Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

6
Vaginal Birth After Cesarean for Cephalopelvic Disproportion Effect of Birth Weight Difference on Success Lorie M. Harper, MD, David M. Stamilio, MD, MSCE, Anthony O. Odibo, MD, MSCE, Jeffrey F. Peipert, MD, PhD, and George A. Macones, MD, MSCE OBJECTIVE: To estimate the effect of birth weight differ- ence between the current and index pregnancy on vaginal birth after cesarean (VBAC) failure in patients whose prior cesarean was for cephalopelvic disproportion (CPD). METHODS: This was a retrospective cohort study of women with one cesarean for CPD, comparing the rate of VBAC failure in women whose infant was smaller, the same, or larger in the VBAC attempt compared with cesarean. The primary outcome was VBAC attempt failure, defined as a patient who attempted VBAC but subsequently required a cesarean delivery for any indication. Univariable, stratified, and multivariable analyses were used. RESULTS: Of 13,706 patients attempting VBAC, 1,511 had one prior cesarean delivery for CPD. Compared with patients with the same birth weight, a lower birth weight was associated with fewer failed VBAC attempts (29.6% compared with 37.8%, adjusted odds ratio OR 0.7, 95% confidence interval CI 0.5–1.0) and a higher birth weight was associated with more failed VBAC attempts (54.5% compared with 37.8%, adjusted OR 2.0, 95% CI 1.5–2.8). CONCLUSION: Birth weight difference has a moderate effect on the rate of VBAC success in patients whose prior cesarean delivery was for CPD. (Obstet Gynecol 2011;117:343–8) DOI: 10.1097/AOG.0b013e31820776fd LEVEL OF EVIDENCE: II T he indication for prior cesarean delivery has an effect on the success of a subsequent attempt at vaginal birth. Although a cesarean delivery for non- recurring indications, such as malpresentation, is as- sociated with relatively high success rates (80%) in vaginal birth after cesarean (VBAC) attempts, success rates fall when the prior cesarean delivery was for a recurring indication, such as cephalopelvic dispropor- tion (CPD) or failed induction. 1 In these cases, pa- tients may rightfully ask what is the likelihood of a successful VBAC if the current baby is bigger or smaller than the baby for whom the cesarean delivery was performed; they may also request that repeat ultrasound examinations be performed to determine the difference in fetal weight between this pregnancy and the previous one. The majority of previous publications in this area have focused on the absolute birth weight of the infant in the VBAC attempt and the probability of a successful VBAC. 2–4 However, limited clinical research is available to inform the patient or clinician on the effect of fetal size in this pregnancy compared with the previous birth in the decision of delivery mode. In this study, we attempt to estimate the effect of birth weight difference between the current and index pregnancy on VBAC failure in patients whose prior cesarean delivery was for cephalopelvic disproportion. METHODS This was a secondary analysis of a retrospective, multicenter cohort study of women with a prior cesarean delivery. 5 Seventeen tertiary and commu- nity hospitals, all with institutional review board approval, participated in a study designed to estimate the rate of and risk factors for uterine rupture in VBAC attempts. The study was conducted from 1995 to 2000. Methods of the study have been published in detail previously, but a brief description follows. 5 From the Department of Obstetrics and Gynecology, Washington University in St. Louis, St. Louis, Missouri. Supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (RO1 HD 35631 to G.A.M.; T32HD to L.M.H. [PI: Macones]) and UL1RR024992 (PI: Evanoff). Corresponding author: Lorie M. Harper, MD, Department of Obstetrics and Gynecology, Washington University in St. Louis, 4911 Barnes Jewish Hospital Plaza, Campus Box 8064, St. Louis, MO 63110; e-mail: [email protected]. Financial Disclosure The authors did not report any potential conflicts of interest. © 2011 by The American College of Obstetricians and Gynecologists. Published by Lippincott Williams & Wilkins. ISSN: 0029-7844/11 VOL. 117, NO. 2, PART 1, FEBRUARY 2011 OBSTETRICS & GYNECOLOGY 343

Transcript of Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

Page 1: Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

Vaginal Birth After Cesarean forCephalopelvic DisproportionEffect of Birth Weight Difference on Success

Lorie M. Harper, MD, David M. Stamilio, MD, MSCE, Anthony O. Odibo, MD, MSCE,Jeffrey F. Peipert, MD, PhD, and George A. Macones, MD, MSCE

OBJECTIVE: To estimate the effect of birth weight differ-ence between the current and index pregnancy on vaginalbirth after cesarean (VBAC) failure in patients whose priorcesarean was for cephalopelvic disproportion (CPD).

