American Journal of Obstetrics and Gynecology (2005) 192, 153e60
www.ajog.org
Comparing McRoberts’ and Rubin’s maneuvers for initial management of shoulder dystocia: An objective evaluation Edith D. Gurewitsch, MD,a,* Esther J. Kim, BS,b Jason H. Yang, BS,b Katherine E. Outland,b Mary K. McDonald, BS,b Robert H. Allen, PhDb Departments of Gynecology and Obstetricsa and Biomedical Engineeringb Johns Hopkins University, Baltimore, Md
KEY WORDS Shoulder dystocia Birth trauma Brachial plexus McRoberts’ maneuver Rubin’s maneuver Rotation Laboratory model Injury prevention
Objective: This study was undertaken to objectively compare delivery traction force, fetal neck rotation, and brachial plexus elongation after 3 different initial shoulder dystocia maneuvers: McRoberts’, anterior Rubin’s, and posterior Rubin’s. Study design: We developed a laboratory birthing simulator comprised of a maternal model with a 3-dimensional bony pelvis, an instrumented fetal model, a force-sensing glove, and a computerbased data acquisition system. A single operator performed 30 simulated shoulder dystocia deliveries using standard downward traction after 1 maneuver was performed. Ten deliveries simulated McRoberts’ maneuver with fetal shoulders in the anteroposterior diameter. Ten deliveries involved approximately 30-degree oblique rotation of the anterior shoulder with the spine oriented anteriorly (anterior Rubin’s maneuver). Ten deliveries involved approximately 30degree rotation of the posterior shoulder to the opposite oblique pelvic diameter, with the spine oriented posteriorly (posterior Rubin’s maneuver). Peak traction force, brachial plexus elongation, and neck rotation were compared between groups using analysis of variance, with P ! .05 considered significant. Results: Rubin’s maneuvers were found to require less traction force than McRoberts’: 16.2 G 2.1 lbs for McRoberts’ compared with 8.8 G 2.2 lbs and 6.5 G 1.8 lbs for posterior and anterior Rubin’s respectively (P ! .0001). Brachial plexus extension was significantly lower after anterior Rubin’s maneuver compared with McRoberts’ or posterior Rubin’s maneuvers. Conclusion: In a laboratory model of initial maneuvers for shoulder dystocia, anterior Rubin’s maneuver requires the least traction for delivery and produces the least amount of brachial plexus tension. Further study is needed to validate these results clinically. Ó 2005 Elsevier Inc. All rights reserved.
The contents of this publication are the sole responsibility of the authors and do not necessarily reflect the opinions of the NCIPC. Supported in part by the Centers for Disease Control and Prevention National Center for Injury Prevention and Control Grants for Traumatic Injury Biomechanics Research No. 1-R49-CE00433-01. Presented at the Twenty-Fourth Annual Meeting of the Society of Maternal Fetal Medicine, New Orleans, La, February 2-7, 2004. * Reprint requests: Edith Diament Gurewitsch, MD, Gynecology and Obstetrics, Johns Hopkins HospitalePhipps 217, 600 N Wolfe St, Baltimore, Md 21287. E-mail:
[email protected] 0002-9378/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.05.055
154
Gurewitsch et al
Figure 1 Schematic of anterior and posterior Rubin’s maneuvers for shoulder dystocia. (A) Shoulder dystocia occurring after restitution of the fetal head to left occiput transverse position. Fetal shoulders are aligned and obstructed in the anteroposterior diameter (10.5 cm here) of a gynecoid pelvis (maternal soft tissue not shown). (B) Same fetus after the anterior shoulder has been rotated counterclockwise about 30 degrees into the oblique diameter of the pelvis (anterior Rubin’s maneuver). (C) Alternative 30degree clockwise rotation of the posterior shoulder to the opposite oblique diameter of the pelvis (posterior Rubin’s maneuver). After either anterior or posterior rotation, the fetal shoulders fit into the 12.5 cm oblique dimension.
