Diagnosis of station and rotation of the fetal head in the second ...

Ultrasound Obstet Gynecol 2009; 33: 331–336Published online 9 February 2009 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/uog.6313Diagnosis of station and rotation of the fetal headin the second stage of labor with intrapartum translabialultrasoundT. GHI, A. FARINA, A. PEDRAZZI, N. RIZZO, G. PELUSI and G. PILUDepartment of Obstetrics and Gynecology, University of Bologna, Bologna, ItalyKEYWORDS:fetus; operative delivery; prenatal diagnosis; second stage of labor; translabial ultrasoundABSTRACTObjective To investigate the ability of intrapartumtranslabial sonography to diagnose fetal station in thesecond stage of labor.Methods Patients with uncomplicated pregnancies atterm gestation with fetuses in vertex presentation inthe second stage of labor underwent serial translabialsonography and digital examinations. In a sagittal sectionof the maternal pelvis, the direction of the head was notedand categorized as downward, horizontal or upward.By rotating the transducer in the transverse plane thecerebral midline echo was also visualized and the rotationof the head was noted. Clinical and ultrasound data werecompared using Somer’s d-test.Results Sixty patients underwent a total of 168 clinicaland sonographic examinations. When on the sonogramthe fetal head was directed downward, the station assessedclinically was most frequently ≤+1 cm from the ischialspines (44/57 (77.2%) cases); when the direction washorizontal, the station was most frequently ≤+2 cm(53/59 (89.8%) cases); when the fetal head was directedupward, the station was usually ≥+3 cm (46/52 (88.5%)cases). Failure to visualize the cerebral midline or arotation ≥ 45 ◦ were associated with a station of +2 cmor less in 98/103 (95.1%) examinations. Conversely, arotation of < 45 ◦ was associated with a station of +3 cmor more in 45/65 (69.2%) examinations. All comparisonsbetween clinical and sonographic findings demonstrateda statistically significant relationship (P < 0.0001). Theprobability of a station +3 cm or more was particularlyhigh when an upward direction of the head wasseen in combination with a rotation of < 45 ◦ (40/42(95.2%) examinations). The interobserver variability(Cohen’s kappa 0.795 and 0.727 for station and rotation,respectively; P < 0.001) and intraobserver variability(0.845 for both station and rotation, P < 0.001) suggestedgood reproducibility of the method.Conclusions Translabial sonography allows a diagnosisof fetal station with an accuracy comparable to that ofdigital examination and may provide useful informationfor diagnosing obstructed labor in the second stage aswell as assisting in the choice of instrumental delivery.Copyright © 2009 ISUOG. Published by John Wiley &Sons, Ltd.INTRODUCTIONIn the second stage of labor fetal station is assessedclinically to evaluate descent, to identify dystocia and todecide how to expedite delivery when this is indicated.Evaluation of fetal head rotation is also of value, asan angle < 45 ◦ to the anteroposterior diameter of thematernal pelvis suggests that the base of the skullis at or lower than the level of the ischial spines,which corresponds to the smallest diameter of thepelvis 1,2 . There is a paucity of studies on the accuracyof digital examination but the general consensus isthat reproducibility is low and diagnostic uncertaintyremains high even for operators with much experience 1–3 .Nevertheless, clinicians are frequently challenged withthe diagnosis of obstructed labor in the second stage.Furthermore, because of either failure to progress orfetal distress a choice often needs to be made whetherto perform an operative vaginal delivery or a Cesareansection. Failed vaginal extraction followed by Cesareansection is associated with an increased decision-to-deliveryinterval and with a much increased risk of fetal trauma 4–7 .On the other hand, Cesarean delivery when the fetal headCorrespondence to: Dr T. Ghi, Clinica Ostetrica e Ginecologica, Università degli Studi di Bologna, Via Massarenti 13, 40138 Bologna, Italy(e-mail: tullioghi@yahoo.com)Accepted: 7 November 2008Copyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd.ORIGINAL PAPER