METHODS: This was a retrospective cohort study ofwomen with one cesarean for CPD, comparing the rate ofVBAC failure in women whose infant was smaller, the same,or larger in the VBAC attempt compared with cesarean. Theprimary outcome was VBAC attempt failure, defined as apatient who attempted VBAC but subsequently required acesarean delivery for any indication. Univariable, stratified,and multivariable analyses were used.

RESULTS: Of 13,706 patients attempting VBAC, 1,511 hadone prior cesarean delivery for CPD. Compared withpatients with the same birth weight, a lower birth weightwas associated with fewer failed VBAC attempts (29.6%compared with 37.8%, adjusted odds ratio �OR� 0.7, 95%confidence interval �CI� 0.5–1.0) and a higher birth weightwas associated with more failed VBAC attempts (54.5%compared with 37.8%, adjusted OR 2.0, 95% CI 1.5–2.8).

CONCLUSION: Birth weight difference has a moderateeffect on the rate of VBAC success in patients whose priorcesarean delivery was for CPD.(Obstet Gynecol 2011;117:343–8)DOI: 10.1097/AOG.0b013e31820776fd

LEVEL OF EVIDENCE: II

The indication for prior cesarean delivery has aneffect on the success of a subsequent attempt at

vaginal birth. Although a cesarean delivery for non-recurring indications, such as malpresentation, is as-sociated with relatively high success rates (80%) invaginal birth after cesarean (VBAC) attempts, successrates fall when the prior cesarean delivery was for arecurring indication, such as cephalopelvic dispropor-tion (CPD) or failed induction.1 In these cases, pa-tients may rightfully ask what is the likelihood of asuccessful VBAC if the current baby is bigger orsmaller than the baby for whom the cesarean deliverywas performed; they may also request that repeatultrasound examinations be performed to determinethe difference in fetal weight between this pregnancyand the previous one. The majority of previouspublications in this area have focused on the absolutebirth weight of the infant in the VBAC attempt andthe probability of a successful VBAC.2–4 However,limited clinical research is available to inform thepatient or clinician on the effect of fetal size in thispregnancy compared with the previous birth in thedecision of delivery mode.

In this study, we attempt to estimate the effect ofbirth weight difference between the current and indexpregnancy on VBAC failure in patients whose priorcesarean delivery was for cephalopelvic disproportion.

METHODSThis was a secondary analysis of a retrospective,multicenter cohort study of women with a priorcesarean delivery.5 Seventeen tertiary and commu-nity hospitals, all with institutional review boardapproval, participated in a study designed to estimatethe rate of and risk factors for uterine rupture inVBAC attempts. The study was conducted from 1995to 2000. Methods of the study have been published indetail previously, but a brief description follows.5

From the Department of Obstetrics and Gynecology, Washington University inSt. Louis, St. Louis, Missouri.

Supported by grants from the Eunice Kennedy Shriver National Institute ofChild Health and Human Development (RO1 HD 35631 to G.A.M.; T32HDto L.M.H. [PI: Macones]) and UL1RR024992 (PI: Evanoff).

Corresponding author: Lorie M. Harper, MD, Department of Obstetrics andGynecology, Washington University in St. Louis, 4911 Barnes JewishHospital Plaza, Campus Box 8064, St. Louis, MO 63110; e-mail:[email protected].

Financial DisclosureThe authors did not report any potential conflicts of interest.

