Keywords: compression stockings, deep venous thrombosis, exercise, post-thrombotic syndrome, randomized cross-over trial.

Transcription

1 Journal of Thrombosis and Haemostasis, 1: ORIGINAL ARTICLE Effect of graduated elastic compression s on leg symptoms and signs during exercise in patients with deep venous thrombosis: a randomized cross-over trial 1 S. R. KAHN, y L. AZOULAY, A. HIRSCH,y M. HABER, C. STRULOVITCH and I. SHRIER Center for Clinical Epidemiology and Community Studies; and the ydepartment of Medicine, Sir Mortimer B. Davis Jewish General Hospital, McGill University, Montréal, Canada Summary. Graduated elastic compression s (ECS) are often prescribed after deep venous thrombosis (DVT) to alleviate acute symptoms and to prevent and treat post-thrombotic syndrome (PTS). In patients with DVT, leg symptoms tend to worsen with exercise. The effects of ECS use during exercise have not been studied. Objectives were to determine whether ECS improve symptoms and signs and increase exercise capacity when worn during treadmill exercise by patients with prior DVT, with or without PTS. The methods employed a randomized cross-over trial. We recruited subjects who had a first episode of unilateral DVT at least 1 year earlier and categorized them as having, or not having, the PTS using a validated scale. Subjects underwent two identical treadmill exercise sessions at least 1 week apart, and were randomly assigned to wear knee-length 30 mmhg ECS on the affected leg during one of the two sessions. Venous symptoms, leg volume, leg circumference and calf muscle flexibility were measured in the affected leg before and after both exercise sessions. Subjects achieved similar percentage maximum predicted heart rates during both sessions. Comparing the ECS to no ECS session, there were no significant differences in treadmill time (21.2 vs min, P ¼ 0.94), gain in leg volume (71 vs. 73 ml, P ¼ 0.83), or change in soleus or gastrocnemius flexibility, whether or not PTS was present. Symptoms in general worsened slightly with exercise regardless of whether or not ECS were worn and did not differ according to PTS status. Per-subject analysis showed that use of ECS resulted in global improvement of symptoms in 25% of subjects, global worsening in 33% of subjects, and had no or Correspondence: Dr S. R. Kahn, Center for Clinical Epidemiology and Community Studies, Sir Mortimer B. Davis Jewish General Hospital, 3755 Cote Ste. Catherine Rm. A-114, Montreal, Quebec H3T 1E2, Canada. Tel.: þ ext. 4667; fax: þ ; susan.kahn@ mcgill.ca 1 A portion of this work was presented as a poster at the American Society of Hematology Annual meeting, Orlando, Florida, USA, in December Received 28 August 2002, resubmitted 1 October 2002, accepted 3 October 2002 inconsistent effects in 42% of subjects. Whether or not PTS was present, the use of ECS during exercise by patients with prior DVT did not improve symptoms and signs during exercise or increase exercise capacity. Keywords: compression s, deep venous thrombosis, exercise, post-thrombotic syndrome, randomized cross-over trial. The post-thrombotic syndrome (PTS), a chronic sequel of deep venous thrombosis (DVT), occurs in up to 20 50% of patients with symptomatic DVT, and usually becomes established by the first 2 years after the acute thrombotic episode [1,2]. Typical clinical features of PTS include leg pain and heaviness, swelling, venous ectasia, skin induration and occasionally, venous ulceration [3]. Graduated elastic compression s (ECS) are often prescribed to patients after DVT to prevent PTS and/or to alleviate leg symptoms and swelling. They are worn daily during waking hours, i.e. while the patient is upright and ambulatory. ECS are thought to be beneficial via reducing venous hypertension, improving tissue microcirculation, and assisting the calf muscle pump, thereby reducing edema and venous symptoms [4,5]. However, there is conflicting evidence regarding the true effectiveness of ECS in preventing and treating PTS [2,6], and there is little data regarding the optimum compression to achieve benefit or the patterns of use that produce maximal effect. For example, venous symptoms and leg swelling tend to worsen with upright posture and exercise, but whether use of ECS during exercise by patients with previous DVT relieves or, alternately, aggravates symptoms is unknown. This is of practical concern to these patients who are often eager to carry on with their usual activities, including exercise. If ECS are effective in reducing exercise-induced leg swelling or leg symptoms, their use might allow patients to increase their activity level and thereby obtain greater overall benefit from exercise. The objectives of this study were to determine if ECS worn during treadmill exercise in patients with prior DVT (at least one year previously) improve symptoms and signs and increase exercise capacity, and whether the presence or absence of the PTS influences these effects.

