California College of Midwives
June 2001 Principles of Mother-Friendly Childbearing Services
The following information on Shoulder Dystocia is reprinted from the
Association of Radical Midwives (A.R.M.) at http://www.midwifery.org.uk
Additional Info on Shoulder Dystocia on the A.R.M. site is at http://www.radmid.demon.co.uk/
Shoulder dystocia and birth injuries, including brachial plexus injuries
Angela Horn www.homebirth.org.uk
The research I've read suggests that brachial plexus (B.P.) injury is the most common after-effect of shoulder dystocia, but that normally the injury is just transient. Most studies find that few injuries persist after a couple of months [4, 7, 17, 23]. The highest figures quoted are in the range of 15% of shoulder dystocia babies suffering some injury, of which 5% of the injuries are permanent (I think this means 5% of the 15%, but it's not clear) .
More examples: Discussion of  says: "Shoulder dystocia occurs in < 1% of vaginal births and 12% to 15% of cases are associated with some evidence of injury to the brachial plexus. Injury to the C5-6 nerve roots leads to Erb's-Duchenne palsy, whereas injury to the nerve root C8-T1 leads to Klumpke's palsy Fortunately, 80% to 90% of Erb's palsies and up to 50% of Klumpke's resolve without leaving any persistent neurologic deficits."
Although approx. half of all shoulder dystocia cases involve babies of normal weight, large babies with shoulder dystocia are more likely to suffer Erb's palsy (a specific B.p. injury) than small babies with shoulder dystocia. They appear not to be any more likely to suffer other serious consequences though, eg oxygen deprivation. 
The causes of B.P. injury are controversial, but all the evidence is that they are caused mainly, not by shoulder dystocia itself, but by some delivery methods attempted to resolve shoulder dystocia. The big no-no is pulling on the head, which stretches the neck and nerves in it. However, this can't account for all cases as some babies born by elective c/s have B.P. injuries. See  for a discussion.
Elective c/s to avoid birth injuries in large babies?
Elective c/s to avoid the possibility of shoulder dystocia is NOT recommended for large babies on the basis of several recent studies, eg:
For non-diabetic women, for each permanent brachial plexus injury prevented by the policy of elective c/s for babies estimated to weigh over 4,000g (8lb 13 oz), 3695 cesarean deliveries were performed at an additional cost of $8.7 million...For the policy of elective c/s for babies estimated at over 4500g (9lbs 15 oz), it takes 2345 Cesarean deliveries and $4.9 million to prevent one permanent brachial plexus injury. 
For non-diabetic women, another study found that : "Under the most plausible assumptions, a prophylactic
cesarean policy with either a 4000- or 4500-g macrosomia threshold would require more than 1000 cesarean
deliveries and millions of dollars to avert a single permanent brachial plexus injury." 
Doctors with higher c/s rates for babies over 4000g do not achieve better outcomes than those with lower rates  - but various outcomes for mothers were worse with 'liberal' use of c/s for large babies.
A retrospective study  of 227 babies weighing over 4,500g (9lb 15oz), average 4,706g (10lb 6oz) found that : "Shoulder dystocia occurred 29 times, for an incidence of 18.5% in vaginal deliveries for macrosomia. There were seven cases each of Erb palsy and clavicular fracture, and one humeral fracture. By 2 months of age, all affected infants were without permanent sequelae. There was no birth asphyxia or perinatal mortality related to delivery for macrosomia. ..There was no statistically significant difference with respect to hemorrhage or hospital stay for women who had a vaginal delivery (with or without shoulder dystocia) compared with women who had a cesarean delivery."
A dissenting study is , which suggests (on the basis of 77 cases of shoulder dystocia) that "A policy for
elective caesarean section for birthweights in excess of 4000 g (97 percentile) would prevent 44% of shoulder dystocias, increase the caesarean section rate by 2% and half the perinatal mortality among births with shoulder dystocia.".
However, note that this study is from Singapore where, as the authors say,
"The local birthweight distribution is very different from the West..." It says that 4000g (8lb 13 oz) is the 97th percentile there, but in the UK I think the 97th percentile is about 4,500g. A brief look at a growth chart suggests that 4000g is only 75th percentile here.
Interesting ... but perhaps even more interesting is the fact that two obstetricians in the USA  are taking these figures and applying them to shoulder dystocia generally, saying without qualification that elective c/s for estimated weights over 4000g would "would prevent 44% of shoulder dystocias......" -
This is a classic case of abuse of data. There is no reason to suppose that a policy of elective c/s for babies above the 4000g would achieve the same results in other populations, and that could mean elective c/s for 25% of babies in some countries! For these conclusions to be applicable to the USA or UK, you would need to have comparable c/s rates generally, comparable obstetric practice, and the average sizes of mothers and babies would have to be similar... none of these applies as far as I know.
I found a couple of other papers suggesting (few actually recommend) elective c/s for babies over a certain weight. I noticed a trend in the research; those studies which involve retrospective case controls reach different conclusions than those which look at births for large groups. The retrospective case control studies usually take, say, 100 cases of shoulder dystocia, or of births of large babies, and look at complications which occurred, and consider whether they would have been prevented by elective c/s. They do not generally look at any other outcomes (eg maternal morbidity or mortality) - just what cut-off policy for c/s would have prevented the shoulder dystocia. If *all* you cared about was shoulder dystocia then all babies would be born by c/s...
The problem with this is that they reach a conclusion for a known birthweight - so they might recommend that all babies weighing 4,500g or over should be born by c/s to prevent 20% of shoulder dystocia cases, *but* that doesn't help much with decisions before the baby is born, because of the inaccuracy of weigh estimation. So retrospectively you might be able to say that having had such a weight as the cut-off point for c/s would have achieved these results, eg say that 10 c/s would have been necessary to prevent one case of s.d.... but if you then start to do elective c/s for an *estimated* weight of that level, you might find that 20 c/s were necessary to achieve the same results... or 5. The point is that the 'tool' of ultrasound weight estimation is so flawed, it is hard to extrapolate guidelines from retrospective analysis.
Anyway, I digress. The studies which start out looking at large numbers of births of all weights, and then consider what would happen if you imposed a cut-off point for elective c/s for predicted large babies, tend to find that elective c/s is *not* recommended, eg  and others cited above.
I suppose there is an element of human psychology reflected in this too. If you start out looking at the cases which went wrong, and think about what you might have done differently with the benefit of hindsight, you may reach very different conclusions than if you start out thinking about what the best decision would be overall. Which seems to me to be the basic difference between retrospective case-control studies and the others.
Induction of labour at term for suspected large babies?
Again, the evidence suggests that induction at term is NOT recommended to avoid birth trauma for suspected large babies. It increases the c/s rate and associated problems for mother and baby, but does *not* reduce the shoulder dystocia rate, nor does it improve outcomes. See [13, 14, 24, 25, 26].
I'm afraid reading all this is just confirming all my prejudices...
Inaccuracy of fetal weight estimation
The problems of intervention such as induction or c/s based on estimated fetal weights are of course compounded by the notorious inaccuracy of ultrasound weight estimation. For example,  found that ""Ultrasonography and labor induction for patients at risk for fetal macrosomia should be discouraged". Predictions of macrosomia > 4,200g increased induction rates and c/s rates, but "There was no significant difference in the incidence of shoulder dystocia or the occurrence of birth trauma."
Ultrasound may overestimate fetal weight in general, and is pretty hit-and-miss at the best of times - eg  found that "In 66 of 86 women (77%) delivering within 3 days of ultrasound examination, estimated fetal weight (EFW) exceeded birth weight. In only 41 of these 86 women (48%) were the EFWs within the corresponding 500-g category of birth weight" Bear in mind that 500g is over 1lb!! That's a big margin for error...The 6th Annual CESDI report looked at deaths of large babies (4000g and over), and it concluded that ultrasound estimation of fetal weight was *NOT* recommended where a large baby was suspected, because "the inaccuracy of ultrasound estimates have been well documented. Indeed, it is possible that estimating fetal weight by late ultrasound may do more harm than good by increasing intervention rates" (p47). They also quote research concluding that elective induction and elective c/s are not recommended, although suggest that large randomised controlled trials are needed. Their main recommendation is that, where a large baby is suspected, the attendants should be on the alert for a delay in late labour, which could be a warning sign for shoulder dystocia.
What about death or brain damage from shoulder dystocia?
Note that the above study is talking about permanent brachial plexus injuries, not the things that worry us more, like death... but death or brain damage from shoulder dystocia is rarer than nerve damage. An article online  is a review of birth injuries arising from shoulder dystocia, and the author says that "Recent investigations have noted no clear link between shoulder dystocia and hypoxic brain damage". If you read this article, by the way, note that some of the stats quoted are confusing as the study actually looked at 107 'potential or actual' malpractice litigations where shoulder dystocia was involved. This means that presumably only the more serious injuries were considered, and of course the more serious the injury, the more likely it is to be the subject of a malpractice suit.
So when the researchers say that "Permanent hypoxic or traumatic cerebral damage was documented in almost one-third of the cases. ", they mean that one third of their malpractice cases involved brain damage, not that one third of shoulder dystocia cases involve brain damage. This study also says that "Five cases of cerebral palsy followed intentional cutting or incidental breaking of the umbilical cord, wrapped around the neck once, prior to delivery of the shoulders. " - interesting as elsewhere I found warnings against cutting or clamping nuchal cords where there is any possibility of shoulder dystocia.
The 5th Annual CESDI (Confidential Enquiry into Stillbirths and Deaths in Infancy, published by UK government) report on shoulder dystocia looked at 56 deaths in the two years (1994/5) where some degree of shoulder dystocia had been documented. They note that there is a real problem with definitions, and it appears that in many of these cases, shoulder dystocia may not have been the main problem. It says that in 47% of deaths due to shoulder dystocia, the interval between birth of the head and body was 5 minutes or less, and that "one would expect a previously healthy fetus to survive a period of cerebral hypoxia of that duration" - and that other factors were often involved: perhaps "a preceding hypoxic stress renders a fetus less able to withstand the additional asphyxial impact of shoulder impaction". It seems at least possible that some of these babies would have died even if there had not been any difficulty delivering the shoulders, and as noted before, we do not know what degree of shoulder dystocia any of them exhibited. So the figure of 56 deaths over two years does not mean that shoulder dystocia was responsible for all of these deaths.
Article  is a detailed summary of issues in shoulder dystocia cases, although perhaps outdated and conservative. The authors say: "A normal term fetus can endure up to 10 minutes of asphyxia before permanent neurologic injury occurs. During the first 5 minutes, the major harm to the neonate is most likely to be iatrogenic....Although the fetal pH declines at a rate of 0.04 units/min between delivery of the head and trunk, shoulder dystocia rarely results in death or asphyxial injury... "
Intervention causing shoulder dystocia?
I understand from a previous post that the rate of B.p. injuries was significantly higher in the USA than in the UK. I would not find this surprising, given the high level of intervention which seems normal in births there - the studies I've looked at found that shoulder dystocia was more likely where labours were induced, accelerated with oxytocin, and particularly where forceps or, especially, ventouse deliveries are attempted [6, 19].
Perhaps this is because
a) These interventions were employed because progress was slow, but progress was slow for a reason? Baby needed more time to turn, mother needed more time to move?
b) If you're having oxytocin you're probably being continuously monitored, and thus are probably lying down/semi-recumbent...great position for trapping a baby. Discussions on web articles [2,3] listed below give more info about specific increases in pelvic diameters when mother moves to all-fours.
c) Other factors about assisted deliveries - eg the speed with which the head is brought down, or the angle - which makes shoulder dystocia more likely. Any thoughts, anyone? eg Article  says one of the situations where s.d. is more likely is "when truncal rotation does not occur (as with precipitous labor)." - could this also apply to assisted deliveries? There is also mention on the USA Midwife Archives  about the need for the shoulders to rotate properly to fit through the pelvis, and that with assisted deliveries the head may be pulled down before this has happened. Apparently shoulder dystocia linked to assisted delivery is more common in smaller babies; in larger babies, it's more likely to happen even with spontaneous birth of the head.