Shoulder dystocia is a complication of vaginal delivery that requires a maneuver or series of maneuvers beyond customary traction on the fetal head to relieve the impaction created by the fetal shoulders and the maternal pelvis. Although algorithms for maneuver sequences exist,1,2 these are not based on prospectively derived clinical evidence. Results of retrospective studies comparing shoulder dystocia maneuvers are conflicting: Findings range from no maneuver being superior to another in averting untoward outcome3,4 to fetal maneuvers (eg, Woods’ screw and posterior arm release) being associated with worse outcome than maternal maneuvers (eg, suprapubic pressure and McRoberts’)5 to fetal maneuvers being associated with better outcomes than maternal maneuvers.6,7 The only shoulder dystocia maneuver that has been objectively evaluated to date has been the McRoberts’
maneuver.8 A laboratory model was used to demonstrate that, compared with lithotomy position, reorientation of the pelvis by McRoberts’ maneuver can reduce traction force, clavicle fracture, and brachial plexus elongation.8 By rotating the pelvis, the pubic symphysis is raised approximately 1 cm.9,10 Considered from this geometric perspective, the reported effectiveness of McRoberts’ maneuver in resolving approximately 40% of shoulder dystocia cases11 is apparent, because a fetal bisacromial diameter of nearly average dimension or below12 should subsequently fit within the anteroposterior diameter of the pelvic inlet with about this frequency. However, the same laboratory experiments also showed that the benefit of McRoberts’ maneuver was limited to bisacromial widths up to 12 cm. As the fetal shoulder width increased (representing greater fetopelvic disproportion and more severe shoulder
Gurewitsch et al
155
Figure 2
Schematic of laboratory model.
dystocia), McRoberts’ maneuver failed to achieve atraumatic delivery.8 The limitation of McRoberts’ positioning in such a scenario is owed principally to the fetal shoulders still being maintained in the smaller anteroposterior diameter of the pelvis. Another way to reduce shoulder impaction is by rotating the fetal shoulders within the pelvic inlet about 30 degrees.13,14 This can be accomplished by applying pressure behind either the anterior or the posterior shoulder. It is known more commonly as Rubin’s maneuver. As shown in Figure 1, either anterior or posterior Rubin’s maneuver will achieve a 2-cm increase in available space in the gynecoid pelvis. Effectively restoring the more physiologic shoulder position, the deliberate diagonal reorientation of the fetal shoulders will accommodate a bisacromial diameter of up to the oblique diameter of the pelvis (12.5 cm, on average). Simultaneous adduction of the shoulders reduces the bisacromial diameter. In addition, rotation of the shoulders applies the same physical principle of ‘‘winding a screw caught in threads’’ described by Woods
in 1943,15 achieving forward progression of the shoulders. On the basis of the geometric and mechanical advantages previously described, we hypothesized that, after performance of the Rubin’s technique as an initial maneuver for shoulder dystocia, the force subsequently required to complete delivery and the resultant tension produced in the brachial plexus would be less than the corresponding force required and tension produced after performance of the McRoberts’ maneuver. However, objective comparison of shoulder dystocia maneuvers is difficult to accomplish clinically. Maneuver sequences become more variable and complex, particularly as shoulder dystocia severity increases. The inability to standardize fetal weight, fetal shoulder width, maternal pelvic dimensions, or soft tissue resistance between subjects also thwarts objective comparison of shoulder dystocia maneuvers in a clinical setting. Brachial plexus tension produced during delivery cannot be measured in vivo. We therefore chose to evaluate the effect of the McRoberts’ and Rubin’s maneuvers on clinician-applied
156
Figure 3 Representative traction time histories for 3 different initial maneuvers for shoulder dystocia: McRoberts’ maneuver, anterior Rubin’s and posterior Rubin’s. The head-to-body interval in these, which was typical of all the deliveries, was about 5 seconds.
force, fetal neck rotation, and brachial plexus extension by using a laboratory model of simulated shoulder dystocia deliveries.
