During cardiac arrest and after cardiopulmonary resuscitation, activation of blood coagulation occurs, with a lack of adequate endogenous fibrinolysis. The aim of the present study was to determine whether the serum D-dimer concentration on admission is an independent predictor of all-cause mortality in patients with out-of-hospital cardiac arrest. We enrolled 182 consecutive patients (122 men, mean age 64.3 ± 15 years), who had presented to the emergency department from January 2007 to July 2012 because of out-of-hospital cardiac arrest. Information about the initial arrest rhythm, biochemical parameters, including the D-dimer concentration on admission, neurologic outcomes, and 30-day all-cause mortality were retrospectively collected. Of the 182 patients, 79 (43.4%) had died. The patients who died had had lower systolic (100 ± 39.6 vs 120.5 ± 26.9 mm Hg; p = 0.0004) and diastolic (58.3 ± 24.1 vs 74 ± 16.3 mm Hg; p <0.0001) blood pressure on admission. The deceased patients more often had had a history of myocardial infarction (32.9% vs 25.2%; p = 0.04) and less often had had an initial shockable rhythm (41.8% vs 60.2%; p = 0.02). The patients who died had had a significantly higher mean D-dimer concentration (9,113.6 ± 5,979.2 vs 6,121.6 ± 4,597.5 μg/L; p = 0.005) compared with patients who stayed alive. On multivariate logistic regression analysis, an on-admission D-dimer concentration >5,205 μg/L (odds ratio 5.7, 95% confidence interval 1.22 to 26.69) and hemoglobin concentration (odds ratio 1.66, 95% confidence interval 1.13 to 2.43) were strong and independent predictors of all-cause mortality. In conclusion, patients with a higher D-dimer concentration on admission had a poorer prognosis. The D-dimer concentration was an independent predictor of all-cause mortality.
The survival rates for patients after out-of-hospital cardiac arrest (OHCA) have been tremendously low. The percentage of survival to discharge after OHCA in Europe has been 9%, in North America 6%, in Australia 11%, and in Asia 2%. The differences have not only been related to variant healthcare quality. Population characteristics, cardiac or noncardiac OHCA etiology, and initial cardiac arrest rhythm have also been extremely important. In the treatment of patients with OHCA, the process of decision making and adjusting suitable therapy must be quick and efficient, because a delay in therapy worsens the prognosis. Currently, insufficient data are available to introduce a proper algorithm for risk of death stratification. Numerous attempts to develop simple, predictive of mortality biomarkers have failed. We wondered whether studies have now gone too far, and methods that are too sophisticated are being examined. Perhaps, we have overlooked the simplest, most obvious, and already easily accessible markers, such as fibrinolysis products (e.g., D-dimers). The aim of the present study was to determine whether the serum D-dimer concentration on admission is an independent predictor of all-cause mortality in patients with OHCA.
Methods
The study population consisted of consecutive patients who had experienced nontraumatic OHCA, in whom restoration of spontaneous circulation was achieved, and who were transported alive to the hospital from January 2008 to June 2012. We excluded patients with OHCA due to pulmonary embolism and those <18 years old. All patients had their medical records retrospectively reviewed for the circumstances of the OHCA (e.g., the cause, whether it was witnessed, who introduced cardiopulmonary resuscitation [CPR], the duration of the CPR delay, if any, and the initial heart rhythm) and the history of previous disease, including the cardiovascular risk factors (e.g., age, hypertension, diabetes mellitus, smoking, dyslipidemia, and major adverse cardiac events). The biochemical parameters were assessed in blood samples taken on admission to the hospital. The D-dimer concentration was measured in all study patients. The primary end point of the present study was all-cause mortality within 30 days after OHCA. The regional ethics committee (Medical University of Warsaw, Warsaw, Poland) approved the study protocol.
