 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Prophylaxis against Deep Venous Thrombosis after Total Knee Arthroplasty. Pneumatic Plantar Compression and Aspirin Compared with Aspirin Alone* , NEW YORK, N.Y.Investigation performed at The Hospital for Special Surgery, Affiliated with The New York Hospital—Cornell University Medical College, New York City
A prospective, randomized study was conducted to assess the efficacy of pulsatile pneumatic plantar compression for prophylaxis against deep venous thrombosis after total knee arthroplasty performed with use of regional anesthesia. One hundred and twenty-two patients (164 knees) who were scheduled to have a unilateral or a one-stage bilateral total knee arthroplasty were separately randomized to be managed with either aspirin alone or the pulsatile pneumatic plantar-compression device and aspirin. The prevalence of deep venous thrombosis was 27 per cent (twenty-two of eighty-one knees) in the group treated with pneumatic plantar compression compared with 59 per cent (forty-nine of eighty-three knees) in the patients managed with aspirin alone (the control group) (p < 0.001). A significant difference was also noted in the group that had had a unilateral arthroplasty (a prevalence of 27 per cent [eleven of forty-one knees] in the group treated with pneumatic plantar compression, compared with 67 per cent [twenty-six of thirty-nine knees] in that treated with aspirin alone; p < 0.006) and in the group that had had a one-stage bilateral procedure (a prevalence of 28 per cent [eleven of forty knees] in the group treated with pneumatic plantar compression, compared with 52 per cent [twenty-three of forty-four knees] in that treated with aspirin alone; p < 0.03). No proximal thrombi were noted in any patient who used the pulsatile pneumatic plantar-compression device, while the prevalence of proximal thrombosis in the popliteal or femoral veins was 14 per cent (twelve of eighty-three knees) in the group treated with aspirin alone (p < 0.0003). In the group treated with a unilateral procedure and aspirin alone the prevalence of proximal thrombosis was 13 per cent (five of thirty-nine knees; p < 0.02), while in the group treated with a bilateral procedure and aspirin alone it was 16 per cent (seven of forty-four knees; p < 0.01). Only in the patients who had had a unilateral procedure was use of the compression device associated with significantly less edema postoperatively than was use of aspirin alone. The change between the preoperative and postoperative circumferences of the thigh and leg was significantly less (9 ± 4.1 millimeters [mean and standard deviation] less for the thigh [p < 0.01] and 6 ± 3.9 millimeters less for the leg [p < 0.049]) with the compression device than with aspirin alone. In addition, there was significantly less mean drainage (98 ± 61.1 milliliters) in the group treated with a unilateral procedure and pneumatic compression, compared with that treated with a unilateral procedure and aspirin alone (p < 0.041). An internal timer of the compression device was used to assess the compliance of the patient with use of the device, and a relationship between deep venous thrombosis and the total duration of treatment with the device was found. The patients in whom deep venous thrombosis did not develop used the device for a mean of 96 ± 23.4 hours (range, sixty to 164 hours) postoperatively, or 19.2 ± 5.1 hours a day, while those in whom thrombosis developed used it for a mean of 67 ± 21.1 hours (range, twenty-six to 101 hours), or 13.4 ± 4.3 hours a day (p < 0.001). No untoward effects were noted in any patient who used the device. This study confirms the safety and efficacy of pulsatile pneumatic plantar compression and aspirin compared with aspirin alone and supports the use of mechanical compression for prophylaxis against deep venous thrombosis and for reduction of edema in patients who have had a total knee arthroplasty. In addition, we found a direct relationship between compliance with the use of this device and its efficacy in reducing deep venous thrombosis.
Deep venous thrombosis following total knee arthroplasty occurs commonly in the calf and is more frequent after bilateral than after unilateral procedures34. In contrast to the prevalence following total hip arthroplasty, isolated proximal venous thrombosis is infrequent after total knee arthroplasty25,31,34,42,43,55. Propagation of thrombi in the veins of the calf proximally into the popliteal and femoral veins is well documented and is the causative factor for most fatal pulmonary emboli after total knee arthroplasty10,20,22,36,46. The need for some form of prophylaxis therefore is paramount to avoid or decrease the prevalence of thromboembolic disease and its associated morbidity and mortality after total knee arthroplasty. The etiology of thromboembolic disease is multifactorial and it is related to multiple risk factors, including the type and duration of anesthesia34,44,52,53 as well as the means of postoperative prophylaxis. Prophylaxis includes the use of pharmacological agents, such as aspirin9,23,25,26,29,31,34,40,41,43,57, low-molecular-weight heparin6,12, and warfarin4,8,11,12,21,23,24,29,34. Prophylaxis can also be carried out with mechanical devices, such as pneumatic compression boots4,5,7,11,16-18,21,22,24,25,27,30-35,37,39,42-45,48-51,55,59-61. The use of intermittent pneumatic-compression devices accelerates the emptying of the venous system, decreases venous stasis, and effects fibrinolysis1,14-17. Multiple studies have documented the safety and efficacy of intermittent pneumatic compression after total joint arthroplasty5,22,25,30-35,43,48-51,59-61. Pulsatile pneumatic plantar compression is a relatively new modality that utilizes the physiological venous pump mechanism in the sole of the foot as described by Gaskell and Parrott19 as well as by Gardner and Fox14-18. In contrast to traditional external pneumatic-compression devices that increase venous pressure by only twenty to thirty millimeters of mercury (2.67 to 4.00 kilopascals), pulsatile pneumatic plantar-compression devices produce a forceful ejection of blood from the foot into the calf with a pressure of more than 100 millimeters of mercury (13.33 kilopascals). This pulse wave of venous blood that originates in the plantar venous plexus flows proximally into the popliteal and femoral veins and is demonstrable with both venography and Doppler ultrasound14-18. Pulsatile pneumatic plantar compression has been accepted as an efficacious form of mechanical prophylaxis against deep venous thrombosis after total hip arthroplasty5,13,56. We hypothesized that it would also be a useful modality for prophylaxis against deep venous thrombosis after total knee arthroplasty and that it would have the added advantages, compared with traditional external pneumatic-compression devices, of ease of application (as the foot is readily accessible) and, we hoped, of improved compliance by the patient. We used aspirin as a control in this study, as it was used as a control in a previous study from our institution in which a thigh-high external pneumatic-compression device was evaluated25. The purpose of the current study was to compare the efficacy of pulsatile pneumatic plantar compression and aspirin with that of aspirin alone for prophylaxis against deep venous thrombosis after primary unilateral and one-stage bilateral total knee arthroplasty performed with use of regional anesthesia. In addition, we sought to evaluate the reduction of edema and to assess objectively the compliance by the patient.
Patients scheduled for a primary unilateral or a one-stage bilateral total knee arthroplasty on the Surgical Arthritis Service at The Hospital for Special Surgery were eligible for inclusion in the study. The criteria for exclusion included a history of deep venous thrombosis, pulmonary emboli, peptic ulcer disease, or allergy to aspirin. Any patient who was receiving anticoagulation therapy was also excluded. The study was approved by the Institutional Review Board, and informed written consent was obtained from each patient. Between May 1993 and March 1995, 164 knees (122 patients) were studied. Eighty patients had a primary unilateral total knee arthroplasty and forty-two, a primary one-stage bilateral total knee arthroplasty. All procedures were performed with use of regional anesthesia. There were forty male and eighty-two female patients. The primary diagnosis was osteoarthrosis in 109 patients (145 knees) and rheumatoid arthritis in thirteen patients (nineteen knees). The patients were separately randomized by hospital number into two groups: those who were managed with aspirin alone (the control group) and those who were managed with the pulsatile pneumatic plantar-compression device and aspirin (the study group). The control group had eighty-three arthroplasties (thirty-nine patients [thirty-nine knees] had a unilateral procedure and twenty-two patients [forty-four knees], a bilateral procedure), while the study group had eighty-one arthroplasties (forty-one