Weight reduction drives improvement in insulin resistance independent of OSA phenotype and CPAP treatment in the RICCADSA cohort
(2026) In Sleep Medicine 144.- Abstract
Background Insulin resistance is common in adults with obstructive sleep apnoea (OSA), yet the extent to which longitudinal metabolic improvement is driven by OSA phenotype, continuous positive airway pressure (CPAP) treatment, or changes in body weight remains unclear. We examined 1-year changes in insulin resistance across OSA and non-OSA subgroups in the non-diabetic RICCADSA cohort. Methods In the RICCADSA study, non-diabetic participants (n = 332) were categorized into four predefined subgroups: non-sleepy OSA randomized to CPAP (n = 72), non-sleepy OSA randomized to no CPAP (n = 81), sleepy OSA treated with CPAP (n = 98), and non-OSA controls (n = 81). Insulin resistance was assessed using log-transformed homeostasis model... (More)
Background Insulin resistance is common in adults with obstructive sleep apnoea (OSA), yet the extent to which longitudinal metabolic improvement is driven by OSA phenotype, continuous positive airway pressure (CPAP) treatment, or changes in body weight remains unclear. We examined 1-year changes in insulin resistance across OSA and non-OSA subgroups in the non-diabetic RICCADSA cohort. Methods In the RICCADSA study, non-diabetic participants (n = 332) were categorized into four predefined subgroups: non-sleepy OSA randomized to CPAP (n = 72), non-sleepy OSA randomized to no CPAP (n = 81), sleepy OSA treated with CPAP (n = 98), and non-OSA controls (n = 81). Insulin resistance was assessed using log-transformed homeostasis model assessment of insulin resistance (ln-HOMA-IR) at baseline and 12 months. Longitudinal changes were analysed using linear mixed-effects models including subgroup, time, and subgroup-by-time interaction, with adjustment for age, sex, body mass index (BMI), hypertension, and cardiovascular medications. Secondary linear regression analyses were performed to identify independent predictors of change in insulin resistance (Δln-HOMA-IR). Results Ln-HOMA-IR decreased significantly over time across the entire cohort (p < 0.001). Neither OSA/CPAP subgroup (p = 0.44) nor the subgroup-by-time interaction (p = 0.89) was significant, indicating parallel improvements in insulin resistance irrespective of OSA phenotype or CPAP treatment. In mixed-effects models, higher BMI and male sex were independently associated with higher ln-HOMA-IR levels (both p < 0.01). In secondary linear regression analyses, reduction in BMI independently predicted improvement in insulin resistance (Δln-HOMA-IR), whereas sex was not associated with metabolic improvement. Conclusions In this non-diabetic RICCADSA cohort, insulin resistance improved over 1 year independent of OSA phenotype and CPAP treatment. Weight reduction, rather than sex or sleep apnoea–related factors, emerged as the principal determinant of metabolic improvement, underscoring the central role of weight management in cardiometabolic risk reduction among adults with and without OSA.
(Less)
- author
- Pihtili, Aylin ; Celik, Yeliz ; Thunström, Erik and Peker, Yüksel LU
- organization
- publishing date
- 2026-08
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Sleep Medicine
- volume
- 144
- article number
- 108976
- publisher
- Elsevier
- external identifiers
-
- scopus:105036187004
- pmid:42030837
- ISSN
- 1389-9457
- DOI
- 10.1016/j.sleep.2026.108976
- language
- English
- LU publication?
- yes
- id
- 413e0641-965f-46d8-ae0f-794994cfd09b
- date added to LUP
- 2026-05-26 14:20:14
- date last changed
- 2026-09-03 04:31:11
@article{413e0641-965f-46d8-ae0f-794994cfd09b,
abstract = {{<p>Background Insulin resistance is common in adults with obstructive sleep apnoea (OSA), yet the extent to which longitudinal metabolic improvement is driven by OSA phenotype, continuous positive airway pressure (CPAP) treatment, or changes in body weight remains unclear. We examined 1-year changes in insulin resistance across OSA and non-OSA subgroups in the non-diabetic RICCADSA cohort. Methods In the RICCADSA study, non-diabetic participants (n = 332) were categorized into four predefined subgroups: non-sleepy OSA randomized to CPAP (n = 72), non-sleepy OSA randomized to no CPAP (n = 81), sleepy OSA treated with CPAP (n = 98), and non-OSA controls (n = 81). Insulin resistance was assessed using log-transformed homeostasis model assessment of insulin resistance (ln-HOMA-IR) at baseline and 12 months. Longitudinal changes were analysed using linear mixed-effects models including subgroup, time, and subgroup-by-time interaction, with adjustment for age, sex, body mass index (BMI), hypertension, and cardiovascular medications. Secondary linear regression analyses were performed to identify independent predictors of change in insulin resistance (Δln-HOMA-IR). Results Ln-HOMA-IR decreased significantly over time across the entire cohort (p < 0.001). Neither OSA/CPAP subgroup (p = 0.44) nor the subgroup-by-time interaction (p = 0.89) was significant, indicating parallel improvements in insulin resistance irrespective of OSA phenotype or CPAP treatment. In mixed-effects models, higher BMI and male sex were independently associated with higher ln-HOMA-IR levels (both p < 0.01). In secondary linear regression analyses, reduction in BMI independently predicted improvement in insulin resistance (Δln-HOMA-IR), whereas sex was not associated with metabolic improvement. Conclusions In this non-diabetic RICCADSA cohort, insulin resistance improved over 1 year independent of OSA phenotype and CPAP treatment. Weight reduction, rather than sex or sleep apnoea–related factors, emerged as the principal determinant of metabolic improvement, underscoring the central role of weight management in cardiometabolic risk reduction among adults with and without OSA.</p>}},
author = {{Pihtili, Aylin and Celik, Yeliz and Thunström, Erik and Peker, Yüksel}},
issn = {{1389-9457}},
language = {{eng}},
publisher = {{Elsevier}},
series = {{Sleep Medicine}},
title = {{Weight reduction drives improvement in insulin resistance independent of OSA phenotype and CPAP treatment in the RICCADSA cohort}},
url = {{http://dx.doi.org/10.1016/j.sleep.2026.108976}},
doi = {{10.1016/j.sleep.2026.108976}},
volume = {{144}},
year = {{2026}},
}