Acute Kidney Injury and Rapidly Progressive Glomerulonephritis in Acute Post-Streptococcal and Post-Infectious Glomerulonephritis: Pooled Prevalence, Age-Related Risk and Prognostic Determinants — A Systematic Review and Meta-Analysis

Abstract

Background: Acute post-streptococcal glomerulonephritis (APSGN) commonly causes acute nephritic syndrome in children, yet how often it causes acute kidney injury (AKI) or a rapidly progressive course is unquantified.

Objective: To pool the prevalence of AKI, kidney replacement therapy (KRT) and rapidly progressive or crescentic glomerulonephritis (RPGN), and to explain the variation between cohorts.

Methods: PubMed and Europe PMC were searched. Observational studies reporting AKI, KRT or RPGN in post-streptococcal or post-infectious glomerulonephritis were eligible. Prevalence was pooled by random-intercept logistic regression, determinants as odds ratios. Risk of bias used ROBINS-E, certainty GRADE.

Result: Thirty studies were included, 26 pooled. How a study defined AKI dominated its reported risk: 62.3% (95% CI 44.7-77.2) under a bare creatinine or eGFR threshold, 35.0% (21.3-51.7) under consensus criteria and 16.0% (8.6-27.8) where none was stated (p = 0.0001). Definition class and income setting explained 66.7% of variance. Across 21 cohorts AKI complicated 34.7% (23.1-48.5) of presentations (n = 2,454; I2 = 95.2%; prediction interval 3.2-89.4%), rising to 49.6% (35.5-63.7) in the 13 cohorts stating a definition. KRT was required in 3.3% (1.4-7.8) and RPGN in 6.8% (3.4-13.2). Among children each year of age raised the odds of an adverse outcome by 17% (OR 1.17, 1.09-1.24); heavy proteinuria gave OR 2.66 (1.50-4.73). Certainty was very low.

Conclusion: Reported AKI risk depends more on how a centre defines AKI than on the population served. Between a third and a half of children develop AKI; severe outcomes are uncommon. A minimum reporting standard is proposed.