(Chest. 2005;128:3568-3571.)
© 2005
American College of Chest Physicians
The Prevalence of Pulmonary Embolism and Pulmonary Hypertension in Patients With Type II Diabetes Mellitus*
Mohammad-Reza Movahed, MD, PhD;
Mehrtash Hashemzadeh, MS and
M. Mazen Jamal, MD, MPH
* From the Division of Cardiology (Dr. Movahed), University of California, Irvine, Medical Center, Orange; and Long Beach Veteran Administration Medical Center (Mr. Hashemzadeh and Dr. Jamal), Long Beach, CA.
Correspondence to: Mohammad-Reza Movahed, MD, PhD, Assistant Clinical Professor, University of California, Irvine, Medical Center, Department of Medicine, Division of Cardiology, 101 The City Drive, Bldg 53, Rm 100, Orange, CA 92868-4080; e-mail: rmova{at}aol.com
 |
Abstract
|
|---|
Background: Patients with diabetes mellitus (DM) have a hypercoagulable state that may increase their risk for thromboembolism. However, the data about this association are contradictory in the literature. The goal of this study was to evaluate the occurrence of pulmonary embolism (PE) and pulmonary hypertension (PHT) in patients with DM after adjusting for coronary artery disease (CAD), congestive heart failure (CHF), hypertension, and smoking using a large database.
Method: We used patient treatment file documents to inpatient hospital admissions containing discharge diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes) from Veterans Health Administration Hospitals. The patients were classified into two groups: a DM group with an ICD-9-CM code for DM (293,124), and a control group with an ICD-9-CM code for hypertension but no DM (552,623). The ICD-9-CM code for PE (415.19) and the ICD-9-CM code for PHT (416.0) were used to study prevalence of these diseases in DM patients vs control patients. We performed univariate and multivariate analyses adjusting for CAD, CHF, and smoking. Continuous variables were analyzed by unpaired t test. Binary variables were analyzed by 
2 and Fisher exact tests.
Results: PE was present in 2,011 patients with DM (0.7%) vs 2,759 patients (0.5%) in the control group. PHT was present in 3,356 patients with DM (1.1%) vs 3,357 patients (0.6%) in the control group. Using multivariate analysis, DM remained independently associated with PE (odds ratio [OR], 1.27; 95% confidence interval [CI], 1.19 to 1.35; p < 0.001) and with PHT (OR, 1.53; 95% CI, 1.45 to 1.60; p < 0.001).
Conclusion: Patients with DM have significantly higher prevalence of PE and PHT independent of CAD, hypertension, CHF, or smoking. The pathogenesis of this association is not known at this time.
Key Words: deep vein thrombosis • diabetes mellitus • pulmonary embolism • pulmonary hypertension • sudden death
|
Introduction
|
|---|
Patients with diabetes mellitus (DM) are at higher risk for cardiovascular disease, myocardial infarction, and sudden death.1234 Furthermore, patients with DM have a hypercoagulable state contributing to an increase of cardiovascular events. Hypercoagulability is multifactorial. Many coagulation factors, such as fibrinogen, d-dimer, and von Willebrand factor have been elevated, and fibrinolysis is decreased along with platelet hyperaggrebility and endothelial dysfunction.56789 Despite the proven association of DM with atherosclerosis involving the arterial system, the involvement of veins as the cause of thromboembolism has been controversial. In a large study10 involving women, DM was not associated with pulmonary thromboembolism. However, Tsai, et al11 found DM as an independent risk for venous thromboembolism in a large population including both genders. Furthermore, there is no study available reporting any association between DM and pulmonary hypertension (PHT). In order to address this controversy, we used a very large database of patients with type II diabetes for the occurrence of pulmonary embolism (PE) or PHT using univariate and multivariate analyses adjusting for coronary artery disease (CAD), congestive heart failure (CHF), and smoking.
|
Materials and Methods
|
|---|
Data Collection and Data Sources
We used patient treatment files (PTFs) from the Austin Automation Center available since July 1969. The PTFs document inpatient diagnoses from Veterans Health Administration hospitals, extended care discharges, and non-Veterans Affairs hospital discharges at the expense of the Veterans Health Administration hospital. The PTFs record demographics of patients and their discharge diagnosis. Since 1984, a primary diagnosis and up to nine secondary diagnoses have been available and were documented according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).12 This study was approved by the Institutional Review Board.
Identification of Study Groups
Patients with type II DM but free of hypertension were stratified in one group in comparison to patients with hypertension but no DM as the control group. In order to have a comparative population as a control, we used hypertension. This enabled us to have a control population that was larger than the diabetes population. Patients with an ICD-9-CM code of 250.01314 were identified from the PTFs for 1990 through 2000. The date of a patient’s first appearance in the PTFs with type II DM was considered to be the date of diagnosis. Nonmatched control patients with hypertension but without DM were selected from the same PTFs during the same period. A total of 845,748 patients were studied. There were 293,124 patients with DM and 552,624 patients with hypertension but no DM as the control group.
