The Continuing Controversy Over Screening for Gestational Diabetes - PowerPoint PPT Presentation

6/5/2014 The Continuing Controversy Over Screening for Gestational Diabetes I have nothing to disclose. Kirsten E. Salmeen, MD Assistant Professor Obstetrics, Gynecology & Reproductive Sciences Maternal-Fetal Medicine GDM & Controversy The Continuing Controversy Over Screening for Gestational Diabetes • The nature of screening tests • Why screening for GDM matters • The major controversies • Possible sources of those controversies • What I think you should do 1

6/5/2014 The Nature of Screening Tests • Screening is the identification of an asymptomatic disease, harmful condition or risk factor. How great is the burden of suffering • When deciding how to screen, the following must be caused by GDM? considered: - Burden of suffering caused by the condition - Therapeutic interventions available - Performance of available screening tests Fletcher et al. Clinical Epidemiology: The Essentials, 5 th Ed, Lippincott Williams & Wilkins 2013 Why should we be concerned with GDM at all? Overall % RR/OR Macrosomia 20 RR ~1.4 Blinded study of ~25,000 women at 15 centers, 9 countries Pre-Eclampsia 15 RR ~1.7 Primary predictor: Levels of hyperglycemia Cesarean Section Varies RR ~ 1.2 Shoulder Dystocia 3-5 OR ~ 1.2 Primary outcomes: Birth weight > 90%ile, primary CD, IUFD ~ 0.05 RR ~ 2 neonatal hypoglycemia, cord-blood C-peptide level HAPO Study Cooperative Research Group. N Engl J Med. 2008;358(19):1991-2002. Schmidt M et al. Diabetes Care. 2001;24(7):1151-5. Wendland E et al. BMC Pregnancy Childbirth. 2012;31(12):23-36. HAPO Study Cooperative Research Group. N Engl J Med. 2008;358(19):1991-2002. 2

6/5/2014 HAPO Results Increasing maternal glycemia is associated with increased risk of maternal and fetal complications. How good is the therapeutic intervention for GDM? HAPO Study Cooperative Research Group. N Engl J Med. 2008;358(19):1991-2002. Crowther – Trial of Treatment for GDM Landon – Trial of Treatment for GDM Intervention Routine Adjusted RR or Adjusted Intervention Control Group Group N= 490 Care N= 510 Treatment Effect p-value Group N = 485 N = 473 Relative Risk p-value (%) (%) (%) (%) *Any serious perinatal 1 4 0.33 (0.14 – 0.75) 0.01 NICU Admission 9 11.6 0.77 (0.51 – 1.18) 0.19 complication Admission to NICU 71 61 1.13 (1.03 – 1.23) 0.04 Macrosomia 5.9 14.3 0.41 (0.26 – 0.66) < 0.001 Neonatal Macrosomia 10 21 0.47 (0.34 – 0.64) < 0.001 5.3 6.8 0.77 (0.44 – 1.36) 0.32 Hypoglycemia Neonatal hypoglycemia 7 5 1.42 (0.87 – 2.32) 0.16 Shoulder Dystocia 1.5 4.0 0.37 (0.14 – 0.97) 0.02 Cesarean Delivery 26.9 33.8 0.79 (0.64 – 0.99) 0.02 Preeclampsia 12 18 0.7 (0.51 – 0.95) 0.02 Preeclampsia or 8.6 13.6 0.63 (0.42 – 0.96) 0.01 Cesarean Delivery 31 32 0.97 (0.81 – 1.16) 0.73 GHTN * One or more of: death, shoulder dystocia, bone fracture, nerve palsy Landon et al. N Eng J Med. 2009;361:1339-48. Crowther et al. N Engl J Med. 2005;352:2477-86. 3

6/5/2014 How good are the screening Increasing maternal glycemia is associated with worse perinatal outcomes. tests for GDM? Treatment improves outcomes. (How good is too good?) What’s the controversy?! GDM Controversies GDM Controversies One-Step Testing v. Two-Step Testing One-Step Testing v. Two-Step Testing Carpenter Coustan v. National Diabetes Data Group Carpenter Coustan v. National Diabetes Data Group Universal Screening v. Risk-Based Screening Universal Screening v. Risk-Based Screening Early Screening v. 24-28 Week Screening Early Screening v. 24-28 Week Screening Hemoglobin A1c v. No Hemoglobin A1c Hemoglobin A1c v. No Hemoglobin A1c Blood sugar testing for 1 abnormal value v. No testing Blood sugar testing for 1 abnormal value v. No testing 4

6/5/2014 One-Step vs. Two-Step Testing GDM Controversies Two-Step One-Step Testing v. Two-Step Testing One-Step Step 1: Carpenter Coustan v. National Diabetes Data Group Non-Fasting, 50 g, 1 hr serum Fasting, 75 g, 1 & 2 hr serum Universal Screening v. Risk-Based Screening glucose measurement glucose measurement ≥ 130/140 mg/dL � Step 2 Early Screening v. 24-28 Week Screening 1+ abnormal value � GDM Step 2: Hemoglobin A1c v. No Hemoglobin A1c Fasting, 100 g, 3 hr glucose test GDM prevalence ~ 20% 2+ abnormal values � GDM Blood sugar testing for 1 abnormal value v. No testing GDM prevalence ~ 5-10% GDM Controversies Carpenter-Coustan v. NDDG One-Step Testing v. Two-Step Testing Fasting 1 hr 2 hr 3 hr GDM Carpenter Coustan v. National Diabetes Data Group (mg/dL) (mg/dL) (mg/dL) (mg/dL) Prevalence Universal Screening v. Risk-Based Screening National Diabetes Data 105 190 165 145 3-4% Early Screening v. 24-28 Week Screening Group Carpenter- Hemoglobin A1c v. No Hemoglobin A1c Coustan 95 180 155 140 5-7% Criteria Blood sugar testing for 1 abnormal value v. No testing 5

6/5/2014 Universal vs. Risk-Based Screening Universal vs. Risk-Based Screening 1 st - 3 rd International Workshop on GDM (1979, 1984, 1990): Universal Screening 4 th & 5 th International Workshop on GDM: (1997 & 2005): Risk-Based Screening “All pregnant patients should be screened for GDM, whether by the patient’s medical history, clinical risk factors, or laboratory screening test results to determine blood glucose levels.” Added in 5 th Workshop Metzger et al. Diabetes 1991(40) Suppl 2: 197-201. Metzger et al. Diabetes Care 2007(30);Suppl 2:S251-260. Universal vs. Risk-Based Screening GDM Controversies One-Step Testing v. Two-Step Testing Carpenter Coustan v. National Diabetes Data Group January 2014 Universal Screening v. Risk-Based Screening “[There is] adequate evidence that screening for and treatment of GDM can significantly reduce the risk for Early Screening v. 24-28 Week Screening preeclampsia, fetal macrosomia, and shoulder dystocia…as a result of the evidence… Hemoglobin A1c v. No Hemoglobin A1c Blood sugar testing for 1 abnormal value v. No testing The USPSTF recommends screening for gestational diabetes mellitus in asymptomatic pregnant women after 24 weeks of gestation (B recommendation).” http://www.uspreventiveservicestaskforce.org/uspstf13/gdm/gdmfinalrs.htm 6

6/5/2014 Early Screening • Detecting women with pre-existing diabetes or glucose intolerance (pre-diabetes) January 2014 • ACOG: History of GDM, known impaired “The USPSTF concludes that the current glucose metabolism, obesity evidence is insufficient to assess the balance of benefits and harms of screening for GDM in • ADA: Severe obesity, strong family history, asymptomatic pregnant women before 24 personal history of GDM, impaired glucose weeks of gestation.” metabolism, glucosuria http://www.uspreventiveservicestaskforce.org/uspstf13/gdm/gdmfinalrs.htm Hemoglobin A1c GDM Controversies One-Step Testing v. Two-Step Testing A1c ≥ 6.5 � DM2 Carpenter Coustan v. National Diabetes Data Group A1c 5.7 – 6.5 � Glucose Intolerance Diagnosing Type 2 DM: A1c < 5.7 � Normal Universal Screening v. Risk-Based Screening Early Screening v. 24-28 Week Screening Average HbA1c Values Non-Diabetic Women 2 nd 3 rd Non- Hemoglobin A1c v. No Hemoglobin A1c 1 st Trimester Pregnant Trimester Trimester Blood sugar testing for 1 abnormal value v. No testing 4.8 – 5.5 4.3 – 5.4 4.4 – 5.4 4.7 – 5.7 HbA1c % (5.2) (5.0) (4.9) (5.1) http://www.diabetes.org/diabetes-basics/diagnosis/?loc=DropDownDB-diagnosis O’Connor et al. Clin Chem Lab Med 2012;50(5):905-9. 7

6/5/2014 GDM Controversies Pregnancy Outcomes for Women with One-Step Testing v. Two-Step Testing 1 Abnormal Value on 3 hour Carpenter Coustan v. National Diabetes Data Group Universal Screening v. Risk-Based Screening Early Screening v. 24-28 Week Screening Hemoglobin A1c v. No Hemoglobin A1c Blood sugar testing for 1 abnormal value v. No testing McLaughlin et al AJOG 2006;194:e16-19. Treatment for Patients With 1 Abnormal Value Sources of Controversy Fassett et al. AJOG 2007;196:597.e1-597.e4 8

6/5/2014 Sensitivity v Specificity What constitutes disease ? One-Step Two-Step Carpenter Coustan National Diabetes Data Group Universal Screening Risk-Based Screening Early Screening 24-28 Week Screening Hemoglobin A1c No Hemoglobin A1c Testing for 1 abnormal value No f/u for 1 abnormal value Dichotomization of a continuous process is More Sensitive, Less Specific Less Sensitive, More Specific bound to result in disagreement More women with disease test positive Fewer women with disease test positive More women WITHOUT disease test positive Fewer women WITHOUT disease test positive Diagnosing women who might not actually Missing a clinically important diagnosis have clinically important disease What primary cesarean section rate defines a bad outcome from disease ? Who Decides? HAPO Study Cooperative Research Group. N Engl J Med. 2008;358(19):1991-2002. 9

6/5/2014 Lack of unambiguous evidence that aggressive diagnosis improves clinically important pregnancy outcomes • The Landon study included women with 2 abnormal values on a 3-hour “Parachutes reduce the risk of injury after • Studies of treatment are within the gravitational challenge, but confines of strict clinical trials their effectiveness has not been proved with • No study has compared outcomes randomised controlled between women who rule-in by 1-step approach but rule out by 2-step trials.” approach Differences in Perceived Goals of Testing Worry about the over-medicalization of pregnancy and increased anxiety about diagnosis 10