CLINICAL INVESTIGATION
D-dimer
as a Prognostic Biomarker for Mortality in Chronic Obstructive Pulmonary Disease Exacerbation Oren Fruchter, MD, Mordechai Yigla, MD and Mordechai R. Kramer, MD
Abstract: Background: A major barrier to chronic obstructive pulmonary disease (COPD) research and management is lack of easily obtained biomarkers that are predictive of clinically important outcome measures. Objectives: We sought to investigate in patients admitted for acute exacerbation of COPD (AECOPD) the association of D-dimer (fibrin degradation product) obtained upon admission with in-hospital mortality and postdischarge prognosis. Methods: Clinical and laboratory data were evaluated in 61 patients admitted for AECOPD in whom D-dimer levels were obtained and in whom venous thromboembolism/ pulmonary embolism was excluded. Receiver operating characteristics curve was used to determine the optimal cutoff level for D-dimer that discriminated survivors versus nonsurvivors during index hospitalization, and during follow-up that extended to a median observation period of 62.6 months. Results: Mean (6SD) age of the study cohort was 71.2 6 10.5 years. Mean D-dimer level in nonsurvivors (n 5 12) was significantly higher than in survivors (n 5 49): 3.18 6 0.97 mg/L versus 1.45 6 1.18 mg/L, respectively, P 5 0.0006. D-dimer level .1.52 mg/L predicted in-hospital mortality with a sensitivity and specificity of 100% and 63.6%, respectively. After discharge, median survival of patients with D-dimer above and below 1.52 mg/L were 9.6 and 62.6 months, respectively (hazard ratio 5 2.636; 95% confidence interval 5 1.271– 6.426, P 5 0.0111). Conclusions: Elevated D-dimer is a reliable prognostic marker for both short-term and long-term survival in patients admitted for AECOPD. Prospective studies are required to further establish the appropriate role of D-dimer as a prognostic biomarker in patients with COPD. Key Indexing Terms: Chronic obstructive pulmonary disease; Prognosis; D-dimer; Mortality. [Am J Med Sci 2015;349(1):29–35.]
C
hronic obstructive pulmonary disease (COPD) represents a significant and growing health care concern as a leading cause of morbidity and mortality worldwide.1,2 The natural course of COPD is characterized by a progressive decline in pulmonary function tests and recurrent exacerbations requiring hospitalizations.3–5 Acute exacerbation of COPD (AECOPD) is associated with both substantial in-hospital mortality and may impact long-term prognosis. The frequency and severity of exacerbation are the most important factors determining overall prognosis in COPD4,5; hence, accurate individual risk assessment during exacerbation is of pivotal importance for clinical management and rational allocation of medical resources. Consequently, reliable predictors for in-hospital mortality, which are easily obtained upon admission, namely biomarkers, are urgently needed.6–10
From the Pulmonary Division (OF, MRK), Rabin Medical Center, Petah Tiqwa, Israel; The Sackler School of Medicine (OF, MRK), Tel Aviv University, Tel Aviv, Israel; and Pulmonary Division (MY), Rambam Medical Center, Haifa, Israel. Submitted February 9, 2014; accepted in revised form July 22, 2014. The authors have no financial or other conflicts of interest to disclose. Correspondence: Oren Fruchter, MD, The Pulmonary Division, Rabin Medical Center, Beilinson Hospital, Petach Tikva, 49100 Israel (E-mail: [email protected]).
The American Journal of the Medical Sciences
Some clinical evidence shows that the clinical course of patients with COPD may be complicated by thrombosis in the pulmonary vessels and a hypercoagulability state in particular during acute exacerbation.11 Pulmonary embolism (PE) may be a trigger of acute dyspnea in patients with COPD.12–14 D-dimer is a marker of in vivo thrombin and plasmin activation that has been used in the diagnostic workup of venous thromboembolism (VTE) in patients with COPD admitted for acute respiratory distress.15 The predictive value for hospital course and in-hospital mortality of elevated D-dimer levels in patients with AECOPD in whom VTE/PE is excluded has never been evaluated. In this study, we sought to determine whether D-dimer levels obtained upon admission in patients with AECOPD correlates with both in-hospital mortality and long-term prognosis. Our hypothesis was that identifying an enhanced prothrombotic state in patients with COPD during acute exacerbation by using D-dimer level could be used to distinguish a subgroup of patients at increased risk for both early and late mortality.
MATERIALS AND METHODS Study Population Approval for this study was obtained from the Rambam Medical Center Review Board. No personally identifiable information was used. All consecutive patients who were admitted to a General Medicine ward at the Rambam University Hospital, Haifa, Israel, over the period between February 3, 2000, and July 22, 2007, for the primary diagnosis of AECOPD were retrospectively studied. All hospital admissions of patients with a primary ICD-10 (International Statistical Classification of Diseases and Related Health Problems, 10th Revision) discharge coded diagnosis of COPD, COPD with acute exacerbation or COPD with acute lower respiratory tract infection. Patients were included if the following criteria were met: diagnosis of COPD, according to the criteria set by The Global Initiative for Chronic Obstructive Lung Disease (GOLD)16 (all patients had postbronchodilator forced expiratory volume in the first second (forced expiratory volume in 1 second [FEV1]/forced vital capacity [FVC] ,70% and symptoms indicating COPD, that is, history of chronic progressive symptoms such as dyspnea, cough and wheeze and a history of smoking). Patients treated with long-term oxygen treatment were excluded. AECOPD was defined by the presence of an increase in at least 2 of the 3 symptoms, dyspnea, cough and sputum purulence severe enough to warrant hospital admission without concomitant evidence of pneumonia. Subjects with acute coronary syndrome (electrocardiogram changes with elevated troponin I) were excluded. Patients with pulmonary edema, pneumothorax and pneumonia (based on chest x-ray) were excluded. Patients who were initially admitted to intensive care unit or subsequently transferred to such unit were excluded. Patients who required either invasive or noninvasive ventilatory support upon hospitalization, and were directly transferred to the intensive care unit, were also excluded.
Volume 349, Number 1, January 2015
29
Fruchter et al
D-dimer
Testing According to hospital guidelines, every patient was evaluated with a standardized protocol, including physical examination, chest x-ray, electrocardiogram and arterial blood gases. In addition, The Modified Wells criteria17 were assessed in patients with suspected PE. The decision to perform D-dimer determination was left to the discretion of the attending physician. Blood samples were collected by venipuncture into 3.2% citrate tubes. D-dimer was performed on fresh platelet-poor plasma, prepared by centrifugation at 2000g for 15 minutes. All the assays were measured on D-DI kit (Diagnostica STAGO). For the D-dimer assay, normal values were considered below the cutoff level of 0.5 mg/L. Patients in whom D-dimer levels were above 0.5 mg/L underwent diagnostic workup to diagnose or exclude VTE/PE. Diagnostic Algorithm for Diagnosis or Exclusion of VTE/PE The diagnosis of deep vein thrombosis (DVT) was established using Doppler’s ultrasound technique with an ALT 5000 Ultrasound device (Philips; Advanced Technology Laboratories, Bothell, WA). The veins were examined with and without compression maneuver, and color and spectral Doppler were also applied. DVT was diagnosed if the vein was noncompressible and blood flow compromised. A high probability lung scan for PE was defined when ventilationperfusion mismatch was observed in 2 or more of the lung segments. Computed tomography (CT) pulmonary angiography was performed with 16-detector row scanners (Brilliance; Philips Medical Systems, Cleveland, OH) using low-osmolar nonionic contrast material. Diagnosis of PE by spiral chest CT was based on direct visualization of a thrombus in 1 or more of the main or segmental or subsegmental pulmonary arteries. In patients with elevated D-dimer, VTE/PE was excluded by
normal Doppler ultrasound and either normal/low-probability lung scan or normal spiral chest CT. Patients in whom D-dimer levels were obtained and VTE/PE was excluded by the above algorithm were included in the study. Epidemiological and Baseline Laboratory Data Data were collected retrospectively. The age, gender and smoking status of the patients were documented. For all included patients, the following data were assessed: Medical history and comorbid conditions (diabetes mellitus, ischemic heart disease [IHD] and congestive heart failure [CHF]) were recorded on standardized forms from patient charts. IHD was defined as patients with history of acute myocardial infarction, coronary artery bypass surgery or percutaneous coronary intervention. An index case of CHF was based on the following: 1 inpatient hospitalization under diagnosis-related group 127 or 124 and 1 of the above ICD-9CM codes for CHF in the principal position. Active smoking status was defined as having smoked within the past 6 months. Results of laboratory analyses performed within 24 hours of hospital admission were retrieved from the hospital’s laboratory database. If multiple results were available, average levels were used. Outcome and Follow-up Hospital discharge date was used as the starting point regarding the follow-up period. End of follow-up was December 31, 2007. In-hospital mortality and length of hospital stay were determined for each patient. Patients were followed up for 6 years after admission by review of the clinical note sent through the death registration record in the event of out-ofhospital death. State regulations mandate computerized death registration ensuring no follow-up data were missing.