Midwife management of shoulder dystocia
Someone asked about midwife management to avoid brachial plexus injuries. There are a couple of studies
showing very good outcomes for midwife management of shoulder dystocia. See for example 
The Farm article on Shoulder Dystocia and use of the all-fours position http://www.thefarm.org/lifestyle/dystocia.html
"From 1971 to the present, the midwives have attended 1750 births. Thirty-five of these were complicated by shoulder dystocia, and all of them were managed by midwives .. Three early births were managed with traditional maneuvers, resulting in some birth injuries. The remaining 32 were managed by having the mother assume the all-fours position, with no mortality, no birth injuries, and with excellent Apgar scores. All the babies for whom follow-up was possible (29 of 35) were developmentally normal (ages 9 months to 15 years). These statistics compare favorably with the reported mortality rates of 21% to 29% and morbidity rates of 16% to 48%. In addition, despite frequent recommendations that any maneuvers to deliver the shoulders be preceded by a generous episiotomy or proctoepisiotomy, 23 of the babies were delivered over an intact perineum, and there were no 3rd or 4th-degree lacerations. Finally, though some authors recommend the time-consuming step of administering general anesthesia to the mother before attempting alternative maneuvers, 23 these babies were all delivered without anesthesia."
This is a study on The Farm's results for shoulder dystocia, co-authored by Ina Mae Gaskin and a doctor who is trying to 'educate the masses' amongst obstetricians:
All-fours maneuver for reducing shoulder dystocia during labor. Bruner JP, Drummond SB, Meenan AL, Gaskin IM http://www.ncbi.nlm.nih.gov/htbin-post/Entrez/query?uid=9610468&form=6&db=m& Dopt=b
Study  found comparable outcomes for nurse-midwife practices compared to general medical practice in the USA, but a trend towards fewer incidences of s.d. was noticed when mothers were side-lying (as opposed to the 'obvious' recumbent/lithotomy position I suppose)
Shoulder Dystocia and home birth
Something I have learned from all this is that shoulder dystocia is far less likely to happen at a home birth than a hospital birth - or at least, if it does happen at home, it's likely to be in a known high-risk situation.
Shoulder dystocia figures for large babies are complicated by the inclusion of children of diabetic mothers, and obese mothers. The body shape of babies born to diabetic mothers appears different, making shoulder dystocia more likely for any given circumstance. For non-diabetic mothers, rates of s.d. for large babies are much lower,and for non-diabetic women of normal weight, lower still. [20, 16]
So, the reasons why I believe shoulder dystocia to be a lesser risk at home births are:
Most cases of serious s.d. involve women with gestational diabetes, and to a lesser extent women of great size. If a mother doesn't fall into either of these categories then you can cut your worrying by about 90%. Home births will not involve oxytocin augmentation - a known risk factor for s.d.... if the mother needs oxytocin then she will be transferring to hospital anyway. Ditto assisted deliveries. A mother at home is less likely to be lying on her back in the first place, so baby has more room to get out.
REFERENCES  Shoulder Dystocia LCDR Robert B. Gherman, MC USNR; T. Murphy Goodwin, MD
 USA Midwife ARchives page on Shoulder Dystocia - lots of information and discussion.
 The Farm article on Shoulder Dystocia and use of the all-fours position
 Fetal macrosomia: risk factors and outcome. A study of the outcome concerning 100 cases >4500 g. (9lbs 15 oz) AUTHORS: Berard J; Dufour P; Vinatier D; Subtil D; Vanderstichele S; Monnier JC; Puech F SOURCE: Eur J Obstet Gynecol Reprod Biol 1998 Mar;77(1):51-9..Shoulder dystocia occurred fourteen times (22% of vaginal deliveries) ...There were five cases of Erb's palsy, one of which was associated with humeral fracture, and four cases of clavicular fracture. By three months of age, all affected infants were without sequelae. There was no related perinatal mortality and only two cases of birth asphyxia..
 The effectiveness and costs of elective cesarean delivery for fetal macrosomia diagnosed by ultrasound.