Continuous data are presented as the mean ± SD and were compared using the Mann-Whitney U test or Student’s t test. The categorical variables were compared using either the chi-square or Fisher exact test. The results were included in a receiver operating characteristic curve analysis. The multivariate logistic regression analysis, including hemoglobin, D-dimer, estimated glomerular filtration rate (eGFR), uric acid, potassium, lactate, and systolic and diastolic blood pressure, was performed to identify the predictors of all-cause mortality. Survival curves were constructed using the Kaplan-Meier method, and differences in survival were assessed using the log-rank test. A p value <0.05 was considered statistically significant, and the 95% confidence intervals (CIs) were recorded. All analyses were performed using Statistical Analysis Systems statistical software, version 8.02 (SAS Institute, Cary, North Carolina).
Results
The present analysis included data from 182 patients (122 men, mean age 64.3 ± 15.0 years). The mean body mass index in the study group was 28.4 ± 4.8 kg/m 2 , and the mean systolic and diastolic blood pressure was 111.3 ± 34.7 and 66.9 ± 21.5 mm Hg, respectively. The prevalence of cardiovascular diseases and risk factors was 46.2% for hypertension, 28.6% for diabetes, 28.9% for previous myocardial infarction, 7.1% for previous stroke, 18.7% for arrhythmias, 8.8% for chronic kidney disease, and 20.3% for dyslipidemia. The most common cause of cardiac arrest in this group of patients was acute coronary syndrome (140 patients, 76.9%). The OHCA was caused by heart failure in 23 patients (12.6%), electrolyte abnormalities in 11 (6.1%), and other causes in 8 (4.4%).
During the first 30 days of observation, 79 patients (43.4%) died ( Table 1 ). Of the 79 deaths, 31 (39.2%) were associated with cardiogenic shock, 30 (38%) with multiorgan failure, 15 (19%) with anoxic encephalopathy, and 3 (3.8%) with pneumonia.
| Parameter | Value |
|---|---|
| Male gender | 122 (67.0) |
| Age (yrs) | 64.3 ± 15 |
| Body mass index (kg/m 2 ) | 28.4 ± 4.8 |
| Systolic blood pressure (mm Hg) | 111.3 ± 34.7 |
| Diastolic blood pressure (mm Hg) | 66.9 ± 21.5 |
| Heart rate (beats/min) | 94.7 ± 28.9 |
| 30-Day mortality | 79 (43.4) |
| OHCA witnessed by medical personnel | 45 (24.7) |
| Shockable initial rhythm | 95 (52.2) |
| Previous myocardial infarction | 52 (28.6) |
| Previous stroke | 13 (7.1) |
| Heart failure | 42 (23.1) |
| Hypertension | 84 (46.2) |
| Diabetes mellitus | 52 (28.6) |
| Dyslipidemia | 37 (20.3) |
| Smoking | 44 (24.2) |
| Familiar history of cardiovascular disease | 13 (7.1) |
| Laboratory results | |
| Troponin I (ng/ml) | |
| First measurement | 5.3 ± 21.9 |
| Second measurement | 32.9 ± 53.5 |
| Third measurement | 27.9 ± 39.1 |
| Creatine kinase-MB mass (ng/ml) | 38.9 ± 122.5 |
| Hemoglobin (g/dl) | 13.3 ± 2.3 |
| Prothrombin time (s) | 13.5 ± 5.7 |
| International normalized ratio | 1.2 ± 0.5 |
| Activated partial thromboplastin time (s) | 43.8 ± 31.5 |
| Fibrinogen (mg/dl) | 419.2 ± 154.5 |
| D-dimers (μg/L) | 7,692.4 ± 5,542 |
| N-terminal pro-brain natriuretic peptide (pg/ml) | 5,586.6 ± 9,085.6 |
| eGFR (ml/min/1.73) | 45.5 ± 14.4 |
| Uric acid (mg/dl) | 7.3 ± 2.7 |
| Sodium (mmol/L) | 139.2 ± 5.6 |
| Potassium (mmol/L) | 4.4 ± 1 |
| Lactate (mmol/L) | 7.0 ± 4.8 |
| Echocardiographic parameters | |
| Left ventricle diastolic dimension (cm) | 5.2 ± 0.8 |
| Left ventricle systolic dimension (cm) | 4.0 ± 1.7 |
| Interventricular septum thickness (cm) | 1.2 ± 0.2 |
| Posterior wall diastolic thickness (cm) | 1.0 ± 0.2 |
| Aorta (cm) | 3.1 ± 0.4 |
| Right ventricle (cm) | 2.6 ± 0.5 |
| Left ventricle ejection fraction (%) | 41.1 ± 14.0 |
| Left atrium (cm) | 3.8 ± 0.6 |
We divided the patients into 2 groups according to the 30-day mortality. The patients who died had had a lower systolic (100.0 ± 39.6 vs 120.5 ± 26.9 mm Hg; p = 0.0004) and diastolic (58.3 ± 24.1 vs 74.0 ± 16.3 mm Hg; p <0.0001) blood pressure on admission than those who survived. The patients who died had more often had a history of myocardial infarction (32.9% vs 25.2%; p = 0.04). With no differences in the baseline characteristics between the 2 groups (including gender, age, body mass index, history of hypertension, diabetes, or previous cardiac arrest).
