Angiology. Doppler Ultrasound in the Diagnosis of Venous Thrombosis

Transcription

1 Angiology The Journal of Vascular Diseases VOLUE 46 JANU/VRY 1995 NUBER 1 Doppler Ultrasound in the Diagnosis of Venous Thrombosis Arie arkel,.d.,.a.c.a.* Yehuda Weich,.D.t and Diana Gaitini,.D.t HAIA, ISRAEL ABSTRACT Deep vein thrombosis (DVT) has known morbidity and mortality. New noninvasive techniques such as Bmode scanning and Doppler ultrasonography (duplex) are highly accurate in the diagnosis of this problem but are relatively expensive and time consuming. Continuouswave Doppler, a precursor noninvasive technique, is simple, cheap, and easy to perform at the patient's bedside. To test the effectiveness of this technique the authors prospectively studied patients with clinical suspicion of DVT by Doppler ultrasound and compared the results with those from venography. During fourteen consecutive months, patients with a clinical suspicion of DVT underwent continuouswave Doppler examination of both inferior limbs. Each case was diagnosed as positive, negative, or inconclusive. In addition, the patients underwent a questionnaire regarding risk factors, symptoms, and mean relevant physical findings. Doppler examination was blinded to venography results. A total of 116 patients with clinical suspicion of DVT were examined by Doppler ultrasound. Their mean age was fiftyfive years (range: eighteen to eightyeight). There were 57 men and 59 women, and from this group a total of 40 patients underwent both Doppler ultrasound examination and venography in the course of fortyeight consecutive hours. When cases with an inconclusive result were excluded, Doppler ultrasound showed a sensitivity of 89%, a specificity of 100%, and an accuracy of 94% for the diagnosis of DVT when compared with venography. When a similar analysis was done for the proximal named veins, continuouswave Doppler examination showed a very high specificity and positive predictive value and a moderate sensitivity and negative predictive value. (continued on next page) rom the Departments of *Internal edicine "D" and tradiology, Rambam edical Center, Technion, aculty of edicine, Haifa, Israel. Presented at the 40th Annual eeting, American College of Angiology, Orlando, lorida, October, Westminster Publications, Inc., 708 Glen Cove Avenue, Glen Head, NY 11545, U.S.A. 65

2 Angiology The Journal of Vascular Diseases January 1995 (Abstract continued) In the entire group of patients with a Doppler ultrasound examination (n=116), 48 cases had a positive study. Prolonged bed rest, previous surgery, and previous DVT were the most frequent risk factors, pain and swelling the most frequent symptoms, and edema and tenderness the main clinical signs. In patients without DVT (n = 57), trauma, mahgnant obstructing tumors, and cellulitis were the most frequent final diagnoses. The study was inconclusive in 11 patients. In conclusion, continuouswave Doppler is accurate and may be used as the initial technique in the diagnosis of DVT. Cases that are inconclusive, or cases in which clinical findings are in evident contradiction to Doppler findings, should undergo duplex examination or venography. Introduction Deep vein thrombosis (DVT) is an important cause of morbidity and mortality. Since the clinical diagnosis of DVT is unreliable, objective methods are necessary to demonstrate the presence of thrombi. 1 During recent years new noninvasive techniques such as Bode scanning and echodoppler (duplex) have been developed. These methods have been demonstrated to be highly accurate and reliable for the diagnosis of venous thrombosis,.3 but they are rather expensive and time consuming. Continuouswave Doppler, one of the initial techniques used for the diagnosis of DVT,"* is cheap, portable, and easy to perform. To test the effectiveness of this technique we prospectively studied patients with a clinical suspicion of DVT with Doppler ultrasound and compared the results from this method with those from venography in a blinded fashion. aterials and ethods During fourteen consecutive months, patients referred to rule out DVT were subjected to a Doppler ultrasound examination. The study was performed with a continuouswave Doppler with a 4 Hz transducer. Principles involved in Doppler ultrasound design have been previously described by Strandness and Sumner.^ Normal venous Doppler findings include the presence of spontaneous phasic flow and the appearance of augmentation after distal compression or proximal release. DVT is considered to be present if flow is markedly diminished or absent, continuous flow is present, or augmentation after distal compression or proximal release is absent or sig 66 nificantly reduced. Examination of the common femoral vein (CV), superficial femoral vein (SV), and posterior tibial vein (PTV) at the ankle was performed with the patient in the supine position. The popliteal vein (PV) was examined with the patient prone, whenever possible, and the knee slightly flexed, with the leg resting over a pillow. Both legs were examined, and the findings in the suspected leg were compared with those in the contralateral one. All Doppler studies were done by the same investigator (A..). Results for each of the venous segments examined and the respective leg were reported as positive or negative. Studies that gave equivocal results were considered inconclusive. Ascending contrast venography was performed according to the standard technique with the contrast media injected through a needle inserted in one of the dorsal veins of the foot of the symptomatic limb. Acute thrombi were defined as a constant filling defect, abrupt cutoff of veins, and segment diversion of blood. Venography was usually carried out subsequent to the Doppler examination. The lapse of time between both tests had to be less than fortyeight hours for patients to be included in the study. In any case, the results of the Doppler and venographic examinations were reported without knowledge of the result of the other modality. Correlation between the results of venography and Doppler ultrasound examination was done by crosstab, and only definitive results were considered for analysis. Sensitivity, specificity, accuracy, and positive and negative predictive values of the Doppler examination, compared with the gold standard, venography, were determined for the involved leg and for the following venous segments: CV, SV, and PV. In addition, each patient was subjected to a questionnaire that in

3 arkel Doppler Ultrasound for DVT Diagnosis eluded personal data, risk factors for venous thrombosis, symptoms, signs, or other relevant clinical information. Subjects who underwent continuouswave Doppler examination, but no venography, during the same period of time were also included. ost of these patients were examined during the last part of this trial, and they were included for frequency analysis of risk factors, symptoms, physical findings, and other pertinent data. Results A total of 116 patients with clinical suspicion of DVT were examined by Doppler ultrasound. Their mean age was fiftyfive years (range: eighteen to eightyeight). There were 57 men and 59 women, and from this group a total of 4 patients underwent both Doppler ultrasound examination and venography. In cases the Doppler and venographic examinations were performed more than fortyeight hours apart (three and six days, respectively), and they were not considered for the analysis. When cases with an inconclusive result are excluded, Doppler ultrasound showed a sensitivity of 89% (17/19), a specificity of 100% (14/14), a positive predictive value of 100% (17/17), a negative predictive value of 88% (14/16), and an accuracy of 94% (31/33) for the diagnosis of DVT, when compared with venography (Table I). When a similar analysis was done for the CV, SV, and PV, continuouswave Doppler examination showed a very high specificity and positive predictive value and a moderate sensitivity and negative predictive value (Table II). Sensitivity and negative predictive value were better for the SV than for the other Table I Comparison of Doppler indings with Phlebography in 40 Patients with Suspected DVT Doppler indings Positive Phlebography Negative Total Positive Negative Total =nonconclusive. Table U Comparison of Doppler indings with Phlebography on the Different Venous Segments CV n (%) SV n (%) PV n (%) Sensitivity 11/15 (73) 16/17 (94) 13/16 (81) Specificity 18/18 (100) 16/16 (100) 18/18 (100) PPV 11/11 (100) 16/16 (100) 13/13 (100) NPV 18/ (8) 16/17 (94) 18/1 (86) Accuracy 9/33 (88) 3/33 (97) 31/34 (91) *CV=common femoral vein; SV= superficial femoral vein; PV=popliteal vein; PPV=positive predictive value; NPV=negative predictive value. 67

4 Angiology The Journal of Vascular Diseases January 1995 Table III Distribution of Thrombi in the Different Venous Segments According to Doppler Ultiasound and Venography Doppler Venography Patient Sex Age CV SV PV CV SV PV Calf 1..A.. G.A. 3. L.A. 4. Y.A. 5. E.A. 6. LA. 7. A.A. 8. D.B. 9. S.B. 10. CC. 11. R.C. 1. V.C. 13. P.D. 14. S.D. 15. S.D. 16..D. 17. N R.G. 19. L.H. 0. E.I. 1. R.K.. E.K. 3. S.L. 4. Y.L R.. 7. N.. 8. C. 9. P B.O. 31. S.O. 3. Z.P. 33. A.R. 34..S. 35. D.S. 36. CT. 37. N.T. 38. T.T. 39. R.V. 40. S.V h ( H f f f ) f t 1 1 \ \ f _ V H _ h 4 h 1 t l=positive; =negative; = nonconclusive; CV=common femoral vein; SV=superficial femoral vein; PV=popliteal vein. 68

