Clinical characteristics and prognosis prediction in patients with AECOPD and type 2 diabetes mellitus: A multicenter observational study

Abstract

Objectives

Diabetes is a common comorbidity in COPD population. This study aimed to explore the impacts of T2DM on clinical characteristics and outcomes of patients with exacerbation of COPD, as well as develop a specified prognostic model for these patients.

Methods

AECOPD patients were enrolled from a prospective, noninterventional, multicenter cohort study. Propensity score matching with a 1:2 ratio was performed to compare the characteristics and prognosis between patients with and without T2DM. Predictors for short-term mortality were determined by logistic regression analysis and a prediction nomogram were established and further validated in another cohort.

Results

A total of 1804 AECOPD patients with T2DM and 3608 matched patients without T2DM were included. AECOPD patients with T2DM presented with worse disease profile and prognosis. Eight independent predictors for short-term mortality were determined, including advanced age, disturbance of consciousness, chronic cardiac disease, low blood pressure, high heart rate, elevated neutrophil, urea nitrogen and random blood glucose. A prognostic nomogram was established with an AUC of 0.878 (95%CI: 0.842-0.915) in derivation cohort and 0.834 (95% CI: 0.767-0.901) in validation cohort, which was superior to DECAF (0.647 [95%CI: 0.535-0.760]) and BAP-65 score (0.758 [95%CI: 0.666-0.850]). The calibration curve and decision curve analysis also indicated its accuracy and applicability. Besides, a web calculator based on the nomogram was constructed to simplify the use of prognostic nomogram in clinical practice.

Conclusions

Comorbid diabetes is significantly associated with severe disease profile and worse prognosis in AECOPD population. Our nomogram may help to facilitate early risk assessment and proper decision-making among patients with AECOPD and T2DM.

Keywords

AECOPD Type 2 diabetes mellitus prognostic model

Introduction

Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide with heavy economic and social burden. In 2019, the global prevalence of COPD among population aged 30-79 years was 10.3%, which translates to 391.9 million people.¹ COPD caused 3.2 million deaths annually, accounting for 81.7% of chronic respiratory disease related deaths globally.² Moreover, the prevalence of COPD are projected to keep increasing over the coming decades.³ Acute exacerbation of COPD (AECOPD) is the primary reason for hospitalization among COPD patients, which contribute to compromised quality of life and increased mortality, responsible for the greatest COPD related burden on the healthcare system.⁴

Type 2 diabetes mellitus (T2DM) is one of the most common comorbidities among COPD population.⁵ Due to systemic inflammation, oxidative stress, hypoxia, acidosis, and irrational application of systemic glucocorticoid, COPD patients are at a significantly higher risk of T2DM than general population.^6–9 Besides, there are also reports that diabetes will deteriorate the clinical outcomes of AECOPD patients.¹⁰ Several prognostic tools have been established for AECOPD patients. Among these, DECAF and BAP-65 score are most commonly accepted predictive models for prognosis among general AECOPD patients.^11,12 The DECAF Score is a simple yet effective predictor of in-hospital mortality in AECOPD patients which is composed of dyspnea, eosinopenia, consolidation, acidaemia and atrial fibrillation.¹¹ BAP-65 score only contains four parameters (blood urea nitrogen ≥ 25 mg/dL, altered mental status, pulse > 109 beats/min, age > 65 years) and was proved to be effective in predicting in-hospital mortality and the need for mechanical ventilation in general AECOPD population.¹² However, the predictive performance of these models in subgroups of patients with AECOPD and T2DM is unclear. Besides, currently, there are few researches specifically on diabetic AECOPD patients, and most of them are retrospective design with small sample sizes, and no further analysis were conducted to explore the factors related to poor prognosis in these population. Thus, given the high prevalence of T2DM and its potential adverse impacts on prognosis among AECOPD patients, as well as the heavy medical burden caused by both T2DM and AECOPD, it is crucial to establish a specific prognostic tool for these population.

In this study, we aimed to confirm the influence of T2DM on clinical outcomes in AECOPD population and establish a specified prediction model for short-term mortality, as well as evaluate the performance of existing general prognostic models (including DECAF and BAP-65 score) among patients with AECOPD and T2DM.