patients [forty-one knees] had a unilateral procedure and twenty patients [forty knees], a bilateral procedure). The aspirin was administered postoperatively on the night of the operation and was continued twice a day in oral doses of 325 milligrams. The pulsatile pneumatic plantar-compression device (PlexiPulse; NuTech, San Antonio, Texas) was applied in the recovery room. Postoperatively, the device was used while the patient was recumbent, and it was removed only for walking and daily hygiene. The cuff of the device has a single chamber that inflates for two seconds and cycles every twenty seconds. The pressure in the chamber was set to the intermediate level of 160 millimeters of mercury (21.33 kilopascals), as suggested by the manufacturer. All patients had epidural anesthesia with intravenous sedation. After the lower extremity had been exsanguinated with an elastic bandage, all patients had a total knee arthroplasty with insertion of a posterior stabilized device with cement, under tourniquet control. Before closure of the wound, a closed drainage system was inserted. At the end of the operation, a bulky dressing was applied and was reinforced with medial and lateral plaster splints to keep the knee in full extension. On the first postoperative day, the drain was removed, the plaster splints were discarded, and progressive, continuous passive motion was started. A standard physical-therapy protocol, including full weight-bearing on the extremity with the assistance of a walker, was initiated on the second postoperative day. Risk factors for deep venous thrombosis and pulmonary embolism were determined for each patient and included an age of more than forty years, female gender, obesity, varicose veins, smoking, hypertension, diabetes mellitus, and coronary artery disease. Preoperative determinations included the serum hematocrit and the circumferences of the thigh and leg, measured ten centimeters proximal to the superior pole of the patella and ten centimeters distal to the inferior pole of the patella, respectively. On the third postoperative day, the serum hematocrit, the need for transfusion, and the amount of drainage from the wound were assessed. On the fifth postoperative day, the circumferences of the thigh and leg were measured again. A gross assessment of blood loss was determined by comparing each patient's preoperative serum hematocrit with that noted on the third postoperative day. The amount of postoperative edema was assessed by comparing the preoperative circumferences of the thigh and leg with those measured on the fifth postoperative day. In accordance with the protocol at our institution, a venogram of the affected extremity was made for all patients on the fourth to the seventh postoperative day unless this was contraindicated (as in patients who had an allergy to contrast medium) or unattainable, in which case venous Doppler ultrasound was performed. The radiologist who interpreted the study was blinded with regard to the prophylactic regimen. Thrombi were classified as proximal if the popliteal or femoral veins were involved and as distal if only the veins of the calf were involved. Distal thrombi were then subclassified as major (if they were six centimeters long or longer or were multiple) or minor (if they were less than six centimeters long). If the result of the screening study was positive, the prophylactic treatment was considered to have failed, the use of aspirin was discontinued, and therapy with warfarin was instituted. If the patient had been receiving pneumatic plantar compression, that treatment was also discontinued. If the result of the screening study was negative, the use of aspirin was continued but the use of the pneumatic plantar compression device was discontinued, as the patient was then able to walk. Patients who had a negative result on the screening study were discharged from the hospital and continued to use aspirin for one month. Although treatment with low-dose warfarin for a thrombosis that is distal to the knee and is not causing symptoms is controversial, we managed all patients who had a positive result on the study with warfarin, regardless of the presence of symptoms. All patients who had a deep venous thrombosis were discharged from the hospital and received warfarin for a minimum of six weeks. The compliance of the patient with the study protocol was assessed with a unique aspect of the PlexiPulse pneumatic compression device. An internal timing device (hour meter), one for each extremity, was used to record the over-all amount of time that pneumatic plantar compression was employed. After completion of treatment with the device (at a mean of five days postoperatively), the total amount of hours was recorded and the association with a positive or a negative result on the screening study was determined to assess further the efficacy of the device. Statistical evaluation was performed with use of chi-square analysis, the Fisher exact test, and the Student t test.