Material and methods We designed and implemented a birthing simulator specifically for shoulder dystocia that was more mechanically and anatomically biofidelic than other commercially available simulators because of the following innovations: the system included a maternal model with an embedded 3-dimensional bony replica of the gynecoid pelvis (Health Edco, Waco, Tex), a birth canal, and an expulsing ‘‘uterus’’ connected to a variable pneumatic pump that can simulate the natural pattern of intermittent uterine contractions and maternal expulsive effort. Flexible legs were attached in such a manner that when hyperflexed, the pelvis rotated 16 degrees cephalad. Pelvic rotation could be measured with a built-in rotary potentiometer, allowing objective assessment of proper execution of McRoberts’ positioning. Pleather ‘‘skin’’ and other soft tissue components were made of polyurethane, carpet foam, and foam sealer. This not only allowed better simulation of maternal soft tissue than the less pliable hard plastic of most simulators, but the user also could gain better access to the fetal shoulders and arms within the birth canal for performance of fetal manipulation techniques, such as the Rubin’s maneuver, Woods’ screw, and posterior arm release. The instrumented fetal model mimicked the movement of the head and neck. A joystick device consisting of a rod fixed about an anchor running along 2 Ushaped tracks was attached to a wooden dowel representing the cervical vertebrae, which was surrounded by a wooden block representing the volume of the neck.
Gurewitsch et al
Figure 4 Representative neck rotation time histories for 3 different initial maneuvers for shoulder dystocia: McRoberts’ maneuver, anterior Rubin’s and posterior Rubin’s.
Two 4-cm wooden dowels affixed to the bottom of the wooden block represented larger-than-average clavicles, allowing for simulation of impaction of the fetal shoulders behind the pubic symphysis. The wooden construction mimicked the properties of neonatal bone16 and could break either spontaneously or purposely. To allow simulation of neck rotation, the top of the ‘‘neck’’ was attached to a wooden block that rotated about a dowel, which in turn was anchored to another wooden block representing the head. The dowel also moved up and down (axially) within the wooden block to simulate neck extension. A piece of string attached to an area on the neck corresponding to the C5 vertebra on each side (left and right) of the model was used to represent nerves of the upper tracts of the brachial plexus. Although the brachial plexus nerves exit the spinal cord from C5 to T1 vertebrae, physical constraints of the model allowed representation of only 1 nerve tract on each side. We chose to represent the upper portion of the brachial plexus because it is the locus where tension is incurred first during actual shoulder dystocia events. A second piece of string was also connected to the T1 vertebra, but this was used to stabilize the device. To measure neck rotation and brachial plexus extension, the model incorporated rotary and linear potentiometers, respectively. When neck rotation actuated the rotary potentiometer, it generated an output voltage proportional to the amount of torsion. Similarly, when lateral flexion induced brachial plexus elongation on 1 side, this actuated 1 of the 2 linear potentiometers and generated an output voltage proportional to the amount of tension. The rotary potentiometer used to measure neck rotation was linear in a range beyond G90 degrees, effectively covering the true range of motion for neck rotation, which is about G80 degrees. The anterior and posterior brachial plexus representations measured stretch accurately up to 2.3 cm, about the length of
Gurewitsch et al
Figure 5 (A) Representative brachial plexus tension in anterior shoulder for 3 different initial maneuvers for shoulder dystocia: McRoberts’ maneuver, anterior Rubin’s, and posterior Rubin’s. (B) Representative brachial plexus tension in posterior shoulder for 3 different initial maneuvers for shoulder dystocia: McRoberts’ maneuver, anterior Rubin’s, and posterior Rubin’s.
