exhibit significant facial and skull edema, which usually resolves within 24-48 hours. Trauma during labor may cause tracheal and laryngeal edema immediately after delivery, which can result in neonatal respiratory distress. In addition, fetal anomalies or tumors, such as fetal goiters that may have contributed to fetal malpresentation, may make intubation difficult. Physicians with expertise in neonatal resuscitation should be present at delivery in the event that intubation is required. When a fetal anomaly has been previously diagnosed by ultrasonographic evaluation, the appropriate pediatric specialists should be consulted and informed at time of labor.

Bellussi F, Ghi T, Youssef A, et al. The use of intrapartum ultrasound to diagnose malpositions and cephalic malpresentations. Am J Obstet Gynecol . 2017 Dec. 217 (6):633-41. [QxMD MEDLINE Link] .

[Guideline] Ghi T, Eggebø T, Lees C, et al. ISUOG Practice Guidelines: intrapartum ultrasound. Ultrasound Obstet Gynecol . 2018 Jul. 52 (1):128-39. [QxMD MEDLINE Link] . [Full Text] .

Shaffer BL, Cheng YW, Vargas JE, Laros RK Jr, Caughey AB. Face presentation: predictors and delivery route. Am J Obstet Gynecol . 2006 May. 194(5):e10-2. [QxMD MEDLINE Link] .

Borell U, Fernstrom I. The mechanism of labour. Radiol Clin North Am . 1967 Apr. 5(1):73-85. [QxMD MEDLINE Link] .

Borell U, Fernstrom I. The mechanism of labour in face and brow presentation: a radiographic study. Acta Obstet Gynecol Scand . 1960. 39:626-44.

Gardberg M, Leonova Y, Laakkonen E. Malpresentations--impact on mode of delivery. Acta Obstet Gynecol Scand . 2011 May. 90(5):540-2. [QxMD MEDLINE Link] .

Collaris RJ, Oei SG. External cephalic version: a safe procedure? A systematic review of version-related risks. Acta Obstet Gynecol Scand . 2004 Jun. 83(6):511-8. [QxMD MEDLINE Link] .

Verspyck E, Bisson V, Gromez A, Resch B, Diguet A, Marpeau L. Prophylactic attempt at manual rotation in brow presentation at full dilatation. Acta Obstet Gynecol Scand . 2012 Nov. 91(11):1342-5. [QxMD MEDLINE Link] .

Johnson JH, Figueroa R, Garry D. Immediate maternal and neonatal effects of forceps and vacuum-assisted deliveries. Obstet Gynecol . 2004 Mar. 103(3):513-8. [QxMD MEDLINE Link] .

Benedetti TJ, Lowensohn RI, Truscott AM. Face presentation at term. Obstet Gynecol . 1980 Feb. 55(2):199-202. [QxMD MEDLINE Link] .

BROWNE AD, CARNEY D. OBSTETRICS IN GENERAL PRACTICE. MANAGEMENT OF MALPRESENTATIONS IN OBSTETRICS. Br Med J . 1964 May 16. 1(5393):1295-8. [QxMD MEDLINE Link] .

Campbell JM. Face presentation. Aust N Z J Obstet Gynaecol . 1965 Nov. 5(4):231-4. [QxMD MEDLINE Link] .

Contributor Information and Disclosures

Teresa Marino, MD Assistant Professor, Attending Physician, Division of Maternal-Fetal Medicine, Tufts Medical Center Disclosure: Nothing to disclose.

Francisco Talavera, PharmD, PhD Adjunct Assistant Professor, University of Nebraska Medical Center College of Pharmacy; Editor-in-Chief, Medscape Drug Reference Disclosure: Received salary from Medscape for employment. for: Medscape.

Carl V Smith, MD The Distinguished Chris J and Marie A Olson Chair of Obstetrics and Gynecology, Professor, Department of Obstetrics and Gynecology, Senior Associate Dean for Clinical Affairs, University of Nebraska Medical Center Carl V Smith, MD is a member of the following medical societies: American College of Obstetricians and Gynecologists , American Institute of Ultrasound in Medicine , Association of Professors of Gynecology and Obstetrics , Central Association of Obstetricians and Gynecologists , Society for Maternal-Fetal Medicine , Council of University Chairs of Obstetrics and Gynecology , Nebraska Medical Association Disclosure: Nothing to disclose.

Chitra M Iyer, MD, Perinatologist, Obstetrix Medical Group, Fort Worth, Texas.

Chitra M Iyer, MD is a member of the following medical societies: American College of Obstetricians and Gynecologists , Society of Maternal-Fetal Medicine .

Disclosure: Nothing to disclose.

What would you like to print?

• Print this section
• Print the entire contents of
• Print the entire contents of article

• HIV in Pregnancy
• Pulmonary Disease and Pregnancy
• Kidney Disease and Pregnancy
• Vaccinations/Immunizations During Pregnancy
• Anemia and Thrombocytopenia in Pregnancy
• Cardiovascular Disease and Pregnancy
• Common Pregnancy Complaints and Questions
• Is immunotherapy for cancer safe in pregnancy?
• Labetalol, Nifedipine: Outcome on Pregnancy Hypertension
• The Pregnancy Challenges of Women With Chronic Conditions

• Drug Interaction Checker
• Pill Identifier
• Calculators

• 2020/viewarticle/immunotherapy-cancer-safe-pregnancy-2024a100083dnews news Is immunotherapy for cancer safe in pregnancy?

• 2002246123-overviewDiseases & Conditions Diseases & Conditions Kidney Disease and Pregnancy

Fetal Presentation, Position, and Lie (Including Breech Presentation)

• Key Points |

Abnormal fetal lie or presentation may occur due to fetal size, fetal anomalies, uterine structural abnormalities, multiple gestation, or other factors. Diagnosis is by examination or ultrasonography. Management is with physical maneuvers to reposition the fetus, operative vaginal delivery , or cesarean delivery .

Terms that describe the fetus in relation to the uterus, cervix, and maternal pelvis are

Fetal presentation: Fetal part that overlies the maternal pelvic inlet; vertex (cephalic), face, brow, breech, shoulder, funic (umbilical cord), or compound (more than one part, eg, shoulder and hand)

Fetal position: Relation of the presenting part to an anatomic axis; for vertex presentation, occiput anterior, occiput posterior, occiput transverse

Fetal lie: Relation of the fetus to the long axis of the uterus; longitudinal, oblique, or transverse

Normal fetal lie is longitudinal, normal presentation is vertex, and occiput anterior is the most common position.

Abnormal fetal lie, presentation, or position may occur with

Fetopelvic disproportion (fetus too large for the pelvic inlet)

Fetal congenital anomalies

Uterine structural abnormalities (eg, fibroids, synechiae)

Multiple gestation

Several common types of abnormal lie or presentation are discussed here.

Transverse lie

Fetal position is transverse, with the fetal long axis oblique or perpendicular rather than parallel to the maternal long axis. Transverse lie is often accompanied by shoulder presentation, which requires cesarean delivery.

Breech presentation

There are several types of breech presentation.

Frank breech: The fetal hips are flexed, and the knees extended (pike position).

Complete breech: The fetus seems to be sitting with hips and knees flexed.

Single or double footling presentation: One or both legs are completely extended and present before the buttocks.

Types of breech presentations

Breech presentation makes delivery difficult ,primarily because the presenting part is a poor dilating wedge. Having a poor dilating wedge can lead to incomplete cervical dilation, because the presenting part is narrower than the head that follows. The head, which is the part with the largest diameter, can then be trapped during delivery.

Additionally, the trapped fetal head can compress the umbilical cord if the fetal umbilicus is visible at the introitus, particularly in primiparas whose pelvic tissues have not been dilated by previous deliveries. Umbilical cord compression may cause fetal hypoxemia.

Predisposing factors for breech presentation include

Preterm labor

Uterine abnormalities

Fetal anomalies

If delivery is vaginal, breech presentation may increase risk of

Umbilical cord prolapse

Birth trauma

Perinatal death

Face or brow presentation

In face presentation, the head is hyperextended, and position is designated by the position of the chin (mentum). When the chin is posterior, the head is less likely to rotate and less likely to deliver vaginally, necessitating cesarean delivery.

Brow presentation usually converts spontaneously to vertex or face presentation.

Occiput posterior position

The most common abnormal position is occiput posterior.

The fetal neck is usually somewhat deflexed; thus, a larger diameter of the head must pass through the pelvis.

Progress may arrest in the second phase of labor. Operative vaginal delivery or cesarean delivery is often required.

Position and Presentation of the Fetus

If a fetus is in the occiput posterior position, operative vaginal delivery or cesarean delivery is often required.

In breech presentation, the presenting part is a poor dilating wedge, which can cause the head to be trapped during delivery, often compressing the umbilical cord.

For breech presentation, usually do cesarean delivery at 39 weeks or during labor, but external cephalic version is sometimes successful before labor, usually at 37 or 38 weeks.

Copyright © 2024 Merck & Co., Inc., Rahway, NJ, USA and its affiliates. All rights reserved.

• Cookie Preferences

• Physician Physician Board Reviews Physician Associate Board Reviews CME Lifetime CME Free CME MATE and DEA Compliance
• Student USMLE Step 1 USMLE Step 2 USMLE Step 3 COMLEX Level 1 COMLEX Level 2 COMLEX Level 3 96 Medical School Exams Student Resource Center NCLEX - RN NCLEX - LPN/LVN/PN 24 Nursing Exams
• Nurse Practitioner APRN/NP Board Reviews CNS Certification Reviews CE - Nurse Practitioner FREE CE
• Nurse RN Certification Reviews CE - Nurse FREE CE
• Pharmacist Pharmacy Board Exam Prep CE - Pharmacist
• Allied Allied Health Exam Prep Dentist Exams CE - Social Worker CE - Dentist
• Point of Care
• Free CME/CE

Delivery, Face and Brow Presentation

Introduction.

The term presentation describes the leading part of the fetus or the anatomical structure closest to the maternal pelvic inlet during labor. The presentation can roughly be divided into the following classifications: cephalic, breech, shoulder, and compound. Cephalic presentation is the most common and can be further subclassified as vertex, sinciput, brow, face, and chin. The most common presentation in term labor is the vertex, where the fetal neck is flexed to the chin, minimizing the head circumference. Face presentation is an abnormal form of cephalic presentation where the presenting part is the mentum. This typically occurs because of hyperextension of the neck and the occiput touching the fetal back. Incidence of face presentation is rare, accounting for approximately 1 in 600 of all presentations. [1] [2] [3]  In brow presentation, the neck is not extended as much as in face presentation, and the leading part is the area between the anterior fontanelle and the orbital ridges. Brow presentation is considered the rarest of all malpresentation, with a prevalence of 1 in 500 to 1 in 4000 deliveries. [3]

Both face and brow presentations occur due to extension of the fetal neck instead of flexion; therefore, conditions that would lead to hyperextension or prevent flexion of the fetal neck can all contribute to face or brow presentation. These risk factors may be related to either the mother or the fetus. Maternal risk factors are preterm delivery, contracted maternal pelvis, platypelloid pelvis, multiparity, previous cesarean section, and black race. Fetal risk factors include anencephaly, multiple loops of cord around the neck, masses of the neck, macrosomia, and polyhydramnios. [2] [4] [5]  These malpresentations are usually diagnosed during the second stage of labor when performing a digital examination. Palpating orbital ridges, nose, malar eminences, mentum, mouth, gums, and chin in face presentation is possible. Based on the position of the chin, face presentation can be further divided into mentum anterior, posterior, or transverse. In brow presentation, the anterior fontanelle and face can be palpated except for the mouth and the chin. Brow presentation can then be further described based on the position of the anterior fontanelle as frontal anterior, posterior, or transverse. Diagnosing the exact presentation can be challenging, and face presentation may be misdiagnosed as frank breech. To avoid any confusion, a bedside ultrasound scan can be performed. [6]  Ultrasound imaging can show a reduced angle between the occiput and the spine or the chin is separated from the chest. However, ultrasound does not provide much predictive value for the outcome of labor. [7]

Anatomy and Physiology

Register for free and read the full article, learn more about a subscription to statpearls point-of-care.

Before discussing the mechanism of labor in the face or brow presentation, it is crucial to highlight some anatomical landmarks and their measurements. 

Planes and Diameters of the Pelvis

The 3 most important planes in the female pelvis are the pelvic inlet, mid-pelvis, and pelvic outlet. Four diameters can describe the pelvic inlet: anteroposterior, transverse, and 2 obliques. Furthermore, based on the landmarks on the pelvic inlet, there are 3 different anteroposterior diameters named conjugates: true conjugate, obstetrical conjugate, and diagonal conjugate. Only the latter can be measured directly during the obstetric examination. The shortest of these 3 diameters is obstetrical conjugate, which measures approximately 10.5 cm and is the distance between the sacral promontory and 1 cm below the upper border of the symphysis pubis. This measurement is clinically significant as the fetal head must pass through this diameter during the engagement phase. The transverse diameter measures about 13.5 cm and is the widest distance between the innominate line on both sides. The shortest distance in the mid pelvis is the interspinous diameter and usually is only about 10 cm. 

Fetal Skull Diameters

There are 6 distinguished longitudinal fetal skull diameters:

• Suboccipito-bregmatic: from the center of anterior fontanelle (bregma) to the occipital protuberance, measuring 9.5 cm. This is the diameter presented in the vertex presentation. 
• Suboccipito-frontal: from the anterior part of bregma to the occipital protuberance, measuring 10 cm 
• Occipito-frontal: from the root of the nose to the most prominent part of the occiput, measuring 11.5 cm
• Submento-bregmatic: from the center of the bregma to the angle of the mandible, measuring 9.5 cm. This is the diameter in the face presentation where the neck is hyperextended. 
• Submento-vertical: from the midpoint between fontanelles and the angle of the mandible, measuring 11.5 cm 
• Occipito-mental: from the midpoint between fontanelles and the tip of the chin, measuring 13.5 cm. It is the presenting diameter in brow presentation. 

Cardinal Movements of Normal Labor

• Neck flexion
• Internal rotation
• Extension (delivers head)
• External rotation (restitution)
• Expulsion (delivery of anterior and posterior shoulders)

Some key movements are impossible in the face or brow presentations. Based on the information provided above, it is obvious that labor be arrested in brow presentation unless it spontaneously changes to the face or vertex, as the occipito-mental diameter of the fetal head is significantly wider than the smallest diameter of the female pelvis. Face presentation can, however, be delivered vaginally, and further mechanisms of face delivery are explained in later sections.

