QJM Advance Access originally published online on January 17, 2005
QJM 2005 98(2):139-145; doi:10.1093/qjmed/hci020
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
QJM vol. 98 no. 2 © Association of Physicians 2005; all rights reserved.
Lower-limb deep-vein thrombosis in a general hospital: risk factors, outcomes and the contribution of intravenous drug use
From the Tropical and Infectious Disease Unit, Royal Liverpool University Hospital, Liverpool, UK
Received 27 May 2004 and in revised form 1 November 2004
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
Background: Much of the morbidity associated with deep-vein thrombosis (DVT) is absent from the evidence base used to define best clinical practice. Intravenous (IV) drug use is an increasingly frequent cause of DVT.
Aim: To obtain a profile of DVT patients in routine clinical care, and determine the contribution of IV drug use.
Design: Retrospective case-note review.
Methods: We reviewed 232 episodes of lower-limb DVT in a large district general hospital during 1996.
Results: Patients had mean (range) age 62.8 (21–97) years, with 43.9% aged > 70 years. A large proportion would have been excluded from prospective studies that have contributed to current DVT guidelines. Risk factors included smoking (33.0%), immobility (26.5%), previous DVT (23.6%), surgery in the last 3 months (18.2%), malignancy (16.5%), varicose veins (10.5%) and IV drug use (6.9%). Forty-five (19.4%) had multiple risk factors. Postsurgical DVT commonly presented from the community following initial hospital discharge. Intravenous drug use accounted for 48.4% of episodes in patients aged 
40 years. Thrombosis was right-sided in 68.8% of IV drug users, compared to 38.2% in others (p = 0.034).
Discussion: Our patients differed from those in most of the prospective studies used to develop routine clinical care pathways for DVT. Intravenous drug use is an important cause of community-acquired DVT in young adults.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Lower-limb deep-vein thrombosis (DVT) is common. Prospective studies involving rigorously selected patient groups have played a major role in defining the natural history and providing the evidence base for optimum management strategies.1–6 However, generalizability of these studies to typical hospital patients with DVT is uncertain, as they excluded substantial numbers of patients, with different inclusion criteria and risk factors for DVT in each study. Several of these risk factors carry significant co-morbidity, such as increased age, heart failure, stroke and in particular malignancy, the presence of which increased risk of death eight-fold after DVT in one 8-year follow-up study.6
In the early 1990s, we noticed that a large number of patients admitted to our unit were intravenous drug users (IDUs), and that a significant number had DVT or arterial injury.7 There are few systematic data on the numbers of IDUs among DVT patients, while diagnosis and immediate and long-term clinical management in these patients are difficult and poorly defined. The intravenous use of recreational drugs is associated with considerable morbidity, a significant proportion of which may be from the drug itself,8 transmission of blood-borne disease9 and localized complications of injecting habits.10
The aim of this study was to obtain a profile of DVT patients in a routine district general hospital setting, examining risk factors, presentation, management and outcome of a retrospective-identified patient group diagnosed with lower-limb DVT over a year. We also hoped to determine the overall contribution of IV drug use to lower-limb DVT and to highlight problems in management of IDU. Difficulties in extracting records of DVT patients from our hospital system were also identified. At the time of this study, patients with both proximal and distal DVT were managed as in-patients.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The medical records of all patients with lower-limb DVT discharged from University Hospital Aintree in 1996 were reviewed. To maximize patient ascertainment, three sets of patient lists were used: hospital computer records of discharge diagnoses, radiology departmental lists of venograms and Doppler ultrasound scans, and accident and emergency admission lists for possible DVT. Episodes were excluded if they were direct presentations of pulmonary embolism (PE, 26 episodes), thrombophlebitis (22 episodes), non-limb venous thrombosis (10 episodes) or upper-limb DVT (5 episodes). Episodes were not considered discrete if the patient was readmitted within 7 days of discharge.
A pre-piloted proforma was used. Data extracted from case records included patient demographics and risk factors, clinical presentation, use of diagnostic procedures, management, outcome, complications and recurrence at 6 months. Patient premorbidity on admission was scored using the Karnofsky Performance Index.11 Development of PE was accepted if recorded as such. Cause of death was recorded as DVT-related when resulting from PE, bleeding during anticoagulation or extension and associated sepsis.
Selected studies on diagnostic management,1 comparison of IV heparin with low-molecular-weight heparin (LMWH),2,3 use of TED stockings,4 duration of warfarin therapy5 and long-term clinical course6 were compared for their patient exclusion profiles. A particular focus was whether IDUs were included, and we contacted the authors of each published study specifically to ask about this.
Statistical analysis was conducted with SPSS 10.0 for Windows. The 
2 test was used to test differences between nominal data, Student's t-test for scaled data and Mann-Whitney U-test for ordinal data. Spearman's rho was used for correlations. Significance was taken at a 2-tailed p value of 0.05.
This study was initiated and performed with the assistance of the clinical audit department. In accordance with Trust guidelines at the time, ethical approval was not required for a retrospective case-note audit study.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
The hospital computer administration system identified 266 episodes with thrombosis-related discharge diagnoses, of which 184 episodes were lower-limb DVT. A further 48 episodes were identified from the additional sources (Figure 1), 28 (58.3%) of which were hospital-acquired DVT. In total, 232 episodes in 223 patients were studied. The diagnosis of DVT was confirmed in all but seven cases by venography, Doppler ultrasound scanning or autopsy; these seven were diagnosed on clinical criteria alone.
|
Most DVTs were proximal, and proximal DVT was more common with IDUs and malignancy. There was no difference in distribution between community and hospital acquired cases (Table 1). There was a significant difference in laterality between IDUs and non-IDUs: 11/16 (68.8%) vs. 71/186 (38.2%), respectively, had the right leg affected (p = 0.034).
|
Patient characteristics and risk factors are summarized in Table 2. Male:female ratio was 1:1.4, with no difference in mean age between males (61.8 years) and females (63.3 years). There was a bimodal pattern of ages, with peaks seen in the age groups 26–30 and 71–75 years; 98/223 (43.9%) of patients were aged >70 years. Median Karnofsky Performance Index score at presentation was 7 (range 2–10), reflecting ‘an inability to carry out normal activity or to do active work', and scores inversely correlated with age (correlation coefficient –0.549, p < 0.0005). Significant differences in risk factor prevalence between community and hospital presentations are highlighted in Table 2, although there was no significant difference in the overall number of risk factors (Figure 2). Multiple risk factors were present in 45 (19.4%) patients. For hospital-acquired cases, median (range) time from admission to presentation was 15.5 (1–369) days. Trauma (21.9%), stroke (17.2%) and elective surgery (12.5%) were the most common reasons for admission.
|
|
Comparison with selected prospective studies1–6 on clinical course and management (Table 3) demonstrates an under-representation in those studies of patients with previous DVT history, cancer, varicose veins, pregnancy and IV drug use, necessitated largely through study exclusion criteria, while other quoted exclusion criteria (e.g. short life expectancy, bilateral or IVC thromboses, very recent surgery or anaesthesia, bleeding diatheses or nidus, leg paresis, arterial insufficiency or venous ulceration, unavailability for follow-up) would have excluded significant additional numbers.
|
History of previous DVT was more common in IDUs (46.7%) vs. non-IDUs (21.5%) (p = 0.027) and in patients with thrombophilia (5/6). A history of surgery in the preceding 3 months was found in 40/220 (18.2%) patients, of whom 19 (47.5%) presented following discharge from the operation, constituting 19/168 (11.3%) of community acquired DVTs. A malignancy was present in 36/212 (17.0%), and in these patients multiple risk factors were common, including mean age of 66.8 years, recent surgery in 11/34 (32%), other immobility 7/34 (20.5%), low cardiac output states 5/35 (14.3%) and previous DVT 4/32 (12.5%).
In-patient mortality in our DVT patients was 3.9% (9/232) from DVT-related causes and 6.0% (14/232) from DVT-unrelated causes, while one cause of death was undetermined. Excluding IDUs, the mean (range) age of patients with DVT-unrelated deaths was significantly higher than for those who survived, but the age difference between DVT-related death and survival was not significant (74.0 (50–89) and 73.7 (51–95) vs. 64.1 (25–97) years, p = 0.046 and 0.085, respectively). Risk factors associations with mortality are listed in Table 4, of which malignancy was the most significant.
|
The DVT progressed to PE in 15/227 (6.6%), representing 12/164 (7.3%) of community- and 3/63 (4.8%) of hospital-acquired DVT, respectively (p = 0.693). Mortality was 8/15 (53.3%), directly from PE in 5/15 and from DVT-unrelated causes in 3/15, while PE was significantly associated with both DVT-related (5/9, p < 0.0005) and unrelated mortality (3/14, p = 0.001) compared to survival (7/206). Karnofsky Performance Index (median 8, range 2–10) was not significantly different from non-PE patients (p = 0.718).
There were two extensions to the inferior vena cava and 12 to the iliac veins. Post-thrombotic syndrome (PTS) developed within 6 months in three patients, although a further three presented with existing leg ulcerations and a history of DVT. Bleeds during in-patient anticoagulation were recorded for 17/211 (8.1%) episodes, of which 3/17 (17.6%) were fatal. Bleeds during home anticoagulation were recorded for 4/187 (2.1%) episodes; none was fatal. DVT recurrence was seen in 19/209 (9.1%) surviving episodes at 6 months, and was associated with IDU (6/19, 31.6%, p < 0.0005) and venous trauma in non-injecting, non-surgical patients (5/18, p = 0.005); no association with malignancy or thrombophilia was demonstrated.
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Our efforts to identify DVT cases and obtain case notes were extensive. Problems encountered were miscoded hospital computer records (4.9% of episodes) and the confusion or combination of other thrombosis-related diagnoses with DVT (21.8%). The hospital activity index identified only 79.3% of confirmed episodes, as determined by including accident and emergency and radiology indexes. A significant number of missed episodes were hospital-acquired, which were frequently complicated with an array of medical problems, giving rise to coding difficulty. We can speculate over the significance of this finding with respect to other similar studies on DVT, which may have missed these more morbid patients in whom management complications are also more likely. We recommend that future studies on DVT where cases are identified retrospectively use multiple sources. Radiology departmental records in particular identified a large proportion of the missed cases.
There is a marked difference in the characteristics of this group of patients compared to those used in the prospective studies analysed, indicating the patients used in the evidence-base for formulating DVT management are not representative of those seen in clinical practice. The effect of this can be seen in the difference in overall in-patient mortality rate between this study (9.9%) and randomized controlled trials of DVT treatment (
4%),12 with a greater number dying of DVT-unrelated causes (6.0%). This is particularly pertinent to the need for subgroup-orientated therapy.13
There was a high proportion of IDUs in our DVT patients. The association of IV drug misuse with DVT is well recognized.7,14 Their representation among hospital DVT patients is less well appreciated, but accounted for 9.6% of all community-acquired DVT and 48.4% of patients aged <40 years in our study. This is similar to a study of women with DVT in Glasgow, where 21% of all DVTs and 52% of those aged <40 years were IDUs.14 The association of IDU with right-sided DVT has not been described before. Previous studies in non-IDUs have shown proximal DVT to occur more often on the left, with little difference in laterality of distal DVT15 IDUs have been universally excluded from the prospective studies we reviewed (personal communications),1,3–6,16,17 owing largely to their erratic behaviour. They are often not mentioned in general DVT reviews.18
Deep-vein thrombosis is one of a number of lower-limb complications of groin injecting10 and may arise or be complicated by coincident sepsis.19,20 With the high rate of recurrence of 31.6% at 6 months, there is increased risk of PTS. Evaluation is additionally complicated by soft-tissue pathology and venous access difficulties, which would hamper diagnostic ultrasonography and venography. Further work is required to define the clinical course, natural history and optimum approach to investigation and treatment of swollen limbs21 and DVT in this group, especially as a subsequent study in our hospital has confirmed the importance of IDUs over a larger period of time, with high rates of recurrence, irrespective of the type of treatment offered in hospital.22
Malignancy was present in 17% of patients, and was associated with multiple coincident DVT risk factors. It was the most significant risk factor associated with mortality and, perhaps consequently, was not associated with recurrence.
A significant number of patients were postsurgical, and 46.3% of these presented with DVT from the community following discharge. Risk of DVT may persist for a considerable time following discharge from surgery.23 The advantage of continuing prophylactic anticoagulation upon discharge to the community with LMWH has been suggested before.24
Approximately 10% of proximal DVT embolized (86.7% of PE). Mortality from PE was 53.3%, with a third dying from the PE and a fifth from other causes. This is much higher than the 15–17.5% mortality rate quoted for direct presentations of PE,25 and perhaps indicates the significant morbidity of patients who present with embolization despite adequate anticoagulation therapy.
This study describes DVT as a condition associated with considerable morbidity, much of which is absent from the evidence base used to define best clinical practice management. Intravenous drug use is an important cause of community-acquired DVT in young adults.
|
Acknowledgments |
|---|
We thank our colleagues for permission to review case notes of patients admitted under their care, Keith Jones for assistance with hospital activity indexes, Sandra Corfe and her colleagues in the clinical audit department for assistance with study design and case note retrieval, and Tracey Walker for her patient assistance in retrieving radiology reports.
|
Footnotes |
|---|
Address correspondence to Dr N.J. Beeching, Tropical and Infectious Disease Unit, Royal Liverpool University Hospital, Prescot Street, Liverpool L7 8XP. e-mail: nicholas.beeching{at}rlbuht.nhs.uk
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. Heijboer H, Buller HR, Lensing AW, Turpie AG, Colly LP, ten Cate JW. A comparison of real-time compression ultrasonography with impedance plethysmography for the diagnosis of deep-vein thrombosis in symptomatic outpatients. N Engl J Med 1993; 329:1365–9.
2. Luomanmaki K, Grankvist S, Hallert C, Jauro I, Ketola K, Kim HC, Kiviniemi H, Koskivirta H, Sorskog L, Vilkko P. A multicentre comparison of once-daily subcutaneous dalteparin (low molecular weight heparin) and continuous intravenous heparin in the treatment of deep vein thrombosis. J Intern Med 1996; 240:85–92.[CrossRef][Web of Science][Medline]
3. The Columbus Investigators. Low-molecular-weight heparin in the treatment of patients with venous thromboembolism. N Engl J Med 1997; 337:657–62.
4. Brandjes DP, Buller HR, Heijboer H, Huisman MV, de Rijk M, Jagt H, ten Cate JW. Randomised trial of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet 1997; 349:759–62.[CrossRef][Web of Science][Medline]
5. Schulman S, Rhedin A-S, Lindmarker P, Carlsson A, Larfars G, Nicol P, Loogna E, Svensson E, Ljungberg B, Walters H, Viering S, Nordlander S, Leijd B, Jonsson K-A, Hjorth M, Olle L, Boberg J. A comparison of six weeks with six months of oral anticoagulation therapy after a first episode of venous thromboembolism. N Engl J Med 1995; 332:1661–5.
6. Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, Cattelan AM, Polistena P, Bernardi E, Prins MH. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996; 125:1–7.
7. Fox R, Beeching NJ, Morrison C, Ruben S, Garvey T. Misuse of temazepam. Br Med J 1992; 305:253.
8. Killam AL. Cardiovascular and thrombosis pathology associated with cocaine use. Hematol Oncol Clin North Am 1993; 7:1143–51.[Web of Science][Medline]
9. Lamden KH, Kennedy N, Beeching NJ, Lowe D, Morrison CL, Mallinson H, Mutton KJ, Syed Q. Hepatitis B and hepatitis C virus infections: risk factors among drug users in Northwest England. J Infection 1998; 37:260–9.[CrossRef][Web of Science][Medline]
10. Roszler MH, McCarroll KA, Donovan KR, Rashid T, Kling GA. The groin hit: complications of intravenous drug abuse. Radiographics 1989; 9:487–508.[Abstract]
11. Karnofsky DA, Buchenal JH. The clinical evaluation of chemical therapeutic agents in cancer. In: Mackad CM, ed. Evaluation of Chemotherapeutic Agents. New York, Columbia University Press, 1949.
12. Leizorovicz A, Simmonneau G, Decousus H, Boissel JP. Comparison of efficacy and safety of low molecular weight heparins and unfractionated heparin in initial treatment of deep venous thrombosis: a meta analysis. Br Med J 1994; 309:299–304.
13. Lensing AW, Prandoni P, Prins MH, Buller HR. Deep-vein thrombosis. Lancet 1999; 353:479–85.[CrossRef][Web of Science][Medline]
14. McColl MD, Tait RC, Greer IA, Walker ID. Injecting drug use is a risk factor for deep vein thrombosis in women in Glasgow. Br J Haematol 2001; 112:641–3.[CrossRef][Web of Science][Medline]
15. Ouriel K, Green RM, Greenberg RK, Clair DG. The anatomy of deep venous thrombosis of the lower extremity. J Vasc Surg 2000; 31:895–900.[CrossRef][Web of Science][Medline]
16. Lindmarker P, Schulman S, et al. The risk of ipsilateral versus contralateral recurrent deep vein thrombosis in the leg. J Intern Med 2000; 247:601–6.[Web of Science][Medline]
17. Boccalon H, Elias A, Chale JJ, Cadene A, Gabriel S. Clinical outcome and cost of hospital vs home treatment of proximal deep vein thrombosis with a low-molecular-weight heparin: the Vascular Midi-Pyrenees study. Arch Intern Med 2000; 160:1769–73.
18. Gorman WP, Davis KR, Donnelly R. ABC of arterial and venous disease. Swollen lower limb—1: general assessment and deep vein thrombosis. Br Med J 2000; 320:1453–6.
19. Kniemeyer HW, Grabitz K, Buhl R, Wust HJ, Sandmann W. Surgical treatment of septic venous thrombosis. Surgery 1995; 118:49–53.[CrossRef][Web of Science][Medline]
20. Mackenzie AR, Laing RB, Douglas JG, Greaves M, Smith CC. High prevalence of iliofemoral venous thrombosis with severe groin infection among injecting drug users in North East Scotland: successful use of low molecular weight heparin with antibiotics. Postgrad Med J 2000; 76:561–5.
21. Pieper B, Templin T. Chronic venous insufficiency in persons with a history of injection drug use. Res Nurs Health 2001; 24:423–32.[CrossRef][Web of Science][Medline]
22. Jenkins NF, Syed FF, Beadsworth MB, Nye FJ, Beeching NJ. Deep vein thrombosis in injecting drug users—common but neglected. Unpublished data, 2004.
23. Huber O, Bounaneaux H, Borst F, Rohner A. Post-operative pulmonary embolism after discharge: an underestimated risk. Arch Surg 1992; 127:310–13.
24. Kakkar VV, de Lorenzo F. Prevention of venous thromboembolism in general surgery. Baillieres Clin Haematol 1998; 11:605–19.[CrossRef][Web of Science][Medline]
25. Goldhaber SZ. Pulmonary embolism. N Engl J Med 1998; 339:93–104.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||