TABLE 1. Demographic and clinical characteristics of patients with COPD with/without D-dimer analysis stratified according to the presence or absence of VTE/PE Patients with AECOPD Patients with AECOPD (D-dimer measured), n 5 73 Patients characteristics (D-dimer not measured), (mean 6 SD) n 5 1017 VTE/PE confirmed, n 5 12 VTE/PE excluded, n 5 61 P Age (yrs) Gender (male/female) Female gender (%) Current smoker, n (%) Wells score ,2 Wells score .2 and ,6 Wells score .6 GOLD stage I/II Gold stage III/IV Ischemic heart disease, n (%) Diabetes mellitus, n (%) Congestive heart failure, n (%) Previous thromboembolism Hemoglobin level (g/dL) White blood cell count (3103/mL) PaCO2 (mm Hg) PaO2 (mm Hg) D-dimer level (mg/L); range
73.2 6 11.3 666/351 34.5 620 (61) 875 (86%) 111 (10.9%) 31 (3%) 732 (72%) 285 (28%) 315 (30.9) 264 (26) 467 (46) 23 (2.2%) 12.5 6 2.1 8.21 6 1.1 45.7 6 11.3 87.6 6 22.3 NA
71.2 6 10.5 4/8 33.3 7 (58.3) 1 (8.3%) 3 (25%) 8 (6.6%) 9 (75%) 3 (25%) 4 (33.3) 3 (25) 6 (50) 2 (16%) 12.2 6 3.2 9.71 6 3.1 47.7 6 13.4 88.9 6 32.1 2.53 6 0.62; 1.51–5.31
72.2 6 10.6 40/21 34.4 29 (47.5) 43 (70.5%) 14 (23%) 4 (6.5%) 42 (68.8%) 19 (31.2%) 18 (29.5) 17 (27.8) 15 (24.5) 1 (1.6%) 13.2 6 2.2 10.71 6 4.2 56.3 6 25.1 95.7 6 51.6 1.72 6 1.31; 0–4.00
NS NS NS NS
NS NS NS NS 0.022 NS NS NS NS ,0.0001
AECOPD, acute exacerbation of COPD; COPD, chronic obstructive pulmonary disease; GOLD, global initiative for chronic obstructive lung disease; VTE/PE, venous thromboembolism/pulmonary embolism.
30
Volume 349, Number 1, January 2015
D-dimer
TABLE 2. Comparison of clinical and laboratory data between patients stratified according to survival during index hospitalization Patients characteristics Survivors Nonsurvivors (mean 6 SD) (n 5 49) (n 5 12) P Age (yrs) Female gender, n (%) Current smoker, n (%) Ischemic heart disease, n (%) GOLD stage I/II Gold stage III/IV Diabetes mellitus, n (%) Congestive heart failure, n (%) Median length of hospitalization (d) Hemoglobin level (g/dL) White blood cell count (3103/mL) Antibiotic treatment during hospitalization PaCO2 (mm Hg) PaO2 (mm Hg) pH D-dimer level (mg/L) (range)
73.0 6 10.3 15 (30.6) 25 (52.5) 16 (32.6)
72.1 6 11.7 6 (50) 5 (41.6) 4 (33.3)
0.3408 0.3534 0.164 0.4622
40 (81.6%) 9 (18.3%) 15 (30.6) 15 (30.6)
2 (16.6%) 10 (83.3%) 3 (25) 4 (33.3)
,0.0001 ,0.0001 0.9769 0.0668
6
10
0.7619
13.3 6 2.3 10.7 6 4.4
12.9 6 1.8 10.6 6 3.4
0.6520 0.7652
40 (81%)
10 (83.3%)
0.99
49.8 6 15.8 94.5 6 55.8 7.39 6 0.07 1.45 6 1.18 (0–4.00)
75.7 6 38.8 99.5 6 41.4 7.31 6 0.18 3.18 6 0.97 (1.56–4.00)
0.0972 0.8614 0.4321 0.0006
GOLD, global initiative for chronic obstructive lung disease.
Statistical Analysis Descriptive data are presented as mean (6SD) or median (range). Comparisons between groups were made by using the Mann-Whitney’s U test (for continuous variables) or x2 test (for categorical variables), where appropriate. Receiver operating characteristics (ROC) analysis was used to determine the optimal cutoff level for D-dimer that predicted in-hospital mortality.
and Mortality in COPD
Survival curves after discharge were estimated by the KaplanMeier method and compared using the log rank test. A Cox proportional multivariate hazards model using all potential predictors of mortality was performed. Two-sided P value of ,0.05 was considered statistically significant. All of the statistical analyses were performed using a statistical software package (MedCalc Version 9.3.0.0; MedCalc Software, Ostend, Belgium).
RESULTS Between February 3, 2000, and July 22, 2007, 1090 patients were admitted to the General Medicine ward with a primary admission diagnosis of AECOPD. Of these patients, D-dimer levels were obtained in 73 patients upon admission. VTE/PE was confirmed in 12 patients (16.4%) and excluded in 61 patients who constituted the study cohort. Baseline characteristics of 1017 patients with COPD in whom D-dimer was not measured compared with patients in whom D-dimer levels were obtained stratified according to the presence of VTE/PE are presented in Table 1. As a group, patients in whom D-dimer levels were obtained did not differ in clinical or laboratory parameters compared with COPD patients in whom D-dimer levels were not measured. Furthermore, there were no significant differences between the 3 groups with respect to clinical and laboratory parameters apart from a higher percentage of subjects with CHF among patients in whom VTE/PE was diagnosed compared with patients in whom VTE/PE was excluded. In addition, D-dimer levels were significantly higher, as expected in patients with VTE/PE than in patients with AECOPD. The mean (6SD) age of the study population was 72.2 6 10.6 years. Most patients (65.6%) were male. Nearly half the patients (47.5%) were still actively smoking. The vast majority of patients (68.8%) were at GOLD stages I–II. The mean (6SD) FEV1, FVC and FEV1/FVC ratio were 65 6 21%, 72 6 11% and 0.64 6 0.11, respectively. Regarding comorbidities, 29.5% had IHD, 24.5% had CHF and 27.8% had diabetes mellitus. During index hospitalization, 12 patients (19.6%) died. The etiologies for death as stated on the hospital records were respiratory failure (n 5 1), CHF (n 5 4), sepsis (n 5 5) and multiorgan failure (n 5 2). The differences in main parameters
FIGURE 1. Comparison of mean (6SD) D-dimer level upon admission between patients with COPD who survived and deceased during index hospitalization. COPD, chronic obstructive pulmonary disease.
