Arch Gynecol Obstet
DOI 10.1007/s00404-016-4067-0
MATERNAL-FETAL MEDICINE
Risk factors for brachial plexus injury in a large cohort
with shoulder dystocia
Katherine A. Volpe1 • Jonathan M. Snowden2 • Yvonne W. Cheng3,4
Aaron B. Caughey2
Received: 13 February 2016 / Accepted: 1 March 2016
Ó Springer-Verlag Berlin Heidelberg 2016
Abstract
Objective To examine birthweight and other predictors of
brachial plexus injury (BPI) among births complicated by
shoulder dystocia.
Study design A retrospective cohort study of term births
complicated by shoulder dystocia in California between
1997 and 2006. Birthweight at time of delivery was strat-
ified into 500-g intervals. Women were further stratified by
diabetes status, parity, and race/ethnicity. The perinatal
outcome of BPI was assessed.
Results This study included 62,762 deliveries complicated
by shoulder dystocia, of which 3168 (5 %) resulted in BPI. The
association between birthweight and BPI remained significant
regardless of confounders. Each increasing birthweight interval
was associated with an increasing risk of BPI compared with
3000–3499-g birthweight. Race/ethnicity, diabetes, and parity
were also independently associated with BPI.
This research was presented as a poster presentation at the 32nd
Annual Meeting of the Society for Maternal Fetal Medicine, February
11, 2012.
& Katherine A. Volpe
kvolpe@salud.unm.edu
1
Department of Obstetrics and Gynecology, The University of
New Mexico School of Medicine, MSC 10 5580,
1 University of New Mexico, 87131-0001 Albuquerque,
NM, Mexico
2
Department of Obstetrics and Gynecology, Center for
Women’s Health, Oregon Health and Science University,
Portland, OR, USA
3
Department of Obstetrics and Gynecology, California Pacific
Medical Center, Davis, CA, USA
4
Department of Surgery, University of California, Davis,
CA, USA
•
Conclusion Increasing birthweight increases the risk of
BPI among births with shoulder dystocia, independent of
advanced maternal age, race, parity, gestational diabetes, or
operative vaginal delivery.
Keywords Birthweight
Brachial plexus injury

Shoulder dystocia
Introduction
Most cases of brachial plexus injury (BPI) resolve spon-
taneously within a year; however, the 5–8 % of injuries
that persist are a leading cause of birth related litigation [1,
2]. The overall incidence of BPI in the United States is
between 1 and 2 cases per 1000 deliveries [3]. Major risk
factors for BPI include shoulder dystocia, operative vaginal
delivery, macrosomia, gestational diabetes, and breech
presentation [1].
Historically, BPI was thought to result solely from
excessive downward traction by the birth attendant after
delivery of the fetal head and was, thus, associated with the
force required to resolve shoulder dystocia. However,
because of the incidence of BPI in births without shoulder
dystocia, it has been hypothesized that BPI may be caused
by two broad mechanisms of injury: (1) injury sustained in
utero and during descent and (2) injury sustained at the
time of expulsion [4]. The presence of BPI after cesarean
delivery supports this hypothesis [5–7].
In the setting of shoulder dystocia, the most common
complication of interest is BPI. In smaller studies of births
complicated by shoulder dystocia, the rate of BPI has been
reported between 3.8 and 19 % [8–10]. Macrosomia has
also been associated with both increased incidence and
severity of BPI [11, 12]. However, to the best of our
123
 Arch Gynecol Obstet

knowledge, no large studies have examined the association
between increasing birthweight categories and BPI in the
setting of shoulder dystocia (Pubmed MEDLINE search
keywords ‘‘birthweight,’’ ‘‘macrosomia,’’ ‘‘brachial plexus
injury,’’ ‘‘shoulder dystocia’’). Our study examines risk
factors for BPI in a population of births complicated by
shoulder dystocia.
Multivariable logistic regression was used to estimate
adjusted odds ratios (AORs) and respective 95 % confi-
dence intervals of BPI according to birthweight intervals,
while adjusting for gestational age, diabetes status, race/
ethnicity, parity, sex, maternal age, operative vaginal
delivery, and induction of labor.