the unstretched brachial plexus. Because each of the potentiometers varied linearly between position and output, the slope of the interpolated equation provided a discrete method of calculating the rotation or displacement at each potentiometer, given some reference point. For each potentiometer, calibration tests were performed to obtain trendlines between output voltage and degrees of neck rotation and length of brachial plexus extension.17 Linearity, measured in terms of R2, ranged from 0.95 to 0.99. Once assembled, the model was housed in a commercial neonatal model (Childbirth Graphics, Waco, Tex). Cotton stuffing within the model was initially removed to allow room for fit, and then was reinserted in part once the instrumentation was properly positioned. Finally, the outer felt covering was sutured to enclose the model, except for wires from the potentiometers that were allowed to protrude as an ‘‘umbilical cord.’’
157
Figure 6 Photograph of rotational technique performed routinely at delivery in 1940s and 1950s. Shown is the operator’s right hand inserted anteriorly along the posterior aspect of the fetus’ right shoulder. Sliding the hand down the back of the neck to the posterior shoulder, the operator guides the posterior shoulder into the hollow of the sacrum. In a continuous sweep of the hand in a clockwise direction about 30 degrees, the posterior shoulder is rotated to an oblique position within the pelvis. The remainder of the delivery can thus proceed with minimal traction on the head and neck. (When the fetus is in right occiput-anterior position, the left hand is used and the rotation is counterclockwise.) (Reprinted with permission from Elsevier courtesy of Bager B.24)
A separate system for measuring clinician-applied forces used a piezo-resistive pressure sensor embedded in a nylon-Lycra glove worn by the clinician. This system was similar to those described previously.18-20 The fetal neck motion and brachial plexus tension parameters, as well as the clinician forces, were transmitted to a data acquisition system developed in LabView (National Instruments, Austin, Tex), which featured a graphic user interface that displayed realtime information recorded from multiple channels of input (2 for brachial plexus tension [in the anterior and posterior shoulders], 1 for neck rotation, and 1 for clinician traction). The interface also displayed the instantaneous, real-time voltage readings from each channel and allowed the user to scroll the time history of the measurements to find the peak values. The system also outputted and saved the data in a text file, making it available for mathematical and statistical analyses. A schematic of the maternal model, fetal model and
158 Table
Gurewitsch et al Summary of peak value averages of time history data for the 3 initial shoulder dystocia management schemes
Traction force (pounds) Fetal Neck Rotation (degrees) Brachial plexus stretch in anterior shoulder (mm) Brachial plexus stretch in posterior shoulder (mm)
McRoberts (n = 10)
Posterior Rubin (n = 10)
Anterior Rubin (n = 10)
P value
16.2 G 2.1 14.5 G 6.7
8.8 G 2.2 9.6 G 4.3
6.5 G 1.8 13.2 G 3.3
! .0001 .09
7.3 G 2.5
6.9 G 2.9
2.9 G 1.0
.0003
2.3 G 0.7
1.5 G 0.4
0.9 G 0.1
! .0001
Values expressed as mean G SD.
force-sensing glove is depicted in Figure 2, along with the associated data-acquisition system.
Experimental design With approval from the Institutional Review Board, a single operator (E.D.G.) performed 30 simulated shoulder dystocia deliveries by using downward traction applied to the head after 1 maneuver was performed. Ten deliveries simulated McRoberts’ maneuver with fetal shoulders in the anteroposterior diameter of the pelvis and fetal head in the LOT position externally. Ten deliveries involved rotating the anterior shoulder approximately 30 degrees counterclockwise to align with the oblique pelvic diameter with the spine oriented anteriorly and the head in LOA position externally (anterior Rubin’s maneuver). Ten deliveries involved rotating the posterior shoulder approximately 30 degrees clockwise to the opposite oblique diameter, with the spine oriented posteriorly and the head in LOP position externally (posterior Rubin’s maneuver). The modeled fetal shoulder width was specifically chosen so that after performance of a single maneuver, whether McRoberts’ maneuver or either Rubin’s maneuver, the fetal model would be delivered through the maternal pelvis once otherwise ‘‘customary’’ traction was subsequently applied to the fetal head. The operator applied only as much traction to the head as was needed to deliver the shoulders and the trunk. In this way, we designed our experiment to measure that subsequent traction force and assess its resultant effects on fetal neck rotation and brachial plexus elongation; our experiment was not designed to assess which maneuver was more likely to relieve shoulder dystocia. For each delivery, clinician traction, neck rotation and brachial plexus elongation were blindly recorded by the data acquisition system. Peak values of traction force, extension in the brachial plexus of the anterior and posterior shoulders, and fetal neck rotation were compared between the different groups of delivery maneuvers with analysis of variance. A P value !.05 was considered significant.
