Background Serum creatinine level has been reported to be related to the incidence of strokes.

Objective To examine the relationship between serum creatinine level and the severity of extracranial carotid artery atherosclerosis.

Design and methods This is a secondary analysis of data from 88 patients (59 men and 29 women) who had previously had transient ischaemic attacks or minor strokes and had been included in intervention trials of symptomatic carotid disease. Narrowing of internal carotid artery was estimated by angiography. Both internal carotid arteries were measured and the severity was expressed as the sum of percentage stenoses on both sides. The risk profiles of patients with moderate and severe internal carotid artery stenosis were compared.

Results The sex-adjusted mean serum creatinine concentration in those with severe carotid disease was significantly higher than that in those with moderate disease (106.3 ± 3.3 versus 91.3 ± 3.7 μmol/l, P = 0.003), but still within the normal range. The risk of having severe disease was compared with risk of having moderate disease for people ranked by their serum creatinine levels. Univariate logistic regression showed that the odds ratio (OR) for having severe involvement of internal carotid artery was greater for patients in mid-tertile of creatinine values than it was for those in the lowest tertile (OR 6.3, 95% confidence interval 2.0-20.4, P = 0.002). The creatinine levels of patients in these two tertiles were within the normal range. The OR was no greater for patients in the highest tertile of creatinine values, which were slightly elevated above the normal range. These OR did not change after adjustment for age, sex, hypertension or systolic blood pressure, diabetes, smoking and lipid levels.

Conclusion Results of this study demonstrate for the first time that serum creatinine level, even within the range of upper normal or mildly elevated levels, is related to the angiographic severity of internal carotid artery disease in patients who have previously had transient ischaemic attacks and that this relationship is independent of classic cardiovascular risk factors.