Results
Materials and methods
During the 10-year period, there were 62,672 deliveries
We designed a retrospective cohort study of term singleton complicated by shoulder dystocia. Demographics and
births complicated by shoulder dystocia in California from characteristics of this population are shown in Table 1.
1997 to 2006. Birthweight at the time of delivery was Within this cohort of births complicated by shoulder dys-
stratified into 500-g intervals. Women were further strati- tocia, 3168 (5.0 %) resulted in BPI. The risk of BPI
fied by diabetes status, parity, and race/ethnicity. For the increased with each incremental increase in birthweight
stratification by diabetes status, women with either gesta- (Table 2, p \ 0.001). There was a significant difference in
tional diabetes or diabetes prior to pregnancy were grouped the prevalence of BPI among pregnancies complicated by
together and compared with women with no diabetes. For shoulder dystocia, between women with diabetes as com-
all groups, the perinatal outcome of BPI was assessed. pared with women without. Considering all birthweights,
The data source was the California Vital Statistics Birth 9.2 % of births in women with diabetes were further
Certificate Data linked with the California Patient Dis- complicated by BPI, while 4.7 % of births in women
charge Data as well as Vital Statistics Death Certificate without diabetes were complicated by BPI. Within each
Data and Vital Statistics Fetal Death File from 1997 to group, birthweight remained associated with the risk of BPI
2006. The California Office of Statewide Health Planning (p \ 0.001), and for each birthweight interval, women with
and Development (OSHPD) Healthcare Information diabetes had a significantly higher risk of BPI (Table 2;
Resource Center under the State of California Health Fig. 1).
Human Services (HHS) Agency performed the linkage of Births complicated by shoulder dystocia were more
data. Maternal antepartum and postpartum hospital records likely to be associated with BPI in nulliparous women.
for the 9 months prior to delivery and 1 year post delivery, Multiparous women had an average BPI incidence of
as well as, birth records and all infant admission and 4.9 %, while nulliparous women had an average of 5.4 %
readmissions occurring within the first year of life were (Table 2; Fig. 2). Again, within each group, birthweight
included in the resultant linked data sets. Linkage for the remained associated with the risk of BPI (p \ 0.001).
mother/baby pair was achieved using the ‘‘record linkage
number’’, a unique alphanumeric encrypted code unique to
the mother and the baby. Institutional Review Board (IRB) Table 1 Maternal, fetal, and delivery characteristics
approval was obtained from the Committee on Human N %
Research at the University of California, San Francisco, the
Maternal characteristics
IRB at Oregon Health & Science University, and the
Advanced maternal age 10,978 17.5
California OSHPD and the Committee for the Protection of
Nulliparous 20,130 32.1
Human Subjects (CPHS).
Diabetes 5419 8.7
Women with a diagnosis of shoulder dystocia were
White 26,425 42.2
identified using the International Statistical Classification
of Diseases and Related Health Problems, revision 9 (ICD- Hispanic 26,034 41.6
9) codes. ICD-9 codes used for the identification of women Asian 6158 9.8
with shoulder dystocia included 660.4. The ICD-9 code African–American 2976 4.8
767.6 was used for the identification of BPI. Exclusion Fetal characteristics
criteria were multiple gestations, births with congenital Male 34,317 54.8
anomalies, and small for gestational age neonates. Large for gestational age (C4500 g) 8045 12.8
Statistical calculations were performed with Stata (ver- Delivery characteristics
sion 12, StataCorp, College Station, TX). Dichotomous Operative vaginal delivery 10,498 16.8
outcomes were compared using the Chi-square test with This table shows maternal fetal and delivery characteristics of the
p \ 0.05 was used to indicate statistical significance. study cohort: women with births complicated by shoulder dystocia

