SEARCH

SEARCH BY CITATION

Keywords:

  • anticoagulants;
  • cancer;
  • pulmonary embolism;
  • survival

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

Summary.  Background:  Little is known about the risk factors and outcome of unsuspected pulmonary embolism (UPE) in cancer patients.

Objectives:  To assess the risk factors and outcome of UPE in cancer patients.

Methods:  The charts of 66 patients diagnosed with UPE were reviewed. Two control groups were selected: 132 cancer patients without pulmonary embolism (PE) and 65 cancer patients with clinically suspected PE. Variables associated with UPE were identified by multivariable analysis. Six-month survival and recurrent venous thromboembolism were compared by use of Cox proportional analysis.

Results:  Twenty-seven (40.9%) patients with UPE had symptoms suggesting PE. Adenocarcinoma (odds ratio [OR] 4.45; 95% confidence interval [CI] 1.98–9.97), advanced age (OR 1.18; 95% CI 1.02–1.38), recent chemotherapy (OR 4.62; 95% CI 2.26–9.44), performance status > 2 (OR 7.31; 95% CI 1.90–28.15) and previous venous thromboembolism (OR 4.47; 95% CI 1.16–17.13) were associated with UPE. When adjusted for tumor stage and performance status, 6-month mortality did not differ between patients with UPE and patients without PE (hazard ratio 1.40; 95% CI 0.53–3.66; P = 0.50). Patients with UPE were more likely to have central venous catheters and chemotherapy and less likely to have proximal clots than patients with clinically suspected PE. Recurrent venous thromboembolism occurred in 6.1% and 7.7% of patients with UPE and symptomatic PE, respectively.

Conclusion:  UPE is not associated with an increased risk of death. Patients with clinically suspected PE and those with UPE have similar risks of recurrent venous thromboembolism.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

Clinically unsuspected pulmonary embolism (UPE) is being increasingly diagnosed in cancer patients who have undergone thoracic multidetector computed tomography for other diagnostic purposes [1]. Only a few studies have evaluated the risk factors and the natural history of UPE in patients with cancer. Whether the risk factors for UPE differ from those of clinically suspected pulmonary embolism (PE) in patients with cancer is not known. The survival of cancer patients with UPE has rarely been evaluated, and few data are available on the risk of recurrent venous thromboembolism (VTE) in patients with UPE [2–4].

We reviewed the charts of cancer patients diagnosed with UPE, of cancer patients without PE and of cancer patients with clinically suspected PE to identify the risk factors for UPE, to evaluate the survival of cancer patients with UPE, and to compare the risks of recurrent VTE and the survival rates of patients with UPE and patients with clinically suspected PE.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

We designed a case–control study to evaluate the risk factors and the clinical consequences of UPE in cancer patients. We also assessed the presentation, risk of recurrence and risk of death of UPE as compared with clinically suspected PE. Three groups of patients were defined: a first group of patients with cancer and UPE (cases), a second group with cancer but without PE, and a third group with cancer and clinically suspected PE. The study was approved by the Institutional Review Board of the French Learned Society for Respiratory Medicine (Société de Pneumologie de Langue Française).

Patients

Cancer patients with UPE (cases)  We retrospectively identified all patients followed at our institution for a solid cancer who were diagnosed with a PE on multidetector computed tomography (MDCT) of the chest performed during routine cancer staging. The computerized reports of all thoracic MDCT scans performed from 1 August 2005 to 31 December 2010 including the words ‘cancer’ or ‘tumor’ or ‘neoplasm’ and ‘pulmonary embolism’ were retrieved. This period was chosen because all MDCT examinations were performed on a 64-detector MDCT unit. In addition, patients were identified from the Clinical Data Warehouse of the hospital, with the following diagnostic codes from the International Classification of Diseases version 10 (ICD-10): ‘pulmonary embolism’ (I26*) AND (‘malignant neoplasm’ [C00*–C97*] OR ‘Neoplasms of uncertain or unknown behavior’ [D37*–D48*]). Patients’ charts were then retrieved to ascertain that the diagnosis of PE was not clinically suspected before MDCT.

