5.3.3.0 Two-‍Step Strat­e­gy

In 2013, the Na­tion­al In­sti­tutes of Health (NIH) con­vened a con­sen­sus de­vel­op­ment con­fer­ence to con­sid­er di­ag­nos­tic cri­te­ria for di­ag­nos­ing GDM (81). The 15-‍mem­ber panel had repre­sentatives from obstetrics/‍gynecology, ma­ter­nal-‍fetal medicine, pe­di­atrics, di­a­betes re­search, bio­statis­tics, and other re­lat­ed fields. The panel rec­om­mend­ed a two-‍step ap­proach to screen­ing that used a 1-h 50-g glu­cose load test (GLT) fol­lowed by a 3-h 100-g OGTT for those who screened pos­i­tive. The Amer­i­can Col­lege of Ob­ste­tri­cians and Gy­ne­col­o­gists (ACOG) rec­om­mends any of the com­monly used thresh­olds of 130, 135, or 140 mg/dL for the 1-h 50-g GLT (82). A sys­tematic re­view for the U.S. Pre­ven­tive Ser­vices Task Force com­pared GLT cut-‍offs of 130 mg/dL (7.2 mmol/‍L) and 140 mg/dL (7.8 mmol/‍L) (83). The high­er cut­off yield­ed sen­si­tiv­i­ty of 70–88% and specificity of 69–89%, while the lower cut­off was 88–99% sen­si­tive and 66– 77% specific. Data re­gard­ing a cut­off of 135 mg/dL are lim­it­ed. As for other screen­ing tests, choice of a cut­off is based upon the trade-‍off be­tween sen­si­tiv­i­ty and specificity. The use of A1C at 24–28 weeks of ges­ta­tion as a screen­ing test for GDM does not func­tion as well as the GLT (84).

Key fac­tors cited by the NIH panel in their de­ci­sion-‍mak­ing pro­cess were the lack of clin­i­cal trial data demon­strat­ing the benefits of the one-‍step strat­e­gy and the po­ten­tial neg­a­tive conse­quences of iden­tifying a large group of women with GDM, in­clud­ing med­i­calization of preg­nan­cy with in­creased health care uti­liza­tion and costs. More­over, screen­ing with a 50-g GLT does not re­quire fast­ing and is there­fore eas­i­er to ac­com­plish for many women. Treat­ment of high­er-‍thresh­old ma­ter­nal hy­per­glycemia, as iden­tified by the two-‍step ap­proach, re­duces rates of neona­tal macro­so­mia, large-for-ges­ta­tional-age births (85), and shoul­der dys­to­cia, with­out in­creas­ing small-for-ges­ta­tional-age births. ACOG cur­rently sup­ports the two-‍step ap­proach but notes that one el­e­vat­ed value, as op­posed to two, may be used for the di­ag­no­sis of GDM (82). If this ap­proach is im­ple­ment­ed, the in­ci­dence of GDM by the two-‍step strat­e­gy will like­ly in­crease marked­ly. ACOG rec­om­mends ei­ther of two sets of di­ag­nos­tic thresh­olds for the 3-h 100-g OGTT (86,87). Each is based on dif­ferent math­e­mat­i­cal con­ver­sions of the orig­i­nal rec­om­mend­ed thresh­olds, which used whole blood and nonen­zy­mat­ic meth­ods for glu­cose de­ter­mi­na­tion. A sec­ondary anal­y­sis of data from a ran­domized clin­i­cal trial of iden­tification and treat­ment of mild GDM (88) demon­strat­ed that treat­ment was sim­i­larly beneficial in pa­tients meet­ing only the lower thresh­olds (86) and in those meet­ing only the high­er thresh­olds (87). If the two-‍step ap­proach is used, it would ap­pear ad­van­ta­geous to use the lower di­ag­nos­tic thresh­olds as shown in step 2 in Table 2.6.