The effect of antibiotic therapy on clinical outcome in patients hospitalized with moderate COVID-19 disease: a prospective multi-center cohort study

Abstract

Purpose: The benefit of antibiotic treatment (ABT) for patients with moderate COVID-19 is unclear and overtreatment poses the risk of adverse effects such as Clostridioides difficile infection and antibiotic resistance. This multi-center study compares health status improvement between patients with and without ABT at hospital admission.

Methods: Between March 2020 and May 2023, hospitalized adults with confirmed SARS-CoV-2 infection were recruited from the German National Pandemic Cohort Network (NAPKON), which includes patients from various hospitals across Germany. The study population included patients with moderate or severe COVID-19 at baseline. The primary objective was to compare health improvement or decline after two weeks between patients who received ABT at baseline and those who did not in the moderate COVID-19 population. The statistical analysis adjusted for confounders such as gender, age, vaccination status, clinical condition, and comorbidities. The severe COVID-19 population was investigated as a secondary objective.

Results: A total of 1,317 patients (median age 59 years; 38% women) were eligible for analysis, of whom 1,149 had moderate and 168 severe COVID-19 disease. ABT for pneumonia was administered to 467 patients with moderate and 117 with severe COVID-19. ABT at baseline was significantly associated with a higher deterioration rate after two weeks in patients with moderate COVID-19 (ABT: 292 improvement, 61 deterioration; no ABT: 429 improvement, 14 deterioration). A similar result was obtained in the multiple regression analysis where an odds ratio of 5.00 (95% confidence interval: 2.50 - 10.93) for ABT was observed.

Conclusion: We found no benefit of antibiotic therapy in patients with moderate COVID-19. Use of ABT was associated with a higher likelihood of clinical deterioration.

Keywords: COVID-19; Clinical improvement; Clinical outcome; Moderate disease; Rational antibiotic therapy.

Fig. 1

Flowchart of study participants. ABT antibiotic treatment, NAPKON National Pandemic Cohort Network, SUEP Cross-sectoral cohort, WHO World Health Organization

Fig. 2

Forest plots for moderate COVID-19 at baseline. Results from multiple logistic regressions. Shown are logarithmic (base 10) odds ratios and 95% confidence intervals of the final multiple model after model selection. A Factors associated with ABT at baseline; variables in the full model but excluded during model selection: body mass index, smoking, Clinical Frailty Scale. B Factors associated with improvement vs. deterioration after 14 days; variables in the full model but excluded during model selection: Charlson Comorbidity Index, body mass index, smoking, Clinical Frailty Scale, baseline WHO (World Health Organisation) score. Results from a multiple logistic regression. ref: Reference category for the variable in the multiple logistic regression model, CCI: Charlson comorbidity index, WHO: baseline WHO (World Health Organization) score

Fig. 3

Kaplan-Meier curves for outcome death during observation period (hospital stay and follow-up). A Baseline WHO score 4 - 5; B Baseline WHO score 6-9. Shown are the survival probabilities and the 95% confidence intervals (shaded). p: P-value for differences in the distribution of survival times between patients with and without ABT (log rank test). Note that the y axis range is different in A and B