Background: Subclinical hypothyroidism (SCH) in pregnancy is associated with adverse pregnancy and perinatal outcomes. However, few studies have investigated the evolution of postpartum thyroid function in these women. This study aimed to determine the postpartum outcomes of SCH during pregnancy and the clinical and biochemical factors related to the evolution of long-term hypothyroidism. Methods: A total of 393 women diagnosed with SCH during pregnancy (defined as thyrotropin [TSH] >4.0 μIU/mL with normal free thyroxine levels according to the 2017 American Thyroid Association guidelines) were prospectively followed up after delivery. Among them, 216 underwent long-term follow-up [median (interquartile range) follow-up time: 11 (7–19) months] postpartum. The clinical and biochemical characteristics of the women with long-term postpartum hypothyroidism and euthyroidism were compared. Linear mixed model (LMM) was used to explore the risk factors for longitudinal changes of TSH, and logistic regression analysis was employed to identify the independent predictors of long-term postpartum hypothyroidism. Results: The probability of long-term hypothyroidism after delivery in SCH during pregnancy was 38.9%. Among the subjects with normal thyroid function 6-week postpartum, 28.2% developed hypothyroidism during long-term follow-up. The LMM showed that gestational age at the time of SCH diagnosis (estimate: −0.018, p = 0.004) and thyroid peroxidase antibodies (TPOAb) (estimate: 0.001, p = 0.020) were significantly associated with longitudinal changes of TSH. The logistic regression model showed that TPOAb positive both during pregnancy and six-week postpartum was a risk factor for long-term hypothyroidism after delivery (odds ratio = 4.686 [95% confidence interval 1.242 to 17.680], p = 0.023). Conclusions: More than one-third of patients with SCH during pregnancy had persistent hypothyroidism after delivery. We recommend that patients with TPOAb positive both during pregnancy and six-week postpartum undergo close follow-up to detect persistent hypothyroidism, especially before the next pregnancy.
Objectives: To investigate the predictors influencing the advancement of 10-year cardiovascular disease (CVD) risk after infection with Corona Virus Disease 2019 (COVID-19) in type 2 diabetes patients, and to provide a theoretical basis for an early intervention program for the cardiovascular dimension of the cardiovascular-kidney-metabolic syndrome (CKM syndrome).Methods: A cohort of 378 individuals diagnosed with type 2 diabetes was analyzed retrospectively. The progression of 10-year CVD risk was characterized by an elevation in the 10-year CVD risk category, as determined by the SCORE2-Diabetes scoring system, in type 2 diabetic infected with COVID-19. Factors influencing 10-year CVD risk progression were evaluated through univariate and multivariate stepwise logistic regression. Nonlinear relationships between predictors and 10-year CVD progression were assessed using restricted cubic spline (RCS) analysis, subsequently followed by an analysis of threshold effects. Finally, the predictive performance of various factor combinations for 10-year CVD risk progression during the post-acute COVID-19 phase in type 2 diabetes mellitus cohorts was measured by area under roc curve (AUC).Results: After infection with COVID-19, 12.2% (n=46) experienced progression in their 10-year CVD risk category. In multifactorial stepwise logistic regression, alcohol consumption [odds ratio (OR) 2.10, 95% confidence interval (CI) 1.02-4.34], estimated glomerular filtration rate (eGFR) (OR 0.96, 95% CI 0.94-0.99) and high-sensitivity C-reactive protein (hs-CRP) (OR 1.33, 95% CI 1.13-1.57), were found to be significantly linked to the progression of 10-year CVD risk. Restricted cubic spline analysis (RCS) demonstrated a nonlinear correlation between hs-CRP and 10-year CVD risk progression with a threshold of 3.0 mg/L. 10-year CVD risk was significantly higher with increasing hs-CRP levels at hs-CRP < 3.0 mg/L (OR 2.28, 95% CI 1.48-3.55), and the two-stage model significantly superior to a single linear model (P = 0.028 for log-likelihood ratio). Among the different combinations of models for alcohol consumption, hs-CRP, and eGFR, the full model combination of all three had the best predictive effect (AUC = 0.749).Conclusion: Alcohol consumption and elevated hs-CRP were associated with increased cardiovascular risk progression, while higher eGFR levels were inversely associated with risk progression.