Bacterial co-infection raises in-hospital mortality of COVID-19 patients: a retrospective study.

Abstract

We aim to explore whether the bacterial co-infection with COVID-19 will raise the in-hospital mortality. COVID-19 patients' information were collected for analysis in our retrospective study. Neutrophil count and procalcitonin (PCT) were used to estimate whether there was a suspected bacterial co-infection. The main baselines between the suspected bacterial infection (SBI) and no evidence of bacterial infection (NBI) groups were no significant differences. In SBI group, patients required more therapies than NBI group. There was significantly higher in-hospital mortality (26% vs.9%, P < 0.001) between SBI and NBI groups in overall population. And in each subgroup based on pneumonia inflammation index (PII), it also showed higher in-hospital mortality of COVID-19 patients with bacterial co-infection. With logistic regression models, it showed that bacterial co-infection was associated with significantly higher in-hospital mortality in overall population (OR 1.694, 95% CI 1.179-2.434, p = 0.004) and mild subgroup (OR 2.374, 95% CI 1.249-4.514, p = 0.008). The rate of bacterial co-infection in overall population was 51%. At the same time, it showed a significantly higher rate of bacterial co-infection in critical subgroup than severe subgroup (63% vs. 49%, p = 0.003), and than that in moderate subgroup (63% vs. 48%, p = 0.002) based on clinical classification. It showed a significantly higher rates of bacterial co-infection in severe subgroup than moderate subgroup (66% vs. 49%, p = 0.001) based on PII. The result showed that the risk factor associated with significantly higher in-hospital mortality was PII (OR 1.018, 95%CI 1.012 to 1.024, P < 0.001) with logistic regression models. Bacterial co-infection estimated by Neutrophil count and procalcitonin significantly raises in-hospital mortality of COVID-19 patients in overall population in our study. Its impact is more significant in mild and moderate PII subgroups. PII based on CT imaging combined with neutrophil count and PCT is beneficial for accurate differentiation of bacterial co-infection of COVID-19.