Materials and Methods
Study population
This was a retrospective cohort study involving all pregnant women who underwent screening for GDM at St Michael’s Hospital in Toronto, ON, Canada, a tertiary university affiliated medical center, between January 2008 and December 2011. The study protocol received approval from the institutional research ethics board (number 12-177 c ).
Women with multiple gestations, cases complicated by major anomalies, women who did not deliver in our medical center, women who underwent a 75 g OGTT in the presence of a negative 50 g glucose challenge test (GCT), and women who were diagnosed with GDM solely on the basis of an abnormal 50 g GCT were excluded from the study ( Figure 2 ).
Definitions
During the study period, the GDM screening and diagnosis strategy in our institution followed the CDA 2008 guidelines, which are based on a 2-step screening approach ( Figure 1 ). The first step involves universal screening at 24-28 weeks of gestational age using the 1 hour, 50 g GCT. Results are interpreted as screen negative (result <140 mg/dL; <7.8 mmol/L), screen positive (result between 140 and 184 mg/dL; 7.8-10.2 mmol/L), or diagnostic for GDM (result ≥185 mg/dL; ≥10.3 mmol/L). Of note, the new CDA 2013 guidelines raised the diagnostic threshold of the 50 g GCT to 200 mg/dL (11.1 mmol/L).
Patients screening positive undergo a fasting state 2 hour, 75 g OGTT with fasting, with 1 hour and 2 hour threshold values of 95 mg/dL (5.3 mmol/L), 191 mg/dL (10.6 mmol/L), and 160 mg/dL (8.9 mmol/L), respectively ( Figure 1 ). The diagnostic 75 g OGTT is interpreted as positive for GDM if 2 or more values exceed their corresponding thresholds. Patients with only 1 abnormal value are diagnosed with impaired glucose tolerance (IGT) and were managed similarly to GDM patients in our medical center. The 2013 CDA guidelines endorse 1 abnormal value as diagnostic of GDM.
In addition, in accordance with the CDA guidelines at the time of the study, women with multiple risk factors could be screened using a 50 g GCT or undergo direct testing using the 75 g OGTT earlier in the pregnancy to decrease the risk of false-negative results in high-risk individuals.
According to the IADPSG recommendations, the screening and diagnosis of GDM should be based on a 1-step strategy using the fasting-state 75 g OGTT with the fasting, 1 hour and 2 hour threshold values of 92 mg/dL (5.1 mmol/L), 180 mg/dL (10.0 mmol/L), and 153 mg/dL (8.5 mmol/L), respectively ( Figure 1 ). The test is interpreted as positive for GDM if 1 or more values exceed their corresponding thresholds.
All women diagnosed with IGT or GDM in our medical center are followed up in a multidisciplinary clinic that includes an endocrinologist experienced in the management of diabetes, an obstetrician experienced in the management of high-risk pregnancies, a diabetic nurse, and a nutritional consultant experienced in the management of diabetic patients. Patients attend the clinic from the time of diagnosis until the time of delivery on a weekly or biweekly basis.
All patients receive education regarding a glucometer use, and they record glucose values 4 times daily (fasting and 2 hours following each meal). Patients failing to meet the target glucose values (fasting <95 mg/dL; <5.3 mmol/L and 2 hours postprandially <121 mg/dL; <6.7 mmol/L) after 1-2 weeks of diet management are treated with insulin. Oral hypoglycemic agents (eg, glyburide and metformin) are not used routinely in our clinic for the management of GDM. Labor induction at 38-40 weeks of gestation is considered on the basis of clinical factors such as the degree of glycemic control and the estimated fetal weight.
Macrosomia is defined as a birthweight greater than 4000 g, and large for gestational age (LGA) is defined as a birthweight greater than the 90th percentile for gestational age, based on national growth curves. Respiratory morbidity was defined as the diagnosis of respiratory distress syndrome, transient tachypnea of the newborn, or a need for respiratory support (intubation of continuous positive airway pressure).
Composite adverse outcome was defined as the presence of one of the following conditions: hypertensive complications of pregnancy (ie, gestational hypertension or preeclampsia), shoulder dystocia, third- to fourth-degree perineal tears, LGA neonates, admission to the neonatal intensive care unit (NICU), neonatal respiratory morbidity, neonatal hypoglycemia, or neonatal jaundice. We believed including a composite outcome containing both maternal and neonatal adverse outcome measures would serve to better quantify the overall risk of adverse outcome in each of the groups.