123
Arch Gynecol Obstet

Table 2 Incidence of BPI in births complicated by shoulder dystocia

N BPI %

Birthweight (g)
2500–2999 662 14 2.1
3000–3499 7632 184 2.4
3500–3999 24,234 939 3.9
4000–4499 22,099 1214 5.5
4500–4999 6976 661 9.4
C5000 1069 156 14.6
Race/ethnicity
White 26,425 1132 4.3
Fig. 2 Incidence of BPI stratified by parity. This figure graphically
Hispanic 26,034 1469 5.6 compares the incidence of BPI in the setting of shoulder dystocia in
African–American 2676 251 8.4 nulliparous women with multiparous women for each birthweight
Asiana 6158 260 4.2 interval
Parity
Nulliparous 20,130 1083 5.4
Multiparous 42,491 2083 4.9
Diabetes status
No diabetes 57,246 2671 4.7
Diabetes in pregnancyb 5426 497 9.2
a
Considering all birthweights, there was not a significant difference
in the incidence of BPI between Asian and white women
b
Includes gestational diabetes and diabetes diagnosed prior to
pregnancy

Fig. 3 Incidence of BPI stratified by race/ethnicity. This figure graph-

ically compares the incidence of BPI in the setting of shoulder
dystocia in Black, Hispanic, Asian, and white women for each
birthweight interval

When investigated with multivariable regression analy-

Fig. 1 Incidence of BPI stratified by diabetes status. This fig-
ure graphically compares the incidence of BPI in the setting of
shoulder dystocia in women with no diabetes to women with either
diabetes preceding pregnancy or gestational diabetes for each
birthweight interval
sis, the association between birthweight and BPI in women
with shoulder dystocia remained significant regardless
of advanced maternal age (p = 0.046), nulliparity
(p \ 0.001), and operative delivery (p \ 0.001). When
controlling for potential confounders, increasing birth-
weight was associated with an increasing risk of BPI in
women with shoulder dystocia when compared with
3000–3499-g birthweight (Table 3).

We also found that births complicated by shoulder

dystocia were more likely to be associated with BPI in non-
white women (Table 2; Fig. 3). Hispanic women had an
average BPI incidence of 5.6 %, and African American
women had and incidence of 8.4 %, while white women
had an average of 4.3 %. Within each group, birthweight
remained associated with the risk of BPI (p \ 0.001).
Discussion
The association between birthweight and shoulder dystocia
is well described [9]. Our study advances current under-
standing by describing the association between increasing
birthweights starting in a normal range and neonatal injury

