The clinical significance of lower extremity deep vein thrombus (DVT) propagation in the setting of anticoagulation therapy remains unclear. The purpose of this study is to compare results of thrombus outcome found with repeat duplex ultrasonography to the incidence of pulmonary embolism and mortality. During a recent 18-month period, 457 patients were diagnosed with lower extremity DVT with duplex ultrasonography and their data were retrospectively analyzed. Repeat examinations were available for review in 118 patients (51 men, 67 women). Results of repeat duplex exams were divided into 4 groups: resolved, improved, unchanged, or extended proximally. All patients received heparin and warfarin therapy. Ventilation-perfusion (V/Q) scans were obtained only for signs and symptoms of pulmonary embolism (n = 30). Mortality, the prevalence of high-probability V/ Q scans, frequency of intracaval filter insertion, gender, mean age, mean prothrombin time (PT), mean partial thromboplastin time (PTT), mean number of repeat ultrasounds per patient, and mean time over which the repeat ultrasounds took place were compared among the 4 groups. Patients who had proximal extension of DVT (19%) on repeat duplex ultrasound had an increased prevalence of pulmonary embolism (p
Introduction
Several studies have prospectively assessed deep vein thrombosis (DVT) in regard to thrombus resolution and propagation using repeat duplex ultrasound1-7 However, little is known about the clinical significance of proximal DVT extension documented by repeat duplex exam in the setting of adequate anticoagulation. Furthermore, routine monitoring of DVT can be justified in this era of cost- containment medicine only if therapy is changed as a result of such testing. The purposes of this study are to retrospectively review our experience with repeat duplex imaging of lower extremity DVT and compare results of thrombus outcome to level of anticoagulation, incidence of pulmonary embolism, and mortality.
Patients and Methods
During an 18-month period, 457 patients were diagnosed with lower extremity DVT with venous duplex ultrasonography at the Maimonides Medical Center Vascular Laboratory. Repeat duplex exams were available for review in 118 (26%) patients. There were 67 women and 51 men ranging in age from 23 to 91 years (mean 62 17 years). All patients were referred for evaluation by either the surgical or medical service. Total follow-up time with ultrasound and the length of time each patient spent in the study were calculated.
Duplex imaging studies were performed with an Acuson 128, or a Toshiba SSH-140A, or a Quantum 2000 scanner using a 5 MHz probe. Deep vein segments evaluated for thrombosis included the external iliac vein when possible, the common femoral vein, the superficial and deep femoral veins, the popliteal vein, and the calf veins. All veins were examined in the transverse and longitudinal views. All exams were performed by vascular technologists, recorded on videotape, and interpreted by one attending vascular surgeon.
Criteria for diagnosis of DVT included absence or diminished color Doppler signals with respiration or augmentation maneuvers, echogenie signals within the vein lumen either partially or completely occluding the vessel, inability to compress the vein by pressure on the transducer, and dilated veins where thrombus was suspected. Once the diagnosis of DVT was made, all vein segments involved were recorded.
Patients were stratified into 4 groups (extended, unchanged, improved, or resolved) based on the results of the repeat duplex ultrasound as compared to the original exam. Thrombus extension was defined as propagation to the next proximal vein segment. Clots that propagated only distally were not included in this group. Thrombus was determined to be unchanged if the clot remained in the same vein segment without evidence of recanalization or only propagated distally. Improvement of thrombus was recorded if there was evidence of recanalization or regression in the overall size or number of vein segments involved. A thrombus was determined to be resolved if intraluminal echogenic signals disappeared, the vein was fully compressible with pressure on the transducer, and color Doppler signals returned with augmentation maneuvers and respiration.
All patients diagnosed with acute DVT were begun on an intravenous heparin regimen and subsequently placed on a warfarin regimen after 24 hours of heparin therapy. Routine prothrombin time (PT) and partial thromboplastin time (PTT) were obtained and heparin was discontinued once the PT was 1.5 to 2.0 control levels. (International normalized ratio values were not yet the standard when these data were collected.) The interval for oral anticoagulation ranged from 3 months to the entire length of the study. Average PT and PTT were calculated from hospital laboratory records for each patient when available.
Ventilation-perfusion (V/Q) scans and chest roentgenograms were obtained only when patients presented with signs and symptoms of pulmonary embolism (PE). Briefly, with the patient in the supine position, ventilation scans were performed with inhaled aerosolized diephilene triamine pentatate acid (DTPA) and perfusion scans were obtained by intravenous injection of technetium Tc 99m macroaggregated albumin. Pulmonary ventilation and perfusion images were then created with wide-field scintillation cameras collecting a determined number of counts. Scans were interpreted to be high- probability if there were large segmentai perfusion defects without corresponding ventilation or roentgenographic abnormalities. Low- probability scans demonstrated perfusion defects with matching ventilation or roentgenographic abnormalities. Normal V/Q scans had no perfusion defects present. Intermediate scans did not meet criteria for normal, low, or high probability of PE.8 Only high- probability V/Q scans were determined to depict pulmonary emboli.