Indications

As mentioned previously, spontaneous vaginal delivery can be successful in face presentation. However, the main indication for vaginal delivery in such circumstances would be a maternal choice. It is crucial to have a thorough conversation with a mother, explaining the risks and benefits of vaginal delivery with face presentation and a cesarean section. Informed consent and creating a rapport with the mother is an essential aspect of safe and successful labor.

Contraindications

Vaginal delivery of face presentation is contraindicated if the mentum is lying posteriorly or is in a transverse position. In such a scenario, the fetal brow is pressing against the maternal symphysis pubis, and the short fetal neck, which is already maximally extended, cannot span the surface of the maternal sacrum. In this position, the diameter of the head is larger than the maternal pelvis, and it cannot descend through the birth canal. Therefore, the cesarean section is recommended as the safest mode of delivery for mentum posterior face presentations. Attempts to manually convert face presentation to vertex, manual or forceps rotation of the persistent posterior chin to anterior are contraindicated as they can be dangerous. Persistent brow presentation itself is a contraindication for vaginal delivery unless the fetus is significantly small or the maternal pelvis is large.

Continuous electronic fetal heart rate monitoring is recommended for face and brow presentations, as heart rate abnormalities are common in these scenarios. One study found that only 14% of the cases with face presentation had no abnormal traces on the cardiotocograph. [8]  External transducer devices are advised to prevent damage to the eyes. When internal monitoring is inevitable, monitoring devices on bony parts should be placed carefully. 

Consultations that are typically requested for patients with delivery of face/brow presentation include the following:

• Experienced midwife, preferably looking after laboring women 1:1
• Senior obstetrician 
• Neonatal team - in case of need for resuscitation 
• Anesthetic team - to provide necessary pain control (eg, epidural)
• Theatre team  - in case of failure to progress, an emergency cesarean section is required.

Preparation

No specific preparation is required for face or brow presentation. However, discussing the labor options with the mother and birthing partner and informing members of the neonatal, anesthetic, and theatre co-ordinating teams is essential.

Technique or Treatment

Mechanism of Labor in Face Presentation

During contractions, the pressure exerted by the fundus of the uterus on the fetus and the pressure of the amniotic fluid initiate descent. During this descent, the fetal neck extends instead of flexing. The internal rotation determines the outcome of delivery. If the fetal chin rotates posteriorly, vaginal delivery would not be possible, and cesarean section is permitted. The approach towards mentum-posterior delivery should be individualized, as the cases are rare. Expectant management is acceptable in multiparous women with small fetuses, as a spontaneous mentum-anterior rotation can occur. However, there should be a low threshold for cesarean section in primigravida women or women with large fetuses.

The pubis is described as mentum-anterior when the fetal chin is rotated towards the maternal symphysis. In these cases, further descent through the vaginal canal continues, with approximately 73% of cases delivering spontaneously. [9]  The fetal mentum presses on the maternal symphysis pubis, and the head is delivered by flexion. The occiput is pointing towards the maternal back, and external rotation happens. Shoulders are delivered in the same manner as in vertex delivery.

Mechanism of Labor in Brow Presentation

As this presentation is considered unstable, it is usually converted into a face or an occiput presentation. Due to the cephalic diameter being wider than the maternal pelvis, the fetal head cannot engage; thus, brow delivery cannot occur. Unless the fetus is small or the pelvis is very wide, the prognosis for vaginal delivery is poor. With persistent brow presentation, a cesarean section is required for safe delivery.

Complications

As the cesarean section is becoming a more accessible mode of delivery in malpresentations, the incidence of maternal and fetal morbidity and mortality during face presentation has dropped significantly. [10]  However, some complications are still associated with the nature of labor in face presentation. Due to the fetal head position, it is more challenging for the head to engage in the birth canal and descend, resulting in prolonged labor. Prolonged labor itself can provoke fetal distress and arrhythmias. If the labor arrests or signs of fetal distress appear on CTG, the recommended next step in management is an emergency cesarean section, which in itself carries a myriad of operative and post-operative complications. Finally, due to the nature of the fetal position and prolonged duration of labor in face presentation, neonates develop significant edema of the skull and face. Swelling of the fetal airway may also be present, resulting in respiratory distress after birth and possible intubation.

Clinical Significance

During vertex presentation, the fetal head flexes, bringing the chin to the chest, forming the smallest possible fetal head diameter, measuring approximately 9.5 cm. With face and brow presentation, the neck hyperextends, resulting in greater cephalic diameters. As a result, the fetal head engages later, and labor progresses more slowly. Failure to progress in labor is also more common in both presentations compared to the vertex presentation. Furthermore, when the fetal chin is in a posterior position, this prevents further flexion of the fetal neck, as browns are pressing on the symphysis pubis. As a result, descending through the birth canal is impossible. Such presentation is considered undeliverable vaginally and requires an emergency cesarean section. Manual attempts to change face presentation to vertex or manual or forceps rotation to mentum anterior are considered dangerous and discouraged.

Enhancing Healthcare Team Outcomes

A multidisciplinary team of healthcare experts supports the woman and her child during labor and the perinatal period. For a face or brow presentation to be appropriately diagnosed, an experienced midwife and obstetrician must be involved in the vaginal examination and labor monitoring. As fetal anomalies, such as anencephaly or goiter, can contribute to face presentation, sonographers experienced in antenatal scanning should also be involved in the care. It is advised to inform the anesthetic and neonatal teams in advance of the possible need for emergency cesarean section and resuscitation of the neonate. [11] [12]

Gardberg M, Leonova Y, Laakkonen E. Malpresentations--impact on mode of delivery. Acta obstetricia et gynecologica Scandinavica. 2011 May:90(5):540-2. doi: 10.1111/j.1600-0412.2011.01105.x. Epub     [PubMed PMID: 21501123]

Tapisiz OL, Aytan H, Altinbas SK, Arman F, Tuncay G, Besli M, Mollamahmutoglu L, Danışman N. Face presentation at term: a forgotten issue. The journal of obstetrics and gynaecology research. 2014 Jun:40(6):1573-7. doi: 10.1111/jog.12369. Epub     [PubMed PMID: 24888918]

Zayed F, Amarin Z, Obeidat B, Obeidat N, Alchalabi H, Lataifeh I. Face and brow presentation in northern Jordan, over a decade of experience. Archives of gynecology and obstetrics. 2008 Nov:278(5):427-30. doi: 10.1007/s00404-008-0600-0. Epub 2008 Feb 19     [PubMed PMID: 18283473]

Bashiri A,Burstein E,Bar-David J,Levy A,Mazor M, Face and brow presentation: independent risk factors. The journal of maternal-fetal     [PubMed PMID: 18570114]

Shaffer BL, Cheng YW, Vargas JE, Laros RK Jr, Caughey AB. Face presentation: predictors and delivery route. American journal of obstetrics and gynecology. 2006 May:194(5):e10-2     [PubMed PMID: 16647888]

Bellussi F, Ghi T, Youssef A, Salsi G, Giorgetta F, Parma D, Simonazzi G, Pilu G. The use of intrapartum ultrasound to diagnose malpositions and cephalic malpresentations. American journal of obstetrics and gynecology. 2017 Dec:217(6):633-641. doi: 10.1016/j.ajog.2017.07.025. Epub 2017 Jul 22     [PubMed PMID: 28743440]

Ghi T, Eggebø T, Lees C, Kalache K, Rozenberg P, Youssef A, Salomon LJ, Tutschek B. ISUOG Practice Guidelines: intrapartum ultrasound. Ultrasound in obstetrics & gynecology : the official journal of the International Society of Ultrasound in Obstetrics and Gynecology. 2018 Jul:52(1):128-139. doi: 10.1002/uog.19072. Epub     [PubMed PMID: 29974596]

Benedetti TJ, Lowensohn RI, Truscott AM. Face presentation at term. Obstetrics and gynecology. 1980 Feb:55(2):199-202     [PubMed PMID: 7352081]

Ducarme G, Ceccaldi PF, Chesnoy V, Robinet G, Gabriel R. [Face presentation: retrospective study of 32 cases at term]. Gynecologie, obstetrique & fertilite. 2006 May:34(5):393-6     [PubMed PMID: 16630740]

Cruikshank DP, Cruikshank JE. Face and brow presentation: a review. Clinical obstetrics and gynecology. 1981 Jun:24(2):333-51     [PubMed PMID: 7307363]

Domingues AP, Belo A, Moura P, Vieira DN. Medico-legal litigation in Obstetrics: a characterization analysis of a decade in Portugal. Revista brasileira de ginecologia e obstetricia : revista da Federacao Brasileira das Sociedades de Ginecologia e Obstetricia. 2015 May:37(5):241-6. doi: 10.1590/SO100-720320150005304. Epub     [PubMed PMID: 26107576]

. Intrapartum care for healthy women and babies. 2022 Dec 14:():     [PubMed PMID: 32212591]

Use the mouse wheel to zoom in and out, click and drag to pan the image

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed, delivery, face and brow presentation, affiliations.

• 1 Vilnius University, Lithuania, Imperial London Healthcare NHS Trust
• 2 University of Health Sciences, Rawalpindi Medical College
• PMID: 33620804
• Bookshelf ID: NBK567727

The term presentation describes the leading part of the fetus or the anatomical structure closest to the maternal pelvic inlet during labor. The presentation can roughly be divided into the following classifications: cephalic, breech, shoulder, and compound. Cephalic presentation is the most common and can be further subclassified as vertex, sinciput, brow, face, and chin. The most common presentation in term labor is the vertex, where the fetal neck is flexed to the chin, minimizing the head circumference. Face presentation is an abnormal form of cephalic presentation where the presenting part is the mentum. This typically occurs because of hyperextension of the neck and the occiput touching the fetal back. Incidence of face presentation is rare, accounting for approximately 1 in 600 of all presentations. In brow presentation, the neck is not extended as much as in face presentation, and the leading part is the area between the anterior fontanelle and the orbital ridges. Brow presentation is considered the rarest of all malpresentation, with a prevalence of 1 in 500 to 1 in 4000 deliveries.

Both face and brow presentations occur due to extension of the fetal neck instead of flexion; therefore, conditions that would lead to hyperextension or prevent flexion of the fetal neck can all contribute to face or brow presentation. These risk factors may be related to either the mother or the fetus. Maternal risk factors are preterm delivery, contracted maternal pelvis, platypelloid pelvis, multiparity, previous cesarean section, and black race. Fetal risk factors include anencephaly, multiple loops of cord around the neck, masses of the neck, macrosomia, and polyhydramnios. These malpresentations are usually diagnosed during the second stage of labor when performing a digital examination. Palpating orbital ridges, nose, malar eminences, mentum, mouth, gums, and chin in face presentation is possible. Based on the position of the chin, face presentation can be further divided into mentum anterior, posterior, or transverse. In brow presentation, the anterior fontanelle and face can be palpated except for the mouth and the chin. Brow presentation can then be further described based on the position of the anterior fontanelle as frontal anterior, posterior, or transverse. Diagnosing the exact presentation can be challenging, and face presentation may be misdiagnosed as frank breech. To avoid any confusion, a bedside ultrasound scan can be performed. Ultrasound imaging can show a reduced angle between the occiput and the spine or the chin is separated from the chest. However, ultrasound does not provide much predictive value for the outcome of labor.

Copyright © 2024, StatPearls Publishing LLC.

PubMed Disclaimer

Conflict of interest statement

Disclosure: Julija Makajeva declares no relevant financial relationships with ineligible companies.

Disclosure: Mohsina Ashraf declares no relevant financial relationships with ineligible companies.

• Continuing Education Activity
• Introduction
• Anatomy and Physiology
• Indications
• Contraindications
• Preparation
• Technique or Treatment
• Complications
• Clinical Significance
• Enhancing Healthcare Team Outcomes
• Review Questions

Similar articles

• Sonographic diagnosis of fetal head deflexion and the risk of cesarean delivery. Bellussi F, Livi A, Cataneo I, Salsi G, Lenzi J, Pilu G. Bellussi F, et al. Am J Obstet Gynecol MFM. 2020 Nov;2(4):100217. doi: 10.1016/j.ajogmf.2020.100217. Epub 2020 Aug 18. Am J Obstet Gynecol MFM. 2020. PMID: 33345926
• Sonographic evaluation of the fetal head position and attitude during labor. Ghi T, Dall'Asta A. Ghi T, et al. Am J Obstet Gynecol. 2024 Mar;230(3S):S890-S900. doi: 10.1016/j.ajog.2022.06.003. Epub 2023 May 19. Am J Obstet Gynecol. 2024. PMID: 37278991 Review.
• Leopold Maneuvers. Superville SS, Siccardi MA. Superville SS, et al. 2024 Jul 27. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan–. 2024 Jul 27. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2024 Jan–. PMID: 32809649 Free Books & Documents.
• Intrapartum sonographic assessment of the fetal head flexion in protracted active phase of labor and association with labor outcome: a multicenter, prospective study. Dall'Asta A, Rizzo G, Masturzo B, Di Pasquo E, Schera GBL, Morganelli G, Ramirez Zegarra R, Maqina P, Mappa I, Parpinel G, Attini R, Roletti E, Menato G, Frusca T, Ghi T. Dall'Asta A, et al. Am J Obstet Gynecol. 2021 Aug;225(2):171.e1-171.e12. doi: 10.1016/j.ajog.2021.02.035. Epub 2021 Mar 4. Am J Obstet Gynecol. 2021. PMID: 33675795
• Labor with abnormal presentation and position. Stitely ML, Gherman RB. Stitely ML, et al. Obstet Gynecol Clin North Am. 2005 Jun;32(2):165-79. doi: 10.1016/j.ogc.2004.12.005. Obstet Gynecol Clin North Am. 2005. PMID: 15899353 Review.
• Gardberg M, Leonova Y, Laakkonen E. Malpresentations--impact on mode of delivery. Acta Obstet Gynecol Scand. 2011 May;90(5):540-2. - PubMed
• Tapisiz OL, Aytan H, Altinbas SK, Arman F, Tuncay G, Besli M, Mollamahmutoglu L, Danışman N. Face presentation at term: a forgotten issue. J Obstet Gynaecol Res. 2014 Jun;40(6):1573-7. - PubMed
• Zayed F, Amarin Z, Obeidat B, Obeidat N, Alchalabi H, Lataifeh I. Face and brow presentation in northern Jordan, over a decade of experience. Arch Gynecol Obstet. 2008 Nov;278(5):427-30. - PubMed
• Bashiri A, Burstein E, Bar-David J, Levy A, Mazor M. Face and brow presentation: independent risk factors. J Matern Fetal Neonatal Med. 2008 Jun;21(6):357-60. - PubMed
• Shaffer BL, Cheng YW, Vargas JE, Laros RK, Caughey AB. Face presentation: predictors and delivery route. Am J Obstet Gynecol. 2006 May;194(5):e10-2. - PubMed

Publication types

• Search in PubMed
• Search in MeSH
• Add to Search

Related information

Linkout - more resources, full text sources.

