0022-534 7/92/14 74-1056$03.00/0 THE JOURNAL OF UROLOGY Copyright© 1992 by AMERICAN UROLOGICAL ASSOCIATION, INC.

Vol. 147, 1056-1059, April 1992

Printed in U.S.A.

PROSPECTIVE RANDOMIZED TRIAL OF WARFARIN AND INTERMITTENT PNEUMATIC LEG COMPRESSION AS PROPHYLAXIS FOR POSTOPERATIVE DEEP VENOUS THROMBOSIS IN MAJOR UROLOGICAL SURGERY PARAMJIT S. CHANDHOKE,*'t GRETCHEN A. W. GOODING

AND

PERINCHERY NARAYAN+

From the Departments of Urology and Radiology, University of California School of Medicine, San Francisco, California

ABS'l'R-ACT

Postoperative deep venous thrombosis and pulmonary embolus remain a major source of morbidity and mortality for the urological surgery patient. We report the results of the first 100 patients in a prospective, randomized trial of low dose warfarin and intermittent pneumatic leg compression for deep venous thrombosis prophylaxis. All patients underwent preoperative and postoperative realtime ultrasound imaging and Doppler flow studies of the popliteal, femoral and iliac veins for the evaluation of deep venous thrombosis. Our results indicate that low dose warfarin is as effective as intermittent pneumatic leg compression for prophylaxis of deep venous thrombosis. Low dose warfarin can be used effectively without any significant bleeding complications. We recommend the use of low dose warfarin as an alternative to intermittent pneumatic leg compression for deep venous thrombosis prophylaxis of the urological patient undergoing a major urological operation. KEY WORDS: thrombophlebitis, ultrasonic diagnosis, warfarin, urology

Patients undergoing urological surgery are estimated to have a 25% incidence of postoperative deep venous thrombosis. 1 However, the incidence of deep venous thrombosis is believed to be much higher for a major pelvic operation. 125Iodine (1251) fibrinogen leg scanning studies have shown that as many as 40% of the patients undergoing open prostatectomy have subclinical deep venous thrombosis. 1 Therefore, along with knee and hip arthroplasty, major pelvic surgery is considered to be one of the highest risk factors for postoperative deep venous thrombosis. 2 In the absence of prophylaxis these patients have a 10 to 20% risk of proximal vein thrombosis, and a 1 to 5% risk of fatal pulmonary emboli. 2 The goal of prophylaxis is to prevent deep venous thrombosis and thereby minimize its associated complications (the postphlebitic syndrome, recurrent venous thrombosis and pulmonary embolism). Of the thrombi detected by 1251 fibrinogen leg scanning 80% are considered small, asymptomatic and confined to the calf, and they are not associated with subsequent complications.3 Most clinically significant and fatal emboli arise from venous thrombi that occur or extend proximal to the calf veins (that is the popliteal, femoral or iliac veins). Thus, a deep venous thrombosis prophylaxis regimen may be considered effective only if it prevents the formation of proximal vein thrombosis. A variety of regimens is available for the prevention of deep venous thrombosis in postoperative patients, including heparin, dihydroergotamine, oral anticoagulants (warfarin), dextran, aspirin, intermittent pneumatic leg compression, graduated compression stockings or a combination of these modalities. Although effective in the deep venous thrombosis prophylaxis of the general surgical patient, low dose heparin has been shown to be ineffective for the patient undergoing urological surgery. 4- 8 In a controlled, randomized trial of low dose heparin and calf intermittent pneumatic leg compression in patients undergoing Accepted for publication September 27, 1991. * Dornier/American Foundation for Urologic Disease Scholar. t Present address: Division of Urologic Surgery, Washington University School of Medicine, 4960 Audubon Ave., St. Louis, Missouri 63110. :j: Requests for reprints: Department of Urology, University of California at San Francisco, U575-533 Parnassus Ave., San Francisco, California 94143-0738.

