CSA Big data summary | ResMed
CSA Big data summary

Central Sleep Apnoea during CPAP therapy: first insights from a big data analysis

Big data is a promising and innovative way to explore questions of clinical relevance, identify disease patterns and characteristics, and generate hypotheses in healthcare. A tremendous amount of data is now available—and growing exponentially—from a number of sources, including telemonitored medical devices that are connected to databases and provide information on device performance and patient status. Analysis of such data may provide new insights and support new approaches to healthcare management.

In a cutting-edge analysis, real-world, de-identified data were used to characterise Central Sleep Apnoea (CSA) during CPAP therapy of US telemonitored patients. The analysis was able to identify 3 categories of CSA during CPAP therapy, all of which negatively affected CPAP therapy compliance and increased therapy termination risk. 1 A second analysis performed on the same database, found that switching patients with persistent or emergent CSA from CPAP to ASV therapy* may improve compliance and thus, potentially patient outcomes.2

Three categories of CSA during CPAP therapy were identified: emergent, transient, and persistent CSA

The study “Trajectories of CSA during CPAP therapy” looked at de-identified data from 133,000 telemonitored patients treated for sleep disordered breathing (SDB) with ResMed Positive Airway Pressure (PAP) devices in the US in 2015.1 New information about the natural history of CSA during CPAP therapy was discovered using repeated measures based on real-life telemonitoring data rather than single “snapshots” of CSA.

CSA occurred in 3.5% of patients; 3 categories of treatment associated CSA were identified1: emergent (20%), transient (55%), and persistent (25%) CSA.



Each category is associated with decreased compliance and increased therapy drop-out risk1

The presence of CSA was associated with decreased CPAP usage hours and increased likelihood of treatment discontinuation, as compared with OSA. The probability of continuing CPAP therapy on day 300 was 83% for OSA, and 79%, 76% and 72% for transient, persistent and emergent CSA, respectively.

The hazard ratios for therapy termination for the 3 CSA groups were 1.3, 1.5, and 1.7, respectively.

These findings were consistent using either the European Respiratory Society or the US definition of persistent CSA (AHI ≥15/h or CAI ≥5/h).

Switching from CPAP to ASV in patients with emergent or persistent CSA may improve adherence2

A secondary analysis showed that compliance in patients with emergent or persistent CSA who switched from CPAP to ASV improved immediately after the switch was made. There was a +22% adherence improvement in the two patient subgroups that switched from CPAP to either fixed (n = 127, p < 0.05) or variable (n = 82, p < 0.01) EPAP ASV.2 The average AHI before the CPAP to ASV switch among patients with emergent or persistent CSA was 17.34/hour as compared with 4.1/hour after the switch.

The data suggest that if CSA persists after 2 weeks, the patient fits within the trajectory of emergent or persistent CSA and may benefit from a switch to ASV.*



The study was led by an external international committee of sleep and respiratory experts: Jean-Louis Pépin (France), Holger Woehrle (Germany), Atul Malhotra (USA), and Peter Cistulli (Australia).



* ASV therapy is contraindicated in patients with chronic, symptomatic heart failure (NYHA 2-4) with reduced left ventricular ejection fraction (LVEF ≤ 45%) and moderate to severe predominant central sleep apnoea.

  1. Liu et al. Trajectories of Emergent Central Sleep Apnea During CPAP therapy. Chest. 2017;152(4):751-60.
  2. Pépin et al. Adherence to Positive Airway Therapy After Switching From CPAP to ASV: A Big Data Analysis. J Clin Sleep Med. 2018 Jan 15;14(1):57-63. doi: 10.5664/jcsm.6880.