Impaired sleep efficiency predicts adverse prognosis in elderly pancreatic cancer patients: a retrospective case - control study based on wearable smart devices.

BACKGROUND: Pancreatic cancer remains one of the most aggressive and lethal malignancies, with particularly poor prognoses in the elderly. Recent research has highlighted the role for lifestyle factors, including sleep, in cancer prognosis. This study aimed to evaluate the effect of sleep duration and quality, as measured by wearable smart devices, on the prognosis of pancreatic cancer in the elderly. METHODS: This retrospective case-control study included 200 elderly patients diagnosed with stage I pancreatic cancer who underwent first-line chemotherapy and Whipple surgery. Sleep metrics were recorded using the WHOOP Strap 2.0 device, and subjective sleep quality was assessed via the Richards-Campbell Sleep Questionnaire (RCSQ) and Pittsburgh Sleep Quality Index (PSQI). Patients were divided into good and poor prognosis groups based on postoperative complications. RESULTS: Significant differences were observed in several sleep merasures between good and poor prognosis groups. The poor prognosis group exhibited longer wake after sleep onset (27.60 ± 4.14 minutes vs. 25.90 ± 3.28 minutes, P = 0.002) and reduced sleep efficiency (75 ± 0.20% vs. 74.5 ± 0.30%, P < 0.001). Additionally, time in bed was longer in the poor prognosis group (8.76 ± 0.21 hours vs. 8.60 ± 0.18 hours, P < 0.001). In correlational analysis, sleep efficiency significantly correlated with days of hospitalization (rho = 0.724, P < 0.001). Multivariate logistic regression identified days of hospitalization (OR 6.914, P = 0.018), time in bed (OR 5.489, P = 0.012), first time out of bed (OR 4.414, P = 0.041) and sleep efficiency (OR 26.595, P < 0.001) as independent predictors of prognosis. CONCLUSION: Sleep duration and quality are significantly associated with prognosis in elderly pancreatic cancer patients. Continuous sleep monitoring may inform individualized care strategies to improve clinical outcomes.