After evaluation of the risk factors for each group, it was noted that a significantly greater number of cigarette smokers, both over-all (p < 0.03) and in the group that had had a bilateral procedure (p < 0.01), used the pulsatile pneumatic plantar-compression device. With the numbers available, no other significant difference in risk factors was noted between the study and the control group. The rate of deep venous thrombosis was significantly lower in the patients who used the pulsatile pneumatic plantar-compression device and aspirin than in those who used aspirin alone. Over-all, the prevalence of deep venous thrombosis was 27 per cent (twenty-two of eighty-one knees) in the study group, compared with 59 per cent (forty-nine of eighty-three knees) in the control group (p < 0.001). A significant difference was also noted in the group treated with unilateral total knee arthroplasty; the prevalence of deep venous thrombosis was 27 per cent (eleven of forty-one knees) in the group treated with pneumatic plantar compression and aspirin, compared with 67 per cent (twenty-six of thirty-nine knees) in the control group (p < 0.006). In the group that had had a one-stage bilateral procedure, there was also a significant difference: the prevalence of deep venous thrombosis was 28 per cent (eleven of forty knees) in the study group, compared with 52 per cent (twenty-three of forty-four knees) in the control group (p < 0.03) (Fig. 1).
No proximal thrombi were noted in any patient who used the pulsatile pneumatic plantar-compression device, while proximal thrombi in the popliteal or femoral veins were seen after 14 per cent (twelve) of the eighty-three arthroplasties in the control group (p < 0.0003). The prevalence of proximal thrombosis was 13 per cent (five of thirty-nine knees) in the group treated with a unilateral procedure and aspirin alone (p < 0.02), while it was 16 per cent (seven of forty-four knees) in the group treated with a bilateral procedure and aspirin alone (p < 0.01). The analysis of deep venous thrombosis in the calf revealed that multiple, larger thrombi were more common in the control group; in the study group, the thrombi tended to be isolated and smaller. Over-all, major deep venous thrombosis in the calf was observed after 10 per cent (eight) of the eighty-one arthroplasties in the study group, compared with 39 per cent (thirty-two) of the eighty-three arthroplasties in the control group (p < 0.0001). The prevalence of major deep venous thrombosis in the calf was 15 per cent (six of forty-one knees) after unilateral procedures and 5 per cent (two of forty knees) after bilateral procedures in the study group, compared with 44 per cent (seventeen of thirty-nine knees) after unilateral procedures (p < 0.006) and 34 per cent (fifteen of forty-four knees) after bilateral procedures (p < 0.0009) in the control group (Fig. 2).
A venogram was made, on the fourth to the seventh postoperative day (mean, five days), for sixty-one (75 per cent) of the eighty-one extremities in the study group and for sixty-nine (83 per cent) of the eighty-three extremities in the control group (p > 0.25). In the remaining extremities, for which venograms were contraindicated or unattainable, venous Doppler ultrasound was used. Venography was performed after 71 per cent (twenty-nine) of the forty-one unilateral arthroplasties and after 80 per cent (thirty-two) of the forty arthroplasties in patients who had a bilateral procedure in the study group, compared with 77 per cent (thirty) of the thirty-nine unilateral arthroplasties (p > 0.60) and after 89 per cent (thirty-nine) of the forty-four arthroplasties in patients who had a bilateral procedure (p > 0.40) in the control group. With the numbers available, no significant difference was noted, between the study and the control group, with regard to the use of venography as opposed to Doppler ultrasound. Over-all, there was a large amount of variability in the amount of drainage collected in the closed drainage system. No significant difference was noted between the over-all group treated with pneumatic plantar compression and the control group (mean, 665 compared with 694 milliliters; p > 0.16). However, after the unilateral total knee arthroplasties, significantly less drainage was collected in the study group than in the control group (mean, 612 compared with 710 milliliters; p < 0.041). No major difference was noted between the study and control groups after the bilateral total knee arthroplasties. Over-all, the mean reduction in the serum hematocrit in the group treated with plantar compression was 8.3 per cent (a mean of 35.9 per cent preoperatively compared with a mean of 27.6 per cent postoperatively); in the control group, the mean reduction was 9.4 per cent (a mean of 36.7 per cent preoperatively