Abdominal Visceral Adiposity in the First Trimester Predicts Glucose Intolerance in Later Pregnancy

We completed a prospective cohort study at a single outpatient ultrasound clinic at St. Michael's Hospital in Toronto, Ontario, Canada, between January and May 2008. Women with a singleton pregnancy were eligible for the study at 11–14 weeks' gestation. Women with type 1 or type 2 diabetes prior to pregnancy or with a previous history of GDM were excluded. Written and informed consent was obtained, and the study was approved by the hospital research ethics board.

We used the technique of Armellini et al. (6) to measure subcutaneous and visceral abdominal adipose tissue. A total of 62 patients were scanned on a Philips iU22 ultrasound machine using a 5-2 or 9 MHz probe. Subcutaneous fat depth was measured from the subcutaneous fat layer to the outer border of the rectus abdominus muscle at the level of the linea alba (Fig. 1). Visceral fat depth was measured from the inner border of the rectus abdominus muscle at the level of the linea alba to the anterior wall of the abdominal aorta.

Two sonographers—one a perinatal obstetrician and the other an experienced ultrasound technologist—each performed three measurements of the subcutaneous and visceral fat depths. Each rater was masked to the other's assessment, and the measurements were recorded on separate data collection sheets.

Intrarater reliability of ultrasonography was determined for each rater using three images of subcutaneous and three images of visceral adipose tissue depth per participant, respectively. Interobserver reliability was separately calculated for subcutaneous and visceral adiposity.

Mean subcutaneous and visceral adipose tissue depths were determined for each participant, with the measures of both sonographers pooled. The upper-quartile value for each was considered elevated, and the three lowest quartile values were treated as the referent. All participants underwent a 50-g GCT at 24–28 weeks' gestation. An abnormal 50-g GCT was defined at a conventional cut point: ≥7.8 mmol/l. Unadjusted and adjusted ORs and 95% CIs expressed the risk of an abnormal GCT in association with an elevated subcutaneous and an elevated visceral adipose tissue depth, respectively.

All P values were two-sided, and significance was set at a value of 0.05. Statistical analyses were performed using SAS (version 9; SAS Institute, Cary, NC).

There were 62 women recruited in total at mean ± SD gestation 12.4 ± 0.60 weeks and age 31.7 ± 5.0 years. The median gravidity was 2.0, and 31 (50.0%) were of nonwhite ethnicity. Mean prepregnancy BMI was 23.9 ± 5.2 kg/m². Mean subcutaneous and visceral adiposity tissue depths by ultrasonography were 1.8 ± 0.72 cm (range 0.63–3.7) and 4.0 ± 1.4 cm (1.3–8.2), respectively. The 50-g GCT was completed at 27.4 ± 1.4 weeks' gestation.

Reliability measures were based on all 62 women. The intraclass correlation coefficient for intraobserver agreement of visceral adiposity tissue measurement was 0.94 (95% CI 0.91–0.96) for the physician and 0.97 (0.95–0.98) for the technologist. Similar results were seen for subcutaneous adiposity measurement. Lin's concordance correlation coefficient for interobserver reliability (between physician vs. technologist) was 0.79 (0.69–0.88) for subcutaneous adiposity and 0.87 (0.82–0.93) for visceral adiposity.

Fifty-eight women formed the sample used to analyze the relationship between adiposity and subsequent GCT positivity. No significant associations were observed between the upper-quartile subcutaneous adipose tissue depth and a positive GCT (Table 1). However, an elevated visceral adipose tissue depth was significantly associated with a positive GCT (unadjusted OR 17.3 [95% CI 1.8–163.8]). Even after adjusting for maternal age and prepregnancy BMI, the association remained significant (16.9 [1.5–194.6]).

Since we included only 62 women, our risk estimates were imprecise. We used a 50-g GCT as an indicator of glucose intolerance later in pregnancy rather than a more definitive 2-h 75-g oral glucose tolerance test. However, a positive GCT is a reasonable predictor of GDM-related adverse perinatal outcomes (7). A strength of this cohort study is that we prospectively assessed abdominal adiposity at around the same gestational age using a standardized protocol. All sonographers were masked to each other's measurements, and the GCTs were carried out without knowledge of the abdominal depths.

Visceral adiposity predicts insulin resistance (8) and diabetes (4) independent of BMI, so it was logical for us to use ultrasonography to measure visceral fat in relation to glucose intolerance in pregnancy. When used in nonpregnant patients, ultrasound has a correlation coefficient of between 0.55 (9) and 0.81 (10) and a diagnostic concordance of 74% with computed tomography (9) in the assessment of visceral adiposity.

Maternal obesity, routinely defined as an elevated prepregnancy BMI, is associated with adverse pregnancy outcomes (1,–3,11). BMI may not accurately differentiate between the contributions of muscle and fat to body weight or of subcutaneous versus visceral abdominal fat. Epidemiological and metabolic studies have found that the adverse metabolic consequences of excess fat depend largely on the location of the fat (12,13), with centrally located visceral fat more pathogenic than subcutaneous adipose tissue (14). Our results are consistent with this concept.

Measurement of visceral adiposity during a routine 11–14 weeks' gestation ultrasound might improve the performance of screening for GDM (15). Moreover, identifying women at high risk for GDM because of elevated visceral adiposity could lead to either earlier screening or earlier dietary and lifestyle modifications. Clearly, this opens up a new avenue for research.

No potential conflicts of interest relevant to this article were reported.