Materials and methods

Study design and participants

This was a multicenter observational study. In this research, patients for model development were enrolled from a prospective, noninterventional, multicenter, real-world cohort study in China, the MAGNET AECOPD (MAnaGement aNd advErse ouTcomes in inpatients with acute exacerbation of COPD) Registry study.¹³ For model development, a minimal sample size of 1474 participants was needed to minimize model overfitting and ensure precise estimation with approximately 20 candidate predictors, a Cox-Snell R² value of 0.236, and a reported short-term mortality of 3%–14% among diabetic AECOPD patients^9,14 according to the formulae by Riley et al.¹⁵ The diagnosis of AECOPD was based on the following criteria¹⁶: (1) a medical history of COPD diagnosed by personal history, clinical symptoms and post-bronchodilator forced expiratory volume in one second/forced vital capacity <70% during stable stage of COPD or confirmed by previous medical diagnosis records and (2) an acute worsening of respiratory symptoms requiring additional treatment. The diagnosis of T2DM was verified by previous diagnosis records or made if one of the following conditions is met based on the American Diabetes Association guidelines¹⁷: (1) fasting plasma glucose ≥7.0 mmol/L; (2) 2-h plasma glucose ≥11.1 mmol/L during an 75-g oral glucose tolerance test; (3) Glycosylated hemoglobin A1C ≥6.5%; or (4) random plasma glucose ≥11.1 mmol/L in patients with classic symptoms of hyperglycemia or hyperglycemic crisis. Exclusion criteria included: (1) discharged within 72h of admission, (2) failed to follow up for 2 months. For model external validation, consecutive patients were prospectively enrolled from six of the 10 institutions who participated in the MAGNET study between August 2021 and December 2022 according to the same criteria.

Data collection and study outcomes

A standardized case report form was completed for every participant, including baseline demographics, comorbidities, clinical manifestations and auxiliary examinations within 24h after admission. Chronic kidney disease (CKD) is defined as abnormalities of kidney structure or function, present for a minimum of 3 months, according to the KDIGO guideline.¹⁸ The primary outcome was short-term mortality (60-day mortality), the secondary outcomes included in-hospital mortality, length of hospital stay, intensive care unit (ICU) admission, need for invasive mechanical ventilation (IMV) and hospitalization costs.

Statistical analysis

Categorical variables were presented as number and percentage and compared using the chi-squared test. Continuous variables were summarised as median and interquartile range and analyzed using the Wilcoxon’s rank-sum test.

Considering age, sex and comorbidities might be influence factors for clinical outcomes among AECOPD patients,^5,19,20 propensity score matching (PSM) analysis was performed to balance the differences in the above variables between groups, matching at a ratio of 1:2.

Multiple imputations were conducted to impute missing data if the missing values were less than 20%, and variables with a missing rate of more than 20% were excluded.

Clinical parameters which might have impacts on prognosis of AECOPD patients and were available in our dataset were chosen for evaluation according to existing literature and clinical experience, including demographic characteristics, previous frequency of exacerbations of COPD and routine treatment, clinical manifestations, vital signs, comorbidities, and indicators reflecting organ function status.^5,20 Univariate and multivariate logistic regression analysis were performed to identify independent predictors for 60-day mortality. Multicollinearity among the variables was tested using variance inflation factor (VIF). For better clinical interpretation and more convenient application, continuous variables were categorized (except age) according to the clinically relevant cutoff values. Then, a prognostic nomogram and web calculator was developed based on the results of multivariate logistic regression model.

The prognostic nomogram was internally validated by bootstrap resampling and external validated in another cohort. The overall performance of the model was tested by Brier score, and its discrimination was determined by the area under the Receiver Operating Characteristic (ROC) curve (AUC). Calibration curve and decision curve analysis were conducted to assess the goodness of fit and clinical applicability of the nomogram. Meanwhile, ROC curve analyses of DECAF and BAP-65 score were also performed to compare their predictive capacity for short-term mortality with our nomogram.

All statistical analyses were conducted using R version 4.0.3. or SPSS version 22.0. All statistical analyses were two tailed and p values <0.05 were considered statistically significant.

Results

1804 AECOPD patients with T2DM and 3608 matched patients without T2DM were included in our analysis after PSM (Figure 1(A)). Covariates including age, sex and comorbidities were all balanced between the groups after PSM (Table 1 and Figure S1).