• NCBI Bookshelf

• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Face presentation at term: incidence, risk factors and influence on maternal and neonatal outcomes

• Maternal-Fetal Medicine
• Published: 09 April 2024
• Volume 310 , pages 923–931, ( 2024 )

Cite this article

• Yongqing Zhang 1   na1 ,
• Tiantian Fu 1   na1 ,
• Luping Chen 1 ,
• Yinluan Ouyang 1 ,
• Xiujun Han 1 &
• Danqing Chen   ORCID: orcid.org/0000-0002-0201-7215 1  

143 Accesses

Explore all metrics

The incidence, diagnosis, management and outcome of face presentation at term were analysed.

A retrospective, gestational age-matched case–control study including 27 singletons with face presentation at term was conducted between April 2006 and February 2021. For each case, four women who had the same gestational age and delivered in the same month with vertex position and singletons were selected as the controls (control group, n = 108). Conditional logistic regression was used to assess the risk factors of face presentation. The maternal and neonatal outcomes of the face presentation group were followed up.

The incidence of face presentation at term was 0.14‰. After conditional logistic regression, the two factors associated with face presentation were high parity (adjusted odds ratio [aOR] 2.76, 95% CI 1.19–6.39)] and amniotic fluid index > 18 cm (aOR 2.60, 95% CI 1.08–6.27). Among the 27 cases, the diagnosis was made before the onset of labor, during the latent phase of labor, during the active phase of labor, and during the cesarean section in 3.7% (1/27), 40.7% (11/27), 11.1% (3/27) and 44.4% (12/27) of cases, respectively. In one case of cervical dilation with a diameter of 5 cm, we innovatively used a vaginal speculum for rapid diagnosis of face presentation. The rate of cesarean section and postpartum haemorrhage ≥ 500 ml in the face presentation group was higher than that of the control group (88.9% vs. 13.9%, P  < 0.001, and 14.8% vs. 2.8%, P  = 0.024), but the Apgar scores were similar in both sets of newborns. Among the 27 cases of face presentation, there were three cases of adverse maternal and neonatal outcomes, including one case of neonatal right brachial plexus injury and two cases of severe laceration of the lower segment of the uterus with postpartum haemorrhage ≥ 1000 ml.

Conclusions

Face presentation was rare. Early diagnosis is difficult, and thus easily neglected. High parity and amniotic fluid index > 18 cm are risk factors for face presentation. An early diagnosis and proper management of face presentation could lead to good maternal and neonatal outcomes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Prognosis for deliveries in face presentation: a case–control study

Prenatal factors related to face presentation: a case–control study.

Critical analysis of risk factors for intrapartum fetal death

Explore related subjects.

• Medical Ethics
• Medical Imaging

Data availability

The original data can be provided by E-mail if needed.

Code availability

Cunningham FG, Williams JW (2018) Williams obstetrics, 25th edn. McGraw-Hill Medical, New York, pp 450–452

Google Scholar  

Arsène E, Langlois C, Clouqueur E, Deruelle P, Subtil D (2019) Prognosis for deliveries in face presentation: a case-control study. Arch Gynecol Obstet 300:869–874. https://doi.org/10.1007/s00404-019-05241-6

Article   PubMed   Google Scholar  

Cruikshank DP, Cruikshank RN (1981) Face and brow presentation. A review. Clin Obstet Gynecol 24:333–351. https://doi.org/10.1097/00003081-198106000-00003

Article   CAS   PubMed   Google Scholar  

Schwartz Z, Dgani R, Lancet M, Kessler I (1986) Face presentation. Aust N Z J Obstet Gynaecol 26:172–176. https://doi.org/10.1111/j.1479-828x.1986.tb01560.x

Bashiri A, Burstein E, Bar-David J, Levy A, Mazor M (2008) Face and brow presentation: independent risk factors. J Matern Fetal Neonatal Med 21:357–360. https://doi.org/10.1080/14767050802037647

Arsene E, Langlois C, Garabedian C, Clouqueur E, Deruelle P, Subtil D (2016) Prenatal factors related to face presentation: a case–control study. Arch Gynecol Obstet 294:279–284. https://doi.org/10.1007/s00404-015-3992-7

Tapisiz OL, Aytan H, Altinbas SK, Arman F, Tuncay G, Besli M et al (2014) Face presentation at term: a forgotten issue. J Obstet Gynaecol Res 40:1573–1577. https://doi.org/10.1111/jog.12369

Lau WL, Cho LY, Leung WC (2011) Intrapartum translabial ultrasound demonstration of face presentation during first stage of labor. J Obstet Gynaecol Res 37:1868–1871. https://doi.org/10.1111/j.1447-0756.2011.01650.x

Shaffer BL, Cheng YW, Vargas JE, Laros RK Jr, Caughey AB (2006) Face presentation: predictors and delivery route. Am J Obstet Gynecol 194:e10-12

Benedetti TJ, Lowenshon RI, Truscott AM (1980) Face presentation at term. Obstet Gynecol 55:199–202

CAS   PubMed   Google Scholar  

Pilliod RA, Caughey AB (2017) Fetal malpresentation and malposition: diagnosis and management. Obstet Gynecol Clin N Am 44:631–643. https://doi.org/10.1016/j.ogc.2017.08.003

Article   Google Scholar  

Bellussi F, Ghi T, Youssef A, Salsi G, Giorgetta F, Parma D et al (2017) The use of intrapartum ultrasound to diagnose malpositions and cephalic malpresentations. Am J Obstet Gynecol 217:633–641. https://doi.org/10.1016/j.ajog.2017.07.025

Ducarme G, Ceccaldi PF, Chesnoy V, Robinet G, Gabriel R (2006) Face presentation: retrospective study of 32 cases at term. Gynecol Obstet Fertil 34:393–396. https://doi.org/10.1016/j.gyobfe.2005.07.042

De Bernardo G, Svelto M, Giordano M, Sordino D (2017) Face presentation in delivery room: what is strategy? BMJ Case Rep. https://doi.org/10.1136/bcr-2016-219114

Article   PubMed   PubMed Central   Google Scholar  

Manning JB, Tolcher MC, Chandraharan E, Rose CH (2015) Delivery of an impacted fetal head during cesarean: a literature review and proposed management algorithm. Obstet Gynecol Surv 70:719–724. https://doi.org/10.1097/OGX.0000000000000248

Download references

Acknowledgements

The authors wish to acknowledge Menglin Zhou, Zhengyun Chen and Guohui Yan for their valuable assistance for the manuscript.

No specific funding was obtained for this study.

Author information

Yongqing Zhang and Tiantian Fu have contributed equally to this work.

Authors and Affiliations

Department of Obstetrics, School of Medicine, Women’s Hospital, Zhejiang University, 1st Xueshi Road, Hangzhou, 310006, Zhejiang, People’s Republic of China

Yongqing Zhang, Tiantian Fu, Luping Chen, Yinluan Ouyang, Xiujun Han & Danqing Chen

You can also search for this author in PubMed   Google Scholar

Contributions

YZ: conceptualization, methodology, writing—original draft. TF: conceptualization, formal analysis, writing—original draft. LC: data collection, follow-up. YO: investigation, resources. XH: investigation, formal analysis, supervision. DC: conceptualization, writing—review and editing, supervision. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Xiujun Han or Danqing Chen .

Ethics declarations

Conflict of interest.

All authors have no conflicts of interest to disclose.

Ethics approval

Our study was planned in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Local Ethics Committee of the Women’s hospital, school of medicine, Zhejiang university ( Ethical No. IRB-20210211-R ).

Consent for publication

Consent to participate.

As this was a retrospective study, written informed consents were not obtained, but all patients’ records/information were anonymized before analysis. The images used in this paper (Figs. 1 , 2 , 3 ) were given informed consent by the pregnant woman.

Human and animal rights

This article does not contain any studies with animals performed by any of the authors.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Zhang, Y., Fu, T., Chen, L. et al. Face presentation at term: incidence, risk factors and influence on maternal and neonatal outcomes. Arch Gynecol Obstet 310 , 923–931 (2024). https://doi.org/10.1007/s00404-024-07406-4

Download citation

Received : 09 July 2023

Accepted : 28 January 2024

Published : 09 April 2024

Issue Date : August 2024

DOI : https://doi.org/10.1007/s00404-024-07406-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Face presentation
• Risk factors
• Brachial plexus injury
• Postpartum haemorrhage
• Find a journal
• Publish with us
• Track your research

• 2022 New Pearls of Exxcellence Articles

Management of Brow, Face, and Compound Malpresentations

Author: Meera Kesavan, MD

Mentor: Lisa Keder MD Editor: Daniel JS Martingano DO MBA PhD

Registered users can also download a PDF or listen to a podcast of this Pearl. Log in now , or create a free account to access bonus Pearls features.

Fetal malpresentation, including brow, face, or compound presentations, complicates around 3-4% of all term births. Because these abnormal fetal presentations still are cephalic, many such cases result in vaginal deliveries, yet there are increased risks for adverse outcomes, including cesarean delivery resultant surgical complications, persistent malpresentation precluding vaginal delivery, and abnormal labor resulting in arrest of dilation or descent.

These fetal malpresentation are differentiated in the following ways:

• In face presentations, the presenting part is the mentum, which is further divided based on its position, including mentum posterior, mentum transverse or mentum anterior positions. This typically occurs because of hyperextension of the neck and the occiput touching the fetal back. Mentum anterior malpresentations can potentially achieve vaginal deliveries, whereas mentum posterior malpresentations cannot.
• In brow presentations, there is less extension of the fetal neck as in face presentations making the leading fetal part being the area between the anterior fontanelle and the orbital ridges. These presentations are uncommon and are managed similarly to face presentations. Brow presentation can be further described based on the position of the anterior fontanelle as frontal anterior, posterior, or transverse.
• Compound presentation is defined as the leading fetal part, including a fetal extremity, alongside a cephalic or breech presentation. Management of compound presentations is expected (and often incidentally noted following delivery) because the extremity will often either retract as the head descends or will feasibly allow for delivery in its current position, with manipulation attempts to reduce the compound presentation usually avoided.

Risk factors for brow and face presentations include fetal CNS malformations, congenital or chromosomal anomalies, advanced maternal age, low birthweight, abnormal maternal pelvic anatomy (e.g. contracted pelvis, cephalopelvic disporotion, platypelloid pelvis, etc.) and nulliparity. non-Hispanic White women have the highest risk for malpresentation, whereas non-Hispanic Black women have the lowest risk.

Diagnosis usually is made during the second stage of labor while performing routine vaingla examinations and involves palpation of the abnormal leading fetal part (forehead, orbital ridge, orbits, nose, etc.) Obstetric ultrasound can additionally provide complimentary information to support these diagnoses and distinguish from other fetal malpresentations or malpositions. In face presentation, the mentum (chin) and mouth are palpable.

Management considerations for face, brow, and compounds presentations are unique with compound presentations having higher rates of vaginal delivery and lower complications as compared to either brow or face presentations.

• For brow presentations, approximately 30-40% of brow presentations will convert to a face presentation, and about 20% will convert to a vertex presentation. Anterior positions have the possibility of vaginal deliveries and can be managed by usual labor management principles, whereas mentum posterior positions are indications for cesarean delivery.
• For face presentations, the likelihood of vaginal delivery depends on the orientation of the mentum, with mentum anterior being most suitable for vaginal delivery. If the fetus is mentum posterior, flexion of the neck is precluded and results in the inability of fetal descent.
• For compound presentations, management is expectant and manipulation of the leading extremities should be avoided. Most cases of compound presentation result in vaginal deliveries. For term deliveries, compound presentations with parts other than the hand are unlikely to result in safe vaginal delivery.

Labor management for brow and face presentation overall involves continuous fetal heart rate monitoring and repeat clinical assessments, given the increased potential of fetal complications as noted. Caution should be used with internal monitoring devices, which can cause ophthalmic injury or trauma to the presenting fetal parts, with the use of fetal scalp electrodes discouraged and intrauterine pressure catheters acceptable with appropriate clinical judgment and feasibility.

Midforceps, breech extraction, and manual manipulation are not recommended and increase the risk of maternal and neonatal morbidity. 

Neonatal outcomes for both face and brow presentations include facial edema, bruising, and soft tissue trauma. Complications of compound presentation specifically include umbilical cord prolapse and injury to the presenting limb. With appropriate management, neonatal and maternal morbidity for face, brow, and compound presentations are low.

Further Reading:

Bar-El L, Eliner Y, Grunebaum A, Lenchner E, et al. Race and ethnicity are among the predisposing factors for fetal malpresentation at term. Am J Obstet Gynecol MFM. 2021 Sep;3(5):100405. doi: 10.1016/j.ajogmf.2021.100405. Epub 2021 Jun 4. PMID: 34091061.

Bellussi F, Ghi T, Youssef A, et al. The use of intrapartum ultrasound to diagnose malpositions and cephalic malpresentations. Am J Obstet Gynecol. 2017 Dec;217(6):633-641. doi: 10.1016/j.ajog.2017.07.025. Epub 2017 Jul 22. PMID: 28743440 . 

Pilliod RA, Caughey AB. Fetal Malpresentation and Malposition: Diagnosis and Management. Obstet Gynecol Clin North Am. 2017 Dec;44(4):631-643. doi: 10.1016/j.ogc.2017.08.003. PMID: 29078945 .

Zayed F, Amarin Z, Obeidat B, et al. Face and brow presentation in northern Jordan, over a decade of experience. Arch Gynecol Obstet. 2008 Nov;278(5):427-30. doi: 10.1007/s00404-008-0600-0. Epub 2008 Feb 19. PMID: 18283473 . 

Initial Approval: August 2013; Revised: 11/2016; Revised July 2018; Reaffirmed January 2020; Revised September 2021. Revised July 2023.

This site uses cookies.

This feature is only available for registered users.

Log in here:, not a registered user.

Becoming a registered user gives you access to special features like PDF downloads and podcast episodes of each SASGOG Pearl of Exxcellence.

Create a Free Account

Are you sure you want to remove this Pearl from your favorites list?

Fastest Obstetric, Gynecology and Pediatric Insight Engine

• Abdominal Key
• Anesthesia Key
• Basicmedical Key
• Otolaryngology & Ophthalmology
• Musculoskeletal Key
• Obstetric, Gynecology and Pediatric
• Oncology & Hematology
• Plastic Surgery & Dermatology
• Clinical Dentistry
• Radiology Key
• Thoracic Key
• Veterinary Medicine
• Gold Membership

Diagnosis and Management of Abnormalities of the Face

Key Points • The fetal face can be visualised by ultrasound from 9 weeks’ gestation onwards. • After 9 weeks’ gestation, only proportional changes occur in the fetal face. • Clefts and micrognathia are the most common facial anomalies. • In many genetic disorders, the face has a deviant appearance. • At the routine anomaly scan the profile, the lip and the eyes should be examined. • A detailed ultrasound investigation in high-risk patients requires complete assessment of the face in all three orthogonal planes. Introduction Of all ultrasound views, the fetal face is highly appreciated by parents and frequently imaged during ultrasound examinations. The face is anatomically and functionally a complex structure which poses several challenges for prenatal imaging. The complexity is caused by its particular varied three-dimensional morphology and curved nature. Furthermore, the face includes various sensory organs, each with a function of vital importance. A lot of information can be obtained by ultrasound examination of the fetal face. Besides obvious and clinically relevant anomalies, such as clefts or microphthalmia, changes in shape, subtle dysmorphic features or markers, can provide clues to specific genetic syndromes. Facial anomalies are frequently associated with other anomalies or part of syndromes or sequences. The finding of a facial anomaly therefore requires a thorough examination of the entire fetus. In high-risk patients or when anomalies are found, a facial segment-specific analysis should be performed in all three orthogonal planes. The first characteristics of the fetal face become visible at 9 weeks’ gestation (postmenstrual age) when the jawbones become ossified ( Fig. 35.1 ). After 9 weeks’ gestation, growth and development of the face are dominated by proportional changes and changes in the relative positions of the facial elements until long after birth. Imaging and diagnosis of some facial anomalies such as bilateral clefts and severe variants of retrognathia are possible from 11 weeks’ gestation, but with a high-resolution equipment and transvaginal approach, they may also become visible at an earlier stage. • Fig. 35.1 Two-dimensional ultrasound image of an embryo of 9 weeks and 4 days showing the jawbones as bright with spots. In each trimester, the face has specific characteristics ( Fig. 35.2 ). The first trimester face has a triangular shape. The orbital plane surmounted by the relatively large neurocranium forms the base and the two rami of the mandible, the sides, ending in the pointed chin. Between 16 and 36 weeks’ gestation, the facial height grows relatively more than the width, especially before 25 weeks’ gestation. In the third trimester, the face looks round because of subcutaneous fat accumulation, especially below the zygomatic bones, and because of the widening of the mandible. As opposed to the adult face, the fetal face is relatively small, which is the result of the rudimentary upper and lower jaws, the small nasal cavities and sinuses and the unerupted primary teeth. • Fig. 35.2 Three-dimensional rendered ultrasound images of a first ( A ), second ( B ) and third trimester ( C ) fetus. When looking at the midsagittal view, the fetal profile (FP) is concave throughout gestation as opposed to the relative flat profile in postnatal life ( Fig. 35.3 ). The concavity, mainly caused by the rather retrognathic position of the small mandible, with respect to the maxilla, is probably meant to facilitate for the newborn the concomitance of sucking movements and breathing during breast feeding. • Fig. 35.3 Prenatal two-dimensional ultrasound image of a fetus ( A ) with the maxilla–nasion–mandible angle ( MNM ) and postnatal cephalogram of a child ( B ) with point A–nasion–point B angle (ANB); the MNM angle is larger than the ANB angle, indicating a more convex profile in prenatal life. Ultrasound Investigation of the Fetal Face The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) has published guidelines for the routine midtrimester ultrasound scan. The recommended minimum requirements for basic midtrimester anatomical examination of the fetal face should include an attempt to visualise the upper lip for exclusion of cleft lip. If technically feasible, other facial features that can be assessed include the median facial profile and evaluation of the orbits, nose and nostrils. Many countries have developed guidelines for the routine midtrimester scan, usually recommending visualisation of the profile, eyes and lips. In a screening setting, the midsagittal (profile) plane, the coronal plane (orbits) and the anterior coronal (nose–mouth) plane are usually sufficient to visualise these structures with two-dimensional (2D) ultrasound ( Fig. 35.4 ). • Fig. 35.4 Ultrasound images showing the median profile view ( A ), the orbits in an axial view ( B ) and an anterior oblique coronal nose–mouth view ( C ). A median facial profile provides important diagnostic clues for the diagnosis of cleft lip, bossing or sloping forehead, micrognathia and nasal (bone) anomalies. The coronal or axial views can be used to assess the orbits that should appear symmetrical and intact. The distance between the orbits is about the same as the diameter of one orbit. The symmetrical nose and nostrils, mouth and lips are evaluated in an anterior oblique coronal view to detect in particular cleft lip. When anomalies are encountered at a routine scan, the pregnant woman is commonly referred to a fetal medicine unit for a more detailed ultrasound examination. In this setting, examination of the fetal face usually starts with a subjective evaluation with 2D ultrasound. A more systematic examination of the fetal face should include sagittal, axial and coronal planes. In the midsagittal plane profile view, the forehead, nasal bones, prenasal thickness, soft tissue of the nose, philtrum, tongue, palatal bone, vomer, lower lip and chin can be evaluated, and clear anomalies or dysmorphic features may be observed. The oropharynx with the uvula (equal sign) can be informative and indicate for the existence of a cleft palate. The equal sign can also be obtained in both the axial or coronal planes. In the paramedian sagittal planes the orbits, eyelids, lenses and the ears can be visualised. Of the coronal planes, the slightly tilted nose-mouth plane is the most used for evaluation of the nose (tip, alae nasi and nostrils), upper lip and mouth. More tilted views, towards the deeper structures of the face, allows for visualisation of the maxilla, palate, both eyelids, orbits with lenses and frontal bones. Serial axial images of the fetal face are particularly useful to analyse the maxilla and mandible with the tooth buds but also to view the orbits and the lenses, cheekbones, choanae, lips and tongue. A detailed facial segmental analysis is presented later in this chapter. The use of three-dimensional (3D) ultrasound examination is the next step. This technique has been a major breakthrough in the study of the fetal face and has given rise to a new discipline called ‘fetal dysmorphology’ (see later discussion). Suspected anomalies can be validated by objective measures of the craniofacial features at a single point in time as well serially at different gestational ages. The combination of typical facial features combined with additional findings such as associated structural anomalies affecting all organs but especially the central nervous system (CNS), poly- or oligohydramnios, fetal size, information obtained from family history and a genetic workup can lead to recognition of a known syndrome. Three-Dimensional Ultrasound of the Fetal Face Three-dimensional ultrasound is especially useful in demonstrating curved structures ( Fig. 35.5 ) and surface malformations ( Fig. 35.6 ) and improves accurate topographic depiction of structures. Therefore, this technique is a major addition in the assessment of the fetal face. • Fig. 35.5 Three-dimensional surface-rendered image with high definition technique of a normal second trimester fetus, showing a symmetrical face; clearly an intact upper lip; and a sharp, definable cupid’s bow. • Fig. 35.6 Three-dimensional surface-rendered image of fetus with nasofrontal encephalocele (arrow) . In 2D ultrasound, a single plane is imaged at a time, and the sonographer has to construct in her or his mind the complex 3D anatomy of the face. 3D ultrasound enables to collect a volume of ultrasound information consisting of multiple 2D slices. The examination starts by placing a box (region of interest) of variable size on the 2D image of the object that we want to investigate. Then a sweep is produced by a motor within the probe, whereby all the adjacent 2D slices within the box are stored. When the fetus is quiet, a slow sweep can be used, which improves spatial resolution. Ideally, there should only be amniotic fluid in front of the face and no other structures such as limbs. All stored sectional planes are integrated to form the volume. After the volume is digitally stored, the volume can be manipulated by modifying the colours and the image settings and by using different modes. These options can extract different information from the same dataset. Volumes of the fetal face are usually analysed by using cross-sectional images through the volume or by rendered images. Both approaches improve understanding of the complex anatomy of the face; cross-sectional imaging by enhancing spatial awareness and rendering imaging by facilitating a lifelike 3D view of the face (see Fig. 35.5 ). Cross-Sectional Imaging When cross-sectional imaging is used, usually the three orthogonal planes are displayed simultaneously, named multiplanar imaging ( Fig. 35.7 ). The ultrasound volume of the fetal face can be rotated and reviewed millimetre by millimetre by scrolling through the volumes. • Fig. 35.7 Example of multiplanar imaging with the fetal face turned in the standard position. R is the right side, and L is the left side of the fetus. It is advisable to turn the face in the standard position with the coronal view in the upper left box, the profile view with the nose pointing to the left in the upper right box and the axial view with the nose pointing downwards in the lower left box (see Fig. 35.7 ). In this way, it is clear what is the right and left side of the fetus, and using standard positions improves subjective pattern recognition. The reference dot, which marks the intersection of the three orthogonal planes, is very helpful in identifying structures. Multiplanar imaging is extremely helpful in defining the exact midsagittal plane. When examining the fetal profile (FP) subjectively, the true midsagittal plane is usually assumed to be present. However clear landmarks to define the exact midsagittal plane are missing. It is shown that presumed 2D profile images are significantly oblique in 30% of cases. The multiplanar imaging of 3D ultrasound provides the ultrasonographer with a unique tool: the possibility to visualise contemporarily the three orthogonal planes. A deviation from the exact midsagittal plane is easily recognised and can be corrected to the true midsagittal plane. Evaluation of the profile in an incorrect midsagittal plane can lead to diagnostic inaccuracies. For example, in a view deviating from the midsagittal plane, the most protruding part of the chin will not be visualised, creating the impression of retrognathia. Also, the forehead will suggest more bossing and the nose will seem small ( Fig. 35.8 ). • Fig. 35.8 Ultrasound images of the fetal profile. A, A deviating midsagittal profile view, creating an image of retrognathia, flat nose and slightly bossing forehead. B, The exact midsagittal profile view obtained with multiplanar imaging of the same fetus, showing a normal profile. Another way to display a volume cross-sectionally is by tomographic ultrasound imaging or the ‘multislice method’. Several slices of a volume, which are parallel to each other, are simultaneously displayed with predefined number and spacing of the slices ( Fig. 35.9 ). This offers the examiner a more complete picture. The mandible and maxilla, for example, can be viewed at the same time. • Fig. 35.9 Example of tomographic ultrasound imaging display of the fetal face. Rendered Images In rendering mode, the image includes information from the entire volume. The images from the three orthogonal planes are combined to obtain a realistic 3D picture (see Fig. 35.5 ). It is possible to turn and rotate the volume and view the volume from various positions. Technical options such as the electronic scalpel used to remove unwanted structures can improve the image quality and the diagnostic value of the ultrasound examinations ( Figs. 35.10 and 35.11 ). • Fig. 35.10 Three-dimensional rendered frontal view of the fetal face before ( A ) and after ( B ) use of the electronic scalpel to remove the hand in front of the face. • Fig. 35.11 Three-dimensional rendered frontal view of the face with high definition technique ( A ). Half of the picture is deleted with the electronic scalpel ( B ). After turning the head around the y-axis of the fetus, the profile is visible against a clear black background ( C ). By choosing various threshold values, the rendered volume can be studied in a variety of ways. In the surface view, the outer surface of the fetus is highlighted. The fetus is displayed as a 3D sculpture. There are various technical options for editing this sculpture, for example, adjusting the gain or the position of the light or using high-definition techniques that provide skin-like pictures with sharp contrasts ( Fig. 35.12 ). By observing the fetal face in rendering mode, we will have a general subjective visual impression comparable with the ‘gestalt’ approach of clinical genetics, offering the possibility to suspect dysmorphologies, especially later in pregnancy ( Fig. 35.13 ). Rendering mode is also helpful in assessing tumours, clefts and ear anomalies. In the maximum mode, the bones are emphasised. The strongest echoes are kept, and the echoes from the soft tissue are eliminated ( Fig. 35.14 ). This especially allows visualisation of curved skeletal structures such as sutures and fontanels of the skull, hard palate and nasal bones. • Fig. 35.12 A–E, Three-dimensional rendered ultrasound pictures in several settings showing various facial expressions. • Fig. 35.13 Surface-rendered images of syndromic fetuses, trisomy 21 ( A ), Binder syndrome ( B ) and Apert syndrome ( C ). (Courtesy Dr B. Benoit) • Fig. 35.14 Examples of three-dimensional rendered ultrasound pictures of a normal face using surface ( A ) and maximum ( B ) mode. Finally, the possibility to store 3D ultrasound volumes and edit them offline facilitates communication and research. Realistic 3D images improve communication with the parents and health professionals involved in the management of the pregnancy. Four-dimensional (4D) ultrasound added the fourth dimension ‘time’ to the 3D picture. This dramatically improved dynamic assessment of the fetal face. Movements of the mouth (yawning), tongue, eyelids and lenses can be visualised with 2D, but with 4D ultrasound, these movements can be visualised with greater ease and more details. Complex movement as seen in facial expression can be observed (see Fig. 35.12 ). 4D ultrasound might be an important modality for future prenatal behaviour studies. Facial Clefts Of all congenital anomalies involving the face, clefts are the most common, affecting about 1 or 2 per 1000 live births. The incidence of facial cleft varies by gender, ethnicity and geographic location, being more frequent in boys (twice as much as in girls), in Asians and in central–north Europe compared to the European average. In contrast, isolated cleft palate occurs equally often in all races (about 4 in 1000 births) and is more common in females. The distribution of clefts is estimated to be 25% for cleft lip, 50% for cleft lip and palate and 25% for isolated cleft palate. Facial clefts may either be isolated or associated with other anomalies. The incidence of associated structural anomalies, chromosomal aberrations (mainly trisomy 13 and 18) or an underlying genetic syndrome or sequence varies with the anatomical cleft type. Therefore adequate characterisation of the cleft is important not only to counsel parent appropriately on the severity of the defect but also on its likely association with chromosomal or syndromal anomalies. Isolated clefts are associated with low mortality and morbidity rates and are primarily an aesthetic and functional problem. The percentage of isolated cases is the highest in the cleft lip (without a defect in the alveolar ridge or palate) and close to 0% in midline and atypical cleft groups. Bilateral clefts have lower percentages of isolated cases than unilateral clefts. The most frequent associated defects are musculoskeletal anomalies (polydactyly and limb reductions) followed by malformations of the CNS and malformations of the cardiovascular system. For all clefts grouped together, the reported incidence of isolated clefts varies between 31% and 71%. It should be noted that the incidence in reality may be lower because of unidentified syndromes or late manifestation of developmental delays (e.g., learning difficulties). About 350 syndromes are associated with facial clefting. The most common syndromes and sequences identified after birth in cleft patients (including isolated cleft palate) are Pierre Robin sequence, Van der Woude syndrome, Stickler syndrome, CHARGE (coloboma, heart defects, atresia choanae, retardation of growth, genital abnormalities, and ear abnormalities) association, ectrodactyly-ectodermal dysplasia-cleft syndrome, Goldenhar syndrome, Apert syndrome, Treacher Collins syndrome, Wolf-Hirschhorn syndrome, velocardiofacial syndrome (22q11deletion), Smith-Lemli-Opitz syndrome, Fryns syndrome and oral-facial-digital syndrome. In a small percentage of cases, facial clefts are atypical and occur in different regions of the face. According to the well-established Tessier classification, the defects are numbered from 0 to 14 and classified based on the anatomical localisation of the cleft, with the orbit as reference structure. The most common atypical facial cleft found at birth is Tessier 7, a lateral cleft between the corner of the mouth and the ear. When there is suspicion of a fetal cleft with associated anomalies the parents should be referred for genetic counselling and additional genetic investigations. Usually a multidisciplinary team consisting of a medical specialist (fetal medicine), a plastic surgeon (ear, nose and throat) and a specialised nurse or social worker will explain the surgical options, aesthetic outcomes and specific feeding needs of a newborn with a cleft lip or palate and will take care of the psychological impact of the anomaly for the family. The recurrence risk for isolated cases is 5% if one sibling or parent is affected and 10% if two siblings are affected. In syndromic cases, all kinds of inheritance have been described (autosomal dominant, autosomal recessive, X-linked). Ultrasound Examination of Facial Clefts In the first trimester, the following findings are clues to the presence of a facial cleft: the premaxillary protrusion ( Fig. 35.15 ), an abnormal maxilla–nasion–mandible (MNM) angle, an abnormal retronasal triangle ( Fig. 35.16A ) and the maxillary gap ( Fig. 35.15 and 35.16B ). Also, an enlarged nuchal translucency increases the risk for facial clefts. • Fig. 35.15 Two- and three-dimensional images of a first trimester fetus with bilateral clefts, premaxillary protrusion, maxillary gap and micrognathia. A, Sagittal 2D scan. B, 3D rendered image. • Fig. 35.16 A, Coronal view (posterior to the nose) of a normal first trimester fetus, showing a normal retronasal triangle with intact palate (arrow) and a normal mandibular gap (asterisk) . The retronasal triangle is formed by three echogenic lines formed by the two frontal processes of the maxilla and the palate. B, Sagittal view of a normal first trimester fetus, showing an intact palate and no maxillary gap (arrow) . The ultrasonographic assessment usually starts by 2D ultrasound examination. The slightly tilted coronal nose–mouth view is an important and obligatory plane for the diagnosis of facial clefts. In this plane, the nostrils, philtrum and upper lip can be evaluated ( Fig. 35.17 ). • Fig. 35.17 Two-dimensional image of frontal nose–mouth view showing an intact upper lip ( A ), a unilateral cleft lip ( B ) and a bilateral cleft lip ( C ). In the axial plane, the alveolar ridge and the upper lip can also be evaluated ( Fig. 35.18 ). • Fig. 35.18 Two-dimensional image of axial plane through maxilla with unilateral cleft lip and palate. Arrow indicates the cleft. In the midsagittal plane, the premaxillary protrusion can be seen, especially in bilateral clefts, although all types of clefts show some protrusion of the premaxilla except when there is an intact alveolar ridge ( Fig. 35.19 ). The protrusion is caused by the missing restraining effect of an intact musculus orbicularis oris, and its severity can be quantified with the MNM angle (see Fig. 35.18 ). In unilateral clefts, the profile may look flat because of missing soft tissue (in front of the maxilla) and because the muscularis orbicularis oris pulls the lip to the noncleft side. ( Fig. 35.18A ). • Fig. 35.19 Ultrasound pictures of a case with isolated unilateral cleft lip and palate ( A ), a case with trisomy 13 and bilateral cleft lip and palate ( B ) and a case with an isolated bilateral cleft lip and palate ( C ). In all three cases, the maxilla–nasion–mandible angles were enlarged. Prenatal diagnosis of a cleft palate is a challenge because of the shadows produced by surrounding osseous structures interfering with a good visualisation of the palate. When the fetal head is tilted backwards and the mouth is slightly open, the whole palate can be visualised from the hard palate to the uvula (soft palate). Care has to be taken to scan in the exact midsagittal plane ( Fig. 35.20 ). • Fig. 35.20 Two-dimensional ultrasound picture of a fetus with a backward-tilted head and open mouth, showing an intact hard and soft palate. If the fetus makes breathing movement, adding colour may reveal a cleft in the palate by showing bidirectional flow of amniotic fluid over the palate. The equal sign described by Wilhelm is an important marker for an intact uvula ( Fig. 35.21 ). It can be visualised in in all three planes, and its visualisation proves an intact palate. If the uvula cannot be visualised in its typical presentation (equal sign), there should be a strong suspicion of a cleft palate. • Fig. 35.21 Coronal view through the pharynx showing the equal sign (circle) cranial to the vocal cords (arrows) . There were great expectations for the role of 3D ultrasound in the diagnosis of facial clefts, especially in the challenging diagnosis of the palate. 3D ultrasound does not seem to improve the detection rate of cleft lip and palate; however, a more precise and reliable diagnosis can be achieved. For example, rendered images of the face can show unequivocally that the upper lip is intact (see Fig. 35.5 ) or may be useful in defining the extent of the cleft(s) in the lip (complete or incomplete, uni- or bilateral) ( Fig. 35.22 ). • Fig. 35.22 Three-dimensional frontal rendered views, showing an incomplete unilateral cleft lip ( A ), a complete unilateral cleft lip ( B ) and a bilateral complete cleft lip and palate with protrusion of the premaxilla ( C ). Several techniques have been proposed to improve detection of especially cleft palate. First the ‘reverse-face’ view was introduced by Campbell. With this technique, a frontal 3D-rendered view of the face is obtained and rotated 180 degrees around the y-axis of the fetus. Then by looking from the oropharynx (from inside out), the palate, tongue, nasal cavity and orbits can be seen by scrolling through the volume ( Fig. 35.23 ). • Fig. 35.23 Three-dimensional ultrasound image of reversed-face view showing an intact palate (arrow) above the tongue (asterisk) . Hereafter different strategies using 3D volumes have been developed to visualise the palate; these techniques include the ‘flipped-face’ view ( Fig. 35.24 ), the ‘flipped face’ view with angled insonation, the anterior axial 3D view reconstruction and variants of these. They differ mainly in starting plane, insonation angles, the way the volume is rotated and the position of the view bar. The best images are obtained when the head of the fetus is slightly extended, the mouth is open and amniotic fluid is present in the oral cavity (see Fig. 35.20 ). In everyday practice, fetuses have very different positions, and the image quality differs in each patient. Therefore the best strategy must be considered on a case-by-case basis. • Fig. 35.24 Three-dimensional multiplanar image with flipped-face technique of a fetus with an intact palate. The profile image in the left upper quadrant is turned upside down. The render box is adjusted to the size of the palate. In the box in the right lower quadrant , a rendered image of the palate is visible. Recently, new 3D tools such as the ‘polyline’ enable tracing a line along the curved 2D sagittal view of the palate and to obtain a rendered image of it ( Fig. 35.25 ). Also in these cases, it is important that the fetus has a slightly open mouth and that the palate lies as much as possible perpendicular to the scanning plane. • Fig. 35.25 Three-dimensional rendered image of intact hard palate and alveolar ridge. Micrognathia Micrognathia refers to a hypoplastic mandible. In retrognathia, the anteroposterior axis is mostly affected. Both conditions usually coexist, but frequently only the term micrognathia is used to refer to the combination micrognathia and retrognathia. After facial clefts, micrognathia is the most common congenital facial anomaly. Micrognathia can be isolated but is often associated. Conditions presenting micrognathia can be categorised in syndromic conditions primarily involving the mandible (e.g. Pierre Robin sequence, Treacher Collins syndrome, acrofacial dysostosis, orofaciodigital syndromes), skeletal and neuromuscular diseases (e.g. Pena–Shokeir syndrome, multiple pterygium syndrome, achondrogenesis, campomelic dysplasia), chromosomal aberrations (e.g. trisomy 18, trisomy13, triploidy, cri du chat syndrome, Wolf-Hirschhorn syndrome, Pallister Killian syndrome) and other nonchromosomal syndromic conditions (e.g. Meckel-Gruber syndrome, Fryns syndrome, Goldenhar syndrome, Peters plus syndrome). A search in the OMIM website (Online Mendelian Inheritance in Man, a database of human genes and genetic disorders) retrieved 564 hits for ‘micrognathia’ and 180 for ‘retrognathia, (in 2017). Table 35.1 summarises publications on the incidence of associated conditions. The very low percentage of isolated cases illustrates the severity of this facial anomaly, although it is likely that mild isolated cases escape prenatal identification. TABLE 35.1 Overview of Published Data on the Incidence of Associated Structural Anomalies, Chromosomal Anomalies, Skeletal or Neuromuscular Diseases, Syndromes or Sequences and Isolated Cases in Fetuses With Micrognathia

