Abstract
Introduction
The multiple sleep latency test (MSLT) is used to diagnose disorders of hypersomnolence. Although internationally-recognised protocols do not stipulate whether patients should be woken from the preceding overnight polysomnography (PSG), many labs wake their patients for logistic reasons. This study analyses the impact on PSG and MSLT parameters of forced wake (FW) from the overnight PSG compared with unrestricted sleep (US).
Methods
400 consecutive patients (FW=200; US=200) undergoing PSG/MSLT were included and the following parameters were compared: Epworth Sleepiness Scale (ESS), Morningness-Eveningness Questionnaire score (MEQ), PSG total sleep time (TST), wake-up time from the PSG, overall MSLT sleep latency (MSL), individual nap latencies (SLNap 1–4), number of MSLT naps with sleep-onset REM periods (#SOREMP), and percentage of MSLTs with overall MSL<8 minutes (%MSLT<8).
Results
The 2 groups were well-matched for ESS and MEQ. The FW group had more males (49% vs 39%). When compared to FW, patients with US had longer TST (+38 minutes; p=<0.0001), later wake-up time (+52 minutes; p<0.0001), longer MSL (+1.9 minutes; p=0.0049), 50% fewer #SOREMP (p=0.0224), and 16% fewer %MSLT<8 (p=0.0018). SLNap1 increased by 1.5 minutes (p=0.0623), SLNap2 increased by 2.0 minutes (p=0.0067), SLNap3 increased by 0.75minutes (p=0.0533) and SLNap4 increased by 2.5 minutes (p=0.0059).
Discussion
Allowing patients to have unrestricted sleep on the night prior to the MSLT resulted in significantly longer TST, longer sleep latencies during the MSLT, fewer SOREMP and fewer tests with MSL<8 minutes. International protocols should stipulate unrestricted sleep on the PSG prior to the MSLT to improve diagnostic accuracy.