Use of digital pressure measurements for the diagnosis of AV access-induced hand ischemia
Department of Surgery
Medical Subject Headings
Adult; Aged; Arteriovenous Shunt, Surgical; *Blood Pressure; Brachial Artery; Case-Control Studies; Female; Fingers; Hand; Humans; Ischemia; Logistic Models; Male; Middle Aged; Photoplethysmography; Predictive Value of Tests; Prospective Studies; ROC Curve; *Renal Dialysis; Sensitivity and Specificity; *Signal Processing, Computer-Assisted
Diagnosis of hand ischemia secondary to an arteriovenous (AV) hemodialysis access is predominantly based on clinical findings. Occasionally, this diagnosis is unclear and more objective diagnostic testing is needed. No validated method of objective testing currently exists. The purpose of this study is to evaluate digital pressure measurements in the diagnosis of AV access-induced hand ischemia. A case-control study of 40 patients with an upper extremity AV access consisted of 15 patients (38%) with clinical AV access-induced hand ischemia and 25 patients (62%) without clinical evidence of ischemia. Arterial pressure in the third finger ipsilateral to the AV access was measured by digital photoplethysmography. Basal digital pressure (BDP), digital to contralateral brachial index (DBI), and change in digital pressure with access compression (CDP) were determined. Logistic regression modeling with gains tables and ROC calculations were used to determine optimal test thresholds and to test model discrimination for predicting AV access-induced hand ischemia. The mean BDP (30 vs 102 mmHg; p < 0.001) and mean DBI (0.3 vs 0.8; p < 0.001) were significantly lower in patients with AV access-induced hand ischemia than in those without. The accuracy for determining hand ischemia using a threshold-adjusted BDP of 60 mmHg was 92% (sensitivity 100%, specificity 87%) and using DBI of 0.4 was 94% (sensitivity 92%, specificity 96%). While the mean CDP was significantly higher in patients with hand ischemia (85 vs 40; p < 0.001), the accuracy for predicting ischemia was marginal (78%). In conclusion, a BDP less than 60 mmHg or a DBI less than 0.4 in a patient with an AV access were highly associated with hand ischemia. CDP's marginal discrimination ability likely stems from the presence of asymptomatic physiologic steal in the group without clinical steal. Nonetheless, CDP may be useful in identifying patients with ischemia who will not benefit from an intervention that specifically targets the AV access.
Rights and Permissions
Citation: Vasc Med. 2006 Nov;11(4):227-31.