Results Figure 3 depicts typical force time histories for 1 McRoberts’ delivery and 2 Rubin’s maneuver deliveries (anterior and posterior). Peak clinician-applied traction approached 18 pounds for this simulated shoulder dystocia delivery in McRoberts’ position, about 8 pounds for posterior Rubin maneuver and about 5 pounds for the anterior Rubin maneuver. The head-tobody interval for each delivery was about 5 seconds, which was the norm for the entire data set. Figure 4 shows typical neck rotation for the 3 types of delivery. Positive values correspond to posteriorly directed rotation. Figure 5 shows (1) brachial plexus extension in the anterior shoulder and (2) brachial plexus elongation in the posterior shoulder during the head-to-body interval after the 3 different types of maneuvers. As can be seen, tension in the brachial plexus in the anterior shoulder is about 4 times higher than the brachial plexus tension in the posterior shoulder. Table I shows the mean peak values of the 10 trials of each category of delivery. Rubin’s maneuvers were found to require less peak traction force for delivery than McRoberts’ maneuver: 16.2 G 2.1 pounds for McRoberts’ compared with 8.8 G 2.2 pounds and 6.5 G 1.8 pounds for posterior and anterior Rubin’s respectively (P ! .0001). Brachial plexus tension was significantly lower after anterior Rubin’s maneuver compared with McRoberts’ or posterior Rubin’s maneuvers.
Comment To date, none of the standard shoulder dystocia maneuvers has been proven to be superior in averting mechanical birth trauma. McRoberts’ maneuver is probably favored because delivery is usually accomplished through the same manual technique used in routine deliveriesddownward traction applied to the fetal headdwithout requiring mastery of additional skill in manipulating the fetus within the birth canal. Indeed, McRoberts’ maneuver will successfully resolve many cases of shoulder dystocia without trauma to the infant.8,11 However, persistent use of this method during
Gurewitsch et al less common but more severe shoulder dystocia deliveries raises the risk for mechanical injury.6,8,20 By contrast, manipulation of the fetal trunk (as opposed to the head) to deliver the posterior shoulder or arm has been shown to resolve severe shoulder dystocia with a lower incidence of neurologic sequelae,4,6 especially when used as the sole management technique.6,7,21 When considered historically, rotational maneuvers for shoulder dystocia are not only efficacious, but appear to reduce neonatal morbidity. Fetal manipulation techniques were described and incorporated into clinical practice about 20 years before maternal-based maneuvers were advocated. Remarkably, during the 2 decades that followed implementation of fetal manipulation techniques, the incidence of permanent obstetric brachial palsy declined by 400%.22 In more recent years, the incidence of shoulder dystocia-associated obstetric brachial plexus palsy has not declined and may even be rising,23,24 despite widespread use of McRoberts’ positioning and more liberal use of cesarean delivery. Prudence dictates that traction should be minimized to whatever extent possible during any delivery, but especially during a shoulder dystocia. Here again, history provides valuable insight: Rotation of the fetal trunk after delivery of the head by application of pressure behind the posterior aspect of the posterior shoulder, as shown in Figure 6, was practiced routinely in vaginal deliveries during the 1940s and 1950s.25 The specific intent of this practice was to maximize the space for the fetal shoulders to pass through and thereby reduce traction on the fetal head. In the current experiment of laboratory-simulated shoulder dystocia deliveries, we have demonstrated that, compared with McRoberts’ maneuver, the simple 30degree rotation of the fetal shoulders to the