332 Ghi et al.is impacted deeply in the maternal pelvis is associatedwith increased risk of maternal trauma, excessive bloodloss and infection, as well as of neonatal trauma andadmission to intensive care 1,7–9 .It has recently been suggested that translabial intrapartumultrasonography may be useful in assessing fetalhead engagement and station 10,11 . As the traditional clinicalevaluation has many limitations, a new tool capableof increasing diagnostic objectivity and accuracy wouldbe of great interest. The aim of our study was to compare,in a prospective study, clinical and sonographic evaluationsto test the hypothesis that ultrasonography providesvaluable information on fetal head descent in the secondstage of labor.PATIENTS AND METHODSThe study group included uncomplicated singletonpregnancies at term gestation (37 weeks or more) withfetuses in cephalic presentation in the second stage oflabor. Women attempting vaginal birth after Cesareansection were excluded. These patients were evaluatedconsecutively by two perinatologists with extensiveexperience in both ultrasonography and the managementof labor. Ultrasound is commonly used in our labor anddelivery unit. Patients were informed of the experimentalnature of the translabial scans and consented to theexamination. Starting from full dilatation of the cervix upto delivery serial translabial sonograms followed by digitalexaminations were performed at intervals of 15–30 min.Head station was expressed in cm from the level ofthe ischial spines; rotation was categorized as ≥ 45 ◦ or< 45 ◦ with regard to the anteroposterior diameter ofthe pelvis 1,2 . Digital and sonographic examinations wereperformed with the patient in a resting position andwith an empty bladder. The ultrasound examinationswere performed with a Voluson I (GE Medical Systems,Zipf, Austria) equipped with a multifrequency convextransabdominal transducer. The transducer was firstpositioned suprapubically to identify the position ofthe occiput by demonstrating the fetal orbits and/orcervical spine 12 . Then it was positioned translabiallyalong the sagittal plane to assess the direction ofthe fetal head (Figure 1), which was categorized asdownward, horizontal or upward as previously suggested(Figure 2) 10,11 . Eventually the transducer was rotated inthe transverse plane to identify the midline of the fetalhead, defined as the echogenic line interposed betweenthe two cerebral hemispheres. The angle formed by themidline and the anteroposterior diameter of the pubiswas assessed and was divided into two categories: ≥ 45 ◦and < 45 ◦ (Figure 3). Although digital and sonographicexaminations were performed by the same operators,labor and delivery were managed exclusively on the basisof the former evaluations and were not influenced by thesonographic findings. Operative vaginal deliveries wereperformed with vacuum application (Mytivac, Fismedical,Milan, Italy). The following information was recordedin each case: duration of second stage, type and modeof delivery (spontaneous, vacuum extraction, Cesareansection), maternal and fetal complications and, whenapplicable, indication for operative delivery. Operativeprocedures were also subjectively subdivided by theoperators who performed them into easy and difficult.The correlation between clinical and sonographicfindings was assessed by univariable analysis usingSomers’ d-test, which gives a measure of association forcontingency tables with ordered categories 13 .(a)PubisCaputFetal headFigure 1 Schematic representation (a) and translabial sonographic image (b) of the fetal position in the second stage of labor, with thetransducer aligned along the sagittal plane of the maternal pelvis, showing the pubis and fetal head. The caput can be easily differentiatedfrom the skull table.Copyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2009; 33: 331–336.

