Incidence of ipsilateral postoperative deep venous thrombosis in the amputated lower extremity of patients with peripheral obstructive arterial disease
Article Outline
Objective
Patients undergoing amputation of the lower limb due to peripheral arterial disease (PAD) are at risk of developing deep venous thrombosis (DVT). Few studies in the research literature report the incidence of DVT during the early postoperative period or the risk factors for the development of DVT in the amputation stump. This prospective study evaluated the incidence of DVT during the first 35 postoperative days in patients who had undergone amputation of the lower extremity due to PAD and its relation to comorbidities and death.
Methods
Between September 2004 and March 2006, 56 patients (29 men), with a mean age of 67.25 years, underwent 62 amputations, comprising 36 below knee amputations (BKA) and 26 above knee amputations (AKA). Echo-Doppler scanning was performed preoperatively and on postoperative days 7 and 31 (approximately). All patients received acetylsalicylic acid (100 mg daily) preoperatively and postoperatively, but none received prophylactic anticoagulation.
Results
DVT occurred in 25.8% of extremities with amputations (10 AKA and 6 BKA). The cumulative incidence in the 35-day postoperative period was 28% (Kaplan-Meier). There was a significant difference (P = .04) in the incidence of DVT between AKA (37.5%) and BKA (21.2%). Age ≥70 years (48.9% vs 16.8%, P = .021) was also a risk factor for DVT in the univariate analysis. Of the 16 cases, 14 (87.5%) were diagnosed during outpatient care. The time to discharge after amputation was averaged 6.11 days in-hospital stay (range, 1-56 days). One symptomatic nonfatal pulmonary embolism occurred in a patient already diagnosed with DVT. There was no relation between other comorbidities and DVT. The multivariate analysis showed no association between risk factors and the occurrence of DVT in the amputated extremity. DVT ipsilateral to the amputation did not influence the mortality rate (9.7%).
Conclusion
The incidence of DVT in the early postoperative period (≤35 days) was elevated principally in patients aged ≥70 years and for AKA. Patients with PAD who have recently undergone major amputations should be considered at high risk for DVT, even after hospital discharge. Given the high rate of postoperative DVT observed in this study, we now recommend prophylactic anticoagulation for these patients, but further study is needed to determine the optimal duration and efficacy of this treatment.
Although many studies have considered venous thromboembolism (VTE) in clinical and surgical patients, few have focused on patients who have undergone amputation of the lower extremities because of peripheral arterial disease (PAD). To our knowledge, no prospective research studies with a sufficiently large amputee patient sample have considered the incidence of deep venous thrombosis (DVT) in the early postoperative period. Data from existing studies are divergent, and the incidence of VTE is highly variable, ranging from 0% to 66.66%.1, 2 Amputation involves considerable surgical trauma, including bone and soft tissue dissection and division, and many surgeons do not administer pharmacologic DVT prophylaxis because of concerns about bleeding complications.
Studies of patients undergoing hip and knee prosthesis procedures show a high incidence of DVT, not only in the early postoperative period but also up to 30 days after surgery.3, 4 Studies involving amputation of ischemic extremities have been limited to the hospitalization period and provide no data on the occurrence of DVT after discharge.5, 6, 7, 8, 9, 10
In the present prospective study, we included patients with PAD who underwent major amputation of the lower extremity, and aimed to evaluate (1) the postoperative incidence of DVT in the amputated extremity during hospitalization and after discharge, (2) the relation between comorbidities and DVT incidence, and (3) the relation of DVT to symptomatic pulmonary embolism (PE) and operative mortality.
Methods
A prospective and descriptive study of patients admitted consecutively to the Emergency Department and the Vascular Surgery Ward of the Clinics Hospital of the Faculty of Medicine of São Paulo University, Brazil, was conducted from September 2004 through March 2006. The study was reviewed and approved by the local Research Ethics Committee.
The following variables were studied: age, sex, heart disease, smoking, bedridden status, female hormone therapy, obesity (body mass index >30 kg/m2), systemic arterial hypertension, diabetes mellitus, chronic renal failure, varicose veins of the lower extremities, paresthesia or paraplegia of the limbs, malignancy, level of arterial occlusion, secondary infection of the ischemic extremity, and recent arterial revascularization surgery. We analyzed the association between the incidence of VTE and the level of amputation, general medical postoperative complications, amputation stump complications, and postoperative mortality.