an open urological operation Coe et al reported a venographic determined deep venous thrombosis rate of 25% in the control group, 21 % in the heparin group and 3.5% in the intermittent pneumatic leg compression group. 4 They concluded that low dose heparin was ineffective, whereas intermittent pneumatic leg compression provided significant deep venous thrombosis prophylaxis for the urological surgery patient. W arfarin has been conclusively shown to be a highly effective form of deep venous thrombosis prophylaxis 9 and it has been considered by some as the prophylaxis of choice for the orthopedic patient undergoing hip arthroplasty. 2 However, its routine use has been limited by the perceived risk of postoperative bleeding complications. Two studies have shown that low dose warfarin can be used as effective deep venous thrombosis prophylaxis without any added risk of bleeding complications in the perioperative period. 10• 11 To be most effective for deep venous thrombosis prophylaxis, intermittent pneumatic leg compression should be applied sequentially on the calf and thigh, instituted intraoperatively and continued postoperatively until the patient becomes fully ambulatory.12 Potential drawbacks of intermittent pneumatic leg compression are occasional patient discomfort, possible noncompliance by the patient or hospital staff, and its occasional unavailability for the entire intraoperative and postoperative period. Intermittent pneumatic leg compression is also not routinely available at all hospitals caring for the urological surgery patient. Furthermore, a recent study evaluating the effectiveness of intermittent pneumatic leg compression in patients undergoing total hip replacement indicated that intermittent pneumatic leg compression may not be as effective as warfarin prophylaxis. 13 To evaluate efficacy of an alternative prophylaxis for the urological patient undergoing a major urological procedure, we began a prospective, controlled, randomized clinical trial of intermittent pneumatic leg compression versus low dose warfarin in January 1987. We report the results of the study in our first 100 patients. MATERIALS AND METHODS

A total of 100 patients hospitalized between January 1987 and December 1990 for a major open urological operation lasting more than 2 hours was randomized (by coin toss) to either intermittent pneumatic leg compression or low dose

1056

DEEP VENOUS THROMBOSIS PROPHYLAXIS IN UROLOGICAL SURGERY

warfarin prophylaxis. None of the patients had a history of venous thrombosis, prolonged immobility or paralysis, congestive heart failure in the perioperative period or an inherited risk of thromboembolic disorders. All patients undergoing pelvic surgery had Penrose drains placed bilaterally in the perivesical area. These drains were removed gradually in the postoperative period as the drainage subsided. All patients underwent a preoperative and a 1 to 2-week postoperative realtime ultrasound examination of the iliac, common femoral, superficial and deep femoral, and popliteal veins. The transducers used for imaging were linear array of 3.5 or 5 MHz. The common femoral veins were examined for visible thrombus, for venous size at rest and after a Valsalva maneuver, and for venous response to compression by the transducer on the skin surface over the vein. Veins were considered normal if they were compressible, increased by 50% during a Valsalva maneuver, augmented freely and had aphasic venous signal on Doppler examination. The superficial femoral, profunda femoris and popliteal veins were examined for the same findings without a Valsalva maneuver done consistently. The iliac veins were commonly obscured postoperatively but a normal response to compression, augmentation and phasicity of the common femoral vein was considered indicative of normalcy. If the common femoral vein had completely normal responses except for a decreased or lack of response to a Valsalva maneuver postoperatively, and the patient was able to accomplish an adequate Valsalva maneuver (which was sometimes not possible because of postoperative discomfort), that was considered either an indication of deep venous thrombosis of the iliac veins, or of an adjacent compressing hematoma, lymphocele or other space-occupying mass. When the iliac veins were visualized, the waveform on spectral analysis was characterized and response to augmentation was assessed. The Valsalva maneuver and compression were not routinely performed on the iliac veins. Numerous studies in recent years have shown that real-time ultrasound imaging is equivalent to venography for the diagnosis of deep venous thrombosis. 14- 19 In our study protocol all patients suspected of having deep venous thrombosis by ultrasound also underwent subsequent venography. Low dose warfarin prophylaxis was begun on the night of the operation and continued postoperatively until the patient was discharged from the hospital (1 to 2 weeks). The goal of low dose warfarin prophylaxis was to achieve a prothrombin time of approximately 1.5 times the preoperative value by 3 or 4 days postoperatively. The dose of warfarin was adjusted thereafter to maintain the prothrornbin time at this level (approximately 16 to 18 seconds). Sequential leg and thigh intermittent pneumatic leg compression was instituted intraoperatively and continued postoperatively for 5 days or until the patient became fully ambulatory. RESULTS

Our patient population may be considered in the high risk category of venous thromboembolic disease. 14 The major risks of deep venous thrombosis associated with increased patient age, malignancy and major pelvic surgery were inherent in this patient population. These risks were distributed proportionally among the intermittent pneumatic leg compression and low dose warfarin groups (table 1). The mean age of the warfarin group was 66.1 ± 6.4 years (standard deviation), whereas for the intermittent pneumatic leg compression group it was 67.5 ± 7.1 years. Of the 100 patients 90 underwent surgery for a urological malignancy. Most of our patients had a major pelvic operation. In the other pelvic surgery group (table 1) 2 patients underwent pelvic lymph node dissection and 1 underwent an open bladder neck suspension procedure. Kidney surgery was either radical or partial nephrectomy for renal cancer. Two patients (4%) who received intermittent pneumatic leg

TABLE

L Patient characteristics and distribution of operations Low Dose Warfarin

Pt. characteristics: Mean age (yrs.) ± standard deviation No. men/No. women Urological malignancy(%) Type of surgery: Radical prostatectomy Radical cystectomy Other pelvic surgery Kidney surgery

TABLE 2.

1057

66.1 ± 6.4

Intermittent Pneumatic Leg Compression 67.5

±

53/0 100

46/1

44

37 4 3 3

7.1

98

5 0 4

Complications of prophylaxis for deep venous thrombosis Complication

Low Dose Warfarin (%)

Deep venous thrombosis Pulmonary embolus Pelvic fluid collection Postop. bleeding Deaths

Intermittent Pneumatic Leg Compression

0 0 7.5

2 0

(%)

4* 2* 15* O* 0

* P value not statistically significant by Fisher's exact test.

compression prophylaxis had deep venous thrombosis postoperatively. A 66-year-old woman underwent open bladder neck suspension and she presented clinically with a pulmonary embolus before the routinely scheduled postoperative ultrasound examination. Left femoral deep venous thrombosis was documented by ultrasound criteria and venography. A 63-year-old man underwent radical retropubic prostatectomy. In addition to the diagnosis of right femoral deep venous thrombosis on postoperative ultrasound examination, a large left pelvic lymphocele developed that required percutaneous drainage and sclerosis. Both of these patients were treated with anticoagulation therapy and recovered fully from the thromboembolic complications. There were no cases of postoperative deep venous thrombosis in the low dose warfarin group. Deep venous thrombosis was ruled out by subsequent venography in 2 patients, 1 each in the low dose warfarin and intermittent pneumatic leg compression groups, who had an abnormal postoperative ultrasound study. An interesting observation in our study was an increased occurrence of postoperative pelvic fluid collections in patients undergoing intermittent pneumatic leg compression (table 2). Whereas 7.5% of the low dose warfarin group had postoperative pelvic fluid collections, this complication occurred in 15% of the intermittent pneumatic leg compression group (statistically not significant). Although most of these were believed to be lymphoceles by ultrasound imaging, the smaller ones were asymptomatic, untreated and could possibly have been hematomas. The 10 patients with this complication are listed in table 3. In 6 of these patients the fluid collections were small and asymptomatic, and required no subsequent intervention. The other 4 patients required specific treatment 2 to 6 weeks postoperatively. One patient in the low dose warfarin group required percutaneous drainage of an infected lymphocele. Three patients in the intermittent pneumatic leg compression group had symptomatic pelvic lymphoceles that had to be either aspirated or sclerosed. There was 1 patient in the low dose warfarin group with dark, heme-positive stools 5 days after radical retropubic prostatectomy. He had a history of cirrhosis of the liver, a preoperative hematocrit of 33, prothrombin time of 12.6 and normal liver function tests but pitting edema of the lower extremities. He received a warfarin dose of 2.5 mg. on the day of surgery and 2.5 mg. on subsequent postoperative days. At the time of the hematochezia prothrombin time was 21.3 and hematocrit was 31 (which was unchanged from the hematocrit measured 1