Cancer patients without PE  For each case patient with UPE, two patients followed at our institution for a solid cancer who underwent a thoracic MDCT study within the same week as the case patient were selected to constitute the control group of cancer patients without PE. These patients were identified from the MDCT reports corresponding to this time period. As one of our aims was to identify the risk factors for UPE, the controls were not matched with the cases except for the time period. When more than two controls were available, the two patients with the closest date or time of examination were selected.

Cancer patients with symptomatic PE  Patients followed at our institution for a solid cancer and admitted for clinically suspected PE during the same period were selected from the Clinical Data Warehouse of the hospital with the same ICD-10 codes as used for patients with UPE. Patients’ charts were retrieved to ascertain that the diagnosis of PE was clinically suspected before MDCT was performed. When possible, one control with clinically suspected PE was selected for each UPE case within the 3 months preceding or following the date of diagnosis of the case. Because one of our aims was to determine whether the UPE characteristics were different from those of clinically suspected PE, cases and controls were not matched except for the time period.

Exclusion criteria

Patients with the following conditions were excluded from the study: age < 18 years, hematologic malignancy, cancer in remission (without disease or treatment for ≥ 1 year), and anticoagulant treatment at full therapeutic dosage at the time of MDCT.

Data extraction and follow-up

Medical records were reviewed and the data were extracted by two authors (M.S.H. and B.P.). Age, sex, risk factors for VTE, previous VTE, primary tumor site, stage, histologic type (adenocarcinoma or other), date of diagnosis, evolution of the cancer (remission, progression, or stability), performance status, specific anticancer treatment administered within 30 days of MDCT and clinical symptoms suggesting PE before MDCT were recorded. Performance status was graded according to the World Health Organization criteria: 0, asymptomatic (fully active, able to carry on all predisease activities without restriction); 1, symptomatic but completely ambulatory (restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature); 2, symptomatic, in bed for < 50% of the time in bed during the day (ambulatory and capable of all self-care, but unable to carry out any work activities); 3, symptomatic, in bed for > 50% of the time, but not bedbound (capable of only limited self-care; confined to a bed or chair for ≥50% of waking hours); and 4, bedbound (completely disabled; cannot carry out any self-care) [5].

The medical records of all patients were reviewed to ascertain the patients’ outcome during the 6 months following the initial MDCT examination. We recorded:

  • 1
    Initial anticoagulant treatment (unfractionated heparin or low molecular weight heparin [LMWH]), long-term anticoagulant treatment (LMWH or vitamin K antagonists), the use of vena cava filters, and specific antitumor treatment.
  • 2
    Evolution of the underlying cancer.
  • 3
    Death, objectively confirmed recurrent VTE defined according to guidelines, and major bleeds defined according to ISTH guidelines [6,7].

MDCT

MDCT scans were all performed with a 64-detector MDCT unit (Lightspeed VCT; GE Healthcare, Milwaukee, WI, USA). The MDCT acquisition and injection protocols were different for PE diagnosis and cancer follow-up, mainly with regard to the slice reconstruction thicknesses and the contrast medium injection protocol. The slice reconstruction thickness was 0.625 mm for PE diagnosis and 1.25 mm for cancer follow-up studies.

For PE diagnosis, a volume of 100–120 mL of iodinated contrast medium was injected at a rate of 4 mL s−1, and a bolus tracking technique was used to start the acquisition. For cancer follow-up, 90 mL of iodinated contrast was injected at a rate of 3 mL s−1, and the acquisitions were started with a fixed delay of 30 s after the beginning of injection.

The same contrast medium concentration was used for both protocols, with an iodine concentration of 300 mg mL−1.

The diagnosis of PE or UPE was systematically confirmed by an experienced thoracic radiologist. The level of the most proximal clot (subsegmental, segmental, lobar or main pulmonary artery) was recorded.

Statistical analysis

Continuous variables are presented as mean ± standard deviation or as medians (25th to 75th percentile range) when not normally distributed. Categorical variables are presented as numbers and percentages.