© 2011 by The American College of Obstetricians and Gynecologists. Publishedby Lippincott Williams & Wilkins.ISSN: 0029-7844/11

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 OBSTETRICS & GYNECOLOGY 343

Page 2: Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

Participants were identified at each site usingInternational Classification of Disease, Ninth Revi-sion (ICD-9) codes for “previous cesarean delivery,delivered”; trained research nurses used standardized,closed-end data collection forms to extract data frommedical charts. Three percent of charts were re-extracted for quality control. Data collected includedmaternal demographics, medical and obstetric his-tory, social history, family history, details of the indexpregnancy, antepartum course, labor and deliveryevents, complications, and maternal outcomes.

For the primary analysis, women with one priorcesarean delivery for CPD who attempted VBACwere divided into three groups: birth weight in theVBAC attempt was the same as in the prior cesareandelivery (within 100 g), birth weight in the VBACattempt was more than 100 g smaller than in thecesarean delivery, or birth weight in the VBACattempt was more than 100 g larger than in thecesarean delivery. The weight of 100 g was chosen asa prior study demonstrated a decrease in the odds ofsuccessful VBAC for every 100-g increase in birthweight difference.6 The primary outcome evaluatedwas VBAC attempt failure, defined as a patient whoattempted VBAC but subsequently required a cesar-ean delivery for any indication. Receiver operatingcharacteristics curves were generated to evaluate theutility of birth weight difference to predict VBACfailure. Secondary analyses were performed using anexposure of birth weight in the VBAC attempt of250 g and 500 g greater or less than the cesareandelivery birth weight.

Women were excluded if they had a prior classi-cal cesarean or more than one prior cesarean deliv-ery. As we were primarily interested in viable preg-nancies, women were excluded from the analysis ifthe VBAC birth weight was documented as less than500 g. A documented birth weight of less than 500 gfor the cesarean was excluded as this was likely torepresent a classical cesarean.

Because cephalopelvic disproportion can be dif-ficult to diagnose,7 a secondary analysis of this data setwas performed, including women whose diagnosescould have included CPD. These indications werefailed induction and nonreassuring fetal status. Afailed induction may actually represent CPD, forexample in the case of a macrosomic fetus that fails todescend into the pelvis. A prior cesarean delivery fornonreassuring fetal status may represent a patientwith a protracted labor curve and a category 2 tracing(according to Eunice Kennedy Shriver National Instituteof Child Health and Human Development criteria)8

where the physician suspected CPD. Because it is

difficult to clarify these subtleties from a retrospectivestudy, we elected to include these indications insecondary analyses.

Secondary outcomes examined include uterinerupture, blood transfusions, postpartum fever, and acomposite of complications that included uterine rup-ture, bladder injury, uterine artery laceration, andbowel injury. Definitions of each outcome may befound in prior publications.5

The exposure and comparison groups were com-pared with descriptive and univariable statistics usingANOVA for continuous variables and �2; or Fisherexact tests for categorical variables. Potentially con-founding variables of the exposure-outcome associa-tion were identified in the stratified analyses. Multi-variable logistic regression models for the primaryoutcome were then developed to estimate the effect ofchange in birth weight between the current and indexpregnancy on VBAC failure. Covariates for initialinclusion in multivariable statistical models were se-lected using results of the univariable and stratifiedanalyses and based on historical known confoundingfactors for VBAC success (prior vaginal delivery,diabetes, and spontaneous labor).1,9 Factors were re-moved in a backward stepwise fashion, based onsignificant changes (10%) in the exposure adjustedodds ratio or significant differences between hierar-chical models using the likelihood ratio test. Thestatistical analysis was performed using Stata 10 Spe-cial Edition.

RESULTSThe initial chart review identified 25,076 patients withthe ICD-9 code “previous cesarean delivery, deliv-ered,” 13,706 of whom attempted VBAC. Birthweight data for both the index and current pregnancywere available in 91.0% of these patients. Of these,1,511 had one prior cesarean delivery for CPD. Anadditional 1,175 had one prior cesarean delivery forfailed induction and 1,292 had one prior cesareandelivery for nonreassuring fetal status. Patients in thethree exposure groups (smaller birth weight, samebirth weight, and larger birth weight) were similarwith respect to age, gravidity, race, presence of ahypertensive disorder, spontaneous labor, and oxyto-cin use (Table 1). Women in the same birth weightcategory were slightly more likely to deliver at auniversity hospital and to have their labor augmentedand were less likely to have a prior vaginal delivery,have diabetes, or be induced.