Figure 1.

Patient inclusion flowchart. (A) Derivation cohort. (B) Validation cohort. AECOPD, acute exacerbation of chronic obstructive pulmonary disease; T2DM, type 2 diabetes mellitus.

Table 1.

Comparisons of clinical characteristics and outcomes among AECOPD patients with or without T2DM.

Variables	Overall (n = 5412) n (%) or median (IQR)	AECOPD patients with T2DM (n = 1804) n (%) or median (IQR)	AECOPD patients without T2DM (n = 3608) n (%) or median (IQR)	p Value
Demographics
Age (year)	75.0 (68.0, 82.0)	75.0 (68.0, 82.0)	75.0 (69.0, 82.0)	0.526
Male	3900 (72.1)	1306 (72.4)	2594 (71.9)	0.700
BMI (kg/m²)	22.0 (19.8, 25.0)	22.9 (20.2, 25.8)	21.9 (19.4, 24.4)	<0.001
Current smoking	344 (19.1)	330 (19.1)	14 (17.7)	0.755
≥2 acute exacerbations of COPD in the last 12 months	1553 (28.7)	546 (30.3)	1007 (27.9)	0.071
LTOT	256 (14.2)	243 (14.1)	13 (16.5)	0.555
Long-term OCS therapy	175 (3.2)	53 (2.9)	122 (3.4)	0.385
Clinical manifestations
Fever	675 (12.5)	260 (14.4)	415 (11.5)	0.002
Disturbance of consciousness	242 (4.5)	92 (5.1)	150 (4.2)	0.113
Heart rate (beat/min)	88 (78, 99)	88 (78, 100)	88 (78, 98)	0.164
SBP (mmHg)	133 (122, 148)	135 (123, 149)	133 (122,147)	0.068
DBP (mmHg)	78 (69, 86)	78 (69, 86)	78 (69, 86)	0.149
Comorbidities
Hypertension	3196 (59.1)	1055 (58.5)	2141 (59.3)	0.545
Chronic cardiac disease	2667 (49.3)	883 (48.9)	1784 (49.4)	0.729
Chronic kidney disease	437 (8.1)	164 (9.1)	273 (7.6)	0.052
Chronic hepatic disease	109 (2.0)	41 (2.3)	68 (1.9)	0.338
Malignancy	291 (5.4)	103 (5.7)	188 (5.2)	0.443
Connective tissue disease	73 (1.3)	28 (1.6)	45 (1.2)	0.359
Auxiliary examinations
pH	7.41 (7.38, 7.45)	7.42 (7.38, 7.45)	7.41 (7.37, 7.44)	0.015
Oxygenation index	346.0 (251.0, 429.0)	318.0 (232.0, 416.0)	353.0 (263.0, 436.0)	<0.001
PaCO₂ (mmHg)	41.8 (35.8, 49.9)	42.1 (35.8, 51.6)	41.7 (35.9, 49.2)	0.049
Hemoglobin (g/L)	127.0 (113.0, 140.0)	125.0 (111.0, 139.0)	129.0 (114.0, 141.0)	<0.001
Platelet (×10⁹/L)	189.0 (142.0, 243.0)	183.0 (136.0, 238.0)	193.0 (145.0, 246.0)	<0.001
Leukocyte (×10⁹/L)	7.68 (5.90, 10.17)	8.03 (6.12,10.79)	7.47 (5.80, 9.86)	<0.001
Neutrophil (×10⁹/L)	5.55 (3.96, 8.11)	5.96 (4.20, 8.75)	5.31 (3.86, 7.79)	<0.001
Eosinophil (×10⁹/L)	0.08 (0.01,0.18)	0.07 (0.01, 0.17)	0.08 (0.01, 0.19)	0.001
BUN (mmol/L)	5.9 (4.4, 8.2)	6.3 (4.6, 8.8)	5.7 (4.3, 7.8)	<0.001
Creatinine (µmol/L)	79.0 (63.0, 104.0)	79.0 (64.0, 110.0)	79.0 (63.0, 102.0)	0.034
Albumin (g/L)	37.1 (33.7, 40.4)	37.1 (33.5, 40.3)	37.0 (33.8, 40.4)	0.206
NT-proBNP (pg/mL)	282.0 (83.0, 1244.0)	430.0 (136.0, 1933.0)	227.0 (68.0, 1037.0)	<0.001
D-dimer (mg/L)	0.77 (0.41, 1.64)	0.84 (0.45, 1.70)	0.72 (0.39, 1.61)	<0.001
Clinical outcomes
Length of hospital stay	10.0 (7.0, 14.0)	11.0 (7.0, 16.0)	9.0 (6.0, 14.0)	<0.001
ICU Admission	487 (9.0)	217 (12.0)	270 (7.5)	<0.001
Need for invasive mechanical ventilation	242 (4.5)	100 (5.5)	142 (3.9)	0.007
Length of invasive mechanical ventilation	5.0 (3.0, 11.0)	5.0 (3.0, 11.0)	6.0 (4.0, 10.0)	0.955
In-hospital mortality	121 (2.2)	53 (2.9)	68 (1.9)	0.013
60-day mortality	172 (3.2)	79 (4.4)	93 (2.6)	<0.001
Hospitalization costs (¥)	13668.0 (9651.3, 20829.8)	15547.2 (10220.7, 25785.0)	13082.3 (9392.5, 19423.1)	<0.001

