Why this matters

Pulse oximeters based on photoplethysmography (PPG) have been shown to overestimate oxygen saturation in people with darker skin, particularly at lower saturation levels. This can lead to missed (occult) hypoxemia and delayed treatment. Demonstrating accurate performance across skin pigmentations and across the clinically critical 70 to 100% range is essential for safe and equitable patient monitoring.

How the study was done

27 healthy adult volunteers (14 female, 13 male) wore the CardioWatch 287-2 on the left wrist while breathing a controlled gas mixture that lowered their oxygen saturation stepwise from 100% to about 70%. At each plateau, arterial blood samples were drawn and analysed on three CO-oximeters to provide the gold-standard reference (SaO₂). Pulse rate was compared against a 3-lead ECG. Skin pigmentation was assessed with the Monk Skin Tone Scale and the Individual Typology Angle, giving 11 light, 10 medium and 6 dark-skinned participants. In total, 651 paired data points were analysed. The study design followed current FDA and ISO expectations for pulse oximeter testing.

Key results

For SpO₂, the watch achieved an accuracy (ARMS) of 2.16% with a mean bias of just 0.12% versus arterial blood, and a correlation of R² = 0.92. For pulse rate, ARMS was 2.63 bpm with a mean bias of -0.31 bpm and R² = 0.97.

Performance was consistent across all skin tone groups, with SpO₂ A_RMS below 3% and pulse rate ARMS below 3 bpm in every group:

Accuracy also held up at low saturation. In the 70 to 80% SaO₂ range, where the most critical clinical decisions are made, A_RMS ranged from 2.16% to 2.48% across the three skin tone groups.

In line with earlier evidence

These findings are consistent with a previously published study in which the CardioWatch 287-2 achieved an SpO₂ A_RMS of 1.28% and a pulse rate A_RMS of 2.51 bpm during high-intensity interval training, a setting with significant motion (Vijgeboom et al., BioMedical Engineering OnLine, 2024).

Continuous monitoring beyond the bedside

The CardioWatch 287-2 is a wrist-worn, CE marked and FDA 510(k)-cleared monitor for continuous pulse rate, SpO₂ and skin temperature and intermittent respiratory rate. By providing reliable vital signs across patient populations, it supports earlier recognition of deterioration on general wards and at home, without tethering patients to bedside equipment.

Study limitations: testing was performed in healthy adults aged 18 to 50 in a controlled laboratory setting, which may not fully reflect patient physiology or real-world care settings. The CardioWatch 287-2 is intended only as an adjunct in patient assessment and should not be used as the sole basis for diagnosis or therapy. It is not intended for high-acuity environments, for acutely ill cardiac patients at risk of life-threatening arrhythmias, or for SpO₂ monitoring under high motion or low perfusion.