P404 can the baseline nt–probnp level be used as a prognostic marker in patients hospitalized for covid–19? a single centre experience

Abstract Background NT–proBNP is commonly used a reliable prognostic biomarker in heart failure. Although SARS–CoV–2 is primarily a respiratory virus, it can also cause a myocardial injury. Previous observations indicate that COVID–19 patients can show a pathological rise of NT–proBNP during the disease course. Aim To assess the in–hospital prognostic significance of baseline NT–proBNP levels in COVID–19 patients. Methods We retrospectively analysed the data of one–hundred and ninety–two consecutive patients (mean age 70±15, 54.6 % males), hospitalized in our institution for COVID–19 disease. Demographic parameters, clinical history, pharmacological treatments and laboratory data at the admission were analysed. According to the baseline NT–proBNP levels, the whole population was divided into normal (Group A) and elevated (Group B) NT–proBNP, considering ≥ 125 pg/mL level as the pathological cut off. The length–of–stay, the orotracheal intubation rate, non–invasive ventilation and in–hospital mortality were taken into account as prognostic parameters. Results Forty–seven patients and one–hundred and forty–five patients belonged to Group A and Group B, respectively. Group A patients were significantly younger (57±13 vs 74±13 yrs, p < 0.001), with a lower rate of previous cardiac disease (6.4% vs 39.3%, p < 0.001) and atrial fibrillation (4.3% vs 16.7%, p < 0.033) and a better eGFR (94±20 vs 71±29 ml/m’, p < 0.001). No differences were noted between the two groups in the prevalence of diabetes, hypertension, ACE/ARBs treatment. The length–of–stay was similar (20±13 days in Group A vs 22±19 days in Group B, respectively, p=ns). Although patients of Group B showed a higher rate for orotracheal intubation (4.3% vs 13.8%) and non–invasive ventilation (13.8% vs 32.4%,), these differences were not significantly different. The in–hospital mortality was considerably lower in patients with normal baseline NT–proBNP level, as compared to Group B patients (2.1% vs 23.4% p < 0.001). When stratified by quartiles of NT–proBNP, the subgroups showed a prognosis clearly related to the expression of the biomarker. Conclusion In patients hospitalized for COVID–19, normal baseline NT–proBNP level identifies a population with a short–term better outcome. This widely diffuse biomarker could be used in the initial phase of admission as a prognostic tool to characterize the in–hospital prognosis.