Background: Rapid Response System (RRS) was designed as a safety tool for early detection and intervention of a deteriorating patient on a general floor in a hospital and Modified Early Warning System (MEWS) scores can be used to identify these patients. Obstructive Sleep Apnea (OSA) has been described as an independent risk factor for long term morbidity and mortality, and STOP-Bang questionnaire has been widely used to screen patients at risk for OSA. The patients with OSA, including patients at high risk for OSA identified by STOP-Bang questionnaire, have higher rates of RRS events, however their outcomes have rarely been studied. This study is an attempt to identify STOP-Bang score-based risk classification as a predictor of outcome of such RRS events.
Methods: This retrospective cohort study was conducted on patients admitted between November, 2014 and June, 2017 at Presence Saint Francis Hospital. STOP-Bang questionnaire which includes four subjective (STOP: Snoring, Tiredness, Observed apnea and high blood Pressure) and four demographic (Bang: BMI, Age, Neck circumference, Gender) items is calculated by nursing staff on each admission to hospital. Data was collected by accessing the charts of patients above the age of 18 years, who had STOP-Bang questionnaire filled at the time of admission and had RRS event during the admission. Patients with STOP-Bang scores between 0 and 2 were identified as low-Risk OSA (LR-OSA) and between 3 and 8 as High-Risk OSA (HR-OSA) and outcomes were compared using these groups. Primary outcome studied was death from any cause or hospice enrollment during the hospital stay. Secondary outcomes included ICU transfer, intubation events, code blue events, duration of mechanical ventilation, duration of ICU stay and duration of hospital stay. Analysis was done with SPSS using chi-square test, Fisher’s exact test and Student’s t test. 2-tailed p value < 0.05 was considered significant.
Results: Out of 519 patients who had RRS events in the study period, 297 patients were eligible for this study. 147(49.49%) were males and 150(50.51%) females with an overall mean age of 67.88±18.65 years. 97(32.66%) patients were identified as LR-OSA and 200(67.34%) were HR-OSA. Patients who were HR-OSA were significantly older than LR-OSA (71.31±15.5 yrs. vs 61.21±22.33 yrs. p<0.0001). Mean STOP-Bang score for LR-OSA was 1.57±0.63 and HR-OSA 3.97±1.19 (p<0.0001). MEWS at the time of RRS and time to RRS from admission was non-significantly lower for HR-OSA vs LR-OSA ((3.86±1.95 vs 4.06±2.10, p=0.426), (76.15±94.14 hrs. vs 82.97±100.14 hrs. p=0.567)). Primary outcome i.e. death or hospice enrollment occurred in 58 (19.53%) patients with non-significantly higher number in HR-OSA vs LR-OSA (38 vs 20, p=0.741). Non-significantly higher number of patients were transferred to ICU (91 vs 37, p=0.230), got intubated (26 vs 12, p=0.879) and had code blue (9 vs 4, p=0.573) in HR-OSA vs LR-OSA groups respectively. Patients with HR-OSA also had non-significant higher duration of mechanical ventilation (0.49±2.04 days vs 0.36±1.46 days, p=0.555), duration of stay in ICU (1.44±2.82 days vs 1.32±2.45 days, p=0.726) and duration of stay in hospital (9.42±8.04 days vs 8.27±5.86 days, p=0.213).
Conclusions: From these results it is apparent that though patients with HR-OSA are older and have more RRS events, their outcomes after RRS do not vary based on their OSA risk assessment including their risk of death or hospice enrollment, transfer or stay in ICU, intubation events, code blue or duration of stay in hospital.