Calculation of Comorbidity Index
A comorbidity index was calculated for study groups, and an adaptation of the Charlson comorbidity index1314 was applied to administrative databases. Demographics were found from the computerized records for both study groups. Social security numbers were used to obtain the inpatient files (1990 to 2000) for the following ICD-9-CM codes: PHT (416.0), PE (415.19), CHF (428.0), CAD (414.9), and smoking (305.1, V15.82).12
Statistical Analysis
The statistical analysis was performed using statistical software (SAS/STAT; SAS Institute; Cary, NC; and SPSS version 13.0; SPSS, Chicago, IL). Any p value < 0.05 was accepted as statistically significant. Continuous variables were analyzed using unpaired t tests. Binary variables were studied using 2 test and Fisher Exact Test. Quantitative variables were mean ± SD. In the multivariate analysis, a logistic regression model was utilized to calculate the odds of PHT and PE in patients with DM vs control patients while adjusting for age, gender, and comorbid conditions. Odds ratios (ORs) and their 95% confidence intervals (CIs) were used to indicate the strength of influence.
|
Results
|
|---|
We studied 293,124 patients with DM and 552,624 control patients hospitalized from 1990 to 2000. Demographics were similar between the two groups. The mean ages for DM patients and control patients were 65.8 ± 11.3 years and 64.8 ± 12.6 years, respectively (p = not significant). Race (white race, 65.3% in the DM group and 68.3% in the control group) and gender (male gender, 97.8% in the DM group and 97.4% in the control group) were similar in both groups. DM was found to be significantly associated with PE and PHT using univariate and multivariate analyses. PE was present in 2,011 DM patients (0.7%) vs 2,759 patients (0.5%) in the control group. PHT was present in 3,356 DM patients (1.1%) vs 3,357 patients (0.6%) in the control group. Using multivariate analysis adjusting for CAD, CHF, and smoking, DM remained independently associated with PE (OR, 1.27; 95% CI, 1.19 to 1.35; p < 0.001) and PHT (OR, 1.53; 95% CI, 1.45 to 1.60; p < 0.001) [Table 1
, Fig 1
].
|
Discussion
|
|---|
In our study, we found DM to be strongly associated with PE and PHT independent of CAD, CHF, or smoking. Patients with DM have a hypercoagulable state with many coagulations and metabolic abnormalities, involving lipids, endothelial dysfunction, fibrinolytic systems, and platelets.56789 These abnormalities could explain the higher association of DM with PE and PHT. However, there is a controversy about this association in the literature. While some studies1115161718 are consistent with our findings, others10192021 did not find an association. The reason for this discrepancy in the literature is not known. Most of the studies with negative findings were very small in size, except for one study10 that studied only relatively young women. In contrast, Tsai et al11 found DM to be an independent risk for thromboembolism in a large population involving both genders, which is consistent with our findings. Furthermore, our study is the largest and the only report showing a significant independent association of DM with PHT in addition to PE. The cause of this association is not known at this time. Hypercoagulability in patients with DM may contribute to the risk of thromboembolism in this population. PHT could be secondary to underdiagnosed PE. Furthermore, the hypercoagulable state could have a direct thrombogenic effect on the lung vasculature. We suggest that the occurrence of venous thromboembolism and PHT should be considered in patients with DM presenting with unexplained dyspnea or hypoxia.
Limitations
This study is a retrospective study and is limited to inpatient hospital admissions. There is no information available about the standardization of the diagnoses that limit this study. Our control population consisted of hypertensive, nondiabetic patients who are not representative of all nondiabetic patients. However, patients with hypertension are at increase risk of cardiovascular disease; therefore, our association is most likely underestimated. There are no data available concerning the duration of DM limiting our study. Women were significantly underrepresented due to the small number of female patients at Veterans Affairs hospitals. Furthermore, we did not have any data about the weight or height of our patients. It is known that obese patients are at higher risk for DM and thromboembolism; therefore, we cannot exclude the effect of obesity in our study. However, we compared DM patients to a patient population with hypertension who may also have a higher prevalence of obesity; therefore, body mass index data may have been similar in both groups. We cannot exclude an impact of potential differences in the clinical evaluation of patients with hypertension vs diabetes, particularly with regard to the sensitivity of clinical tests chosen to detect CAD, CHF, PHT, and PE, or the consistency with which clinical tests (vs clinical judgment) were employed to support the discharge diagnoses in the two groups.
|
Conclusion
|
|---|
Based on our findings, PE and PHT occur more often in patients with DM independent of other cardiovascular abnormalities. The cause of this finding is not known at this time and needs to be further investigated. We suggest that the occurrence of venous thromboembolism and PHT should be considered in patients with DM presenting with unexplained dyspnea or hypoxia.
|
Acknowledgements
|
|---|
We thank Ms. Gale Good and Dr. Mehrnoosh Hashemzadeh for their support and editing of this manuscript.
|
Footnotes
|
|---|
Abbreviations: CAD = coronary artery disease; CHF = congestive heart failure; CI = confidence interval; DM = diabetes mellitus; ICD-9-CM = International Classification of Diseases, Ninth Revision, Clinical Modification; OR = odds ratio; PE = pulmonary embolism; PHT = pulmonary hypertension; PTF = patient treatment file
Received for publication March 15, 2005.
