A new clinical paradigm for the interpretation of NSAIDs-induced laboratory abnormalities: Threshold-symptom...

Nonsteroidal anti-inflammatory drugs (NSAIDs) like Ibuprofen and Diclofenac are often employed but cause minor laboratory test abnormalities in routine examination tests, such as inflammatory markers and liver-renal function tests. They can generate pseudodisease and lead to unnecessary investigation. The present study presents a novel "interference threshold-clinical symptom correlation" model that associates lab abnormality with patient complaints or clinical symptoms after NSAID usage to differentiate drug-induced effect from actual disease in primary care. A mixed retrospective-prospective observational study was conducted in 426 patients from three major primary care centers. Retrospective laboratory data and drug exposure history were determined and clinical symptoms were prospectively monitored after withdrawal of NSAID. Logistic regression and threshold modeling established interference ranges for significant laboratory indices. Model performance was assessed by receiver operating characteristic (ROC) analysis with an area under the curve (AUC) of 0.91 (95% CI: 0.87-0.95). The model reduced unnecessary intervention by 46% in an externally validated cohort. Shortcomings include heterogeneity of NSAID type and dose, no control group and difficulty in standardizing correlation between symptom and threshold. This model, however, provides an efficient, pragmatic tool to improve interpretation of laboratory changes in association with NSAID and it enhances patient safety in primary care.