2 A randomized cross-over trial 495 Methods Subjects Study subjects were recruited from our hospital s Thrombosis Clinic, a clinic that follows all patients diagnosed with venous thromboembolism (VTE). Consecutive patients with a first diagnosis of unilateral DVT established by objective testing at least one year prior were approached for participation in the study. This time period was chosen to avoid, for safety reasons, subjecting patients with a recent, fresh DVT to vigorous treadmill exercise, and to allow the passage of adequate time since the diagnosis of DVT to be able to differentiate early reversible symptoms and swelling from established PTS. Subjects were excluded if they had bilateral DVT, more than one episode of DVT, symptomatic pulmonary embolism, any medical condition that precluded treadmill exercise, if they were incapable of responding to a questionnaire in French or English, or if they were unable or refused to provide informed consent. Prior to its initiation, the hospital s Research Ethics Committee approved the study. Study procedures The study protocol required two study visits, which took place at least one week apart, always in the late afternoon. Each visit had three components: baseline (pre-exercise) measures, a treadmill exercise protocol, and postexercise measures, as outlined below. The procedures for both treadmill sessions were identical, except that subjects were randomly assigned (by coin toss) to wear a brand new, custom-fitted, 30 mmhg knee-length compression on their affected leg during either the first or second session. Baseline measures Data were collected about the subjects age, gender, and site of DVT from the hospital records by one of two research assistants. Proximal DVT was defined as DVT involving the popliteal or more proximal venous segments, with or without associated calf (distal) DVT. The supervising physician, who was blinded to data from the hospital record, ascertained the presence and severity of PTS at baseline using the Villalta PTS scale [7]. The Villalta PTS scale assigns a severity score from 0 (not present or minimal) to 3 (severe) for five symptoms (pain, cramps, heaviness, pruritus and paresthesia) and six signs (edema, skin induration, hyperpigmentation, venous ectasia, redness, pain during calf compression). A summative score of 4 or less indicates absence of PTS, a score of 5 14 indicates mild/moderate PTS, and a score of 15 or more or the presence of a venous ulcer indicates severe PTS. This scale has high interobserver agreement for individual components, total score, and classification of PTS severity [7]. Subjects were unaware of the study hypothesis and rated their symptoms of PTS (heaviness, aching, swelling, heat or burning, restless legs, throbbing, itching, tingling) using 10-cm visual analog scales. Leg volume, circumference and calf muscle flexibility were then measured in the affected leg by one of two research assistants who were blinded to the subject s PTS status. These measures were chosen because leg swelling is a known consequence of DVTand could be attenuated by ECS use [4], and reduced flexibility is associated with exercise-induced symptoms such as stiffness and cramping in healthy individuals [8]. Calf circumference at 10 cm above floor level was measured using a standardized, reliable tape measure device [9]. To obtain a more accurate measure of swelling, leg volume was measured by placing the lower leg into a container of water and measuring the volume of water displaced. In order to assess the reliability of this instrument, in a separate experiment we repeated volume measurements 10 times in one subject and the standard deviation of the measures (74 ml) represented 2.6% of the mean volume (2800 ml), indicating good reliability, similar to that reported by others [4]. Calf muscle flexibility of the soleus muscle (a postural muscle used in standing and walking) and the gastrocnemius muscle (a non-postural muscle, i.e. used in running and jumping) were assessed with a gravity-based goniometer [10,11] as follows: for the gastrocnemius muscle, the subject stood with the knee straight, and the foot was planted on the ground with the heel fixed and the toes pointing straight ahead. The leg was moved forward, keeping the heel firmly planted on the ground. Flexibility is defined as the angular rotation of the tibia away from the perpendicular to the floor (08) during the maneuver. For the soleus muscle, the same procedure was used but the knee was bent. For both maneuvers, a larger angle indicates greater flexibility. Exercise protocol After the baseline measures were recorded, subjects rested for 30 min before beginning exercise. Exercise was performed on a treadmill set at a 58 incline. During a 5-min warm-up, subjects progressively increased treadmill speed (all changes in speed were recorded) in order to reach a comfortable speed that still resulted in mild tachypnea and sweating. Following the 5-min warm-up, subjects exercised at a constant speed (i.e. speed at the end of warm-up) for as long as they were able, up to a maximum of 30 min (total of 35 min). Thirty minutes was chosen because this duration of brisk walking each day will provide much of the total health benefits of exercise [12]. In order to standardize the two tests as closely as possible, the research assistant set the treadmill speed for the second test to exactly the same speed as the first test, both during warm-up and for the test itself. For both treadmill tests, the total treadmill time, maximum speed and mean heart rate achieved by the subject were documented. Post-exercise measures After exercise, leg volume, calf circumference and calf muscle flexibility were re-measured in the affected leg. Without access to their pre-exercise symptom scores, subjects were asked again to score the severity of individual symptoms in the affected leg using the same visual analog scales.