In women attempting VBAC whose prior cesar-ean delivery was for CPD, 586 (38.9%) had a failedVBAC attempt. When grouped according to lower

344 Harper et al Birth Weight Difference and VBAC Success OBSTETRICS & GYNECOLOGY

Page 3: Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

birth weight, same birth weight, or higher birth weightin the VBAC attempt compared with the cesareandelivery, a modest effect of birth weight difference onVBAC failure rate is seen (Table 2). Compared withpatients in the same birth weight category, womenwhose infant had a lower birth weight had a moder-ately lower rate of failed VBAC (29.6% comparedwith 37.8%, adjusted odds ratio �OR� 0.7, 95% confi-dence interval �CI� 0.5–1.0). Higher birth weightcategory had a moderately higher rate of failedVBAC compared with the same birth weight (54.5%compared with 37.8%, adjusted OR 2.0, 95% CI1.5–2.8). Birth weight was used as a continuous vari-able to predict the failure of VBAC. Figure 1 displays

the generated ROC curve for birth weight differenceused to predict failed VBAC; the area under the curveis 0.68.

The risk of uterine rupture, composite complica-tions, and blood transfusion was not significantlydifferent when lower or higher birth weight wascompared with the same birth weight group. The riskof postpartum fever was lower in the lower birthweight group compared with the same birth weightgroup (10.0% compared with 17.3%, adjusted OR 0.5,95% CI 0.3–0.8). The risk of postpartum fever wassimilar between the higher and same birth weightcategories (13.6% compared with 17.3%, adjusted OR0.8, 95% CI 0.5–1.1).

Table 2. Risk of Failed Vaginal Birth After Cesarean in Patients With a Prior Cesarean Delivery forCephalopelvic Disproportion

Lower BirthWeight(n�802)

RR(95% CI)

AdjustedOR

(95% CI)

Same BirthWeight (Ref)

(n�225)

Higher BirthWeight(n�484)

RR(95% CI)

AdjustedOR

(95% CI)

Failed VBAC 237 (29.6) 0.8 (0.6–1.0) 0.7* (0.5–1.0) 85 (37.8) 264 (54.5) 1.4 (1.2–1.7) 2.0* (1.5–2.8)Uterine rupture 8 (1.0) 0.4 (0.2–1.4) — 5 (2.2) 11 (2.2) 1.0 (0.4–2.9) —Blood transfusion 4 (0.5) — — 0 5 (1.0) — —Postpartum fever 80 (10.0) 0.6 (0.4–0.8) 0.5† (0.3–0.8) 39 (17.3) 66 (13.6) 0.8 (0.5–1.1) 0.7† (0.5–1.2)Composite complications‡ 19 (2.4) 0.7 (0.3–1.5) 0.7§ (0.3–1.6) 8 (3.5) 22 (4.5) 1.3 (0.6–2.8) 1.3§ (0.6–3.0)

RR, relative risk; CI, confidence interval; OR, odds ratio; VBAC, vaginal birth after cesarean; —, unable to perform calculation eitherdue to small number of cases or reference group of 0.

Prior cesarean delivery for cephalopelvic disproportion only (n�1,515). Data are n (%).* Adjusted for prior vaginal delivery, labor induction, and African-American race.† Adjusted for prior vaginal delivery and African-American race.‡ Includes uterine rupture, bladder injury, uterine artery laceration, and bowel injury.§ Adjusted for prior vaginal delivery.

Table 1. Characteristics of Patients by Birth Weight Difference

Lower Birth Weight(n�1,756)

Same Birth Weight*(n�575)

Higher Birth Weight(n�1,644) P

Maternal age (y) 30.1�5.5 29.9�5.6 29.8�5.6 .24Gravidity 3.2�1.4 3.0�1.4 3.2�1.5 .20University hospital 911 (51.9) 262 (45.5) 850 (51.7) .02Prior vaginal delivery 476 (27.1) 127 (22.0) 444 (27.0) .04African-American race 545 (31.0) 178 (30.9) 557 (33.9) .16Labor type

