Wenwen Li, Youhong Kang, Huale Zhang, Wenxin Tan, Feng Li, Kunhai Wu, Zhuanji Fang
In this hospital-based cohort, early- and mid-pregnancy GWG patterns were associated with HDP. The estimates for the smallest groups were imprecise, and the observational design does not establish causality or predictive utility. Studies with prospectively recorded HDP diagnosis timing and external validation are required before clinical use.
OBJECTIVE: To examine the association between trajectories of gestational weight gain (GWG) during early and mid-pregnancy and hypertensive disorders of pregnancy (HDP). A trajectory approach was used because total GWG or a single measurement cannot represent the timing and pattern of weight change.
METHODS: This retrospective cohort included 11,278 women who attended the nutrition clinic and delivered at Fujian Provincial Maternity and Child Health Hospital, Fuzhou, China, between March 2014 and May 2018. Interval-specific GWG through 28 weeks was characterized using latent class growth modeling. Logistic regression estimated associations with HDP; the fully adjusted model included maternal age, pre-pregnancy body mass index, parity, platelet count, lactate dehydrogenase, and alanine aminotransferase.
RESULTS: The current analysis used four reported GWG patterns: stable-low (Group 1; n=9,314, 82.6%), early-slow/mid-fast (Group 2; n=70, 0.6%), early-fast/mid-slow (Group 3; n=301, 2.7%), and moderate-growth (Group 4; n=1,593, 14.1%). Compared with Group 1, the fully adjusted odds of HDP were higher in Group 3 (odds ratio [OR]=2.285, 95% confidence interval [CI] 1.370-3.613; P=0.001) and Group 4 (OR=1.356, 95% CI 1.037-1.752; P=0.023), but not Group 2 (OR=1.084, 95% CI 0.176-3.528; P=0.911).
CONCLUSION: In this hospital-based cohort, early- and mid-pregnancy GWG patterns were associated with HDP. The estimates for the smallest groups were imprecise, and the observational design does not establish causality or predictive utility. Studies with prospectively recorded HDP diagnosis timing and external validation are required before clinical use.