123
 Arch Gynecol Obstet

Table 3 Multivariable logistic regression analysis differences arise from access issues, nutritional issues
Risk factors AOR 95 % CI
or regional practice variations remains in need of
further exploration.
BW 3000–3499 g Reference While we have assembled a large cohort of women
BW 3500–3999 g 1.84 1.55–2.18 who experienced shoulder dystocia, our study is limited
BW 4000–4499 g 2.95 2.48–3.50 by the inherent features of a retrospective cohort design
BW 4500–4999 g 5.35 4.45–6.43 and our data set. These include selection bias, and reli-
BW C5000 g 9.36 7.32–11.98 ance on data collected for purposes other than study
Asian 1.02 0.88–1.18 outcomes. Specifically, since our data rely on ICD-9
Hispanic 1.35 1.24–1.47 codes, we were only able to determine the presence or
African–American 2.39 2.05–2.78 absence of an outcome and not the severity. For example,
Operative vaginal delivery 1.67 1.52–1.84 we were unable to use common indicators for the severity
Nulliparous 1.31 1.20–1.43 of the shoulder dystocia such as how many maneuvers
Gestational diabetes 1.77 1.58–2.00 were required for resolution. Furthermore, we stratified
IOL 0.98 0.86–1.04 births at risk for BPI by birthweight as measured at the
Male fetus 0.72 0.66–0.77 time of delivery. Unfortunately, accurate prediction of
AMA 1.11 1.00–1.22 birthweight prior to delivery is currently inaccessible.
This limits the application of our findings to current
IOL induction of labor
practice. Finally, we used multivariable logistic regression
to control for potential bias due to confounding. However,
among women who experienced shoulder dystocia. It is there may be residual confounding from uncontrolled
possible that this association exists, because even in the factors. For example, we did not have information on
setting of shoulder dystocia, greater force may have been maternal prepregnancy weight/body mass index (BMI),
needed to resolve the shoulder dystocia with a larger labor dystocia, or malposition, which could have
neonate. Possible increase in force might result in more increased the likelihood of BPI.
stretch and trauma to the brachial plexus. Alternatively, By describing the maternal and neonatal characteristics
larger neonates with longer labors may be more likely to associated with BPI in the setting of shoulder dystocia, we
develop a brachial plexus injury related to abnormal des- set the stage for comparison with BPI in births without
cent or prolonged labor. delivery complications. Because of the limitations associ-
Even when controlling for birthweight, we found an ated with using ICD-9 data and the retrospective nature of
association between diabetes in pregnancy and BPI. the study, we did not feel that we could draw meaningful
Although the association between diabetes in pregnancy conclusions in this regard from our data set. Two smaller
and BPI has been described [7, 11, 13–15], the contribution studies have looked at this issue. In a study of 39 cases of
of shoulder dystocia to the relationship is not fully known. BPI stratified by shoulder dystocia, characteristics of the
It may be that some of the observed effect is due to the two groups differed in terms of birthweight, maternal age,
difference in weight distribution in neonates of pregnancies and parity [20]. A recent study of 329 cases of BPI strat-
complicated by diabetes. This difference in weight distri- ified by shoulder dystocia, found that BPI in women with
bution may influence the force required to resolve the shoulder dystocia as compared with those without was
shoulder dystocia. Alternatively, research into whether more likely to be associated with obesity and macrosomia
neurologic injury that occurs in the setting of hyper- and less likely to be associated with operative vaginal
glycemia is less likely to resolve deserves in vitro labora- delivery, cord pH \ 7.10 and involvement of the posterior
tory and in vivo animal studies. shoulder [21]. Larger studies in the future may be able to
We additionally found an association between BPI better analyze risk factors for BPI in the absence of
and race/ethnicity. For each birthweight category, shoulder dystocia.
Hispanic and African–American women had a higher
Acknowledgments Dr. Caughey and Dr. Cheng have full control of
risk of BPI (Fig. 3). However, an increased risk of all primary data, they will allow review if requested.
shoulder dystocia in Hispanic and African American
women has been reported [16, 17]. This is consistent Compliance with ethical standards
with poorer outcomes in these populations across a
Conflict of interest Author Volpe declares that she has no conflict
range of obstetric outcomes, including success of trial of interest. Author Cheng declares that she has no conflict of interest.
of labor after cesarean, postpartum hemorrhage, and Author Snowden declares that he has no conflict of interest. Author
peripartum infection [18, 19]. Whether these Caughey declares that he has no conflict of interest.