AUTHORS: Rouse DJ; Owen J; Goldenberg RL; Cliver SP SOURCE: JAMA 1996 Nov 13;276(18):1480-6 )
 This study looked at 100 cases of babies predicted over 4,500g, whose average birth weight turned out to be 4,730g - about 10lb 7 oz...
TITLE: Shoulder dystocia. A complication of fetal macrosomia and prolonged second stage of labor with midpelvic delivery. AUTHORS: Benedetti TJ; Gabbe SG SOURCE: Obstet Gynecol 1978 Nov;52(5):526-9 In the absence of prolonged second stage (PSS) and midpelvic delivery, the incidence of shoulder dystocia was 0.16%. However, with PSS and midpelvic delivery (ie assisted delivery), the incidence of shoulder dystocia was 4.57% ..When PSS occurred and midpelvic delivery was attempted, the incidence of shoulder dystocia was 21% in infants weighing in excess of 4 kg.. All shoulder dystocias and failed vaginal deliveries occurred after use of the vacuum extractor. Immediate neonatal injury was apparent in 47% of infants with shoulder dystocia after PSS with midpelvic delivery. There were no maternal or fetal deaths related to shoulder dystocia during the study period.
 TITLE: Shoulder dystocia: its incidence and associated risk factors. AUTHORS: Sandmire HF; O'Halloin TJ SOURCE: Int J Gynaecol Obstet 1988 Feb;26(1):65-73 ABSTRACT: ..Among the 73 shoulder dystocia cases there were no perinatal deaths and all birth-related injuries associated with shoulder dystocia were temporary except for two cases of mild muscular weakness among 12 brachial palsy cases.
Common Intrapartum Denominators of Shoulder Dystocia Related Birth Injuries Leslie Iffy, Valeria Varadi, A. Jakobovits Department of Obstetrics and Gynecology and Pediatrics UMDNJ -New Jersey Medical School, Newark, New Jersey, USA and Universitats-Frauenklinik Aachen, Germany
 Macrosomia--maternal, fetal, and neonatal implications. AUTHORS: Modanlou HD; Dorchester WL; Thorosian A; Freeman RK SOURCE: Obstet Gynecol 1980 Apr;55(4):420-4
 http://www.shoulderdystocia.com/causes.html Brachial Plexus Causation: An Old Problem Revisited by
James A. O'Leary & James L. O'Leary, II
 An analysis of risk factors for the prediction of shoulder dystocia in 16,471 consecutive births. AUTHORS: Yeo GS; Lim YW; Yeong CT; Tan TC AUTHOR AFFILIATION: Department of Maternal Fetal Medicine, Kandang Kerbau Hospital, Singapore. SOURCE: Ann Acad Med Singapore 1995 Nov;24(6):836-40
 Maternal and infant complications in high and normal weight infants by method of delivery. AUTHORS:
Gregory KD; Henry OA; Ramicone E; Chan LS; Platt LD AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, Cedars Sinai Medical Center, University of California at Los Angeles, School of Medicine, 90048, USA. SOURCE: Obstet Gynecol 1998 Oct;92(4 Pt 1):507-13
 TITLE: Induction of labor versus expectant management in macrosomia: a randomized study. AUTHORS: Gonen O; Rosen DJ; Dolfin Z; Tepper R; Markov S; Fejgin MD AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, Meir General Hospital, Kfar-Saba, Israel. SOURCE: Obstet Gynecol 1997 Jun;89(6):913-7
 Labor induction with a prenatal diagnosis of fetal macrosomia. AUTHORS: Leaphart WL; Meyer MC; Capeless EL AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, University of Vermont, Burlington, USA. SOURCE: J Matern Fetal Med 1997 Mar-Apr;6(2):99-102
 Prophylactic cesarean delivery for fetal macrosomia diagnosed by means of ultrasonography--A Faustian bargain? AUTHORS: Rouse DJ; Owen J AUTHOR AFFILIATION: Division of Maternal-Fatal Medicine, Department of Obstetrics and Gynecology, University of Alabama at Birmingham, Alabama, USA. SOURCE: Am J Obstet Gynecol 1999 Aug;181(2):332-8
 Shoulder Dystocia LCDR Robert B. Gherman, MC USNR; T. Murphy Goodwin, MD
http://www.shoulderdystocia.com/reviews4.html A thorough overview of the subject although quite conservative (no mention of all fours, assumes women on back) and possibly outdated.