Comparisons between groups were performed first with univariate analysis. For continuous variables, Student’s t-test was used for normally distributed variables and the Wilcoxon test for non-normally distributed variables. The chi-square test (or chi-square exact test, when appropriate) was used for categorical variables.

A multivariate analysis was performed, keeping variables with a P-value of < 0.2 in univariate analysis, by use of a multivariate logistic model with stepwise variable selection, to identify factors independently associated with UPE (UPE vs. no PE) or with clinically suspected PE (UPE vs. clinically suspected PE).

Overall survival and cancer progression-free survival in each group were computed with a Kaplan–Meier model, and a log-rank test was used to compare the three groups. Cox proportional hazard models were used to estimate unadjusted and adjusted hazard ratios (HRs) for overall and progression-free survival and their 95% confidence intervals (CIs). Candidate risk factors for mortality, including age, prognostic cancer group [8], tumor stage, performance status, and delay from the onset of cancer, were entered into a univariate Cox model. Risk factors with a P-value of < 0.2 in the univariate Cox model were entered into the final multivariate model.

Cox proportional hazard models were used to estimate unadjusted and adjusted HRs for recurrent VTE-free survival and their 95% CIs in patients with UPE and in patients with clinically suspected PE. Candidate risk factors for recurrent VTE, including previous VTE, sex, nature of the underlying cancer, and tumor stage, were entered into a univariate Cox model. Risk factors with a P-value of < 0.2 in the univariate Cox model were entered into the final multivariate model.

To show the adjusted survival curves graphically and to generate survival probability estimates for the stratified Cox proportional hazard models, we used the methods described by Zhang et al. [9].

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

Patients

A total of 66 cancer patients with UPE were identified during the study period: 46 were identified from the MDCT reports, and an additional 20 patients were identified from the Clinical Data Warehouse of the hospital. We included 132 patients in the control group of cancer patients without PE and 66 patients in the control group of patients with cancer and clinically suspected PE (Table 1). One patient with clinically suspected PE did not consent to participate in the study, so only 65 patients with clinically suspected PE were analyzed. In this group, only 53 patients were identified during the 3 months preceding or following the date of identification of a case. Therefore, 12 additional patients with clinically suspected PE during the study period but outside the predefined 6-month time range were included.

Table 1.   Characteristics of the three groups at the time of inclusion
 Cases (1)Patients without PE (2)Patients with clinically suspected PE (3) P (1) vs. (2) P (1) vs. (3)
  1. BMI, body mass index; CVC, central venous catheter; PE, pulmonary embolism; SD, standard deviation; VTE, venous thromboembolism. *Data on BMI are lacking for 12 patients (cases), for 30 patients (control 1), and for 17 patients (control 2).

Age (years), mean (SD)67 (11)63 (12)64 (11)0.0280.179
Male sex, n (%)33 (50)70 (53)34 (52)0.6870.792
BMI (kg m−2*), mean (SD)23 (4)23 (4)26 (5)0.9180.001
Cancer, n (%)
 Lung20 (30)45 (34)16 (25)0.0510.612
 Colorectal9 (14)19 (14)7 (11)
 Breast7 (11)12 (9)12 (18)
 Prostate4 (6)7 (5)5 (8)
 Uterus5 (8)3 (2)4 (6)
 Ovary6 (9)6 (5)6 (9)
 Kidney0 (0)6 (5)1 (1)
 Stomach5 (8)3 (2)3 (5)
 Esophagus1 (1)7 (5)0 (0)
 Pancreas2 (3)2 (1)2 (3)
 Larynx0 (0)14 (11)5 (8)
 Other4 (6)6 (5)2 (3)
 Unknown3 (4)2 (1)2 (3)
Cancer prognosis, n (%)
 Good11 (17)21 (16)17 (26)0.9890.223
 Intermediate26 (39)52 (39)28 (43)
 Poor29 (44)59 (45)20 (31)
Adenocarcinoma, n (%)51 (77)67 (51)47 (72)< 0.0010.327
Cancer stage, n (%)
 Localized3 (4)19 (14)3 (5)0.0130.210
 Locally advanced14 (21)25 (19)14 (21)
 Metastatic46 (70)68 (51)38 (58)
 Remission3 (4)20 (15)10 (15)
Performance status, n (%)
 1–258 (88)127 (96)58 (89)0.0260.808
 3–48 (12)5 (4)7 (11)
Anticancer therapy, n (%)
 Surgery5 (8)9 (7)15 (23)0.8450.004
 Radiotherapy2 (3)6 (4)2 (3)0.7211.000
 Chemotherapy43 (65)42 (32)31 (48)< 0.0010.044
 Antiangiogenic drugs10 (15)9 (7)3 (5)0.0740.076
 Antihormonal therapy3 (4)4 (3)3 (5)0.6881.000
 None13 (20)66 (50)17 (26)< 0.0010.379
 Other0 (0)4 (3)0 (0)0.303 
Surgery within 1 month, n (%)5 (8)11 (8)17 (26)0.8540.014
Hospitalization within 1 month, n (%)10 (15)19 (14)28 (43)0.887< 0.001
Chronic cardiac or respiratory disease, n (%)3 (4)4 (3)0 (0)0.6880.244
Previous VTE, n (%)8 (12)4 (3)10 (15)0.0220.588
CVC, n (%)48 (73)72 (54)35 (54)0.0140.025