Copyright © 2014 by the Southern Society for Clinical Investigation. Unauthorized reproduction of this article is prohibited.
31
Fruchter et al
FIGURE 2. Dot plot presenting the distribution of individual D-dimer levels in patients who survived and deceased during index hospitalization.
between survivors and nonsurvivors are listed in Table 2. There were no significant differences between survivors and nonsurvivors with respect to age, smoking status and comorbidities apart from a trend for a higher proportion of patients with CHF among nonsurvivors. A significantly higher proportion of nonsurvivors were in GOLD stages III–IV compared with survivors (83.3% versus 18.3%, P , 0.0001). Patients who died during index hospitalization had a higher PaCO2 upon admission, although the difference was not statistically significant (75.7 6 38.8 mm Hg versus 49.8 6 15.8 mm Hg, P 5 0.0972). When D-dimer level upon admission was compared between patients who survived and patients who died during hospitalization (Figure 1), a significantly higher D-dimer level was detected in nonsurvivors versus survivors (3.18 6 0.97 mg/L versus 1.45 6 1.18 mg/L,
P 5 0.0006). The distribution of D-dimer level among patients who survived and deceased during index hospitalization is presented in Figure 2. ROC analysis (Figure 3) identified D-dimer .1.52 mg/L upon admission as the optimal cutoff level to discriminate survivors from deceased subjects during index hospitalization. (area under the ROC 5 0.862, standard error 5 0.080; 95% confidence interval [CI] 5 0.746–0.938, P 5 0.0001). The overall prevalence of patients with D-dimer .1.52 mg/L was 49.1%. The proportion of patients with D-dimer .1.52 mg/L among patients who survived and deceased during index hospitalization were 36.7% and 100%, respectively. D-dimer level above 1.52 mg/L had a sensitivity and specificity of 100% and 63.3%, respectively, for predicting in-hospital mortality (Table 3). After discharge, overall median survival for the entire cohort was 62.6 months. Survival rates at 1, 2 and 3 years were 68%, 57% and 52%, respectively. ROC analysis based on
TABLE 3. Test characteristics of D-dimer upon admission as a parameter for predicting in-hospital mortality and longterm mortality for cutoff level with the optimal sensitivity and specificity scorers In-hospital Long-term Criteria mortality mortality Cutoff point for D-dimer (mg/L) AUC 95% CI Subjects with D-dimer .1.52 mg/L (%) Sensitivity (%) Specificity (%) LR FIGURE 3. Receiver operating characteristics (ROC) curve for D-dimer as marker of in-hospital mortality in patients with COPD. COPD, chronic obstructive pulmonary disease.
32
1.52
1.52
0.862 0.746–0.938 12/12 (100)
0.671 0.535–0.789 32/49 (65.3)
100 63.3 2.72
65.3 68.7 2.09
AUC, area under the curve; CI, confidence interval; LR, likelihood ratio.
Volume 349, Number 1, January 2015
D-dimer
and Mortality in COPD
(DM), GOLD stage III–IV, PaCO2 and D-dimer level upon admission. The only independent predictors of mortality were advanced GOLD stage (P 5 0.012, HR 5 3.521; 95% CI 5 2.314–7.531) and D-dimer levels above 1.52 mg/L upon admission (P 5 0.0165, HR 5 2.728; 95% CI 5 1.205–6.170).
DISCUSSION
FIGURE 4. Receiver operating characteristics (ROC) curve for Ddimer as marker of postdischarge mortality in patients with COPD. COPD, chronic obstructive pulmonary disease.
postdischarge survival data (Figure 4) identified D-dimer .1.52 mg/L upon admission as the optimal cutoff level to predict postdischarge mortality with a sensitivity and specificity of 65.3% and 68.7%, respectively (Table 3). When patients were stratified according to D-dimer level upon admission, Kaplan-Meier analysis demonstrated a significant survival disadvantage for patients with D-dimer level above 1.52 mg/L (Figure 5). Median survival for patients with D-dimer levels below and above 1.52 mg/L were 61 and 9.6 months, respectively (P 5 0.0111, hazard ratio [HR] 5 2.636; 95% CI 5 1.271–6.426, by log rank test). A Cox proportional multivariate hazards model using all potential predictors of mortality was performed. Potential predictors for long-term mortality analyzed in the model were age, CHF, IHD, diabetes mellitus
Although GOLD stage has been shown in our report to be an independent predictor of mortality in accordance with previous studies,3–5 our main finding is that elevated D-dimer obtained upon admission in patients with AECOPD is an additional independent predictor for both increased risk for inhospital mortality and poor long-term prognosis. The diagnostic workup of patients with suspected VTE/ PE has been simplified by the use of D-dimer testing. D-dimer is a degradation product of cross-linked fibrin produced by plasmin.15 It has been showed that D-dimer testing is a reliable marker for VTE/PE in patients with AECOPD with a high negative predictive value.14 In our study, the prevalence of DVT/PE in patients with AECOPD was 16.4%, which is similar to that reported by Agkun et al18 who diagnosed VTE in 13.3% of 120 patients with COPD. D-dimer concentration is known to be increased not only by VTE/PE but also by solid tumors, leukemias, chronic liver diseases, severe infections, after trauma and recent operations, disseminated intravascular coagulation, pregnancy, preeclampsia, exercise, vasculitis, sickle cell anemia crisis, myocardial infarction and unstable angina pectoris.19 D-dimer is also elevated in smokers and patients with stable COPD probably reflecting increased coagulation activation.20 The significance of elevated D-dimer levels in patients presenting with AECOPD in whom thrombosis (of peripheral veins or pulmonary arteries) has been excluded, is unknown. Our observation that D-dimer levels above 1.52 mg/L (roughly 3 times the upper limit of normal in our laboratory) is associated with increased inhospital mortality and decreased long-term probability of survival is novel and warrants an explanation. Raimondi et al19 reported that plasma D-dimer levels increased in various infectious diseases. Shilon et al21 and Querol-Ribelles et al22 showed that plasma D-dimer levels increase with the severity of the
FIGURE 5. Kaplan-Meier survival probability of patients with COPD by D-dimer level. D-dimer level above 1.52 mg/L (- - -) and below 1.52 mg/L (──). P 5 0.0111 by log rank test. COPD, chronic obstructive pulmonary disease.
Copyright © 2014 by the Southern Society for Clinical Investigation. Unauthorized reproduction of this article is prohibited.