The accuracy of each patient's history and clinical examination was tested, comparing the data with color-flow Doppler findings, and checking clinical examination sensitivity and specificity.
Duplex ultrasound (DUS) scanning was chosen as the diagnostic method for DVT due to its noninvasive characteristic and ability to provide reproducible results in repeated examinations,11 as well as its high sensitivity in examining the proximal deep venous system of the lower extremity.12, 13, 14 Patients underwent DUS imaging once before amputation and twice after surgery, between postoperative days 3 and 15, and between days 16 and 35.
All examinations were performed using a Logiq 5 ultrasound system (GE Healthcare, Milwaukee, Wisc). The superficial femoral veins and common veins were examined, as well as the popliteal vein and saphenofemoral junction, for both lower extremities.15 B-mode was used to detect intraluminal thrombi and vein compressibility, color mode was used to evaluate flow, and Doppler spectral mode was used for evaluation of the curve pattern. The examination included the iliac veins whenever alteration in respiratory flow phasicity was present in the common femoral vein. The examination result was considered positive for DVT when thrombus was identified and venous flow absent.12, 13, 16
All patients received acetylsalicylic acid (100 mg, daily) preoperatively and postoperatively, but none received prophylactic anticoagulation. Patients underwent stimulation for early mobilization combined with motor and respiratory physical therapy, and were referred to an outpatient rehabilitation center at discharge. Patients diagnosed with DVT were treated with nonfractionated heparin or low-molecular-weight heparin, followed by warfarin sodium for a period of 6 months.14, 17, 18
All patients investigated in the study presented with chronic arterial obstruction of the lower extremities, with or without previous arterial revascularization, and were indicated for above knee amputation (AKA) or below knee amputation (BKA). Patients who received any type of anticoagulation therapy before surgery or who were unwilling to participate in the study were not included. The treatment regimen in patients excluded due to previous anticoagulation was not discontinued.
Patients were excluded from the study if (1) DVT was identified during the preoperative or intraoperative period (all divided veins were assessed for the presence of intraluminal thrombi and the findings were reported by the surgeon), (2) the postoperative DUS scan was not performed for any reason, (3) DVT was diagnosed only in the contralateral extremity during the postoperative period, and (4) postoperative anticoagulation was administered for any reason (treatment or prophylaxis).
Statistical analysis
The cumulative incidence of DVT of the stump and mortality were determined by the Kaplan-Meier method, with an accepted standard error of up to 0.1. Univariate analysis of the risk factors for DVT was obtained by the log-rank test method, and Cox regression was applied for multivariate analysis. Statistical significance was set at P < .05.
Results
During the study period, 82 patients with PAD underwent 88 amputations. As summarized in Table I, 26 patients were excluded from the study; thus, 62 amputations were studied in 56 patients, three of which were bilateral amputations. In three other patients, a new proximal amputation was performed 30 days after the first surgery. The mean age of the patients was 67.25 ± 11.7 years (range, 43-88 years), and 51.7% were men. The prevalence of associated diseases is listed in Table II. Of the 56 patients studied, 37 (66.0 %) were unable to walk before amputation, and 12 (21.4%) were smokers.
Table I. Patients excluded from the study and causes (n = 26)
| Exclusion criteria | No. |
|---|---|
| Pre-op presence of DVT | 6 |
| Intra-op diagnosis of DVT | 2 |
| Contralateral DVT in post-op period | 2 |
| No post-op echo-Doppler (due to lack of conditions/death) | 9 |
| Post-op anticoagulant treatment | 4 |
| Complications of amputation stump with hip disarticulation | 1 |
| Lost to follow-up | 2 |
Table II. Prevalence of associated diseases (n = 60)
| Disease | No. (%) |
|---|---|
| Systemic arterial hypertension | 48 (80.0) |
| Diabetes mellitus | 42(70.0) |
| Heart disease | 16(26.7) |
| Varicose veins of lower limbs | 14(23.3) |
| Cerebral vascular accident | 13(21.7) |
| Obesity | 7(11.7) |
Six patients had prior amputations of contralateral limbs (4 BKA and 2 AKA), and two patients underwent bilateral BKA in the same surgery. In all cases the primary indication for amputation was critical limb ischemia associated with tissue necrosis. In 76.7% there was also an associated local infection. Eight patients (12.9%) had undergone an arterial revascularization in the previous 30-day period.
The predominant location of obstructive arterial disease in the amputated limb was the femoropopliteal segment (71%). The prevalence of both iliofemoral and infrapopliteal disease was 14.5%. The most frequent level of amputation was BKA, in 36 (58.1%) cases. The left lower extremity was the site of amputation in 51.6%. None of the patients underwent knee disarticulation.
Mortality in the 30-day period was 9.7%. Most deaths had well-established causes, except for one patient in which autopsy presented no specific diagnosis (no evidence of PE).
Medical complications occurred in 20.0% of the patients as follows: bronchopneumonia in 11.7%, peritonitis and acute myocardial infarction in 3.3% each, and sepsis in 1.7%. One patient (1.6%) presented with a symptomatic nonfatal PE. Amputation stump complications occurred in 45.3% of the amputations. In five cases of BKA, an early (≤30 days) more proximal amputation was necessary.
During recovery, 16 cases of DVT were diagnosed in the ipsilateral amputated extremity (Table III), 12 within the first 15 days (75%) and four after this period. Of the 16 cases, 14 (87.5%) were diagnosed during outpatient care, the average time to discharge after amputation was 6.11 days (range, 1-56 days). No iliac DVT was diagnosed, and none of the patients who underwent bilateral amputation had evidence of DVT. DVT also occurred in the contralateral limb in three patients. The cumulative incidence of DVT ipsilateral to the amputation was 24.4% ≤30 days and 28.0% ≤35 days, with respective standard errors of 0.057 and 0.064 (Fig 1).
Table III. Diseased venous segment and level of amputation
| Venous segment | Amputation level | Total | |
|---|---|---|---|
| Above knee | Below knee | ||
| Femoral | 10 | 4 | 14 |
| Popliteal | 0 | 1 | 1 |
| Femoropopliteal | 0 | 1 | 1 |
| Total | 10 | 6 | 16 |
Fig 1.
Cumulative incidence of venous thrombosis of the amputation stump by the Kaplan-Meier method. The arrowhead indicates the average day of discharge after amputation.6, 11
Clinical evaluation of the signs and symptoms of DVT in the amputated limb had sensitivity of 56.2%, specificity 82.6%, negative-predictive value of 84.4%, and a positive-predictive value of 52.9%.
Univariate analysis (Table IV) demonstrated a significant difference in the incidence of DVT ipsilateral to the amputation when different levels were compared. AKA presented a cumulative incidence of venous thrombosis of 37.5%, compared with 21.2% in patients who underwent BKA, within the period up to day 35 after surgery (P = .04; Fig 2).
Table IV. Univariate analysis of risk factors for deep venous thrombosis of the amputation stump (log rank)
| Factor | No. | DVT per 100 individuals/d | Rate of events | P |
|---|---|---|---|---|
| Sex | ||||
| Male | 31 | 1.506 | 1 | .13 |
| Female | 31 | 0.725 | 0.48 | |
| Arterial hypertension | ||||
| Yes | 50 | 1.063 | 0.95 | .89 |
| No | 12 | 1.115 | 1 | |
| Diabetes mellitus | ||||
| Yes | 44 | 0.810 | 0.44 | .09 |
| No | 18 | 1.837 | 1 | |
| Smokers | ||||
| Yes | 13 | 0.929 | 0.84 | .69 |
| No | 49 | 1.112 | 1 | |
| Bedridden | ||||
| Yes | 39 | 0.819 | 0.53 | .20 |
| No | 23 | 1.557 | 0 | |
| Amputation level | ||||
| Thigh | 26 | 1.792 | 1 | .044 |
| Leg | 36 | 0.643 | 0.36 | |
| Age, y | ||||
| ≤69 | 36 | 0.590 | 0.28 | .012 |
| ≥70 | 26 | 2.105 | 1 | |
| Side | ||||
| Left | 32 | 0.960 | 0.81 | .55 |
| Right | 30 | 1.180 | 1 | |
| Obesity | ||||
| Yes | 8 | 0.885 | 0.62 | .66 |
| No | 54 | 1.419 | 1 | |
| Ischemia/limb infection | ||||
| Yes | 48 | 1.025 | 0.83 | .76 |
| No | 14 | 1.242 | 1 | |
| Reamputation | ||||
| Yes | 5 | 2.273 | 1 | .66 |
| No | 57 | 1.036 | 0.46 | |
| Contralateral amputation | ||||
| Yes | 6 | 1.460 | 1 | .67 |
| No | 56 | 1.033 | 0.71 | |
| Stroke | ||||
| Yes | 14 | 1.163 | 1 | .94 |
| No | 48 | 1.045 | 0.90 | |
| Heart disease | ||||
| Yes | 17 | 1.626 | 1 | .37 |
| No | 45 | 0.890 | 0.55 | |
| Revascularization | ||||
| Yes | 8 | 1.031 | 0.96 | .99 |
| No | 54 | 1.078 | 1 | |
| Stump complications | ||||
| Yes | 27 | 1.250 | 1 | .67 |
| No | 35 | 0.939 | 0.75 | |
| Clinical complications | ||||
| Yes | 11 | 0.725 | 0.63 | .44 |
| No | 51 | 1.151 | 1 | |
| Death | ||||
| Yes | 6 | 3.278 | 1 | .20 |
| No | 56 | 0.978 | 0.3 | |
| Varicose veins | ||||
| Yes | 15 | 1.312 | 1 | .72 |
| No | 47 | 0.990 | 0.75 |
Fig 2.
Cumulative incidence by long-rank analysis of venous thrombosis of amputation stump divided by amputation level above knee (top line) and below knee (bottom line). The arrowhead indicates the average day of discharge after amputation.6, 11
Patients aged ≥70 years had significantly higher risk for DVT in the amputated limb than younger patients within the 35-day period (48.9% vs 16.8%, respectively; P = .012; Fig 3). The multivariate analysis showed no association between risk factors and the occurrence of DVT in the amputated extremity.
Fig 3.
Cumulative incidence by log-rank analysis of venous thrombosis of the amputation stump divided by age ≥70 years (top line) and age <70 (bottom line). The arrowhead indicates the average day of discharge after amputation.6, 11
Discussion
Analysis of the general patient characteristics revealed no difference in the indications for amputation and complications in the amputated extremity with regard to previous related data.19, 20, 21, 22, 23 The most frequent site of chronic arterial occlusion in this investigation was the superficial femoral artery, in accordance with Roon et al.24 The percentage of patients who had undergone recent prior revascularization of the amputated limb was within the limits found in the literature.25, 26, 27, 28 We also observed that the concomitant bilateral amputations were in accordance with other reports that stated values of 10.68% to 15%.23, 29
Previous studies30, 31 reported that failed BKA occurs in approximately 15% of patients, similar to the incidence observed in the present study. All new amputations to a more proximal level were performed in the BKA group. BKA failure was probably related to the surgery being performed distal to a femoropopliteal obstruction (absence of a palpable popliteal pulse) and close to an infection source. Most patients had combined ischemia and local infection preoperatively; these are contributing factors for infection as the major cause of BKA complication (54.5%). The occurrence of these complications and new amputation to a more proximal level did not influence the incidence of DVT.
Previous studies report that the incidence of nonfatal symptomatic PE varies from 2.1% to 8%.23, 24 We believe that the low incidence of symptomatic PE in the present study was due to the diagnostic accuracy of the DUS series, which resulted in immediate treatment of DVT and a consequent decrease in the incidence of PE. Another possible reason was our choice not to investigate patients without pulmonary symptoms. When diagnostic examinations were performed for routine asymptomatic PE in previous studies, results varied from 11% to 14.28%.32, 33 Although the study by Williams et al34 attributed a PE risk four times higher for amputations at the AKA level, we had no cases of PE for AKA amputation in our study and only one PE in a BKA.
Previous reports described perioperative mortality for major amputation varying from 0% to 37%.24, 35, 36, 37 We also observed a mortality rate within this range. Univariate statistical analysis revealed no relationship between DVT and death. Nearly all DVT was diagnosed in the first 15 days after surgery, in agreement with previous studies, independent of sample and methodology used.5, 6, 7, 8, 9, 10
The patients who required amputation were not solely at risk in the early postoperative period, they were also at risk after hospital discharge. In this study, 87.5% of the DVT diagnoses were made at the scheduled outpatient return appointment for the postoperative DUS examination. This finding differs from that of others studies because most researchers have studied the incidence of DVT only during the hospitalization.1, 2, 5, 6, 7
A relationship was found between the type of amputation performed and the incidence of DVT. The high incidence of DVT in patients who underwent AKA is probably because the femoral vein is a conduction vessel with few tributaries, a feature that leads to poor venous flow in its remnant segments and consequent thrombosis. Old age (≥70 years) was another factor related to DVT ipsilateral to the amputation, but no difference was observed in the distribution of AKA and BKA, or bedridden individuals.
The presence of the other studied risk factors and variables showed no relation to DVT ipsilateral to the amputation, even after multivariate analysis; this finding is in contrast to data from previous studies for clinical and surgical patients in general.38, 39, 40, 41, 42, 43, 44, 45, 46, 47 This discrepancy may reflect the relatively small number of patients in the present study.
Yager et al5 demonstrated DVT risk factors to be chronic venous insufficiency or previous history of DVT, or both, and the presence of prior amputation of the contralateral extremity. Their analysis also included patients diagnosed in the preoperative period (6 of 9, 66.7%) and in the contralateral limb (4 of 9, 44.4%), and was conducted using a methodology distinct from that used in our study, in which we only considered postoperative cases of DVT in the amputated extremity stump.
Conclusion
The incidence of DVT ipsilateral to the amputation is elevated during the early postoperative period (35 days), mainly in AKA and at ages of ≥70 years. There was no relationship between the presence of the other comorbidities studied and postoperative DVT occurrence. All patients with PAD scheduled to undergo major amputation should be considered at high risk for the development of DVT, even during the period after discharge from the hospital. On the basis of these results, we recommend prophylactic anticoagulation (if not contraindicated) and surveillance to all patients undergoing this type of procedure. Further studies are required to determine the optimal method and duration for prophylaxis treatment.
Author contributions
References
- . Prospective screening for deep vein thrombosis in high risk patients. Am J Surg. 1977;134:187–190
- . Prophylaxis in iliofemoral venous thrombosis: the major amputee as a clinical research model. Br J Surg. 1973;60:831
- Extended out-of-hospital low-molecular-weight heparin prophylaxis against deep venous thrombosis in patients after elective hip arthroplasty: a systematic review. Ann Int Med. 2001;135:858–869
- . Short-durantion prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of prospective studies investigation symptomatic outcomes. Arch Intern Med. 2002;162:1465–1471
- . Deep vein thrombosis associated with lower extremity amputation. J Vasc Surg. 1995;22:612–615
- . Incidence of deep vein thrombosis following vascular surgery. Int Angiol. 1997;16:65–69
- . Deep venous thrombosis and pulmonary embolism in bilateral lower extremity amputee patients. Arch Phys Med Rehabil. 1999;80:509–511
- . Deep vein thrombosis after lower limb amputation. Am J Phys Med Rehabil. 2000;79:145–149
- . High resolution real-time ultrasound for the diagnosis of venous thrombosis in the rehabilitation setting. Am J Phys Med rehabil. 1991;70:3–4
- . Deep venous thrombosis as a cause of stump swelling in two lower extremity amputee patients. Arch Phys Med Rehabil. 1993;74:1002–1003
- . Recent clinical trials in the diagnosis of deep vein thrombosis. Curr Opin Hematol. 1999;6:275–279
- Color-flow duplex scanning for the surveillance and diagnosis of acute deep venous thrombosis. J Vasc Surg. 1992;15:366–376
- . Ultrasonography and diagnosis of venous thromboembolism. Circulation. 2004;109(12 suppl 1):I9–I14
- Triplex US in the diagnosis of asymptomatic deep venous thrombosis. Acta Radiol. 1997;38:327–331
- . International interdisciplinary Consensus Committee on Venous Anatomical Terminology (Nomenclature of the veins of the lower limbs: an international interdisciplinary consensus statement). J Vasc Surg. 2002;36:416–422
- . Distribution of thrombi in acute lower extremity deep venous thrombosis: implications for sonography and CT and MR venography. AJR Am J Roentgenol. 2000;175:1299–1301
- . A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. N Engl J Med. 1995;332:1661–1665
- . Predictors of recurrente after deep vein thrombosis and pulmonary embolism. Arch Inter Med. 2000;160:761–768
- . Risk factors in healing of below-knee amputation (Appraisal of 64 amputations with vascular disease). Am J Surg. 1981;141:718–720
- . Rehabilitation outcome 5 years after 100 lower-limb amputations. Br J Surg. 1994;81:1596–1599
- Results of lower extremity amputations in patients with end-stage renal disease. J Vasc Surg. 1994;20:14–19
- Cardiac morbidity and operative mortality following lower-extremity amputation: the significance of multiple eagle criteria. Ann Vasc Surg. 1999;13:204–208
- Functional outcome in a contemporany series of major lower extremity amputations. J Vasc Surg. 2003;38:7–14
- . Below-Knee amputation: a modern approach. Am J Surg. 1977;134:153–158
- . Amputation level following unsuccessful distal limb salvage operations. Surgery. 1980;87:683–687
- . Morbid implications of above-knee amputations. Arch Surg. 1980;115:165–167
- . Late outcomes of limb loss after failed infrainguinal bypass. J Vasc Surg. 1995;21:613–622
- . Do patients with critical limb ischaemia undergo multiple amputations after infrainguinal bypass surgery?. Eur J Vasc Endovasc Surg. 2001;21:427–431
- . The role of vascular laboratory criteria in the selection of patients for lower extremity amputation. Ann Surg. 1982;195:469–473
- Prospective study of 713 below-knee amputations for ischaemia and the effect of a prostacyclin analogue on healing. Br J Surg. 1994;81:33–37
- . Surgical results of leg amputation according to Ghormley's technique in the treatment of chronic lower limb ischaemia. Acta Chir Belg. 2006;106:52–54
- . Pulmonary embolism associated with surgically proved deep venous thrombosis. Am J Surg. 1975;129:500–502
- . Pulmonar embolism after amputation of the lower extremity. Surg Gynecol Obstet. 1975;140:246–248
- Failure of low dose heparin to prevent pulmonary embolism after hip surgery or above the knee amputation. Ann Surg. 1978;188:468–474
- . Preoperative risk factors for 30-day mortality after elective surgery of vascular disease in Department of Veterans Affairs hospitals: is race important?. J Vasc Surg. 2001;34:634–640
- Perioperative cardiovascular risk stratification of patients with diabetes who undergo elective major vascular surgery. J Vasc Surg. 2002;35:894–901
- . Relative risks of limb revascularization and amputation in the modern era. Cardiovasc Surg. 1994;2:754–759
- Terapeutic and clinical course of deep vein thrombosis. Am J Surg. 1983;146:581–585
- Prevalence and distribution of calf vein thrombosis in patients with symptomatic deep venous thrombosis: a color-flow duplex study. J Vasc Surg. 1996;24:738–744
- . Extent of thrombus evaluated in patients with recurrent and first vein thrombosis. J Vasc Surg. 2002;36:357–360
- . Incidence of diagnosed deep vein thrombosis in the general population: systematic review. Eur J Vasc Endovasc Surg. 2003;25:1–5
- . Deep vein thrombosis of leg: is there a “high risk” group?. Am J Surg. 1970;120:527–530
- . Recurrent venous thromboembolism after deep vein thrombosis: incidence and risk factors. Arch Intern Med. 2000;160:769–774
- Comparison of 3 and 6 months of oral anticoagulant therapy after a first episode of proximal deep vein thrombosis or pulmonary embolism and comparison at 6 and 12 weeks of therapy after isolated calf deep calf vein thrombosis. Circulation. 2001;103:2453–2460
- Relative impact of risk factors for deep vein thrombosis and pulmonary embolism: a population-based study. Arch Intern Med. 2002;162:1245–1248
- . Early outcome after isolated calf vein thrombosis. J Vasc Surg. 1997;26:749–756
- . Risk factors for deep vein thrombosis and pulmonary embolism: a population-based case-control study. Arch Inter. Med. 2000;160:809–815
Competition of interest: none.
PII: S0741-5214(08)01193-2
doi:10.1016/j.jvs.2008.07.055
© 2008 Published by Elsevier Inc.