Patients with UPE and patients without PE at the time of MDCT examination  Among the 66 patients with UPE, 27 (41%) had one or more symptoms suggesting PE before MDCT, 23% had dyspnea, 9% had chest pain, 1% had hemoptysis, and 8% had leg pain. Patients with UPE were significantly older and were more likely to have advanced disease and adenocarcinoma than patients without PE. They were more likely to have had chemotherapy during the month before MDCT, to have a lower performance status, to have had previous VTE and to have a long-term central venous line than control patients without PE (Table 1). There were no significant differences between cases and control patients without PE in their type of tumor, mean BMI, recent surgery, chronic cardiac or respiratory disease, or recent hospitalization (Table 1). In multivariable analysis, age, adenocarcinoma, recent chemotherapy, performance status > 2 and previous VTE were independently associated with UPE (Table 2).

Table 2.   Multivariable analysis of clinical variables associated with clinically unsuspected pulmonary embolism in patients with cancer
VariableOR (95% CI)
  1. CI, confidence interval; MDCT, multidetector computed tomography; OR, odds ratio. *OR for each increment of 5 years.

Age at the time of MDCT*1.18 (1.02–1.38)
Adenocarcinoma vs. other histology4.45 (1.98–9.97)
Current chemotherapy4.62 (2.26–9.44)
Previous venous thromboembolism4.47 (1.16–17.13)
Performance status (3 or 4 vs. 1 or 2)7.31 (1.90–28.15)

Patients with UPE and patients with clinically suspected PE at the time of MDCT  Patients with UPE had a lower BMI, were more likely to have had chemotherapy and were less likely to have had surgery for cancer within the last month of MDCT than patients with clinically suspected PE. Patients with UPE were more likely to have long-term central venous catheters and were less likely to be hospitalized before MDCT (Table 1). Age, type of tumor, tumor histology and performance status did not differ significantly between patients with UPE and patients with clinically suspected PE (Table 1). A dedicated study to detect deep vein thrombosis (DVT) was performed in 30 patients with UPE, and DVT was diagnosed in 14 of them (47%), as compared with 40 among 58 patients with clinically suspected PE (69%). Patients with clinically suspected PE had more proximal clots than patients with UPE (= 0.001) (Fig. 1).

image

Figure 1.  Level of the most proximal clot in patients with clinically unsuspected pulmonary embolism (UPE) and patients with clinically suspected pulmonary embolism (SPE).

