Feasibility of respiratory rate assessment using the photoplethysmographic signal for risk stratification in the general population

Respiration is an essential autonomic function, with disturbances in respiratory rate potentially indicating an elevated risk profile. Despite its established prognostic significance, respiratory signals have been largely overlooked in risk stratification over the past decades. With the widespread availability of the photoplethysmographic (PPG) signal - an optical measure of blood volume changes in the skin through light absorption - PPG-derived respiratory rate measurement presents a promising, non-invasive method for assessing respiratory function and potential risk. In this study, we investigated the feasibility of using the PPG signal to reliably estimate respiratory rate and its application in risk assessment.

Within a population-based survey in Germany, the INVADE study, participants aged ≥60 years were enrolled between August 2013 and February 2015 and followed for a median of 4 years. A total of 1,332 subjects were examined using a simultaneous recording of respiration with an impedance sensor via ECG and a photoplethysmographic sensor. Amplitude variations and baseline wander were utilized to determine respiratory rate from the photoplethysmogram.

A strong correlation between the conventional impedance-based approach and the PPG-based approach could be observed (r = 0.802, p < 0.001. Figure 1). The corresponding Bland-Altman plot confirmed a sufficient agreement between both approaches (Figure 2). Prognostic analyses consistently demonstrated that individuals with a respiratory rate >18.6 brpm have an increased mortality risk, with the PPG signal slightly overestimating survival rates for higher respiratory rates (Figure 3 Kaplan Meier analysis).

Our results confirm a good agreement between the conventional impedance-based and the PPG-based approach to determine respiratory rate. Prognostic analyses are feasible also with the PPG-based approach. The widespread availability of the photoplethysmographic signal renders it a promising tool for risk stratification in the general population.

Figure 1: Scatter plot showing a strong correlation between the conventional impedance-based approach via ECG and the PPG-based approach (r = 0.802, p < 0.001, N = 1258)

Figure 2: Bland-Altman plot showing a sufficient agreement between respiratory rate measured via ECG signal and via PPG signal (mean = -0,8666, SD = 1,54629)

Figure 3: Kaplan-Meier curves showing survival probability as a function of respiratory rate (<18.6 breaths per minute [brpm] vs. ≥18.6 brpm) for (A) ECG-based measurements and (B) PPG-based measurements. The PPG-signal (B) shows a slight overestimation of survival rates for higher respiratory rates, though it follows the same trend as the ECG signal.