compared with a mean of 27.3 per cent postoperatively) (p > 0.07). After the unilateral total knee arthroplasties, the mean reduction in the serum hematocrit was 7.2 per cent (a mean of 35.4 per cent preoperatively compared with a mean of 28.2 per cent postoperatively) in the study group and 9.1 per cent (a mean of 36.4 per cent preoperatively compared with a mean of 27.3 per cent postoperatively) in the control group; this difference was significant (p < 0.02). However, no appreciable difference was noted between the study and control groups after the bilateral procedures. Also, no important difference was noted, between the study and the control group, with regard to the number of units of blood transfused postoperatively. After the unilateral total knee arthroplasties, there was significantly less swelling in the patients who were managed with the pneumatic plantar-compression device than in the control group. The changes in the mean circumferences of the thigh and leg were both significantly less (9 ± 4.1 and 6 ± 3.9 millimeters less [p < 0.01 and p < 0.049], respectively) in the former group compared with the latter. However, this difference in postoperative edema was not clinically important. In the group treated with bilateral total knee arthroplasty, we observed no appreciable difference in the edema between the study and the control group. There was a significant relationship between the number of hours that the patient used the compression device postoperatively and the development of deep venous thrombosis. The patients in whom deep venous thrombosis did not develop used the device for a mean (and standard deviation) of 96 ± 23.4 hours (range, sixty to 164 hours), or 19.2 ± 5.1 hours a day, while those in whom a thrombosis did develop used the device for a mean of 67 ± 21.1 hours (range, twenty-six to 101 hours), or 13.4 ± 4.3 hours a day (p < 0.001) (Fig. 3).
No untoward effects were noted in any patient who used the pneumatic plantar-compression device. There was no evidence of major or minor bleeding, and there were no problems with the skin associated with the use of the device.
Patients who have an elective total joint arthroplasty of a lower extremity are at high risk for the development of thromboembolic disease2,5,8,9,11-13,22-35,37-45,48-61. Without prophylaxis after total knee arthroplasty, the over-all prevalence of deep venous thrombosis has been between 40 and 84 per cent, while proximal thrombosis has been observed in 8 to 23 per cent of patients in large series2,4,9,25,31,32,35,37-40,42,43,45,50,55,58,60. Pulmonary embolism has been documented in 2 to 5 per cent of patients who have not had prophylaxis, and fatal pulmonary embolism has been reported in 1 to 2 per cent25,34. Since isolated proximal thrombosis is rare after total knee arthroplasty, propagation of thrombi in the calf proximally into the popliteal and femoral veins is the main etiology of most fatal pulmonary emboli. Propagation of thrombi in the calf was demonstrated by Doouss10 and by Kakkar et al.36; more recently, Grady-Benson et al., using serial venous Doppler ultrasound, reported a 23 per cent rate of propagation of thrombi in the calf (in five of twenty-two patients)22. Prophylaxis after total knee arthroplasty is essential to prevent the known morbidity and mortality associated with thromboembolic disease. Although regional anesthesia has been shown to reduce the prevalence of thromboembolic disease in association with total hip arthroplasty, it appears less effective in association with total knee arthroplasty when a tourniquet is used44,53. Sharrock et al. demonstrated a reduction in both the over-all rate of deep venous thrombosis and the rate of proximal deep venous thrombosis after total knee arthroplasty in which regional anesthesia had been used (48 per cent [109 of 227 knees] and 4 per cent [nine of 227 knees], respectively), compared with the rates associated with the use of general anesthesia (64 per cent [169 of 264 knees] and 9 per cent [twenty-four of 264 knees], respectively)53. However, some form of prophylaxis against thromboembolic disease is still required, since the prevalence of proximal and distal thrombosis was notable. Prophylatic regimens include the use of pharmacological agents, such as aspirin9,23,25,26,29,31,34,40,41,43,57, low-molecular-weight heparin6,12, and warfarin4,8,11,12,21,23,24,29,34, and mechanical methods, such as pneumatic compression4,5,7,11,16-18,21,22,24,25,27,30-35,37,39,42-45,48-51,55,59-61. After a review of the results for more than 9000 patients, the Antiplatelet Trialists' Collaboration concluded that antiplatelet therapy produced a highly significant reduction in deep venous thrombosis (p < 0.00001) and pulmonary embolism (p < 0.00001)2. Although that study did not deal specifically with patients