oblique diameter of the pelvis described by Rubin reduces clinician-applied force subsequently applied to the fetal head and produces less brachial plexus tension. A somewhat unexpected finding was a difference between anterior and posterior Rubin’s maneuvers, with our results demonstrating least traction and brachial plexus elongation when the fetal spine ends up oriented anteriorly within the pelvis. While lateral traction that follows either McRoberts’ or Rubin’s maneuvers results in flexion of the neck, it is possible that the direction of pull after an anterior Rubin’s maneuver is modified by a component of forward neck flexion (owing to the relative occiput-anterior positioning of the head), which may be more favorable with regard to brachial plexus tension than a component of backward neck extension that would be added after a posterior Rubin’s maneuver (resulting from the relative posterior positioning of the fetal occiput). However, artifact of the laboratory model may also be a possible explanation for the apparent difference and substantiation of our finding through additional investigation is needed.
159 We recognize that restricting our analysis to 1 ‘‘delivering clinician’’ introduces a potential bias because the single operator was aware of the maneuvers performed each time and therefore could have adjusted the subsequently applied force on the basis of preconceived notions about the effects of each maneuver. Had we tested with multiple operators, this effect might have been reduced, but could have introduced other confounding issues, such as variation in delivery style. Instead of multiple operators, we used multiple trials and blinded measurement to reduce the risk of bias. Because the instruments of the model brachial plexus were enclosed within the fetal model and thus hidden from the operator, elongation produced by the operator’s actions also could not be seen as they were occurring. Post hoc, we note that the average ‘‘headto-body’’ interval for all deliveries was about 5 seconds, a time frame that would have made it difficult to consciously or unconsciously alter the extraction forces or twists applied to the fetal head. The differences in measured forces and neck motion parameters, though significantly different between interventions, were only a few pounds, millimeters, or degrees of rotationdand of magnitudes that would have been difficult to alter consciously or unconsciously. Furthermore, the unexpected result concerning anterior versus posterior Rubin’s maneuver supports the idea that bias, even if present, did not likely affect our results. The incidence of brachial plexus injury appears to be rising,24 a phenomenon that may require re-evaluation of present-day shoulder dystocia management. From historical and mechanical perspectives, implementing Rubin’s maneuver initially in shoulder dystocia management algorithms apparently has merit. The current laboratory model validates this concept. Additional objective evaluation, both clinical and experimental, is needed. While it is premature to offer specific clinical recommendations for shoulder dystocia management based on limited clinical observation6,7,21,22 and these experimental results, fetal manipulation is not new to obstetric management. We would encourage greater emphasis on its practice during training so that clinicians are more familiar with its use. Like McRoberts’ maneuver, a simple 30-degree manual rotation of the fetal shoulders can be practiced even during nonshoulder dystocia deliveries, in the same manner performed half a century ago. By doing so, interns and residents can be trained in Rubin’s maneuver well before it is truly needed for a severe shoulder dystocia.
References 1. Acker DB. Shoulder dystocia. In: Friedman EA, Acker DB, Sachs BP, editors. Obstetrical decision making. Toronto: BC Decker; 1987.
160 2. Dildy GA. Shoulder dystocia. In: Hankins GDV, Clark SL, Cunningham FG, Gilstrap LC, editors. Operative obstetrics. Norwalk (CT): Appleton & Lange; 1995. p. 233-56. 3. Nocon JJ, McKenzie DK, Thomas LJ, Hansell RS. Shoulder dystocia: an analysis of risks and obstetric maneuvers. Am J Obstet Gynecol 1993;168:1732-9. 4. Gherman RB, Ouzounian JG, Goodwin TM. Obstetric maneuvers for shoulder dystocia and associated fetal morbidity. Am J Obstet Gynecol 1998;178:1126-30. 5. McFarland MB, Langer O, Piper JM, Berkus MD. Perinatal outcome and the type and number of maneuvers in shoulder dystocia. Int J Gynecol Obstet 1996;55:219-24. 6. Baskett TF, Allen AC. Perinatal implications of shoulder dystocia. Obstet Gynecol 1995;86:14-7. 7. Gross SJ, Shime J, Farine D. Shoulder dystocia: predictors and outcome. Am J Obstet Gynecol 1987;156:334-6. 8. Gonik B, Allen R, Sorab J. Objective evaluation of the shoulder dystocia phenomenon: effect of maternal pelvic orientation on force reduction. Obstet Gynecol 1989;74:44-8. 9. Gonik B, Stringer C, Held B. An alternate maneuver for management of shoulder dystocia. Obstet Gynecol 1983;145:882-4. 10. Poggi SH, Spong CY, Allen RH. Prioritizing posterior arm delivery during severe shoulder dystocia. Obstet Gynecol 2003;101:1068-72. 11. Gherman RB, Goodwin M, Souter I, Neumann K, Ouzounian JG, Paul RH. The McRoberts’ maneuver for the alleviation of shoulder dystocia: how successful is it? Am J Obstet Gynecol 1997;176:656-61. 12. Verspyck E, Goffinet F, Hellot M, Milliez J, Marpeau L. Newborn shoulder width: a prospective study of 2222 consecutive measurements. BJOG 1999;106:589-93. 13. Rubin A. Management of shoulder dystocia. JAMA 1964;189:835-7.
Gurewitsch et al 14. Ramsey PS, Ramin KD, Field CS, Rayburn WF. Shoulder dystociadrotational maneuvers revisited. J Reprod Med 2000; 45:85-8. 15. Woods C. A principle of physics as applicable to shoulder dystocia. Am J Obstet Gynecol 1943;45:796-804. 16. Bankoski BR, Allen RH, Nagey DA, Miller F. Toward preventing newborn injury [abstract] Ann Biomed Eng 1994;22:66. 17. Sorab J, Allen RH, Gonik B. Modeling human birth to study the effects of clinician-applied forces. Rubinsky B. 29-30. 1988. 1988. 18. Sorab J, Allen RH, Gonik B. Tactile sensory monitoring of clinician-applied forces during delivery of newborns. IEEE Trans Biomed Eng 1988;35:1090-3. 19. Poggi S, Spong C, Patel C, Ghidini A, Pezzullo J, Allen R. Randomized trial of prophylactic McRoberts’ versus lithotomy to decrease force applied to the fetus during delivery. Am J Obstet Gynecol 2003;189:S191. 20. Allen R, Sorab J, Gonik B. Risk factors for shoulder dystocia: an engineering study of clinician-applied forces. Obstet Gynecol 1991;77:352-5. 21. Schwartz B, Dixon D. Shoulder dystocia. Obstet Gynecol 1958;11:468-71. 22. Adler J, Patterson RL. Erb’s palsy: long term results of treatment in eighty-eight cases. J Bone Jt Surg 1967;49-A:1052-74. 23. Wolf H, Hoeksma AF, Oei SL, Bleker OP. Obstetric brachial plexus injury: risk factors related to recovery. Eur J Obstet Gynecol Reprod Biol 2000;88:133-8. 24. Bager B. Perinatally acquired brachial plexus palsy: a persisting challenge. Acta Paediatr 1997;86:1214-9. 25. Reid D. Conduct of normal labor and the puerperium. In: Reid D, editor. Textbook of obstetrics. Philadelphia: WB Saunders; 1962. p. 448-89.