Human birth observed in real-time open magnetic resonance imaging

OBJECTIVE: Knowledge about the mechanism of labor is based on as- sumptions and radiographic studies performed decades ago. The goal of this study was to describe the relationship between the fetus and the pelvis as the fetus travels through the birth canal, using an open mag- netic resonance  imaging  (MRI) scanner. 

RESULTS: Delivery occurred  by progressive head extension.  However, extension was a very late movement that was observed when the occiput was in close contact with the inferior margin of the symphysis pubis, occurring simultaneously with gliding downward of the fetal head.
CONCLUSION: This  observational  study  shows,  for  the  first  time, that birth can be analyzed with real-time  MRI. MRI technology allows assessment of maternal and fetal anatomy during labor and delivery. 

Key words:  delivery, imaging, mechanism of labor, magnetic resonance imaging

Thorough understanding of maternal and fetal anatomy and physiology is essential for proper management of labor and delivery. The mechanism of labor is generally understood as the movements of the fetus in relation to the bony structures and soft tissues of the birth canal during labor. More than 1000 radiographic examinations of the pelvis and fetal head before, during, and after labor were performed in early studies to better understand the mechanism of birth.1-3   However, prenatal x-ray expo- sure  has  been  associated  with  an  increased risk of childhood cancer in case- control studies.4-6 
Nonionizing radiation is preferable in pregnant women. Numerous trials of magnetic resonance imaging (MRI) have not revealed any experimental or clinical evidence of fetal harm7-10;  thus, MRI is considered safe for the mother and fe- tus.11,12 MRI has been used to elucidate fetal anatomy,13-16 placental morphol- ogy,17 amniotic fluid volume,18  and bio- chemical assessment (ie, magnetic reso- nance spectroscopy).19  Furthermore, MRI pelvimetry used to be performed to predict consistently women at risk for cephalopelvic disproportion.20-23 Open-configuration MRI systems were de- signed to facilitate interventional proce- dures and functional MRI examinations and to increase patient comfort.24 A field strength of 1 Tesla or more is desirable for obtaining high-quality images in open MRI systems.
Ultrasound is increasingly used to doc- ument fetal head position and station within the maternal pelvis at various stages of labor,25-31 but it also has some limitations. We used an open MRI scanner to take images of a human delivery. Our main
goal was to describe the relationship be- tween fetal movements and position 

as the fetus passage through the birth canal, using an open MRI scanner.

MATERIA L AN D METHOD S
We designed the observational study to
maximize safety for the mother and fetus. Rupture of the amniotic membranes was not planned in early labor because it was suggested previously that the amniotic fluid could lower the intrauterine acoustic sound pressure by 30 dB.32 This is enough to reduce acoustic sound pressure to an ac- ceptable level (<90 dB). In the late second stage, as the fetal head extended and the perineum distended, cinematic MRI a quisition was terminated to ensure that the ears of the newborn were still covered by maternal soft tissue, thereby avoiding ex- posure to MRI noise. During the delivery a midwife (G.R.) and an obstetrician (C.B.) stayed in the magnet room (Figure 1). There were 2 screens inside, 1 to monitor the fetal heart tracing and the second to observe the MRI images. A neonatologist and an anesthetist were also asked to be present in the magnet room. If there had been an abnormal labor course or an emergency, we would have been able to in- terrupt the MRI birth immediately and transfer to the delivery unit. The MRI suite and the delivery unit are on the same floor, and the distance between them is less than

50 m.

The patient underwent intermittent elec- tronic fetal heart monitoring with a proto- type of a MRI-compatible telemetric system with the exception of the image acquisition time. This MRI-compatible wireless elec- tronic fetal heart rate monitoring prototype system, developed by us and modified from the Philips Avalon Cordless Transducer Sys- tem (Philips Healthcare, Best, The Nether-

 
 

Figure 1: Photograph of the open MRI scanner with the patient and the health care personnel before delivery 

 

lands), allows for continuous cardiotoco- graph tracing with few artifacts. MRI was performed on a 1.0 Tesla open high-field MRI scanner with vertical field orientation (Panorama; Philips Healthcare) using a BodySP-Xl receiver coil.

