Outcomes in hospitalised patients with penicillin allergy: a systematic review and meta-analysis protocol

TL;DR

This study aims to determine if a reported penicillin allergy affects mortality rates in hospitalized patients.

Contribution

The study introduces a systematic review and potential meta-analysis protocol to assess the impact of self-reported penicillin allergy on hospital mortality.

Findings

Eligible studies will compare hospitalized patients with and without penicillin allergy.

The study will extract mortality data to assess the impact of penicillin allergy on hospital outcomes.

Abstract

When patients report a penicillin allergy (self-reported penicillin allergy, SRPA), clinicians alter their antibiotic management of the patient. However, fewer than 5% of SRPA have life-threatening reactions, and receiving narrower spectrum antibiotics may result in under treatment. Reports from community studies suggest that patients with SRPA are more likely to come to harm, but as SRPA implies exposure to healthcare, it is unclear whether additional harm is as a result of co-morbidities and the need to receive healthcare, rather than a harm in itself. This raises the question of how great an impact SRPA has on hospital and ICU mortality for patients admitted to hospital.

A literature search will be performed using MEDLINE, Embase and CINAHL (via OVID) in October 2024. The eligible studies will be of adult hospitalised patients with penicillin allergy compared with their non-allergic counterparts. We will extract demographic and outcome data including whether the patients were treated in ICU and mortality (28-days, in-ICU or in-hospital).

If possible, the results will be pooled for a meta-analysis by combining the all-cause mortality of patients at 28 days, in-ICU or in-hospital. Heterogeneity will be assessed using the I2 or H2 statistics if the number of included studies is less than 10.

Previous reports of poor outcomes from SRPA are often based around increased risk of increased antibiotic use and antibiotic resistance. The requirement of SRPA to develop where there is exposure to healthcare and co-morbidities may mean that these outcomes are derived from poorer health rather than from SRPA perse. The impact of SRPA on hospitalised patients is less understood, with outcomes that are well-defined, such as mortality. This review will outline the differences that exist, and a meta-analysis may define the size of any differences.

PROSPERO CRD42024537658.

The online version contains supplementary material available at 10.1186/s13643-025-03038-0.