Campos-Rodriguez · Sleep medicine 2016 · observational cohort study · n=2055

Impact of different hypopnea definitions on obstructive sleep apnea severity and cardiovascular mortality risk in women and elderly individuals.

Cited 33 times in the scientific literature.

Level 3 - non-randomized controlled study

Clinical cohort study assessing prognostic associations across diagnostic criteria

PubMed 27938920 · doi:10.1016/j.sleep.2016.05.020 · record verified 2026-08-29

What was done

Researchers evaluated two Spanish clinical cohorts referred for suspected obstructive sleep apnea (OSA) between 1998 and 2007: 1,116 women and 939 elderly individuals (2,055 total). Using a calibration model, they compared three hypopnea definitions (30–90% flow reduction for ≥10 s): AHI 4% (≥4% oxygen desaturation), AHI 3% (≥3% desaturation), and AHI 3%a (≥3% desaturation or an event-related arousal). They assessed the impact on OSA severity classification and multivariable-adjusted cardiovascular mortality risk.

What was found

In both cohorts, the prevalence of severe OSA (AHI ≥30 events/h) increased by 14% under AHI 3%a compared to AHI 4%. In women, the percentage with AHI <5 events/h decreased from 13.9% with AHI 4% to 1.1% with AHI 3%a. In fully adjusted multivariable analyses, AHI ≥30 events/h was associated with increased cardiovascular mortality risk in women across all three definitions. In elderly individuals, AHI ≥30 events/h was associated with cardiovascular mortality under AHI 4% and AHI 3%, but not under AHI 3%a. Exact hazard ratios, confidence intervals, and p-values were not reported in the abstract.

Why it matters

Adopting more liberal hypopnea scoring criteria inflates OSA severity prevalence and appears to dilute the prognostic association between severe OSA and cardiovascular mortality in elderly populations.

Limits

The abstract reports no numerical effect sizes, confidence intervals, p-values, or follow-up duration. Hypopnea definitions were applied using a calibration model rather than direct manual re-scoring. Findings are derived from clinic-referred Spanish cohorts and may not generalize to broader community populations.

Cited by