- Email: [email protected]
Monocyte count associated with subsequent symptomatic venous thromboembolism (VTE) in hospitalized patients with solid tumors
Thrombosis Research 130 (2012) e279–e282
Contents lists available at SciVerse ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Monocyte count associated with subsequent symptomatic venous thromboembolism (VTE) in hospitalized patients with solid tumors Ponlapat Rojnuckarin ⁎, Noppacharn Uaprasert, Virote Sriuranpong Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand King Chulalongkorn Memorial Hospital, Bangkok 10330, Thailand
a r t i c l e
i n f o
Article history: Received 8 May 2012 Received in revised form 24 August 2012 Accepted 19 September 2012 Available online 15 October 2012 Keywords: Venous thromboembolism Cancer Risk factor Monocyte Cancer of unknown primary Biliary carcinoma
a b s t r a c t Background: Solid tumor is the strongest risk factor for VTE in Thai medical in-patients. This study aimed to identify the markers predicting symptomatic VTE in this group. Methods: Solid tumor patients admitted to the medical wards from June 2007 to December 2009 were monitored for VTE symptoms, excluding patients with VTE on admission. Anticoagulant prophylaxis was not given. Cases were all symptomatic VTE within 6 weeks after discharge. The controls were active solid tumor in-patients admitted in 2009 and did not develop VTE. The cases and controls were compared for the risk factors of VTE and complete blood count (CBC) on admission. Results: There were 28 radiology-confirmed VTE cases during the 2.5-year study period. There were 280 solid tumor patients without VTE as the controls. There was no difference in age (58.4 vs. 61.6 years), sex (53.6% vs. 64.3% male), presence of leg paralysis, acute infection and obesity between cases and controls, respectively. The cases showed higher absolute monocyte counts compared with the controls (0.76 vs. 0.56×109/L, p 0.013), but there were no differences in other CBC parameters. In a multivariate analysis, cancer of unknown primary (Odds ratio [OR] 13.7, 95% confidence interval [CI] 2.74-68.7, p 0.001), biliary cancer (OR 6.6, 95% CI 1.80-24.3, p 0.004) and a monocyte count over 0.5 ×109/L (OR 5.0, 95% CI 1.62-15.5, p 0.005) significantly associated with VTE. Conclusion: Metastatic diseases with obscured primary sites, biliary carcinomas and higher monocyte counts on admission are related to subsequent VTE in hospitalized cancer patients. © 2012 Elsevier Ltd. All rights reserved.
Introduction Without thromboprophylaxis, venous thromboembolism (VTE) causes significant morbidity and mortality to hospitalized medical patients. The American College of Chest Physicians, and other guidelines, recommends the use of anticoagulants in these high-risk medical patients [1]. The problem is, therefore, emerging in Asian countries where the VTE incidence is not low [2], but the preventive measures are scarcely employed [3]. One of the reasons of the under-utilization of the prophylaxis is the notion that the incidence of VTE in Asians is, although not trivial, not as high as that of the Caucasian [4,5]. Supporting this idea, a recent large randomized trial comparing enoxaparin versus placebo in mainly-Asian medical patients shows a remarkably low rate of symptomatic VTE in the placebo arm [6]. Due to a definite risk of bleeding from anticoagulants [7], this lower incidence may shift the risk-benefit ratio of heparins in Asian medical patients.
⁎ Corresponding author at: Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand. Tel.: +66 2 256 4564; fax: +66 2 253 9466. E-mail address: [email protected] (P. Rojnuckarin). 0049-3848/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.thromres.2012.09.015
The other main reason is the questionable benefit of low molecular weight heparin in this group. Although pharmacologic prophylaxis can reduce the incidence of asymptomatic and symptomatic VTE, it is unable to decrease the overall mortality, which is the most important outcome [8,9]. This conclusion is further corroborated in a recent randomized trial containing a large number of Asian patients [6]. A potential approach to this problem is to identify subgroups of patients who have the greatest risks for hospital-acquired VTE. Focusing on the prophylaxis for this group may benefit the most to the patients. Our previous study has shown that the risk scores generated from Western patients could not categorize the probability of VTE in Thais [10]. Furthermore, we found that solid tumors, not hematological cancer, and autoimmune diseases are the 2 strongest VTE risk factors in Thai hospitalized patients. In addition, cancer was more common and attributed to over half of the hospitalacquired VTE. Therefore, this group should be the first target for preventive strategies [10]. The aim of this study is to further sub-classify the risk of VTE in solid tumor patients. Previous investigations of VTE risk scores were performed in Caucasian cancer patients receiving chemotherapy in ambulatory settings [11,12]. They revealed that certain sites of cancer, complete blood count (CBC) parameters and body mass indices (BMI)
e280
P. Rojnuckarin et al. / Thrombosis Research 130 (2012) e279–e282
were independently predictive of VTE. In this study, we analyzed the VTE risks in Thai cancer patients who were admitted to the hospital.
Methods Patients who were admitted to the internal medicine wards of Chulalongkorn Hospital for 3 days or more from June 2007 to December 2009 were monitored for VTE symptoms as part of the 2 previously published studies [2,10]. The purpose is to identify subgroups of patients who are the potential targets for VTE prophylaxis. Therefore, patients with symptoms of VTE before or on admission, although the diagnosis may be confirmed later, were excluded because these incidents were considered unpreventable. The study was approved by the Ethics Committee of the Faculty of Medicine, Chulalongkorn University before enrollment. The objective of the first study from June 2007 to December 2008 was to determine the incidence of VTE. All 42 hospital-acquired VTE cases were enrolled and evaluated for potential VTE risk factors [2]. The denominator was the total number of medical patients who were admitted during the same period. The data on VTE risk factors were not prospectively gathered in patients who did not develop VTE. The objective of the second study in 2009 was to identify the VTE risk factors. There were 2495 medical inpatients prospectively evaluated for VTE risk factors [10]. One hundred and fifty-three patients received anticoagulants and were excluded from the cohort. The common anticoagulant indications were cardiogenic embolism prevention, coronary disease or previously diagnosed VTE. Only 2 cases, which were not cancer, were excluded due to anticoagulant uses for VTE prophylaxis. Consequently, 1335 cases with at least one VTE risk factor were followed for signs and symptoms of VTE until 6 weeks after discharge. All patients were contacted by nursing staff on telephone and interviewed for VTE symptoms. If there were suspected DVTs or PEs, they would be asked to come for confirmatory investigations. The patients, whom we lost contact with, were excluded from the analysis. The flow diagram of the study was previously published [10]. Patients who did not develop VTE from this cohort were used as the controls to determine VTE risk factors because the complete information was available. Deep vein thrombosis (DVT) was diagnosed by Doppler ultrasonography, while pulmonary embolism (PE) was by computerized tomographic angiography (CTA) or ventilation/perfusion (V/Q) scan. The subgroups with active solid tumors from these 2 studies were subjected to a case–control analysis. Because hematological malignancy was not a significant VTE risk factor in our study [10], it was not included. The cases were cancer patients with VTE developing during or within 6 weeks after hospitalization from June 2007 to December 2009. The controls were cancer patients in the 2009 cohort who did not develop VTE or expired without symptomatic VTE up to 6 weeks after discharge. The recorded risk factors for VTE were acute infection, congestive heart failure (New York Heart Association's functional class III or IV), chronic respiratory diseases with acute exacerbations, hemiparesis or paraparesis (motor power grade 0–3), a bed ridden or vegetative state, respiratory failure requiring invasive or non-invasive assisted ventilation, compression fracture, arthritis of lower extremities, a previous history of VTE, a family history of VTE, a previous history of cancer, presence of varicose veins, an estrogen use, history of thrombophilia, thrombocytosis (platelet count > 600 × 10 9/L), obesity (BMI ≥ 30 kg/m 2) and nephrotic syndrome. Acute infections also included febrile neutropenic episodes that required antibiotics irrespective of culture results. In addition, the sites and cell types of cancer, CBC parameters and administrations of chemotherapy during admission were evaluated. Automated CBC was performed at the hospital laboratory using ADVIA120 automated hematology analyzer on the day of admission.