Data collection
Data were extracted from the institutional electronic databases. Initially all women undergoing a 50 g GCT and/or a 75 g OGTT were identified using the laboratory database, and the dates and results of these tests were extracted. For women having had more than 1 pregnancy in the study period, only the first pregnancy was included. Next, we cross-linked this database with the institutional comprehensive perinatal database, from which the following information was extracted: maternal demographic, obstetric, and medical history; complications during the current pregnancy; gestational age at delivery; the need for labor induction; the mode of delivery, labor and delivery complications; neonatal birthweight; and the short-term neonatal outcome. Information regarding chronic hypertension, a previous diagnosis of GDM or hypertensive complications, type of glycemic management, maternal weight, body mass index (BMI), and ethnicity were not available.
Data analysis
All women undergoing a 75 g OGTT were classified into 3 groups based on whether they met the CDA 2008 and the IADPSG criteria: the OGTT-NEGATIVE group, the GDM-IADPSG group, and the GDM-CDA group ( Figure 2 ). Pregnancy outcomes were compared between these 3 groups and a control group consisting of women with a negative 50 g GCT (GCT-NEGATIVE group).
The OGTT-NEGATIVE group included women whose 75 g OGTT results were below IADPSG and CDA 2008 thresholds and therefore considered negative for GDM. The GDM-CDA women included those with IGT or GDM based on the CDA 2008 criteria, meaning 1 or more values were above the threshold in the 75 g OGTT ( Figure 1 ). All the GDM-CDA women were followed up in our diabetes-in-pregnancy clinic and managed similarly, regardless of an IGT or GDM diagnosis, as described in the previous text. The GDM-IADPSG group included women with 75 g OGTT values below the CDA 2008 thresholds but who had at least 1 value exceeding the IADPSG thresholds ( Figures 1 and 2 ). During the study period, this group was managed in the same manner as the OGTT-NEGATIVE and GCT-NEGATIVE groups.
Statistical analysis
Statistical analysis was performed using the SPSS version 21 software (SPSS Inc, Chicago, IL). The χ2 and the independent-sample Student t test were used for the comparison of categorical and continuous variables, respectively. A multivariable logistic regression analysis was used to assess the association between the degree of glucose intolerance and the risk of adverse outcome while adjusting for potential confounders.
Based on previous analyses in our institution, we estimated the rate of composite adverse outcome in a nondiabetic population in our institution to be approximately 15%. In addition, preliminary analysis of our data suggested the adoption of the IADPSG criteria in our institution would increase the proportion of women diagnosed with GDM by approximately 3%. Based on this assumption, we calculated that for a type I error of 5% and type II error of 20%, the minimum sample size required to detect a 50% increase in the rate of composite adverse outcome in women positive only for the IADPSG criteria compared with women with negative GCT would be approximately 3500 women.
A value of P < .05 or 95% confidence interval not crossing 1.0 was considered statistically significant.
Results
Effect of the IADPSG criteria on the rate of GDM
Overall, 6074 women underwent screening for GDM during the study period, of whom 5249 were eligible for the study ( Figure 2 ). The details of each of the study groups are presented in Figure 2 .
Overall, 4183 women had a negative 50 g GCT (GCT-NEGATIVE group). Of the 1066 women who were 50 g GCT positive and underwent a 75 g OGTT, 526 (49.3%) had no abnormal values (OGTT-NEGATIVE group), 155 (14.5%) had 1 or more abnormal values according only to the IADPSG criteria but not according to the CDA 2008 criteria (OGTT-IADPSG group), and 385 (36.1%) had 1 or more abnormal values according to the CDA 2008 criteria (OGTT-CDA group). Of the latter group, 219 women (20.5%) had only 1 abnormal value (defined as IGT according to the CDA 2008 criteria ) and 166 (15.6%) had 2 or more abnormal values (defined as GDM according to the CDA 2008 criteria ).