Spontaneous 773 (44.0) 257 (44.6) 757 (46.0) .49Augmented 352 (20.0) 141 (24.5) 355 (21.6) .07Induced 632 (36.0) 178 (30.9) 533 (32.4) .03Oxytocin use 783 (44.6) 269 (46.7) 719 (43.7) .46

Medical problemsDiabetes 115 (6.6) 11 (1.9) 98 (6.0) �.01Any hypertension† 96 (5.5) 26 (4.5) 84 (5.1) .66

Indication for prior cesarean delivery �.01Cephalopelvic disproportion 802 (45.7) 225 (39.1) 484 (29.4)Failed induction 525 (29.9) 183 (31.8) 467 (28.4)Nonreassuring fetal status 430 (24.5) 168 (29.2) 694 (42.2)

Data are mean�standard deviation or n (%) unless otherwise specified.* Same birth weight�within 100 g of previous birth weight.† Includes chronic hypertension, gestational hypertension, and preeclampsia.

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 Harper et al Birth Weight Difference and VBAC Success 345

Page 4: Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

A similar pattern was seen when patients wereincluded whose prior cesarean delivery was performedfor failed induction (Table 3). Again, patients with alower birth weight in their VBAC attempt comparedwith their cesarean delivery had a 30% lower rate offailed VBAC, while patients with a higher birth weighthad a 40% higher rate of failed VBAC. Patients in thelower birth weight group had a 50% lower rate ofcomposite complications and a 40% lower rate of post-partum fever compared with the same birth weightgroup; the risk of uterine rupture and blood transfusionwas similar. The risk of complications in the higher birthweight group was similar to the same birth weight group.

When including patients whose prior cesarean de-livery was for CPD, failed induction or nonreassuring

fetal status, the risk of failed VBAC was moderatelylower in patients whose infant was smaller comparedwith women whose infants were the same size (Table 4).The risk of failed VBAC attempt was slightly increasedin patients in the higher birth weight category comparedwith same birth weight (42.1% compared with 38.7%,adjusted OR 1.2, 95% CI 1.0–1.5). Patients in the lowerbirth weight group had a 40% lower rate of postpartumfever but were similar to the same birth weight groupwith respect to uterine rupture, blood transfusion, andcomposite complications. The risk of complications inthe higher birth weight group was similar to the samebirth weight.

Similar results were obtained when using anexposure based on birth weight differences of 250 gand 500vg. A modest decrease in failed VBAC at-tempts was seen for the smaller birth weight group(adjusted OR 0.5, 95% CI 0.4–0.6 for 250 g, adjustedOR 0.5, 95% CI 0.4–0.7) and a modest increase infailed VBAC attempts was seen in the larger birthweight group (adjusted OR 1.3, 95% CI 1.1–1.4 for250 g, adjusted OR 1.4, 95% CI 1.2–1.6 for 500 g) inparticipants with a prior cesarean delivery for CPD(data not shown, available upon request).

In patients whose prior cesarean delivery was forCPD, the effect of incremental increases in birth weightwas examined (Table 5). A distinct pattern of increasedrisk of VBAC failure as birth weight gradually increasedwas not seen. Thus, these data do not support thepresence of a dose-response relationship between birthweight difference and failed VBAC attempt.

DISCUSSIONIn this large retrospective cohort, birth weight differ-ence between the VBAC attempt and a prior cesarean

Fig. 1. Receiver Operator Characteristics (ROC) curve ofbirth weight difference to predict a failed vaginal birth aftercesarean (VBAC) attempt in patients with a history ofcephalopelvic disproportion, failed induction, or cesareanfor non-reassuring fetal status.Harper. Birth Weight Difference and VBAC Success. ObstetGynecol 2011.

