Association Between Tonsillectomy and Outcomes in Patients With Immunoglobulin A Nephropathy

Abstract

Importance: Immunoglobulin A nephropathy is a major cause of end-stage renal disease worldwide; previous methods of medical management, including use of renin-angiotensin system inhibitors and corticosteroids, remain unproven in clinical trials.

Objective: To investigate the possible association between tonsillectomy and outcomes in patients with IgA nephropathy.

Design, setting, and participants: This cohort study included 1065 patients with IgA nephropathy enrolled between 2002 and 2004 and divided into 2 groups, those who underwent tonsillectomy and those who did not. Initial treatments (renin-angiotensin system inhibitors or corticosteroids) within 1 year after renal biopsy were also evaluated. A 1:1 propensity score matching was performed to account for between-group differences and 153 matched pairs were obtained. Follow-up concluded January 31, 2014. Analysis was conducted between September 11, 2017, and July 31, 2018.

Exposure: Tonsillectomy.

Main outcomes and measures: The primary outcome was the first occurrence of a 1.5-fold increase in serum creatinine level from baseline or dialysis initiation. Secondary outcomes included additional therapy with renin-angiotensin system inhibitors or corticosteroids initiated 1 year after renal biopsy and adverse events.

Results: In 1065 patients (49.8% women; median [interquartile range] age, 35 [25-52] years), the mean (SD) estimated glomerular filtration rate was 76.6 (28.9) mL/min/1.73 m2 and the median (interquartile range) proteinuria was 0.68 (0.29-1.30) g per day. In all, 252 patients (23.7%) underwent tonsillectomy within 1 year after renal biopsy and 813 patients (76.3%) did not undergo tonsillectomy. The primary outcome was reached by 129 patients (12.1%) during a median (interquartile range) follow-up of 5.8 (1.9-8.5) years. In matching analysis, tonsillectomy was associated with primary outcome reduction (hazard ratio, 0.34; 95% CI, 0.13-0.77; P = .009). In subgroup analyses, benefit associated with tonsillectomy was not modified by baseline characteristic differences. Those undergoing tonsillectomy required fewer additional therapies 1 year following renal biopsy (adjusted hazard ratio, 0.37; 95% CI, 0.20-0.63; P < .001) without increased risks for adverse events, except transient tonsillectomy-related complications.

Conclusions and relevance: This study found that tonsillectomy was associated with a lower risk of renal outcomes in patients with IgA nephropathy. The potential role of tonsillectomy should be considered for preventing end-stage renal disease in patients with IgA nephropathy.

Figure 1.. Comparison of the Cumulative Rates of Renal Survival Between the T1 and T0 Groups After Propensity Score Matching

T1 indicates patients who underwent tonsillectomy; T0, patients who did not undergo tonsillectomy.

Figure 2.. Association Between Tonsillectomy and Primary Outcome in Various Multivariate Models

Hazard ratios (HRs) and 95% confidence intervals for primary events consistently and significantly showed that tonsillectomy was associated with a favorable renal outcome in various multivariate models, including a 1:1 propensity matching model, the inverse probability of treatment–weighted (IPTW) matching model, and a simple multivariate model of the entire cohort. The HRs and 95% CIs (tonsillectomy vs no tonsillectomy) are presented as a log-log plot. The matching model was created by a propensity score using a caliper (a = 0.30 and c = 0.51) (Table 2; eTable 2 in the Supplement). The IPTW model was the mean treatment effect on those treated (eTable 4 in the Supplement). We adjusted the standard Cox proportional hazards model according to age, sex, body mass index, diabetes presence or absence, mean arterial pressure, estimated glomerular filtration rate, proteinuria, urine occult, uric acid, total cholesterol, IgA, complement 3, renin-angiotensin system inhibitor use, and corticosteroid therapy. Supplementary predictive values of these factors are shown in eTable 5 in the Supplement. T1 indicates patients who underwent tonsillectomy; T0, patients who did not undergo tonsillectomy.

Figure 3.. Association Between Tonsillectomy and Primary Outcome in Subgroups for the Baseline Characteristics and Initial Treatments

A, No interaction between baseline clinical parameters and tonsillectomy was found in the overall cohort. Stratified analyses suggest that almost all clinical subgroups tended to have an event-free survival benefiting from tonsillectomy, independently of their estimated glomerular filtration rate (eGFR), proteinuria, hematuria extent, or renin-angiotensin aldosterone system inhibitors (RASi) use. B, Despite the wide 95% CI, the combined tonsillectomy hazard ratios (HRs) consistently exhibited an association with favorable outcomes across various corticosteroid subgroups. Each combined HR in oral steroid therapy without pulse regimen (S1), oral steroid therapy with pulse regimen (S2), or S1 and S2 did not significantly differ from that in no steroid therapy (S0). T1 indicates patients who underwent tonsillectomy; T0, patients who did not undergo tonsillectomy.