'Improved' Creatinine Clearance in a Diabetic, hypertensive Patient: Marker of Renal Disease- or Lab Error?

I would like to comment on Dr Eric Orzeck's response to Dr Higinio Fuentes' question about a diabetic, hypertensive man with an apparent "improvement" in creatinine clearance (CONSULTANT, July 1999, page 1906). The patient's initial creatinine clearance was 41 mL/min/1.73² with a urine volume of 1,070 mL; 3 years later, it was 144 mL/min/1.73 ² with a urine volume of 2,500 mL.

Specifically, Dr Fuentes asked whether the creatinine clearance "improved" because of a larger urine sample. Dr Orzeck responded that the results reflect the progression of renal disease in a diabetic patient. I believe that the findings stem from an error in the collection of urine samples.

To determine whether a 24-hour urine sample is accurately collected. one should measure the total creatinine in the sample. Men produce and excrete roughly 20 mg of creatinine per kilogram of lean body weight over 24 hours. If a 24-hour urine sample contains substantially more or less creatinine than anticipated, the collection may be incomplete.

Since Dr Fuentes' patient weighs 87 kg (192 lb), one would expect a 24-hour urine sample to contain a total creatinine level of about 1.7g. In addition, because of the patient's age (70 years,) , his lean body mass is likely reduced, which allows for even less creatinine production and hence excretion. In the first urine sample, the total creatinine excretion can be calculated to be

1.06 g {(clearance_creatinine X [serum_creatinine]/urine ^{-1 min)} X urine volume}.

However, the second sample contains 3.7g of creatinine, which suggests that it was collected for nearly twice the duration of the first sample.

Other features of this case point to the same conclusion. The patient's creatinine clearance allegedly increased from 41 to 144 mL/min, but the serum creatinine concentration remained the same (1.8 mg/dL,). Serum creatinine levels are logarithmic but inversely proportional to changes in creatinine clearance. Therefore, because the creatinine clearance appears to increase by a factor of nearly 3, one would expect the serum creatinine level to be nearly one-third the baseline value. In this case, the patient's serum level should have decreased from 1.8 to 0.6 mg/dL. The only other way that the serum creatinine concentration could remain unchanged despite a threefold improvement in creatinine clearance would be for the patient to raise his creatinine production (by increasing muscle mass) threefold, which is unlikely for a 7O year-old man.

There is probably little need to repeat this patient's 24-hour urine collection. It is most important to achieve good glucose control-with a goal hemoglobin A_1c of 7% --and optimal blood pressure control.

Dr Perry's creatinine clearance calculations are valid when glomerular filtration rates (GFRs) are in the normal range; they show linear relationships until the GFR falls to 50 mL/min or less. In this range, the urinary creatinine levels diverge, accentuating the differences between low, normal, and high muscle mass.

However, my answer to Dr Fuentes' question focused on the patient's increased GFR, which reflected hyperfiltration. Patients with diabetes mellitus who have worsening proteinuria enter a phase in which renal clearance of creatinine increases before the GFR eventually declines. Hyperfiltration accounts for the increased urinary creatinine level; in this setting, it is impossible to calculate a patient's serum creatinine level using urinary creatinine values that are entered into the equation for determining creatinine clearance.

This patient's two creatinine clearance values were obtained 3 years apart. During this time, as shown by the worsening proteinuria, his renal function significantly changed. The effects of hyperfiltration preclude the reverse correlation of urine and serum creatinine levels.

I agree with Dr Perry that it is extremely important not only to control ambient blood sugar levels by maintaining glycosylated hemoglobin levels at 7% or less, hut also-as the United Kingdom Prospective Diabetes Study has shown-to control blood pressure to no higher than 135/85 mm Hg in all persons with diabetes mellitus¹.

REFERENCE:
1. United Kingdom Prospective Diabetes Study Group. 'Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. Br Med J. 1998:313:703-713.

This article appeared in the January 2000 issue of Consultants: Consultations in Primary Care Vol 40. Issue 1 pp. 21-22.