Table 3. Risk of Failed Vaginal Birth After Cesarean in Patients With a Prior Cesarean Delivery forCephalopelvic Disproportion or Failed Induction

Lower BirthWeight

(n�1,327)RR

(95% CI)

AdjustedOR

(95% CI)

Same BirthWeight (Ref)

(n�408)

HigherBirth Weight

(n�951)RR

(95% CI)

AdjustedOR

(95% CI)

Failed VBAC 417 (31.4) 0.8 (0.7–0.9) 0.7* (0.5–0.8) 168 (41.2) 479 (50.4) 1.2 (1.1–1.4) 1.4* (1.2–1.9)Uterine rupture 10 (0.8) 0.4 (0.2–1.1) — 7 (1.7) 15 (1.6) 0.9 (0.4–2.2) —Blood transfusion 8 (0.6) 0.8 (0.2–3.1) — 3 (0.7) 11 (1.2) 1.6 (0.4–5.6) —Postpartum fever 143 (10.8) 0.6 (0.5–0.8) 0.6† (0.4–0.8) 71 (17.4) 133 (14.0) 0.8 (0.6–1.0) 0.8† (0.6–1.1)Composite complications‡ 30 (2.3) 0.5 (0.3–0.9) 0.5§ (0.3–0.9) 18 (4.4) 42 (4.4) 1.0 (0.6–1.7) 1.0§ (0.6–1.8)

RR, relative risk; CI, confidence interval; OR, odds ratio; VBAC, vaginal birth after cesarean; —, unable to perform calculation eitherdue to small number of cases or reference group of 0.

Prior cesarean delivery for cephalopelvic disproportion or failed induction (n�2,691). Data are n (%).* Adjusted for prior vaginal delivery, labor induction, and African-American race.† Adjusted for prior vaginal delivery and African-American race.‡ Includes uterine rupture, bladder injury, uterine artery laceration, and bowel injury.§ Adjusted for prior vaginal delivery.

346 Harper et al Birth Weight Difference and VBAC Success OBSTETRICS & GYNECOLOGY

Page 5: Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

delivery for cephalopelvic disproportion had a mod-erate, although statistically significant, effect on VBACfailure; however, examination of the receiver operatingcharacteristics curve reveals that birth weight differenceis not a useful predictor of VBAC failure based on themodest area under the curve and the lack of a cleardiscriminatory point that could discern those that aredestined for VBAC failure. A similar, although dimin-ished, effect of birth weight difference was seen inpatients whose prior cesarean delivery was performedfor failed induction and nonreassuring fetal status.Therefore, we conclude that the effect of birth weightdifference is small and should not be used as a criticalfactor in deciding whether VBAC should be attempted.Allowing women with a larger infant than their priorpregnancy to attempt VBAC does not seem to increasethe risk of complications.

We used actual birth weights in our analysis;however, clinicians must utilize estimated fetal weightfrom either physical examination or ultrasound exam-ination, which has an error of 15–20% at term.10,11

Given the error in estimating fetal weight at term, themisclassification bias created by using estimated fetalweight would further decrease the ability to predictVBAC failure. Because actual birth weight does notgenerate a clinically useful prediction model for pre-dicting failed VBAC, we anticipate that ultrasonogra-phy-based estimated fetal weight will not either.

Several previous studies have examined the effectof macrosomia on VBAC success. Zelop et al foundthat a macrosomic fetus was associated with a de-creased rate of VBAC success and no change in therisk of uterine rupture.4 Elkousy et al confirmed thefinding that macrosomia negatively affects the rate ofVBAC success, a finding that is exacerbated in thepopulation with a prior indication of CPD.2 Althoughinformative about the effect of macrosomia onVBAC, these studies do not include informationabout the effect of birth weight difference.

In a secondary analysis of a large prospectivecohort study, Peaceman et al found that as the birthweight difference between the VBAC attempt andprior cesarean delivery increased, the rate of VBACsuccess fell.6 Every 100-g increase in infant birthweight resulted in a 3.8% decrease in the odds of avaginal delivery. One of the main differences betweentheir study and ours was the definition of the expo-sure; Peaceman et al included failed induction as partof CPD, whereas we analyzed this in a secondaryanalysis as failed inductions may or may not representa subset of CPD patients. Additionally, we also in-

Table 4. Risk of Failed Vaginal Birth After Cesarean in Patients With a Prior Cesarean Delivery forCephalopelvic Disproportion, Failed Induction, or Nonreassuring Fetal Status

Lower BirthWeight

(n�1,757)RR

(95% CI)

AdjustedOR

(95% CI)