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related posts:

• Placental Function in Maternofetal Exchange
• Noninvasive Prenatal Diagnosis for Single-Gene Disorders
• Fetal Growth and Growth Restriction

Stay updated, free articles. Join our Telegram channel

Comments are closed for this page.

Full access? Get Clinical Tree

A Beginner’s Guide to Ultrasound

• 📖 Geeky Medics OSCE Book
• ⚡ Geeky Medics Bundles
• ✨ 1300+ OSCE Stations
• ✅ OSCE Checklist PDF Booklet
• 🧠 UKMLA AKT Question Bank
• 💊 PSA Question Bank
• 💉 Clinical Skills App
• 🗂️ Flashcard Collections | OSCE , Medicine , Surgery , Anatomy
• 💬 SCA Cases for MRCGP

To be the first to know about our latest videos subscribe to our YouTube channel 🙌

Table of Contents

Suggest an improvement

• Hidden Post Title
• Hidden Post URL
• Hidden Post ID
• Type of issue * N/A Fix spelling/grammar issue Add or fix a link Add or fix an image Add more detail Improve the quality of the writing Fix a factual error
• Please provide as much detail as possible * You don't need to tell us which article this feedback relates to, as we automatically capture that information for you.
• Your Email (optional) This allows us to get in touch for more details if required.
• Which organ is responsible for pumping blood around the body? * Enter a five letter word in lowercase
• Email This field is for validation purposes and should be left unchanged.

What is ultrasound?

Ultrasound is a common imaging modality that allows visualisation in real-time. As such it is becoming increasingly popular on the wards for diagnosis and management purposes. You should be familiar with its operation and know in which situations it may help your clinical decision making.

What is ultrasound used for?

Ultrasound can be used for:

• Assessment of jugular venous pressure (JVP)
• Venepuncture
• Focused assessment for screening in trauma (FAST)
• Lumbar puncture
• Thoracentesis
• Paracentesis
• Evaluation of abdominal organs

Ultrasound basics

How does ultrasound work¹.

1 . High-frequency sound waves are transmitted from a transducer.

2 . These sound waves are then reflected by different tissue types in different ways.

3 . The reflected sound waves are then picked up by the ultrasound transducer.

4 . The sound waves are then transformed into an image by special software.

How do tissue types differ in their reflection of sound waves?

Bones, fat and stones.

Bones , fat and stones produce a hyperechoic signal .

A hyperechoic signal is bright as most ultrasound waves are reflected.

Cartilage and muscle

Cartilage and muscle produce a hypoechoic signal .

A hypoechoic signal appears  dark as most waves pass through the tissue.

Fluid and fluid-filled structures

Fluid and fluid-filled structures produce an anechoic signal .

An anechoic signal appears black as there is no reflection of ultrasound waves.

A shadow may be noted on an ultrasound when a hypoechoic area is located behind a hyperechoic structure .

Getting started

The first steps of performing an ultrasound involve:

• Turning on the machine (easy, but often overlooked; often a button in the upper left or right corner of the keypad).
• Entering the patient’s information (e.g. name, date of birth, hospital number).
• Selecting an appropriate ultrasound probe for the area being examined.

Probe basics

How do i know which probe i should use.

Typically there are 3 different types of ultrasound probe: linear, curvilinear and phased.

Linear probe :

• High frequency (7-15MHz):
• High resolution but superficial (1-6cm) depth
• Good for vascular access, nerve blocks, assessment of testes and superficial lung tissue

Curvilinear :

• Low frequency (2-5MHz)
• Low resolution, but greater depth (10-20cm)
• Useful for abdominal, pelvic, obstetric and deep lung tissue
• The lowest frequency (1-3MHz)
• Useful for echocardiography

How do I hold the probe?

The image below demonstrates how to appropriately hold an ultrasound probe .

Probe orientation

Typically, there is a dot or a cross on the probe, this correlates with a dot on the left side of the screen.

This marker should be toward the patient’s right in transverse and head in longitudinal.

If you are unsure, it is best to place your finger on one side of the probe and look for movement on the screen (the side that shows movement by the dot is the side that should face the patient’s right).

Once you’ve chosen an appropriate probe and are holding it right, the next steps of performing an ultrasound involve:

• Applying gel to the probe and patient.
• Placing the probe onto the patient and observing the images on the screen.
• Adjusting the settings to achieve an optimal view.

Common settings for achieving an optimal view

• Adjusting the gain of an ultrasound changes the brightness of the image.
• Gain is typically controlled by a knob.
• The gain should be adjusted until fluid appears black and soft tissue appears mid-grey with some parts of the image appearing white
• Depth measures are shown in cm on the side of the ultrasound monitor.
• It is often best to begin deep to orientate yourself and then work more superficially to bring the object of interest into the middle of the screen.

General tips for achieving an optimal view

Some general tips for achieving an optimal view include:

• Use lots of gel
• Make good contact between the probe and skin (whilst ensuring the patient is comfortable)
• Dim the lights to improve your view of the monitor
• Ensure the probe is perpendicular to the skin

Measuring the JVP with ultrasound

1 . Position your patient as you would when assessing the jugular venous pressure (JVP) in a clinical exam (e.g. supine, head of the bed at 45°, patient’s head laterally rotated to the side not being scanned)

2 . Set the gain of the ultrasound to mid-range.