33
Fruchter et al
community-acquired pneumonia. The observation of Piras et al23 that demonstrated chronic systemic inflammatory syndrome in the vast majority of patients presenting to the emergency department with AECOPD is consistent with this hypothesis. Because bacterial and viral respiratory tract infection is thought to play a major role in the pathogenesis of AECOPD, elevated D-dimer level may represent an increased inflammatory burden during exacerbation and consequently be associated with increased in-hospital mortality. There is increasing awareness that IHD and cardiac comorbidity are major contributors to mortality in patients with COPD.24–26 Chronic inflammation in patients with COPD promotes endothelial dysfunction and interacts with the fibrinogenesis, thrombogenesis and fibrinolysis sequence in the atherosclerosis lesion.26,27 Furthermore, increased inflammatory response during AECOPD may increase the ongoing inflammatory process associated with atherosclerosis and result in escalation of atherosclerosis and coronary artery luminal narrowing possibly through processes that involve fibrinolysis releasing D-dimer into the circulation thereby increasing the likelihood of atheroma plaque rupture and subsequent ischemia and myocardial events.26,27 Hence, elevated D-dimer detected during AECOPD could be a useful marker for increased long-term probability of developing cardiovascular events and death. Elevated D-dimer in some patients with AECOPD may indeed signify that enhanced inflammatory-mediated thrombogenic processes prevail and subsequently lead to cardiovascular events and death. Our study has several limitations. The survival analysis in this retrospective study was performed using all-cause mortality as the outcome. The correct coding of cause of death in out-patients in general and in COPD in particularly is problematic since death certificates do not necessarily reflect the precise cause of death. Nevertheless, all-cause mortality is considered to be an acceptable end point in most long-term COPD survival studies.28 In-hospital mortality in our cohort (19.2%) is considerably higher than that cited by previous studies.3–5 We speculate that since the medical facility in which the current study was conducted serves as a tertiary referral center, the population of patients with COPD admitted to our emergency department represents a high-risk population of patients with COPD, in whom in-hospital mortality rate after exacerbation is particularly high. Because of the nature of this retrospective study, 1 obvious limitation is the selection of patients who underwent D-dimer level determination. Of 1090 patients with COPD in the initial cohort, D-dimer was analyzed in 73, 12 of whom were subsequently diagnosed with VTE/PE. The baseline clinical and laboratory characteristics were similar between patients in whom D-dimer levels were obtained and those it was not, reducing to a minimum the risk of selection bias. Consequently, we believe that the cohort in our report truly represents the entire population of patients admitted for AECOPD with respect to all known confounders such as GOLD stage, the severity of exacerbation and comorbidities. An additional potential limitation is the fact that VTE/PE was excluded by imaging modalities only in patients who presented with D-dimer level above 0.5 mg/L (the upper limit of the normal range in our laboratory). It is hypothetically possible that VTE/PE was present and not detected in some patients with COPD with low D-dimer levels. Since, according to previous data, D-dimer has a very high negative predictive value for the exclusion of VTE/PE in patients with COPD,14 we assume that
34
an insignificant proportion of patients with low D-dimer level had undiagnosed clinically important VTE/PE. Furthermore, the fact that patients with low D-dimer had significantly higher survival rates both in hospital and after discharge strengthens this assumption.
CONCLUSION In conclusion, D-dimer obtained upon admission could be used as a valuable biomarker for predicting both short- and long-term prognosis in patients admitted for AECOPD. Further prospective studies are needed to establish its exact role as a possible novel prognostic marker in patients with COPD. REFERENCES 1. Halpern MT, Stanford RH, Borker R. The burden of COPD in the USA: results from the confronting COPD survey. Respir Med 2003;97 (suppl C):S81–9. 2. Chiang CH. Cost analysis of chronic obstructive pulmonary disease in a tertiary care setting in Taiwan. Respirology 2008;13:689–94. 3. Roberts CM, Lowe D, Bucknall CE, et al. Clinical audit indicators of outcome following admission to hospital with acute exacerbation of chronic obstructive pulmonary disease. Thorax 2002;57:137–41. 4. Miravitlles M, Guerrero T, Mayordomo C, et al. Factors associated with increased risk of exacerbation and hospital admission in a cohort of ambulatory COPD patients: a multiple logistic regression analysis. Respiration 2000;67:495–501. 5. Groenewegen KH, Schols AM, Wouters EF. Mortality and mortalityrelated factors after hospitalization for acute exacerbation of COPD. Chest 2003;124:459–67. 6. Karadag F, Karul AB, Cildag O, et al. Biomarkers of systemic inflammation in stable and exacerbation phases of COPD. Lung 2008; 186:403–9. 7. Snell N, Newbold P. The clinical utility of biomarkers in asthma and COPD. Curr Opin Pharmacol 2008;8:222–35. 8. O’Reilly P, Bailey W. Clinical use of exhaled biomarkers in COPD. Int J Chron Obstruct Pulmon Dis 2007;2:403–8. 9. Fabbri LM, Luppi F, Beghé B, et al. Complex chronic comorbidities of COPD. Eur Respir J 2008;31:204–12. 10. Sin DD, Man SF. Biomarkers in COPD: are we there yet? Chest 2008; 133:1296–8. 11. Alessandri C, Basili S, Violi F, et al. Hypercoagulability state in patients with chronic obstructive pulmonary disease. Chronic obstructive bronchitis and haemostasis group. Thromb Haemost 1994;72: 343–6. 12. Tillie-Leblond I, Marquette CH, Perez T, et al. Pulmonary embolism in patients with unexplained exacerbation of chronic obstructive pulmonary disease: prevalence and risk factors. Ann Intern Med 2006;144: 390–6. 13. Rutschmann OT, Cornuz J, Poletti PA, et al. Should pulmonary embolism be suspected in exacerbation of chronic obstructive pulmonary disease? Thorax 2007;62:121–5. 14. Hartmann IJ, Hagen PJ, Melissant CF, et al. Diagnosing acute pulmonary embolism: effect of chronic obstructive pulmonary disease on the performance of D-dimer testing, ventilation/perfusion scintigraphy, spiral computed tomographic angiography, and conventional angiography. ANTELOPE Study Group. Advances in new technologies evaluating the localization of pulmonary embolism. Am J Respir Crit Care Med 2000;162:2232–7. 15. Adam SS, Key NS, Greenberg CS. D-dimer antigen: current concepts and future prospects. Blood 2009;113:2878–87.
Volume 349, Number 1, January 2015
D-dimer
16. Rabe KF, Hurd S, Anzueto A, et al;GOLD Scientific Committee. Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2001;163:1256–76. 17. Gibson NS, Sohne M, Kruip MJ, et al. Christopher study investigators. Further validation and simplification of the Wells clinical decision rule in pulmonary embolism. Thromb Haemost 2008;99:229–34. 18. Akgun M, Meral M, Onbas O, et al Comparison of clinical characteristics and outcomes of patients with COPD exacerbation with or without venous thromboembolism. Respiration 2006;73:428–33. 19. Raimondi P, Bongard O, de Moerloose P, et al. D-dimer plasma concentration in various clinical conditions: implication for the use of this test in the diagnostic approach of venous thromboembolism. Thromb Res 1993;69:125–30.
and Mortality in COPD
22. Querol-Ribelles JM, Tenias JM, Grau E, et al. Plasma D-dimer levels correlate with outcomes in patients with community-acquired pneumonia. Chest 2004;126:1087–92. 23. Piras O, Travaglino F, Autunno A, et al. Chronic systemic inflammatory syndrome in patients with AECOPD presenting to emergency department. Eur Rev Med Pharmacol Sci 2012;16(suppl 1):57–61. 24. Sin DD, Man SF. Impact of cancers and cardiovascular diseases in chronic obstructive pulmonary disease. Curr Opin Pulm Med 2008; 14:115–21. 25. Sin DD, Anthonisen NR, Soriano JB, et al. Mortality in COPD: role of comorbidities. Eur Respir J 2006;28:1245–57. 26. Fimognari FL, Scarlata S, Conte ME, et al Mechanisms of atherothrombosis in chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis 2008;3:89–96.
20. Rohde G, Borg I, Wiethege A, et al. Inflammatory response in acute viral exacerbations of COPD. Infection 2008;36:427–33.
27. Kunter E, Ilvan A, Ozmen N, et al. Effect of corticosteroids on hemostasis and pulmonary arterial pressure during chronic obstructive pulmonary disease exacerbation. Respiration 2008;75:145–54.
21. Shilon Y, Shitrit AB, Rudensky B, et al. A rapid quantitative D-dimer assay at admission correlates with the severity of community acquired pneumonia. Blood Coagul Fibrinolysis 2003;14:745–8.
28. Hansell AL, Walk JA, Soriano JB. What do chronic obstructive pulmonary disease patients die from? A multiple cause coding analysis. Eur Respir J 2003;22:809–14.
Copyright © 2014 by the Southern Society for Clinical Investigation. Unauthorized reproduction of this article is prohibited.
35
D-dimer
as a Prognostic Biomarker for Mortality in Chronic Obstructive Pulmonary Disease Exacerbation Oren Fruchter, MD, Mordechai Yigla, MD and Mordechai R. Kramer, MD
Abstract: Background: A major barrier to chronic obstructive pulmonary disease (COPD) research and management is lack of easily obtained biomarkers that are predictive of clinically important outcome measures. Objectives: We sought to investigate in patients admitted for acute exacerbation of COPD (AECOPD) the association of D-dimer (fibrin degradation product) obtained upon admission with in-hospital mortality and postdischarge prognosis. Methods: Clinical and laboratory data were evaluated in 61 patients admitted for AECOPD in whom D-dimer levels were obtained and in whom venous thromboembolism/ pulmonary embolism was excluded. Receiver operating characteristics curve was used to determine the optimal cutoff level for D-dimer that discriminated survivors versus nonsurvivors during index hospitalization, and during follow-up that extended to a median observation period of 62.6 months. Results: Mean (6SD) age of the study cohort was 71.2 6 10.5 years. Mean D-dimer level in nonsurvivors (n 5 12) was significantly higher than in survivors (n 5 49): 3.18 6 0.97 mg/L versus 1.45 6 1.18 mg/L, respectively, P 5 0.0006. D-dimer level .1.52 mg/L predicted in-hospital mortality with a sensitivity and specificity of 100% and 63.6%, respectively. After discharge, median survival of patients with D-dimer above and below 1.52 mg/L were 9.6 and 62.6 months, respectively (hazard ratio 5 2.636; 95% confidence interval 5 1.271– 6.426, P 5 0.0111). Conclusions: Elevated D-dimer is a reliable prognostic marker for both short-term and long-term survival in patients admitted for AECOPD. Prospective studies are required to further establish the appropriate role of D-dimer as a prognostic biomarker in patients with COPD. Key Indexing Terms: Chronic obstructive pulmonary disease; Prognosis; D-dimer; Mortality. [Am J Med Sci 2015;349(1):29–35.]
C
hronic obstructive pulmonary disease (COPD) represents a significant and growing health care concern as a leading cause of morbidity and mortality worldwide.1,2 The natural course of COPD is characterized by a progressive decline in pulmonary function tests and recurrent exacerbations requiring hospitalizations.3–5 Acute exacerbation of COPD (AECOPD) is associated with both substantial in-hospital mortality and may impact long-term prognosis. The frequency and severity of exacerbation are the most important factors determining overall prognosis in COPD4,5; hence, accurate individual risk assessment during exacerbation is of pivotal importance for clinical management and rational allocation of medical resources. Consequently, reliable predictors for in-hospital mortality, which are easily obtained upon admission, namely biomarkers, are urgently needed.6–10
From the Pulmonary Division (OF, MRK), Rabin Medical Center, Petah Tiqwa, Israel; The Sackler School of Medicine (OF, MRK), Tel Aviv University, Tel Aviv, Israel; and Pulmonary Division (MY), Rambam Medical Center, Haifa, Israel. Submitted February 9, 2014; accepted in revised form July 22, 2014. The authors have no financial or other conflicts of interest to disclose. Correspondence: Oren Fruchter, MD, The Pulmonary Division, Rabin Medical Center, Beilinson Hospital, Petach Tikva, 49100 Israel (E-mail: [email protected]).
The American Journal of the Medical Sciences
Some clinical evidence shows that the clinical course of patients with COPD may be complicated by thrombosis in the pulmonary vessels and a hypercoagulability state in particular during acute exacerbation.11 Pulmonary embolism (PE) may be a trigger of acute dyspnea in patients with COPD.12–14 D-dimer is a marker of in vivo thrombin and plasmin activation that has been used in the diagnostic workup of venous thromboembolism (VTE) in patients with COPD admitted for acute respiratory distress.15 The predictive value for hospital course and in-hospital mortality of elevated D-dimer levels in patients with AECOPD in whom VTE/PE is excluded has never been evaluated. In this study, we sought to determine whether D-dimer levels obtained upon admission in patients with AECOPD correlates with both in-hospital mortality and long-term prognosis. Our hypothesis was that identifying an enhanced prothrombotic state in patients with COPD during acute exacerbation by using D-dimer level could be used to distinguish a subgroup of patients at increased risk for both early and late mortality.
MATERIALS AND METHODS Study Population Approval for this study was obtained from the Rambam Medical Center Review Board. No personally identifiable information was used. All consecutive patients who were admitted to a General Medicine ward at the Rambam University Hospital, Haifa, Israel, over the period between February 3, 2000, and July 22, 2007, for the primary diagnosis of AECOPD were retrospectively studied. All hospital admissions of patients with a primary ICD-10 (International Statistical Classification of Diseases and Related Health Problems, 10th Revision) discharge coded diagnosis of COPD, COPD with acute exacerbation or COPD with acute lower respiratory tract infection. Patients were included if the following criteria were met: diagnosis of COPD, according to the criteria set by The Global Initiative for Chronic Obstructive Lung Disease (GOLD)16 (all patients had postbronchodilator forced expiratory volume in the first second (forced expiratory volume in 1 second [FEV1]/forced vital capacity [FVC] ,70% and symptoms indicating COPD, that is, history of chronic progressive symptoms such as dyspnea, cough and wheeze and a history of smoking). Patients treated with long-term oxygen treatment were excluded. AECOPD was defined by the presence of an increase in at least 2 of the 3 symptoms, dyspnea, cough and sputum purulence severe enough to warrant hospital admission without concomitant evidence of pneumonia. Subjects with acute coronary syndrome (electrocardiogram changes with elevated troponin I) were excluded. Patients with pulmonary edema, pneumothorax and pneumonia (based on chest x-ray) were excluded. Patients who were initially admitted to intensive care unit or subsequently transferred to such unit were excluded. Patients who required either invasive or noninvasive ventilatory support upon hospitalization, and were directly transferred to the intensive care unit, were also excluded.
Volume 349, Number 1, January 2015
29
Fruchter et al
D-dimer
Testing According to hospital guidelines, every patient was evaluated with a standardized protocol, including physical examination, chest x-ray, electrocardiogram and arterial blood gases. In addition, The Modified Wells criteria17 were assessed in patients with suspected PE. The decision to perform D-dimer determination was left to the discretion of the attending physician. Blood samples were collected by venipuncture into 3.2% citrate tubes. D-dimer was performed on fresh platelet-poor plasma, prepared by centrifugation at 2000g for 15 minutes. All the assays were measured on D-DI kit (Diagnostica STAGO). For the D-dimer assay, normal values were considered below the cutoff level of 0.5 mg/L. Patients in whom D-dimer levels were above 0.5 mg/L underwent diagnostic workup to diagnose or exclude VTE/PE. Diagnostic Algorithm for Diagnosis or Exclusion of VTE/PE The diagnosis of deep vein thrombosis (DVT) was established using Doppler’s ultrasound technique with an ALT 5000 Ultrasound device (Philips; Advanced Technology Laboratories, Bothell, WA). The veins were examined with and without compression maneuver, and color and spectral Doppler were also applied. DVT was diagnosed if the vein was noncompressible and blood flow compromised. A high probability lung scan for PE was defined when ventilationperfusion mismatch was observed in 2 or more of the lung segments. Computed tomography (CT) pulmonary angiography was performed with 16-detector row scanners (Brilliance; Philips Medical Systems, Cleveland, OH) using low-osmolar nonionic contrast material. Diagnosis of PE by spiral chest CT was based on direct visualization of a thrombus in 1 or more of the main or segmental or subsegmental pulmonary arteries. In patients with elevated D-dimer, VTE/PE was excluded by
normal Doppler ultrasound and either normal/low-probability lung scan or normal spiral chest CT. Patients in whom D-dimer levels were obtained and VTE/PE was excluded by the above algorithm were included in the study. Epidemiological and Baseline Laboratory Data Data were collected retrospectively. The age, gender and smoking status of the patients were documented. For all included patients, the following data were assessed: Medical history and comorbid conditions (diabetes mellitus, ischemic heart disease [IHD] and congestive heart failure [CHF]) were recorded on standardized forms from patient charts. IHD was defined as patients with history of acute myocardial infarction, coronary artery bypass surgery or percutaneous coronary intervention. An index case of CHF was based on the following: 1 inpatient hospitalization under diagnosis-related group 127 or 124 and 1 of the above ICD-9CM codes for CHF in the principal position. Active smoking status was defined as having smoked within the past 6 months. Results of laboratory analyses performed within 24 hours of hospital admission were retrieved from the hospital’s laboratory database. If multiple results were available, average levels were used. Outcome and Follow-up Hospital discharge date was used as the starting point regarding the follow-up period. End of follow-up was December 31, 2007. In-hospital mortality and length of hospital stay were determined for each patient. Patients were followed up for 6 years after admission by review of the clinical note sent through the death registration record in the event of out-ofhospital death. State regulations mandate computerized death registration ensuring no follow-up data were missing.
