Predictive value of different comorbidity indices for hospitalization due to acute exacerbations in chronic obstructive pulmonary disease patients with comorbidities
10.3760/cma.j.cn114798-20250122-00054
- VernacularTitle:不同评分系统对伴共患疾病慢性阻塞性肺疾病患者急性加重住院风险的预测价值
- Author:
Qinglin CHEN
1
;
Ruoyan ZHANG
1
;
Xiaofang LIU
1
;
Xiujuan YAO
1
;
Yanyun HE
1
;
Ran LI
1
;
Xichun ZHANG
1
Author Information
1. 首都医科大学附属北京同仁医院呼吸与危重症医学科，北京100176
- Publication Type:Journal Article
- Keywords:
Pulmonary disease, chronic obstructive;
Comorbidity;
Hospitalization;
Risk prediction
- From:
Chinese Journal of General Practitioners
2025;24(7):823-833
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To evaluate the predictive efficacy of different comorbidity indices for hospitalization due to acute exacerbations in chronic obstructive pulmonary disease (COPD) patients with comorbidities (CO-COPD).Methods:This retrospective cohort study included 259 stable COPD patients with comorbidities from Beijing Tongren Hospital, Capital Medical University, between October 2021 and September 2023, all with ≥1-year follow-up. Patients were categorized into hospitalized ( n=75) and non-hospitalized ( n=184) groups based on acute exacerbation events. Clinical characteristics, comorbidities, and comorbidity indices, including Charlson Comorbidity Index (CCI), COPD-specific Comorbidity Test (COTE), and comorbidities in chronic obstructive lung disease index (COMCOLD) were compared between two goups. Risk facors of hospitalization due to acute exacerbations were analyzed by Cox regression. Modified indices were developed by incorporating additional respiratory comorbidities (asthma, bronchiectasis, lung cancer) weighted by hazard ratios (HRs) from Cox reguression. The predictive performance of different comorbidity indices for hospitalization was assessed by receiver operating characteristic (ROC) curves. Results:Hospitalized patients exhibited lower BMI, FEV 1% predicted, and FEV 1/FVC (all P<0.05), alongside higher modified British Medical Research Coucil (mMRC) scores and COPD assessment test (CAT) scores, eosinophil counts, and Global Initiative for Chronic Obstructive Lung Disease, (GOLD)severity ( t=3.73, Z=-3.43, Z=-2.43, Z=-11.10, Z=-11.32, Z=-1.80, χ2=17.62, all P<0.05); and also higher use rates of inhaled corticosteroid (ICS) and systemic oral corticosteroid (OCS) ( χ2=5.48, 7.15, all P<0.05). The comorbidities of asthma, bronchiectasis, lung cancer, hypertension, coronary atherosclerotic heart disease, anxiety and depression in hospitalized group were significantly higher ( χ2=22.49, 18.30, 15.63, 5.10, 4.68, 7.46, 5.16, all P<0.05), along with the increased CCI and COTE index ( P<0.05). Comorbid asthma, bronchiectasis, and lung cancer were independent risk factors for hospitalization ( HR=1.841, 2.924, and 2.076, respectively; all P<0.05). Original CCI and COTE showed moderate predictive value ( AUC=0.609 and 0.655), while modified CCI, COTE, and COMCOLD demonstrated improved performance ( AUC=0.730, 0.760, and 0.713, respectively). At optimal cutoffs (modified CCI>3.5, COTE>4.5, COMCOLD>6.5), sensitivities were 61.3%, 76.0%, and 58.7%, with specificities of 70.1%, 61.4%, and 72.3%. Age-stratified analysis revealed enhanced predictive utility of modified indices across age groups. Conclusions:CCI, COTE, and COMCOLD provide modest predictive value for hospitalization in CO-COPD. Modified indices incorporating respiratory comorbidities significantly improve risk stratification, offering clinical utility for identifying high-risk patients in primary care settings.
