Validation of heart rate extracted from wrist-based photoplethysmography (PPG) in the perioperative setting: a prospective observational study. (Preprint)
BACKGROUND Measurement of heartrate (HR) through an unobtrusive, wrist-worn optical HR monitor (OHRM) in the form of a wristband could enable earlier recognition of patient deterioration and timely intervention. OBJECTIVE The goal of this study is to assess the agreement between HR extracted from an OHRM and the golden standard patient monitor during surgery and in the recovery period. METHODS The agreement in beats per minute (bpm) was evaluated between the gold standard electrocardiogram (ECG) from the patient monitor and the HR extracted from the photoplethysmography (PPG) sensor by the OHRM. RESULTS A total of 271.8 hours of data in 99 patients was recorded simultaneously by the OHRM and patient monitor. Median coverage was 86% (interquartile range: 65% to 95%) and did not differ significantly between surgery and recovery (Wilcoxon paired difference test P .17). Agreement analysis showed the limits of agreement (LoA) of the difference between the OHRM and the ECG HR were within the range ± 5 bpm. The mean bias was -0.14 bpm (LoA between -3.08 and 2.79) and -0.19 % (LoA between -4.79 and 4.39) for the PPG measured HR compared to the ECG measured HR during surgery and -0.11 bpm (LoA between -2.79 and 2.59) and -0.15 % (LoA between -3.92 and 3.64) during recovery. CONCLUSIONS An OHRM can obtain good quality signals for HR in the majority of the patients for most of the time in the perioperative setting. OHRM HR monitoring is within the acceptable range compared to ECG derived heartrate, which implies that an OHRM can be considered clinically acceptable for heart rate monitoring in low acuity hospitalized patients. CLINICALTRIAL This non-randomized trial was not registered in a trial registration.