The use of the IADPSG criteria compared with the CDA 2008 criteria would increase the rate of GDM from 7.3% (385 of 5249) to 10.3% ([155+385] of 5249) (ie, absolute increase of 3.0%, or relative increase of 41%) ( Table 1 ). When considering only the rate of true GDM according to the CDA 2008 criteria (ie, ≥2 abnormal values), the use of the IADPSG criteria would result in an absolute increase of 7.1% in the rate of GDM (ie, from 3.2% to 10.3%), which represents a relative increase of 222% ( Table 1 ).
| Criteria | GDM rate |
|---|---|
| CDA 2008 | |
| GDM overall (≥1 abnormal values, GDM plus IGT) | 7.3% (385/5249) |
| GDM (≥2 abnormal values) | 3.2% (166/5249) |
| IGT (1 abnormal value) | 4.1% (219/5249) |
| IADPSG | |
| GDM overall (≥1 abnormal values) | 10.3% ([155+385]/5249) |
| Proportion attributed to the lower threshold values only (≥2 abnormal values) | 5.0% (263/5249) |
| Proportion attributed to using only ≥1 abnormal values to define GDM | 5.3% (277/5249) |
We next assessed the relative impact of using the lower IADPSG thresholds alone and using only 1 abnormal value for diagnosis on the rate of GDM in the study population. For this purpose we compared the proportion of women with 2 or more abnormal values according to the IADPSG criteria (5.0%; 263 of 5249) with that of women with 2 or more abnormal values according to the CDA 2008 criteria (3.2%; 166 of 5249). This implies that the lower IADPSG thresholds alone would increase the rate of true GDM from 3.2% to 5.0%, a relative increase of 56%. The remainder of the increase in the rate of GDM according to the IADPSG criteria (from 5.0% to 10.3%), a 106% increase, is attributable to the use of a single abnormal value to define GDM ( Table 1 ).
We also calculated the contribution of each of the new IADPSG thresholds to the increase in the rate of GDM ( Figure 3 ). The rate of an abnormal fasting value would increase by 7.4% (50% relative increase) (22.2% vs 14.8%), the rate of an abnormal 1 hour value would increase by 12.6% (56.5% relative increase) (34.9% vs 22.3%), and the rate of an abnormal 2 hour value would increase by 7.0% (36.3% relative increase) (26.7% vs 19.3%). The lower 1 hour threshold was the most significant contributor to the higher rate of GDM according to the IADPSG criteria ( Figure 3 ).
Effect of the IADPSG criteria on pregnancy outcomes
To determine whether the use of the IADPSG criteria for GDM diagnosis would identify additional women at risk of adverse pregnancy outcomes, we analyzed the perinatal outcomes of women diagnosed as GDM based on the IADPSG criteria alone (OGTT-IADPSG group; Figure 2 ) and excluded those diagnosed by the CDA 2008 criteria only. The outcomes of this group were compared with our control group that had a negative 50 g GCT (GCT-NEGATIVE group). To determine whether perinatal outcome had a continuous relationship with the degree of glucose intolerance, we also compared the outcome of women with a negative 75 g OGTT (OGTT-NEGATIVE group) and a positive 75 g OGTT according to the CDA 2008 criteria (OGTT-CDA group) with the same control group (GCT-NEGATIVE group; Figure 2 ).
The characteristics of the study groups are presented in Table 2 . Women in the GCT-NEGATIVE group were significantly younger and had a significantly lower 50 g GCT result compared with women in the other groups.
| Characteristic | GCT-NEGATIVE group (n = 4183) | OGTT-NEGATIVE group (n = 526) | OGTT-IADPSG group (n = 155) (no treatment) | OGTT-CDA group (n = 385) (treated as GDM) | |||
|---|---|---|---|---|---|---|---|
| Value | Value | P value a | Value | P value a | Value | P value a | |
| Maternal age, y | 31.5 ± 5.4 | 32.7 ± 4.9 | < .001 b | 33.6 ± 4.7 | < .001 b | 33.9 ± 5.2 | < .001 b |
| >35 | 975 (23.3) | 153 (29.1) | .003 b | 52 (33.5) | .003 b | 148 (38.4) | < .001 b |
| Male fetus | 2080 (50.3) | 238 (46.5) | .5 | 84 (56.0) | .2 | 181 (51.4) | .7 |
| GCT value, mg/dL (mmol/L) | 108 ± 18 (6.0 ± 1.0) | 155 ± 11 (8.6 ± 0.6) | < .001 b | 157 ± 11 (8.7 ± 0.6) | < .001 b | 162 ± 16 (9.0 ± 0.9) | < .001 b |
| OGTT values | N/A | N/A | N/A | N/A | |||
| Fasting, mg/dL (mmol/L) c | 81 ± 5.4 (4.5 ± 0.3) | 85 ± 7.2 (4.7 ± 0.4) | 92 ± 13 (5.1 ± 0.7) | ||||
| 1 hour, mg/dL (mmol/L) c | 149 ± 20 (8.3 ± 1.1) | 175 ± 14.4 (9.7 ± 0.8) | 194 ± 27 (10.8 ± 1.5) | ||||
| 2 hour, mg/dL (mmol/L) c | 124 ± 20 (6.9 ± 1.1) | 140 ± 18 (7.8 ± 1.0) | 158 ± 31 (8.8 ± 1.7) | ||||
a Refers to comparison with the GCT-NEGATIVE group
b Statistically significant at a P value < .05
c The fasting, 1 hour, and 2 hour mean GTT results were significantly different between the 3 groups ( P < .001).