Translabial ultrasound in the second stage of labor 333Pubis Pubis PubisCaputFetal skullSkullFigure 2 Categorization of fetal head direction (indicated by white arrows) in longitudinal translabial sonograms compared with schematicrepresentations: (a) downward direction; (b) horizontal direction; (c) upward direction.To assess interobserver variability 15 sonogramsdepicting the direction of the fetal head and 15 sonogramsdemonstrating rotation of the fetal head were chosenrandomly from the study group and were independentlycategorized by the two operators. The Cohen’s kappastatistic was used to measure the univariable agreementbetween the evaluations of the two operators. A value of1 indicates perfect agreement, and a value of 0 indicatesthat agreement is no better than chance.RESULTSThe study group included 60 women who underwent atotal of 168 ultrasound and clinical examinations (range1–6, median 3). The patients were mostly nulliparous(47/60 (78.3%)) at a median gestational age of 40 (range,37–41) weeks.Assessment of the position of the fetal occiput wasdifficult with translabial ultrasound, as this only rarelyallows visualization of the orbits or cervical spine, and asuprapubic scan was usually necessary. Most fetuses hadan anterior occiput position both at the first ultrasoundexamination and at the time of delivery. Of the nine fetusesthat were found to have a posterior occiput positionduring the second stage, five were delivered as such andthe remaining converted spontaneously to an anteriorocciput position.Forty-three patients (71.7%) had a spontaneous vaginaldelivery. Vacuum extractions and Cesarean sectionswere performed in 11 (18.3%) and six (10.0%) cases,respectively, owing to arrest of descent and/or nonreassuringfetal monitoring. The mean duration of thesecond stage and mean birth weight were 63.1 ± 36.0 minand 3336 ± 401 g, respectively. No adverse outcomeswere registered in this group of patients.Summaries of the sonographic and clinical data aregiven in Tables 1 and 2. As previously described, the fetalskull table and the caput were always easily differentiatedon the sonograms (Figure 1) 10,11 . The head direction wasvisualized and usually it was easily categorized. Theexception was represented by cases with a posteriorocciput position, in which head direction was at timesdifficult to define, and in general tended to remainhorizontal throughout the entire second stage. When thefetal head was directed downward the station clinicallyestablished was most frequently ≤+1 cm (44/57 (77.2%)cases); when the direction was horizontal, the stationwas most frequently ≤+2 cm (53/59 (89.8%) cases);Copyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2009; 33: 331–336.

334 Ghi et al.(a)(c)Midline21Figure 3 Translabial ultrasound in the axial plane of the maternal pelvis. The transducer is rotated by 90 ◦ (curved arrow) (a), to visualize themidline of the fetal head (b). (c) Rotation was subdivided into two categories: ≥ 45 ◦ (1) and < 45 ◦ (2).Table 1 Correlation between fetal head direction established sonographically and station determined by digital examinationStation (digital examination) (n (%))*Ultrasound findings (sagittal view)* ≤+0 cm +1 cm +2 cm ≥+3 cmHead direction downward (n = 57) 14 (24.6) 30 (52.6) 13 (22.8) —Head direction horizontal (n = 59) 2 (3.4) 9 (15.3) 42 (71.2) 6 (10.2)Head direction upward (n = 52) — — 6 (11.5) 46 (88.5)*P < 0.0001 for all comparisons (Somers’ d-test).Table 2 Correlation between fetal head rotation established sonographically and station determined by digital examinationStation (digital examination) (n (%))*Ultrasound findings (transverse view)* ≤+0 cm +1 cm +2 cm ≥+3 cmMidline not visualized (n = 28) 11 (39.3) 17 (60.7) — —Rotation ≥ 45 ◦ (n = 75) 7 (9.3) 16 (21.3) 47 (62.7) 5 (6.7)Rotation < 45 ◦ (n = 65) 4 (6.2) 3 (4.6) 13 (20) 45 (69.2)Rotation ≥ 45 ◦ occiput anterior (n = 71) 6 (8.5) 25 (35.2) 39 (54.9) 1 (1.4)Rotation < 45 ◦ occiput anterior (n = 49) — 1 (2.0) 7 (14.3) 41 (83.7)*P < 0.0001 for all comparisons (Somers’ d-test).when the fetal head was directed upward, the stationwas usually ≥+3 cm (46/52 (88.5%) cases). There was astatistically significant relationship between the results ofclinical examination and ultrasound findings according tothe Somers’ d-test (P < 0.0001 for all comparisons).Most vacuum extractions were only performed witha clinical station of +2 cm or more. In most of thesecases the fetal head was directed upward (9/11). In the sixcases in which Cesarean section was performed the headdirection was horizontal in four cases and downward intwo cases. In two cases the vacuum was applied withthe fetal head directed horizontally. In one of these, thefetus had an occiput posterior position and the procedurewas described as difficult, and was complicated by twodetachments of the cup. In all the remaining cases vacuumextractions and Cesarean sections were described as easy.The midline of the fetal head was seen on ultrasoundin only about half the cases in which the station was 0 or+1 (30/58 (51.7%) cases) because of acoustic shadowingfrom the pubic bones, while it was well identified withstation +2 and more (110/110 (100%) cases). Amongthe 140 ultrasound examinations that resulted in a clearvisualization of the midline, rotation of ≥45 ◦ was usuallyfound with a station of +2 cm or less (70/75 (93.3%)cases) whereas rotation was mostly < 45 ◦ with a station of≥+3 cm (45/65 (69.2%) cases). The correlation betweenhead station assessed clinically and rotation was statisticallysignificant according to Somers’ d-test (P < 0.001).The correlation between clinical station and sonographicrotation was stronger when only fetuses withanterior occiput position were included in the evaluation.In these cases a rotation of ≥ 45 ◦ was found at a stationCopyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2009; 33: 331–336.