A T2-weighted multislice turbospin echo (TSE) single-shot sequence was used to visualize the midsagittal, coronal, and axial planes with the following settings:

1000 milliseconds time of repetition (TR),

100 milliseconds time of echo (TE), flip angle 90°, 40 slices of 6 mm with 1 mm gap, voxel size 1.4 X 1.6 mm, field of view (FOV) 300 X 262 mm, with constant level appearance (CLEAR) correction. The se- quences were repeated every 10 minutes during the second stage of labor. Real-time cinematic MRI series were acquired from the midsagittal plane for representation of the extension phase using an interactive TSE single-shot sequence (TR 1600 milli- seconds, TE 150 milliseconds, flip angle

90°, single slice of 6 mm, voxel size 1.4 X

1.5 mm, FOV 380 X 285 mm, with CLEAR

correction).

RESULT S

In November 2010, a 24 year woman at 37 5/7 weeks of gestation was admitted with regular con- tractions to the Department of Obstet- rics of the Charité University Hospital in Berlin, Germany. The patient received an epidural and was transferred to the open MRI suite. In addition, the cervix was fully dilated, and the presenting part was engaged. Eight MRI studies were performed over a period of 45 minutes: 7 antepartum studies (Figure 2) and 1 postpartum study. First, the woman was examined in the supine position with legs outstretched. In the active second stage, when the mother began expulsive efforts with the valsalva manoeuver, her legs were slightly abducted and sup- ported by padding. This period was eval- uated by real-time cinematic MRI series (Video Clip).

A 2585 gram appropriate-for-gesta- tional age boy with Apgar scores of 9, 9, and 10 at 1, 5, and 10 minutes. Umbilical artery and umbilical vein pH measure- ments are routinely assessed as part of our daily practice. However, because of technical difficulties with the umbilical artery blood sample in this case, only the umbilical vein pH was available, which was 7.32. A neonatologist assessed the condition of the baby. Immediately after childbirth, the maternal anatomy was imaged before and after expulsion of the placenta, using a BFFE sequence (Figure

3). The total individual study time in the magnet room was less than 1 hour. The woman tolerated the discomfort during labor well and her postpartum course was uneventful. She was discharged with her newborn 2 days after delivery. The pediatric screening examinations, in- cluding auditory tests, did not reveal any abnormalities.

Figure 2:View of the midsagittal MRI plane of the maternal pelvis before the expulsion phase without pushing

Figure 3: MRI examination of the maternal pelvis in the third stage of labor

COMMENTS

The mechanical factors that  influence the progress of labor are of interest to obstetricians, but they are often difficult to investigate. For many years, digital ex- amination was the only method that was used during labor to provide information about the mother’s bony pelvis and soft tis- sue and the fetus. This method has the dis- advantage that only limited areas of the fe- tus and birth canal can be assessed.

Because mechanical factors are pri- marily involved in the seven cardinal movements of labor33,34 (engagement, descent, flexion, internal rotation, exten- sion, external rotation, and expulsion) elucidation of the process of labor was also investigated by experimental studies on preserved pelvises and pelvic models. Anatomically correct models are imper-

ative for accurate simulations of normal and complicated deliveries.

Clearly, it is impossible to faithfully re- produce labor conditions in experiments involving models; therefore, conclusions based on the results of such studies remain hypothetical.35  In other words, it is diffi- cult to generate models that mirror accu- rately the in vivo relationships during the labor and delivery process, and hence, re- sults derived from simulation are often based on untested assumptions.

Ultrasound is the imaging modality of choice for pregnant women.36 Today the cardinal movements can be studied with sonography.37-40   Transperineal ultrasound is rapidly becoming an established method to assess progression of labor and the likelihood of a successful opera- tive vaginal delivery.41 However, specific bony landmarks of the maternal pelvis, such as the ischial spines, cannot be visu- alized by intrapartum ultrasound.25 Furthermore, it is impossible to evaluate the fetal attitude which is described as the degree of flexion or extension of the fetal

head in relation to the fetal spine34  because the fetal cervical spine is not visible by transperineal ultrasound.

Fetal MRI has become more widely available and has become an accepted and powerful complementary method for evaluation of the fetus.13  MRI can help to increase knowledge about mater- nal and fetal anatomy during labor. Bony structures as well as soft tissue could be assessed in great detail. MRI sequences of the birth process should be as quite as possible and resistant to movement arte- facts. Our preliminary experiments con- firmed the single-shot TSE sequence to be well suited for fetal imaging for these tasks; sequence optimization also aimed at minimizing the sound pressure level, which was given by the scanner in the range of 10.5–11.3 dB (Table).

Attitude-fetal  flexion or extension

In 1913, the German obstetrician Hugo Sellheim42  found that extension of the fetal head takes place after impingement of the suboccipital region of the fetal head on the mother’s symphysis. This was thought to be caused by rotation around a transverse axis running through the lower border of the symphysis, resulting in de- flexion of the fetal head at the atlantooccipital joint and extension of the cervical spine. This movement begins before crowning. In 1957, the radiological inves- tigations of Borell and Fernström43,44 sug- gested that Sellheim’s explanation of the extension was incorrect. Borell and Fern- ström suggested that the fetal head is flexed and glides continuously downward. The thoracic and cervical spine undergoes ex- tension during the last stage of labor, whereas the fetal head remains flexed until after expulsion. Furthermore, they stated that there is a relatively large distance be- tween the head and the lower border of the symphysis.

An intermediate position between these 2 theories was developed by Murray (1890)45  and Jones (1906).46  The view of Murray45  was that the head continues to glide downward at the same time as the movement of extension. According to Jones,46  the movement of extension does not occur merely at the articulation be- tween the occiput and the atlas but is pre- ceded by an extension of the entire cervical spine.

Our visualization of the normal mechanism of late second-stage labor by MRI shows that extension started as soon as the occiput was in close contact with the inferior margin of the symphysis pubis. Thereafter, extension was simultaneous with gliding downward of the fetal head. At this point, the birth canal curved 90° upward and the fetal head was delivered by extension and rotated around the symphysis pubis. To the best of our knowledge, this is the first time that this mechanism has been clearly visualized. Thus, our investigation of the mecha- nism of labor using real time MRI pro- duced results that are in line with Mur- ray’s and Jones’s theory, which were described more than a century ago.45,46

MRI-compatible cardiotocography Poutamo et al47 published a comparison of electronic fetal heart monitoring before and after MR imaging in 16 preg- nant women. The authors showed that MRI acquisition does not influence the fetal heart rate or fetal activity. Shake- speare et al48 and Vadeyar et al49 were the first to assess fetal well-being during MRI examination. They had to remove ferro- magnetic parts from the device to be able to use within the MRI scanner. These au- thors also found no visible effect of MRI on fetal heart rate patterns. In the present observation, the same approach regarding the cardiotocography was chosen, namely removing ferromagnetic parts to allow monitoring of fetal heart rate during MRI examination without interference.

Future research

Future research should include visual- ization of the first stage of labor by MRI. Arrest of labor, necessitating a cesarean delivery, is a major cause of maternal morbidity and mortality.50,51  An im- proved understanding of the mechanism of labor will help clinicians towards a more individualized approach to labor, allowing them to more easily distinguish normal and abnormal courses of labor. This knowledge would allow clinicians to intervene in a timely and effective fashion to ensure a favorable outcome. Furthermore, a basic knowledge of the attitude of the fetal head at the time of its passage through the lower part of the birth canal is of practical value in opera- tive vaginal deliveries. Future studies might also provide a basis for virtual re- ality computer programs to teach health care personnel in training.