Abbreviations: AECOPD, acute exacerbation of chronic obstructive pulmonary disease; COPD, chronic obstructive pulmonary disease; T2DM, type 2 diabetes mellitus; BMI, body mass index; LTOT, long-term home oxygen therapy; OCS, oral corticosteroids; SBP, systolic blood pressure; DBP, diastolic blood pressure; PaCO₂, arterial partial pressure of carbon dioxide; BUN, blood urea nitrogen; NT-proBNP, N-terminal pro-B type natriuretic peptide; ICU, intensive care unit; IQR, interquartile range.

As shown in Table 1, demographic characteristics and clinical manifestations were similar between patients with or without T2DM, but diabetic patients have higher proportion of fever. Patients with T2DM tended to present lower oxygenation index, hemoglobin, platelet and eosinophil but higher PaCO₂, leukocyte, neutrophil, blood urea nitrogen (BUN), creatinine, NT-proBNP and D-dimer, indicating more severe disease profiles. Moreover, substantial differences in clinical outcomes were observed between these two groups. Patients with T2DM were significantly associated with higher in-hospital and 60-day mortality and longer length of hospital stay. The proportion of ICU admission, need for IMV and overall hospitalization costs were also considerably higher in diabetic group.

Development of a specified prediction model for short-term mortality in AECOPD patients with T2DM

A total of 1804 AECOPD patients with T2DM were eventually included in the derivation cohort and 79 (4.4%) of them died within 60 days after admission. In the univariate analysis, 17 factors were found to be associated with 60-day mortality (Table S1) and eight of these variables were identified as independent predictors in multivariate analysis, including older age, disturbance of consciousness, chronic cardiac disease, low blood pressure, heart rate >100beats/min, neutrophil >6.3 × 10⁹/L, BUN >7.2 mmol/L, and random blood glucose >11.1 mmol/L (Figure 2). The VIF (1.00 to 1.11) was less than 5, indicating no existing multicollinearity. These eight factors were eventually included in the prognostic model and a predictive nomogram was established (Figure 3). As the nomogram shown, patients with older age, disturbance of consciousness, comorbid chronic cardiac disease, low blood pressure, high heart rate, elevated neutrophil, BUN and random blood glucose will get a higher total score, indicating a higher risk of short-term mortality.

Figure 2.

Independent risk factors for 60-day mortality among AECOPD patients with T2DM in the multivariate logistic regression analysis. Blood pressure <90/60mmHg: systolic blood pressure <90 mmHg or/and diastolic blood pressure <60 mmHg. AECOPD, acute exacerbation of chronic obstructive pulmonary disease; T2DM, type 2 diabetes mellitus; BUN, blood urea nitrogen; OR, odds ratio; CI, confidence interval; Ref, reference.

Figure 3.

The nomogram for 60-day mortality among AECOPD patients with T2DM. Blood pressure <90/60mmHg: systolic blood pressure <90 mmHg or/and diastolic blood pressure <60 mmHg. AECOPD, acute exacerbation of chronic obstructive pulmonary disease; T2DM, type 2 diabetes mellitus; BUN, blood urea nitrogen.