Translabial ultrasound in the second stage of labor 335Table 3 Correlation between sonographically determined fetal head rotation and directionUltrasound findings (sagittal view) (n (%))*Ultrasound findings (transverse view)* Head downward Head horizontal Head upwardMidline not visualized (n = 28) 28 (100) — —Rotation ≥ 45 ◦ (n = 75) 29 (38.7) 44 (58.7) 2 (2.7)Rotation < 45 ◦ (n = 65) 5 (7.7) 15 (23.1) 45 (69.2)Rotation ≥ 45 ◦ occiput anterior (n = 71) 27 (38.0) 42 (59.2) 2 (2.8)Rotation < 45 ◦ occiput anterior (n = 49) 1 (2.0) 6 (12.2) 42 (85.7)*P < 0.0001 for all comparisons (Somers’ d-test).of ≤ 2 cm in 70/71 (98.6%) cases and a rotation of < 45 ◦was found at a station of ≥+3 cm in 41/49 (83.7%) cases(P < 0.001). Indeed, some fetuses with a posterior occiputwere found to have a rotation < 45 ◦ already at the pelvicinlet.Vacuum was always applied in fetuses with a rotationof < 45 ◦ . In the six Cesarean sections, rotation wasundetectable or ≥ 45 ◦ in two and four cases, respectively.The correlation between sonographically determinedfetal head rotation and direction is shown on Table 3.When the midline was not visualized, the fetal headwas always directed downward (28/28 (100%) cases).Among the 120 studies where rotation could be assessedand the occiput was anterior, an angle of ≥ 45 ◦ (71cases) was usually found if fetal head direction wasdownward or horizontal (69/71 (97.2%) cases), whereaswhen the rotation was < 45 ◦ (49 cases) fetal head directionwas mostly upward (42/49 (85.7%) cases). When headrotation was < 45 ◦ and the head direction was upward,the station was ≥+3cmin40of42(95.2%)cases.There was good agreement between the two operatorsin categorizing the sonograms. Cohen’s kappa yieldeda value of 0.795 and 0.727 for station (stratified intothree levels) and rotation (< 45 ◦ or ≥ 45 ◦ ), respectively.Associated P-values were both statistically significant.The intraobserver variability yielded a kappa of 0.845(P < 0.001) for both rotation and station.DISCUSSIONOur results are in agreement with those of previousstudies and confirm that translabial sonography providesvaluable information on fetal station in the second stageof labor 10,11 . On translabial sonograms, the descent ofthe fetal head in the maternal pelvis describes a curvedpath (Figure 4). In the downward slope of the curve thehead is consistently in the upper third of the pelvis. Whenthe direction is horizontal the head is usually in themidpelvis. Eventually, an upward direction indicates thatthe head is most likely in the lower third of the pelvis.Demonstration of fetal rotation is also of value. The useof a transverse view of the fetal head to assess the anglebetween the cerebral midline and the anteroposteriordiameter of the maternal pelvis is a novel approach thatin our hands provided significant additional information.The combination of an upward direction of the fetal headPubisIschial spineFigure 4 Schematic representation of the curved path that the fetalhead describes (curved arrow) during a normal second stage oflabor.with a rotation of < 45 ◦ indicates with almost completecertainty that the station is +3 cmormore.We do acknowledge some limitations of our study.We have compared sonographic findings with clinicalfindings, which are imprecise by universal agreement 1–3 .Another limitation of our study is that ultrasound findingswere more predictive of head station when the occiput wasin the anterior position. Categorizing fetal head directionwas more difficult with a posterior occiput. In these casesthe head tended to remain horizontal even when in thelower part of the pelvis. Similarly, some of these fetuseswere seen to have a rotation < 45 ◦ already at the pelvicinlet. Clearly, the process of labor is more complex andless predictable for fetuses with a posterior occiput. Thismalpresentation continues to represent a challenge formodern obstetrics 14,15 and our experience suggests that inthese cases ultrasound may prove less helpful than whenthe occiput is in a normal anterior position.In our series, nearly one third of the patients underwentoperative delivery, which is certainly a higher proportionthan expected in a group of normal patients in the secondstage of labor. This is probably the consequence of thestudy design. Our patients were consecutively recruitedamong those assisted by two senior obstetricians who aremostly involved in difficult deliveries.Copyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2009; 33: 331–336.