Download figure to PowerPoint

Six-month follow-up

Outcome of patients with UPE and patients without PE  Eleven patients (17%) with UPE and nine patients (7%) without PE died during the 6-month follow-up period (= 0.02 by the log-rank test). After adjustment for performance status and tumor stage with Cox regression analysis, no significant difference in survival was observed between the two groups (HR 1.40; 95% CI 0.53–3.66; = 0.50) (Fig. 2A). Progression-free survival did not differ between patients with and without UPE after adjustment for tumor stage and performance status (HR 1.06; 95% CI 0.65–1.70; = 0.81) in the Cox regression analysis (Fig. 2B).

image

Figure 2.  Survival probability curves for 66 cancer patients with unsuspected pulmonary embolism (UPE) and 132 cancer patients without UPE. P-values are based on the Cox proportional hazard model comparing UPE patients with controls. The model was stratified for stage of cancer and performance status. (A) Global survival. (B) Progression-free survival. CI, confidence interval; HR, hazard ratio; PE, pulmonary embolism.

Download figure to PowerPoint

Outcome of patients with UPE and patients with clinically suspected PE  All patients with UPE or clinically suspected PE were given anticoagulant treatment, mostly LMWH (in 83% of patients with UPE and in 84% of patients with clinically suspected PE). Objectively confirmed recurrent VTE was diagnosed in four patients with UPE (6%) and in five patients with clinically suspected PE (8%). After adjustment for previous VTE, no difference was observed in the time to recurrent VTE between patients with UPE and patients with clinically suspected PE (HR 0.77; 95% CI 0.21–2.73; = 0.70) (Fig. 3A). Major hemorrhage was observed in three patients with UPE (4%) and in five patients with clinically suspected PE (8%). A total of 11 patients (17%) with clinically suspected PE died during the follow-up period. No difference in the risk of death at 6 months was observed between patients with UPE and patients with clinically suspected PE after adjustment for performance status and tumor stage (HR 0.73; 95% CI 0.31–1.70; = 0.47) (Fig. 3B).

image

Figure 3.  Survival probability curves for 66 cancer patients with unsuspected pulmonary embolism (UPE) and 66 cancer patients with clinically suspected pulmonary embolism (SPE). P-values are based on the Cox proportional hazard model comparing patients with UPE with patients with SPE. The model was stratified for previous venous thromboembolism (VTE) for survival without recurrent VTE (A) and for the stage of tumor and performance status for global survival (B).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

The results of this study suggest that UPE is associated with poor performance status, adenocarcinoma, previous VTE, and recent chemotherapy. After adjustment for performance status and cancer stage, UPE was not associated with a decrease in survival rate. Patients with UPE were more likely to have had recent chemotherapy and central venous catheters than patients with symptomatic PE. More common risk factors for PE, such as high BMI, recent surgery, or recent hospitalization, were less often observed in patients with UPE than in patients with symptomatic PE. Patients with UPE also had fewer proximal clots than patients with symptomatic PE, but no differences in the rates of recurrent VTE or death were observed between the two groups of patients with PE.

The diagnosis of PE is difficult, because the signs and symptoms are neither sensitive nor specific, which may explain why clinically unsuspected PE is often discovered at autopsy [10–12]. Symptoms suggesting PE, such as dyspnea, leg pain, and chest pain, were found retrospectively in 41% of the patients with UPE. This has been previously reported by O’Connell et al. [13], who also found that a prior history of VTE was more common among patients with UPE, a finding confirmed by the present study. The high incidence of UPE discovered on MDCT performed for cancer staging may thus result from the high incidence of VTE in the setting of cancer, from the frequent use of MDCT in these patients, or from the fact that PE is frequently not suspected in cancer patients with non-specific respiratory or cardiac symptoms. An increase in the threshold for PE suspicion in patients with previous VTE, adenocarcinoma or recent chemotherapy and who complain of dyspnea or chest discomfort may help in the earlier diagnosis of PE in patients with cancer.

The increased incidence of UPE in patients with poor performance status, adenocarcinoma, previous VTE and recent chemotherapy may be attributable to more frequent use of MDCT in these patients, but there is no evidence for an increased rate of cancer staging with MDCT in these patients.

Our results do not confirm previous observations suggesting that UPE is associated with a decrease in overall survival in patients with cancer [3]. The survival model was adjusted for age, sex, type and stage of cancer and location of metastases in a previous study [3] and for performance status and cancer stage in our study. Performance status remained the only variable that was significantly associated with an increase in mortality risk in the present study. When multivariable analysis was performed with all variables that were significantly associated with mortality in the univariate analysis, UPE was, again, not significantly associated with mortality (data not shown). Thus, our results may suggest that the higher mortality associated with UPE is mainly related to the advanced cancer stage and the poor performance status of these patients rather than UPE itself.

Patients with UPE were less likely to have the usual risk factors for PE, such as recent surgery or recent hospitalization, than patients with clinically suspected PE. Conversely, recent chemotherapy and central venous catheters were more often recorded in patients with UPE. UPE may be more frequently related to the embolization of small clots when central venous catheters are used for chemotherapy, but this remains purely speculative and deserves appropriate investigation.

All patients with UPE were given anticoagulant treatment, mainly LMWH, as recommended for patients with VTE and cancer [14]. The rates of recurrent VTE at 6 months in the two groups of patients with PE are similar to the rate previously observed in randomized controlled trials using long-term LMWH in patients with VTE and cancer [15,16]. The lack of difference in the overall mortality and recurrent VTE rates between patients with clinically suspected PE and patients with UPE suggests that UPE has a significant recurrence risk. It was previously shown that, although overall mortality did not differ between patients with UPE and patients with clinically suspected PE, patients with isolated subsegmental UPE had a lower mortality rate [3]. We did not find any difference in recurrence rate between patients with subsegmental UPE and patients with clinically suspected PE, but the number of patients with isolated subsegmental PE was low in our study. Some data do indeed suggest that patients with cancer and unsuspected subsegmental PE who do not receive anticoagulation have a low risk of recurrent PE [3,17].

Our study has several limitations. Owing to its retrospective design, some data are lacking: for example, associated DVT was not systematically searched for in patients with UPE. Data on respiratory symptoms were not systematically recorded in the charts of patients without PE, precluding any meaningful comparison between the groups with and without UPE with regard to the symptoms associated with UPE.

In conclusion, in the context of a retrospective study, a high proportion of patients with UPE have respiratory symptoms preceding the diagnosis of PE. This may prompt the attention of the oncologist when a patient at risk for UPE complains of dyspnea or chest pain. The increased mortality rate of patients with UPE seems to be mainly related to their underlying condition and, especially, poor performance status rather than to UPE itself. Clinically suspected PE and UPE lead to similar risks of recurrent VTE during anticoagulant treatment, suggesting that they should be treated in the same manner. Prospective and/or interventional studies are required to confirm these findings.

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

M. Sahut D’Izarn, and B. Planquette: acquisition of data, conception and design, critical revision of the manuscript, and final approval of the version to be published; A. Caumont Prim and G. Chatellier: conception and design, statistical analysis, critical revision of the manuscript, and final approval of the version to be published; M. P. Revel and P. Avillach: acquisition of data, conception and design, critical revision of the manuscript, and final approval of the version to be published; O. Sanchez and G. Meyer: conception and design, drafting of the manuscript, and final approval of the version to be published.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References