123
Arch Gynecol Obstet
Ethical standards This article does not contain any studies with
human participants or animals performed by any of the authors.
References
1. Doumouchtsis SK, Arulkumaran S (2009) Are all brachial plexus
injuries caused by shoulder dystocia? Obstet Gynecol Surv
64(9):615–623
2. Jakobovits A (1996) Medico-legal aspects of brachial plexus
injury: the obstetrician’s point of view. Med Law 15(1):175–182
3. Foad SL, Mehlman CT, Ying J (2008) The epidemiology of
neonatal brachial plexus palsy in the US. J Bone Joint Surg Am
90(6):1258–1264
4. Jennett RJ, Tarby TJ (1997) Brachial plexus palsy: an old prob-
lem revisited again. II. Cases in point. Am J Obstet Gynecol
176:1354–1356 (discussion 1356–1357)
5. Graham EM, Forouzan I, Morgan MA (1997) A retrospective
analysis of Erb’s palsy cases and their relation to birth weight and
trauma at delivery. J Matern Fetal Med 6(1):1–5
6. Bar J, Dvir A, Hod M, Orvieto R, Merlob P, Neri A (2001)
Brachial plexus injury and obstetrical risk factors. Int J Gynaecol
Obstet 73(1):21–25
7. Ecker JL, Greenberg JA, Norwitz ER, Nadel AS, Repke JT
(1997) Birth weight as a predictor of brachial plexus injury.
Obstet Gynecol 89(5 Pt 1):643–647
8. Gregory KD, Henry OA, Ramicone E, Chan LS, Platt LD (1998)
Maternal and infant complications in high and normal weight
infants by method of delivery. Obstet Gynecol 92(4 Pt
1):507–513
9. Nesbitt TS, Gilbert WM, Herrchen B (1998) Shoulder dystocia
and associated risk factors with macrosomic infants born in
California. Am J Obstet Gynecol 179(2):476–480
10. Ouzounian JG, Korst LM, Miller DA, Lee RH (2013) Brachial
plexus palsy and shoulder dystocia: obstetric risk factors remain
elusive. Am J Perinatol 30(4):303–307. doi:10.1055/s-0032-
1324698 (Epub 2012 Aug 16)
11. Gilbert WM, Nesbitt TS, Danielsen B (1999) Associated factors
in 1611 cases of brachial plexus injury. Obstet Gynecol
93(4):536–540
12. Pondaag W, Allen RH, Malessy MJ (2011) Correlating birth-
weight with neurological severity of obstetric brachial plexus
lesions. BJOG 118(9):1098–1103
13. Mollberg M, Hagberg H, Bager B, Lilja H, Ladfors L (2005)
High birthweight and shoulder dystocia: the strongest risk factors
for obstetrical brachial plexus palsy in a Swedish population-
based study. Acta Obstet Gynecol Scand 84(7):654–659
14. McFarland LV, Raskin M, Daling JR, Benedetti TJ (1986) Erb/
Duchenne’s palsy: a consequence of fetal macrosomia and
method of delivery. Obstet Gynecol 68(6):784–788
15. Dawodu A, Sankaran-Kutty M, Rajan TV (1997) Risk factors and
prognosis for brachial plexus injury and clavicular fracture in
neonates: a prospective analysis from the United Arab Emirates.
Ann Trop Paediatr 17(3):195–200
16. Berggren EK, Boggess KA, Funk MJ, Stuebe AM (2012) Racial
disparities in perinatal outcomes among women with gestational
diabetes. J Womens Health (Larchmt) 21(5):521–527
17. Cheng YW, Norwitz ER, Caughey AB (2006) The relationship of
fetal position and ethnicity with shoulder dystocia and birth
injury. Am J Obstet Gynecol 195(3):856–862
18. Cahill AG, Stamilio DM, Odibo AO, Peipert J, Stevens E,
Macones GA (2008) Racial disparity in the success and compli-
cations of vaginal birth after cesarean delivery. Obstet Gynecol
111(3):654–658
19. Grobman WA, Bailit JL, Rice MM, Wapner RJ, Reddy UM,
Varner MW, Thorp JM Jr, Leveno KJ, Caritis SN, Iams JD, Tita
AT, Saade G, Rouse DJ, Blackwell SC, Tolosa JE, VanDorsten
JP, Eunice Kennedy Shriver National Institute of Child Health
and Human Development (NICHD) Maternal-Fetal Medicine
Units (MFMU) (2015) Racial and ethnic disparities in maternal
morbidity and obstetric care. Obstet Gynecol 125(6):1460–1467
20. Jennett RJ, Tarby TJ, Kreinick CJ (1992) Brachial plexus palsy:
an old problem revisited. Am J Obstet Gynecol 166(6 Pt
1):1673–1676
21. Gurewitsch ED, Johnson E, Hamzehzadeh S, Allen RH (2006)
Risk factors for brachial plexus injury with and without shoulder
dystocia. Am J Obstet Gynecol 194(2):486–492
123