No abstract

Introduction Polysomnography (PSG) is the gold standard for the diagnosis of obstructive sleep apnea (OSA). Given cost, insurance restrictions and in some cases limited access to sleep center testing, the use of home based sleep apnea testing is becoming increasingly more common. A proportion of patients with technically adequate HSAT who are negative end up having significant disease on PSG. The characteristics of patients who are found to have moderate to severe sleep apnea on polysomnogram (PSG) after a negative home sleep apnea test (HSAT) are not known. We aim to phenotype these patients. Methods We conducted a retrospective chart review from March 2018 to February 2020. A total of 953 adult patients (18 years old and older) underwent HSAT, 248 tests resulted negative (apnea-hypopnea index <5/h). Out of the negative HSAT, 17 patients had moderate to severe obstructive sleep apnea on PSG. Those were included for analysis. Data on patient characteristics such as age, body mass index (BMI), gender, STOP-BANG, ESS and comorbidities was gathered. Respiratory disturbance index, recording time, flow time, oximetry time on HSAT was recorded. PSG recording time, baseline AHI, supine AHI and non-supine AHI were also noted. Technically inadequate HSAT were excluded from analysis. Results The percentage of patients with negative HSAT who were found to have moderate to severe sleep apnea on PSG and were included for analysis was 6.85% (n17). Mean age was 41 years. Mean BMI was 33 kg/m2. Common comorbidities were hypertension (29%), asthma (17.6 %), depression (17.6%), anxiety (11.7%) and reflux (5.9%). Average ESS was 11.7 and STOP-BANG was 3.8. The mean recording time was 477 minutes, flow time 391 minutes and oximetry time was 426 minutes on HSAT. Average PSG recording time was 433 minutes. Average AHI was 24 with supine being 33.2/h and non-supine 17.9/h. Conclusion A proportion of patients with negative HSAT have moderate to severe OSA on follow-up polysomnogram. These patients were young, with lower-class obesity, more positional OSA, and no associated complex comorbidities. Re-evaluation of current diagnostic algorithms and further research is needed to phenotype this at-risk group, as first-line PSG may be more cost-effective and efficient. Support (if any):