TABLE 1. Demographic and clinical characteristics of patients with COPD with/without D-dimer analysis stratified according to the presence or absence of VTE/PE Patients with AECOPD Patients with AECOPD (D-dimer measured), n 5 73 Patients characteristics (D-dimer not measured), (mean 6 SD) n 5 1017 VTE/PE confirmed, n 5 12 VTE/PE excluded, n 5 61 P Age (yrs) Gender (male/female) Female gender (%) Current smoker, n (%) Wells score ,2 Wells score .2 and ,6 Wells score .6 GOLD stage I/II Gold stage III/IV Ischemic heart disease, n (%) Diabetes mellitus, n (%) Congestive heart failure, n (%) Previous thromboembolism Hemoglobin level (g/dL) White blood cell count (3103/mL) PaCO2 (mm Hg) PaO2 (mm Hg) D-dimer level (mg/L); range
73.2 6 11.3 666/351 34.5 620 (61) 875 (86%) 111 (10.9%) 31 (3%) 732 (72%) 285 (28%) 315 (30.9) 264 (26) 467 (46) 23 (2.2%) 12.5 6 2.1 8.21 6 1.1 45.7 6 11.3 87.6 6 22.3 NA
71.2 6 10.5 4/8 33.3 7 (58.3) 1 (8.3%) 3 (25%) 8 (6.6%) 9 (75%) 3 (25%) 4 (33.3) 3 (25) 6 (50) 2 (16%) 12.2 6 3.2 9.71 6 3.1 47.7 6 13.4 88.9 6 32.1 2.53 6 0.62; 1.51–5.31
72.2 6 10.6 40/21 34.4 29 (47.5) 43 (70.5%) 14 (23%) 4 (6.5%) 42 (68.8%) 19 (31.2%) 18 (29.5) 17 (27.8) 15 (24.5) 1 (1.6%) 13.2 6 2.2 10.71 6 4.2 56.3 6 25.1 95.7 6 51.6 1.72 6 1.31; 0–4.00
NS NS NS NS
NS NS NS NS 0.022 NS NS NS NS ,0.0001
AECOPD, acute exacerbation of COPD; COPD, chronic obstructive pulmonary disease; GOLD, global initiative for chronic obstructive lung disease; VTE/PE, venous thromboembolism/pulmonary embolism.
30
Volume 349, Number 1, January 2015
D-dimer
TABLE 2. Comparison of clinical and laboratory data between patients stratified according to survival during index hospitalization Patients characteristics Survivors Nonsurvivors (mean 6 SD) (n 5 49) (n 5 12) P Age (yrs) Female gender, n (%) Current smoker, n (%) Ischemic heart disease, n (%) GOLD stage I/II Gold stage III/IV Diabetes mellitus, n (%) Congestive heart failure, n (%) Median length of hospitalization (d) Hemoglobin level (g/dL) White blood cell count (3103/mL) Antibiotic treatment during hospitalization PaCO2 (mm Hg) PaO2 (mm Hg) pH D-dimer level (mg/L) (range)
73.0 6 10.3 15 (30.6) 25 (52.5) 16 (32.6)
72.1 6 11.7 6 (50) 5 (41.6) 4 (33.3)
0.3408 0.3534 0.164 0.4622
40 (81.6%) 9 (18.3%) 15 (30.6) 15 (30.6)
2 (16.6%) 10 (83.3%) 3 (25) 4 (33.3)
,0.0001 ,0.0001 0.9769 0.0668
6
10
0.7619
13.3 6 2.3 10.7 6 4.4
12.9 6 1.8 10.6 6 3.4
0.6520 0.7652
40 (81%)
10 (83.3%)
0.99
49.8 6 15.8 94.5 6 55.8 7.39 6 0.07 1.45 6 1.18 (0–4.00)
75.7 6 38.8 99.5 6 41.4 7.31 6 0.18 3.18 6 0.97 (1.56–4.00)
0.0972 0.8614 0.4321 0.0006
GOLD, global initiative for chronic obstructive lung disease.
Statistical Analysis Descriptive data are presented as mean (6SD) or median (range). Comparisons between groups were made by using the Mann-Whitney’s U test (for continuous variables) or x2 test (for categorical variables), where appropriate. Receiver operating characteristics (ROC) analysis was used to determine the optimal cutoff level for D-dimer that predicted in-hospital mortality.
and Mortality in COPD
Survival curves after discharge were estimated by the KaplanMeier method and compared using the log rank test. A Cox proportional multivariate hazards model using all potential predictors of mortality was performed. Two-sided P value of ,0.05 was considered statistically significant. All of the statistical analyses were performed using a statistical software package (MedCalc Version 9.3.0.0; MedCalc Software, Ostend, Belgium).
RESULTS Between February 3, 2000, and July 22, 2007, 1090 patients were admitted to the General Medicine ward with a primary admission diagnosis of AECOPD. Of these patients, D-dimer levels were obtained in 73 patients upon admission. VTE/PE was confirmed in 12 patients (16.4%) and excluded in 61 patients who constituted the study cohort. Baseline characteristics of 1017 patients with COPD in whom D-dimer was not measured compared with patients in whom D-dimer levels were obtained stratified according to the presence of VTE/PE are presented in Table 1. As a group, patients in whom D-dimer levels were obtained did not differ in clinical or laboratory parameters compared with COPD patients in whom D-dimer levels were not measured. Furthermore, there were no significant differences between the 3 groups with respect to clinical and laboratory parameters apart from a higher percentage of subjects with CHF among patients in whom VTE/PE was diagnosed compared with patients in whom VTE/PE was excluded. In addition, D-dimer levels were significantly higher, as expected in patients with VTE/PE than in patients with AECOPD. The mean (6SD) age of the study population was 72.2 6 10.6 years. Most patients (65.6%) were male. Nearly half the patients (47.5%) were still actively smoking. The vast majority of patients (68.8%) were at GOLD stages I–II. The mean (6SD) FEV1, FVC and FEV1/FVC ratio were 65 6 21%, 72 6 11% and 0.64 6 0.11, respectively. Regarding comorbidities, 29.5% had IHD, 24.5% had CHF and 27.8% had diabetes mellitus. During index hospitalization, 12 patients (19.6%) died. The etiologies for death as stated on the hospital records were respiratory failure (n 5 1), CHF (n 5 4), sepsis (n 5 5) and multiorgan failure (n 5 2). The differences in main parameters
FIGURE 1. Comparison of mean (6SD) D-dimer level upon admission between patients with COPD who survived and deceased during index hospitalization. COPD, chronic obstructive pulmonary disease.