The shape and direction of the birth canal has generally been investigated by palpation during labor and in frozen sec- tions from women who died during la- bor. There is no doubt that the human fetuses must negotiate a curve to be born.33  Nevertheless, there are conflict- ing results about the level of the curved part of the birth canal, the so-called knee. It is widely reported in textbooks, but not supported by evidence, that the knee lies at the level of the ischial spines.52  In contrast, Borell and Fernström1 stated that the curve of the birth canal lies lower and outside the bony pelvis and is entirely formed by the soft parts. MRI visualization of the ischial spine level during labor in comparison with the fetal head station during the extension phase may help to shed more light on this discussion.

In conclusion, collecting images of the fetus during delivery using an open MRI is feasible. We showed that MRI technol- ogy is useful for visualizing normal ma- ternal and fetal anatomy during labor. This observation opened a new way to study the mechanism of birth.              

REFERENCES

1. Borell U, Fernström I. The mechanism of la

bour. Radiol Clin North Am 1967;5:73-85.

2. Caldwell W, Moloy H. Anatomical variations in the female pelvis and their effect in labor with a suggested classification. Am J Obstet Gyne• col 1933;26:479.

3. Caldwell W, Moloy H, D’Esopo D. Further studies on the mechanism of labor. Am J Obstet Gynecol 1935;30:763-814.

4. Stewart A, Webb J, Giles B, Hewitt D. Malig• nant disease in childhood and diagnostic irradi• ation in utero. Lancet 1956;2:447.

5. Stewart A, Webb J, Hewitt D. A survey of childhood malignancies. Br  Med  J  1958;1:

1495-508.

6. Bithell JF, Stewart AM. Pre-natal irradiation and childhood malignancy: a review of British data from the Oxford Survey. Br  J  Cancer

1975;31:271-87.

7. Clements H, Duncan KR, Fielding K, Gow• land PA, Johnson IR, Baker PN. Infants ex• posed to MRI in utero have a normal paediatric assessment at 9 months of age. Br J Radiol

2000;73:190-4.

8. Kok RD, de Vries MM, Heerschap A, van den Berg PP. Absence of harmful effects of mag• netic resonance exposure at 1.5 T in utero dur• ing the third trimester of pregnancy: a follow-up study. Magn Reson Imaging 2004;22:851-4.

9. Baker PN, Johnson IR, Harvey PR, Gowland PA, Mansfield P. A three-year follow-up of chil• dren imaged in utero with echoplanar magnetic resonance. Am J Obstet Gynecol 1994;170:

32-3.

10. Myers C, Duncan KR, Gowland PA, John• son IR, Baker PN. Failure to detect intrauterine growth restriction following in utero exposure to MRI. Br J Radiol 1998;71:549-51.

11. Chen MM, Coakley FV, Kaimal A, Laros RK Jr. Guidelines for computed tomography and magnetic resonance imaging use during preg• nancy and lactation. Obstet Gynecol 2008;112:

333-40.

12. Kawabata  I,  Takahashi Y,  Iwagaki  S, Tamaya T. MRI during pregnancy. J Perinat Med 2003;31:449-58.

13. Estroff JA. The growing role of MR imaging in the fetus. Pediatr Radiol 2009;39(Suppl 2): S209-10.

14. Johnson IR, Symonds EM, Kean DM, et al. Imaging the pregnant human uterus with nu• clear magnetic resonance. Am J Obstet Gyne• col 1984;148:1136-9.

15. Garden AS, Griffiths RD, Weindling AM, Martin PA. Fast-scan magnetic resonance im• aging in fetal visualization. Am J Obstet Gynecol

1991;164:1190-6.

16. Timor-Tritsch  IE, Monteagudo A. Magnetic resonance imaging versus ultrasound for fetal central nervous system abnormalities. Am J Obstet Gynecol 2003;189:1210-1; author reply

1211-2.

17. Warshak CR, Eskander R, Hull AD, et al. Accuracy of ultrasonography and magnetic res• onance imaging in the diagnosis of placenta ac• creta. Obstet Gynecol 2006;108:573-81.

18. Zaretsky MV,  McIntire DD,  Reichel TF, Twickler DM. Correlation of measured amniotic fluid volume to sonographic and magnetic res• onance  predictions.  Am  J  Obstet  Gynecol

2004;191:2148-53.

