Marginal benefit of renal biopsy in systemic lupus erythematosus.
The predictive information provided by renal biopsy was assessed for four systemic lupus erythematosus (SLE) patient populations and compared with the predictive information from clinical data without benefit of biopsy, both measured against actual outcome. Renal biopsy results, whether studied by light or electron microscopy, contain important prognostic information. However, the prognostic information from renal biopsies in these patient groups is generally less than that of even the simplest clinical classifications; and when combined with clinical information, the total prognostic content is essentially that of the clinical information alone. Thus, judged by presently available data, the renal biopsy in SLE provides mainly redundant prognostic information. The marginal benefit is the difference between what is known before and after a test. Quantitation of predictive accuracy allows assessment of marginal benefit, that is, the increment in accuracy afforded by an additional test. Costly and potentially hazardous procedures, such as renal biopsy, require reassessment in terms of marginal rather than absolute predictive ability.