# Uropathogens and prognosis among patients with hospital-diagnosed acute pyelonephritis: insights from a 19-year population-based cohort study

## TL;DR

This study analyzed trends in hospital-diagnosed acute pyelonephritis over 19 years, finding that Escherichia coli remained the main cause, while non-E. coli infections had worse outcomes.

## Contribution

The study provides a long-term population-based analysis of uropathogen trends and their association with clinical outcomes in acute pyelonephritis.

## Key findings

- Microbiological testing rates increased from 75.1% to 92.9% between 2000-2006 and 2013-2018.
- Non-E. coli APN was linked to longer hospital stays and higher 30-day mortality compared to E. coli APN.
- E. coli remained the predominant pathogen (77.3%-81.9%) across all study periods.

## Abstract

The microbial aetiologies of acute pyelonephritis (APN) may change over time. We aimed to describe long-term trends in microbiological diagnostics and pathogen distribution in patients with hospital-diagnosed APN, and to characterise clinical outcomes by pathogens.

We conducted a population-based, serial cross-sectional and cohort study of patients with hospital-diagnosed APN in North Denmark across three periods covering 2000-2018. National health registries were linked with microbiological data to describe temporal trends in microbiological diagnostics and pathogen distribution, and to provide a descriptive comparison of median length of stay (LOS) with interquartile range (IQR) and 30-day cumulative mortality with 95% confidence interval (CI) between Escherichia coli and non-E. coli APN.

We identified 5338 APN episodes among 4773 patients. The proportion with urine culture increased from 75.1% in 2000-2006 to 92.9% in 2013-2018, with a concomitant increase in the proportion with a positive urine culture from 44.4% to 56.7%. The median LOS declined by 2 days across calendar periods. E. coli remained the predominant pathogen with a prevalence in the range 77.3%-81.9%. Non-E. coli APN was more common in male, older, and comorbid patients, and was characterised by longer LOS (median 5 days [IQR: 3-8] vs. 4 days [IQR: 2-6]) and higher 30-day mortality (3.7% [95% CI 2.3%-5.2%] vs. 1.0% [95% CI 0.6%-1.5%]) compared with E. coli.

Microbiological testing increased during the study period, and the pathogen distribution remained largely stable with E. coli as the predominant uropathogen. Non-E. coli infections were associated with slightly less favourable short-term outcomes.

The online version contains supplementary material available at 10.1007/s15010-026-02729-7.