Copyright © 2014 by the Southern Society for Clinical Investigation. Unauthorized reproduction of this article is prohibited.
31
Fruchter et al
FIGURE 2. Dot plot presenting the distribution of individual D-dimer levels in patients who survived and deceased during index hospitalization.
between survivors and nonsurvivors are listed in Table 2. There were no significant differences between survivors and nonsurvivors with respect to age, smoking status and comorbidities apart from a trend for a higher proportion of patients with CHF among nonsurvivors. A significantly higher proportion of nonsurvivors were in GOLD stages III–IV compared with survivors (83.3% versus 18.3%, P , 0.0001). Patients who died during index hospitalization had a higher PaCO2 upon admission, although the difference was not statistically significant (75.7 6 38.8 mm Hg versus 49.8 6 15.8 mm Hg, P 5 0.0972). When D-dimer level upon admission was compared between patients who survived and patients who died during hospitalization (Figure 1), a significantly higher D-dimer level was detected in nonsurvivors versus survivors (3.18 6 0.97 mg/L versus 1.45 6 1.18 mg/L,
P 5 0.0006). The distribution of D-dimer level among patients who survived and deceased during index hospitalization is presented in Figure 2. ROC analysis (Figure 3) identified D-dimer .1.52 mg/L upon admission as the optimal cutoff level to discriminate survivors from deceased subjects during index hospitalization. (area under the ROC 5 0.862, standard error 5 0.080; 95% confidence interval [CI] 5 0.746–0.938, P 5 0.0001). The overall prevalence of patients with D-dimer .1.52 mg/L was 49.1%. The proportion of patients with D-dimer .1.52 mg/L among patients who survived and deceased during index hospitalization were 36.7% and 100%, respectively. D-dimer level above 1.52 mg/L had a sensitivity and specificity of 100% and 63.3%, respectively, for predicting in-hospital mortality (Table 3). After discharge, overall median survival for the entire cohort was 62.6 months. Survival rates at 1, 2 and 3 years were 68%, 57% and 52%, respectively. ROC analysis based on
TABLE 3. Test characteristics of D-dimer upon admission as a parameter for predicting in-hospital mortality and longterm mortality for cutoff level with the optimal sensitivity and specificity scorers In-hospital Long-term Criteria mortality mortality Cutoff point for D-dimer (mg/L) AUC 95% CI Subjects with D-dimer .1.52 mg/L (%) Sensitivity (%) Specificity (%) LR FIGURE 3. Receiver operating characteristics (ROC) curve for D-dimer as marker of in-hospital mortality in patients with COPD. COPD, chronic obstructive pulmonary disease.
32
1.52
1.52
0.862 0.746–0.938 12/12 (100)
0.671 0.535–0.789 32/49 (65.3)
100 63.3 2.72
65.3 68.7 2.09
AUC, area under the curve; CI, confidence interval; LR, likelihood ratio.
Volume 349, Number 1, January 2015
D-dimer
and Mortality in COPD
(DM), GOLD stage III–IV, PaCO2 and D-dimer level upon admission. The only independent predictors of mortality were advanced GOLD stage (P 5 0.012, HR 5 3.521; 95% CI 5 2.314–7.531) and D-dimer levels above 1.52 mg/L upon admission (P 5 0.0165, HR 5 2.728; 95% CI 5 1.205–6.170).
DISCUSSION
FIGURE 4. Receiver operating characteristics (ROC) curve for Ddimer as marker of postdischarge mortality in patients with COPD. COPD, chronic obstructive pulmonary disease.
postdischarge survival data (Figure 4) identified D-dimer .1.52 mg/L upon admission as the optimal cutoff level to predict postdischarge mortality with a sensitivity and specificity of 65.3% and 68.7%, respectively (Table 3). When patients were stratified according to D-dimer level upon admission, Kaplan-Meier analysis demonstrated a significant survival disadvantage for patients with D-dimer level above 1.52 mg/L (Figure 5). Median survival for patients with D-dimer levels below and above 1.52 mg/L were 61 and 9.6 months, respectively (P 5 0.0111, hazard ratio [HR] 5 2.636; 95% CI 5 1.271–6.426, by log rank test). A Cox proportional multivariate hazards model using all potential predictors of mortality was performed. Potential predictors for long-term mortality analyzed in the model were age, CHF, IHD, diabetes mellitus
Although GOLD stage has been shown in our report to be an independent predictor of mortality in accordance with previous studies,3–5 our main finding is that elevated D-dimer obtained upon admission in patients with AECOPD is an additional independent predictor for both increased risk for inhospital mortality and poor long-term prognosis. The diagnostic workup of patients with suspected VTE/ PE has been simplified by the use of D-dimer testing. D-dimer is a degradation product of cross-linked fibrin produced by plasmin.15 It has been showed that D-dimer testing is a reliable marker for VTE/PE in patients with AECOPD with a high negative predictive value.14 In our study, the prevalence of DVT/PE in patients with AECOPD was 16.4%, which is similar to that reported by Agkun et al18 who diagnosed VTE in 13.3% of 120 patients with COPD. D-dimer concentration is known to be increased not only by VTE/PE but also by solid tumors, leukemias, chronic liver diseases, severe infections, after trauma and recent operations, disseminated intravascular coagulation, pregnancy, preeclampsia, exercise, vasculitis, sickle cell anemia crisis, myocardial infarction and unstable angina pectoris.19 D-dimer is also elevated in smokers and patients with stable COPD probably reflecting increased coagulation activation.20 The significance of elevated D-dimer levels in patients presenting with AECOPD in whom thrombosis (of peripheral veins or pulmonary arteries) has been excluded, is unknown. Our observation that D-dimer levels above 1.52 mg/L (roughly 3 times the upper limit of normal in our laboratory) is associated with increased inhospital mortality and decreased long-term probability of survival is novel and warrants an explanation. Raimondi et al19 reported that plasma D-dimer levels increased in various infectious diseases. Shilon et al21 and Querol-Ribelles et al22 showed that plasma D-dimer levels increase with the severity of the
FIGURE 5. Kaplan-Meier survival probability of patients with COPD by D-dimer level. D-dimer level above 1.52 mg/L (- - -) and below 1.52 mg/L (──). P 5 0.0111 by log rank test. COPD, chronic obstructive pulmonary disease.
Copyright © 2014 by the Southern Society for Clinical Investigation. Unauthorized reproduction of this article is prohibited.
