Risk factors for mortality in patients with upper extremity and internal jugular deep venous thrombosis

From the Eastern Vascular Society

Risk factors for mortality in patients with upper extremity and internal jugular deep venous thrombosis Anil Hingorani, MD, Enrico Ascher, MD, Natalia Markevich, MD, RVT, William Yorkovich, RPA, Richard Schutzer, MD, Manikyam Mutyala, MD, Suresh Nahata, MD, and Theresa Jacob, PhD, Brooklyn, NY Objective: To elucidate the natural history of upper extremity deep venous thrombosis (UEDVT), we examined factors that may contribute to the high mortality associated with UEDVT. Methods: Five hundred forty-six patients were diagnosed with acute internal jugular/subclavian/axillary deep venous thrombosis from January 1992 to June 2003 by duplex scanning at our institution. There were 329 women (60%). The mean age ⴞ SD was 68 ⴞ 17 years (range, 1-101 years). Risk factors for UEDVT were the presence of a central venous catheter or pacemaker in 327 patients (60%) and a history of malignancy in 119 patients (22%). Risk factors for mortality within 2 months of the diagnosis of UEDVT that were analyzed included age, sex, presence of a central venous catheter or pacemaker, history of malignancy, location of UEDVT, concomitant lower extremity deep venous thrombosis, systemic anticoagulation, placement of a superior vena caval filter, and pulmonary embolism. Results: The overall mortality rate at 2 months was 29.6%. The number of patients diagnosed with pulmonary embolism by positive ventilation/perfusion scan or computed tomographic scan was 26 (5%). The presence of a central venous catheter or pacemaker (P < .001), concomitant lower extremity deep venous thrombosis (P ⴝ .04), not undergoing systemic anticoagulation (P ⴝ .002), and the placement of a superior vena caval filter (P ⴝ .02) were associated with mortality within 2 months of the diagnosis of UEDVT by univariate analysis. Pulmonary embolism (P ⴝ .42), sex (P ⴝ .65), and a history of malignancy (P ⴝ .96) were not. Conclusions: These data suggest that the high associated mortality of UEDVT may be due to the underlying characteristics of the patients’ disease process and may not be a direct consequence of the UEDVT itself. ( J Vasc Surg 2005;41:476-8.)

Prior literature examining upper extremity deep venous thrombosis (UEDVT) suggested that it is a benign process associated with little mortality and morbidity. However, in our prior review of 170 patients diagnosed with UEDVT at our institution over a 5-year period, we noted a significant mortality associated with UEDVT. The 1- and 3-month mortality rates for the entire study group were 16% and 34%, respectively.1 In an effort to further explore this issue, we compared the mortality associated with UEDVT and that associated with lower extremity deep venous thrombosis (LEDVT) and analyzed the risk factors associated with this observed mortality.2 We reviewed records and performed interviews of 430 patients with LEDVT and 52 patients with UEDVT who presented to our institution between January 1994 and June 1995. Pulmonary embolism was documented by ventilation/perfusion lung scan in 9 (17%) of 52 patients with UEDVT and 33 (8%) of 430 patients with LEDVT (P ⬍ .05). Twenty-five (48%) of the UEDVT patients died within 6 months of the diagnosis of From the Division of Vascular Surgery, Department of Surgery, Maimonides Medical Center. Competition of interest: none. Presented at the Eastern Vascular Society, Philadelphia, Pa, Apr 30, 2004. Reprint requests: Enrico Ascher, MD, Division of Vascular Surgery, Maimonides Medical Center, 4802 Tenth Ave, Brooklyn, NY 11219 (e-mail: [email protected]). 0741-5214/$30.00 Copyright © 2005 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2004.12.038

476

UEDVT. Conversely, 14 patients (13%) in the LEDVT group died within 6 months of the diagnosis of LEDVT (P ⬍ .0002). In an attempt to further investigate the high mortality associated with UEDVT, we examined various subgroups of the total population. However, in an analysis of the various subgroups of these prior reports, it was noted that small numbers of some of the groups limited the conclusions that could be drawn.3 To further investigate these issues, we herein expanded the analysis to included data from our 11-year experience with UEDVT. PATIENTS AND METHODS A total of 2451 patients underwent duplex ultrasonography to rule out UEDVT at the Vascular Laboratory at the Maimonides Medical Center (accredited by the Intersocietal Commission for the Accreditation of Vascular Laboratories) from January 1992 to June 2003. This examination included visualization of the subclavian, axillary, brachial, internal jugular, radial, and ulnar veins. The criteria for making the diagnosis of acute UEDVT were an absence of augmentation of flow with respiration and augmentation maneuvers, an inability to compress the vein, whenever applicable, and hypoechoic signals within the lumen of the vein. All examinations were recorded on videotape and interpreted by attending vascular surgeons in conjunction with registered vascular technologists. Data were reviewed from medical records, office charts, databases from the

JOURNAL OF VASCULAR SURGERY Volume 41, Number 3

Hingorani et al 477

Maimonides Medical Center radiology department, the tumor registry, the Social Security Death Index, the New York City Department of Vital Statistics, telephone interviews, and clinical examinations. When pulmonary embolism was clinically suspected, ventilation/perfusion scans or computed tomography with intravenous contrast was used to confirm the diagnosis. Scans were considered to be positive if there were large segmental perfusion defects without corresponding ventilation or roentgenographic abnormalities. Intermediateprobability scans were considered negative. Statistical analysis was performed by using ␹2 analysis and Student t tests with the Texasoft Winks 4.21 program and Graphpad Instat 2.05a program (SPSS Inc, Chicago, Ill). RESULTS Of the 2451 upper extremity venous duplex examinations, 546 (22%) were found to be consistent with acute UEDVT and formed the basis of the study. UEDVT was detected bilaterally in 46 patients (8%). A total of 329 patients were female (60%). Ages ranged from 1 to 101 years (mean ⫾ SD, 68 ⫾ 17 years). Risk factors for UEDVT were the presence of a central venous catheter or pacemaker in 327 patients (60%) and a history of malignancy in 119 patients (22%). The overall mortality rate at 2 months was 29.6%. The number of patients diagnosed with pulmonary embolism by positive ventilation/perfusion scan or computed tomographic scan was 26 (5%). Figure demonstrates mortality as a function of time after the diagnosis of UEDVT. In an effort to further explore this high associated mortality, we further stratified the patients. Table I lists the sites of deep venous thrombosis (DVT) and the associated mortality. Table II shows the associated mortality as a function of age and sex. Table III shows the incidence of various factors and their associated mortality. No correlation was found with mortality by year of diagnosis, side of UEDVT, or bilaterality of UEDVT. Twenty-one percent of the patients who died within 2 months had cancer. Of the 546 patients, the primary treatment was systemic anticoagulation with intravenous heparin in 307 patients (56%). Anticoagulation consisted of the administration of heparin to obtain an activated partial thromboplastin time of 2.0 to 3.0 times that of control, followed by the administration of coumadin to obtain an international normalized ratio of 2.0 to 3.0 for 3 to 6 months. Seventy patients (13%) had a complication from anticoagulation or a contraindication to anticoagulation and underwent placement of a superior vena caval filter.4-6 Three patients (0.5%) were diagnosed with thoracic outlet syndrome and underwent thrombolysis and operative decompression. The remaining patients did not undergo anticoagulation, according to the decision of the primary care physician. DISCUSSION Previous literature has suggested that UEDVT affects mostly young men and that, because it is associated with little morbidity and mortality, it is therefore of little conse-

Mortality by month. IJ DVT, Internal jugular deep venous thrombosis.

Table I. Association of 2-month mortality with the location of DVT Variable No. patients 2-mo mortality

Group 1

Group 2

Group 3

304 (56%) 29%

80 (15%) 26%

162 (30%) 33%

DVT, Deep venous thrombosis. Group 1, subclavian/axillary DVT; group 2, isolated internal jugular DVT; group 3, combined subclavian/axillary and internal jugular DVT. P ⫽ .45 between groups.

quence.7-10 Much of this review challenges these views and supports our previous theory that UEDVT is associated with a significant incidence of morbidity and mortality. Although the patients did not seem to be dying from clinically apparent pulmonary embolism, many of the patients were extremely ill with multiorgan system dysfunction. For example, an elderly septic patient in the intensive care unit with a history of cancer and respiratory failure and in need of a central line would be expected to have a high mortality.11-14 On the basis of these data, it seems that the location of the UEDVT does not matter, because even isolated internal jugular DVT was associated with a high mortality as compared with subclavian or axillary DVT. This tends to support the concept that the high mortality associated with UEDVT may not necessarily be related to the DVT itself but rather to the underlying characteristics of the patient. This also suggests that patients with even an isolated internal jugular DVT need to be treated as aggressively as those with more extensive thrombosis and that internal jugular DVT is not a benign entity. Many patients were not anticoagulated by their primary medical doctor. This often was out of fear of complications of anticoagulation in patients who often had sepsis, cancer, or respiratory failure. Therefore, the effect of anticoagulation on the incidence of clinically evident pulmonary em-

JOURNAL OF VASCULAR SURGERY March 2005

478 Hingorani et al

Table II. Association of mortality at 2 months after diagnosis of UEDVT with age and sex Variable

Female sex

Age ⬎60 y

Age ⬎70 y

Age ⬎80 y

No. with factor present 2-mo mortality with factor present 2-mo mortality with factor absent P value

329 (60%) 26% 32% .65

405 (74%) 28% 11% ⬍.001

316 (58%) 35% 17% ⬍.001

178 (32%) 41% 22% ⬍.001

UEDVT, Upper extremity deep venous thrombosis.

Table III. Association of mortality at 2 months after diagnosis of UEDVT with various factors

Variable

PE

Concomitant LEDVT

No. with factor present 2-mo mortality with factor present 2-mo mortality with factor absent P value

26 (5%) 35% 27% .42

119 (22%) 34% 25% .04

History of neoplasm

Presence of central line*

Anticoagulation

Placement of SVC filter

119 (22%) 28% 28% .96

327 (60%) 33% 19% ⬍.001

307 (56%) 23% 36% .002

63 (12%) 37% 16% ⬍.001

UEDVT, Upper extremity deep venous thrombosis; PE, pulmonary embolism; LEDVT, lower extremity deep venous thrombosis; SVC, superior vena cava. *The presence of a central venous catheter included 60 (11%) patients with a transvenous pacemaker.

bolism and mortality cannot be accurately assessed with these data because of a possible selection bias. However, we do suggest that UEDVT should be considered as a marker for severity of illness rather than just a possible source of thromboembolism. Indeed, in a prior series, we documented increased APACHE III scores of patients with UEDVT as compared with LEDVT.2 Because the incidence of UEDVT seems to be much less than that of LEDVT, it would seem that the upper extremity veins are less prone to DVT. Thereby, the underlying processes (hypercoagulable states, low flow, inflammatory states, and so on)15 that result in the thrombosis of the upper extremity veins might be more severe and might result in the high associated mortality. From year to year, no difference was detected between the mortality rates despite recent advances in critical care medicine. This suggests that the mechanisms that underlie this high associated mortality are poorly understood and that there is much room for further study and improvement in the care of this subset of patients. We thank Anne Ober for editorial assistance. REFERENCES 1. Hingorani A, Ascher E, Lorenson E, DePippo P, Salles-Cunha S, Scheinman M, et al. Upper extremity deep venous thrombosis and its impact on morbidity and mortality rates in a hospital-based population. J Vasc Surg 1997;26:853-60. 2. Hingorani A, Ascher E, Hanson J, Scheinman M, Yorkovich W, Lorenson E, et al. Upper extremity versus lower extremity deep venous thrombosis. Am J Surg 1997;174:214-7. 3. Hingorani A, Ascher E, Ward M, Mazzariol F, Gunduz Y, Ramsey PJ, et al. Combined upper and lower extremity deep venous thrombosis. Cardiovasc Surg 2001;9:472-7.

4. Ascher E, Hingorani A, Mazzariol F, Jacob T, Yorkovich W, Gade P. Clinical experience with superior vena caval Greenfield filters. J Endovasc Surg 1999;6:365-9. 5. Ascher E, Hingorani A, Tsemekhin B, Yorkovich W, Gunduz Y. Lessons learned from a 6-year clinical experience with superior vena cava Greenfield filters. J Vasc Surg 2000;32:881-7. 6. Spence LD, Gironta MG, Malde HM, Mickolick CT, Geisinger MA, Dolmatch BL. Acute upper extremity deep venous thrombosis: safety and effectiveness of superior vena caval filters. Radiology 1999;210: 53-8. 7. Tilney NL, Griffiths HJG, Edward EA. Natural history of major venous thrombosis of the upper extremity. Arch Surg 1970;101:792-96. 8. Prescott SM, Tikoff G. Deep venous thrombosis of the upper extremity: a reappraisal. Circulation 1951;59:350-5. 9. Gloviczki P, Kazmier F, Hollier L. Axillary-subclavian venous occlusion: the morbidity of a non lethal disease. J Vasc Surg 1986;4:333-7. 10. Painter TD, Karpf M. Deep venous thrombosis of the upper extremity five years experience at a university hospital. Angiology 1984;35:743-9. 11. Knaus WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG, et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991; 100:1619-36. 12. Marinella MA, Kathula SK, Markert RJ. Spectrum of upper-extremity deep venous thrombosis in a community teaching hospital. Heart Lung 2000;29:113-7. 13. Karabay O, Yetkin U, Onol H. Upper extremity deep vein thrombosis: clinical and treatment characteristics. J Int Med Res 2004;32:429-35. 14. Sheikh MA, Topoulos AP, Deitcher SR. Isolated internal jugular vein thrombosis: risk factors and natural history. Vasc Med 2002;7: 177-9. 15. Hingorani A, Ascher E, Yorkovich W, Mazzariol F, Jacob T, Gunduz Y, et al. Upper extremity deep venous thrombosis: an underrecognized manifestation of a hypercoagulable state. Ann Vasc Surg 2000;14: 421-6.

Submitted Aug 29, 2004; accepted Dec 13, 2004.