Descriptive data were expressed as means ± standard deviations (SD) or percents. The differences between continuous variables were determined using Student's t test. Risk factors of VTE and mortality were expressed as odds ratios with 95% confidence interval (CI) and p values from Chi square tests. The factors that showed the p values under 0.10 from univariate analyses were subsequently put in a multivariate analysis using a binary logistic regression model. All statistical analyses were computed using SPSS 16.0 program for Windows. Results Baseline characteristics and outcomes of the patients There were a total of 308 patients. The mean (±SD) age was 61.3 ± 14.6 years ranging from 16 to 94 years and 63.3% of them were male. The most common primary sites were lung (25.6%), gastrointestinal tract (20.8%), hepatocellular (15.3%) and biliary tract cancers (7.8%). Biliary carcinomas comprised 21 cholangiocarcinomas and 3 gallbladder cancers. Prostate cancer was not included in the urogenital tract tumors due to the lower risk. Tumors of unknown primary sites were found in 2.9% (Table 1). Among the 264 patients who had staging information, 181 (68.6%) showed distant metastasis. Stage 1, 2 and 3 diseases were reported in 6.1%, 7.6% and 17.8%, respectively. During the 2.5-year study period, there were a total of 28 confirmed VTE patients. Thirteen were PE, 12 were DVT and 3 were DVT with PE. The median time to develop VTE was 9.5 days after admission, ranging from 2 to 68 days. Three VTE cases were diagnosed on day 10, 22 and 40 after discharges. In the last year of the study period, there were 280 cancer patients who did not develop VTE as a control group. Twelve symptomatic VTE cases were diagnosed in 2009 yielding the VTE incidence rate of 4.1% in this solid tumor cohort. Comparing cases and controls, the age (58.4 ± 12.1 vs. 61.6 ± 14.9 years, p 0.27) and sex (53.6% vs. 64.3% male, p 0.26) were not statistically different. The in-hospital mortality rate was 28.6% (88/308). The VTE cases showed longer mean hospital stay (25.5 ± 10.8 vs. 18.1 ± 23.9 days, p 0.11) and higher mortality rates (35.7% vs. 27.9%, p 0.38) than those of the controls, but there was no statistical significance. Primary sites in the cases and the controls are shown in Table 1. Metastatic cancer from obscured primary sites and biliary carcinoma appeared to be over-represented in the VTE cases. Notably, 23.8% (5/21) of cholangiocarcinoma and 33.3% (1/3) of gallbladder cancer developed VTE. On the other hand, hepatocellular carcinoma showed a low VTE rate. The 3 other primary sites that had VTE were hemangiopericytoma, melanoma and thymoma. Factors associated with venous thromboembolism (VTE) The comparisons of body mass index (BMI) and CBC parameters on the days of admission between cases and controls were shown in Table 2. Obese patients were uncommon. Only 1 patient in the control group had BMI of over 30 kg/m 2. There were no significant Table 1 The primary sites of cancer and venous thromboembolism (VTE). Primary sites
Total (%)⁎
VTE (%)†
Primary sites
Total (%)⁎
VTE (%)†
Lung Gastrointestinal tract Hepatoma Biliary tract Breast Head and neck
79 (25.6) 64 (20.8) 47 (15.3) 24 (7.8) 22 (7.1) 14 (4.5)
8 (10.1) 3 (4.7) 1 (2.1) 6 (25.0) 1 (4.5) 0 (0.0)
Urogenital tract Pancreas Unknown Brain Other Total
13 (4.2) 12 (3.9) 9 (2.9) 4 (1.3) 20 (6.5) 308
1 (7.7) 0 (0) 5 (55.6) 0 (0) 3 (15.0) 28
⁎ Percentage of the primary sites out of the total patients. † Percentage of VTE out of the patients with respective primary sites.
P. Rojnuckarin et al. / Thrombosis Research 130 (2012) e279–e282 Table 2 The differences in body mass index (BMI) and complete blood count (CBC) parameters between cases and controls.
Body mass index (kg/m2) Hemoglobin (g/dL) Mean corpuscular volume (fL) White blood count (x109/L) Neutrophil count (x109/L) Lymphocyte count (x109/L) Monocyte count (x109/L) Eosinophil count (x109/L) Platelet (x109/L)
Cases
Control
P value
23.2 ± 3.36 10.9 ± 1.98 83.8 ± 9.68 12.6 ± 6.64 10.0 ± 6.19 1.60 ± 0.91 0.759 ± 0.312 0.192 ± 0.254 305 ± 123
21.9 ± 3.70 11.0 ± 2.20 84.2 ± 9.43 10.8 ± 7.81 8.79 ± 7.34 1.29 ± 0.93 0.555 ± 0.416 0.117 ± 0.257 275 ± 176
0.242 0.873 0.838 0.237 0.397 0.094 0.013 0.138 0.385
e281
Factors associated with mortality Significant factors related to in-hospital mortality were displayed in Table 4. The cases showing higher white blood cell counts and lower eosinophils showed higher mortality rates in univariate analyses. In a multivariate analysis, the only factors associated with the death rate were acute infection, assisted ventilation and metastatic diseases with the odds ratios of 2.90 (95% CI 1.55-5.42, p 0.001), 2.59 (95% CI 1.10-6.12, p 0.03) and 2.08 (95% CI 1.06-4.10, p 0.034), respectively. The CBC parameters were not significantly related to mortality after adjusting for these clinical factors.
