Polygraphy vs. polysomnography: missing osas in symptomatic snorers-a reminder for clinicians

The purposes of this study are to investigate the usefulness of polygraphy (PG) in diagnosing obstructive sleep apnea (OSA) in sleepy/tired snorers compared to polysomnography (PSG) and, further, to search for suspected respiratory arousals in the PG. One hundred eighty-seven adults suffering from sleepiness/tiredness and snoring had undergone ambulant PG and were considered to be normal, using American Academy of Sleep Medicine's 2007 hypopnea criteria A. After approximately 7 months, in-lab PSG was performed using hypopnea criteria B, where arousals are also recognized. Validated questionnaires (Hospital Anxiety and Depression Scale, self-rated general health) were answered. In a subgroup, the sensitivity and specificity were calculated for flow limitation index (FLI) and flattening index (FlatI) in PG compared with the respiratory distress index (RDI) in PSG. Despite the normal PG, at PSG, the median RDI was 11.0 (range, 0-89.1). One hundred sixty-eight out of one hundred seventy-eight (90 %) were found to have at least mild OSA and 119/187 (64 %) with moderate-severe OSA according to the RDI values. The sensitivity and specificity were low (< 70 %) for FLI and FlatI. Forty-nine percent of the patients rated anxiety at borderline or pathological levels, 35 % rated corresponding depression levels, and 45 % rated poor or fair general health. PG was insufficient to rule out OSA when the respiratory events were mainly associated with arousals. Almost half of these patients experience low general health and psychiatric problems. We recommend a full-night PSG when PG is "normal", and patients have symptoms of snoring and sleepiness/tiredness.