Same BirthWeight (Ref)

(n�576)

Higher BirthWeight

(n�1,645)RR

(95% CI)

AdjustedOR

(95% CI)

Failed VBAC (%) 534 (30.4) 0.8 (0.7–0.9) 0.7* (0.6–0.8) 223 (38.7) 693 (42.1) 1.1 (1.0–1.2) 1.2* (1.0–1.5)Uterine rupture 19 (1.1) 0.9 (0.4–2.1) — 7 (1.2) 19 (1.1) 1.0 (0.4–2.3) —Blood transfusion 11 (0.6) 1.2 (0.3–4.3) — 3 (0.5) 15 (0.9) 1.8 (0.5–6.0) —Postpartum fever 180 (10.2) 0.7 (0.5–0.8) 0.6† (0.5–0.8) 88 (15.3) 210 (12.8) 0.8 (0.7–1.1) 0.8† (0.6–1.1)Composite complications‡ 46 (2.6) 0.7 (0.4–1.2) 0.7§ (0.4–1.2) 21 (3.6) 56 (3.4) 0.9 (0.6–1.5) 1.0§ (0.6–1.6)

RR, relative risk; CI, confidence interval; OR, odds ratio; VBAC, vaginal birth after cesarean; —, unable to perform calculation eitherdue to small number of cases or reference group of 0.

Data are n (%) unless otherwise specified.Prior cesarean delivery for cephalopelvic disproportion, failed induction, or nonreassuring fetal status (n�3,984).* Adjusted for prior vaginal delivery, African-American race, and labor induction.† Adjusted for prior vaginal delivery and African-American race.‡ Includes uterine rupture, bladder injury, uterine artery laceration, and bowel injury.§ Adjusted for prior vaginal delivery.

Table 5. Risk of Failed Vaginal Birth AfterCesarean for Patients by IncrementalIncreases in Birth Weight Difference

FailedVBAC�n (%)�

RR(95% CI)

AOR(95% CI)

0–100 g (n�119) 49 (41.2) Ref Ref101–250 g (n�148) 80 (54.1) 1.3 (1.0–1.7) 1.7* (1.1–2.9)251–500 g (n�169) 89 (52.7) 1.3 (1.0–1.7) 1.5† (0.9–2.5)501–750 g (n�105) 56 (53.3) 1.3 (1.0–1.7) 1.6* (0.9–2.7)751–1,000 g (n�37) 25 (67.6) 1.6 (1.2–2.2) 3.1* (1.4–6.8)More than 1,000 g

(n�25)14 (56.0) 1.4 (0.9–2.0) 1.9* (0.8–4.6)

VBAC, vaginal birth after cesarean; RR, relative risk; CI,confidence interval; AOR, adjusted odds ratio.

* Adjusted for prior vaginal delivery.† Adjusted for prior vaginal delivery and diabetes.

VOL. 117, NO. 2, PART 1, FEBRUARY 2011 Harper et al Birth Weight Difference and VBAC Success 347

Page 6: Vaginal_Birth_After_Cesarean_for_Cephalopelvic.21.pdf

cluded nonreassuring fetal status as a subset of CPD,and we examined complication rates in these groups.

The strengths of this study are its large size andcomprehensive clinical data available, allowing us toexamine outcomes, including maternal complica-tions, in this very specific subset of patients. Also, weanalyzed patients with a prior cesarean delivery forfailed induction or nonreassuring fetal status sepa-rately, as these may or may not represent cases ofcephalopelvic disproportion.

An inherent limitation of a retrospective study isthe possibility of selection bias. Specifically, physi-cians may encourage women who are “good” VBACcandidates (ie, prior vaginal delivery, spontaneouslabor) to attempt VBAC and encourage patients whoare poor candidates to undergo elective repeat cesar-ean delivery. This type of selection bias may havediminished the observed effect of a larger infant onthe rate of VBAC failure. Also, as the birth weightdifference increased, the number of patients attempt-ing VBAC decreased, thus limiting our power todetect a difference.