Evaluation and validation of the prognostic model

As shown in Table 2 and Figure 4(A), in the derivation cohort, the nomogram had good overall performance with a Brier score of 0.034 and an AUC of 0.878 (95%CI: 0.842-0.915). The calibration curve revealed good consistency between the predicted probability of 60-day death by nomogram and the observed probability (Figure S2(A)). Decision curve demonstrated superior overall net benefits obtained from the application of nomogram within a wide range of threshold probabilities (Figure S3(A)). The bootstrap internal validation also generated a satisfying AUC of 0.878 (95%CI: 0.842-0.915). An external validation was further performed in another cohort including 672 diabetic AECOPD patients (Figure 1(B)). Compared with the derivation cohort, patients in validation cohort tended to present less severe disease states with lower proportion of frequent exacerbations history, comorbid chronic cardiac disease, ICU admission, need for IMV and less hospitalization expenses (Table S2). Nevertheless, the nomogram maintained its excellent performance with a Brier score of 0.034 and an AUC of 0.859 (95%CI: 0.792-0.927), and outperformed the DECAF score (0.647 [95%CI: 0.535-0.760]) and BAP-65 score (0.758 [95%CI: 0.666-0.850]) (Table 2 and Figure 4(B)).

Table 2.

Model performance for predicting 60-day mortality among AECOPD patients with T2DM.

	Derivation cohort (n = 1804)	Validation cohort (n = 672)
Brier score	0.034	0.034
AUC (95% CI)	0.878 (0.842-0.915)	0.859 (0.792-0.927)
Calibration intercept	0.00	0.00
Calibration slope	1.00	1.00

Abbreviations: AECOPD, acute exacerbation of chronic obstructive pulmonary disease; T2DM, type 2 diabetes mellitus; AUC, area under the Receiver Operating Characteristic curve; CI, confidence interval.

Figure 4.

ROC curves for 60-day mortality among AECOPD patients with T2DM. (a) ROC curve of nomogram in the derivation cohort. (b) ROC curves of nomogram and DECAF in the validation cohort. ROC, receiver operating characteristic; AECOPD, acute exacerbation of chronic obstructive pulmonary disease; T2DM, type 2 diabetes mellitus.

Ultimately, we constructed a web calculator based on the nomogram, aiming to simplify and facilitate the use of nomogram in clinical practice. This calculator is freely available at https://xiaoqianli.shinyapps.io/AECOPDandDM/, allowing users to input predictor values to obtain the probability output of 60-day death directly (Figure S4).

Discussion

In this study, we found that comorbid T2DM is significantly associated with severe disease profile and worse prognosis in AECOPD patients. Considering the high incidence of T2DM and its negative effects on clinical outcomes among AECOPD population, as well as current little knowledge on the clinical characteristics of patients with AECOPD and T2DM, we further explored the risk factors and developed a specific prognostic tool for for these patients. Eight easily accessible predictors were identified and a predictive nomogram with a web calculator for 60-day mortality was established. This nomogram can effectively predict the clinical outcomes of diabetic AECOPD patients, and is superior to existing general AECOPD prognostic models, which can promote early and accurate prognosis evaluation and facilitate optimal decision disposition among these population.

In our AECOPD cohort, the prevalence of T2DM are approximately 13.5%, considerably higher than the incidence rate of 8% in the general population.²¹ Consistent with previous evidences,²² our study also found that comorbid diabetes are related to worse prognosis in AECOPD patients even after adjusting for confounders. Several factors may account for the complicated relation between diabetes and AECOPD as well as its adverse impacts on clinical outcomes. Firstly, the COPD-related systemic inflammation, oxidative stress, hypoxia could raise the level of glucose-elevating hormones and induce insulin resistance, increasing the risk of T2DM.²³ Besides, reduced physical activity, increased obesity and irrational corticosteroid exposure such as the use of high-dose ICS or long-term use of OCS for patients without indications may also contribute to increased the prevalence of T2DM.²⁴ In addition, T2DM could lead to microvascular and peripheroneural damage and chronic inflammation, resulting in reduced lung elasticity and compromised lung function, which in turn promote the development of COPD.²⁵ In addition, diabetes can cause neutrophil dysfunction, impaired immunity and increased susceptibility to bacterial infection, an important inducement of excerbations of COPD.²⁶

Regarding clinical characteristics, we found that diabetic AECOPD patients tend to have a higher proportion of fever and higher levels of leukocytes and neutrophils, which indicated these patients were more often accompanied by infections or infections may be a common induced factors of exacerbations among them. Besides, diabetic patients presented more severe disease profiles, with higher percentage of IMV demand and ICU admission. Therefore, greater emphasis should be placed on the screening of diabetes in COPD population and intensive blood glucose management among these patients may be of great importance. Timely respiratory support, more intensive and individualized surveillance should also be considered to optimize the management in this population.

Several prognostic models were developed among general AECOPD patients, but no studies published specifically explored predictors for poor prognosis among diabetic AECOPD patients. In this study, Eight easily available indicators were identified as independent risk factors for 60-day mortality among these patients. Consistent with previously studies on general AECOPD patients, advanced age was also recognized as a predictor for short-term mortality among diabetic AECOPD patients. Chronic cardiac disease were confirmed as a independent risk factor in this study. The coexisting chronic cardiac disease may contribute to more severe illness state and complexity of disease management. Besides, the inflammatory response and adverse effects on circulation system due to exacerbations of COPD can also worsen the underlying chronic cardiac disease and lead to increased all-cause mortality.²⁷ Other risk factors for 60-day mortality identified in this study included hypotension, disturbance of consciousness, elevated heart rate and BUN, which are similar to the components of CURB-65 score, a prognostic tool used to guide risk assessment and management in community acquired pneumonia (CAP).²⁸ Previous reports demonstrated that CURB-65 score may work as well in mortality prediction in AECOPD as it does in CAP.²⁹ The underlying reason may be the two conditions share important etiological and pathophysiological features.³⁰ For patients with coexisting T2DM, the correlation between AECOPD and CAP may be more significant given the important role that pneumonia play in the exacerbations of COPD among these patients. Besides, we also found that elevated neutrophils can be used as an effective marker for poor prognosis. Clinically, the elevated neutrophils are mostly associated with infections, and infection-induced AECOPD generally has a worse clinical prognosis.³¹ Therefore, prompt etiological examination and antibiotic treatment are crucial when an exacerbation triggered by infection is suspected among these patients. Previous studies have shown that compared with neutrophol alone, neutrophil to lymphocyte ratio (NLR) can more fully reflect the global involvement of immune system in inflammatory diseases, and NLR is an accurate prognostic predictor for CAP, a common complication of AECOPD.^32,33 Therefore, we further compared the predictive performance of the model which integrate NLR instead of neutrophil with our originally developed nomogram, and found that model with NRL as one of the components seems to provide better prediction efficiency than our nomogram, but the difference was not statistically significant (AUC: 0.882 [95%CI: 0.846-0.919] vs 0.878 [95% CI: 0.842-0.915], p = .459 ).

This is the first prognostic model specific for patients with AECOPD and diabetes incorporating predictors readily available for most patients at admission or can be obtained easily and quickly during hospitalization, allowing clinicians to make early and rapid risk assessment, which may facilitate proper decision disposition and improve the clinical outcomes among these patients..

There are some limitations in our study, and several questions remain unanswered. Firstly, although our study has evaluated the impacts of random blood glucose at admission on prognosis, random blood glucose level is influenced by multiple factors and may not fully reflect glycaemic control status. Future studies are warranted to determine the correlation between premorbid or in-hospital glycaemic glycemic control and poor prognosis among AECOPD population. Monitoring serum glycated hemoglobin/albumin levels may be of great clinical values. Secondly, diabetes was not routinely screened in the present study and our study population was restricted to inpatients who may tend to have more severe exacerbations, which might compromise the generalizability of our nomogram. Thirdly, due to the higher rate of loss to follow-up from the third month, to minimize bias, we only selected 2-month mortality as the prognosis endpoint in this study. But we have further evaluated the performance of our nomogram among patients who were not lost to follow-up at 3 months, 6 months and 12 months, and found that the nomogram exhibit similar predictive accuracy at these time points. However, studies with more complete long-term follow-up data are still needed to further verify the predictive efficacy of our nomogram for longer prognosis. Lastly, due to data missing of pulmonary function tests, we were unable to evaluate the impacts of pulmonary function on prognosis among these population. As in this study, lung function test results are also often unavailable for many AECOPD patients in real clinical practice, our research may help rapid risk assessment and early proper management using routinely available clinical data.

