Chest

Chest

Volume 128, Issue 4, October 2005, Pages 2130-2137
Journal home page for Chest

Clinical Investigations
Prevalence of Positional Sleep Apnea in Patients Undergoing Polysomnography

https://doi.org/10.1378/chest.128.4.2130Get rights and content

Study objectives

The primary aim of this study was to determine the prevalence of positional obstructive sleep apnea using a functional definition. Positional sleep apnea was defined as a total apnea-hypopnea index (AHI) ≥ 5 with a > 50% reduction in the AHI between the supine and nonsupine postures, and an AHI that normalizes (AHI < 5) in the nonsupine posture. A secondary aim was to determine if positional sleep apnea can be diagnosed accurately during a split-night study.

Design

Retrospective chart review.

Setting

Two sleep centers in Buffalo, NY, one a Veterans Affairs Western New York Healthcare System Sleep Center (VAWNY) and the other a freestanding ambulatory center (Associated Sleep Center [ASC]).

Patients

Three hundred twenty-six patients from the VAWNY, including 57 patients who underwent a split-night study and 242 patients from the ASC who underwent polysomnography.

Interventions

None.

Measurements

Patient characteristics and sleep study results.

Results

Positional sleep apnea was seen in 49 of 99 patients (49.5%) with mild sleep apnea (AHI, 5 to 15/h), 14 of 72 patients (19.4%) with moderate sleep apnea (AHI, 15 to 30/h), and 5 of 77 patients (6.5%) with severe sleep apnea (AHI > 30/h). Sufficient sleep (> 15 min) in both postures was not seen in 104 of 269 patients (38.7%) and 80 of 242 overnight studies (33.1%) at the VAWNY and ASC, respectively, and was not seen in 47 of 57 split-night studies (82.5%). The percentage of studies with insufficient sleep in both postures was significantly greater for split-night studies (p < 0.0001).

Conclusions

Positional sleep apnea is common particularly in patients with mild disease. Positional sleep apnea cannot usually be assessed during a split-night study.

Section snippets

Subjects

The sleep records of consecutive patients referred for overnight polysomnography to rule out OSA at the Veterans Affairs Western New York Healthcare System Sleep Center (VAWNY) from January 8, 2001, to November 12, 2002, and from an Associated Sleep Center (ASC) from March 6, 2003, to August 8, 2003, were reviewed for study. Two hundred sixty-nine patients who underwent all-night diagnostic polysomnography and 57 patients who underwent a combined diagnostic and CPAP titration split-night study

Results

The prevalence of positional sleep apnea at the two centers is shown in Table 1 and Figure 1. Not surprisingly, the percentage of studies in which positional sleep apnea could not be evaluated was significantly greater during split-night studies (p < 0.0001) than during a full diagnostic night study. Positional sleep apnea was more common in VAWNY patients who underwent full-night studies compared to ASC patients. However, when patients were classified according to sleep apnea severity, there

Discussion

Positional sleep apnea as defined by a > 50% reduction in the AHI between the supine and nonsupine postures, and an AHI in the nonsupine posture < 5/h was relatively common, being identified in 68 of 248 patients (27.4%) with an overall AHI ≥ 5/h in whom sufficient sleep in both postures was observed. Positional sleep apnea was significantly more common when sleep apnea was mild (49.5%) than when it was moderate (19.4%) or severe (6.5%). When split-night studies were performed, sufficient sleep

References (25)

  • LloydSR et al.

    Physiologic basis of therapy for sleep apnea [letter]

    Am Rev Respir Dis

    (1987)
  • DyonzakJ et al.

    Prevalence of positional differences in obstructive sleep apnea [abstract]

    Sleep Res

    (1993)

    • Cited by (0)

    Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal.org/misc/reprints.shtml).

    View full text
    Copyright © 2005 The American College of Chest Physicians. Published by Elsevier Inc. All rights reserved.