3 496 S. R. Kahn et al Statistical considerations Sample size estimate The primary comparisons were the differences in treadmill time, symptom severity, leg swelling and leg flexibility associated with exercise while wearing or not wearing ECS. An important second objective was to determine if the effects of compression s differed in patients with PTS. In the absence of data from previous studies, we considered a difference in means of one standard deviation to be clinically relevant. Since all of our study outcomes were continuous variables, we calculated that in order to achieve a two-sided type I error of 0.05 and type II error of 0.20 (power ¼ 0.80), we required 17 subjects. As the effects of ECS could be different in patients with and without PTS, we aimed to study enough subjects within each group to ensure adequate statistical power. Therefore, we planned to recruit at least 17 subjects with PTS and 17 subjects without PTS. Statistical analysis For comparison of baseline characteristics in subjects with vs. without PTS, Fisher s exact test of proportions or unpaired t- tests of means were used. For within-patient comparisons of measures in the vs. no session, paired t-tests of means were used. Results Study population Forty-three patients were recruited over a 1-year period ( ). Three patients were excluded (one had a history of bilateral DVT, one was not capable of following the research assistant s instructions, and one did not complete the second half of the study), leaving an evaluable study population of 40 subjects. Post-thrombotic syndrome was present in 19 [47.5%] of subjects, and was classified by the Villalta PTS scale as mild/ moderate in 17 and severe in 2. Age, sex distribution, site of index DVT and number of years since DVT diagnosis among subjects with and without PTS were similar (Table 1). At baseline, subjects with PTS had worse symptoms, higher leg volume and leg circumference, and significantly lower flexibility than subjects without PTS. Exercise protocol The mean treadmill speed at which subjects exercised was miles h 1, and the mean heart rate achieved was beats min 1, or 78% of the age-predicted maximum heart rate (calculated as 220 age). There were no differences in mean treadmill times in the no vs. sessions ( min vs min, P ¼ 0.94). Symptoms Overall, there were no statistically significant differences in exercise-induced changes in any of the leg symptoms between the and no sessions (Table 2). Most symptoms worsened slightly with exercise, whether or not ECS were worn, and whether or not PTS was present, but these changes were not statistically significant. We assessed whether the lack in overall change in symptoms could have been due to improvement in symptoms in some patients and aggravation of symptoms in others. First, we considered that the smallest change in symptom score that was clinically important was 0.5 cm on our 10-cm visual analog Table 1 Characteristics of study population at baseline PTS (n ¼ 19) No PTS (n ¼ 21) P-value Age [years, mean (SD)] 51.7 (12.3) 51.1 (14.7) 0.89 Female sex (%) 52.6% 33.3% 0.22 Proximal DVT (%) 78.9% 61.9% 0.24 Years since DVT diagnosis [mean (SD)] 2.4 (1.4) 2.0 (1.4) 0.33 PTS severity (n) Mild/moderate 17 Severe 2 Symptom severity [mean (SE)] Heaviness 1.88 (.53) 0.86 (.34) 0.11 Aching 1.94 (.49) 0.81 (.39) 0.08 Itching 0.41 (.38) 0.11 (.11) 0.45 Burning 0.62 (.29) 0.50 (.38) 0.82 Tingling 1.18 (.60) 0.07 (.07) 0.06 Restlessness 0.75 (.37) 0.60 (.43) 0.80 Swelling 2.68 (.50) 0.47 (.29) Throbbing 0.98 (.48) 0.41 (.36) 0.34 Leg volume (ml), mean (SE) 3016 (204) 2804 (113) 0.36 Leg circumference (cm), mean (SE) 23.1 (0.7) 22.5 (0.4) 0.41 Gastrocnemius flexibility (degrees), mean (SE) 20.6 (1.8) 27.7 (1.4) Soleus flexibility (degrees), mean (SE) 27.9 (1.4) 33.7 (1.4) DVT, deep venous thrombosis; PTS, post-thrombotic syndrome. Subjects rated the severity of individual symptoms using a 10-cm visual analog scale.