Conclusions

Diabetic AECOPD patients present more severe illness profile and worse prognosis than non-diabetic AECOPD patients. A prognostic nomogram for short-term mortality among patients with AECOPD and T2DM were developed based on eight easily available clinical indicators and externally validated. With good performance and clinical practicability, this model may facilitate early risk assessment and proper decision-making among these patients in clinical practice.

Supplemental Material

Supplemental Material - Clinical characteristics and prognosis prediction in patients with AECOPD and type 2 diabetes mellitus: A multicenter observational study

Supplemental Material for Clinical characteristics and prognosis prediction in patients with AECOPD and type 2 diabetes mellitus: A multicenter observational study by Xiaoqian Li, Qun Yi, Yuanming Luo, Hailong Wei, Huiqing Ge, Huiguo Liu, Jianchu Zhang, Xianhua Li, Xiufang Xie, Pinhua Pan, Mengqiu Yi, Lina Cheng, Hui Zhou, Liang Liu, Chen Zhou, Jiarui Zhang, Lige Peng, Jiaqi Pu and Haixia Zhou in Chronic Respiratory Disease

Footnotes

Author contributions

HZ gave the study concept and design; all authors acquired, analyzed, and interpreted the data, and revised the manuscript for important intellectual content; Xiaoqian Li drafted the manuscript; HZ and QY supervised the study; all authors read and approved the final manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Natural Science Foundation of Sichuan Province [Grant number 2022NSFSC1311], National Natural Science Foundation of China [Grant number 82170013], Sichuan Province Science and Technology Support Program [Grant number 2022YFS0262] and National Key Research and Development Program of China [Grant number 2016YFC1304202].

Ethical statement

Trial registration

Chinese clinical trial registry, ChiCTR2100044625. URL: .

ORCID iD

Haixia Zhou

Supplemental Material

Supplemental material for this article is available online.

References

Adeloye

Song

Zhu

, et al. Global, regional, and national prevalence of, and risk factors for, chronic obstructive pulmonary disease (COPD) in 2019: a systematic review and modelling analysis. Lancet Respir Med 2022; 10(5): 447–458.

GBD Chronic Respiratory Disease Collaborators . Prevalence and attributable health burden of chronic respiratory diseases, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet Respir Med 2020; 8(6): 585–596.

Mathers

Loncar

. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006; 3(11): e442.

Wang

Yang

, et al. Prevalence and risk factors of chronic obstructive pulmonary disease in China (the China Pulmonary Health [CPH] study): a national cross-sectional study. Lancet 2018; 391(10131): 1706–1717.

Putcha

Drummond

Wise

, et al. Comorbidities and chronic obstructive pulmonary disease: prevalence, influence on outcomes, and management. Semin Respir Crit Care Med 2015; 36(4): 575–591.

Antus

Harnasi

Drozdovszky

, et al. Monitoring oxidative stress during chronic obstructive pulmonary disease exacerbations using malondialdehyde. Respirology 2014; 19(1): 74–79.

Oltmanns

Gehring

Rudolf

, et al. Hypoxia causes glucose intolerance in humans. Am J Respir Crit Care Med 2004; 169(11): 1231–1237.

Adrogué

Chap

Okuda

, et al. Acidosis-induced glucose intolerance is not prevented by adrenergic blockade. Am J Physiol 1988; 255(6 Pt 1): E812–E823.

Lin

Shi

Kang

, et al. Analysis of prevalence and prognosis of type 2 diabetes mellitus in patients with acute exacerbation of COPD. BMC Pulm Med 2021; 21(1): 7.

10.

Parappil

Depczynski

Collett

, et al. Effect of comorbid diabetes on length of stay and risk of death in patients admitted with acute exacerbations of COPD. Respirology 2010; 15(6): 918–922.

11.

Steer

Gibson

Bourke

. The DECAF Score: predicting hospital mortality in exacerbations of chronic obstructive pulmonary disease. Thorax 2012; 67(11): 970–976.

