Abstract
BACKGROUND: Thrombocytopenia is the most frequent hemostatic abnormality in sepsis and is closely linked to excess mortality. METHODS: We undertook a retrospective cohort study of 82 septic patients admitted to Changzhou Second People's Hospital between 1 January 2023 and 30 June 2024. Participants were stratified by 28-day survival status. The clinical differences, platelet trends, prognostic value, and risk-factor analysis were analyzed between the two groups. RESULTS: Survivors and non-survivors differed significantly in age, oxygenation index, white-blood-cell count, serum albumin, cholinesterase and in the prevalence of diabetes and hypertension (all comparisons P < .05). Platelet counts measured on intensive care unit (ICU) days 1, 3, 7 and 14 were assessed for their prognostic value. By day 14, platelet counts were markedly higher in survivors than in non-survivors (P < .05). The day-14 platelet count yielded an area under the receiver-operating-characteristic (ROC) curve (AUC) of 0.640, outperforming the day-1 platelet count (AUC 0.514), APACHE II score (AUC 0.488) and Sequential Organ Failure Assessment (SOFA) score (AUC 0.394; all comparisons P < .05). Kaplan-Meier analysis showed that patients whose day-14 platelet count was ≥ 224 × 10(9) L(-1) had significantly better 28-day survival than those below this threshold (P < .05). Platelet count < 224 × 10(9) L(-)(1) in day 14 was an independent predictor of 28-day mortality (P < .05). CONCLUSION: The survival group shares high platelet count and maintaining platelet count at a higher level is associated with improving the 28-day prognosis of sepsis patients.