Share this article on:

White’s Classification of Maternal Diabetes and Vaginal Birth After Cesarean Delivery Success in Women Undergoing a Trial of Labor

Cormier, Clint M. MD; Landon, Mark B. MD; Lai, Yinglei PhD; Spong, Catherine Y. MD; Rouse, Dwight J. MD; Leveno, Kenneth J. MD; Varner, Michael W. MD; Simhan, Hyagriv N. MD; Wapner, Ronald J. MD; Sorokin, Yoram MD; Miodovnik, Menachem MD; Carpenter, Marshall MD; Peaceman, Alan M. MD; O’Sullivan, Mary J. MD; Sibai, Baha M. MD; Langer, Oded MD; Thorp, John M. MD; Mercer, Brian M. MD

doi: 10.1097/AOG.0b013e3181c534ca
Original Research

OBJECTIVE: To estimate the rate of vaginal birth after cesarean delivery (VBAC) success in diabetic women based on White’s Classification.

METHODS: This is a secondary analysis of an observational study conducted at 19 medical centers of women attempting VBAC. Diabetic women with singleton gestations, one prior cesarean delivery, and cephalic presentation who underwent a trial of labor were included. Vaginal birth after cesarean delivery success rates and maternal and neonatal complications were compared based on White’s Classification.

RESULTS: Of 11,856 women who underwent trial of labor, 624 met all study criteria (class A1, 356; A2, 169; B, 70; C, 21; D/R/F, 8). Vaginal birth after cesarean delivery success in each group was: A1, 68.5% (95% confidence interval [CI] 63.4–73.3%); A2, 55% (95% CI 47.2–62.7%); B, 70% (95% CI 57.9–80.4%); C, 47.6% (95% CI 25.7–70.2%); and D/F/R, 12.5% (95% CI 0.3–52.7%). Maternal and neonatal complications were rare and not found to be different among groups.

CONCLUSION: Our study provides estimates for VBAC success based on White’s classification and indicates a relatively low rate of perinatal complications after VBAC attempt for diabetic women.


Our study provides estimates for vaginal birth after cesarean delivery (VBAC) success based on White’s Classification and indicates a relatively low rate of perinatal complications after VBAC attempt for diabetic women. SUPPLEMENTAL DIGITAL CONTENT IS AVAILABLE IN THE TEXT.

From the Departments of Obstetrics and Gynecology University of Texas Health Science Center at Houston, Houston, Texas; The Ohio State University, Columbus, Ohio; University of Alabama at Birmingham, Birmingham Alabama; University of Texas Southwestern Medical Center, Dallas, Texas; University of Utah, Salt Lake City, Utah; University of Pittsburgh, Pittsburgh, Pennsylvania; Thomas Jefferson University, Philadelphia, Pennsylvania; Wayne State University, Detroit, Michigan; University of Cincinnati, Cincinnati, Ohio; Columbia University, New York, New York; Brown University, Providence, Rhode Island; Northwestern University, Chicago, Illinois; University of Miami, Miami, Florida; University of Tennessee, Memphis, Tennessee; University of Texas Health Science Center at San Antonio, San Antonio, Texas; University of North Carolina at Chapel Hill, Chapel Hill, North Carolina; Case Western Reserve University Cleveland, Ohio, The George Washington University Biostatistics Center, Washington, DC, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development.

*For a list of other members of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network, see the Appendix available online at

Supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (HD21410, HD21414, HD27860, HD27861, HD27869, HD27905, HD27915, HD27917, HD34116, HD34122, HD34136, HD34208, HD34210, HD40500, HD40485, HD40544, HD40545, HD40560, HD40512, and HD36801).

Presented in part at the 55th Annual Meeting of the Society for Gynecologic Investigation, March 26–29, 2008, San Diego, California.

The authors thank Francee Johnson, RN, BSN, for protocol development and coordination between clinical research centers and Elizabeth Thom, PhD, and Sharon Gilbert, MS, MBA, for protocol and data management and statistical analysis.

Dr. Spong, Associate Editor of Obstetrics & Gynecology, was not involved in the review or decision to publish this article.

Corresponding author: Clint M. Cormier, MD, Department of Obstetrics, Gynecology, and Reproductive Services, University of Texas Health Sciences Center at Houston, 6431 Fannin Street, Suite 3.430, Houston, TX 77030; e-mail:

Financial Disclosure: The authors did not report any potential conflicts of interest.

More than a fourth of all births in the United States are by cesarean delivery (CD).1 Attempts to lower the cesarean delivery rate have focused on encouraging vaginal birth after cesarean delivery (VBAC). An integral part of achieving a lower cesarean delivery rate through better utilization of VBAC is predicting who is not a good candidate for trial of VBAC.2

The incidence of diabetes among American adults has increased from 4.9% in 1990 to 6.9% in 1999.3 The lifetime risk of diabetes for an American female is approximately 39%.3 Prevalence of gestational diabetes in the United States is reported to be between 1.4% and 14% of pregnancies.4,5 More reproductive-aged women are being diagnosed with diabetes than ever before, as the incidence of both pregestational and gestational diabetes is approaching nearly 5% of all pregnancies.6

Diabetic women are at increased risk of cesarean delivery secondary to labor arrest, failed induction, and fetal intolerance to labor.7,8 Furthermore, diabetic women have increased body mass index and weight gain during pregnancy, both of which have a negative impact on VBAC success.8 Large population-based studies have shown VBAC in the nondiabetic population to be a reasonable option for the patient and provider who are motivated to avoid the maternal morbidity associated with repeat cesarean delivery.9 Currently, there is a paucity of data regarding VBAC outcomes for diabetic women. Coleman et al10 showed that the VBAC success rate was lower in gestational diabetic women when compared with a nondiabetic population, while Grinstead et al11 identified medical conditions, of which diabetes was included, to be a predictor of VBAC failure. There is a relationship between increasing cesarean delivery rates and severity of diabetes.12 Vaginal birth after cesarean delivery success rate based on disease severity has been examined previously, although that study took place in a single center.13

The purpose of this study was to estimate the rate of VBAC success in diabetic women by White’s classification of diabetes in pregnancy.14 Secondary objectives included estimating maternal and neonatal complication rates in diabetic women attempting VBAC.

Back to Top | Article Outline


This is a secondary analysis of a 4-year observational study conducted at 19 academic medical centers of the Eunice Kennedy Shriver National Institute of Child Health and Human Development Maternal–Fetal Medicine Units Network (NICHD-MFMU Network) between 1999 and 2002.9 Details of the study design have been reported previously.9 Inclusion criteria for this analysis were maternal diabetes, singleton gestation, one prior low-transverse cesarean delivery, gestational age 37 0/7 weeks or more, and a VBAC attempt. Women with pregnancies complicated by fetal structural or chromosomal abnormalities were excluded.

Maternal demographic, clinical, and outcome data were abstracted from medical records charts by trained research nurses. Vaginal birth after cesarean delivery success was defined as VBAC attempt that ended in vaginal delivery. A composite outcome of maternal complications, which included uterine rupture, uterine dehiscence, need for hysterectomy (by cesarean delivery), deep vein thrombosis, need for transfusion (intraoperative/postoperative), and maternal death, was reported. In a similar fashion, a composite of neonatal complications, which included 5-minute Apgar scores less than 4, umbilical artery pH less than 7.0, seizures (confirmed), hypoxic-ischemic encephalopathy, and neonatal death, was reported.

Maternal and neonatal characteristics and outcomes were stratified by White’s class (A1, A2, B, C, or D/F/R)14 and compared with the exact Cochran-Armitage test of trend.15 Continuous variables were compared with the use of the Wilcoxon rank sum test and categorical variables were compared with the use of χ2 or Fisher exact test. Multivariable logistic regression was used to adjust for gestational age and labor induction for the primary outcome of successful VBAC. Nominal two-sided P values are reported with statistical significance defined as a P<.05. No adjustment was considered for multiple comparisons. SAS (SAS Institute, Cary, NC) was used for the analysis except the exact tests and confidence intervals (Clopper-Pearson),16 which were computed by Cytel Studio 8 (Cytel Inc., Cambridge, MA). Approval for the original study was obtained at each of the originally participating institutions. Approval for this secondary analysis was obtained from the University of Texas Health Science Center in Houston Institutional Review Board.

Back to Top | Article Outline


In the overall cohort, 11,856 women underwent trial of labor. There were 1,358 women with diabetes; 624 (45.9%) attempted VBAC. The rate of VBAC attempt by White’s class was 48.7% for A1, 45.3% for A2, 44.9% for B, 33.3% for C, and 23.5% for D/F/R. Comparisons of demographic and clinical characteristics by White’s class groups are presented in Table 1. Gestational age at delivery and body mass index (prepregnancy) were different among the White’s class groups. The frequency of chronic hypertension and labor induction also increased with advancing class. In women with unsuccessful VBAC, fetal indications (cord prolapsed, nonreassuring fetal heart rate tracing and abruption) for cesarean delivery were common: A1, 25%; A2, 29%; B, 43%; C, 18%; and D/F/R, 57%. Figure 1 describes the VBAC success rate by White’s class. The overall VBAC success rate was 63.6%. VBAC success in each White’s class group was: A1, 68.5% (95% confidence interval [CI] 63.4–3.3%); A2, 55% (95% CI 47.2–62.7%); B, 70% (95% CI 57.9–80.4%); C, 47.6% (95% CI 25.7–70.2%); D/F/R, 12.5% (95% CI 0.3–52.7%).

Table 1

Table 1



In the initial analysis, an unadjusted test of trend was performed that indicated decreased VBAC success with advancing White’s class (P=.004). However, when adjusted for gestational age and labor induction, the only differences in VBAC success rates were between Class A1 (odds ratio 12.7, 95% CI 1.5–109.1) and B (odds ratio 16.0, 95% CI 1.8–143.5) when compared with the D/F/R group. Maternal and neonatal complications are described in Table 2. There were no differences in major adverse outcomes among groups.

Table 2

Table 2

Back to Top | Article Outline


Our study evaluated VBAC outcomes for more than 600 diabetic women and noted that VBAC success rate was 64%, while in the nondiabetic women from the original cohort the success rate was 73.6%.9 When examining individual White’s class groups, diabetic women in both class C and D/F/R had unadjusted VBAC success rates that were below 50%. Due to the relatively low number of women in these categories studied (n=29), we are unable to draw strong conclusions. Given the lack of prior data and the clinical importance, it is relevant to offer some discussion regarding the potential causes. Prior studies have shown a higher rate of fetal growth restriction for women with advanced diabetes,17 and given the high rate of cesarean delivery for fetal indication in these groups, this may be one reason for the relatively low success rates. Long-standing diabetes, and its associated vasculopathy, could predispose those patients to VBAC failure due underlying uteroplacental insufficiency and fetal intolerance to labor. Additionally, clinicians may be less willing to tolerate abnormal fetal heart rate patterns due to maternal disease status or have bias toward lower threshold for cesarean delivery due to maternal disease status. We also noted that maternal and neonatal complications were rare and not different across White’s class groups. Our findings are similar to those of Blackwell et al13 who also specifically looked at VBAC success in diabetics. Their results showed a higher cesarean delivery rate in diabetics undergoing VBAC (56.3%) compared with diabetic women who did not have a history of cesarean delivery (26.3%). They also showed no significant difference in frequency of major maternal or neonatal complications. However, they did not compare the patients in their study based on White’s class.

Strengths of this study include the considerable sample size, multiple centers, and standardized data collection and methodology. Also, as a multicenter study, our sample population represents a diverse group of patients. Our study has some limitations. The relatively small number of women with White’s class D/F/R diabetes makes point estimation regarding the VBAC success in this group difficult, as evidenced by the large CI surrounding the VBAC success rate. Also, given the overall low frequency of maternal or neonatal complications, we did not have adequate power to detect small differences among groups. While we adjusted for confounders, such as induction of labor and gestational age at time of delivery, we did not adjust for other clinical factors (eg, parity, cervical status). Because this was a multicenter study, center-to-center variability likely occurs. Other factors, such as interphysician variability with regard to decision to attempt VBAC or threshold for converting from VBAC attempt to cesarean delivery, was also not evaluated. Because the original study on which this analysis was performed was observational, no specific directives were given to practitioners with regard to those decisions. Therefore, due to the nature of the dataset, we cannot rule out selection bias.

This study provides useful data for the counseling of diabetic women with previous cesarean delivery who are evaluating the risks and benefits of VBAC. Our study provides estimates for VBAC success based on White’s class and indicates a relatively low rate of perinatal complications after VBAC attempt for diabetic women. The decision for a diabetic woman to attempt VBAC should be made jointly with her and her physician on a case-by-case basis.

Back to Top | Article Outline


1.Martin JA, Sutton PD, Ventura SJ, Menacker F, Munson ML. Births: the final data for 2003. Natl Vital Stat Rep 2005;54:1–116.
2.Durnwald C, Mercer B. Vaginal birth after cesarean delivery: predicting success, risks of failure. J Matern Fetal Neonatal Med 2004;15:388–93.
3.ADA. Economic costs of diabetes in the U.S. in 2002. Diabetes Care 2002;26:917–32.
4.Dabelea D, Snell-Bergeon JK, Hartsfield CL, Bischoff KJ, Hamman RF, McDuffie RS. Increasing prevalence of gestational diabetes mellitus (GDM) over time and by birth cohort: Kaiser Permanente of Colorado GDM screening program. Diabetes Care 2005;28:579–84.
5.Getahun D, Nath C, Ananth CV, Chavez MR, Smulian JC. Gestational diabetes in the United States: temporal trends 1989 through 2004. Am J Obstet Gynecol 2008;198:e1–5.
6.Laolia A, Dalfra MG, Fedele D. Pregnancy complicated by type 2 diabetes: An emerging problem. Diabetes Res Clin Pract 2008;80:2–7.
7.Cousins L. Pregnancy complications among diabetic women: Review 1965–1985. Obstet Gynecol Surv 1987;42:140–9.
8.Juhasz G, Gyamfi C, Gyamfi P, Tocce K, Stone JL. Effect of body mass index and excessive weight gain on success of vaginal birth after cesarean delivery. Obstet Gynecol 2005;106:741–6.
9.Landon M, Hauth JC, Leveno KJ, Spong CY, Leindecker S, Varner MW, et al for The NICHD/MFMU Network. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med 2004;351:2581–9.
10.Coleman TL, Randall H, Graves W, Lindsay M. Vaginal birth after cesarean among women with gestational diabetes. Am J Obstet Gynecol 2001;45:987–90.
11.Grinstead J, Grobman WA. Induction of labor after one prior cesarean: Predictors of vaginal delivery. Obstet Gynecol 2004;103:534–8.
12.Blackwell SC, Hassan SS, Wolfe HW, Michaelson J, Berry SM, Sorokin Y. Why are cesarean delivery rates so high in diabetic pregnancies? J Perinat Med 2000;28:316–20.
13.Blackwell SC, Hassan SS, Wolfe HW, Michaelson J, Berry SM, Sorokin Y. Vaginal birth after cesarean in the diabetic gravida. J Repro Med 2000;45:987–90.
14.White P. Classification of obstetric diabetes. Am J Obstet Gynecol 1978;130:228–30.
15.Mehta CR, Patel NR, Senchaudhuri P. Exact power and sample-size computations for the Cochran-Armitage trend test. Biometrics 1998;54:1615–21.
16.Clopper CJ, Pearson E. The use of confidence or fiducial limits illustrated in the case of binomial. Biometrika 1934;26:404–13.
17.Becker T, Vermuelen MJ, Wyatt PR, Meier C, Ray JG. Prepregnancy diabetes and risk of placental vascular disease. Diabetes Care 2007;30:2496–8.
18.Kruskal WH, Wallis WA. Use of ranks in one-criterion variance analysis. J Am Stat Assoc 1952;47:583–621.


Supplemental Digital Content

Back to Top | Article Outline
© 2010 by The American College of Obstetricians and Gynecologists.