The patients who survived 30 days had more often had an initial shockable rhythm (ventricular fibrillation or ventricular tachycardia) compared with those who had died (60.2% vs 41.8%; p = 0.02). Of the laboratory parameters, the mean eGFR (39.3 ± 15.0 vs 50.2 ± 12.0 ml/min/1.73; p <0.0001) was lower and the uric acid concentration (8.5 ± 3.0 vs 6.4 ± 2.1 mg/dl; p = 0.0001) was higher in the deceased group than in the surviving group. Regarding the ion concentration, the potassium levels were higher in the patients who died than in those who survived (4.7 ± 1.1 vs 4.2 ± 0.9 mmol/L; p = 0.002) and the calcium levels were lower (1.6 ± 0.5 vs 2.0 ± 0.4 mmol/L; p = 0.004). In those who did not survive, the hemoglobin concentration (12.6 ± 2.7 vs 13.9 vs 1.7; p = 0.0009) was lower and the lactate concentration was higher (9.2 ± 5.0 vs 4.9 ± 4.0 mmol/L; p <0.0001) than in those who survived. The mean D-dimer concentration in the surviving group was significantly lower than in the group that died (6,121.6 ± 4,597.5 vs 9,113.6 ± 5,979.2 μg/L; p = 0.005; Table 2 ). The area under the receiver operating characteristic curve for D-dimer, eGFR, systolic blood pressure, diastolic blood pressure, and hemoglobin to detect mortality was 0.686 (95% CI 0.57 to 0.79), 0.679 (95% CI 0.56 to 0.78), 0.587 (95% CI 0.47 to 0.70), 0.607 (95% CI 0.49 to 0.72), and 0.645 (95% CI 0.5 to 0.75), respectively. No statistically significant difference was found between the areas under the receiver operating characteristic curves (D-dimer vs eGFR, p = 0.92; D-dimer vs hemoglobin, p = 0.60; D-dimer vs systolic blood pressure, p = 0.25; D-dimer vs diastolic blood pressure, p = 0.36; Figure 1 ). Figure 2 shows the Kaplan-Meier curves for 30-day mortality according to the D-dimer concentration (p = 0.0002). On multivariate logistic regression analysis, a D-dimer concentration on admission >5,205 μg/L (odds ratio 5.7, 95% CI 1.22 to 26.69) and the hemoglobin concentration on admission (odds ratio 1.66, 95% CI 1.13 to 2.43) were strong and independent predictors of all-cause mortality.
| Parameter | Death Within 30 Days (n = 79) | Alive Through 30 Days (n = 103) | p Value |
|---|---|---|---|
| Male gender | 47 (59.5) | 75 (72.8) | 0.08 |
| Age (yrs) | 66.1 ± 12.7 | 62.9 ± 16.5 | 0.08 |
| Body mass index (kg/m 2 ) | 29.6 ± 4.4 | 27.5 ± 5.0 | 0.09 |
| Systolic blood pressure (mm Hg) | 100.0 ± 39.6 | 120.5 ± 26.9 | 0.0004 ∗ |
| Diastolic blood pressure (mm Hg) | 58.3 ± 24.1 | 74.0 ± 16.3 | <0.0001 ∗ |
| Heart rate (beats/min) | 94.8 ± 30.9 | 94.6 ± 27.3 | 0.34 |
| OHCA witnessed by medical personnel | 18 (27.8) | 27 (26.1) | 0.72 |
| Shockable initial rhythm | 33 (41.8) | 62 (60.2) | 0.02 ∗ |
| CPR delay <5 min | 20 (25.3) | 38 (36.9) | 0.13 |
| Previous myocardial infarction | 26 (32.9) | 26 (25.2) | 0.04 ∗ |
| Previous stroke | 5 (6.3) | 8 (7.8) | 0.80 |
| Heart failure | 17 (21.5) | 25 (24.3) | 0.88 |
| Hypertension | 31 (39.2) | 53 (51.5) | 0.37 |
| Diabetes mellitus | 25 (31.6) | 27 (26.2) | 0.12 |
| Dyslipidemia | 8 (10.1) | 29 (28.2) | 0.02 ∗ |
| Smoking | 18 (22.8) | 26 (25.2) | 0.73 |
| Family history of cardiovascular disease | 3 (3.8) | 10 (9.7) | 0.78 |
| Laboratory results | |||
| Troponin I (ng/ml) | |||
| First measurement | 4.4 ± 15.5 | 6.0 ± 26.0 | 0.61 |
| Second measurement | 26.3 ± 48.5 | 37.2 ± 56.5 | 0.20 |
| Third measurement | 30.7 ± 43.9 | 26.2 ± 36.8 | 0.61 |
| Creatine kinase-MB mass (ng/ml) | 391.8 ± 459.3 | 27.6 ± 59.7 | 0.30 |
| Hemoglobin (g/dl) | 12.6 ± 2.7 | 13.9 ± 1.7 | 0.0009 ∗ |
| International normalized ratio | 1.3 ± 0.7 | 1.2 ± 0.4 | 0.09 |
| Fibrinogen (mg/dl) | 418.6 ± 181.5 | 419.7 ± 131.8 | 0.85 |
| D-dimers (μg/L) | 9,113.6 ± 5,979.2 | 6,121.6 ± 4,597.5 | 0.005 ∗ |
| N-terminal pro-brain natriuretic peptide (pg/ml) | 6,495.8 ± 10,901.1 | 4,950.2 ± 7,704.0 | 0.31 |
| eGFR (ml/min/1.73) | 39.3 ± 15.0 | 50.2 ± 12.0 | <0.0001 ∗ |
| Uric acid (mg/dl) | 8.5 ± 3.0 | 6.4 ± 2.1 | 0.0001 ∗ |
| Sodium (mmol/L) | 138.7 ± 7.9 | 139.6 ± 3.2 | 0.28 |
| Potassium (mmol/L) | 4.7 ± 1.1 | 4.2 ± 0.9 | 0.002 ∗ |
| Lactate (mmol/L) | 9.2 ± 5.0 | 4.9 ± 4.0 | 0.0001 ∗ |
| Echocardiographic parameters | |||
| Left ventricle diastolic dimension (cm) | 5.3 ± 0.8 | 5.1 ± 0.8 | 0.38 |
| Left ventricle systolic dimension (cm) | 5.2 ± 1.4 | 3.7 ± 1.7 | 0.30 |
| Interventricular septum thickness (cm) | 1.1 ± 0.2 | 1.2 ± 0.2 | 0.21 |
| Posterior wall diastolic thickness (cm) | 1.0 ± 0.1 | 1.0 ± 0.2 | 0.78 |
| Aorta (cm) | 2.9 ± 0.3 | 3.1 ± 0.4 | 0.16 |
| Right ventricle (cm) | 2.4 ± 0.3 | 3.1 ± 0.4 | 0.44 |
| Left ventricle ejection fraction (%) | 36.3 ± 13.6 | 42.6 ± 13.8 | 0.06 |
| Left atrium (cm) | 3.7 ± 0.6 | 3.8 ± 0.6 | 0.69 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