19. Kok RD, van den Bergh AJ, Heerschap A, Nijland R, van den Bergh PP. Metabolic infor• mation from the human fetal brain obtained with proton   magnetic  resonance  spectroscopy. Am J Obstet Gynecol 2001;185:1011-5.

20. Fox LK, Huerta-Enochian GS, Hamlin JA, Katz VL. The magnetic resonance imaging- based fetal-pelvic index: a pilot study in the community hospital. Am  J  Obstet  Gynecol

2004;190:1679-85; discussion 1685-8.

21. Hoyte L, Thomas J, Foster RT, Shott S, Jakab M, Weidner AC. Racial differences in pel• vic morphology among asymptomatic nullipa• rous women as seen on three-dimensional magnetic resonance images. Am J Obstet Gy• necol 2005;193:2035-40.

22. Huerta-Enochian GS, Katz VL, Fox LK, Hamlin JA, Kollath JP. Magnetic resonance- based serial pelvimetry: do maternal pelvic di• mensions change during pregnancy? Am J Ob- stet  Gynecol 2006;194:1689-94;  discussion

1694-5.

23. Zaretsky MV, Alexander JM, McIntire DD, Hatab MR, Twickler DM, Leveno KJ. Magnetic resonance imaging pelvimetry and the predic• tion of labor dystocia. Obstet Gynecol 2005;

106:919-26.

24. Hailey D. Open magnetic resonance imag• ing (MRI) scanners. Issues Emerg Health Tech• nol 2006:1-4.

25. Bamberg C, Scheuermann S, Slowinski T, et al. Relationship between fetal head station established using an open magnetic resonance imaging scanner and the angle of progression determined by transperineal ultrasound. Ultra• sound Obstet Gynecol 2011;37:712-6.

26. Barbera A, Becker T, Macfarland H, Hob- bins J. Assessment of fetal head descent in la• bor  with  transperineal ultrasound. Teaching DVD. Washington, DC: American College of Obstetrics and Gynecology; 2003:176.

27. Barbera AF, Pombar X, Perugino G, Lezotte DC, Hobbins JC. A new method to assess fetal head descent in labor with transperineal ultra• sound. Ultrasound Obstet Gynecol 2009;33:

313-9.

28. Ghi T, Farina A, Pedrazzi A, Rizzo N, Pelusi G, Pilu G. Diagnosis of station and rotation of the fetal head in the second stage of labor with intrapartum translabial ultrasound. Ultrasound Obstet Gynecol 2009;33:331-6.

29. Henrich W, Dudenhausen J, Fuchs I, Ka• mena A, Tutschek B. Intrapartum translabial ul• trasound  (ITU): sonographic landmarks and correlation with successful vacuum extraction. Ultrasound Obstet Gynecol 2006;28:753-60.

30. Kalache KD, Duckelmann AM, Michaelis SA, Lange J, Cichon G, Dudenhausen JW. Transperineal ultrasound imaging in prolonged second stage of labor with occipitoanterior pre• senting fetuses: how well does the ‘angle of progression’ predict the mode of delivery? Ul• trasound Obstet Gynecol 2009;33:326-30.

31. Tutschek B, Braun T, Chantraine F, Henrich W. A study of progress of labour using intrapar• tum translabial ultrasound, assessing head sta• tion, direction, and angle of descent. BJOG

2010;118:62-9.

32. Glover P, Hykin J, Gowland P, Wright J, Johnson I, Mansfield P. An assessment of the intrauterine sound intensity level during obstet• ric echo-planar magnetic resonance imaging. Br J Radiol 1995;68:1090-4.

33. Norwitz ER, Robinson JN, Repke JT. Labor and delivery. In: Gabbe SG, Niebyl JR, Simpson JL, eds. Obstetrics: normal and problem preg• nancies, 4th ed. New York: Churchill Living• stone; 2001:353-400.

34. Liao JB, Buhimschi CS, Norwitz ER. Normal labor: mechanism and duration. Obstet Gyne• col Clin North Am 2005;32:145-64, vii.