who had had an arthroplasty, the authors recommended antiplatelet therapy in association with prolonged high-risk orthopaedic operations and they noted that antiplatelet therapy can be conveniently continued after discharge from the hospital, in contrast to many other forms of prophylaxis2,40. Low-dose warfarin has been utilized for prophylaxis against thromboembolic disease, but it is associated with bleeding complications, routine phlebotomy, notable cost, and inconvenience after discharge4,8,11,12,21,23,24,29,34. Recently, Lotke et al. compared the results with use of aspirin or low-dose warfarin in 388 patients who had had a total hip or knee arthroplasty and noted no difference in the prevalence of deep venous thrombosis or pulmonary embolism40. Whether this was due to the effectiveness of aspirin or to the ineffectiveness of low-dose warfarin remains to be clarified. Fitzgerald et al., in a study of 349 patients who had had a total knee arthroplasty, noted a significantly higher rate of thrombosis in a control group that had received low-dose warfarin, compared with the rate in a group that had been treated with low-molecular-weight heparin (45 per cent compared with 25 per cent; p < 0.001)12. We used aspirin as a control in our study because antiplatelet therapy had been employed previously at our institution25 and because, as also noted in the two studies just cited12,40, we had concerns regarding the use of low-dose warfarin for prophylaxis against deep venous thrombosis. Multiple studies have documented the safety and efficacy of intermittent pneumatic compression after total joint arthroplasty5,22,25,30-35,43,48-51,59-61. Haas et al., in a study of 119 patients who had been managed with either a thigh-high sequential external pneumatic-compression device or aspirin after a total knee arthroplasty, found a reduction in deep venous thrombosis—from 47 per cent with the use of aspirin to 22 per cent with the use of compression (p < 0.03)—in patients who had had a unilateral procedure25. However, in the group that had had a bilateral procedure, compression did not lead to a significant reduction, and the rate of proximal thrombosis with that type of prophylaxis was 8 per cent. Compliance by the patients was not documented objectively25. In 1983, Gardner and Fox described their observations of a so-called physiological venous foot-pump in the sole of the foot; this pump involved the venae comitantes of the lateral plantar artery14. The pump had a stroke volume of approximately twenty milliliters that emptied only through the deep venous system and was activated solely on weight-bearing14. On the basis of this description, and that of Gaskell and Parrott19, pneumatic plantar-compression devices were developed to activate the pump in patients who are unable to walk. In addition to increasing venous return, it is hypothesized that plantar impulse pumping may prevent deep venous thrombosis by means of several other mechanisms14-18. First, high-flow states increase turbulence around venous valves, thus decreasing the formation of thrombi. Second, hemodynamic studies17,18 have confirmed increased blood flow and tissue perfusion with the release of endothelial-derived relaxing factor and prostacyclin. Third, Allenby et al. documented enhancement of fibrinolysis with use of external pneumatic compression1, and fibrinolytic activity has been thought to occur with compression of the foot, as with other pneumatic-compression devices17. Plantar compression appears to be a useful modality for prophylaxis against deep venous thrombosis, and its efficacy after total hip arthroplasty has been documented5,13,51,56. Bradley et al. performed a prospective, randomized study of seventy-four patients who had a total hip arthroplasty; half were managed with a plantar compression device5. Venography revealed only two clots in the femoral veins in the patients who had been managed with the pump, compared with eleven clots in the femoral veins and one clot in the veins of the calf in the control group5. Fordyce and Ling evaluated the results of plantar compression after total hip arthroplasty in seventy-nine patients; venography revealed clinically important deep-venous thrombosis (proximal thrombi and major thrombi in the veins of the calf) in two (5 per cent) of the thirty-nine patients who had been managed with the pump and in sixteen (40 per cent) of the forty patients in the control group13. Santori et al. recently demonstrated, with use of Doppler ultrasound, the effectiveness of plantar compression compared with that of unfractionated heparin after total hip arthroplasty in 132 patients; those authors noted deep venous thrombosis in 13 per cent (nine) of the sixty-seven patients in the first group and 35 per cent (twenty-three) of the sixty-five patients