RMD Open (Feb 2024)

Prognostic factors for liver, blood and kidney adverse events from glucocorticoid sparing immune-suppressing drugs in immune-mediated inflammatory diseases: a prognostic systematic review

  • Abhishek Abhishek,
  • Danielle van der Windt,
  • Christopher Carroll,
  • Matthew J Grainge,
  • Richard Riley,
  • Joanna Leaviss,
  • Munira Essat,
  • Tim Card

DOI
https://doi.org/10.1136/rmdopen-2023-003588
Journal volume & issue
Vol. 10, no. 1

Abstract

Read online

Background Immune-suppressing drugs can cause liver, kidney or blood toxicity. Prognostic factors for these adverse-events are poorly understood.Purpose To ascertain prognostic factors associated with liver, blood or kidney adverse-events in people receiving immune-suppressing drugs.Data sources MEDLINE, Web of Science, EMBASE and the Cochrane library (01 January 1995 to 05 January 2023), and supplementary sources.Data extraction and synthesis Data were extracted by one reviewer using a modified CHARMS-PF checklist and validated by another. Two independent reviewers assessed risk of bias using Quality in Prognostic factor Studies tool and assessed the quality of evidence using a Grading of Recommendations Assessment, Development and Evaluation-informed framework.Results Fifty-six studies from 58 papers were included. High-quality evidence of the following associations was identified: elevated liver enzymes (6 studies) and folate non-supplementation (3 studies) are prognostic factors for hepatotoxicity in those treated with methotrexate; that mercaptopurine (vs azathioprine) (3 studies) was a prognostic factor for hepatotoxicity in those treated with thiopurines; that mercaptopurine (vs azathioprine) (3 studies) and poor-metaboliser status (4 studies) were prognostic factors for cytopenia in those treated with thiopurines; and that baseline elevated liver enzymes (3 studies) are a prognostic factor for hepatotoxicity in those treated with anti-tumour necrosis factors. Moderate and low quality evidence for several other demographic, lifestyle, comorbidities, baseline bloods/serologic or treatment-related prognostic factors were also identified.Limitations Studies published before 1995, those with less than 200 participants and not published in English were excluded. Heterogeneity between studies included different cut-offs for prognostic factors, use of different outcome definitions and different adjustment factors.Conclusions Prognostic factors for target-organ damage were identified which may be further investigated for their potential role in targeted (risk-stratified) monitoring.PROSPERO registration number CRD42020208049.