Dr. Dwight P. Cruikshank, Milwaukee, Wisconsin. Shoulder dystocia occurs in < 1% of vaginal births and 12% to 15% of cases are associated with some evidence of injury to the brachial plexus. Injury to the C5-6 nerve roots leads to Erb's-Duchenne palsy, whereas injury to the nerve root C8-T1 leads to Klumpke's palsy Fortunately, 80% to 90% of Erb's palsies and up to 50% of Klumpke's resolve without leaving any persistent neurologic deficits.
 Outcomes of macrosomic infants in a nurse-midwifery service. AUTHORS: Nixon SA; Avery MD; Savik K AUTHOR AFFILIATION: University of Minnesota School of Nursing, Minneapolis 55455, USA. SOURCE: J Nurse Midwifery 1998 Jul-Aug;43(4):280-6 ABSTRACT: ...Large infants had birth outcomes comparable to those reported by others in the medical literature, suggesting that nurse-midwifery management, including consultation with physician colleagues, can be appropriate and safe.
 Shoulder dystocia and associated risk factors with macrosomic infants born in California AUTHORS: Nesbitt TS; Gilbert WM; Herrchen B AUTHOR AFFILIATION: Center for Health Services Research in Primary Care, Department of Obstetrics and Gynecology, University of California, Davis, USA. SOURCE: Am J Obstet Gynecol 1998 Aug;179(2):476-80 CITATION IDS: PMID: 9731856 UI: 98400582
 Anthropometric differences in macrosomic infants of diabetic and nondiabetic mothers. AUTHORS: McFarland MB; Trylovich CG; Langer O AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, University of Texas Health Science Center, San Antonio 78284-7836, USA. SOURCE: J Matern Fetal Med 1998 Nov-Dec;7(6):292-5
 The Green Bay cesarean section study. IV. The physician factor as a determinant of cesarean birth rates for the large fetus. AUTHORS: Sandmire HF; DeMott RK
 Fetal macrosomia: does antenatal prediction affect delivery route and birth outcome? AUTHORS: Weeks JW; Pitman T; Spinnato JA 2nd AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, University of Louisville, School of Medicine, KY 40292, USA. SOURCE: Am J Obstet Gynecol 1995 Oct;173(4):1215-9
 The outcome of macrosomic infants weighing at least 4500 grams: Los Angeles County + University of
Southern California experience. AUTHORS: Lipscomb KR; Gregory K; Shaw K AUTHOR AFFILIATION: Division of Maternal Fetal Medicine, Los Angeles County + University of Southern California. SOURCE: Obstet Gynecol 1995 Apr;85(4):558-64
 Influence of spontaneous or induced labor on delivering the macrosomic fetus. AUTHORS: Friesen CD; Miller AM; Rayburn WF AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, University of Nebraska, College of Medicine, Omaha, USA. SOURCE: Am J Perinatol 1995 Jan;12(1):63-6
 TITLE: Elective induction versus spontaneous labor after sonographic diagnosis of fetal macrosomia.
AUTHORS: Combs CA; Singh NB; Khoury JC AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, University of Cincinnati College of Medicine, Ohio. SOURCE: Obstet Gynecol 1993 Apr;81(4):492-6
 Pregnancy outcome following ultrasound diagnosis of macrosomia. AUTHORS: Delpapa EH; Mueller-Heubach E AUTHOR AFFILIATION: Department of Obstetrics and Gynecology, University of Pittsburgh, Magee-Women's Hospital, Pennsylvania. SOURCE: Obstet Gynecol 1991 Sep;78(3 Pt 1):340-3
 Shoulder dystocia: predictors and outcome. A five-year review. AUTHORS: Gross SJ; Shime J; Farine D SOURCE: Am J Obstet Gynecol 1987 Feb;156(2):334-6
 J Reprod Med 1998 May;43(5):439-43 All-fours maneuver for reducing shoulder dystocia during labor. Bruner JP, Drummond SB, Meenan AL, Gaskin IM