Agreement between noninvasive hemoglobin monitoring (SpHb) and laboratory hemoglobin in patients undergoing free-flap reconstruction for oral squamous cell carcinoma.

BACKGROUND

Surgery for head and neck squamous cell carcinoma (HNSCC) often involves extensive resection with free-flap reconstruction, leading to significant blood loss. Accurate intraoperative hemoglobin (Hb) monitoring is crucial to guide transfusion. Non-invasive Hb monitoring (SpHb) offers continuous Hb measurement; however, its reliability in major oncological surgeries remains uncertain. Thus, this study evaluated the agreement between SpHb and point-of-care testing-measured Hb (POCT-Hb).

METHODS

In this retrospective study, patients undergoing oral squamous cell carcinoma (SCC) resection with free-flap reconstruction (Sep 2023-Oct 2025) were analyzed. Hb levels were measured at the time of anesthesia induction (T1) and surgical closure (T2) using POCT and simultaneous SpHb. The postoperative central-laboratory Hb level validated the POCT accuracy. Correlation (Pearson), intraclass correlation coefficient (ICC), and agreement (Bland-Altman) analyses were performed. Effects of intraoperative blood loss and dopamine infusion on SpHb accuracy were assessed using subgroup analyses, t-tests, and a univariate general linear model.

RESULTS

Fifty-five patients were included. SpHb showed a moderate correlation with POCT-Hb at T1 (r = 0.619, ICC = 0.563) and T2 (r = 0.639, ICC = 0.563), with wider Bland-Altman limits at T2. Furthermore, changes in SpHb underestimated ΔHb compared with POCT (r = 0.439). Intraoperative blood loss did not significantly affect SpHb accuracy, whereas dopamine use was associated with larger absolute differences and wider limits of agreement. The dopamine level remained an independent predictor of SpHb-POCT discrepancy.

CONCLUSION

SpHb showed moderate agreement with POCT-Hb and was affected by dopamine use. Although its absolute accuracy is limited, SpHb may serve as a useful adjunct for trend monitoring and for guiding the timing of laboratory testing.