in the second51. There is limited information, however, with regard to the efficacy of such a device in patients who have had a total knee arthroplasty. Wilson et al. studied a series of fifty-nine patients who had had sixty total knee arthroplasties and compared the results in those who had used a plantar compression device with those in a control group that had not had any prophylaxis60. Those authors noted that use of the device was associated with a reduction in the prevalence of proximal deep venous thrombosis; such thrombi occurred in 19 per cent (six) of the thirty-two knees in the control group and in none of the twenty-eight knees in the study group (p = 0.01)60. Their study lacked statistical power for deep venous thrombosis in the calf and failed to control for many variables, such as the type of anesthesia, risk factors, and compliance by the patient60. A major criticism of pneumatic compression has been the lack of compliance by the patient. In a study of 138 patients, Comerota et al. noted that only 33 per cent had a properly functioning external pneumatic-compression device; 85 per cent of the failures were due to lack of compliance, and only 15 per cent were due to a malfunctioning device7. In the current study, we believed that it was essential to document compliance by the patients to assess more accurately the effectiveness of plantar compression. The PlexiPulse foot pump that we selected for this study had the benefit of an internal timing device to document such compliance. In addition, pneumatic plantar compression can be easily applied, as the foot is readily accessible. We found that patients in whom deep venous thrombosis did not develop had used the plantar compression device postoperatively for an over-all mean of ninety-six hours (a mean of 19.2 hours a day), while those in whom it did develop had used the device for an over-all mean of only sixty-seven hours (a mean of 13.4 hours a day). The most common reason for lack of compliance was failure to restart the pneumatic compression device after treatment was interrupted (such as after walking, physical therapy, or diagnostic testing). One limitation of our study concerns the postoperative screening test for deep venous thrombosis. For the overwhelming majority of patients, we made a venogram of the affected extremity; we used venous Doppler ultrasound only if a venogram was unattainable. Doppler ultrasound may not be as accurate as venography in determining the size of a thrombosis in the calf3; however, we are able to differentiate between major and minor thrombi in the calf as well as between isolated and multiple thrombi with use of Doppler ultrasound at our institution. With the numbers available, we did not observe a significant difference in the use of these two modalities between the study and the control group. In a multicenter trial, Gardner et al. demonstrated a reduction in post-traumatic and postoperative edema with use of artificial activation of the venous foot-pump18; similar findings were noted by Myerson and Henderson after trauma or an operation involving the foot and ankle47. In the current study, we noted a significantly smaller increase in the circumferences of the thigh (p < 0.01) and leg (p < 0.049) in the patients who had been managed with plantar compression after unilateral arthroplasty compared with the patients who had been managed with aspirin after unilateral arthroplasty; however, this difference was not clinically important. Of the patients treated with unilateral arthroplasty, those in the study group also had a significant reduction (ninety-eight milliliters) in the mean amount of drainage collected in the closed drainage system compared with those in the control group; we hypothesized that this was related to the increase in venous and lymphatic return in the affected extremity. In the patients who had had a bilateral procedure, greater blood loss necessitated increased replacement of fluids postoperatively and a net reduction in edema or in the amount of drainage collected was not observed. We hypothesized that this was because of the inability of the foot-pump to increase the venous and lymphatic drainage enough to overcome the excess postoperative replacement of fluids. In conclusion, we found pulsatile pneumatic plantar compression and aspirin to be a safe and effective method of prophylaxis against thromboembolic disease in patients who had had a unilateral or a one-stage bilateral total knee arthroplasty. Furthermore, we demonstrated that effective prophylaxis with this device depends on compliance by the patient in the postoperative period.
NOTE: The authors thank Susan Edleman, P.A., and Carolyn Dessel, P.A., for their assistance in the initial collection of the data.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |