Critical discrepancies in oximetry among patients with Raynaud's phenomenon

Systemic sclerosis (SSc) is associated with a high frequency of pulmonary complications, making oxygen saturation critical for monitoring purposes. However, there are significant difficulties, such as skin thickening and edema, digital ulcers, and Raynaud's phenomenon (RP), which may lead to inaccurate and unreliable conventional digital readings. The objective of the present study is to assess the concordance of oxygen saturation at different anatomical sites, including the forehead, which has been proven to be unaffected by vascular changes, such as RP.

Patients with SSc and RP were included. Pulse oximetry was recorded for each patient using a transmittance pulse oximeter on the index finger and pinna; a reflectance pulse oximeter was also used in the supraciliary region and on the dorsum of the hand. These measurements were obtained both with and without RP, and the concordance between them was subsequently determined using the Bland-Altman plot and Lin's coefficient. Additional potential relationships with other factors that may alter the measurement and the rate of unsuccessful measurements at each site were also analyzed.

Sixty-nine patients were included. One third (37%) had interstitial lung disease. Measurements taken in the presence of RP on the fingers, dorsum of the hand, and pinna showed mean oxygenation values of 5%, 7%, and 1.5% lower than measurements taken on the forehead. Lin's coefficients were between 0.1 and 0.2, indicating low concordance. No variations were observed for finger edema and skin thickness. According to the heteroscedasticity analysis, these differences are more marked when saturation levels fall below 93%. Furthermore, 14% of patients had unsatisfactory measurements on the fingers in the presence of RP, although none were observed on the forehead, associated with diffuse skin disease (OR 11.6, 95% CI 1.46-9.1, p = 0.02).

Concordance, measured using different methods, is low across other sites, with greater significance at lower oximetry levels. This affects the study and follow-up of patients with pulmonary complications, such as interstitial lung disease or pulmonary hypertension. Forehead measurements are not affected by vasoconstriction induced by RP and can be an excellent option for patient follow-up. Key Points • Oximetry values measured on fingers affected by Raynaud's phenomenon differ significantly (5%) from those measured on the forehead. • The difference between finger oximetry measurements varies heterogeneously and is significantly greater when values are below 93%. • In patients with Raynaud's phenomenon, finger oximetry is often not feasible due to low perfusion (14%), a limitation not observed with forehead measurements. • Forehead oximetry is not affected by Raynaud's phenomenon and is therefore recommended in patients with this condition.