In a recent article published in eClinical Medicine, researchers determine whether frequent antibiotic use increased the risk of adverse outcomes, including death, following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection.
Study: Repeated antibiotic exposure and risk of hospitalization and death following COVID-19 infection (OpenSAFELY): a matched case–control study. Image Credit: dturphoto / Shutterstock.com
Table of Contents
Previous studies indicate that repeated and frequent exposure to antibiotics increases the risk of adverse outcomes like autoimmune diseases after infectious diseases. This could be due to disturbances in gut microbiota that ultimately lead to severe immune and metabolic dysregulation, or gut colonization of antibiotic-resistant pathogens that make people more susceptible to adverse consequences of infection.
Some studies have also reported that SARS-CoV-2 infection alters the composition of the gut microbiome, especially in those exposed to antibiotics for prolonged periods. Thus, the researchers of the current study hypothesized that patients who frequently use antibiotics are more susceptible to adverse coronavirus disease 2019 (COVID-19) outcomes.
About the study
In the present population-level matched case-control study, researchers create two distinct cohorts to examine the long-term effects of repeated and intermittent antibiotic use on subsequent COVID-19 severity.
The National Health Services (NHS) in England governs the OPEN SAFELY-TPP platform, from which the researchers retrieved the data of electronic health records (EHRs) of 22 million people representing 40% of the population of England. OPEN SAFELY-TPP data were linked to other databases at the patient level, which integrated primary and secondary care, COVID-19 test, and death registration data.
In Study 1, the researchers investigated the effects of prior antibiotic exposure on COVID-19-related hospitalizations. The study population comprised eligible patients selected from the beginning of the pandemic to the end of 2021. The International Classification of Diseases, 10th revision (ICD-10) codes U07.1 and U07.2 were used to identify COVID-19 cases.
The index date of cases was the incident date of COVID-19 hospital admission from the NHS Digital Secondary Use Service (SUS). Controls were identified from the first COVID-19 record retrieved through the Second-Generation Surveillance System (SGSS).
In Study 2, the severity of outcomes was determined among COVID-19-related hospitalizations identified in Study 1 by measuring the number of deaths.
The researchers matched one case with six controls based on age, gender, and general practice region for both studies. The study was divided into three periods, including February 2020-August 2020, September 2020-April 2021, and May 2022 December 2022.
The maximum antibiotic exposure time frame was set to three years. Since only long-term effects of antibiotic exposure were considered, all prescriptions issued within six weeks of the study index date were excluded.
A total of 55 systemic antibiotics listed in the British National Formulary (BNF) chapter 5.1 for common infectious diseases were evaluated, except for antituberculosis and antileprotic drugs.
Quintile groups were created based on the number of prior antibiotic prescriptions to indicate the frequency of previous antibiotic exposure. For example, the first and fifth quintiles represented low- and high-frequency users. Based on the type of antibiotic prescribed, each quintile was further divided into one to three groups.
A conditional logistic regression model was used to compare the frequency of antibiotic exposure between cases and controls and estimate odds ratios (ORs) with 95% confidence intervals (95% CI). These models were adjusted for all confounding factors including ethnicity, body mass index (BMI), index of multiple deprivations (IMD), care home residents, and smoking status.
The researchers identified 2.47 million COVID-19 patients between February 1, 2020, and December 31, 2021, 98,420 of whom sought hospitalization for COVID-19. After matching with replacements, the study population consisted of 0.67 million patients.
Among all hospitalized patients, 23% died in 30 days, all of whom comprised the Study 2 cohort. As compared to controls, cases had more frequent antibiotic exposure in the prior three years and were at an increased risk of adverse outcomes in crude and adjusted models.
In the quintile with the highest antibiotic exposure, the adjusted OR for hospital admission and death were 1.80 and 1.34 with 95% CI, respectively, as compared to patients without antibiotic exposure. In the sensitivity analyses, adjusted ORs were slightly higher in patients between 40 and 59 years of age with ORs of 2.59 and 2.26 for Studies 1 and 2, respectively.
The adjusted OR (aOR) of quintile five exposure in Study 1 with more than three antibiotic types was nearly double that of quintile one with only one antibiotic type exposure. Comparatively, the OR for the third quintile was the highest at 1.43, in which people used antibiotics moderately. No marked variations in quintiles one and two were observed in those with only one type of antibiotic exposure.
An additional analysis was conducted to understand the effect of potential confounding factors from antibiotic use during the six week exclusion period. To this end, no changes in post-COVID severe outcomes were observed in Study 2; however, overall aOR declined in Study 1.
The current study is the first to examine previous three-year antibiotics exposure and COVID-19 severity and between antibiotic diversity and COVID-19-related hospital admission. The study results indicate a dose-response association between previous antibiotic exposure frequency and severe COVID-19 outcomes.
An association between antibiotic exposure frequency, antibiotic diversity, and COVID-19 clinical outcome severity was observed. Thus, clinicians should not indiscriminately prescribe antibiotics repeatedly or intermittently, as this treatment is associated with considerable risks following infection.