As this cohort was designed to investigate mater-nal risks associated with VBAC, we were unable toexamine infant outcomes. Although we have exten-sive information available regarding maternal comor-bidities, information on maternal body mass index,which has been noted to be a variable affectingVBAC success, is unavailable in this cohort.12,13 Fi-nally, a small number of patients did not have birthweight data available for both pregnancies, leading toanother potential selection bias. However, these pa-tients account for only 9% of the study population andwere characteristically similar to those in the studysample with few exceptions (less likely to be AfricanAmerican, more likely to deliver at a universityhospital, slight differences in labor type).

Despite these limitations, we feel that clinicallyimportant conclusions can be drawn. Although birthweight difference modestly affects VBAC success inpatients with a prior cesarean delivery for CPD,evaluation of birth weight difference incrementallyand the receiver operating characteristics curve didnot reveal a clear cutoff where the risk of failed VBACbecame unacceptable. Therefore, we conclude thatalthough VBAC failure increases moderately as

VBAC birth weight increases over the index cesar-ean birth weight, we cannot make recommenda-tions on a difference over which VBAC should notbe attempted.

REFERENCES1. Landon MB, Leindecker S, Spong CY, Hauth JC, Bloom S,

Varner MW, et al. The MFMU Cesarean Registry: factorsaffecting the success of trial of labor after previous cesareandelivery. Am J Obstet Gynecol 2005;193(3 pt 2):1016–23.

2. Elkousy MA, Sammel M, Stevens E, Peipert JF, Macones G.The effect of birth weight on vaginal birth after cesareandelivery success rates. Am J Obstet Gynecol 2003;188:824–30.

3. Nwokoro CA, Njokanma OF, Orebamjo T, Okeke GC. Vag-inal birth after primary cesarean section: the fetal size factor.J Obstet Gynaecol 2003;23:392–3.

4. Zelop CM, Shipp TD, Repke JT, Cohen A, Lieberman E.Outcomes of trial of labor following previous cesarean deliveryamong women with fetuses weighing �4000 g. Am J ObstetGynecol 2001;185:903–5.

5. Macones GA, Peipert J, Nelson DB, Odibo A, Stevens EJ,Stamilio DM, et al. Maternal complications with vaginal birthafter cesarean delivery: a multicenter study. Am J ObstetGynecol 2005;193:1656–62.

6. Peaceman AM, Gersnoviez R, Landon MB, Spong CY, Lev-eno KJ, Varner MW, et al. The MFMU Cesarean Registry:impact of fetal size on trial of labor success for patients withprevious cesarean for dystocia. Am J Obstet Gynecol 2006;195:1127–31.

7. Dystocia and augmentation of labor. ACOG Practice BulletinNo. 49. American College of Obstetricians and Gynecologists.Obstet Gynecol 2003;102:1445–54.

8. Macones GA, Hankins GD, Spong CY, Hauth J, Moore T. The2008 National Institute of Child Health and Human Develop-ment workshop report on electronic fetal monitoring: updateon definitions, interpretation, and research guidelines. ObstetGynecol 2008;112:661–6.

9. Dharan VB, Srinivas SK, Parry S, Ratcliffe SJ, Macones G.Pregestational diabetes: a risk factor for vaginal birth aftercesarean section failure? Am J Perinatol 2010;27:265–70.

10. Anderson NG, Jolley IJ, Wells JE. Sonographic estimation offetal weight: comparison of bias, precision and consistencyusing 12 different formulae. Ultrasound Obstet Gynecol 2007;30:173–9.

11. Nahum GG, Stanislaw H. Ultrasonographic prediction of termbirth weight: how accurate is it? Am J Obstet Gynecol 2003;188:566–74.

12. Durnwald CP, Ehrenberg HM, Mercer BM. The impact ofmaternal obesity and weight gain on vaginal birth after cesar-ean section success. Am J Obstet Gynecol 2004;191:954–7.

13. Hibbard JU, Gilbert S, Landon MB, Hauth JC, Leveno KJ,Spong CY, et al. Trial of labor or repeat cesarean delivery inwomen with morbid obesity and previous cesarean delivery.Obstet Gynecol 2006;108:125–33.

348 Harper et al Birth Weight Difference and VBAC Success OBSTETRICS & GYNECOLOGY