12.

Tabak

Sun

Johannes

, et al. Mortality and need for mechanical ventilation in acute exacerbations of chronic obstructive pulmonary disease: development and validation of a simple risk score. Arch Intern Med 2009; 169(17): 1595–1602.

13.

Zhou

, et al. Validation of risk assessment models predicting venous thromboembolism in inpatients with acute exacerbation of chronic obstructive pulmonary disease: a multicenter cohort study in China. Thromb Haemost 2022; 122(7): 1177–1185.

14.

Monsour

Rodriguez

Abdelmasih

, et al. Characteristics and outcomes of diabetic patients with acute exacerbation of COPD. J Diabetes Metab Disord 2021; 20(1): 461–466.

15.

Riley

Ensor

Snell

KIE

, et al. Calculating the sample size required for developing a clinical prediction model. BMJ 2020; 368: m441.

16.

Rodriguez-Roisin

. Toward a consensus definition for COPD exacerbations. Chest 2000; 117(5 Suppl 2): 398S–401S.

17.

American Diabetes Association . 2. Classification and diagnosis of diabetes. Diabetes Care 2016; 39(Suppl 1): S13–S22.

18.

Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group . KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int 2024; 105(4S): S117–S314.

19.

Raghavan

Varkey

Bartter

. Chronic obstructive pulmonary disease: the impact of gender. Curr Opin Pulm Med 2017; 23(2): 117–123.

20.

Negewo

Gibson

McDonald

. COPD and its comorbidities: impact, measurement and mechanisms. Respirology 2015; 20(8): 1160–1171.

21.

Zheng

Ley

. Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nat Rev Endocrinol 2018; 14(2): 88–98.

22.

Belligund

Attaway

Lopez

, et al. Diabetes associated with higher health care utilization and poor outcomes after COPD-related hospitalizations. Am J Manag Care 2022; 28(9): e325–e332.

23.

Hughes

McGettrick

Sapey

. Shared mechanisms of multimorbidity in COPD, atherosclerosis and type-2 diabetes: the neutrophil as a potential inflammatory target. Eur Respir Rev 2020; 29(155): 190102.

24.

Cazzola

Rogliani

Calzetta

, et al. Targeting mechanisms linking COPD to type 2 diabetes mellitus. Trends Pharmacol Sci 2017; 38(10): 940–951.

25.

Kinney

Black-Shinn

Wan

, et al. Pulmonary function reduction in diabetes with and without chronic obstructive pulmonary disease. Diabetes Care 2014; 37(2): 389–395.

26.

Alba-Loureiro

Munhoz

Martins

, et al. Neutrophil function and metabolism in individuals with diabetes mellitus. Braz J Med Biol Res 2007; 40(8): 1037–1044.

27.

Kunisaki

Dransfield

Anderson

, et al. Exacerbations of chronic obstructive pulmonary disease and cardiac events. A post hoc cohort analysis from the SUMMIT randomized clinical trial. Am J Respir Crit Care Med 2018; 198(1): 51–57.

28.

Roberts

Lowe

Bucknall

, et al. Clinical audit indicators of outcome following admission to hospital with acute exacerbation of chronic obstructive pulmonary disease. Thorax 2002; 57(2): 137–141.

29.

Chang

Sullivan

Karalus

, et al. Audit of acute admissions of chronic obstructive pulmonary disease: inpatient management and outcome. Intern Med J 2007; 37(4): 236–241.

30.

Chang

Sullivan

Karalus

, et al. Predicting early mortality in acute exacerbation of chronic obstructive pulmonary disease using the CURB65 score. Respirology 2011; 16(1): 146–151.

31.

Steer

Norman

Afolabi

, et al. Dyspnoea severity and pneumonia as predictors of in-hospital mortality and early readmission in acute exacerbations of COPD. Thorax 2012; 67(2): 117–121.

32.

Buonacera

Stancanelli

Colaci

, et al. Neutrophil to lymphocyte ratio: an emerging marker of the relationships between the immune system and diseases. Int J Mol Sci 2022; 23(7): 3636.

33.

de Jager

Wever

Gemen

, et al. The neutrophil-lymphocyte count ratio in patients with community-acquired pneumonia. PLoS One 2012; 7(10): e46561.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.72 MB