33
Fruchter et al
community-acquired pneumonia. The observation of Piras et al23 that demonstrated chronic systemic inflammatory syndrome in the vast majority of patients presenting to the emergency department with AECOPD is consistent with this hypothesis. Because bacterial and viral respiratory tract infection is thought to play a major role in the pathogenesis of AECOPD, elevated D-dimer level may represent an increased inflammatory burden during exacerbation and consequently be associated with increased in-hospital mortality. There is increasing awareness that IHD and cardiac comorbidity are major contributors to mortality in patients with COPD.24–26 Chronic inflammation in patients with COPD promotes endothelial dysfunction and interacts with the fibrinogenesis, thrombogenesis and fibrinolysis sequence in the atherosclerosis lesion.26,27 Furthermore, increased inflammatory response during AECOPD may increase the ongoing inflammatory process associated with atherosclerosis and result in escalation of atherosclerosis and coronary artery luminal narrowing possibly through processes that involve fibrinolysis releasing D-dimer into the circulation thereby increasing the likelihood of atheroma plaque rupture and subsequent ischemia and myocardial events.26,27 Hence, elevated D-dimer detected during AECOPD could be a useful marker for increased long-term probability of developing cardiovascular events and death. Elevated D-dimer in some patients with AECOPD may indeed signify that enhanced inflammatory-mediated thrombogenic processes prevail and subsequently lead to cardiovascular events and death. Our study has several limitations. The survival analysis in this retrospective study was performed using all-cause mortality as the outcome. The correct coding of cause of death in out-patients in general and in COPD in particularly is problematic since death certificates do not necessarily reflect the precise cause of death. Nevertheless, all-cause mortality is considered to be an acceptable end point in most long-term COPD survival studies.28 In-hospital mortality in our cohort (19.2%) is considerably higher than that cited by previous studies.3–5 We speculate that since the medical facility in which the current study was conducted serves as a tertiary referral center, the population of patients with COPD admitted to our emergency department represents a high-risk population of patients with COPD, in whom in-hospital mortality rate after exacerbation is particularly high. Because of the nature of this retrospective study, 1 obvious limitation is the selection of patients who underwent D-dimer level determination. Of 1090 patients with COPD in the initial cohort, D-dimer was analyzed in 73, 12 of whom were subsequently diagnosed with VTE/PE. The baseline clinical and laboratory characteristics were similar between patients in whom D-dimer levels were obtained and those it was not, reducing to a minimum the risk of selection bias. Consequently, we believe that the cohort in our report truly represents the entire population of patients admitted for AECOPD with respect to all known confounders such as GOLD stage, the severity of exacerbation and comorbidities. An additional potential limitation is the fact that VTE/PE was excluded by imaging modalities only in patients who presented with D-dimer level above 0.5 mg/L (the upper limit of the normal range in our laboratory). It is hypothetically possible that VTE/PE was present and not detected in some patients with COPD with low D-dimer levels. Since, according to previous data, D-dimer has a very high negative predictive value for the exclusion of VTE/PE in patients with COPD,14 we assume that
34
an insignificant proportion of patients with low D-dimer level had undiagnosed clinically important VTE/PE. Furthermore, the fact that patients with low D-dimer had significantly higher survival rates both in hospital and after discharge strengthens this assumption.
CONCLUSION In conclusion, D-dimer obtained upon admission could be used as a valuable biomarker for predicting both short- and long-term prognosis in patients admitted for AECOPD. Further prospective studies are needed to establish its exact role as a possible novel prognostic marker in patients with COPD. REFERENCES 1. Halpern MT, Stanford RH, Borker R. The burden of COPD in the USA: results from the confronting COPD survey. Respir Med 2003;97 (suppl C):S81–9. 2. Chiang CH. Cost analysis of chronic obstructive pulmonary disease in a tertiary care setting in Taiwan. Respirology 2008;13:689–94. 3. Roberts CM, Lowe D, Bucknall CE, et al. Clinical audit indicators of outcome following admission to hospital with acute exacerbation of chronic obstructive pulmonary disease. Thorax 2002;57:137–41. 4. Miravitlles M, Guerrero T, Mayordomo C, et al. Factors associated with increased risk of exacerbation and hospital admission in a cohort of ambulatory COPD patients: a multiple logistic regression analysis. Respiration 2000;67:495–501. 5. Groenewegen KH, Schols AM, Wouters EF. Mortality and mortalityrelated factors after hospitalization for acute exacerbation of COPD. Chest 2003;124:459–67. 6. Karadag F, Karul AB, Cildag O, et al. Biomarkers of systemic inflammation in stable and exacerbation phases of COPD. Lung 2008; 186:403–9. 7. Snell N, Newbold P. The clinical utility of biomarkers in asthma and COPD. Curr Opin Pharmacol 2008;8:222–35. 8. O’Reilly P, Bailey W. Clinical use of exhaled biomarkers in COPD. Int J Chron Obstruct Pulmon Dis 2007;2:403–8. 9. Fabbri LM, Luppi F, Beghé B, et al. Complex chronic comorbidities of COPD. Eur Respir J 2008;31:204–12. 10. Sin DD, Man SF. Biomarkers in COPD: are we there yet? Chest 2008; 133:1296–8. 11. Alessandri C, Basili S, Violi F, et al. Hypercoagulability state in patients with chronic obstructive pulmonary disease. Chronic obstructive bronchitis and haemostasis group. Thromb Haemost 1994;72: 343–6. 12. Tillie-Leblond I, Marquette CH, Perez T, et al. Pulmonary embolism in patients with unexplained exacerbation of chronic obstructive pulmonary disease: prevalence and risk factors. Ann Intern Med 2006;144: 390–6. 13. Rutschmann OT, Cornuz J, Poletti PA, et al. Should pulmonary embolism be suspected in exacerbation of chronic obstructive pulmonary disease? Thorax 2007;62:121–5. 14. Hartmann IJ, Hagen PJ, Melissant CF, et al. Diagnosing acute pulmonary embolism: effect of chronic obstructive pulmonary disease on the performance of D-dimer testing, ventilation/perfusion scintigraphy, spiral computed tomographic angiography, and conventional angiography. ANTELOPE Study Group. Advances in new technologies evaluating the localization of pulmonary embolism. Am J Respir Crit Care Med 2000;162:2232–7. 15. Adam SS, Key NS, Greenberg CS. D-dimer antigen: current concepts and future prospects. Blood 2009;113:2878–87.
Volume 349, Number 1, January 2015
D-dimer
16. Rabe KF, Hurd S, Anzueto A, et al;GOLD Scientific Committee. Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med 2001;163:1256–76. 17. Gibson NS, Sohne M, Kruip MJ, et al. Christopher study investigators. Further validation and simplification of the Wells clinical decision rule in pulmonary embolism. Thromb Haemost 2008;99:229–34. 18. Akgun M, Meral M, Onbas O, et al Comparison of clinical characteristics and outcomes of patients with COPD exacerbation with or without venous thromboembolism. Respiration 2006;73:428–33. 19. Raimondi P, Bongard O, de Moerloose P, et al. D-dimer plasma concentration in various clinical conditions: implication for the use of this test in the diagnostic approach of venous thromboembolism. Thromb Res 1993;69:125–30.
and Mortality in COPD
22. Querol-Ribelles JM, Tenias JM, Grau E, et al. Plasma D-dimer levels correlate with outcomes in patients with community-acquired pneumonia. Chest 2004;126:1087–92. 23. Piras O, Travaglino F, Autunno A, et al. Chronic systemic inflammatory syndrome in patients with AECOPD presenting to emergency department. Eur Rev Med Pharmacol Sci 2012;16(suppl 1):57–61. 24. Sin DD, Man SF. Impact of cancers and cardiovascular diseases in chronic obstructive pulmonary disease. Curr Opin Pulm Med 2008; 14:115–21. 25. Sin DD, Anthonisen NR, Soriano JB, et al. Mortality in COPD: role of comorbidities. Eur Respir J 2006;28:1245–57. 26. Fimognari FL, Scarlata S, Conte ME, et al Mechanisms of atherothrombosis in chronic obstructive pulmonary disease. Int J Chron Obstruct Pulmon Dis 2008;3:89–96.
20. Rohde G, Borg I, Wiethege A, et al. Inflammatory response in acute viral exacerbations of COPD. Infection 2008;36:427–33.
27. Kunter E, Ilvan A, Ozmen N, et al. Effect of corticosteroids on hemostasis and pulmonary arterial pressure during chronic obstructive pulmonary disease exacerbation. Respiration 2008;75:145–54.
21. Shilon Y, Shitrit AB, Rudensky B, et al. A rapid quantitative D-dimer assay at admission correlates with the severity of community acquired pneumonia. Blood Coagul Fibrinolysis 2003;14:745–8.
28. Hansell AL, Walk JA, Soriano JB. What do chronic obstructive pulmonary disease patients die from? A multiple cause coding analysis. Eur Respir J 2003;22:809–14.
Copyright © 2014 by the Southern Society for Clinical Investigation. Unauthorized reproduction of this article is prohibited.
35