Discussion differences in CBC parameters, except the VTE cases had higher absolute monocyte counts compared with the controls. The age, cancer disease characteristics, known VTE risk factors and CBC variables were put in the univariate analysis (Table 3). The significant factors (p b 0.05) related to cancer VTE were a tumor of unknown primary, biliary tract cancer, advance stage, monocyte count over the median monocyte count of the whole group and monocytosis. The median monocyte count of the total patients was 0.484 × 10 9/L, while the upper limit of normal monocyte counts of our laboratory was 0.800 × 10 9/L. A platelet count over 600 × 10 9/L was not significantly associated with VTE (data not shown). The risk factors showing p values below 0.10 from univariate analyses (Table 3) were subjected to a multivariate analysis. The factors remained independently related to VTE in cancer were tumors from unknown primary site, biliary cancers and monocyte counts over 0.5 × 10 9/L with the odds ratios of 13.7 (95% confidence interval [95%CI] 2.74-68.7, p 0.001), 6.6 (95%CI 1.80-24.3, p 0.004) and 5.0 (95%CI 1.62-15.5, p 0.005), respectively. The age over 60 years old (p 0.227), hepatocellular carcinoma (p 0.776), advanced cancer (p 0.998), neutrophilia (absolute neutrophil counts over 7.7 × 10 9/L, p 0.707), leukocytosis (white blood cell counts over 11.0 × 10 9/L, p 0.693) were not independent risk factors. An alternative cut-off using a white blood cell count of 10 × 10 9/L also showed no statistical significance in multivariate analysis.
Table 3 Risk factors for venous thromboembolism (VTE) in univariate analysis.
Age ≥60 yr Unknown primary sites Biliary tract cancer Hepatocellular carcinoma Advanced stage (N = 264)⁎ Distant metastasis (N=264) Chemotherapy Acute infection Paraparesis Hemiparesis Bed ridden Assisted ventilation Arthritis of lower limbs History of VTE WBC ≥ 11 × 109/L ANC ≥ 7.7 × 109/L Monocyte > 0.484 × 109/L Monocyte > 0.8 × 109/L Platelet ≥ 350 × 109/L
Cases
Control
Odds ratio (95% CI)
11 (39.3%) 5 (17.9%) 6 (25.0%) 1 (3.6%) 25 (100%) 20 (80%) 1 (3.6%) 9 (32.1%) 2 (7.1%) 1 (3.6%) 3 (10.7%) 2 (7.1%) 1 (3.6%) 1 (3.6%) 15 (53.6%) 17 (60.7%) 23 (82.1%) 12 (42.9%) 10 (35.7%)
157 4 18 46 203 161 21 85 6 11 19 29 2 1 103 123 130 58 74
0.51 (0.23-1.12) 0.09 15.0 (3.77-59.7) b0.001† 3.97 (1.43-11.0) 0.005† 0.19 (0.03-1.42) 0.071 NA 0.037 1.94 (0.70-5.36) 0.195 0.46 (0.06-3.53) 0.442 1.09 (0.47-2.50) 0.845 3.51 (0.68-18.3) 0.113 0.91 (0.11-7.29) 0.926 1.65 (0.46-5.96) 0.442 0.67 (0.15-2.95) 0.590 5.15 (0.45-58.6) 0.142 10.3 (0.63-170) 0.432 1.97 (0.90-4.31) 0.084 1.97 (0.89-4.37) 0.089 5.27 (1.95-14.3) b0.001† 2.86 (1.28-6.38) 0.008† 1.55 (0.68-3.50) 0.293
(56.1%) (1.4%) (6.4%) (16.4%) (84.9%) (67.4%) (7.5%) (30.4%) (2.1%) (3.9%) (6.8%) (10.4%) (0.7%) (0.4%) (36.9%) (73.9%) (46.6%) (20.8%) (26.4%)
P value
95%CI, 95%Confidence interval; WBC, White blood count; ANC, Absolute neutrophil count. ⁎ Stage 3 and 4 diseases. † These factors remain significant under a multivariate analysis.
The novel findings of this study are the predictive roles of the high monocyte counts, tumors of unknown primary sites and biliary carcinomas for VTE in Thai cancer patients who were hospitalized in the medical wards. These factors were available in all patients on admission and thus may help physicians to beware this complication, as well as to consider for prophylactic measures. In addition to cancer-associated VTE, activation of coagulation system by tissue factor is implicated in tumor progression and metastasis [13]. This notion is supported by some of the clinical trials showing that long-term low molecular weight heparin (LMWH) may prolong the survival of cancer patients especially those without metastasis [14–16]. Metastatic cancer with obscured primary sites on admission may suggest a strong hypercoagulability that not only induces venous thrombosis, but also enhances distant tumor spreading. However, this hypothesis remains to be investigated. A positive correlation of monocyte counts and VTE has been reported in a population-based cohort study [17]. In addition, the relation was also found in ambulatory cancer patients [18]. Monocytes are the main source of microparticles in plasma [19]. These microparticles contain tissue factor, as well as negatively-charged phospholipids, and may aggravate hypercoagulability in cancer [17,19]. In contrast to Western data, we discovered that biliary carcinoma showed a higher VTE rate in Thai cancer patients. Because cholangiocarcinoma is much more prevalent in Asian population compared with developed countries [20], this cancer may be underrepresented in Western series. Although portal vein thrombosis is common in liver cancer, only DVT and PE were included in our analysis. Notably, hepatoma patients displayed a trend towards a lower VTE rate in this study. This may be explained by coagulopathy and/or thrombocytopenia in cirrhosis frequently associated with this kind of tumor. Consistent with this hypothesis, we found that hepatocellular carcinoma patients had significantly lower platelet counts compared with the other types of tumors (213 ± 177 × 10 9/L vs. 290 ± 169 × 10 9/L, p 0.005), but there were no statistical differences in hemoglobin and white blood cell counts. On the other hand, there were no differences in CBC parameters of biliary carcinoma compared with other types of cancers (Data not shown).
Table 4 Factors significantly associated with in-hospital mortality in univariate analyses.
Distant metastasis (N = 264) Acute infection Assisted ventilation WBC (x109/L) Neutrophil count (x109/L) Eosinophil count (x109/L) WBC ≥11 × 109/L WBC ≥10 × 109/L
Death (N = 88)
Survive (N = 220)
P value
61 (77.2%) 39 (44.3%) 18 (20.5%) 13.2 ± 8.83 11.12 ± 8.47 0.084 ± 0.148 43 (48.9%) 48 (54.5%)
120 (64.9%) 55 (25.0%) 13 (5.9%) 10.1 ± 7.04 8.01 ± 6.19 0.139 ± 0.288 75 (34.2%) 61 (41.6%)
0.048 0.001⁎ b0.001⁎ 0.003 0.001 0.027 0.017 0.039
WBC, White blood count. ⁎ These factors remain significant under a multivariate analysis.
e282
P. Rojnuckarin et al. / Thrombosis Research 130 (2012) e279–e282
Our study cannot demonstrate the relationship between VTE and high BMI as reported in previous studies [11,12]. The majority of patients in this series were cachectic from widespread tumors. Obese patients were very rare (Table 2). There has been a report showing that baseline leukocytosis and thrombocytosis were predictive for VTE [11]. However, the associations could not be demonstrated in this study. In contrast to previous papers, cancer patients in this series had acute complications requiring hospital admission. These precipitating conditions, such as acute infections, usually cause abnormalities of various blood cell counts. In addition, this is a relatively small study and modest effects of some factors cannot be excluded. As expected, assisted ventilation, acute infections and distant metastasis were associated with higher mortality. Leukocytosis was not a predictive factor in a multivariate analysis and the association detectable by a univariate analysis may reflect concomitant infections. Neither tumors of unknown primary nor higher monocyte counts related to in-hospital death. A number of reports demonstrated the advantage of prophylactic LMWH in hospitalized medical patients. However, only small subsets of these patients had cancer [21–23]. Cancer patients showed very high rates of both VTE and mortality compared with the other groups of medical patients [10]. The risk-benefit ratio of LMWH specifically in hospitalized patients with cancer remains to be determined. One of the limitations of the study is a small sample size. Some other weak risk factors might have been missed. Nevertheless, the detectable predictive factors are probably the strong ones. To enhance statistical power, VTE cases from 2 cohorts were combined. The characteristics of VTE in 2007–2008 may be different from the controls in 2009. However, this is not likely because both cohorts were from a single center and they were followed consecutively. In conclusion, tumors of the biliary tract and unknown primary sites, as well as higher monocyte counts, associated with subsequent VTE in Asian in-patients with cancer. These findings remain to be validated in a larger study. Conflict of interest statement The authors declare no conflict of interest. Acknowledgments The study was supported by the Ratchadapiseksompotch Fund, Faculty of Medicine, Chulalongkorn University, grant number RA21/51. We are thankful to all helpful doctors and staff of the Department of Medicine, King Chulalongkorn Memorial Hospital. References [1] Kahn SR, Lim W, Dunn AS, Cushman M, Dentali F, Akl EA, et al. Prevention of VTE in Nonsurgical Patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl.):e195S-226S.
[2] Aniwan S, Rojnuckarin P. High incidence of symptomatic venous thromboembolism in Thai hospitalized medical patients without thromboprophylaxis. Blood Coagul Fibrinolysis 2010;21:334-8. [3] Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, et al. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. Lancet 2008;371:387-94. [4] White RH, Keenan CR. Effects of race and ethnicity on the incidence of venous thromboembolism. Thromb Res 2009;123:S11-7. [5] Lee KW, Bang SM, Kim S, Lee HJ, Shin DY, Koh Y, et al. The incidence, risk factors and prognostic implications of venous thromboembolism in patients with gastric cancer. J Thromb Haemost 2010;8:540-7. [6] Kakkar AK, Cimminiello C, Goldhaber SZ, Parakh R, Wang C, Bergmann JF, et al. Low-molecular-weight heparin and mortality in acutely ill medical patients. N Engl J Med 2011;365:2463-72. [7] Lloyd NS, Douketis JD, Moinuddin I, Lim W, Crowther MA. Anticoagulant prophylaxis to prevent asymptomatic deep vein thrombosis in hospitalized medical patients: a systematic review and meta-analysis. J Thromb Haemost 2008;6:405-14. [8] Dentali F, Douketis JD, Gianni M, Lim W, Crowther MA. Meta-analysis: anticoagulant prophylaxis to prevent symptomatic venous thromboembolism in hospitalized medical patients. Ann Intern Med 2007;146:278-88. [9] Wein L, Wein S, Haas SJ, Shaw J, Krum H. Pharmacological venous thromboembolism prophylaxis in hospitalized medical patients: a meta-analysis of randomized controlled trials. Arch Intern Med 2007;167:1476-86. [10] Rojnuckarin P, Uaprasert N, Vajragupta L, Numkarunarunrote N, Tanpowpong N, Sutcharitchan P. Risk factors for symptomatic venous thromboembolism in Thai hospitalised medical patients. Thromb Haemost 2011;106:1103-8. [11] Khorana AA, Kuderer NM, Culakova E, Lyman GH, Francis CW. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood 2008;111:4902-7. [12] Ay C, Dunkler D, Marosi C, Chiriac AL, Vormittag R, Simanek R, et al. Prediction of venous thromboembolism in cancer patients. Blood 2010;116:5377-82. [13] van den Berg YW, Osanto S, Reitsma PH, Versteeg HH. The relationship between tissue factor and cancer progression: insights from bench and bedside. Blood 2012;119:924-32. [14] Altinbas M, Coskun HS, Er O, Ozkan M, Eser B, Unal A, et al. A randomized clinical trial of combination chemotherapy with and without low-molecular-weight heparin in small cell lung cancer. J Thromb Haemost 2004;2:1266-71. [15] Lee AY, Rickles FR, Julian JA, Gent M, Baker RI, Bowden C, et al. Randomized comparison of low molecular weight heparin and coumarin derivatives on the survival of patients with cancer and venous thromboembolism. J Clin Oncol 2005;23:2123-9. [16] Klerk CP, Smorenburg SM, Otten HM, Lensing AW, Prins MH, Piovella F, et al. The effect of low molecular weight heparin on survival in patients with advanced malignancy. J Clin Oncol 2005;23:2130-5. [17] Basavaraj MG, Brækkan SK, Brodin E, Østerud B, Hansen JB. Monocyte count and procoagulant functions are associated with risk of venous thromboembolism: the Tromsø study. J Thromb Haemost 2011;9:1673-6. [18] Connolly GC, Khorana AA, Kuderer NM, Culakova E, Francis CW, Lyman GH. Leukocytosis, thrombosis and early mortality in cancer patients initiating chemotherapy. Thromb Res 2010;126:113-8. [19] Owens III AP, Mackman N. Microparticles in hemostasis and thrombosis. Circ Res 2011;108:1284-97. [20] Tyson GL, El-Serag HB. Risk factors for cholangiocarcinoma. Hepatology 2011;54: 173-84. [21] Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med 1999;341:793-800. [22] Leizorovicz A, Cohen AT, Turpie AG, Olsson CG, Vaitkus PT, Goldhaber SZ, et al. Randomized, placebo-controlled trial of dalteparin for the prevention of venous thromboembolism in acutely ill medical patients. Circulation 2004;110:874-9. [23] Cohen AT, Davidson BL, Gallus AS, Lassen MR, Prins MH, Tomkowski W, et al. Efficacy and safety of fondaparinux for the prevention of venous thromboembolism in older acute medical patients: randomised placebo controlled trial. BMJ 2006;332:325-9.