3 . Apply gel to the patient’s neck.

4 . Place the probe in a transverse orientation within 2cm of the clavicle.

5 . Identify the internal jugular vein (IJV) and the carotid artery, assessing the following:

• Wall thickness: arteries have thicker more muscular walls than venous structures.
• Shape: the carotid will be circular whereas the IJV can be oval or irregularly shaped.
• Compressibility: veins are easily compressed (if you only see one vessel, use less pressure as you may have fully compressed the IJV).
• Respiratory variability: central venous structures will fluctuate in size with respiration.
• Position: the IJV is usually (but not always) lateral to the carotid.

6 . Centre the probe so the IJV is centred on the monitor.

7 . Slowly rotate the probe keeping the IJV in the centre until a sagittal view is achieved (ensure you are not foreshortening the vein by carefully moving the probe medially and laterally).

8 . Locate the point of the initial collapse of the IJV (centre the probe over this point).

9 . Measure the JVP height above the sternal angle as normal.

Ultrasound-guided intravenous access

In this section, we will be focusing on peripheral intravenous access, however, similar principles are applied for central venous line insertions. 5,6

Indications

Indications for ultrasound-guided intravenous access include:

• Multiple failed attempts
• History of difficult cannulation

Contraindications

Ultrasound-guided IV access should not supplant intraosseous (IO) access in life-threatening situations.

Probe choice

A high frequency (5-12 MHz) linear transducer is typically used as high frequency permits a better resolution of structures close to the surface of the skin

A lower frequency curved probe may be more effective in obese patients.

1 . Select a vein (as per non-ultrasound guided peripheral IV placement ).

2 . Machine set up:

• Choose appropriate examination pre-set: typically “peripheral vascular venous” or “superficial vascular, venous”

3 . Clean the ultrasound probe.

4 . Apply a tourniquet.

5 . Apply gel to the ultrasound probe.

6 . Identify the target vein in the transverse plane: note the depth of the anterior wall of the vein and pay attention to any adjacent structures.

7 . Once identified, rotate the probe into the longitudinal plane. Hold the probe with the thumb, index, and middle fingers as shown, using the remaining fingers as an anchor.

8 . Clean the patient’s skin.

9 . Align the needle prior to insertion.

10 . Insert the needle just distal to the transducer probe:

• You should be able to see the needle throughout the procedure on the screen, if you cannot, you must realign.
• You can gently rock the probe to help see the flash of the needle tip if needed.
• Do not advance the needle unless you are able to clearly see the tip.

11 . Advance the needle into the vein.

12 . Hold the needle still and watch as you advance the catheter into the vein.

Common pitfalls

Common pitfalls for ultrasound-guided intravenous access include:

• Applying too much pressure and thus collapsing the vein.
• Choosing suboptimal settings on the ultrasound machine.

Focused assessment with sonography for trauma (FAST)

FAST is used in emergency settings to rule out free fluid in the abdomen as well as pericardial effusions using a curvilinear probe.

As discussed previously, fluid shows up as black, so the scanning clinician is inspecting for black lines surrounding organs.

Probe locations

The key probe locations for FAST are:

• Right upper quadrant
• Left upper quadrant
• Suprapubic region
• Sub-xiphoid region

Right upper quadrant view (Morison’s pouch)

The ultrasound probe is positioned in the coronal plane in right mid-axillary line (between rib 11 and 12).

Once positioned correctly, the ultrasonographer inspects for evidence of free fluid between the liver and the kidney.

Ribs are often in the way;  so the probe often requires some tilting/rotating to align it with an intercostal space.

Left upper quadrant view (perisplenic view)

Similar to Morison’s pouch, but the transducer is placed approximately half a probe length more superior and posterior (as the left kidney is more superior and posterior than the right).

The ultrasonographer inspects for free fluid between the kidney and spleen.

Cardiac view

The ultrasound probe is positioned in the transverse plane, within the sub-xiphoid region (with the probe aimed at the heart).

The ultrasonographer inspects for evidence of pericardial effusion.

Pelvic view (suprapubic)

The ultrasound probe is positioned in the suprapubic region, pointing towards the bladder.

The ultrasonographer inspects for free fluid outside of the bladder.

Both transverse and longitudinal images are typically assessed.

• Toronto Notes 2016: Medical Students Essential Med Notes.
• Clay Kurtz (Author’s own work)
• Emergency Ultrasonography. Resources and Tutorials on EM Ultrasound. Available from: [ LINK ].
• Canadian Internal Medicine Ultrasound (CIMUS). Available from: [ LINK ]
• Joing S, Strote S, Caroon L, et al. Ultrasound-guided peripheral IV placement.  N Engl J Med . 2012;366(25):e38.
• UpToDate. Principles of ultrasound-guided venous access. Jeremiah J Sabado, MD and Mauro Pittiruti, MD. Available from: [ LINK ].
• Ltyore. Morison’s pouch. Public Domain.
• The POCUS Atlas. Available from: [ LINK ].

Other pages

• Product Bundles 🎉
• Join the Team 🙌
• Institutional Licence 📚
• OSCE Station Creator Tool 🩺
• Create and Share Flashcards 🗂️
• OSCE Group Chat 💬
• Newsletter 📰
• Advertise With Us

Join the community

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Radiol Case Rep
• v.19(10); 2024 Oct
• PMC11331709

Left superior vena cava's unconventional path to left atrium drainage: A case report

Muhammad idrees.

a Multan Institute of Kidney Diseases, Multan, Pakistan

Waleed Tariq

b Mayo Hospital, Lahore, Pakistan

Rashid Asghar

Muhammad junaid tahir.

c Shaukat Khanum Memorial Cancer Hospital and Research Center, Lahore, Pakistan

Khabab Abbasher Hussien Mohamed Ahmed

d Faculty of Medicine, University of Khartoum, Khartoum, Sudan

Zohaib Yousaf

e Tower Health, Reading, PA, USA

Persistent left superior vena cava (PLSVC) is a rare congenital anomaly. We presented PLSVC in a patient with end-stage renal disease (ESRD) requiring hemodialysis. The left internal jugular vein was utilized for central venous access due to difficult central vascular access, resulting in a diagnosis of PLSVC draining in the left atrium. This case underscores the importance of awareness of anatomical variations before central catheter placement.

Introduction

Arteriovenous (AV) access is the preferred route for hemodialysis in end-stage renal disease (ESRD) [ 1 ]. A central venous catheter (CVC) is commonly used for temporary vascular access in patients requiring urgent HD. Due to direct access to the superior vena cava (SVC) and right atrium, the right internal jugular vein is the preferred site for CVC placement over the left side and the subclavian veins [ 2 ].

Embryologically, anterior and posterior cardinal veins carry deoxygenated blood from the cranial and caudal parts of the embryo, respectively. During the eighth week of development, the anastomosis between the left and right anterior cardinal veins forms the left brachiocephalic vein, shunting blood from left to right. Post-anastomosis, the proximal segment of the right anterior and common cardinal veins evolves into SVC. In contrast, the left anterior and common cardinal veins obliterate with the formation of the ligament of Marshall and coronary sinus, respectively. The failure to develop the ligament of Marshall can result in PLSVC formation [ 3 , 4 ]. PLSVC is the most common thoracic venous anomaly, with a prevalence of 0.2% to 3% in the general population and 1.3% to 11 % in the population with an associated cardiac anomaly. Most cases are discovered incidentally during CVC placement or imaging-like chest computed tomography [ 5 ]. The ideal site for CVC placement in the presence of PLSVC is unclear [ 6 ]. We reported PLSVC identified in a patient with multi-access failure. Access was successfully established through the left internal jugular vein.

Case presentation

A 43-year-old female with a history of ESRD on maintenance HD for the last year presented to the emergency department with progressive shortness of breath and anuria over the previous 4 days. She had an unrestricted intake of fluids for 1 week due to a hot summer. Previously, she had 2 HD sessions per week instead of the recommended 3 sessions per week due to financial constraints. The patient had opted for no arteriovenous (AV) fistula and had had HD sessions from indwelling CVC at various sites. She had her last session of HD 1 week ago from CVC in the right internal jugular vein. Her CVC was removed 1 week before presentation due to concern for central line-associated bloodstream infection (CLABSI) with a plan for tunneled catheter placement after treatment of active infection.

Upon presentation, she was tachycardiac (110 beats per minute), hypertensive (210/100 mmHg), tachypneic (31 breaths per minute), and desaturated to 76% on room air. Examination revealed periorbital oedema and grade III pedal oedema extending up to the shins bilaterally, and crackles were audible all over the chest. On chest X-ray (CXR), broncho-vascular markings, septal thickening, and ground glass appearance were prominent concerning pulmonary oedema. The patient's renal function showed high levels of urea and creatinine, i.e., 121 mg/dL and 9.28 mg/dL, respectively. A working diagnosis of hypertensive emergency with pulmonary oedema secondary to inadequate renal replacement therapy was made, and the patient was planned for urgent hemodialysis.

A point-of-care ultrasound revealed stenosis of the right jugular, bilateral femoral, and subclavian veins, likely secondary to the multiple CVC insertions over the last year. CVC was inserted in the left internal jugular vein for urgent HD. A post-CVC-insertion CXR disclosed an abnormal position of the dialysis catheter tip along the left mediastinal border ( Fig. 1 ). A computed tomography angiography (CTA) showed PLSVC draining into the left atrium following the path along the left sternal border ( Fig. 2 , Fig. 3 ). A transthoracic echocardiogram was unrevealing for any other concurrent cardiac anatomic anomalies. The rest of the hospital stay was uncomplicated. The patient underwent surgery for AV fistula formation and continued to receive hemodialysis sessions through CVC in PLSVC without any complications.

Posterior to anterior (PA) view of post-CVC insertion chest X-ray (Red arrow pointing at the lumen of CVC, Yellow arrow pointing at the tip of CVC).

Coronal view of CT scan (White arrow pointing at right SVC, Yellow arrow pointing at PLSVC, Red arrow pointing at left atrium).

Axial view of CT scan (Yellow arrow pointing at PLSVC, white arrow pointing at right SVC).

The first case of persistent left SVC was reported in 1787, a common congenital venous circulation anomaly [ 7 ]. Lim and H'ng had also reported a PLSVC diagnosed during CVC catheterization in the left internal jugular vein for HD and later confirmed on CTA. HD was done for 5 months without complications [ 2 ]. Jang et al. also reported a left CVC tip seen along the left mediastinal border on chest radiography. CTA led to a diagnosis of PLSVC [ 8 ].

An opaque line along the medial border of the mediastinum in post-CVC chest X-ray strongly suggests suspicion of PLSVC which must be confirmed by further radiologic investigation. Various imaging modalities have been implied for in-depth confirmation, especially contrast-enhanced computed angiogram, magnetic resonance venography, and conventional venography [ 9 ]. After finding a catheter along the medial border of the mediastinum, we performed CTA to explore the tract. We found contrast enhancement on the left side lateral to the aortic arch ( Fig. 1 , Fig. 2 ) and draining into the left atrium ( Fig. 1 ), confirming the presence of PLSVC. At the same time, the contrast-enhanced tract was also illustrated on the right side draining into the right heart which was the right SVC ( Fig. 1 , Fig. 2 ). Sonavane, Sushilkumar K., et al. and Azizova, Aynur, et al. reported similar radiologic findings on cross-sectional and axial views [ 10 , 11 ].

Hana et al. reported a PLSVC draining through the left brachiocephalic vein in the left atrium posteriorly [ 12 ]. This finding is similar to our case. PLSVC draining into the left atrium is the cause of right to left shunt and is usually associated with cardiac anomalies. However, this can be an isolated finding with no other cardiac anomalies [ 13 ].

Association of various cardiac anomalies including ventricular septal defects, atrial septal defects, or tetralogy of Fallot with PLSVC must be screened with transthoracic echocardiography or cardiac MRI to avoid mal-positioning of CVC [ 14 ]. Patients with PLSVC draining into the left atrium may develop a large right-to-left shunt. Among such patients, dialysis catheters must be avoided and the use of the femoral vein is recommended. The precatheter placement echocardiography is advised to avoid any complications. Similarly, femoral access must be used again in patients with PLSVC with absent RSVC.

PLSVC is most commonly diagnosed incidentally on post-CVC chest radiographs [ 15 ]. Use of the left internal jugular vein, as in our case, led to an incidental diagnosis of PLSVC. Authors suggest close hemodynamic monitoring due to a risk of arrhythmias, embolization, shock, and cardiac arrest associated with the placement of CVC in PLSVC [ 16 ]. An echocardiogram may be helpful to evaluate for concurrent cardiac abnormalities.

By European Renal Best Practice (ERBP) guidelines [ 17 ], the right internal jugular vein is the access of choice among patients requiring initiation of urgent Hemodialysis as there is minimal risk of infection if inserted under strict aseptic techniques. The next approach could be adopted via the left internal vein but the subclavian vein is not recommended as the first choice among such patients as there is an increased risk of venous stenosis. In addition to the higher incidence of thrombosis, there is a higher chance of infection and bacteremia among patients using the femoral as an access site for hemodialysis, that's the reason it is not recommended at any cost among patients needing urgent hemodialysis.

The gold standard mode for Hemodialysis for maintenance hemodialysis is Arteriovenous fistula, followed by arteriovenous graft and then tunneled catheter at least. For the patients requiring chronic renal replacement therapy via hemodialysis with a history of catheter infection, the recommendations by the Infectious Disease Society of America (IDSA) must be followed. In case of evident infection of the catheter insertion site without pyrexia, topical antibiotics must be applied. The local site infection resistant to topical therapy must be treated with systemic antimicrobials. Catheter removal is only advised when both steps fail to attain resolution. When catheter-related bloodstream infection (CRBSI) is confirmed by either clinical signs of active disease (fever, tachycardia, chills, or hypotension during dialysis) or sepsis or sepsis-related complications like thrombophlebitis and metastatic infection appear, the catheter must be removed with immediate initiation of empirical antibiotic therapy after sending blood cultures. The patients facing limited access issues with existent CRBSI, remove the catheter only when other potential insertion sites are available. At this time, catheter removal followed by systemic antibiotic therapy and re-insertion after 48-72 hours is recommended. In the absence of another site for re-insertion, guide wire-assisted catheter exchange is advised after 48-72 hours of systemic antibiotic therapy. Accepting the problems of high failure rate, venous sclerosis, and stenosis of guidewire assisted exchange, keep the catheter in situ and start catheter salvage with antibiotic lock. If this fails then remove the catheter. After all, re-consider AVF, Arteriovenous graft, or move towards peritoneal dialysis [ 18 ].

