Accuracy of ultrasonography for renal artery stenosis: a meta-analysis of diagnostic test studies

G.Williams ¹, T.Karplus ², E.Hodson ¹, W.Yung ², P.Macaskill ³, S.Chan³, J Craig ¹
1. The Centre for Kidney Research, The Children’s Hospital at Westmead
2. Dept Vascular Medicine, Concord Hospital
3. Dept Public Health and Community Medicine, University of Sydney

Although duplex ultrasonography (DUS) is a widely done triage test for renal artery stenosis (RAS), the overall accuracy of the test and the individual parameters measured (peak systolic velocity ­ PSV etc) are ill-defined. Medline (1966-2001) and Embase (1980-2001) were searched using a comprehensive search strategy. Studies were also obtained from reference lists and contacts with known investigators. Eligible studies compared any DUS parameter with the reference standard, renal angiography, for the diagnosis of RAS. Standard methods for diagnostic test meta-analysis were used including quality assessment, summary receiver operating characteristic curves, multiple linear regression with covariate modelling for diagnostic odds ratios (DOR = the odds of disease in test positives/odds of disease in test negatives) to determine sources of heterogeneity and pooling of the DOR using a random effects model . All steps involved two independent reviewers. From 1251 studies identified, 63 were eligible. Overall study quality was sub-optimal with verification bias, insufficient detail for methods and failure to state blinding of investigators being common. Much of the considerable variability in DOR between studies could not, however, be explained. Pooled estimates of the DOR for PSV, RAR and AT were 58.4 (95%CI 29.4 - 115.7), RAR 22.4 (10.2 - 49.2), AT 82.4 (20.1 - 338.9) but paired data (4 studies) showed that PSV was more accurate than RAR (relative DOR 2.4; 95%CI 1.7 to 4.9). Combining tests did not improve test performance. In conclusion, PSV at a threshold of Hb> 200cm/s is probably the most accurate test for RAS. In 1000 people with suspected RAS, assuming 100 have disease, PSV would correctly diagnose 87, 13 would be missed, 93 would undergo an unnecessary angiography, and 807 would avoid unnecessary angiography.

Presented at the Australian Epidemiological Association, 11th Annual Scientific Meeting, Wellington, New Zealand, 5-6 Sept 2002.

Correspondence
Gabrielle Williams
Centre for Kidney Research
The Children’s Hospital at Westmead
Locked Bag 4001
Westmead NSW 2145 Sydney Australia
Tel: +61 2 9845 3041
Fax: +6 12 9845 3038
Email:Gabrielle Williams