IRVINE, CALIF. -- Masimo (NASDAQ: MASI) announced the findings of an exploratory analysis for a unique feasibility study published in CHEST Critical Care in which Dr. Andrew Goodwin and colleagues at the Medical University of South Carolina (MUSC) found that Masimo SET® pulse oximetry performed accurately among critically ill adult patients of all skin tones, including those with low perfusion requiring vasopressors - and reported zero occult hypoxemic events.[1]

Such accuracy studies are typically conducted in controlled laboratory conditions with healthy volunteers, or using retrospective data from hospitalized patients. For the first time, the authors demonstrated the feasibility of conducting a technically rigorous accuracy study inside a hospital environment where arterial blood saturation (SaO2) values invasively obtained at clinical discretion were compared with time-synchronized pulse oximetry saturation (SpO2) values. This unique design allowed SaO2-SpO2 accuracy comparisons to be made at time points that were clinically indicated - initiated by treatment teams in keeping with established practice, both during routine measurement and when spontaneous changes in a patient’s condition necessitated a blood draw - which by definition are important moments to include in any rigorous evaluation of a pulse oximeter’s accuracy in real-world use.

Once complete, the MUSC team’s work will be the largest study of its kind to prospectively examine pulse oximetry performance and the frequency of occult hypoxemia among critically ill adults across the spectrum of light and dark skin tones in a real-world hospital environment. The findings are consistent with previous laboratory and retrospective hospital-based studies which demonstrated Masimo SET®’s accuracy on patients with both dark and light skin tones, even during low perfusion, without clinically significant bias between groups.[2-5]

In this analysis, the researchers demonstrated that the technically rigorous and methodologically sound approach of laboratory studies could be applied in a real-world clinical setting, using a unique design referred to as the Investigation of Skin Pigmentation Effect on Performance of Pulse Oximetry, or INSPIRE. Such technical rigor was evaluated using data collected from the first 50 of a planned 502 patients. In this dataset, they found that Masimo SET® pulse oximetry was accurate among patients of all skin tones—well within FDA specifications for laboratory studies—when compared with arterial blood samples. The modest median bias between SpO2 and SaO2 was not clinically significant, and nor was there a clinically significant difference between the median biases for light and dark skin tone patients.

The authors set out to design a study that would address the limitations and methodologic shortcomings often observed in other studies. These include self-reported race as a surrogate for skin pigmentation, retrospective designs that preclude detailed assessment, lag times between SpO2 and SaO2 collection, inability to specify or standardize sensors or device models, and failure to confirm adequate SpO2 signal quality. Laboratory studies performed in healthy volunteers, while technically rigorous, are often conducted without typical real-world accuracy confounders, such as patient motion, low perfusion related to underlying critical illness, medical comorbidities such as anemia, and more. Collectively, the authors note, these limitations “inhibit the ability to make conclusions about the impact of skin pigmentation on device accuracy.” By contrast, INSPIRE was “designed to overcome these challenges while examining the clinically relevant outcome of occult hypoxemia [a condition in which dangerously low oxygen saturation is not reflected in pulse oximetry readings] in a population vulnerable to this diagnostic error.”

In the present analysis, the researchers enrolled 107 adult patients at two ICUs within the MUSC main hospital from January to July 2024, which yielded a sample size of 50 patients who had at least one clinically indicated blood draw collected while the study team was present. Skin pigmentation was assessed using a validated noninvasive hand-held skin colorimeter, with patients grouped into dark and light skin tone categories based on individual topology angle (ITA). SpO2 readings were collected using Masimo RD SET® sensors connected to Rad-97® Pulse CO-Oximeters®.

From among the 50 patients and 84 blood samples in the sample size, the researchers found that 80 paired SpO2-SaO2 readings from 49 patients met their technical criteria for inclusion. The patients also met INSPIRE’s skin pigmentation requirement, representing a wide range of ITA measurements and all 6 ITA categories. 26 of the 50 patients were classified as having dark skin pigmentation, with 41 data pairs from patients with dark skin tone and 39 from patients with light skin tone. 31 data pairs (39%) were collected from patients during low perfusion (< 1.4%).

The researchers found that, among the 49 patients and 80 data pairs, Masimo SET® pulse oximetry had overall accuracy of 1.47% root-mean-squared (ARMS*), in keeping with the FDA accuracy specification for Masimo RD SET® sensors in no motion and motion conditions, 1.5% ARMS, and significantly outperforming the industry-standard specification of 3% ARMS.[6] The median bias between noninvasive SpO2 and invasive SaO2 was -0.6%, which is not clinically significant. There was a modest, statistically but not clinically significant, difference of 0.7% between the median bias for dark skin tone and light skin tone patients (-0.2% vs. -0.9%; p=0.0004). While nearly half of the data pairs were collected from subjects at risk of occult hypoxemia, zero occult hypoxemic events occurred.

The researchers concluded that the INSPIRE trial “can feasibly enroll critically ill patients across the full spectrum of skin pigmentation while facilitating precise data collection using standard-of-care workflows in an ICU setting,” noting that “[t]hese data support the study’s feasibility in a real-world clinical setting, and the appropriateness of the chosen methods.”

Data collection for the full INSPIRE trial is still in progress. In the full study, the result of which the researchers hope to publish next year, they will investigate and analyze the effect of skin pigmentation on pulse oximeter accuracy across the complete dataset, including a comparison of occult hypoxemia rates between patients with dark and light skin tones.

Andrew Goodwin, M.D., Principal Investigator and Section Chief of Critical Care at MUSC, commented, “Masimo SET® was chosen for our INSPIRE trial because of its industry-leading accuracy. We are pleased to add to the growing body of evidence for pulse oximetry research in varying skin tones. Our feasibility study has demonstrated that we are closer to a definitive answer regarding how the Masimo SET® sensor performs in a real-world ICU environment. Although not yet definitive, our preliminary results suggest that Masimo SET® performs accurately on all skin tones.”

Daniel Cantillon, M.D., Chief Medical Officer at Masimo, added, “The concern for dangerously overestimating blood oxygen levels with pulse oximetry is greatest among critically ill and poorly perfused patients with dark skin tone who are often moving. We’re highly encouraged to see no occult hypoxemia events so far and that the unique INSPIRE study design can prospectively measure that performance in a real-world setting. The reported findings continue to demonstrate top-in-class accuracy. We look forward to the results of the full study.”