Estimates of creatinine clearance (Clcr), using five methods, were compared to measured Clcr in morbidity obese patients. Patients whose total body weight (TBW) was greater than or equal to 195% of their ideal body weight (IBW) and who had urinary Clcr measured with 24-hour urine collections were studied. The 12 men weighed an average of 180 kg; the 31 women averaged 138 kg. The methods studied considered a combination of the following factors: weight, age, sex, and serum creatinine. One method was a nomogram; TBW could not be used because of limitations of the nomogram. Both TBW and IBW were used in two methods that considered weight as a variable. Measured Clcr averaged 105.7 ml/min/1.73 sq m body surface area. All estimating methods using IBW significantly underestimated Clcr; all methods using TBW significantly overestimated Clcr. Two methods that did not include body weight as a variable predicted Clcr more accurately than did other methods, but there was still a significant difference between estimated and measured Clcr. None of the Clcr estimation methods is accurate in morbidly obese patients.