Background Acute pancreatitis (AP) is often accompanied by changes in body temperature (Tb), but whether and how Tb at hospital admission is associated with the outcome of the disease remained unclear. Previous studies showed contradictory results regarding its prognostic value, and no meta-analysis was performed. Methods In this systematic review and meta-analysis, we searched the PubMed, Embase, and CENTRAL databases (from inception to April 25, 2024) for studies in adult AP patients with reported admission Tb and mortality rate or a disease severity marker. With a random-effects model, we calculated pooled proportion (for mortality) or pooled mean (for secondary outcomes) with 95% CI and compared them between febrile and afebrile subgroups. With meta-regression, we assessed the correlation between admission Tb and mortality or length of hospital stay (LOS). Results We found 26 eligible studies including data from 40,630 AP patients. The pooled mortality rate was higher (p < 0.001) in febrile patients (0.15, 95% CI 0.11–0.20; I2 = 40%) than in afebrile patients (0.05, 95% CI 0.03–0.09; I2 = 98%). LOS was longer (p = 0.025) in the febrile subgroup (39.1 days, 95% CI 16.8–61.4; I2 = 91%) than in the afebrile subgroup (17.5 days, 95% CI 7.6–27.5; I2 = 96%). With meta-regression, admission Tb significantly and positively correlated with mortality (0.09, 95% CI 0.07–0.12) and LOS (16.5, 95% CI 5.6–27.5). The APACHE II score was significantly (p = 0.030) higher, while the SOFA score tended to be higher (p = 0.168) in the febrile than in the afebrile subgroup. Blood levels of amylase and C-reactive protein did not differ meaningfully between the groups. Conclusions Fever on admission is associated with higher mortality, increased disease severity, and longer hospitalization in AP. High-quality studies with standardized Tb measurements are needed to determine the prognostic performance of admission Tb in AP. Systematic review registration This study was registered with PROSPERO on 16 January 2020 (registration number: CRD42020139154).

Fever on hospital admission is associated with increased disease severity and mortality in acute pancreatitis: a systematic review and meta-analysis

期刊论文
出版模式：

开放获取
发表日期：

2026年7月21日

查看全文

Eszter Pakai,
Krisztian Klonfar,
Anna Walter,
Nelli Farkas,
Anna Sara Bognar,
Jimin Lee,
Peter Hegyi,
Andras Garami

摘要

Background

Acute pancreatitis (AP) is often accompanied by changes in body temperature (T_b), but whether and how T_b at hospital admission is associated with the outcome of the disease remained unclear. Previous studies showed contradictory results regarding its prognostic value, and no meta-analysis was performed.

Methods

In this systematic review and meta-analysis, we searched the PubMed, Embase, and CENTRAL databases (from inception to April 25, 2024) for studies in adult AP patients with reported admission T_b and mortality rate or a disease severity marker. With a random-effects model, we calculated pooled proportion (for mortality) or pooled mean (for secondary outcomes) with 95% CI and compared them between febrile and afebrile subgroups. With meta-regression, we assessed the correlation between admission T_b and mortality or length of hospital stay (LOS).

Results

We found 26 eligible studies including data from 40,630 AP patients. The pooled mortality rate was higher (p < 0.001) in febrile patients (0.15, 95% CI 0.11–0.20; I² = 40%) than in afebrile patients (0.05, 95% CI 0.03–0.09; I² = 98%). LOS was longer (p = 0.025) in the febrile subgroup (39.1 days, 95% CI 16.8–61.4; I² = 91%) than in the afebrile subgroup (17.5 days, 95% CI 7.6–27.5; I² = 96%). With meta-regression, admission T_b significantly and positively correlated with mortality (0.09, 95% CI 0.07–0.12) and LOS (16.5, 95% CI 5.6–27.5). The APACHE II score was significantly (p = 0.030) higher, while the SOFA score tended to be higher (p = 0.168) in the febrile than in the afebrile subgroup. Blood levels of amylase and C-reactive protein did not differ meaningfully between the groups.

Conclusions

Fever on admission is associated with higher mortality, increased disease severity, and longer hospitalization in AP. High-quality studies with standardized T_b measurements are needed to determine the prognostic performance of admission T_b in AP.

Systematic review registration

This study was registered with PROSPERO on 16 January 2020 (registration number: CRD42020139154).