The number of Greenfield filter devices inserted was obtained from the vascular surgery operative database. Indications for insertion of an intracaval filter device were PE despite adequate anticoagulation or a contraindication to anticoagulation therapy. All procedures were done percutaneously under fluoroscopic guidance. There were no postoperative hematomas and all filters were placed in the infrarenal position.
Mortality data were obtained from the New York City Department of Vital Records. Mortality could be documented only if the death occurred in 1 of the boroughs of the city of New York.
One hundred and five patients were then randomly selected from the 341 patients who had a single duplex ultrasound exam demonstrating lower extremity DVT during the same study period. Results for average age, PT, PTT, gender, prevalence of high- probability V/Q scans, and Greenfield filter insertion were compared with those for patients who underwent repeat scanning to determine if there was evidence of a selection bias for these parameters among either group.
Data were analyzed with aid of SPSS entry program. Mortality was compared with Cox’s proportional hazard model adjusting for age, gender, and length of time spent in the study. The Cochran-Armithee trend test was used to examine the relationship between thrombus outcome and prevalence of high-probability V/Q scan. Exact contingency analysis was used to compare the prevalence of Greenfield filter insertion among the 4 groups. Analysis of variance (ANOVA) was used to compare mean age, mean PT, mean PTT, mean number of repeat ultrasounds per patient, and mean time over which the repeat ultrasounds took place.
Results
A total of 342 duplex ultrasounds were performed for lower extremity DVT on 118 patients over an 18 month period. The mean number of exams per patient was 2.9 1.3 (range 2-8). The average length of time to the first follow-up exam was 1.25 months, and the average total follow-up time with ultrasound was 4.3 months. The average total length of time spent in the study, which was calculated from the date of original diagnosis to date of death or the concluding date of the study, was 10.6 5.6 months. Average PT was 18.5 2.4 while average PTT was 49.6 18.3. In total, 30 V/Q scans were obtained: 2 were normal, 18 were low probability, 1 was intermediate probability, and 9 were high probability. Fourteen Greenfield filters were inserted and 14 deaths were recorded. There were no fatal pulmonary emboli.
Twenty-three patients (19%) had thrombus extend proximally, 45 patients (38%) had thrombus remain unchanged, 23 patients (19%) had thrombus improve, and 27 (23%) patients had complete resolution of thrombus. There was no association between the most proximal original vein segment involved and thrombus outcome (Table I). The average length of time todemonstrate proximal extension of DVT was 28 days. There was no statistical difference in the length of time to demonstrate proximal extension between patients with or without high probability V/Q scans.
Men and women were equally represented in all 4 groups. The extended, unchanged, and improved groups all had similar ages, whereas patients whose DVT resolved were younger (p
Patients whose DVT extended proximally had a significantly increased prevalence of highprobability V/Q scans (p
Results for 105 patients with no duplex ultrasound follow-up for lower extremity DVT are detailed in Table VI. These patients were found to be older than those patients with follow-up ultrasound exams (p
Discussion
Duplex ultrasound was accurately able to identify lower extremity DVT and the particular vein segments involved with thrombus in all patients. Nineteen percent of patients who had repeat duplex exams had thrombus extend proximally while 38% of DVTs remained unchanged and 19% of patients had improvement of the DVT. Only 23% of patients had complete resolution of thrombus over the course of the study. Average anticoagulation parameters were all therapeutic although the range included some values that were above and below 1.5 to 2.0 control values. These findings are similar to other studies that have prospectively assessed DVT outcome with serial duplex scanning and demonstrated a 30%,1 19%,2 38%4 and a 20%5 incidence of propagation despite standard anticoagulation measures. No particular vein segment that was originally involved with thrombus had an increased likelihood of propagation and the average length of time to thrombus extension was 28 days, indicating this process tends to occur early.
Table I. Thrombus outcome and original vein segment involved.
Table II. Demographics and thrombus outcome.
Table III. Thrombus outcome and distribution of V/Q scans.
Table IV. Thrombus outcome and anticoagulation parameters.
Table V. Thrombus outcome compared to mortality and Greenfield filter insertion.
Table VI. Ultrasound follow-up verses no ultrasound follow-up.