336 Ghi et al.Even accepting the limitations of our study, we believethat our results indicate a potential role for translabialsonography in the evaluation of labor. Serial sonographicobservations demonstrating that the fetal head directionfollows a curved path and rotation of the midlineecho towards the anteroposterior axis of the pelvis isreassuring for the normal progression of labor. If thereis an indication for an instrumental delivery, informationprovided by sonography may prove useful as well. Theoption of vaginal extraction is discouraged when thefetal head is less than +2 cm below the level of theischial spines. On the other hand, when the stationis +3 cm or more, extraction carries little if any riskabove that of a spontaneous delivery 11 . Our experienceindicates that a downward direction of the fetal head isusually associated with a station of 0–1 cm, thereforein these cases the choice of a Cesarean section shouldbe considered. Conversely, with an upward direction(the ‘head up’ sign previously described) 11 , particularlyif associated with a rotation of < 45 ◦ , vaginal extractionshould be favored. Our experience is as yet insufficientto provide a clear indication in cases with a horizontaldirection of the head. We also recommend caution ininterpreting ultrasound findings when the occiput of thefetus is posterior.In recent years many reports have proposed arole for ultrasound in the evaluation of laboringpatients 10 –12,16 –24 . A general trend towards an increaseduse of this method has been described 16 . Informationprovided by sonography is manifold. Our study confirmsthat translabial sonography may provide a new andvaluable diagnostic tool for the assessment of the secondstage of labor, although the optimal approach to theevaluation of these sonograms is yet to be demonstrated.We have found that a simple qualitative evaluationis feasible. However, the ultrasound images obtainedtranslabially lend themselves to measurement of distancesand angles, and we expect that attempts at a quantitativeapproach will be the subject of further studies in the nearfuture.REFERENCES1. Hale R. Rosen’s Management of Labor. Physician’s Judgementand Patient Care (2 nd edn). Chapman & Hall: New York, NY,1998.2. Operative vaginal delivery. ACOG Technical Bulletin Number196 – August 1994 (replaces No. 152, February 1991). Int JGynaecol Obstet 1994; 47: 179–185.3. Dupuis O, Silveira R, Zentner A, Dittmar A, Gaucherand P,Cucherat M, Redarce T, Rudigoz RC. Birth simulator: reliabilityof transvaginal assessment of fetal head station asdefined by the American College of Obstetricians and Gynecologistsclassification. Am J Obstet Gynecol 2005; 192:868–874.4. Olagundoye V, MacKenzie IZ. The impact of a trial ofinstrumental delivery in theatre on neonatal outcome. BJOG2007; 114: 603–608.5. Towner D, Castro MA, Eby-Wilkens E, Gilbert WM. Effect ofmode of delivery in nulliparous women on neonatal intracranialinjury. NEnglJMed1999; 341: 1709–1714.6. Alexander JM, Leveno KJ, Hauth J, Landon MB, Thom E,Spong CY, Varner MW, Moawad AH, Caritis SN, Harper M,Wapner RJ, Sorokin Y, Miodovnik M, O’Sullivan MJ, Sibai BM,Langer O, Gabbe SG; National Institute of Child Health andHuman Development Maternal–Fetal Medicine Units Network.Fetal injury associated with cesarean delivery. Obstet Gynecol2006; 108: 885–890.7. Murphy DJ, Liebling