The study was conducted without funding. G. Meyer: research grant support from Boehringer Ingelheim and Leo Pharma, uncompensated lectures for Boehringer Ingelheim, Leo Pharma, Bayer Healthcare, and Sanofi-Aventis, and uncompensated consulting and lectures for Bayer Healthcare, Leo Pharma, and Sanofi Aventis. The other authors state that they have no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Disclosure of Conflict of Interests
  9. References
  • 1
    Dentali F, Ageno W, Becattini C, Galli L, Gianni M, Riva N, Imberti D, Squizzato A, Venco A, Agnelli G. Prevalence and clinical history of incidental, asymptomatic pulmonary embolism: a meta-analysis. Thromb Res 2010; 125: 51822.
  • 2
    den Exter PL, Hooijer J, Dekkers OM, Huisman MV. Risk of recurrent venous thromboembolism and mortality in patients with cancer incidentally diagnosed with pulmonary embolism: a comparison with symptomatic patients. J Clin Oncol 2011; 29: 24059.
  • 3
    O’Connell C, Razavi P, Ghalichi M, Boyle S, Vasan S, Mark L, Caton A, Duddalwar V, Boswell W, Grabow K, Liebman HA. Unsuspected pulmonary emboli adversely impact survival in patients with cancer undergoing routine staging multi-row detector computed tomography scanning. J Thromb Haemost 2011; 9: 30511.
  • 4
    Sun JM, Kim TS, Lee J, Park YH, Ahn JS, Kim H, Kwon OJ, Lee KS, Park K, Ahn MJ. Unsuspected pulmonary emboli in lung cancer patients: the impact on survival and the significance of anticoagulation therapy. Lung Cancer 2010; 69: 3306.
  • 5
    Oken MM, Creech RH, Tormey DC, Horton J, Davis TE, McFadden ET, Carbone PP. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol 1982; 5: 64955.
  • 6
    Torbicki A, Perrier A, Konstantinides S, Agnelli G, Galie N, Pruszczyk P, Bengel F, Brady AJ, Ferreira D, Janssens U, Klepetko W, Mayer E, Remy-Jardin M, Bassand JP, Vahanian A, Camm J, De Caterina R, Dean V, Dickstein K, Filippatos G, et al. Guidelines on the diagnosis and management of acute pulmonary embolism: the Task Force for the Diagnosis and Management of Acute Pulmonary Embolism of the European Society of Cardiology (ESC). Eur Heart J 2008; 29: 2276315.
  • 7
    Schulman S, Kearon C. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005; 3: 6924.
  • 8
    Sant M, Allemani C, Santaquilani M, Knijn A, Marchesi F, Capocaccia R. EUROCARE-4. Survival of cancer patients diagnosed in 1995–1999. Results and commentary. Eur J Cancer 2009; 45: 93191.
  • 9
    Zhang X, Loberiza FR, Klein JP, Zhang MJ. A SAS macro for estimation of direct adjusted survival curves based on a stratified Cox regression model. Comput Methods Programs Biomed 2007; 88: 95101.
  • 10
    Le Gal G, Righini M, Roy PM, Sanchez O, Aujesky D, Bounameaux H, Perrier A. Prediction of pulmonary embolism in the emergency department: the revised Geneva score. Ann Intern Med 2006; 144: 16571.
  • 11
    Stein PD, Beemath A, Matta F, Weg JG, Yusen RD, Hales CA, Hull RD, Leeper KV Jr, Sostman HD, Tapson VF, Buckley JD, Gottschalk A, Goodman LR, Wakefied TW, Woodard PK. Clinical characteristics of patients with acute pulmonary embolism: data from PIOPED II. Am J Med 2007; 120: 8719.
  • 12
    Heit JA, Silverstein MD, Mohr DN, Petterson TM, O’Fallon WM, Melton LJ 3rd. Predictors of survival after deep vein thrombosis and pulmonary embolism: a population-based, cohort study. Arch Intern Med 1999; 159: 44553.
  • 13
    O’Connell CL, Boswell WD, Duddalwar V, Caton A, Mark LS, Vigen C, Liebman HA. Unsuspected pulmonary emboli in cancer patients: clinical correlates and relevance. J Clin Oncol 2006; 24: 492832.
  • 14
    Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133: 454S545S.
  • 15
    Lee AY, Levine MN, Baker RI, Bowden C, Kakkar AK, Prins M, Rickles FR, Julian JA, Haley S, Kovacs MJ, Gent M. Low-molecular-weight heparin versus a coumarin for the prevention of recurrent venous thromboembolism in patients with cancer. N Engl J Med 2003; 349: 14653.
  • 16
    Meyer G, Marjanovic Z, Valcke J, Lorcerie B, Gruel Y, Solal-Celigny P, Le Maignan C, Extra JM, Cottu P, Farge D. Comparison of low-molecular-weight heparin and warfarin for the secondary prevention of venous thromboembolism in patients with cancer: a randomized controlled study. Arch Intern Med 2002; 162: 172935.
  • 17
    Pena E, Kimpton M, Dennie C, Peterson R, Le Gal G, Carrier M. Difference in interpretation of computed tomography pulmonary angiography diagnosis of subsegmental thrombosis in patients with suspected pulmonary embolism. J Thromb Haemost 2012; 10: 4968.