This would seem to imply that some patients with lower extremity DVT can have extension of their thrombi despite anticoagulation measures. It seems likely that the effectiveness of anticoagulation on thrombus varies, and factors that determine efficacy of anticoagulant therapy are not completely understood. However, the incidence of recurrent venous thromboembolism is 15 times greater in patients who are inadequately anticoagulated for 24 hours or more during therapy when compared to patients who are continuously anticoagulated.9 When therapeutic, heparin prevents extension of thrombi, accelerates clot organization and lysis,10,11 and decreases the incidence of thromboembolism.12 Yet venographically proven DVTs have a 32% incidence of propagation on repeat contrast study in the setting of adequate anticoagulation.13 Our data and the aforementioned prospective studies substantiate this same finding with serial duplex scanning.
This raises the question of what is the clinical implication of proximal extension of DVT in regard to thromboembolism? The overall incidence of high-probability V/Q scans in our patients with lower extremity DVT assessed by repeat duplex exam was 7.8%. Patients who had DVT extend proximally had a significantly higher prevalence of high-probability V/Q scans (13%) as compared to those patients with thrombi that remained unchanged, improved, or resolved (11%, 0%, and 3.7%, respectively). Furthermore, the difference in pulmonary embolism among these 4 groups could not be attributed to any difference in the number of Greenfield filters inserted or the level of anticoagulation. It may be that thrombi that propagate are less stable and more likely to fragment and cause pulmonary emboli.
The prevalence of pulmonary embolism was determined by the number of high-probability V/Q scans. Some controversy exists over the ability of V/Q scans to accurately diagnose pulmonary emboli. The Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED)8 assessed the sensitivity and specificity of V/Q scans by comparing 755 pulmonary angiograms to 931 V/Q scans in patients suspected of suffering from PE. Eighty-eight percent of patients with high- probability V/Q scans had angiographically proven PE, whereas 33% of patients with intermediate-probability scans and 12% of patients with low-probability V/Q scans had PE confirmed by angiography. Although including only high-probability V/Q scans as a positive outcome sacrifices some sensitivity, it gives the greatest positive predictive value. Furthermore, including intermediate-probability or even low-probability V/Q scans to increase sensitivity of outcome would markedly reduce specificity.
Given these data, the therapy for patients who propagate lower extremity DVT has not been established. Thrombolytic therapy is one alternative, but venous rethrombosis following treatment of DVT with streptokinase has been reported as high as 60%.14 An analysis of published estimates of the probabilities of various adverse outcomes associated with treatment of DVT with heparin verses heparin plus thrombolytic therapy found that patients treated with heparin alone would have a slightly increased life expectancy.15 This brings into question the use of thrombolytic therapy for proximal extension of DVT’. A second alternative would be insertion of a Greenfield filter. Risks and benefits of surgery must be weighed against a 13% prevalence of PE in patients with propagating DVT. In 1 series examining extended indications for placement of an inferior vena cava filter,16 no deaths were related to filter insertion. A 4.5% wound complication rate was found, and there was a 4.5% incidence of pulmonary emboli following filter insertion, which is similar to the 5% incidence commonly reported by others.17 Additionally, 98% caval patency has been found after filter insertion in longterm follow- up.17 A patient with marginal pulmonary function who cannot sacrifice any pulmonary reserve may be the ideal candidate for insertion of an intracaval filter device when DVT has been found to extend proximally.
Since this is a retrospective study, a certain selection bias might exist for patients who undergo repeat duplex ultrasound for lower extremity DVT as compared to those patients who have only a single exam for the same problem. Perhaps patients with persistently swollen extremities with more pain and/or tenderness prompt physicians to obtain repeated duplex exams to assess the thrombus. Furthermore, persistently symptomatic DVT may be associated with more thromboembolic complications. Conversely, patients with multiple duplex exams for DVT may be more “health conscious” and wish to have repeated exams to follow the outcome of their thrombus. In order to account for such a selection bias, we randomly selected 105 patients from the same study period who had a single duplex exam demonstrating lower extremity DVT and compared them to patients with multiple duplex exams for DVT. Although patients without followup were older, there was no difference in anticoagulation parameters, prevalence of intracaval filter insertion, or high-probability V/Q scans. These findings tend to diminish selection bias favoring either poor or improved outcome associated with repeat duplex exams for lower extremity DVT, further substantiating our findings.
In conclusion, repeat duplex ultrasound identifies proximal extension of lower extremity DVT, which may be associated with an increased prevalence of high-probability V/Q scans. To detect this propagation, we have suggested repeat duplex ultrasonography 1 week and 1 month after the initial diagnosis.1821 Those patients with extension of thrombus despite anticoagulation may benefit from insertion of an intracaval filter device. Prospective, multicenter trials are required to obtain necessary numbers to further substantiate these findings.
REFERENCES
1. Meissner MH, Caps MT, Bergelin RO, et al: Propagation, rethrombosis and new thrombus formation after acute deep venous thrombosis. J Vase Surg 22:558-567, 1995.
2. Killewich LA, Bedford GR, Beach KW, et al: Spontaneous lysis of deep venous thrombi: Rate and outcome. J Vase Surg 9:89-97, 1989.
3. Murphy TP, Cronan JJ: Evolution of deep venous thrombosis: A prospective evaluation with ultrasound. Radiology 177:543-548, 1990.
4. Krupski WC, Bass A, Dilley RB, et al: Propagation of deep venous thrombosis identified by duplex ultrasonography. J Vase Surg 12:467-475, 1990.
5. Ramhorst B, Bemmelen PS, Hoeneveld H, et al: Thrombus regression of spontaneous thrombolysis with duplex scanning. Circulation 86:414-419, 1992.
6. Johnson BF, Manzo RA, Bergelin RO, et al: Relationship between changes in the deep venous system and the development of the postthrombotic syndrome after an acute episode of lower limb deep vein thrombosis: A one- to six-year follow-up. J Vase Surg 21: 307- 313, 1995.
7. Markel A, Manzo RA, Bergelin RO, et al: Valvular reflux after deep vein thrombosis: Incidence and time of occurrence. J Vase Surg 15:377-384, 1992.
8. Results of the Prospective Investigation of Pulmonary Embolism Diagnosis(PIOPED). Value of the ventilation/perfusion scan in acute pulmonary embolism. JAMA 263:2753-2759, 1990.
9. Salzman EW, Deykin D, Shapiro RM, et al: Management of heparin therapy: Controlled prospective trial. N Engl J Med 292:1046-1050, 1975.
10. Bauer G: Clinical experiences of a surgeon in the use of heparin. Am J Cardiol 14:29-35, 1964.
11. Carey LC, Williams RD: Comparative effects of dicoumarol, tromexan, and heparin on thrombus propagation. Ann Surg 152:919- 922, 1960.
12. Coon WW, Willis PW III: Thromboembolic complications during anticoagulant therapy. Arch Surg 105:209-212, 1972.
13. Holm HA, Finnanger B, Hartmann A, et al: Heparin treatment of deep venous thrombosis in 280 patients: Symptoms related to dosage. Acta Med Scand 215: 47-53, 1984.
14. Dhall D, Dawson AA, Mavor GE: Problems of resistant thrombolysis and early recurrent thrombosis in streptokinase therapy. Surg Gynecol Obstet 146:15-20, 1978.
15. O’Meara JJ, McNutt RA, Evans AT, et al: A decision analysis of streptokinase plus heparin as compared with heparin alone for deep-vein thrombosis. N Engl J Med 330:1864-1890, 1994.
16. Rohrer MJ, Scheider MG, Wheeler B, et al: Extended indications for placement of an inferior vena cava filter. J Vase Surg 10:44-50, 1989.
17. Greenfield LJ: Current indications for and results of Greenfield filter placement. J Vase Surg 1:502-504, 1984.
18. Berry RE, George JE, Shaver WA: Free-floating deep venous thrombosis. A retrospective analysis. Ann Surg 211:719-712, 1990.
19. Lively SA, Devitt DT, Elder T, et al: Early Repetitive duplex after diagnosis of deep venous thrombosis. J Vase Tech 26:278-280, 2002.
20. Passman MA, Moneta GL, Taylor LM Jr, et al: Pulmonary embolism is associated with the combination of isolated calf vein thrombosis and respiratory symptoms. J Vase Surg 25:39-45, 1997.
21. Wolf B, Nichols DM, Duncan JL: Safety of a single duplex scan to exclude deep venous thrombosis. Br J Surg 87:1525-1528, 2000.
Enrico Ascher, MD, Patrick S. DePippo, MD, A. Hingorani, MD, W. Yorkovich, RPA and S. Salles-Cunha, PhD, Brooklyn, NY
Vase Endovasc Surg 38:525-531, 2004
From the Division of Vascular Surgery, Department of Surgery, Maimonides Medical Center, Brooklyn, NY
Correspondence: Enrico Ascher, MD, Director, Division of Vascular Surgery, Maimonides Medical Center, 4802 Tenth Avenue, Brooklyn, NY 11219
E-mail: [email protected]
2004 Westminster Publications, Inc, 708 Glen Cove Avenue, Glen Head, NY 11545, USA
Copyright Westminster Publications, Inc. Nov/Dec 2004
Comments