Contents lists available at SciVerse ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Monocyte count associated with subsequent symptomatic venous thromboembolism (VTE) in hospitalized patients with solid tumors Ponlapat Rojnuckarin ⁎, Noppacharn Uaprasert, Virote Sriuranpong Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand King Chulalongkorn Memorial Hospital, Bangkok 10330, Thailand
a r t i c l e
i n f o
Article history: Received 8 May 2012 Received in revised form 24 August 2012 Accepted 19 September 2012 Available online 15 October 2012 Keywords: Venous thromboembolism Cancer Risk factor Monocyte Cancer of unknown primary Biliary carcinoma
a b s t r a c t Background: Solid tumor is the strongest risk factor for VTE in Thai medical in-patients. This study aimed to identify the markers predicting symptomatic VTE in this group. Methods: Solid tumor patients admitted to the medical wards from June 2007 to December 2009 were monitored for VTE symptoms, excluding patients with VTE on admission. Anticoagulant prophylaxis was not given. Cases were all symptomatic VTE within 6 weeks after discharge. The controls were active solid tumor in-patients admitted in 2009 and did not develop VTE. The cases and controls were compared for the risk factors of VTE and complete blood count (CBC) on admission. Results: There were 28 radiology-confirmed VTE cases during the 2.5-year study period. There were 280 solid tumor patients without VTE as the controls. There was no difference in age (58.4 vs. 61.6 years), sex (53.6% vs. 64.3% male), presence of leg paralysis, acute infection and obesity between cases and controls, respectively. The cases showed higher absolute monocyte counts compared with the controls (0.76 vs. 0.56×109/L, p 0.013), but there were no differences in other CBC parameters. In a multivariate analysis, cancer of unknown primary (Odds ratio [OR] 13.7, 95% confidence interval [CI] 2.74-68.7, p 0.001), biliary cancer (OR 6.6, 95% CI 1.80-24.3, p 0.004) and a monocyte count over 0.5 ×109/L (OR 5.0, 95% CI 1.62-15.5, p 0.005) significantly associated with VTE. Conclusion: Metastatic diseases with obscured primary sites, biliary carcinomas and higher monocyte counts on admission are related to subsequent VTE in hospitalized cancer patients. © 2012 Elsevier Ltd. All rights reserved.
Introduction Without thromboprophylaxis, venous thromboembolism (VTE) causes significant morbidity and mortality to hospitalized medical patients. The American College of Chest Physicians, and other guidelines, recommends the use of anticoagulants in these high-risk medical patients [1]. The problem is, therefore, emerging in Asian countries where the VTE incidence is not low [2], but the preventive measures are scarcely employed [3]. One of the reasons of the under-utilization of the prophylaxis is the notion that the incidence of VTE in Asians is, although not trivial, not as high as that of the Caucasian [4,5]. Supporting this idea, a recent large randomized trial comparing enoxaparin versus placebo in mainly-Asian medical patients shows a remarkably low rate of symptomatic VTE in the placebo arm [6]. Due to a definite risk of bleeding from anticoagulants [7], this lower incidence may shift the risk-benefit ratio of heparins in Asian medical patients.
⁎ Corresponding author at: Department of Medicine, Faculty of Medicine, Chulalongkorn University, Bangkok 10330, Thailand. Tel.: +66 2 256 4564; fax: +66 2 253 9466. E-mail address: [email protected] (P. Rojnuckarin). 0049-3848/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.thromres.2012.09.015
The other main reason is the questionable benefit of low molecular weight heparin in this group. Although pharmacologic prophylaxis can reduce the incidence of asymptomatic and symptomatic VTE, it is unable to decrease the overall mortality, which is the most important outcome [8,9]. This conclusion is further corroborated in a recent randomized trial containing a large number of Asian patients [6]. A potential approach to this problem is to identify subgroups of patients who have the greatest risks for hospital-acquired VTE. Focusing on the prophylaxis for this group may benefit the most to the patients. Our previous study has shown that the risk scores generated from Western patients could not categorize the probability of VTE in Thais [10]. Furthermore, we found that solid tumors, not hematological cancer, and autoimmune diseases are the 2 strongest VTE risk factors in Thai hospitalized patients. In addition, cancer was more common and attributed to over half of the hospitalacquired VTE. Therefore, this group should be the first target for preventive strategies [10]. The aim of this study is to further sub-classify the risk of VTE in solid tumor patients. Previous investigations of VTE risk scores were performed in Caucasian cancer patients receiving chemotherapy in ambulatory settings [11,12]. They revealed that certain sites of cancer, complete blood count (CBC) parameters and body mass indices (BMI)
e280
P. Rojnuckarin et al. / Thrombosis Research 130 (2012) e279–e282
were independently predictive of VTE. In this study, we analyzed the VTE risks in Thai cancer patients who were admitted to the hospital.
Methods Patients who were admitted to the internal medicine wards of Chulalongkorn Hospital for 3 days or more from June 2007 to December 2009 were monitored for VTE symptoms as part of the 2 previously published studies [2,10]. The purpose is to identify subgroups of patients who are the potential targets for VTE prophylaxis. Therefore, patients with symptoms of VTE before or on admission, although the diagnosis may be confirmed later, were excluded because these incidents were considered unpreventable. The study was approved by the Ethics Committee of the Faculty of Medicine, Chulalongkorn University before enrollment. The objective of the first study from June 2007 to December 2008 was to determine the incidence of VTE. All 42 hospital-acquired VTE cases were enrolled and evaluated for potential VTE risk factors [2]. The denominator was the total number of medical patients who were admitted during the same period. The data on VTE risk factors were not prospectively gathered in patients who did not develop VTE. The objective of the second study in 2009 was to identify the VTE risk factors. There were 2495 medical inpatients prospectively evaluated for VTE risk factors [10]. One hundred and fifty-three patients received anticoagulants and were excluded from the cohort. The common anticoagulant indications were cardiogenic embolism prevention, coronary disease or previously diagnosed VTE. Only 2 cases, which were not cancer, were excluded due to anticoagulant uses for VTE prophylaxis. Consequently, 1335 cases with at least one VTE risk factor were followed for signs and symptoms of VTE until 6 weeks after discharge. All patients were contacted by nursing staff on telephone and interviewed for VTE symptoms. If there were suspected DVTs or PEs, they would be asked to come for confirmatory investigations. The patients, whom we lost contact with, were excluded from the analysis. The flow diagram of the study was previously published [10]. Patients who did not develop VTE from this cohort were used as the controls to determine VTE risk factors because the complete information was available. Deep vein thrombosis (DVT) was diagnosed by Doppler ultrasonography, while pulmonary embolism (PE) was by computerized tomographic angiography (CTA) or ventilation/perfusion (V/Q) scan. The subgroups with active solid tumors from these 2 studies were subjected to a case–control analysis. Because hematological malignancy was not a significant VTE risk factor in our study [10], it was not included. The cases were cancer patients with VTE developing during or within 6 weeks after hospitalization from June 2007 to December 2009. The controls were cancer patients in the 2009 cohort who did not develop VTE or expired without symptomatic VTE up to 6 weeks after discharge. The recorded risk factors for VTE were acute infection, congestive heart failure (New York Heart Association's functional class III or IV), chronic respiratory diseases with acute exacerbations, hemiparesis or paraparesis (motor power grade 0–3), a bed ridden or vegetative state, respiratory failure requiring invasive or non-invasive assisted ventilation, compression fracture, arthritis of lower extremities, a previous history of VTE, a family history of VTE, a previous history of cancer, presence of varicose veins, an estrogen use, history of thrombophilia, thrombocytosis (platelet count > 600 × 10 9/L), obesity (BMI ≥ 30 kg/m 2) and nephrotic syndrome. Acute infections also included febrile neutropenic episodes that required antibiotics irrespective of culture results. In addition, the sites and cell types of cancer, CBC parameters and administrations of chemotherapy during admission were evaluated. Automated CBC was performed at the hospital laboratory using ADVIA120 automated hematology analyzer on the day of admission.
Descriptive data were expressed as means ± standard deviations (SD) or percents. The differences between continuous variables were determined using Student's t test. Risk factors of VTE and mortality were expressed as odds ratios with 95% confidence interval (CI) and p values from Chi square tests. The factors that showed the p values under 0.10 from univariate analyses were subsequently put in a multivariate analysis using a binary logistic regression model. All statistical analyses were computed using SPSS 16.0 program for Windows. Results Baseline characteristics and outcomes of the patients There were a total of 308 patients. The mean (±SD) age was 61.3 ± 14.6 years ranging from 16 to 94 years and 63.3% of them were male. The most common primary sites were lung (25.6%), gastrointestinal tract (20.8%), hepatocellular (15.3%) and biliary tract cancers (7.8%). Biliary carcinomas comprised 21 cholangiocarcinomas and 3 gallbladder cancers. Prostate cancer was not included in the urogenital tract tumors due to the lower risk. Tumors of unknown primary sites were found in 2.9% (Table 1). Among the 264 patients who had staging information, 181 (68.6%) showed distant metastasis. Stage 1, 2 and 3 diseases were reported in 6.1%, 7.6% and 17.8%, respectively. During the 2.5-year study period, there were a total of 28 confirmed VTE patients. Thirteen were PE, 12 were DVT and 3 were DVT with PE. The median time to develop VTE was 9.5 days after admission, ranging from 2 to 68 days. Three VTE cases were diagnosed on day 10, 22 and 40 after discharges. In the last year of the study period, there were 280 cancer patients who did not develop VTE as a control group. Twelve symptomatic VTE cases were diagnosed in 2009 yielding the VTE incidence rate of 4.1% in this solid tumor cohort. Comparing cases and controls, the age (58.4 ± 12.1 vs. 61.6 ± 14.9 years, p 0.27) and sex (53.6% vs. 64.3% male, p 0.26) were not statistically different. The in-hospital mortality rate was 28.6% (88/308). The VTE cases showed longer mean hospital stay (25.5 ± 10.8 vs. 18.1 ± 23.9 days, p 0.11) and higher mortality rates (35.7% vs. 27.9%, p 0.38) than those of the controls, but there was no statistical significance. Primary sites in the cases and the controls are shown in Table 1. Metastatic cancer from obscured primary sites and biliary carcinoma appeared to be over-represented in the VTE cases. Notably, 23.8% (5/21) of cholangiocarcinoma and 33.3% (1/3) of gallbladder cancer developed VTE. On the other hand, hepatocellular carcinoma showed a low VTE rate. The 3 other primary sites that had VTE were hemangiopericytoma, melanoma and thymoma. Factors associated with venous thromboembolism (VTE) The comparisons of body mass index (BMI) and CBC parameters on the days of admission between cases and controls were shown in Table 2. Obese patients were uncommon. Only 1 patient in the control group had BMI of over 30 kg/m 2. There were no significant Table 1 The primary sites of cancer and venous thromboembolism (VTE). Primary sites
Total (%)⁎
VTE (%)†
Primary sites
Total (%)⁎
VTE (%)†
Lung Gastrointestinal tract Hepatoma Biliary tract Breast Head and neck
79 (25.6) 64 (20.8) 47 (15.3) 24 (7.8) 22 (7.1) 14 (4.5)
8 (10.1) 3 (4.7) 1 (2.1) 6 (25.0) 1 (4.5) 0 (0.0)
Urogenital tract Pancreas Unknown Brain Other Total
13 (4.2) 12 (3.9) 9 (2.9) 4 (1.3) 20 (6.5) 308
1 (7.7) 0 (0) 5 (55.6) 0 (0) 3 (15.0) 28
⁎ Percentage of the primary sites out of the total patients. † Percentage of VTE out of the patients with respective primary sites.
P. Rojnuckarin et al. / Thrombosis Research 130 (2012) e279–e282 Table 2 The differences in body mass index (BMI) and complete blood count (CBC) parameters between cases and controls.
Body mass index (kg/m2) Hemoglobin (g/dL) Mean corpuscular volume (fL) White blood count (x109/L) Neutrophil count (x109/L) Lymphocyte count (x109/L) Monocyte count (x109/L) Eosinophil count (x109/L) Platelet (x109/L)
Cases
Control
P value
23.2 ± 3.36 10.9 ± 1.98 83.8 ± 9.68 12.6 ± 6.64 10.0 ± 6.19 1.60 ± 0.91 0.759 ± 0.312 0.192 ± 0.254 305 ± 123
21.9 ± 3.70 11.0 ± 2.20 84.2 ± 9.43 10.8 ± 7.81 8.79 ± 7.34 1.29 ± 0.93 0.555 ± 0.416 0.117 ± 0.257 275 ± 176
0.242 0.873 0.838 0.237 0.397 0.094 0.013 0.138 0.385
e281
Factors associated with mortality Significant factors related to in-hospital mortality were displayed in Table 4. The cases showing higher white blood cell counts and lower eosinophils showed higher mortality rates in univariate analyses. In a multivariate analysis, the only factors associated with the death rate were acute infection, assisted ventilation and metastatic diseases with the odds ratios of 2.90 (95% CI 1.55-5.42, p 0.001), 2.59 (95% CI 1.10-6.12, p 0.03) and 2.08 (95% CI 1.06-4.10, p 0.034), respectively. The CBC parameters were not significantly related to mortality after adjusting for these clinical factors.
Discussion differences in CBC parameters, except the VTE cases had higher absolute monocyte counts compared with the controls. The age, cancer disease characteristics, known VTE risk factors and CBC variables were put in the univariate analysis (Table 3). The significant factors (p b 0.05) related to cancer VTE were a tumor of unknown primary, biliary tract cancer, advance stage, monocyte count over the median monocyte count of the whole group and monocytosis. The median monocyte count of the total patients was 0.484 × 10 9/L, while the upper limit of normal monocyte counts of our laboratory was 0.800 × 10 9/L. A platelet count over 600 × 10 9/L was not significantly associated with VTE (data not shown). The risk factors showing p values below 0.10 from univariate analyses (Table 3) were subjected to a multivariate analysis. The factors remained independently related to VTE in cancer were tumors from unknown primary site, biliary cancers and monocyte counts over 0.5 × 10 9/L with the odds ratios of 13.7 (95% confidence interval [95%CI] 2.74-68.7, p 0.001), 6.6 (95%CI 1.80-24.3, p 0.004) and 5.0 (95%CI 1.62-15.5, p 0.005), respectively. The age over 60 years old (p 0.227), hepatocellular carcinoma (p 0.776), advanced cancer (p 0.998), neutrophilia (absolute neutrophil counts over 7.7 × 10 9/L, p 0.707), leukocytosis (white blood cell counts over 11.0 × 10 9/L, p 0.693) were not independent risk factors. An alternative cut-off using a white blood cell count of 10 × 10 9/L also showed no statistical significance in multivariate analysis.
Table 3 Risk factors for venous thromboembolism (VTE) in univariate analysis.
Age ≥60 yr Unknown primary sites Biliary tract cancer Hepatocellular carcinoma Advanced stage (N = 264)⁎ Distant metastasis (N=264) Chemotherapy Acute infection Paraparesis Hemiparesis Bed ridden Assisted ventilation Arthritis of lower limbs History of VTE WBC ≥ 11 × 109/L ANC ≥ 7.7 × 109/L Monocyte > 0.484 × 109/L Monocyte > 0.8 × 109/L Platelet ≥ 350 × 109/L
Cases
Control
Odds ratio (95% CI)
11 (39.3%) 5 (17.9%) 6 (25.0%) 1 (3.6%) 25 (100%) 20 (80%) 1 (3.6%) 9 (32.1%) 2 (7.1%) 1 (3.6%) 3 (10.7%) 2 (7.1%) 1 (3.6%) 1 (3.6%) 15 (53.6%) 17 (60.7%) 23 (82.1%) 12 (42.9%) 10 (35.7%)
157 4 18 46 203 161 21 85 6 11 19 29 2 1 103 123 130 58 74
0.51 (0.23-1.12) 0.09 15.0 (3.77-59.7) b0.001† 3.97 (1.43-11.0) 0.005† 0.19 (0.03-1.42) 0.071 NA 0.037 1.94 (0.70-5.36) 0.195 0.46 (0.06-3.53) 0.442 1.09 (0.47-2.50) 0.845 3.51 (0.68-18.3) 0.113 0.91 (0.11-7.29) 0.926 1.65 (0.46-5.96) 0.442 0.67 (0.15-2.95) 0.590 5.15 (0.45-58.6) 0.142 10.3 (0.63-170) 0.432 1.97 (0.90-4.31) 0.084 1.97 (0.89-4.37) 0.089 5.27 (1.95-14.3) b0.001† 2.86 (1.28-6.38) 0.008† 1.55 (0.68-3.50) 0.293
(56.1%) (1.4%) (6.4%) (16.4%) (84.9%) (67.4%) (7.5%) (30.4%) (2.1%) (3.9%) (6.8%) (10.4%) (0.7%) (0.4%) (36.9%) (73.9%) (46.6%) (20.8%) (26.4%)
P value
95%CI, 95%Confidence interval; WBC, White blood count; ANC, Absolute neutrophil count. ⁎ Stage 3 and 4 diseases. † These factors remain significant under a multivariate analysis.
The novel findings of this study are the predictive roles of the high monocyte counts, tumors of unknown primary sites and biliary carcinomas for VTE in Thai cancer patients who were hospitalized in the medical wards. These factors were available in all patients on admission and thus may help physicians to beware this complication, as well as to consider for prophylactic measures. In addition to cancer-associated VTE, activation of coagulation system by tissue factor is implicated in tumor progression and metastasis [13]. This notion is supported by some of the clinical trials showing that long-term low molecular weight heparin (LMWH) may prolong the survival of cancer patients especially those without metastasis [14–16]. Metastatic cancer with obscured primary sites on admission may suggest a strong hypercoagulability that not only induces venous thrombosis, but also enhances distant tumor spreading. However, this hypothesis remains to be investigated. A positive correlation of monocyte counts and VTE has been reported in a population-based cohort study [17]. In addition, the relation was also found in ambulatory cancer patients [18]. Monocytes are the main source of microparticles in plasma [19]. These microparticles contain tissue factor, as well as negatively-charged phospholipids, and may aggravate hypercoagulability in cancer [17,19]. In contrast to Western data, we discovered that biliary carcinoma showed a higher VTE rate in Thai cancer patients. Because cholangiocarcinoma is much more prevalent in Asian population compared with developed countries [20], this cancer may be underrepresented in Western series. Although portal vein thrombosis is common in liver cancer, only DVT and PE were included in our analysis. Notably, hepatoma patients displayed a trend towards a lower VTE rate in this study. This may be explained by coagulopathy and/or thrombocytopenia in cirrhosis frequently associated with this kind of tumor. Consistent with this hypothesis, we found that hepatocellular carcinoma patients had significantly lower platelet counts compared with the other types of tumors (213 ± 177 × 10 9/L vs. 290 ± 169 × 10 9/L, p 0.005), but there were no statistical differences in hemoglobin and white blood cell counts. On the other hand, there were no differences in CBC parameters of biliary carcinoma compared with other types of cancers (Data not shown).
Table 4 Factors significantly associated with in-hospital mortality in univariate analyses.
Distant metastasis (N = 264) Acute infection Assisted ventilation WBC (x109/L) Neutrophil count (x109/L) Eosinophil count (x109/L) WBC ≥11 × 109/L WBC ≥10 × 109/L
Death (N = 88)
Survive (N = 220)
P value
61 (77.2%) 39 (44.3%) 18 (20.5%) 13.2 ± 8.83 11.12 ± 8.47 0.084 ± 0.148 43 (48.9%) 48 (54.5%)
120 (64.9%) 55 (25.0%) 13 (5.9%) 10.1 ± 7.04 8.01 ± 6.19 0.139 ± 0.288 75 (34.2%) 61 (41.6%)
0.048 0.001⁎ b0.001⁎ 0.003 0.001 0.027 0.017 0.039
WBC, White blood count. ⁎ These factors remain significant under a multivariate analysis.
e282
P. Rojnuckarin et al. / Thrombosis Research 130 (2012) e279–e282
Our study cannot demonstrate the relationship between VTE and high BMI as reported in previous studies [11,12]. The majority of patients in this series were cachectic from widespread tumors. Obese patients were very rare (Table 2). There has been a report showing that baseline leukocytosis and thrombocytosis were predictive for VTE [11]. However, the associations could not be demonstrated in this study. In contrast to previous papers, cancer patients in this series had acute complications requiring hospital admission. These precipitating conditions, such as acute infections, usually cause abnormalities of various blood cell counts. In addition, this is a relatively small study and modest effects of some factors cannot be excluded. As expected, assisted ventilation, acute infections and distant metastasis were associated with higher mortality. Leukocytosis was not a predictive factor in a multivariate analysis and the association detectable by a univariate analysis may reflect concomitant infections. Neither tumors of unknown primary nor higher monocyte counts related to in-hospital death. A number of reports demonstrated the advantage of prophylactic LMWH in hospitalized medical patients. However, only small subsets of these patients had cancer [21–23]. Cancer patients showed very high rates of both VTE and mortality compared with the other groups of medical patients [10]. The risk-benefit ratio of LMWH specifically in hospitalized patients with cancer remains to be determined. One of the limitations of the study is a small sample size. Some other weak risk factors might have been missed. Nevertheless, the detectable predictive factors are probably the strong ones. To enhance statistical power, VTE cases from 2 cohorts were combined. The characteristics of VTE in 2007–2008 may be different from the controls in 2009. However, this is not likely because both cohorts were from a single center and they were followed consecutively. In conclusion, tumors of the biliary tract and unknown primary sites, as well as higher monocyte counts, associated with subsequent VTE in Asian in-patients with cancer. These findings remain to be validated in a larger study. Conflict of interest statement The authors declare no conflict of interest. Acknowledgments The study was supported by the Ratchadapiseksompotch Fund, Faculty of Medicine, Chulalongkorn University, grant number RA21/51. We are thankful to all helpful doctors and staff of the Department of Medicine, King Chulalongkorn Memorial Hospital. References [1] Kahn SR, Lim W, Dunn AS, Cushman M, Dentali F, Akl EA, et al. Prevention of VTE in Nonsurgical Patients: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141(2 Suppl.):e195S-226S.
[2] Aniwan S, Rojnuckarin P. High incidence of symptomatic venous thromboembolism in Thai hospitalized medical patients without thromboprophylaxis. Blood Coagul Fibrinolysis 2010;21:334-8. [3] Cohen AT, Tapson VF, Bergmann JF, Goldhaber SZ, Kakkar AK, Deslandes B, et al. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. Lancet 2008;371:387-94. [4] White RH, Keenan CR. Effects of race and ethnicity on the incidence of venous thromboembolism. Thromb Res 2009;123:S11-7. [5] Lee KW, Bang SM, Kim S, Lee HJ, Shin DY, Koh Y, et al. The incidence, risk factors and prognostic implications of venous thromboembolism in patients with gastric cancer. J Thromb Haemost 2010;8:540-7. [6] Kakkar AK, Cimminiello C, Goldhaber SZ, Parakh R, Wang C, Bergmann JF, et al. Low-molecular-weight heparin and mortality in acutely ill medical patients. N Engl J Med 2011;365:2463-72. [7] Lloyd NS, Douketis JD, Moinuddin I, Lim W, Crowther MA. Anticoagulant prophylaxis to prevent asymptomatic deep vein thrombosis in hospitalized medical patients: a systematic review and meta-analysis. J Thromb Haemost 2008;6:405-14. [8] Dentali F, Douketis JD, Gianni M, Lim W, Crowther MA. Meta-analysis: anticoagulant prophylaxis to prevent symptomatic venous thromboembolism in hospitalized medical patients. Ann Intern Med 2007;146:278-88. [9] Wein L, Wein S, Haas SJ, Shaw J, Krum H. Pharmacological venous thromboembolism prophylaxis in hospitalized medical patients: a meta-analysis of randomized controlled trials. Arch Intern Med 2007;167:1476-86. [10] Rojnuckarin P, Uaprasert N, Vajragupta L, Numkarunarunrote N, Tanpowpong N, Sutcharitchan P. Risk factors for symptomatic venous thromboembolism in Thai hospitalised medical patients. Thromb Haemost 2011;106:1103-8. [11] Khorana AA, Kuderer NM, Culakova E, Lyman GH, Francis CW. Development and validation of a predictive model for chemotherapy-associated thrombosis. Blood 2008;111:4902-7. [12] Ay C, Dunkler D, Marosi C, Chiriac AL, Vormittag R, Simanek R, et al. Prediction of venous thromboembolism in cancer patients. Blood 2010;116:5377-82. [13] van den Berg YW, Osanto S, Reitsma PH, Versteeg HH. The relationship between tissue factor and cancer progression: insights from bench and bedside. Blood 2012;119:924-32. [14] Altinbas M, Coskun HS, Er O, Ozkan M, Eser B, Unal A, et al. A randomized clinical trial of combination chemotherapy with and without low-molecular-weight heparin in small cell lung cancer. J Thromb Haemost 2004;2:1266-71. [15] Lee AY, Rickles FR, Julian JA, Gent M, Baker RI, Bowden C, et al. Randomized comparison of low molecular weight heparin and coumarin derivatives on the survival of patients with cancer and venous thromboembolism. J Clin Oncol 2005;23:2123-9. [16] Klerk CP, Smorenburg SM, Otten HM, Lensing AW, Prins MH, Piovella F, et al. The effect of low molecular weight heparin on survival in patients with advanced malignancy. J Clin Oncol 2005;23:2130-5. [17] Basavaraj MG, Brækkan SK, Brodin E, Østerud B, Hansen JB. Monocyte count and procoagulant functions are associated with risk of venous thromboembolism: the Tromsø study. J Thromb Haemost 2011;9:1673-6. [18] Connolly GC, Khorana AA, Kuderer NM, Culakova E, Francis CW, Lyman GH. Leukocytosis, thrombosis and early mortality in cancer patients initiating chemotherapy. Thromb Res 2010;126:113-8. [19] Owens III AP, Mackman N. Microparticles in hemostasis and thrombosis. Circ Res 2011;108:1284-97. [20] Tyson GL, El-Serag HB. Risk factors for cholangiocarcinoma. Hepatology 2011;54: 173-84. [21] Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med 1999;341:793-800. [22] Leizorovicz A, Cohen AT, Turpie AG, Olsson CG, Vaitkus PT, Goldhaber SZ, et al. Randomized, placebo-controlled trial of dalteparin for the prevention of venous thromboembolism in acutely ill medical patients. Circulation 2004;110:874-9. [23] Cohen AT, Davidson BL, Gallus AS, Lassen MR, Prins MH, Tomkowski W, et al. Efficacy and safety of fondaparinux for the prevention of venous thromboembolism in older acute medical patients: randomised placebo controlled trial. BMJ 2006;332:325-9.