Clinical examination alone is neither sensitive nor specific for predicting flap necrosis, so several technologies, including indocyanine green angiography, thermal imaging (using the FLIR ONE), and near-infrared spectroscopy, have been developed to supplement perfusion assessment. This study aims to compare the accuracy of these three methods for intraoperatively predicting clinical flap necrosis in a rat perforator flap model. The authors hypothesized that near-infrared spectroscopy, assessing oxygenation rather than direct perfusion, would yield significantly different predictions.
A 10 × 3-cm epigastric perforator flap was elevated in 14 adult male rats weighing 250 ± 50 g. Flap perfusion was assessed immediately after flap elevation using thermal imaging, near-infrared spectroscopy, and indocyanine green angiography. Measurements were correlated to the clinical endpoint and gold standard of flap necrosis on postoperative day 7.
All three technologies detected significant differences in perfusion along flap length (all p < 0.001), and were associated with significant differences in the odds of developing flap necrosis (all p < 0.001). The areas under the receiver operating characteristic curves were 0.948 for indocyanine green angiography as an absolute value, 0.873 for relative changes with thermal imaging, and 0.792 for tissue oxygenation. The sensitivity, specificity, and accuracy for indocyanine green angiography measured as an absolute value were the highest at 97.8, 87.5, and 92 percent, respectively.
Indocyanine green angiography most accurately predicted flap necrosis in this study; however, tissue oximetry and thermal imaging were also capable of predicting necrosis and represented potentially less expensive or more readily available alternatives for objective perfusion assessment. Additional research can further delineate their roles and cost-efficacy in clinical practice.