CHANDHOKE, GOODING AND NARAYAN

1058 TABLE

Pt. TC GR WR JV RB LP FR WR JS

GT

3. Characteristics of postoperative pelvic fluid collections

Deep Venous Thrombosis Prophylaxis Low dose warfarin Low dose warfarin Low dose warfarin Low dose warfarin Intermittent pneumatic leg compression Intermittent pneumatic leg compression Intermittent pneumatic leg compression Intermittent pneumatic leg compression Intermittent pneumatic leg compression Intermittent pneumatic leg compression

Size of Fluid Collection (cm.)

Treatment

Small Small 4X5 7 X 13 2.5 X 2.5

None None None Drainage of infection None

3X2

None

Large

Sclerosed

Small

None

Large

Aspirated

3 X 6,4 X 8,2.5 X 10 Aspirated

day postoperatively). Warfarin was stopped and the patient was given a single 10 mg. dose of vitamin K. There was no clinical indication for significant postoperative blood loss nor were any blood transfusions given. There were no cases of significant postoperative bleeding requiring postoperative blood transfusion in the intermittent pneumatic leg compression group. No deaths occurred during this study. DISCUSSION

The incidence of postoperative proximal vein deep venous thrombosis in the urological surgery patient undergoing a major pelvic operation has been estimated to be 10 to 20% without prophylaxis. 1 With the use of real-time ultrasound for the diagnosis of proximal vein thrombosis, which may be considered equivalent to venography, the results of our study indicate that low dose warfarin and intermittent pneumatic leg compression are highly effective modes of deep venous thrombosis prophylaxis. Although 4 % of the patients in the intermittent pneumatic leg compression group, compared to 0% in the low dose warfarin group, had postoperative deep venous thrombosis, the total number of patients in the study was too small to make this a statistically significant difference between the 2 groups. Nevertheless, our study indicates that low dose warfarin can be used safely and is as effective as intermittent pneumatic leg compression in the prevention of proximal vein thrombosis and pulmonary embolus for the patient undergoing a major open urological operation. There was a 10% incidence of a postoperative pelvic fluid collection. Of these fluid collections 60% were asymptomatic and did not require intervention, while 40% were clinically significant and required a combination of aspiration, tube drainage and/or sclerosis therapy. Although the incidence of these fluid collections was twice as common in the intermittent pneumatic leg compression group than in the low dose warfarin group, this difference was not statistically significant. The efficacy of a particular deep venous thrombosis prophylactic regimen seems to be dependent on the type of operation and surgical subspecialty. Whereas heparin is effective for the general surgical patient, it has been shown to be ineffective for urological surgery. 8 It is interesting that orthopedic and urological operations also have among the highest rates of deep venous thrombosis without prophylaxis. 2 The efficacy of warfarin in the orthopedic patient has been well established during the last 3 decades. 9 Our study is the first prospective clinical trial of warfarin prophylaxis for the urological patient. The major perceived drawback of warfarin is its potential complication for postoperative bleeding, which has limited its clinical use. To overcome this drawback 2 recent studies have shown that by modifying the dose of warfarin in the perioperative period, its effectiveness for deep venous thrombosis prophylaxis can be retained and the associated bleeding complications minimized. 10· 11 In a randomized prospective clinical

trial of a 2-step warfarin versus dextran prophylaxis for elective hip and knee replacement surgery, Francis et al reported that warfarin was more effective while the bleeding complications in the 2 groups were similar. 10 In a randomized trial of less intensive postoperative warfarin, aspirin or no prophylaxis for patients undergoing an operation for a fractured hip, Powers et al reported that warfarin was most effective as deep venous thrombosis prophylaxis and the transfusion rates for all 3 groups were similar. 11 In their study warfarin was started on the day of operation and adjusted to achieve a prothrombin time of 16 seconds. This protocol was similar to the one adopted in our study. In our initial experience with the use of low dose warfarin before beginning this randomized study, we found that a 10 or 5 mg. starting dose-0-f warfarin often caused the p:r-othrombin time to overshoot by postoperative day 3 or 4. When we adopted an initial 2.5 mg. dose for our clinical protocol, we found that adequate postoperative prothrombin time profiles could be obtained as with an initial 5 mg. dose schedule but without the sudden rapid increase in prothrombin time sometimes associated with the 5 mg. dose. The maximum prothrombin time in this subset of patients was 17.7 ± 1.4 seconds (standard error) for the 2.5 mg. group and 17.8 ± 0.8 for the 5 mg. group. Adequate anticoagulation with only a 2.5 mg. warfarin dose may have been due to the concomitant use of cimetidine (which has a well known clinical effect of decreasing the hepatic metabolism ofwarfarin type anticoagulants 20 ) in some patients. Increased bioavailability of the drug due to preoperative bowel preparation or oral intake status of the patients in the postoperative period may have also contributed to an adequate prothrombin time with the 2.5 mg. dose. In the patient who had hematochezia with warfarin prophylaxis the prothrombin time reached 21.3 seconds with the 2.5 mg. regimen. The hematochezia was not accompanied by a decrease in the hematocrit nor did the patient require any blood transfusion. An interesting question that this study raises is whether low dose warfarin actually prevented clot propagation into the proximal veins or whether it dissolved clots already formed in the smaller, more distal calf veins. Further studies of earlier postoperative ultrasonography or 125 I fibrinogen leg scanning may answer this question. Nevertheless, warfarin appears to achieve the primary objective of preventing proximal vein thrombosis and associated pulmonary embolism. In conclusion, this prospective, randomized clinical trial of deep venous thrombosis prophylaxis in 100 patients undergoing a major open urological operation indicates that low dose warfarin is as effective as intermittent pneumatic leg compression. Low dose warfarin prophylaxis can be provided effectively and without any significant bleeding complications. Its alternative use should be considered in the postoperative management of the urological patient undergoing a major pelvic operation. REFERENCES

1. Prevention of venous thrombosis and pulmonary embolism. J.A.M.A., 256: 744, 1986. 2. Hull, R. D., Raskob, G. E. and Hirsh, J.: Prophylaxis of venous thromboembolism. An overview. Chest, suppl. 5, 89: 3748, 1986. 3. Dalen, J. E., Paraskos, J. A., Ockene, I. S., Alpert, J. S. and Hirsh, J.: Venous thromboembolism. Scope of the problem. Chest, suppl. 5, 89: 3708, 1986. 4. Coe, N. P., Collins, R. E. C., Klein, L. A., Bettmann, M. A., Skillman, J. J., Shapiro, R. M. and Salzman, E.W.: Prevention of deep vein thrombosis in urological patients: a controlled, randomized trial oflow-dose heparin and pneumatic compression boots. Surgery, 83: 230, 1978. 5. Miller, E. V. and Butz, G. W.: Preoperative mini-heparin, prostatectomy and clinical pulmonary thromboembolism. Int. Surg., 66: 155, 1981. 6. Williams, H. T.: Prevention of postoperative deep-vein thrombosis with perioperative subcutaneous heparin. Lancet, 2: 950, 1971. 7. Rosenberg, I. L., Evans, M. and Pollock, A. V.: Prophylaxis of

1059

8. 9.

10.

11.

12.

13.

postoperative leg vein thrombosis by low dose subcutaneous heparin or perioperative calf muscle stimulation: a controlled clinical trial. Brit. Med. J., 1: 649, 1975. Brenner, D. W., Fogle, M. A. and Schellhammer, P. F.: Venous thromboembolism. J. Urol., 142: 1403, 1989. Salzman, E. W.: Prevention of venous thromboembolism in highrisk patients by prophylactic administration of oral anticoagulants. Adv. Exp. Med. Biol., 214: 165, 1987. Francis, C. W., Marder, V. J., Evarts, C. M. and Yaukoolbodi, S.: Two-step warfarin therapy. Prevention of postoperative venous thrombosis without excessive bleeding. J .A.M.A., 249: 37 4, 1983. Powers, P. J., Gent, M., Jay, R. M., Julian, D. H., Turpie, A. G. G., Levine, M. and Hirsh, J.: A randomized trial of less intense postoperative warfarin or aspirin therapy in the prevention of venous thromboembolism after surgery for fractured hip. Arch. Intern. Med., 149: 771, 1989. Hartman, J. T., Pugh, J. L., Smith, R. D., Robertson, W. W., Jr., Yost, R. P. and Janssen, H.F.: Cyclic sequential compression of the lower limb in prevention of deep venous thrombosis. J. Bone Joint Surg., 64: 1059, 1982. Hull, R. D., Raskob, G. E., Gent, M., McLaughlin, D., Julian, D., Smith, F. C., Dale, N. I., Reed-Davis, R., Lofthouse, R. N. and Anderson, C.: Effectiveness of intermittent pneumatic leg

compression for preventing deep vein thrombosis after total hip replacement. J.A.M.A., 263: 2313, 1990. 14. Effeney, D. J., Friedman, M. B. and Gooding, G. A. W.: Iliofemoral venous thrombosis: real-time ultrasound diagnosis, normal criteria and clinical application. Radiology, 150: 787, 1984. 15. Langsfeld, M., Hershey, F. B., Thorpe, L., Auer, A. I., Binnington, H. B., Hurley, J. J. and Woods, J. J.: Duplex B-mode imaging for the diagnosis of deep venous thrombosis. Arch. Surg., 122: 587, 1987. 16. Flinn, W.R., Sandager, G. P., Cerullo, L. J., Havey, R. J. and Yao,

J. S. T.: Duplex venous scanning for the prospective surveillance of perioperative venous thrombosis. Arch. Surg., 124: 901, 1989. 17. Rollins, D. L., Semrow, C. M., F.riedell, M. L., Calligaro, K. D. and Buchbinder, D.: Progress in the diagnosis of deep venous thrombosis: the efficacy of real-time B-mode ultrasonic imaging. J. Vase. Surg., 7: 638, 1988. 18. Becker, D. M., Philbrick, J. T. and Abbitt, P. L.: Real-time ultrasonography for the diagnosis of lower extremity deep venous thrombosis. The wave of the future? Arch. Intern. Med., 149: 1731, 1989. 19. Habscheid, W., Hohmann, M., Wilhelm, T. and Epping, J.: Real-

time ultrasound in the diagnosis of acute deep venous thrombosis of the lower extremity. Angiology, 41: 599, 1990. 20. Physicians' Desk Reference. Oradell, New Jersey: Medical Economics Data, p. 2121, 1991.

Prospective randomized trial of warfarin and intermittent pneumatic leg compression as prophylaxis for postoperative deep venous thrombosis in major urological surgery.

Postoperative deep venous thrombosis and pulmonary embolus remain a major source of morbidity and mortality for the urological surgery patient. We rep...
137KB Sizes 0 Downloads 0 Views