Cardiovascular measures display robust phenotypic stability across long-duration intervals involving repeated sleep deprivation and recovery

Bidirectional predictors between baseline and recovery sleep measures and cardiovascular measures during sleep deprivation and psychological stress

For the first time, we investigated bidirectional predictors between baseline and recovery sleep and cardiovascular (CV) measures during total sleep deprivation (TSD) and psychological stress in a five-day experiment with 32 healthy adults (27-53y; 14 females). CV measures were collected in the morning after two baseline nights (B1, B2) and during TSD morning (TSD AM) and evening following psychological stress (TSD PM). Actigraphy assessed sleep during B2 before TSD and the first recovery night (R1) after TSD. Higher B2 wake after sleep onset (WASO) predicted lower TSD PM stroke volume and higher TSD PM systemic vascular resistance index (SVRI), with greater B2 percent sleep predicting inverse relationships, explaining 12.8%-15.9% of the TSD CV variance. Also, higher B2 WASO predicted higher B2 AM SVRI. Furthermore, longer TSD left ventricular ejection time predicted later R1 sleep offset, longer sleep duration, and higher WASO; by contrast, higher TSD AM and TSD PM heart rate predicted earlier R1 sleep offset. TSD CV indices explained 14.8%-24.9% of the R1 sleep variance. Notably, females showed significant predictive bidirectional relationships. Our novel results demonstrate that baseline sleep predicts CV metrics during TSD and psychological stress, and that these metrics predict recovery sleep, underscoring crucial relationships, mechanisms, and biomarkers between sleep and cardiovascular health.

Short-term and long-term phenotypic stability of actigraphic sleep metrics involving repeated sleep loss and recovery

For the first time, we determined whether actigraphic-assessed sleep measures show inter-individual differences and intra-individual stability during baseline (BL) and recovery (REC) phases surrounding repeated total sleep deprivation (TSD). We conducted a 5-day experiment at Months 2 and 4 in two separate studies (N = 11). During each experiment, sleep measures were collected via wrist actigraphy during two BL 8 h time-in-bed (TIB) nights (B1, B2) and during two REC 8-10 h TIB nights (R1, R2). Intraclass correlation coefficients (ICCs) assessed actigraphic measure long-term stability between 2 and 4 months for (1) the pre-experimental phase before BL; and (2) the BL (B1 + B2), REC (R1 + R2), and BL and REC average (BL + REC) phases; and short-term stability at Month 2 and at Month 4; and (3) between B1 versus B2 and R1 versus R2 in each 5-day experiment. Nearly all ICCs during the pre-experimental, BL, REC, and BL + REC phases were moderate to almost perfect (0.446-0.970) between Months 2 and 4. B1 versus B2 ICCs were more stable (0.440-0.899) than almost all R1 versus R2 ICCs (-0.696 to 0.588) at Month 2 and 4. Actigraphic sleep measures show phenotypic long-term stability during BL and REC surrounding repeated TSD between 2 and 4 months. Furthermore, within each 5-day experiment at Month 2 and 4, the two BL nights before TSD were more stable than the two REC nights following TSD, likely due to increased R1 homeostatic pressure. Given the consistency of actigraphic measures across the short-term and long-term, they can serve as biomarkers to predict physiological and neurobehavioral responses to sleep loss.