Resident Physician Medical College of Wisconsin Wauwatosa, Wisconsin, United States
Background: Racial health inequities pose an urgent public health problem and must be addressed to improve health outcomes. A vital piece in reducing inequities in health outcomes is ensuring that our medical devices are reliable for patients of all skin tones. Pulse oximetry is one example that has been shown to be impacted by skin tone. Near-Infrared Spectroscopy (NIRS) monitoring is widely used in anesthesiology and critical care, providing continuous, non-invasive data about the patient’s oxygen delivery. A cerebral NIRS probe placed on the forehead provides data about cerebral oxygen delivery/utilization, and a renal or somatic probe placed on the lower back informs assessment of somatic oxygen delivery. By providing a flow-weighted average of regional oxygen saturation, NIRS provides an effective tool in the early identification of tissue hypoxia/ischemia. Like pulse oximetry, NIRS uses infra-red light to measure oxygenation, so it is possible that NIRS would also be impacted by skin tone. There is need to objectively quantify the patient’s melanin content, rather than dividing patients based on race, an imprecise proxy for skin pigmentation. Colorimeters help provide a more accurate and objective measurement of a patient’s melanin index. The purpose of this study is to better understand the relationship between skin tone and NIRS measurements, so that clinicians can appropriately interpret NIRS data in patients of all skin tones. Objective: To assess the relationship between NIRS measurements and patient skin pigmentation as measured with the melanin index in pediatric patients admitted to the Pediatric Intensive Care Unit (PICU) and the Cardiac Intensive Care Unit (CICU). Design/Methods: Single-center observational cohort study in PICU and CICU patients admitted from January to June 2023 in whom NIRS monitoring was used. We measured the melanin index, a non-invasive measurement of melanin content using a handheld device, the DSM III Colormeter. We correlated the melanin index to the patients’ NIRS values during their ICU stay using the Pearson’s correlation. Results: 46 patients of various racial backgrounds are included in the analysis. The correlation between average melanin index and average cerebral NIRS reading was significantly different from zero (r= -0.51, p=0.0053, Pearson correlation). Similarly, there was a statistically significant inverse relationship between average melanin index and average renal NIRS (r= 0.53, p=0.0002, Pearson correlation).
Conclusion(s): There is a statistically significant inverse relationship between melanin index and cerebral/renal NIRS values.
