Comparison of arterial and venous implantable Doppler postoperative monitoring of free flaps: Systematic review and meta‐analysis of diagnostic test accuracy

Abstract

Early detection of vascular compromise following free flap microsurgical reconstruction is essential. This can lead to timely reoperations and flap salvage. Both arteries and veins are monitored postoperatively with implantable ultrasound Dopplers based on surgeon preference with no consensus as to which method is best. This systematic review and meta-analysis compared arterial to venous implantable Dopplers for postoperative monitoring of microsurgical free flap reconstruction. Five databases (MEDLINE via PubMed, EMBASE, Cochrane Library, Web of Science, and Scopus) were systematically and independently searched. PRISMA and Cochrane guidelines were strictly followed. Clinical characteristics, donor and recipient sites, specific arterial or venous vessels, diagnostic data, outcomes, and complications were recorded. The seven studies included in the meta-analysis were published from 1994 to 2018, with results from 38 of the possible 64 outcomes and complications queried. A total of 763 flaps with implantable Doppler probes on 527 arteries and 388 veins were included in the study. Compared to patients monitored with venous implantable Dopplers, arterial monitoring was associated with a risk of false-positives reduced by 74% (RR:0.26, 95% CI:0.12, 0.55, I2 = 40%, p = .0004) and risk of signal loss reduced by 63% (RR:0.37, 95% CI:0.24, 0.59, I2 = 26%, p < .0001). Patients being monitored with a venous implantable Doppler did not show significantly different sensitivities, specificities, true-positives, false-negatives, true-negatives, positive predictive values, negative predictive values, time to signal loss, take-backs, salvage rates, flap failures, flap failure rates, arterial, venous, vascular, and hematoma compromise compared to an arterial implantable Doppler. These findings suggest arteries may be the best vessels to monitor when using an implantable Doppler following free flap microsurgical reconstruction. Caution is advised when interpreting these findings for clinical significance due to no significant differences in take-backs, SR, flap failures, and FFR.