Polysomnography in Transition: Reassessing Its Role in the Future of Sleep Medicine

Abstract

PSG, a cornerstone diagnostic instrument in sleep medicine, is recommended for the diagnosis of numerous sleep disorders and might be used as a benchmark for evaluating therapeutical effectiveness. However, PSG has its limitations, and its usefulness in the future warrants reappraisal. First, it is a complex test that requires highly-trained personnel to correctly place the electrodes, monitor the patient, and manually analyse the data. This constitutes a significant economic burden for both society and healthcare systems. PSG also presents few technical limitations: variability of data between nights and variable reliability of scoring between readers. The results given to patients are also limited to the macrostructure of sleep, risking the loss of important information that escapes detection when relying solely on polysomnographic evaluation based on macro sleep stages. The current sleep scoring guidelines raise some doubts about their ability to capture the dynamic and complex nature of human sleep in both clinical and physiological contexts. On the other hand, advanced PSG analysis can provide key information for diagnosis particularly through the microstructural analysis of NREM oscillatory pattern, characterisation of spindles and slow waves, eye movement density, spectrum analysis with hypnodensity and sleep propensity with odd ratio products (ORP). These elements provide a better understanding of the differences between insomnia and poor sleep perception. Furthermore, these methods take into account the dynamics of sleep states, going beyond the mere distinction of sleep into macro stages, which poorly reflects the dynamic nature of sleep itself and including in the assessment of sleep function all the complex associations that sleep itself has with autonomic and cardiorespiratory variables. These insights will transform the role of technicians and clinicians in PSG analysis, with a shift towards training in digital data analysis and algorithms to better inform patients about their PSG results.