Occult renal impairment is common in patients with peripheral vascular disease and normal serum creatinine.

OBJECTIVE The incidence of peripheral vascular disease (PVD) and angiography/angioplasty is rising annually. The UK Small Aneurysm Trial and other trials have shown renal function is a predictor of increased mortality and failed infrainguinal bypass despite patent vessels. Renal function is classically assessed by serum creatinine (SCr). However, SCr can be normal despite significant renal impairment. A more sensitive test is creatinine clearance (CrCl) as determined by 24-hour urine collection in combination with SCr. We studied the incidence of renal impairment, as defined by CrCl, in PVD patients with normal SCr. METHODOLOGY All patients with PVD sufficient to necessitate angiography and normal SCr (< or =120 micromol/l - men; < or =97 micromol/l - women) had their CrCl assessed prior to angiography: using both 24-hour urine collection and the Cockcroft-Gault formula. Various blood tests, a detailed history and examination were performed. A control group of arthritic patients, age and sex-matched with similar SCr, also had their CrCl determined. RESULTS 65 of 76 patients (86%) with normal SCr had a subnormal CrCl (<100 ml/min) and 49 (65%) had a CrCl below 60 ml/min. In the control group of arthritic patients, the proportion having impaired CrCl was significantly less - 67% below 100 mls/min (p=0.0471) and only 15% below 60 mls/min (p<0.0001). The median and interquartile range CrCl of 52 [38-81] mls/min for PVD patients was significantly worse than for control patients (80 [68-119] mls/min -p<0.0001). The Cockcroft-Gault formula for calculating CrCl did not correlate well with the urinary CrCl for the control group but did for PVD patients (p<0.0001). Factors associated with a significantly reduced CrCl were age of at least 75 years, SCr of at least 85 micromol/l and a history of coronary heart disease (all p<0.05). This had a sensitivity of 88% and specificity of 82% for identifying subnormal CrCl. Statin use was associated with a significantly improved CrCl (p=0.040). CONCLUSION Most PVD patients with normal serum creatinine have occult, significantly impaired renal function as defined by creatinine clearance. Vascular surgeons should include creatinine clearance in pre-operative assessment of renal function especially in patients over 75 years old, with a history of coronary heart disease or a serum creatinine over 85 micromol/l. The method of determining creatinine clearance could be the Cockcroft-Gault calculation or ideally 24-hour urinary creatinine clearance measurement. This would allow appropriate early referral to a nephrologist for further investigation and management. It is worth noting that statin use seems to be associated with a protective effect on renal function.