Accepted for publication June 13, 2005.
|
References
|
|---|
- Goraya, TY, Leibson, CL, Palumbo, PJ, et al (2002) Coronary atherosclerosis in diabetes mellitus: a population-based autopsy study. J Am Coll Cardiol 40,946-953[Abstract/Free Full Text]
- Blendea, MC, McFarlane, SI, Isenovic, ER, et al Heart disease in diabetic patients. Curr Diabet Rep 2003;3,223-229
- Kannel, WB, McGee, DL Diabetes and cardiovascular disease: the Framingham study. JAMA 1979;241,2035-2038[Abstract]
- Jouven, X, Desnos, M, Guerot, C, et al Predicting sudden death in the population: the Paris Prospective Study I. Circulation 1999;99,1978-1983[Abstract/Free Full Text]
- Calles-Escandon, J, Garcia-Rubi, E, Mirza, S, et al Type 2 diabetes: one disease, multiple cardiovascular risk factors. Coron Artery Dis 1999;10,23-30[ISI][Medline]
- Carmassi, F, Morale, M, Puccetti, R, et al Coagulation and fibrinolytic system impairment in insulin dependent diabetes mellitus. Thromb Res 1992;67,643-654[CrossRef][ISI][Medline]
- Hu, J, Wei, W, Din, G, et al Variations and clinical significance of coagulation and fibrinolysis parameters in patients with diabetes mellitus. J Tongji Med Univ 1998;18,233-235[Medline]
- Ouvina, SM, La Greca, RD, Zanaro, NL, et al Endothelial dysfunction, nitric oxide and platelet activation in hypertensive and diabetic type II patients. Thromb Res 2001;102,107-114[CrossRef][ISI][Medline]
- Carr, ME Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 2001;15,44-54[CrossRef][ISI][Medline]
- Goldhaber, SZ, Grodstein, F, Stampfer, MJ, et al A prospective study of risk factors for pulmonary embolism in women. JAMA 1997;277,642-645[Abstract]
- Tsai, AW, Cushman, M, Rosamond, WD, et al Cardiovascular risk factors and venous thromboembolism incidence: the longitudinal investigation of thromboembolism etiology. Arch Intern Med 2002;162,1182-1189[Abstract/Free Full Text]
- Hart, AC, Hopkins, CA 6th ed. Expert for hospitals: International classification of diseases, ninth revision; clinical modification 2001;Vol 1.,29-30 St. Anthony Publishing. West Valley City, UT::
- Deyo, RA, Cherkin, DC, Ciol, MA Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45,613-619[CrossRef][ISI][Medline]
- Charlson, ME, Pompei, P, Ales, KL, et al A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40,373-383[CrossRef][ISI][Medline]
- Bergqvist, D, Arnadottir, M, Bergentz, SE, et al Juvenile diabetes mellitus: a risk factor for postoperative venous thromboembolism? Acta Med Scand 1985;217,307-308[Medline]
- Auzepy, P, Benhaim, J, Leblanc, J, et al Subacute or chronic massive pulmonary thromboembolism without chronic cor pulmonale in 2 diabetic subjects. Coeur Med Interne 1974;13,367-373[Medline]
- el-Kashef, H Hyperglycemia increased the responsiveness of isolated rabbit’s pulmonary arterial rings to serotonin. Pharmacology 1996;53,151-159[ISI][Medline]
- Russell, MW, Taylor, DC, Cummins, G, et al Use of managed care claims data in the risk assessment of venous thromboembolism in outpatients. Am J Manag Care 2002;8(Suppl),S3-S9[Medline]
- Jones, EW, Mitchell, JR Venous thrombosis in diabetes mellitus. Diabetologia 1983;25,502-505[Medline]
- Coon, WW Risk factors in pulmonary embolism. Surg Gynecol Obstet 1976;143,385-390[ISI][Medline]
- Bonderman, D, Jakowitsch, J, Adlbrecht, C, et al Medical conditions increasing the risk of chronic thromboembolic pulmonary hypertension. Thromb Haemost 2005;93,512-516[Medline]
TOP
Abstract
Introduction