PLSVC is usually an asymptomatic, incidentally discovered congenital anomaly. Right heart access through the left subclavian vein in the case of PLSVC is challenging but possible. Close hemodynamic monitoring and an echocardiogram are prudent in the case of using a CVC inserted through a PLSVC.

Author contributions

MI and RA conceived and designed the case report, and were responsible for data collection and acquisition of data. MI, WT, ZY, and MJT performed the literature review and wrote the manuscript. K.A.H.M.A., ZY and MJT reviewed and critically revised the manuscript. All authors have approved the final manuscript.

Ethical approval

Ethical approval was not required because this is a case report.

Patient consent

Written informed consent was obtained from the patient for publication and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal on request. All patient data were de-identified to maintain confidentiality.

Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments: Not Applicable. There is no funding received for this study.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Data Descriptor
• Open access
• Published: 24 August 2024

A benchmark for 2D foetal brain ultrasound analysis

• Mariano Cabezas   ORCID: orcid.org/0000-0002-4417-1704 1   na1 ,
• Yago Diez 2   na1 ,
• Clara Martinez-Diago 3   na1 &
• Anna Maroto 3   na1  

Scientific Data volume  11 , Article number:  923 ( 2024 ) Cite this article

Metrics details

• Scientific data

Brain development involves a sequence of structural changes from early stages of the embryo until several months after birth. Currently, ultrasound is the established technique for screening due to its ability to acquire dynamic images in real-time without radiation and to its cost-efficiency. However, identifying abnormalities remains challenging due to the difficulty in interpreting foetal brain images. In this work we present a set of 104 2D foetal brain ultrasound images acquired during the 20th week of gestation that have been co-registered to a common space from a rough skull segmentation. The images are provided both on the original space and template space centred on the ellipses of all the subjects. Furthermore, the images have been annotated to highlight landmark points from structures of interest to analyse brain development. Both the final atlas template with probabilistic maps and the original images can be used to develop new segmentation techniques, test registration approaches for foetal brain ultrasound, extend our work to longitudinal datasets and to detect anomalies in new images.

Similar content being viewed by others

Normative spatiotemporal fetal brain maturation with satisfactory development at 2 years

Automated body organ segmentation, volumetry and population-averaged atlas for 3D motion-corrected T2-weighted fetal body MRI

An automatic multi-tissue human fetal brain segmentation benchmark using the Fetal Tissue Annotation Dataset

Background & summary.

Foetal brain development is a complex sequence of events ocurrying throughout gestation. From early stages of embryonic development, the brain undergoes structural changes until several months after birth 1 , 2 . Therefore, understanding normal brain development is essential to identify potential deviations which may lead to neurological disability. Specifically, various studies have described normal milestones within a specific chronology 3 , 4 , 5 . These changes can be observed by experts in prenatal diagnosis with skills in foetal neurology using ultrasound (US) and magnetic resonance imaging (MRI). However, identifying neurodevelopmental deviations is challenging due to the difficulty in image interpretation 5 .

Currently, US is the established technique for screening due to its ability to acquire dynamic images in real-time without radiation and cost-efficiently. The international guidelines recommend the acquisition of a routine first trimester US within the 11th - 14th weeks and a mid-trimester scan within the 19th - 22th weeks of gestation for anatomical evaluation of the foetus 6 . The standard foetal assessment includes the evaluation of planes acquired using 2D-US. Even though 3D-US is considered useful for prenatal diagnosis of some disorders (mainly involving the face, the skeleton, the cardiovascular system or the brain), the major obstacles for 3D-US implementation worldwide as the main routine acquisition type are related to foetal motion artefacts and acoustic shadowing during volume acquisition. Another widely used prenatal imaging technique is MRI. However, MRI is not used as a primary screening tool and is instead used as a complementary acquisition when foetal abnormalities are suspected (even in selected high-risk cases) 7 , 8 . Moreover, foetal MRI performed before the 18th-22nd weeks does not usually provide additional information to that obtained by US. Generally, MRI provides a detailed visualization of structures between the 26th and 32nd weeks, being superior to US and less susceptible to limitations from maternal body conditions and foetal presentation (bones do not produce occlusion artefacts in MRI) 8 . Within this context, a foetal brain US atlas based on 3D scans of healthy foetuses was recently published 9 . This atlas demonstrates the feasibility of assessing structural changes in the cortex and in the subcortical grey matter by US. Furthermore, this chronological US-based atlas of the foetal brain at each gestational age has the potential to become a useful tool to detect development abnormalities when combined with the standard planes of mid-trimester routine scans. However, considering that the primary diagnostic tool in pregnancy remains 2D-US, studies aimed at improving the identification of fetal structures should focus on 2D-US.

As there is a degree of variability in the shapes of anatomical regions between individuals, atlases are typically built by taking into account a number of images considered to fall within normal parameters. A crucial step in this process is finding corresponding regions between different images and warping them to a common space, a process known as image registration. In the case of US atlases, the lower signal-to-noise ratio (SNR), the differences in location for the foetuses due to the lack of a robust localization technique and the presence of image artefacts that may blur or shadow salient features and structure boundaries makes this process particularly challenging.

In this paper we present a set of foetal brain 2D-US images with manual landmark annotations of structures of interest and soft probabilistic maps for those structures based on these landmarks in a common space. The aim of this dataset is to provide a set of US images as a starting point to study registration and segmentation of foetal brain scans and provide tools for researchers focusing on foetal brain development. Furthermore, in combination with the recently published 4D atlas, this dataset can be also used to study 2D to 3D US registration and techniques to determine the “real age” of gestation and potential abnormalities when comparing to the atlas. To define a common space, a registration algorithm based on fitting an ellipse to the skull segmentation of a convolutional neural network (CNN) was used to align all the images to a common space. In order to avoid misalignements caused by image quality, a subject ellipse was used to estimate an affine transformation to a common (atlas) space. We also took advantage of the automatic differentiation capabilities of the pytorch package in three fundamentally different optimization settings (training a CNN for segmentation, fitting an ellipse to a segmentation boundary, and estimating an affine transformation).

In order to create the dataset presented, the data collected was processed as follows: First, the skull was automatically segmented using a UNet network, then an ellipse was automatically fitted to the shape of the skull. These ellipses where used to register each image to a reference image. Soft probabilistic maps where built using the set of registered images. The rest of this section includes details on each of these steps and a summary of other publicly available datasets. A visual summary can be found in Fig.  1 .

Methodology: 1) The skull is automatically segmented using a UNet network (top-left). 2) An ellipse is fitted to the skull segmentation (top-right) 3) to estimate an affine transformation to a reference image (bottom-right). The axes of the fitted ellipse are warped to the image coordinate axes and are re-scaled to fit the ellipse in the reference image (bottom-left).

Ultrasound data

A prospective cohort of low-risk pregnant women was recruited at routine mid-trimester foetal ultrasound scan. All participants initiated antenatal care before the 12th weeks of gestation, underwent the first trimester ultrasound scan between 12th and 14th weeks and had a low risk for aneuploidies in the first trimester combined screening. Written informed consents were obtained from participants. A private dataset of 70 pregnant women with a routine mid-trimester foetal ultrasound scan at [20 ~ 20.6] weeks without detected abnormalities was acquired, totalling 104 scans (8 women were scanned three times, 18 women were scanned twice and the remaining 44 women were only scanned once). The median of the maternal age was 31 (range 18-42). Images were acquired using high-frequency transabdominal probe (C2-9) of Voluson E10 ultrasound system. For each subject, a transverse view of the foetal head demonstrating a standard normative transcerebellar scanning plane was manually annotated by2 trained clinical experts (Fig.  1 top right). Both experts were part of the Prenatal Diagnosis Unit, one with over 10 years of experience and the other with 5 years of experience. The experts jointly participated in the annotations of each imageto highlight common structures related to brain development as follows:

Skull : 4 landmarks from the inner line of the skull were annotated: 2 at the level of the middle line and the other 2 in a perpendicular imagined line at the level of the posterior corners of the cavum septi pellucidi (CSP).

Cerebellar peduncles (Thalami) : 1 landmark marking the edges of both thalami at the middle line and the outer edges of the concavity shape were annotated (3 total points).

Cerebellum : 8 landmarks on the perimeter the cerebellum were annotated. Specifically, 2 points from the midline, 2 points from the cerebellum external edges and 4 points in the middle of each cerebellum hemisphere.

Cavum septi pellucidi (Cavum) : 4 points, each marking one corner of its rectangular shape, were annotated.

Sylvian Fissure (Sylvius) : 2 landmarks, one for each sylvian fissure edge, and 1 landmark in the inflection point of the fissure were annotated. For all the images, only the inferior fissure was visible.

Midline : 1 landmark in the upper edge of the midline and 1 landmark in the upper edge of the CSP were annotated.

An important aspect of acquisition is that the sylvian fissure was always scanned on the lower part of the image, irrespective of the head orientation (left to right or right to left). This phenomena has important implications for registration. When aligning all images to a common 3D space (for example a 3D atlas template) in order to have all images facing the same direction, the transformation can be modeled with 180° rotations over the y axis. If the transformation is limited to the 2D space (image coordinates), a mirroring operation has to be applied to align all images. In the code we provide to process the images, the second option is used for simplicity. Furthermore, our common space is oriented from left to right (anterior to posterior).

Data acquisition was approved by the ethics committee “Comitè d’Ètica d’Investigació amb Medicaments CEIM GIRONA” with the code 2023.067. Furthermore, the subjects were informed and consented to the open publication of the data.

Skull segmentation database

The segmentation dataset used to train a skull segmentation CNN was downloaded from the HC18 grand challenge on “automated measurement of foetal head circumference using 2D ultrasound images” 10 , 11 . The challenge comprises a set of 800 × 540-pixel 2D US images with a pixel sizes ∈ (0.052, 0.326) mm. The data set was split into 999 images for training and 335 for testing. For each image in the training set, an ellipse was manually fitted to the HC by a trained sonographer but precise segmentations of the true skull boundary were not provided. No pre-processing techniques were applied to the images apart from computing the z-score of the intensity values with respect to the mean and standard deviation of the whole image (non-0 intensities) before feeding them to the segmentation network. The network trained with all the training set images was then used to roughly segment the skull of our own 2D US images and provide a starting point to fit an ellipse.

Other public databases

To the best of our knowledge, only 3 other large datasets focusing on 2-D ultrasound have been made publicly available, including HC18. As mentioned on our brief description of HC18, the annotations are limited (only an ellipse roughly representing the skull is given), the subjects present a large distribution of gestational ages and no further information on the acquired plan e is provided. For studies focusing on the segmentation of structures of interest or registration, new annotations would be needed. Furthermore, the gestational age and acquisition plane for each subject can have a large impact on the appearance of the structures. Another public dataset presented by Burgos-Artizzu et al . 12 includes a large set of images (12400) ranging from the 18th to the 40th week of pregnancy. Images were acquired using six different machines and labelled at the image level. Similarly to HC18, the dataset presents a large range of gestational ages and limited annotations that could only be used to develop classification algorithms. Finally, Alzubaidi et al . 13 released a public dataset of 3832 high resolution images. In contrast with the other two datasets, no mention of gestational age is provided and once again rough annotations in the form of bounding boxes are provided. Moreover, the authors also highlight image resampling as an additional shortcoming of their dataset.

In our dataset, we focus on a specific acquisition plane and gestational age as defined by international guidelines and raw images are provided. Furthermore, landmarks for the most salient points of each structure of interest are provided with software tools to estimate finer-grained masks and bounding boxes around the landmarks. In that sense, our dataset provides a useful tool to address multiple image analysis problems including registration that could not be easily approached with other available datasets and no additional annotations.

Registration method

Medical image registration 14 , 15 is a necessary initial step for many medical image processing applications that rely on group-wise analysis. Typically, images are registered in pairs: one of them is defined as the “fixed” image (or reference) and the other as the “moving” image. The moving image is then warped using a transformation function to generate the final “moved” image. The transformation is commonly optimised using a predefined metric that computes image similarity between the reference and the moved image after transformation. Here, we present a coarse registration method to roughly align different foetal ultrasound images as described in the following sections and illustrated in Fig.  1 . We chose one of the images as the common “reference” image (image 10) and registered the remaining images to it.

Automatic Skull segmentation using a Unet

The Unet architecture 16 is one of the most common CNN architectures for image segmentation. Due to its encoder-decoder structure and the use of skip connections, precise segmentations based on multi-scale features can be obtained for a variety of segmentation applications 17 . Consequently, we used a 2D Unet trained on the HC18 challenge dataset to provide a rough segmentation of the skull for all the images in our registration dataset (see Fig.  1 for an example). Specifically, the encoder and decoder blocks are comprised of 6 convolutional layers with a residual connection 18 of 32, 32, 128, 256, 256 and 1024 features each (inverse order for the decoder) and a bottleneck of 1024 features. To optimise the weights of the Unet, the Adam algorithm with default initial learning rate was used to minimise the binary cross-entropy loss.

Ellipse Registration

An ellipse is a planar curve representing the locus of the points with constant added distances to two “focal points”, as expressed by the quadratic equation:

With points that satisfy f E ( X ,  Y ) ≠ 0 being inside the ellipse perimeter ( f E ( X ,  Y ) < 0) or outside of it ( f E ( X ,  Y ) > 0). The general equation’s coefficients can be obtained from known semi-major axis a (represented by the magnitude of the turquoise vector in Fig.  1 ), semi-minor axis b (represented by the magnitude of the cyan vector in Fig.  1 ), centre coordinates ( x 0 ,  y 0 ) (represented by the point where the two vectors meet in Fig.  1 ) and rotation angle θ (the angle from the positive horizontal axis to the ellipse’s major axis as observed in Fig.  1 ) using the formulae:

These expressions can be derived from the canonical equation \(\frac{{x}^{2}}{{a}^{2}}+\frac{{y}^{2}}{{b}^{2}}=1\) by an affine transformation of the coordinates ( x ,  y ) (with a translation (− x 0 , − y 0 ) and an angle θ ).

Given the set of pixel coordinates of the skull segmentation ( \(({X}_{skull},{Y}_{skull})=\,[({x}_{1},{y}_{1}),\ldots ,({x}_{N},{y}_{N})]\) ) we can fit an ellipse using its parameters ( a , b , x 0 , y 0 and t h e t a ) by minimising the following objective function:

where the coefficients A to F are substituted in f E by their definitions in equations ( 2 ) to ( 7 ) and the estimated ellipse parameters.

This process is repeated 5 times, removing erroneous points of the skull segmentation mask that are one standard deviation away from the mean ellipse error according to Eq. ( 10 ). With the parameters of the skull ellipse estimated through optimisation, we can now define an affine transformation (referred to as Ellipse from now on) to move the brain to the centre of the image as:

Affine image registration

For comparison, a regular rigid registration of 6 unrestricted parameters was performed with different initialisations. From the most basic identity initialisation (referred to as Affine ), to an initialisation using the ellipse parameters of the reference image (referred to as Affine (Reference ellipse) ) and a refinement of the ellipse-based affine transformation from Eq. ( 11 ) (referred to as Ellipse + Affine ).

Probabilistic maps

Once the images are co-registered to a common space based on their ellipse, a two-step process is performed to generate probabilistic maps for all the structures with more than 2 landmarks.

First, the concave hull of each structure is computed using the alphashapes package. This concave hull represents a rough segmentation of the location of the structure. Second, the segmentations for all the subjects are averaged per structure to determine the probability of each pixel to belong to that structure.

This approach has some limitations. Namely, some structures have a polygonal shape, even though they are actually curves (e.g. sylvius) and the final masks are only a rough representation of the real boundaries (e.g. cerebellum). However, these polygonal maps can be still used to determine growth milestones and to provide a rough location of the structure of interest.

Data Records

The original images with manual landmark annotations (Gimp image editing tool format) and the co-registered images and probabilistic maps used within this paper can be found on figshare 19 , and are organised with subject id number (1-52) and an additional number for multiple scans (e.g. 36 and 36.1). Co-registered images are saved in JPEG format with the “_registered” suffix. The final probability maps estimated using a combination of co-registered landmarks and the alphashape package are provided as JPEG images with the name of the structure (e.g. sylvius.jpeg), while comma separated value (CSV) files with the point landmarks of all registered subjects are compressed into a single zip file.

In total, the released dataset consists of 104 annotated 2D US images of foetal brains on the 20th week of pregnancy. The manual annotations are described in Ultrasound dataset section.

Technical Validation

To validate the techniques used for co-registration to a common space, we focused on common medical imaging metrics for registration using landmarks. In this section we describe these metrics and provide some qualitative and quantitative results of the alignment (including a visualisation of the probabilistic maps of one of the structures of interest).

Regarding the quality of the US images, all images were acquired with a high frequency ultrasound probe (2-9 MHz) that provides high resolution images following the ISUOG recommendations. The most constraining factor for the quality of the ultrasound images is maternal obesity. However, for this study we excluded women with maternal morbid obesity (body mass index > 40) as it is one of the factors for high-risk pregnancy. To further illustrate that point, we provide a comparative example between an image from the dataset and a lower quality one in Fig.  2 .

Qualitative comparison between a low quality image where structures are not clearly visible and and image from the dataset.

Evaluation metrics

For this study we chose to use the anatomical knowledge defined by expert annotations as the main directive to evaluate the quality of the registration and avoid focusing exclusively on common pixel-metrics that might be unreliable and disconnected from physical properties 20 . Specifically, we used the point annotations described in the Ultrasound Data section with two point-based metrics and one area-based metric for every anatomical structure with more than 2 points (all the structures except the midline). For completeness sake, we also included the structural similarity index metric (SSIM) pixel-based metric.

Point-based metrics

In order to penalise partial matches between anatomical structures, we considered the Hausdorff distance ( d H ), that computes the worst possible Euclidean distance ( d ( p i ,  p j )) between two sets of points P f (| P f | =  N ) and P m (| P m | =  M ). We also considered the average of the minimum Euclidean distances ( d E ) to express global similarity between structures defined by landmarks.

Area-based metric

To evaluate the superposition between two anatomical structures defined as 2D landmarks (points), we first computed their concave hulls and then considered their Dice similarity coefficient (polygon DSC).

Image similarity metric

For completeness, we also considered the structural similarity index measure (SSIM) as a pixel-intensity-based metric.

Comparison of coarse registration approaches

Figures  3 , 4 and 5 summarise the results with boxplots and Wilcoxon signed-rank tests for all the registration methods and metrics considered. Wherever possible, the results for different anatomical areas are presented separately. For the Euclidean and Hausdorff metrics, lower values indicate better registration, while for the SSIM and Dice metric higher results indicate better registration. Moreover, a qualitative example to illustrate misalignments between the reference points and the registered landmarks is provided in Fig.  6 .

Quantitative results for all the methods on each structure (Hausdorff distance, d H , lower values indicate better registration). The upper part of each boxplot figure indicates the results of pairwise statistical Wilcoxon tests: (ns: 5.00e-02  < p ≤ 1.00e+00, * 1.00e-02  < p ≤ 5.00e-02, ** 1.00e-03  < p ≤ 1.00e-02, *** 1.00e-04  < p ≤ 1.00e-03, **** p ≤ 1.00e-04).

Quantitative results for all the methods on each structure with multiple points (polygon DSC, higher values indicate better registration). The upper part of each boxplot figure indicates the results of pairwise statistical Wilcoxon tests: (ns: 5.00e-02  < p ≤ 1.00e+00 * 1.00e-02  < p ≤ 5.00e-02, ** 1.00e-03  < p ≤ 1.00e-02, *** 1.00e-04  < p ≤ 1.00e-03, **** p ≤ 1.00e-04).

Quantitative results for all the methods according to the SSIM metric (higher values indicate better registration). The upper part of each boxplot figure indicates the results of pairwise statistical Wilcoxon tests: (ns: 5.00e-02  < p ≤ 1.00e+00, * 1.00e-02  < p ≤ 5.00e-02, ** 1.00e-03  < p ≤ 1.00e-02, *** 1.00e-04  < p ≤ 1.00e-03, **** p ≤ 1.00e-04).

Qualitative example of the alignment between the reference image and a randomly selected subject (11). The background image corresponds to the warped image after registration, circles denote the reference points for each structure, crosses represent the registered points and lines are used to illustrate misalignment between the reference and the registration.

Regarding point metrics, the ellipse ( E ) and ellipse with affine methods ( E+A ) obtain overall better results than the other methods. In general, the differences observed were found to be statistically significant for both point metrics and most anatomical structures. Exceptions to this are the thalami and the cerebellum where the pixel based affine registration method initialised using the reference ellipse ( AFF+I ), obtains results that appear worse but are not significantly different. Comparing the E and E+A methods, small (and not statistically significant) differences can be observed. Using a refinement registration after ellipse-based method slightly worsens the metrics when applied to the skull but improves them in all other anatomical structures. This is an expected result as the main focus of the first method is to co-register the skulls (ellipses). The uninitialised affine transformation ( AFF ) obtains significantly worse results than these two methods and has a higher variance of values as illustrated by its wide boxplots. Both methods using exclusively pixel-wise affine registration (AFF, AFF+I) achieve results that are worse than the metrics of the original moving image in some cases. This illustrates the disconnect between pixel-based metrics guiding these methods and point-based metrics targeting distances between real anatomical structures (especially for noisy sequences).

Regarding the differences between the Euclidean and Hausdorff metrics (especially the median values shown by the central line in boxplots), slightly higher Hausdorff values indicate that some point pairs are further than the average euclidean distance mean value. This is especially noticeable on the skull for E and A+E and the thalami for AFF .

The other two metrics show the same general tendencies, even though they focus on different aspects of the registration. The low SSIM values observed for all methods (even though E and A+E outperform all methods) illustrats how challenging this registration scenario is. The variations between individuals and acquisitions and low SNR make the intensity values particularly unreliable. On the other hand, the high anatomical DSC results observed for the ellipse-based methods (includding AFF+I) validate our geometric approach that relies on a rough initial skull estimate to determine the general shape and orientation of the brain. Similarly high results obtained for the cerebellum indicate that a correct skull placement is a crucial important step towards the registration of all brain structures.

Finally, Fig.  7 shows the final heatmap generated from the registered concave hulls (E+A) of the cerebellum for all output images. Even though faint outlines of some incorrect registration results outside of the cerebellum region (delimited by orange points) can be observed, the higher probability regions in the heatmap clearly correspond to the cerebellum region of the reference image delimited by the manually annotated orange landmarks.

Probabilistic map for the cerebellum structure based on the averaged concave hull of the landmarks for each image. Landmarks for the reference image (background) are also provided.

Usage Notes

The code provided to analyse the images and perform a rough initial alignment to a common space has been developed using python. A set of Jupyter notebooks detailing the use of each step is also provided in the repository with visualisation examples of each step of the processing pipeline. Furthermore, we provide the trained weights for the skull segmentation network as part of the data records (file unet.pt 19 ).

Regarding the use of python packages, the code heavily relies on pytorch (version 1.12.0 with CUDA 10.2) to do the heavy lifting. However, numpy (version 1.21.6), scipy (version 1.7.3), scikit-learn (version 1.0.2), scikit-image (version 0.18.1), pandas (version 1.3.4), persim (version 0.3.1) and alphashape (version 1.3.1) are also used for some of the processing steps or to analyse different metrics related to the registration. Furthermore, opencv-python (version 4.5.2.52) and gimpformats (version 2021.1.4) where used to open the images with python. In particular, manual landmark annotation where done using the Gimp software and were stored (together with the ultrasound image) using the xcf format. Finally, matplotlib (version 3.7.1), seaborn (version 0.11.0) and statannot (version 0.2.3) were used for results visualisation inside the Jupyter notebooks.

Code availability

All the code used in the study to generate the final atlas and to co-register the images is publicly available at https://github.com/marianocabezas/fetal-brain .

Monteagudo, A. & Timor-Tritsch, I. E. Normal sonographic development of the central nervous system from the second trimester onwards using 2d, 3d and transvaginal sonography. Prenatal Diagnosis 29 , 326–339 (2009).

Article   PubMed   Google Scholar  

Namburete, A. I. et al . Learning-based prediction of gestational age from ultrasound images of the fetal brain. Medical Image Analysis 21 , 72–86 (2015).

Article   PubMed   PubMed Central   Google Scholar  

Cohen-Sacher, B., Lerman-Sagie, T., Lev, D. & Malinger, G. Sonographic developmental milestones of the fetal cerebral cortex: a longitudinal study. Ultrasound in Obstetrics & Gynecology 27 , 494–502 (2006).

Article   CAS   Google Scholar  

Kline-Fath, B. M. Ultrasound and mr imaging of the normal fetal brain. Neuroimaging Clinics of North America 29 , 339–356 (2019).

Guimaraes, C. V. & Dahmoush, H. M. Fetal brain anatomy. Neuroimaging Clinics of North America 32 , 663–681 (2022).

Salomon, L. J. et al . Isuog practice guidelines (updated): performance of the routine mid-trimester fetal ultrasound scan. Ultrasound in Obstetrics & Gynecology 59 , 840–856 (2022).

Griffiths, P. D. et al . Mri in the diagnosis of fetal developmental brain abnormalities: the meridian diagnostic accuracy study. Health Technology Assessment (Winchester, England) 23 , 1 (2019).

Prayer, D. et al . ISUOG practice guidelines (updated): performance of fetal magnetic resonance imaging. Ultrasound in Obstetrics & Gynecology 61 , 278–287 (2023).

Namburete, A. I. et al . Normative spatiotemporal fetal brain maturation with satisfactory development at 2 years. Nature 1–9 (2023).

van den Heuvel, T. L. A., de Bruijn, D., de Korte, C. L. & van Ginneken, B. Automated measurement of fetal head circumference using 2d ultrasound images. PloS one 13 , e0200412 (2018).

van den Heuvel, T. L. A., de Bruijn, D., de Korte, C. L. & van Ginneken, B. Automated measurement of fetal head circumference using 2d ultrasound images [data set]. Zenodo , https://doi.org/10.5281/zenodo.1322001 (2024).

Burgos-Artizzu, X. P. et al . Evaluation of deep convolutional neural networks for automatic classification of common maternal fetal ultrasound planes. Scientific Reports 10 , 10200 (2020).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Alzubaidi, M. et al . Large-scale annotation dataset for fetal head biometry in ultrasound images. Data in Brief 51 , 109708 (2023).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Fu, Y. et al . Deep learning in medical image registration: a review. Physics in Medicine & Biology 65 , 20TR01 (2020).

Article   Google Scholar  

Gholipour, A., Kehtarnavaz, N., Briggs, R., Devous, M. & Gopinath, K. Brain functional localization: A survey of image registration techniques. IEEE Transactions on Medical Imaging 26 , 427–451, https://doi.org/10.1109/TMI.2007.892508 (2007).

Ronneberger, O., Fischer, P. & Brox, T. U-net: Convolutional networks for biomedical image segmentation. In Proceedings of the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) , 234–241 (2015).

Isensee, F., Jaeger, P. F., Kohl, S. A. A. & Maier-Hein, K. H. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nature: Methods (2021).

He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In IEEE Conf. Comput. Vision Pattern Recog ., 770 – 778 (2016).

Cabezas, M., Diez, Y., Martinez-Diago, C. & Maroto, A. A benchmark for 2d foetal brain ultrasound analysis [data set]. Figshare , https://doi.org/10.6084/m9.figshare.c.6984822.v1 (2024).

Rohlfing, T. Image similarity and tissue overlaps as surrogates for image registration accuracy: Widely used but unreliable. IEEE Transactions on Medical Imaging 31 , 153–163 (2012).

Download references

Author information

These authors contributed equally: Mariano Cabezas, Yago Diez, Clara Martinez-Diago, Anna Maroto.

Authors and Affiliations

Brain and Mind Centre, University of Sydney, Sydney, Australia

Mariano Cabezas

Faculty Of Science, Yamagata University, Yamagata, Japan

Hospital Universitari de Girona Doctor Josep Trueta, Girona, Spain

Clara Martinez-Diago & Anna Maroto

You can also search for this author in PubMed   Google Scholar

Contributions

All authors conceived and designed the experiments, C.M. and A.M. collected the images and performed the landmark annotations, M.C. and Y.D. conducted the experiments and analysed the results. All authors reviewed the manuscript.

Corresponding author

Correspondence to Mariano Cabezas .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Cabezas, M., Diez, Y., Martinez-Diago, C. et al. A benchmark for 2D foetal brain ultrasound analysis. Sci Data 11 , 923 (2024). https://doi.org/10.1038/s41597-024-03774-3

Download citation

Received : 27 December 2023

Accepted : 14 August 2024

Published : 24 August 2024

DOI : https://doi.org/10.1038/s41597-024-03774-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.