Table 3 presents pregnancy outcomes for the control and study groups. The rate of adverse pregnancy outcome in women with a positive 50 g GCT but a negative 75 g OGTT (OGTT-NEGATIVE group) was not significantly different from that of the control group (GCT-NEGATIVE group).
| Outcome | GCT-NEGATIVE group (n = 4183) | OGTT-NEGATIVE group (n = 526) | OGTT-IADPSG group (n = 155) (no treatment) | OGTT-CDA group (n = 385) (treated as GDM) | |||
|---|---|---|---|---|---|---|---|
| Value | Value | P value a | Value | P value a | Value | P value a | |
| Composite adverse outcome | 1095 (26.2) | 139 (26.4) | .90 | 52 (33.5) | .04 b | 136 (35.3) | < .001 b |
| Induction of labor | 461 (11.0) | 53 (10.1) | .50 | 17 (11.0) | .90 | 79 (20.5) | < .001 b |
| Hypertensive disorders c | 124 (3.0) | 16 (3.0) | .90 | 13 (8.4) | < .001 b | 17 (4.4) | .10 |
| GHTN | 91 (2.2) | 14 (2.7) | .50 | 9 (5.8) | .003 b | 14 (3.6) | .07 |
| Preeclampsia | 33 (0.8) | 3 (0.6) | .60 | 4 (2.6) | .02 b | 3 (0.8) | .90 |
| GA at delivery, wks | 39.0 ± 1.5 | 38.9 ± 1.5 | .03 | 38.7 ± 1.3 | .007 b | 38.5 ± 1.7 | < .001 b |
| <37 | 209 (5.0) | 31 (5.9) | .40 | 8 (5.2) | .90 | 30 (7.8) | .02 b |
| Mode of delivery | |||||||
| OVD | 381 (9.1) | 46 (8.7) | .80 | 11 (7.1) | .40 | 34 (8.8) | .90 |
| CD | 1093 (26.1) | 154 (29.3) | .10 | 57 (36.8) | .003 b | 131 (34.0) | .001 b |
| Prolonged maternal stay | 859 (20.5) | 102 (19.4) | .50 | 27 (17.4) | .30 | 99 (25.7) | .02 b |
| PPH | 64 (1.5) | 7 (1.3) | .70 | 3 (1.9) | .70 | 4 (1.0) | .80 |
| Third/fourth-degree lacerations | 197 (4.7) | 26 (4.9) | .80 | 7 (4.5) | .90 | 18 (4.7) | .50 |
| Birthweight, g | 3339 ± 506 | 3363 ± 504 | .30 | 3427 ± 514 | .03 b | 3415 ± 511 | .005 b |
| Macrosomia (>4000 g) | 391 (9.3) | 52 (9.9) | .70 | 19 (12.3) | .20 | 30 (7.8) | .30 |
| LGA (>90th centile) | 337 (8.1) | 52 (9.9) | .20 | 21 (13.5) | .02 b | 45 (11.7) | .01 b |
| Infant outcomes | |||||||
| NICU admission | 173 (4.1) | 17 (3.2) | .30 | 11 (7.1) | .07 | 28 (7.3) | .004 b |
| Respiratory morbidity d | 238 (5.7) | 29 (5.5) | .90 | 6 (3.9) | .30 | 23 (6.0) | .80 |
| Shoulder dystocia | 78 (1.9) | 10 (1.9) | .90 | 3 (1.9) | .90 | 7 (1.8) | .90 |
| Birth trauma | 14 (0.3) | 0 (0.0) | .20 | 1 (0.6) | .50 | 0 (0.0) | .30 |
| Jaundice | 267 (6.4) | 39 (7.4) | .40 | 12 (7.7) | .50 | 36 (9.4) | .03 b |
| Hypoglycemia | 47 (1.1) | 7 (1.3) | .70 | 3 (1.9) | .40 | 28 (7.3) | < .001 b |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