35. Dietze M. A re-evaluation of the mechanism of labour for the contemporary midwifery prac• tice. Midwifery Matters 2001:3-8.

36. Levine D. Obstetric MRI. J Magn Reson Im•

aging 2006;24:1-15.

37. Akmal S, Tsoi E, Howard R, Osei E, Nico• laides KH. Investigation of occiput posterior de• livery by intrapartum sonography. Ultrasound Obstet Gynecol 2004;24:425-8.

38. Akmal S, Tsoi E, Kametas N, Howard R, Nicolaides KH. Intrapartum sonography to de• termine fetal head position. J Matern Fetal Neo• natal Med 2002;12:172-7.

39. Gardberg M, Laakkonen E, Salevaara M. Intrapartum sonography and persistent occiput

posterior position: a study of 408 deliveries. Obstet Gynecol 1998;91:746-9.

40. Souka  AP,  Haritos  T,  Basayiannis K, Noikokyri N, Antsaklis A. Intrapartum ultra• sound for the examination of the fetal head po• sition in normal and obstructed labor. J Matern Fetal Neonatal Med 2003;13:59-63.

41. Yeo L, Romero R. Sonographic evaluation in the second stage of labor to improve the as• sessment of labor progress and its outcome. Ultrasound Obstet Gynecol 2009;33:253-8.

42. Sellheim H. Die Geburt des Menschen nach anatomischen, vergleichend-anatomischen, physi• ologischen, physikalischen, entwicklungsmechanis• chen, biologischen und sozialen Gesichtspunkten. Vol 1.  Gießen (Germany):  Deutsche Frauenhei• lkunde, E. Opitz; 1913.

43. Borell U, Fernström I. The movements in the mechanism of disengagement with special ref• erence to the attitude of the foetal head. Acta Obstet Gynecol Scand 1957;36:347-55.

44. Borell U, Fernström I. Shape and course of the birth canal; a radiographic study in the hu• man. Acta Obstet Gynecol Scand 1957;36:166-78.

45. Murray RM The axis-traction forceps: The mechanical principles, construction, and scope. Trans Edinb Obstet Soc 1890;16:58-89.

46. Jones J. Some causes of delay in labour; with special reference to the function of the cer• vical spine of the fetus. J Obstet Gynecol Br Empire 1906;10

47. Poutamo J, Partanen K, Vanninen R, Vainio P, Kirkinen P. MRI does not change fetal cardio• tocographic  parameters. Prenat Diagn 1998;18:1149-54.

48. Shakespeare SA, Moore RJ, Crowe JA, Gowland PA, Hayes-Gill BR. A method for foetal heart rate monitoring during magnetic resonance imaging using Doppler ultrasound. Physiol Meas 1999;20:363-8.

49. Vadeyar SH, Moore RJ, Strachan BK, et al. Effect of fetal magnetic resonance imaging on

fetal heart rate patterns. Am J Obstet Gynecol 2000;182:666-9. 

50. Cheng YW, Shaffer BL, Bryant AS, Caughey AB. Length of the first stage of labor and associ• ated perinatal outcomes in nulliparous women. Obstet Gynecol 2010;116:1127-35.

51. Cheng YW, Hopkins LM,  Laros RK Jr, Caughey AB. Duration of the second stage of labor in multiparous women: maternal and neo• natal outcomes. Am J Obstet Gynecol 2007; 196:585.e1-6.  

52. Oxorn H, Foote WR. Human labor and birth, 5th ed. Norwalk: Appleton-Century- Crofts; 1986.

53. Dudenhausen JW. Praktische geburtshilfe. Berlin, New York: De Gruyter; 2011.

54. Hanretty KP. Obstetrics illustrated. London: Churchill Livingston; 2003.

55. Westbrook C. Handbook of MRI technique. Oxford: Blackwell Science Ltd; 1999.

56. Westbrook C, Kaut C. MRI in practice. Ox•

ford: Blackwell Science Ltd; 1998.