RE, Patel R, Verity L, Swingler R. Cohortstudy of operative delivery in the second stage of labour andstandard of obstetric care. BJOG 2003; 110: 610–615.8. Levy R, Chernomoretz T, Appelman Z, Levin D, Or Y, HagayZJ. Head pushing versus reverse breech extraction in casesof impacted fetal head during Cesarean section. Eur J ObstetGynecol Reprod Biol 2005; 121: 24–26.9. Murphy DJ, Liebling RE, Verity L, Swingler R, Patel R. Earlymaternal and neonatal morbidity associated with operativedelivery in second stage of labour: a cohort study. Lancet2001; 358: 1203–1207.10. Dietz HP, Lanzarone V. Measuring engagement of the fetalhead: validity and reproducibility of a new ultrasoundtechnique. Ultrasound Obstet Gynecol 2005; 25: 165–168.11. Henrich W, Dudenhausen J, Fuchs I, Kamena A, Tutschek B.Intrapartum translabial ultrasound (ITU): sonographic landmarksand correlation with successful vacuum extraction.Ultrasound Obstet Gynecol 2006; 28: 753–760.12. Akmal S, Tsoi E, Kametas N, Howard R, Nicolaides KH.Intrapartum sonography to determine fetal head position.J Matern Fetal Neonatal Med 2002; 12: 172–177.13. Buchanan W. Nominal and ordinal bivariate statistics: thepractitioner’s view. Am J Political Science 1974; 18: 625–646.14. Ponkey SE, Cohen AP, Heffner LJ, Lieberman E. Persistent fetalocciput posterior position: obstetric outcomes. Obstet Gynecol2003; 101(5 Pt 1): 915–920.15. Cheng YW, Shaffer BL, Caughey AB. The association betweenpersistent occiput posterior position and neonatal outcomes.Obstet Gynecol 2006; 107: 837–844.16. Sherer DM. Intrapartum ultrasound. Ultrasound ObstetGynecol 2007; 30: 123–139.17. Sherer DM, Abulafia O. Intrapartum assessment of fetal headengagement: comparison between transvaginal digital andtransabdominal ultrasound determinations. Ultrasound ObstetGynecol 2003; 21: 430–436.18. Sherer DM, Miodovnik M, Bradley KS, Langer O. Intrapartumfetal head position II: comparison between transvaginal digitalexamination and transabdominal ultrasound assessment duringthe second stage of labor. Ultrasound Obstet Gynecol 2002;19: 264–268.19. Sherer DM, Miodovnik M, Bradley KS, Langer O. Intrapartumfetal head position I: comparison between transvaginal digitalexamination and transabdominal ultrasound assessment duringthe active stage of labor. Ultrasound Obstet Gynecol 2002; 19:258–263.20. Sherer DM, Onyeije CI, Bernstein PS, Kovacs P, Manning FA.Utilization of real-time ultrasound on labor and delivery in anactive academic teaching hospital. Am J Perinatol 1999; 16:303–307.21. Akmal S, Kametas N, Tsoi E, Hargreaves C, Nicolaides KH.Comparison of transvaginal digital examination with intrapartumsonography to determine fetal head position beforeinstrumental delivery. Ultrasound Obstet Gynecol 2003; 21:437–440.22. Akmal S, Kametas N, Tsoi E, Howard R, Nicolaides KH.Ultrasonographic occiput position in early labour in theprediction of caesarean section. BJOG 2004; 111: 532–536.23. Akmal S, Tsoi E, Nicolaides KH. Intrapartum sonography todetermine fetal occipital position: interobserver agreement.Ultrasound Obstet Gynecol 2004; 24: 421–424.24. AkmalS,TsoiE,HowardR,OseiE,NicolaidesKH.Investigationof occiput posterior delivery by intrapartum sonography.Ultrasound Obstet Gynecol 2004; 24: 425–428.Copyright © 2009 ISUOG. Published by John Wiley & Sons, Ltd. Ultrasound Obstet Gynecol 2009; 33: 331–336.