4 Table 2 Comparison of after-before exercise change in venous symptom severity ratings (VAS mean SE), no compression vs. compression sessions Subjects with post-thrombotic syndrome (n ¼ 19) Subjects without post-thrombotic syndrome (n ¼ 21) A randomized cross-over trial 497 Symptoms No compression Compression P-value No compression Compression P-value Heaviness Aching Itching Burning Tingling Restlessness Swelling Throbbing VAS, visual analog scale. Subjects rated their symptoms using a 10-cm visual analog scale. A positive value for change indicates that symptom was rated as worse after exercise compared to before exercise. scale [13]. Next, for each symptom, we determined the proportion of patients who reported improvement with ECS (i.e. had a 0.5 cm decrease in symptom severity), were unchanged with ECS (i.e. had a <0.5 cm change in symptom severity), or were worse with ECS (i.e. had a 0.5 cm increase in symptom severity). For all symptoms, most subjects had no change with ECS, 25% or fewer improved with ECS, and up to 25% of subjects worsened with ECS (Fig. 1). We performed a similar analysis to determine the proportion of patients who had global improvement, no change, worsening, or inconsistent change with ECS for the three most common symptoms (pain, heaviness and aching). Global improvement was defined as improvement in at least one symptom and no symptom worse, no change was defined as no change in any symptom, worsening was defined as at least one symptom worse and none improved, and inconsistent change was defined as some symptoms improved and others worse. Overall, 25% of subjects reported global improvement, 33% were worse, and 42% had inconsistent or no change. While 47% of subjects with PTS reported global worsening with ECS compared with 19% of subjects without PTS, due to the small numbers of patients in each category, we were unable to definitively establish whether certain subgroups of patients have a higher likelihood of experiencing improvement, or, alternately, worsening, of symptoms with ECS. Physiological measures Leg volume and circumference Leg volume (Fig. 2a) increased with exercise. However, there were no differences in the magnitude of this increase between the no vs. sessions, either for the population as a whole, or stratified according to PTS (no ECS, ml; ECS, ml; P ¼ 0.83 for difference) vs. no PTS (no ECS, ml; ECS ml; P ¼ 0.92 for difference). Leg circumference (Fig. 2b) also increased with exercise both in subjects with and without PTS. Although s did not significantly attenuate this increase in subjects with PTS (no ECS, gain of cm; ECS, gain of cm; P ¼ 0.57 for difference), subjects without PTS had less increase of circumference when exercising wearing s compared to no s (no ECS, gain of cm; ECS, gain of cm; P ¼ for difference). Flexibility There were no significant differences in change in flexibility with exercise in the no ECS vs. ECS sessions (Fig. 2c). These results were similar whether or not PTS was present. Discussion Fig. 1. Proportion of patients whose symptoms after-before exercise improved, were unchanged, or worsened with ECS. Symptom severity was rated using a 10-cm visual analog scale. Improvement was defined as a 0.5 cm decrease in symptom severity, no change was defined as a <0.5 cm change in symptom severity, and worsening was defined as a 0.5 cm increase in symptom severity. Based on the results of our study, a custom-fitted, 30 mmhg knee-length elastic compression worn on the affected leg during treadmill exercise by patients with previous DVT does not produce a measurable benefit, acutely, on exerciseinduced changes in leg volume, flexibility, or venous symptoms. These findings were true in patients with and without PTS. In

5 498 S. R. Kahn et al Fig. 2. Comparison of after-before exercise change in leg volume (a), circumference (b), and flexibility (c), in the no compression vs. compression exercise sessions (mean SE), in subjects with PTS (n ¼ 19) and without PTS (n ¼ 21). Open bars, no compression session; shaded bars, compression session. In subjects with PTS, comparing no compression vs. compression exercise sessions for exercise-induced change in leg volume, leg circumference, gastrocnemius flexibility and soleus flexibility, P ¼ 0.83, 0.57, 0.35 and 0.22, respectively. In subjects without PTS, for these comparisons P ¼ 0.92, , 0.94 and 0.07, respectively. subjects without PTS only, we found that use of s resulted in a small but statistically significant attenuation of exercise-induced increase in leg circumference. This probably reflects the play of chance, since in this group we did not find a similar attenuation of exercise-induced increase in leg volume, a measure considered to more accurately reflect leg swelling. Examination of individual patient data showed that for any symptom, 25% or fewer patients reported improvement with ECS, and up to 25% reported worsening. Only 25% of patients reported global improvement in the three most common venous symptoms, while 33% were globally worse. We were not able to identify subgroups of patients more likely to experience improvement, or worsening, of symptoms with ECS. Despite the vast amount of research and resulting consensus statements on the prevention and management of acute VTE [14,15], there has been a paucity of studies of the longer-term management of patients with DVT, 20 50% of whom go on to develop PTS [1,2,6]. Our findings are relevant to the longerterm management of patients with DVT. It is common practice to routinely prescribe ECS after DVT. This practice is based primarily on a single, unblinded, randomized trial, which showed that daily use of ECS for at least 2 years after DVT resulted in a reduction, by about 50%, of the incidence of PTS [2]. Several small non-blinded studies have also shown that use of ECS appears to improve venous symptoms [4,5]. In contrast, a recent randomized placebo controlled trial was unable to demonstrate that daily use of ECS was beneficial in preventing (mean follow-up 5 years) or treating (mean follow-up 2 years) PTS, compared to the use of sham s [6]. Compression s are expensive, difficult to apply, and can be warm, uncomfortable, and itchy. Because of the continued uncertainty regarding the role of s after DVT, further research on their short-term and long-term effectiveness is needed. Our results suggest that they are not of benefit to DVT patients when worn during exercise, whether or not PTS is present. To our knowledge, only one other study has examined the effect of ECS on post-thrombotic legs during exercise. O Donnell and colleagues studied 11 patients with venographic evidence of deep venous damage and eight historical controls [16]. Information was not provided as to whether patients had symptomatic PTS, or when prior DVT had occurred. Venous pressures were recorded while upright and during maximal heel raising, then were repeated after an elastic was applied to the affected limb. They found that elastic compression led to lower pressure swings with exercise, which were similar in amplitude to normal controls, and reduced maximum systolic pressure during exercise. However, compression had no effect on resting venous pressure (significantly higher than normal controls) or postexercise time to return to baseline pressure (significantly faster than normal controls). It is difficult to compare their findings to ours, since we performed a more functional test of walking/jogging rather than heel lifts. Also, in their study, symptoms, volume and flexibility were not assessed, hence it is not known which, if any, of their measures best correlate with symptoms or swelling. Our study has several limitations. First, the effect of ECS was measured acutely during a single exercise session. Hence we are unable to comment on the potential impact of regular, longerterm use of ECS on exercise-induced symptoms, calf flexibility, and swelling. Second, no objective, gold standard test to diagnose PTS exists [3], however, we used a validated clinical scale to diagnose and establish the severity of PTS. This scale has been used in prospective studies of PTS [1,2]. Nevertheless, it is possible that misclassification of diagnosis or severity of PTS may have occurred for some subjects. However, this would be unlikely to change our findings, since we found no differences in the ECS vs. no ECS sessions in subjects with vs. without PTS. Third, our study was not blinded (e.g. by use of a sham ), however, patients were blinded to the study hypothesis to avoid influencing their symptom ratings, and all

6 A randomized cross-over trial 499 physiologic measures were performed using objective instruments by observers blinded to PTS status, thus reducing the likelihood of introducing differential measurement bias. Further, this bias might be expected to favor s since patients would tend to believe an active intervention (ECS) would be of benefit. Finally, our study population was a convenience sample of patients attending our Thrombosis Clinic who agreed to participate in this study and who had DVTat least one year previously, hence we may have selected for less severely affected patients, since those with more severe symptoms at baseline or symptoms that worsen with exercise might have declined participation in the study. We are therefore unable to speculate on whether ECS might be of benefit during exercise in patients with more severe forms of PTS, or in the first year after DVT is diagnosed. In conclusion, in patients with prior DVT, use of ECS on the affected leg during treadmill exercise had no beneficial effects on exercise-induced venous symptoms, leg swelling or flexibility, whether or not PTS was present. Whether certain subgroups of DVT patients might experience symptom improvement from use of ECS during exercise, or whether longer term use of ECS could improve exercise-induced symptoms merit further study. Acknowledgements Drs Kahn and Shrier are Clinical Research Scientists supported by the Fonds de Recherche en Santé du Québec. This study was supported by an unrestricted grant-in-aid from the Beiersdorf-Jobst Research Program of the American College of Phlebology. References 1 Prandoni P, Lensing A, Cogo A, Cuppini S, Villalta S, Carta M, Cattelan A, Polistena P, Bernardi E, Prins M. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med 1996; 125: Brandjes DPM, Buller HR, Heijboer H, Hulsman MV, de Rijk M, Jagt H. Randomized trial of effect of compression s in patients with symptomatic proximal-vein thrombosis. Lancet 1997; 349: Kahn SR, