Obstetrics & Gynecology:
Relationship Between 1-Hour Glucose Challenge Test Results and Perinatal Outcomes
Figueroa, Dana MD; Landon, Mark B. MD; Mele, Lisa ScM; Spong, Catherine Y. MD; Ramin, Susan M. MD; Casey, Brian MD; Wapner, Ronald J. MD; Varner, Michael W. MD; Thorp, John M. Jr MD; Sciscione, Anthony DO; Catalano, Patrick MD; Harper, Margaret MD, MSc; Saade, George MD; Caritis, Steve N. MD; Sorokin, Yoram MS; Peaceman, Alan M. MD; Tolosa, Jorge E. MD, MSCE; for the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units (MFMU) Network
Departments of Obstetrics and Gynecology, University of Alabama at Birmingham, Birmingham, Alabama, The Ohio State University, Columbus, Ohio, University of Texas Health Science Center at Houston, Houston, Texas, University of Texas Southwestern Medical Center, Dallas, Texas, Columbia University, New York, New York, University of Utah, Salt Lake City, Utah, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Drexel University, Philadelphia, Pennsylvania, Case Western Reserve University-MetroHealth Medical Center, Cleveland, Ohio, Wake Forest University Health Sciences, Winston-Salem, North Carolina, University of Texas Medical Branch, Galveston, Texas, University of Pittsburgh, Pittsburgh, Pennsylvania, Wayne State University, Detroit, Michigan, Northwestern University, Chicago, Illinois, and Oregon Health & Science University, Portland, Oregon; The George Washington University Biostatistics Center, Washington, DC; and the Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, Maryland.
Corresponding author: Dana Figueroa, MD, 176F 10270C, 619 19th Street S, Birmingham, AL 35249-7333; e-mail: firstname.lastname@example.org.
Supported by grants from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) (HD27915, HD34116, HD40485, HD34208, HD27869, HD40500, HD40560, HD34136, HD40544, HD27860, HD40545, HD53097, HD21410, HD27917, HD40512, HD53118, HD36801), a General Clinical Research Centers Grant (M01-RR00034), and the National Center for Research Resources (UL1-RR024989, M01-RR00080, UL1-RR025764, C06-RR11234) and does not necessarily represent the official views of the NICHD or the National Institutes of Health.
Financial Disclosure The authors did not report any potential conflicts of interest.
Presented in part as a poster at the 31st Annual Meeting of the Society for Maternal-Fetal Medicine, February 9–12, 2011, San Francisco, California.
* For a list of other members of the NICHD MFMU, see the Appendix online at http://links.lww.com/AOG/A386.
The authors thank Alan T. N. Tita, MD, PhD, for assistance with study design and manuscript development; Francee Johnson, RN, Joanne Tillinghast, RN, and Susan Tolivasia for coordination between clinical research centers; and Elizabeth Thom, PhD, for study design, data management, statistical analysis, and manuscript development.
Dr. Spong and Dr. Rouse, Associate Editors of Obstetrics & Gynecology, were not involved in the review or decision to publish this article.
OBJECTIVE: To estimate the relationship between 1-hour 50 g glucose challenge test values and perinatal outcomes.
METHODS: This was a secondary analysis of data from a multicenter treatment trial of mild gestational diabetes mellitus. Women with glucose challenge test values of 135–199 mg/dL completed a 3-hour oral glucose tolerance test. Mild gestational diabetes mellitus was defined as fasting glucose less than 95 mg/dL and two or more abnormal oral glucose tolerance test values: 1-hour 180 mg/dL or more; 2-hour 155 mg/dL or more; and 3-hour 140 mg/dL or more. Our study included untreated women with glucose challenge test values of 135–139 mg/dL and 140–199 mg/dL and a comparison group with values less than 120 mg/dL. Primary outcomes included a perinatal composite (stillbirth, neonatal death, hypoglycemia, hyperbilirubinemia, neonatal hyperinsulinemia, and birth trauma), large for gestational age (LGA, birth weight above the 90th percentile based on sex-specific and race-specific norms), and macrosomia (greater than 4,000 g).
RESULTS: There were 436 women with glucose challenge test values less than 120 mg/dL and 1,403 with values of 135 mg/dL or more (135–139, n=135; 140–199, n=1,268). The composite perinatal outcome occurred in 25.6% of those with glucose challenge test values less than 120 mg/dL compared with 21.1% for values of 135–139 mg/dL and 35.3% for values of 140–199 mg/dL. Rates of LGA by group were 6.6%, 6.8%, and 12.4%, respectively. Rates of macrosomia by group were 7.8%, 6.1%, and 12.1%, respectively. Compared with glucose challenge test values less than 120 mg/dL, the adjusted odds ratios (ORs) (95% confidence intervals [CIs]) for values of 140–199 mg/dL were 1.48 (1.14–1.93) for the composite outcome, 1.97 (1.29–3.11) for LGA, and 1.61 (1.07–2.49) for macrosomia. For glucose challenge test values of 135–139 mg/dL, adjusted ORs and 95% CIs were 0.75 (0.45–1.21), 1.04 (0.44–2.24), and 0.75 (0.30–1.66), respectively. The subcategories with glucose challenge test values of 140–144 mg/dL and 145–149 mg/dL also were associated with an increase in selected outcomes when compared with those with values less than 120 mg/dL.
CONCLUSIONS: Glucose challenge test values of 135–139 mg/dL were not associated with adverse outcomes compared with values less than 120 mg/dL; however, glucose challenge test values of 140 mg/dL or more were associated with an increase in odds of the composite perinatal outcome, LGA, and macrosomia.
LEVEL OF EVIDENCE: II
Gestational diabetes mellitus (GDM), defined as glucose intolerance with onset during pregnancy, affects between 2% and 14% of all gravid women.1,2 Pregnancies complicated by GDM are associated with increased perinatal and maternal morbidity. Fetal risks include macrosomia, shoulder dystocia, birth injuries, hypoglycemia, and potential long-term sequelae such as obesity and impaired intellectual achievement.3–5 Maternal risks include preeclampsia, operative delivery, and subsequent diabetes mellitus.6,7
Despite an association with adverse outcomes, optimal screening and diagnostic criteria for GDM, including the use of a one-step or two-step strategy, remain controversial.8,9 In the two-step diagnostic strategy, a 1-hour 50 g glucose challenge test screen cutoff value of 140 mg/dL identifies 80% of women with GDM diagnosed after a 3-hour 100-g oral glucose tolerance test (OGTT), whereas a value of 130 mg/dL identifies 90% of GDM.10 However, the yield of each glucose challenge test cutoff varies with the criteria applied to the diagnostic OGTT. The Fourth International Workshop Conference on Gestational Diabetes Mellitus and the American College of Obstetricians and Gynecologists suggest that either 1-hour 50-g screening threshold is acceptable.1,10 Relating screening glucose challenge test results to actual perinatal outcomes may help identify the optimal threshold for a positive screen result. We conducted this study to estimate the association between glucose challenge test results and perinatal outcomes.
METHODS AND MATERIALS
This cohort study is a secondary analysis of data from a subgroup of women enrolled in a multicenter clinical trial of mild GDM.11 In the primary study, women whose blood glucose concentration was between 135 mg/dL and 199 mg/dL 1 hour after a 50-g glucose challenge test (at 24–30 weeks of gestation) were invited to participate. Eligible women completed a 3-hour OGTT. Mild GDM was defined as fasting glucose less than 95 mg/dL and two or more abnormal timed OGTT values: 1-hour 180 mg/dL or more; 2-hour 155 mg/dL or more; 3-hour 140 mg/dL or more. Women then were randomized to either treatment with nutritional counseling, diet modification, and insulin if required or usual prenatal care. A cohort of women who had positive glucose challenge test results but normal OGTT results and matched one-to-one with the randomized cohort with respect to clinical center, race or ethnicity, and body mass index (BMI, calculated as weight (kg)/[height (m)]2) also was followed as part of the usual prenatal care group to allow blinding. A third group of women with normal glucose challenge test results (less than 120 mg/dL) and matched one-to-one to the untreated group with respect to clinical center, race or ethnicity, and BMI also was enrolled as an observational cohort. The current analysis consists of untreated women with glucose challenge test values of 135–139 mg/dL and 140–199 mg/dL (with untreated mild GDM or no GDM) and those with normal glucose challenge test values of less than 120 mg/dL. Women with mild GDM who received treatment were excluded from this analysis. The glucose challenge test result category of 140–199 mg/dL was further stratified into categories of 10-unit increments for comparisons. The lower glucose challenge test result range was further subdivided by five-unit increments (135–139 mg/dL, 140–144 mg/dL, 145–149 mg/dL). We separated out the 135–139 mg/dL category because 135 mg/dL and 140 mg/dL are used alternatively as screen-positive glucose challenge test result cutoffs in different settings.
Study outcomes included 1) a perinatal composite adverse outcome comprising stillbirth, neonatal death, hypoglycemia, hyperbilirubinemia, neonatal hyperinsulinemia, and birth trauma; 2) large for gestational age (LGA), defined as birth weight above the 90th percentile based on sex-specific and race-specific norms; and 3) macrosomia, defined as birth weight greater than 4,000 g.12 Neonatal mortality was defined as death before hospital discharge or aged 30 days if still hospitalized. Hypoglycemia was defined as glucose less than 35 mg/dL by heel stick within 2 hours of birth and before the first nonbreast-fed feeding. Hyperinsulinemia was defined as C-peptide from cord blood greater than 95th percentile, as determined from another unselected obstetric population of women in the Maternal-Fetal Medicine Units network. Birth trauma was defined as Erb's palsy, clavicular, humerus, or skull fracture.
Categorical variables were analyzed using the χ2 test or Fisher’s exact test. Continuous variables were analyzed using the Wilcoxon rank-sum test or the Kruskal-Wallis test. The incidence of the study outcomes was computed for each result category of glucose challenge test: less than 120 mg/dL, 135–139 mg/dL, 140–199 mg/dL (and subcategories in additional analyses); and logistic regression was used in multivariable analysis to estimate the independent associations between the glucose challenge test result categories of 135–139 mg/dL and 140–199 mg/dL and those less than 120 mg/dL. Potential confounders adjusted for included gestational age at delivery, self-reported race or ethnicity, smoking status, maternal age, baseline BMI, parity, and gestational age at enrollment. Potential confounders were chosen for their known association with gestational diabetes or are frequently adjusted for in other studies. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were computed relative to the glucose challenge test result category less than 120 mg/dL. The Cochran-Armitage test for trend was used to compare the incidence of study outcome by increasing glucose challenge test result categories. The test for trend was based on a linear trend. Where relevant, P<.05 was considered statistically significant; no adjustments were made for multiple comparisons. The primary study was approved by the institutional review board at each participating center and the biostatistical coordinating center. Analyses were performed with SAS statistical software.
A total of 1,889 women were enrolled in the primary study and an additional 436 control participants with glucose challenge test values less than 120 mg/dL were enrolled for comparison; 485 women with mild GDM were assigned to the study treatment and were thus not included in this secondary analysis. Data were available for 436 in the comparison group (glucose challenge test values less than 120 mg/dL) and 1,403 with glucose challenge test values of 135–199 mg/dL and either untreated mild GDM or no GDM (glucose challenge test values of 135–139 mg/dL, n=135; 140–199 mg/dL, n=1,268). The baseline characteristics by glucose challenge test result category are shown in Table 1. Women in different glucose challenge test result categories differed significantly by age, race or ethnicity, smoking status, mean gestational age at randomization, and, as expected, by mean glucose level.
Among five stillbirths in the study cohort, two were observed in the group with glucose challenge test values less than 120 mg/dL, none in the 135–139 mg/dL group, and three in the 140–199 mg/dL group. All 13 neonates with birth trauma were born to women in the group with glucose challenge test values of 140–199 mg/dL. The study outcomes by main glucose challenge test result categories are presented in Table 2. The composite perinatal outcome occurred in 25.6% of those with glucose challenge test values less than 120 mg/dL, compared with 21.1% for values of 135–139 mg/dL and 35.3% for values of 140–199 mg/dL. Rates of LGA by group were 6.6%, 6.8%, and 12.4%, respectively. Rates of macrosomia by group were 7.8%, 6.1%, and 12.1%, respectively. The unadjusted and adjusted ORs (95% CI) relative to glucose challenge test values less than 120 mg/dL category are presented in Table 3. Glucose challenge test result category 140–199 mg/dL, but not 135–139 mg/dL, was associated with an increase (approximately 50%) in the composite adverse outcome. Among the individual components of the composite, the strongest association was with hyperinsulinemia. Similar findings by glucose challenge test result category were noted for LGA and macrosomia (nearly twofold and 1.6-fold association with glucose challenge test values of 140–199 mg/dL, respectively). The results were unchanged after adjusting for differences in baseline characteristics (Table 3).
Results from women with glucose challenge test values in the range of 140–199 mg/dL further stratified by 10-unit increments are presented in Table 4. Rates of the perinatal composite outcome, LGA, and macrosomia increased with increasing glucose challenge test result category (P for trend test<.01 for the primary composite outcome and LGA and P=.03 for macrosomia). The subgroup 160–169 mg/dL shows a nonsignificant adjusted OR for LGA and macrosomia. In general, the unadjusted and adjusted results for glucose challenge test result subcategories within 140–199 mg/dL are consistent with an increase in the perinatal composite (up to 1.6-fold), LGA (up to 2.5-fold), and macrosomia (up to 2.9-fold). Results from the lower end of available glucose challenge test result data stratified by five-unit increments (135–139 mg/dL, 140–144 mg/dL, and 145–149 mg/dL) compared with those with results less than 120 mg/dL are presented in Table 5. The smaller categories at 140 mg/dL and above remained associated with perinatal outcomes.
Perinatal outcomes by presence or absence of GDM are presented in Table 6. The risk of the primary composite outcome, LGA, and macrosomia among those with no GDM and GDM and glucose challenge test values of 135–139 mg/dL was 20.4% and 23.3% (P=.73), 4.9% and 13.3% (P=.12), and 3.9% and 13.3% (P=.08), respectively. The risk of the primary composite outcome, LGA, and macrosomia among those with no GDM and GDM and glucose challenge test values of 140–199 mg/dL was 33.8% and 38.1% (P=.14), 11.2% and 14.7% (P=.08), and 11.0% and 14.4% (P=.08), respectively. However, when all women with glucose challenge test values of 135–199 mg/dL were considered without stratification by glucose challenge test result category, the risk of LGA (14.6% compared with 10.5%, P=.03) and macrosomia (14.4% compared with 10.2%, P=.02), but not the composite outcome (37.1% compared with 32.9%, P=.08), differed significantly for women with GDM compared with those without GDM, respectively.
Overall, the risks of the primary perinatal composite outcome, LGA, and macrosomia increased with increasing glucose challenge test values. Importantly, glucose challenge test values of 135–139 mg/dL were not associated with adverse outcomes compared with values less than 120 mg/dL; however, values of 140 mg/dL and above were associated with a 1.5-fold, 2.0-fold, and 1.6-fold increase in odds of the composite perinatal adverse outcome, LGA, and macrosomia, respectively. Specifically, selected outcomes remained higher in women with glucose challenge test values of 140–144 mg/dL and 145–149 mg/dL compared with values less than 120 mg/dL. As expected, selected outcomes for patients with GDM differed from those without GDM, albeit without statistical significance within glucose challenge test result subcategories (likely resulting from the smaller numbers).
Two retrospective analyses have examined the screening glucose challenge test results in relation to perinatal outcomes. In one cohort study, 176 women with glucose challenge test values of 135 mg/dL or more but normal OGTT results by the Fourth International Workshop-Conference on Gestational Diabetes Mellitus criteria (fasting 95 mg/dL or more, 1-hour 180 mg/dL or more; 2-hour 155 mg/dL or more; 3-hour 140 mg/dL or more) had an incidence of macrosomia of 11.9% compared with an incidence of 6.4% among 1,854 women who had screening glucose challenge test values less than 135 mg/dL (relative risk 1.99, P=.009).13 In the second retrospective cohort study, 164 patients with glucose challenge test values of 135 mg/dL or more but normal OGTT results by the National Diabetes Data Group Criteria (fasting 100 mg/dL or more, 1-hour 190 mg/dL or more, 2-hour 165 mg/dL or more, and 3-hour 145 mg/dL or more) had increased risk (OR 5.96, 95% CI 1.3–10.32) of adverse perinatal composite outcome including fetal macrosomia, antenatal death, shoulder dystocia, chorioamnionitis, preeclampsia, intensive care nursery admission, and postpartum endometritis when compared with 1,661 patients with a normal glucose challenge test values less than 135 mg/dL.14 These two studies, however, did not define subcategories of glucose challenge test results for comparison to assess alternative screen-positive cutoffs.13,14 The results of our study involving women with milder GDM are consistent but further suggest that the increased risk associated with glucose challenge test result may be limited to those with values 140 mg/dL and above. Importantly, this secondary analysis is a unique opportunity to observe the natural progression of perinatal outcomes in untreated women with and without mild GDM.
Our study is limited by the lack of pregnant women with glucose challenge test values between 121–134 mg/dL and 200 mg/dL or more. This does not allow for evaluation of the continuous glucose challenge test values in the upper or lower ranges. This study also included a select group of pregnant women with mild GDM because those women with elevated fasting glucose (greater than 95 mg/dL) were excluded. These women likely have more severe GDM that may confer a greater risk for adverse perinatal outcomes. Additionally, these cases may be more likely to occur among those with higher glucose challenge test values, thereby strengthening the associations we report. Inclusion of such patients with the full spectrum of GDM in future studies may help validate and better quantify the strength of the association between perinatal outcomes and different subcategories of glucose challenge test values. This study also lacks long-term data on the risk of GDM. Future research on long-term outcomes such as childhood obesity and impaired intellectual achievements with different glucose challenge test value cutoffs may be useful. The adjusted OR for glucose challenge test result categories greater than 160 mg/dL (Table 4) lacked statistical significance. This may be explained by the low number of patients in these groups and corresponding low power for statistical comparisons. However, the Cochran-Armitage test of trend revealed an overall increase in both the primary composite outcome and LGA with increasing glucose challenge test value. The results are also limited by the use of a composite outcome whose individual outcomes have varying clinical significance. This is a common critique of composite outcomes but this composite was also used in the primary study and we provide additional data on other outcomes. Finally, we may not have sufficient power to clearly delineate differences between the glucose challenge test values of 135–139 mg/dL category and the normal category (less than 120 mg/dL). The point estimates, however, suggest that glucose challenge test values of 135–139 mg/dL are unlikely to be associated with increased adverse outcomes.
Screening and diagnostic criteria for GDM constitute a rapidly evolving topic. Whereas the two-step process (1-hour 50-g load and 3-hour 100-g OGTT) currently is applied in the United States, the World Health Organization and several European countries have advocated the use of the single-step 2-hour 75-g OGTT for the diagnosis of GDM.1,15–17 In January 2011, the American Diabetes Association, in response to the proposal of the International Association of Diabetes and Pregnancy Study Groups, adopted new guidelines for the diagnosis of GDM.18,19 These guidelines advocate use of the 2-hour 75-g OGTT without a preceding screening test, but they have not been adopted by the American College of Obstetricians and Gynecologists. When the two-step strategy is used, the optimal screen-positive cutoff remains controversial and values of 130 mg/dL, 135 mg/dL, and 140 mg/dL have been variously advocated based solely on test characteristics for predicting a diagnosis of GDM using the 3-hour OGTT as the gold standard.1,10 Therefore, our prospectively collected data from a well-characterized trial cohort, including untreated pregnant women with mild GDM, add an important dimension: information about the ultimate effect of these cutoffs on pregnancy outcomes. These data suggest that glucose challenge test values less than 139 mg/dL may not be associated with immediate adverse pregnancy outcomes. A positive screen threshold of 140 mg/dL or more may be reasonable and perhaps more cost-effective than lower thresholds. A related analysis of our source cohort also suggests that glucose challenge test values of 135–142 mg/dL also carry an equivalent risk of GDM.20 Therefore, in settings where the two-step diagnosis of GDM remains prevalent, larger studies involving the full range of glucose challenge test categories are needed to validate our findings. Care must be taken to consider the attenuating effects of treatment for GDM in such studies.
1. Gestational diabetes. ACOG Practice Bulletin No. 30. Obstet Gynecol 2001;98:525–38.
2. Coustan DR. Gestational diabetes. In: National Institutes of Diabetes and Digestive and Kidney Diseases. Diabetes in America. 2nd ed. Bethesda (MD): NIDDK; 1995. NIH Publication No. 95-1468:703–717.
3. Pettitt D, Bennett PH, Knowler WC, Baird HR, Aleck KA. Gestational diabetes mellitus and impaired glucose tolerance during pregnancy: long-term effects on obesity and glucose intolerance in the offspring. Diabetes Care 1985;34(suppl 2):119–22.
4. Blank A, Grave G, Metzger BE. Effects of gestational diabetes on perinatal morbidity reassessed: report of the International Workshop on Adverse Perinatal Outcomes of Gestational Diabetes Mellitus, December 3-4, 1992. Diabetes Care 1995;18:127–9.
5. Crowther CA, Hiller JE, Moss JR, McPhee AJ, Jeffries WS, Robinson JS. Effect of treatment of gestational diabetes mellitus on pregnancy outcomes. N Engl J Med 2005;352:2477–86.
6. Sermer M, Naylor CD, Gare DJ, Kenshole AB, Ritchie JW, Farine D, et al.. Impact of increasing carbohydrate intolerance on maternal-fetal outcomes in 3637 women without gestational diabetes. The Toronto Tri-Hospital Gestational Diabetes Project. Am J Obstet Gynecol 1995;173:146–56.
7. Dornhorst A, Rossi M. Risk and prevention of type 2 diabetes in women with gestational diabetes. Diabetes Care 1998;21(suppl):B43–9.
8. Brody SE, Harris R, Lohr K. Screening for gestational diabetes: a summary of the evidence for the U.S. Preventive Services Task Force. Obstet Gynecol 2003;101:380–92.
9. Tuffnell DJ, West J, Walkinshaw SA. Treatments for gestational diabetes and impaired glucose tolerance in pregnancy. The Cochrane Database of Systematic Reviews 2005, Issue 3. Art. No.: CD003395. DOI: 10.1002/14651858.CD003395.
10. Metzger BE, Coustan DR. Summary and recommendations of the Fourth International Workshop-Conference on Gestational Diabetes Mellitus. Diabetes Care 1998;21(suppl):B161–7.
11. Landon MB, Spong CY, Thom E, Carpenter MW, Ramin SM, Casey B, et al.. A multicenter randomized trial of treatment for mild gestational diabetes. N Engl J Med 2009;361:1339–48.
12. Alexander GR, Kogan MD, Himes JH. 1994-1996 U.S. singleton birth weight percentiles for gestational age by race, Hispanic origin, and gender. Matern Child Health J 1999;3:225–31.
13. Leikin EL, Jenkins JH, Pomerantz GA, Klein L. Abnormal glucose screening tests in pregnancy: a risk factor for fetal macrosomia. Obstet Gynecol 1987;69:570–3.
14. Stamilio DM, Olsen T, Ratcliffe S, Sehdev HM, Macones GA. False-positive 1-hour glucose challenge test and adverse perinatal outcomes. Obstet Gynecol 2004;103:148–56.
15. World Health Organization. Definition, diagnosis, and classification of diabetes mellitus and its complications: report of a WHO consultation. Geneva (Switzerland): WHO Document Production Services; 1999:19–20.
16. Per Kautsky-Willer A, Bancer-Todesca D, Birnbacher R. Gestational diabetes mellitus [in German]. Acta Med Austriaca 2004;31:182–4.
17. Guideline Development Group. Management of diabetes from preconception to the postnatal period: summary of NICE guidance. BMJ 2008;336:714–7.
18. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, et al.. International Association of Diabetes and Pregnancy Study Groups Consensus Panel, International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010;33:676–82.
19. American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2011;34(suppl 1):S62–9.
20. Figueroa D, Landon M, Lai Y, Spong C, Ramin S, Casey B, et al.. Predictive characteristics of elevated 1-hour glucose loading test results for gestational diabetes. Am J Obstet Gynecol 2011;204:S106–7.
Supplemental Digital Content
© 2013 The American College of Obstetricians and Gynecologists