Accurate respiratory rate determination using a novel insertable cardiac monitor algorithm: implications for diagnostic and monitoring potentials beyond heart rhythm disorders

Respiratory rate (RR) is a critical vital sign that is highly relevant in patients with cardiopulmonary disorders. The implantable cardiac monitor (ICM) provides useful data pertaining to heart rhythm, but little is known up to this point regarding its potential diagnostic value in the direct measurement of respiratory parameters. The addition of respiration information could improve understanding of overall health status of heart rhythm patients with ICM.

The primary objective of this study was to evaluate the accuracy of the RR detected by an existing implanted ICM as compared to gold-standard polysomnography (PSG) measurement of respiration.

This prospective single center study enrolled 25 patients (17 male, 62.7±12.2 yesrs) with an implanted ICM and suspected sleep-disordered breathing. The ICM was custom configured with research software to collect respiration data (Fig. 1). Simultaneous, time-synchronized PSG and ICM data were evaluated in two-minute epochs episodically during the night. The offline novel prototype RR algorithm was evaluated on episodes collected by the ICM and compared against expert manual adjudicated RR from PSG by two separate investigators. The interobserver agreement was assessed using intraclass correlation coefficient (ICC). The performance of the novel algorithm was assessed using Bland-Altman analysis with 95% limits of agreement (LOA).

A total of 495 epochs were graded by two independent observers, with good ICC of 0.83 (95% C.I. 0.79–0.86). Epochs free of severe sleep disordered breathing/apnea (n=363) were included in this analysis with 106 of these containing periods of hypopnea. The development and validation datasets were comprised of 235 and 128 epochs, respectively. In the development data, the mean RR was 14.99±3 breaths per minute, and the mean RR was 13.44±2 breaths per minute in the test data. Using Bland-Altman analysis, the bias for the novel prototype algorithm was only −0.13 and +0.32 breaths per minute with 95% LOA of −2.24 to +1.98 and −2.56 to +3.19 breaths per minute (Fig. 2), in the in the development and test dataset respectively.

The novel prototype algorithm applied to the ICM data provided accurate determination of respiratory rate as compared to gold-standard PSG data. The capability to determine respiratory rate accurately from an existing ICM platform demonstrates the potential to extend the diagnostic power of ICMs beyond heart rhythm abnormalities to address a broad range of comorbidities including breathing disorders and heart failure.

Type of funding sources: None.