Metamizole-Associated Risks in Decompensated Hepatic Cirrhosis

Abstract

Background: Because of the increased risk of acute renal failure (ARF), the use of cyclooxygenase (COX) inhibitors is not recommended in patients with decompensated hepatic cirrhosis. Metamizole is not a classic COX inhibitor, but there are insufficient data to support its safe use. In this study, we investigate the effect of metamizole on the risk of ARF in these patients.

Methods: Metamizole use, ARF incidence, and patient mortality were examined in a large, retrospective, exploratory cohort and validated with data from a prospective registry.

Results: 523 patients were evaluated in the exploratory cohort. Metamizole use at baseline was documented in 110 cases (21%) and was independently associated with the development of ARF, severe (grade 3) ARF, and lower survival without liver transplantation at follow-up on day 28 (HR: 2.2, p < 0.001; HR: 2.8, p < 0.001; and HR: 2.6, p < 0.001, respectively). Interestingly, the risk of ARF depended on the dose of metamizole administered (HR: 1.038, p < 0.001). Compared to patients who were treated with opioids, the rate of ARF was higher in the metamizole group (49% vs. 79%, p = 0.014). An increased risk of ARF with metamizole use was also demonstrated in the independent validation cohort (p < 0.001).

Conclusion: Metamizole therapy, especially at high doses, should only be used with a high level of caution in patients with decompensated cirrhosis.

Figure 1

Exploratory cohort: a) 28-day incidence of AKI b) 28-day incidence of severe AKI c) 28-day LTx-free survival Univariate competing risk analysis (a, b) or a log-rank test (c) was used to determine the level of significance. Definition of severe AKI/stage 3 AKI : increase of serum creatinine (sCr) > 3-fold relative to baseline or sCr ≥ 4.0 mg/dL (353.6 µmol/L) with an acute increase ≥ 0.3 mg/dL (≥ 26.5 µmol/L) or dialysis AKI, Acute kidney injury; LTx, liver transplantation

Figure 2

Exploratory cohort: 28-day incidence of AKI a) cumulative 7-day dose of metamizole 1 week prior to baseline (beginning of observation period) b) Metamizole use newly started at baseline Univariate competing risk analysis was carried out to determine the level of significance. AKI, Acute kidney injury; LTx, liver transplantation

Figure 3

Exploratory cohort: 28-day incidence of AKI a) only opioids at baseline compared with only metamizole at baseline Internal validation: 28-day incidence of AKI b) internal validation Univariate competing risk analysis was used to determine the level of significance. AKI, Acute kidney injury; LTx, liver transplantation

eFigure 1

Exploratory cohort: 90-day LTx-free survival The log-rank test was used to determine the level of significance. LTx, Liver transplantation

eFigure 2

a) Pooled NGAL plasma levels, categorized according to metamizole treatment. The medians are shown by horizontal lines. The significance level was calculated using a Mann–Whitney test. b) Longitudinal measurement of plasma NGAL in patients with sample availability over time, categorized according to metamizole treatment. Patients with acute kidney injury were excluded. Altogether, five patients are shown with metamizole initiation and four patients as control group. For metamizole initiation the mean time that elapsed between samples was 90 days (8–257 days). For the control group with no metamizole use the mean was calculated as 64 days (6–148 days). The significance level was calculated with the Wilcoxon matched pairs test. NGAL, Neutrophil gelatinase-associated lipocalin

eFigure 3

Pooled NGAL plasma (a) and urine (b) measurements in patients without acute kidney injury, categorized according to metamizole treatment. In the event of multiple samples for one patient in a subgroup, the sample obtained at the earliest time was chosen. The medians are shown by horizontal lines. The significance level was calculated using a Mann–Whitney test. NGAL, Neutrophil gelatinase-associated lipocalin