5 arkel Doppler Ultrasound for DVT Diagnosis venous segments analyzed. A similar analysis for the calf veins was not performed. The patients with a positive venography in this study comprised a high proportion of cases who had multiple segment involvement (Table III): 14/1 cases (67%) had involvement of all three proximal veins: CV, SV, and PV. Although the calf segment was involved in most of the cases, 17/1 (81%), isolated calf vein thrombosis was not present in any one of the patients (Table III). assive thrombosis, involving all the veins of the leg, from the calf to the thigh, was present in 1 cases (5%). Two cases had a positive venography and a negative Doppler study (cases 0 and 30, Table III). In 1 of them the Doppler examination was performed two days after venography and three days after heparin was commenced. In the other case, venography showed a massive DVT with a patchy distribution. Six cases had an inconclusive Doppler study. In cases, the venography was also inconclusive and showed chronic obstruction secondary to an old DVT. The other 4 inconclusive cases included positive and negative venographic studies (Tables I and III). In the entire group of patients with a Doppler ultrasound examination (n=116), 48 cases had a Table V Symptoms in Study Patients Symptoms DVT( ) n (%) DVT() n (%) Pain 38 (79) 41 (7) Swelling 41 (85) 46 (81) 1 () 1 () ever % refer to the total number of cases in each group (DVT( )=48;DVT() = 57). positive study. The left leg was involved in 7 patients and the right leg in 17 patients (information was not available in 4 patients). The mean duration of symptoms until Doppler examination was 7.8 days (±6.5). Prolonged bed rest and surgery were the most frequent risk factors (Table IV), pain and swelling the most frequent symptoms (Table V), and edema and tenderness the main clinical signs (Table VI). The prevalence of risk factors, symptoms, and signs was usually similar in the patients with and without DVT, except for two clinical signs, cyanosis and prominent collateral veins, which were evidently more frequent in the patients with DVT (Table VI). In the patients without DVT, trauma, local obstructing malignant mass, and celluhtis were the most frequent final diagnoses (Table VII). ost cases referred for examination were inpatients (Table VIII), with patients from internal medicine wards more frequently referred for study than those undergoing surgical procedures. Among surgical cases, patients undergoing orthopedic or neurosurgical procedures were the most frequently studied (Table VIII). Table IV Risk actors in 44 Patients with DVT Risk actors Number of Cases* Prolonged bed rest 19 Postsurgery 14 Previous DVT 7 alignancy 6 Postpartum 5 Trauma 5 IVDA 4 IHD 4 Obesity 4 Pregnancy 3 Discussion The clinical diagnosis of venous thrombosis of the lower limbs is unreliable. or a proper diagnosis of this problem objective methods are necessary.^ Venography, considered as the gold standard, is painful, time consuming, and expensive and can result in complications secondary to dye contrast injection.5 Consequently, noninvasive methods were developed during the last twenty to thirty years for the diagnosis of this problem. EchoDoppler (duplex) and, more recently, color Doppler flow imaging (CDI) are the accepted *One or more risk factor can be present in the same patient. IVDA=intravenous drug abuser, IHD=ischemic heart disease. 69

6 Angiology The Journal of Vascular Diseases January 1995 Table VI Clinical Signs in Study Patients Signs Tenderness Redness Local heat Edema Cyanosis Prominent superficial veins PCD DVT() n (%) 31 (65) 4(8) 5 (5) 43 (90) 8(17) 11 (3) 3(6) DVT() n 36 (63) 4(7) 15 (6) 4 (74) 1 () 0(0) 0(0) % refer to the total number of cases in each group (DVT ( ) = 48; DVT () = 57). PCD = phlegmasia cerulea dolens. Table VII inal Diagnoses in Patients Without DVT inal Diagnosis Trauma alignant obstructing mass Cellulitis Varicose veins CH Baker's cyst Hematoma Venous insufficiency Arthritis Number of Cases Table VHI Referral Source of Studied Patients Referring Department Internal medicine Emergency ward Orthopedics Neurosurgery Gynecology General surgery Oncology Number of Cases CH=congestive heart failure. 70

7 arkel Doppler Ultrasound for DVT Diagnosis techniques currently in use for the diagnosis of DVT, and their use has been validated by comparison with venography..6 Duplex is accurate, can precisely define the level of involvement of venous thrombosis,^ and can demonstrate the presence of nonocclusive thrombi, especially when CDI is used.^ On the other hand, the instrumentation is expensive, the examination is prolonged when both legs are examined and both Doppler and imaging are used, and the technique requires a relatively long learning curve. Continuouswave Doppler was one of the earliest methods used for the diagnosis of venous thrombosis.4 This method is simple, cheap, portable as a pocket instrument, and easy to perform at the patient's bedside. The reported accuracy of the method, sensitivity, specificity, and positive and negative predictive values are similar to those presented here when symptomatic patients are considered.8.9 On the other hand, studies performed in asymptomatic patients, such as for screening in postsurgery cases, show lower sensitivity and positive predictive value.1 Regarding the calf or "belowtheknee" thrombosis, we searched for thrombi at the posterior tibial vein at the ankle, but no more proximally in these veins, nor at the peroneal and anterior tibial veins. Examination of these veins by continuouswave Doppler is difficult and has low sensitivity,'*.!! and therefore, results on these segments are not reported. It is important to take into account that other noninvasive methods are also inaccurate for this area..1 Although CDI can reliably visualize the calf veins, as shown by van Bemmelen et al,i3 and has been found by some investigators to be effective for calf DVT diagnosis, i4 others have found this technique to have low sensitivity when compared with venography in asymptomatic patients.i^ The main risk factors shown in our study included bed rest, previous surgery, and previous DVT. This is similar to other, previously reported trials.16 Intravenous drug abuse (IVDA) was one of the important risk factors present in our study. Thrombosis in these cases was secondary to injection of the drug through the femoral vein, usually causing massive ileofemoral thrombosis. Decision for treatment is especially difficult in these patients because of their persistent drug injection practice, low compliance, and the consequent danger from hemorrhage. Swelling was the most frequent symptom, and edema the most frequent sign, present in the majority of the patients (Tables V and VI). Cyanosis and pronounced collateral veins, although not frequent signs, were clearly more frequent in DVT patients than in those cases with a negative Doppler study. Phlegmasia cerulea dolens was present only in DVT patients, showing once more that the clinical diagnosis of venous thrombosis can be based solely on this finding. Regarding other Doppler findings, we found that the left leg was more frequently involved than the right. It has been known for a considerable period of time that thrombosis tends to occur more frequently in the left leg than in the right.i7 One factor that may contribute to this is the compression of the left common iliac vein by the right commmon iliac artery.i^ Among patients with symptoms but without DVT, the most frequent diagnoses included trauma, malignancy, and cellulitis (Table VII). Rupture of a Baker's cyst, a condition that can mimic venous thrombosis, was present in 3 patients. In these cases the diagnosis was confirmed by Bmode ultrasound. ost of the subjects referred for investigation were hospitalized patients. In this group, the administration of subcutaneous heparin or other prophylactic measures can significantly reduce the number of cases that develop DVT. In patients who underwent a Doppler ultrasound study without venography, treatment was administered according to the results of the noninvasive examination. Although duplex scanning is a more accurate method for the diagnosis of DVT in certain areas, such as the profunda femoral vein, external iliac vein, and the greater and lesser saphenous veins, and for the diagnosis of nonocclusive thrombi, we beheve that when the question is simply whether the patient has or has not DVT, Doppler ultrasound should be sufficient and may be used as the first method of diagnosis. This can result in a savings of time and money. Nevertheless, the use of duplex scanning is advised in the following situations: 1. When the Doppler study is inconclusive. If the Doppler examination is negative, but the clinical presentation is very suggestive of DVT (If the duplex result is also negative, but isolated calf vein thrombosis is suspected based on clinical findings, venography may be indicated. We are convinced that, when diagnosed, isolated calf vein thrombosis should be treated 71

8 Angiology The Journal of Vascular Diseases January 1995 with anticoagulants because of the possibility of proximal extension and pulmonary embolism in a significant number of cases.^9) 3. When a malignant mass or tumor, hematoma, or Baker's cyst rupture is suspected 4. In patients with suspicion of pulmonary embolism when confirmation of DVT is particularly important 5. In research or natural history studies of venous thromboembolic disease An algorithm, expressing the previous statements is suggested (ig. 1). Conclusions Doppler ultrasound is an effective and accurate method for the diagnosis of DVT. We believe that Doppler ultrasound should be the first examination done in a patient with a clinical suspicion of venous thrombosis, and if the study result is clearly positive or negative, treatment should be administered accordingly (ig. 1). In negative studies, it is recommended that the test be repeated after three to five days, and again after ten days, to exclude cases with calf vein thrombosis that had extended proximally. When the Doppler ultrasound examination is inconclusive, or when there is disproportion between the severity of symptoms and a negative Doppler study, duplex scanning should be done, as previously suggested. Whenever the duplex study result is not definitive, one should proceed to venography. Arie arkel,.d.,.a.c.a. Department of Internal edicine "D" Rambam edical Center Haifa, Israel Clinical DVT. j C.W. Doppler Ultrasound Positive T Negative Inconclusive Anticoagulant therapy Duplex (If symptoms are very suggestive of DVT) Do not treat (repeat Doppler after 35 days and again after 1 0 days) Positive Negative Anticoagulant therapy Inconclusive (If catf DVT suspected cllnicaiiy) Do not treat (repeat dupiex after 35 days and again after 1 0 days) Venography igure 1. Suggested algorithm for DVT* investigation in patients referred with a clinical suspicion of venous thrombosis. *DVT=deep vein thrombosis; C.W.=continuous wave. 7

9 arkel Doppler Ultrasound for DVT Diagnosis References 1. Cranky JJ, Canos AJ, Sull WJ: The diagnosis of deep venous thrombosis: allibility of clinical symptoms and signs. Arch Surg 111:3436, Killewich LA, Bedford GR, Beach KW, et al: A prospective study comparing duplex scanning to contrast venography. Circulation 79:810814, Barnes RW, Nix L, Barnes L, et al: Perioperative asymptomatic venous thrombosis: Role of duplex scanning versus venography. J Vase Surg 9:5160, Strandness DE Jr, Sumner DS: Ultrasonic velocity detector in the diagnosis of thrombophlebitis. Arch Surg 104:180183, Bettman A, Paulin S: Leg phlebography: The incidence, nature and modification of undesirable side effects. Radiology 1:101104, attos A, Londrey GL, Leutz DW, et al: Colorflow duplex scanning for the surveillance and diagnosis of acute deep venous thrombosis. J Vase Surg 15: , arkel A, anzo RA, Bergelin R, et al: Pattern and distribution of thrombi in acute venous thrombosis. Arch Surg 17:305309, Barnes RW, Russell H, Wilson R: Doppler Ultrasound Evaluation of Venous Disease: A Programmed Audiovisual Instruction. Iowa City, University of Iowa, 1975, pp Yao JST, Gourmos C, Hobbs JT: Detection of proximal vein thrombosis by Doppler ultrasound flow. Lancet 1:14, TurnbuU TJ, Dymowski JJ: Emergency department use of handheld Doppler ultrasonography. Am J Emerg ed 7:0915, Sigel B, Popky GL, app E, et al: Evaluation of Doppler ultrasound examinations: Its use in diagnosis of lower extremity venous disease. Arch Surg 100:535540, Hull RD, Van Aken WG, Hirsh J, et al: Impedance plethysmography using the occlusive cuff technique in the diagnosis of venous thrombosis. Circulation 53:696700, van Bemmelen PS, Bedford G, Strandness DE Jr: Visualization of calf veins by color flow imaging. Ultrasound in ed & Biol 16:1517, Baxter G, Duffy P, Partridge E: Colour flow imaging of calf vein thrombosis. Clin Radiol 46:19801, Rose SC, Zwiebel WJ, urdock LE, et al: Insensitivity of color Doppler flow imaging for detection of acute calf deep venous thrombosis in asymptomatic postoperative patients. J Vase Interv Radiol 4:111117, arkel A, anzo RA, Bergelin RO, et al: Acute deep vein thrombosis: Diagnosis, localization, and risk factors. J Vase ed Biol 3:43439, currich JP: The occurrence of congenital adhesions in the common iliac veins and their relation to thrombosis of the femoral and iliac veins. Am J ed Sci 135:34346, Cockett B, Thomas L: The iliac compression syndrome. Br J Surg 5:81681, Hyers T, Hull RD, Weg JG: Antithrombotic therapy for venous thromboembolic disease. Chest 10 (suppl) :408S45S,

10