Metamizole and Acute Kidney Injury (M-AKI)

Study type

• Study topic:
  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Safety study (incl. comparative)
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medicinal product name, other:
  Dolamizol, Novalgin, Nolotil
• Study drug International non-proprietary name (INN) or common name:
  METAMIZOLE
• Anatomical Therapeutic Chemical (ATC) code:
  (N02BB02) metamizole sodium

  metamizole sodium
• Medical condition to be studied:
  Acute kidney injury

Population studied

• Short description of the study population:
  First ever inpatient administration of metamizole users in a postoperative setting (i.e., up to 7 days after surgery). Patients who were prescribed metamizole up to 14 days prior to surgery were excluded, as they were likely to be prevalent users. ; Patients on multimodal analgesia (i.e., receiving more than one pain relief medication) within 1 hour of surgery were excluded, because it was considered that these medications were intended to be given together. ; Patients with acute kidney injury or acute kidney disease episodes up to one week prior to surgery were excluded. Patients who had acute dialysis 14 days prior to exposure or on chronic dialysis were excluded, too.
• Age groups:
  • Adult and elderly population (≥18 years)

    • Adults (18 to < 65 years)

      • Adults (18 to < 46 years)
      • Adults (46 to < 65 years)
    • Elderly (≥ 65 years)

      • Adults (65 to < 75 years)
      • Adults (75 to < 85 years)
      • Adults (85 years and over)
• Special population of interest:
  Renal impaired
• Estimated number of subjects:
  11000

Study design details

• Study design:
  New user active comparator cohort study. A temporal relationship between exposure (initiation of metamizole) and outcome (incident AKI), and accurate risk estimation. This study design reduces confounding by indication, prevalent user bias, and fits neatly into the target trial emulation framework.
• Main study objective:
  The primary objective of this study is to assess if metamizole, used in inpatient postoperative settings, increases the risk of acute kidney injury (AKI) compared to opioids or non-steroidal anti-inflammatory drugs (NSAIDs). The primary endpoint is AKI, defined using serum creatinine measurements as per the Kidney Disease: Improving Global Outcomes (KDIGO) for AKI.
• Setting:
  This is a retrospective cohort study conducted at Amsterdam University Medical Centre (Amsterdam UMC) in the Netherlands. Time period of study was between January 2018, 1 and December 2024, 31 were included. ; Patients with multiple hospitalisation episodes were assigned the same patient ID. Inclusion criteria: ; 1) Adults admitted to Amsterdam UMC for more than 24 hours; ; 2) Post-operative pain management setting. ; Exclusion criteria: ; 1) Non-invasive diagnostic procedures or minimally invasive procedures; 2) Kidney transplant in the past year from index date; ; 3) On chronic dialysis; ; 4) Acute dialysis episode within the past 14 days; ; 5) Pre-exposure acute kidney injury or acute kidney disease seven days prior to or on index date; ; 6) No metamizole or comparator use 14 days prior to treatment initiation; ; 7) Pregnant women; ; 8) Childbirth; 9) Patients with a treatment switch or add-on within 1-hour of the initial administration of a pain management drug (i.e., metamizole, NSAIDs, or opioids).; 10) Missing baseline serum creatinine.; 11) Invasive respiratory support.
• Comparators:
  Opioids (i.e., morphine, buprenorphine, oxycodone, piritramide, fentanyl, tramadol, pethidine, hydromorphone, and tapentadol), NSAIDs (i.e., aceclofenac, dexketoprofen, diclofenac, phenylbutazone, flurbiprofen, ibuprofen, indomethacin, ketoprofen, meloxicam, nabumetone, naproxen, etoricoxib, parecoxib, celecoxib, piroxicam, propyphenazone, tiaprofenic acid, high-dose acetylsalycilic acid, and diflusinal).
• Outcomes:
  The primary outcome of interest is AKI occurring during hospitalisation, defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria from 2012, based on changes in serum creatinine (urine output will be considered as a subgroup analysis). ; This outcome is clinically meaningful, well-standardised, and captured in hospital electronic health records from Amsterdam UMC, where outpatient and inpatient data is routinely collected.; Acute kidney injury is a frequent and serious complication in postoperative patients, which is associated with increased morbidity, long-term damage of kidney function, and prolonged hospitalisations. Non-steroidal anti-inflammatory agents are known to contribute to AKI through kidney damage and dysfunction. ; On the other hand, opioids are not known to cause AKI, but they have been associated with respiratory depression and addiction as adverse drug events. Metamizole (dipyrone) is an alternative non-opioid analgesic often thought to have a different kidney safety profile.; Thus, outcome was selected to address a clinically and pharmacologically relevant safety question. This is important given the debate regarding the balance between analgesic efficacy and kidney safety in the postoperative setting.
• Data analysis plan:
  (Generalised) propensity score model: multinomial logistic regression. treatment (i.e., metamizole, NSAIDs, opioids) = confounders. The selected confounders are those which affect the exposure-outcome relationship and outcome relationship. Stabilised inverse probability of treatment weights will be calculated from this model.; ; Censoring models: pooled logistic regression models to estimate the probabilities of remaining uncensored and free of intercurrent events over time, conditional on treatment and relevant time-varying covariates. Stabilised inverse probability weights for censoring and intercurrent events will be derived as relevant and truncated at the 99th percentile.; ; Weighting: Inverse probability weights for censoring will be combined with treatment weights to create composite stabilised weights.; ; Outcome model: cumulative incidence curves will be generated for each treatment group using inverse probability weighting with the composite weights to adjust for confounding, censoring, and intercurrent events. For the controlled direct effect, the estimator corresponds to the complement of a weighted Kaplan-Meier estimator. The risk difference and risk ratio comparing cumulative incidences at 14 days will be calculated.