Effect of skin pigmentation on pulse oximetry accuracy in the emergency department

To determine whether pulse oximeter (PO) accuracy and signal quality are affected by level of skin pigmentation.

Observational study in a community hospital ED. Consecutive adult patients undergoing arterial blood gas determination were enrolled into the study. Skin pigmentation was determined by comparison with standardized color swatches under controlled lighting; assigned values were used to stratify patients into 3 groups (light, intermediate, and dark) using predetermined criteria. Simultaneous with arterial blood sampling, staff recorded PO reading of O2 saturation using the Nellcor D-25 oximeter. PO values were compared with criterion standard values measured using a 4-wavelength spectrophotometer or co-oximeter. PO signal quality also was recorded. Bias (the mean difference between PO and co-oximeter-measured values of hemoglobin saturation) and precision (the standard deviation of the bias) were calculated. Groups were compared using one-way ANOVA, Bartlett's test for variances, and chi2 test.

O2 saturation data were obtained for 284 patients. Bias values did not differ between the 3 skin pigment groups (p = 0.79). Precision was of borderline significance (p = 0.05), but there was no dose-response relation between skin pigmentation and precision. Study personnel reported suboptimal PO function most often among patients in the dark group (p = 0.003), but this finding was of no clinical significance. PO signal failure was rare (<1% of all patients).

Although several prior studies suggest the contrary, this study found that skin pigmentation does not affect the bias or precision of pulse oximetry. Furthermore, skin pigmentation has no clinically significant effect on PO signal quality.