Fig. 3.

Thrombosis-free survival and survival in the absence of adverse clinical outcome in patients with COVID-19 with or without TEs, according to diabetes and stress hyperglycemia. Kaplan-Meier thrombosis-free survival estimates for patients with COVID-19 pneumonia (panel a). Survival rate was estimated for the presence of diabetes or stress hyperglycemia. The log-rank test was used to test differences in the estimated survival rate. Crosses indicate censored patients (censoring for death end of follow-up data). The forest plots (panel b) show the hazard ratios for survival in the absence of adverse clinical outcome according to presence/absence of diabetes/stress hyperglycemia. The presence of thrombotic events was considered as a time-varying covariate in Cox proportional hazards models. The effect estimates were reported as Hazard Ratios with the corresponding 95% CI, estimated according to the Wald approximation. Cox regression analysis was adjusted for sex and age. Dots represent the HR, lines represent 95% confidence interval (CI), and * P < 0.01.