Preliminary development and internal validation of penalized models for 12-month baPWV and cIMT in patients with coexisting mild-to-moderate obstructive sleep apnea and hypertension

Predicting longitudinal vascular measurements in patients with obstructive sleep apnea (OSA) and hypertension may help guide future research on vascular risk assessment. We developed and internally validated models for continuous 12-month brachial-ankle pulse wave velocity (baPWV) and carotid intima-media thickness (cIMT), accounting for baseline vascular status. This retrospective cohort included 360 patients with mild-to-moderate OSA and hypertension at one hospital in China. Separate least absolute shrinkage and selection operator models included the corresponding baseline vascular measure without penalization and 34 penalized baseline candidates. Ten-fold cross-validation selected the penalty, and 1,000 bootstrap samples repeated model development and internal evaluation. A composite increase in baPWV of at least 100 cm/s or cIMT of at least 0.1 mm was examined as an exploratory secondary outcome. The full-data primary models retained only the corresponding baseline vascular measurement. Optimism-corrected R-squared values were 0.652 for baPWV and 0.726 for cIMT. Fibulin-1 was not retained in either full-data primary model. In exploratory analyses of a measurement-sensitive binary composite, allowing Fibulin-1 increased the corrected area under the curve by 0.167 in paired resampling. Because original assay quality-control records were unavailable, this apparent increment cannot establish a reproducible biomarker contribution. Baseline vascular measurements accounted for most of the predictive information in the primary models. Fibulin-1 findings remain exploratory because the biomarker’s apparent contribution to the binary endpoint is sensitive to outcome measurement and the assay cannot be independently audited. These internally validated models are not ready for clinical use; standardized measurement and independent validation are needed.

Authors

Institutions

Publication Details

Journal
BMC Cardiovascular Disorders
Published
2026-11-02
DOI
https://doi.org/10.1186/s12872-026-06893-4
Primary Topic
Cardiovascular Health and Disease Prevention
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Preliminary development and internal validation of penalized models for 12-month baPWV and cIMT in patients with coexisting mild-to-moderate obstructive sleep apnea and hypertension

Wei Yuan, Xiang Tang, Xia Sun, Mingfeng Jin et al.
1 links
BMC Cardiovascular Disorders
Cardiovascular Health and Disease Prevention
article

Preliminary development and internal validation of penalized models for 12-month baPWV and cIMT in patients with coexisting mild-to-moderate obstructive sleep apnea and hypertension

Wei Yuan, Xiang Tang, Xia Sun, Mingfeng Jin, Tao Sun
article en

Abstract

Predicting longitudinal vascular measurements in patients with obstructive sleep apnea (OSA) and hypertension may help guide future research on vascular risk assessment. We developed and internally validated models for continuous 12-month brachial-ankle pulse wave velocity (baPWV) and carotid intima-media thickness (cIMT), accounting for baseline vascular status. This retrospective cohort included 360 patients with mild-to-moderate OSA and hypertension at one hospital in China. Separate least absolute shrinkage and selection operator models included the corresponding baseline vascular measure without penalization and 34 penalized baseline candidates. Ten-fold cross-validation selected the penalty, and 1,000 bootstrap samples repeated model development and internal evaluation. A composite increase in baPWV of at least 100 cm/s or cIMT of at least 0.1 mm was examined as an exploratory secondary outcome. The full-data primary models retained only the corresponding baseline vascular measurement. Optimism-corrected R-squared values were 0.652 for baPWV and 0.726 for cIMT. Fibulin-1 was not retained in either full-data primary model. In exploratory analyses of a measurement-sensitive binary composite, allowing Fibulin-1 increased the corrected area under the curve by 0.167 in paired resampling. Because original assay quality-control records were unavailable, this apparent increment cannot establish a reproducible biomarker contribution. Baseline vascular measurements accounted for most of the predictive information in the primary models. Fibulin-1 findings remain exploratory because the biomarker’s apparent contribution to the binary endpoint is sensitive to outcome measurement and the assay cannot be independently audited. These internally validated models are not ready for clinical use; standardized measurement and independent validation are needed.

BMC Cardiovascular Disorders
Affiliated Hospital of Jiangsu University (CN)
Good health and well-being
Openalex Percentile: Top 13%
Cardiovascular Health and Disease Prevention
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon