Objective To observe the relationship between vitamin D3 and the severity as well as prognosis in patients with sepsis, and to explore whether exogenous vitamin D3 can improve the prognosis in patients with sepsis.Methods A prospective randomized double-blind placebo study was conducted. Fifty-seven patients with sepsis admitted to intensive care unit (ICU) of Shengjing Hospital Affiliated to China Medical University from March to November in 2015 were enrolled. Twenty patients with systemic inflammatory response syndrome (SIRS) and 20 healthy volunteers with normal physical examination as control were enrolled during the same time. Patients with sepsis were divided into general sepsis group and severe sepsis group (including septic shock) according to the criteria for the diagnosis of severe sepsis and septic shock in 2012. According to the diagnostic criteria established by the American Endocrine Society, and on the basis of 25-hydroxy vitamin D3 [25(OH)D3], the sepsis patients with deficiency [25(OH)D320-30μg/L] or insufficiency [25(OH)D3 < 20μg/L] of vitamin D were divided into D3 treatment group (supplemented 300 kU vitamin D3) and placebo group (injected 1 mL physiological saline). 28th day was set as the end point, and the patients with sepsis were divided into survival group and death group. The levels of serum 25(OH)D3 in each group were measured by electrochemical luminescence method, and the difference in 25(OH)D3 levels among patients with different severity, gender, and age were recorded. Procalcitonin (PCT), C-reactive protein (CRP), blood routine, liver and kidney function, electrolytes and arterial blood gas analysis, acute physiology and chronic health evaluation Ⅱ (APACHEⅡ) and sequential organ failure score (SOFA), duration of mechanical ventilation, and length of ICU stay of patients with sepsis were observed. Multivariate Cox proportional hazard regression analysis was used to analyze the risk factors of prognosis in patients with sepsis.Results ① In 57 patients with sepsis, there were 15 patients in general sepsis group, and 42 in severe sepsis group; 29 in D3 treatment group, and 28 in the placebo group; 8 patients died within 28 days with mortality rate of 14.04%. ② The levels of serum 25(OH)D3 in sepsis group and SIRS group were significantly lower than those in healthy control group [μg/L: 3.92 (< 3.00, 11.22), 6.99 (3.51, 9.77) vs. 17.25 (13.48, 22.50), both P < 0.01], but there was no significant difference in the serum 25(OH)D3 level between sepsis group and SIRS group as well as patients with different degrees of sepsis. The serum 25(OH)D3 level in female patients with sepsis (n = 24) was significantly lower than that in male (n = 33), and the difference was statistically significant [μg/L: <3.00 (<3.00, 3.87) vs. 11.96 (5.14, 17.29),Z = -4.020,P = 0.000]. There was no significant difference in serum 25(OH)D3 level between the young (age <60 years old,n = 30) and the old (age ≥ 60 years old,n = 27) patients with sepsis [μg/L: 4.54 (<3.00, 9.88) vs. 3.00 (<3.00, 15.08),Z = -0.601,P = 0.548]. ③ In patients with sepsis, there was no significant difference in the duration of mechanical ventilation [hours: 41.00 (7.50, 82.50) vs. 67.00 (4.75, 127.75)], length of ICU stay (days: 5.48±4.08 vs. 6.68±4.87) and 28-day mortality (10.34% vs. 17.86%) between D3 treatment group and placebo group (allP > 0.05). It was shown by Kaplan-Meier survival curve analysis that there was no significance in 28-day accumulated survived rate between the two groups [log-rank test: χ2 = 0.222,P = 0.638]. It was shown by multivariate Cox regression analysis that APACHE Ⅱ score [relative risk (RR) = 8.487, 95% confidence interval (95%CI) = 1.506-47.835, P = 0.015] and 25(OH)D3 < 20μg/L (RR = 0.088, 95%CI = 0.013-0.592,P = 0.012) were the risk factors of prognosis in patients with sepsis.Conclusions The serum 25(OH)D3 level in ICU patients with sepsis was lower than that in healthy people, but there was no significant difference between patients with sepsis and SIRS. The serum 25(OH)D3 level in sepsis patients was related with gender, and the level of the female was lower than that of the male, but was not related with age. Exogenous vitamin D3 supplementation cannot improve the prognosis of ICU patients with sepsis. APACHE Ⅱ score and 25(OH)D3 < 20μg/L were risk factors for the prognosis in ICU patients with sepsis.