GOUT-36 prediction rule for inpatient gout flare in people with comorbid gout: derivation and external validation.

Objectives To develop and validate a gout flare risk stratification tool for people with gout hospitalized for non-gout conditions. Methods The prediction rule for inpatient gout flare was derived from a cohort of 625 hospitalized people with comorbid gout from New Zealand. The rule had four items: no pre-admission gout flare prophylaxis, no pre-admission urate-lowering therapy, tophus and pre-admission serum urate >0.36 mmol/l within the previous year (GOUT-36 rule). Two or more items are required for the classification of high risk for developing inpatient gout flares. The GOUT-36 rule was validated in a prospective cohort of 284 hospitalized people with comorbid gout from Thailand and China. Results The GOUT-36 rule had a sensitivity of 75%, specificity of 67% and area under the curve of 0.71 for classifying people at high risk for developing inpatient gout flares. Four risk groups were developed: low (no items), moderate (one item), high (two items) and very high risk (three or four items). In a population with frequent (overall 34%) in-hospital gout flares, 80% of people with very high risk developed inpatient flares while 11% with low risk had inpatient flares. Conclusion The GOUT-36 rule is simple and sensitive for classifying people with high risk for inpatient gout flares. The rule may help inform clinical decisions and future research on the prevention of inpatient gout flares. [ABSTRACT FROM AUTHOR]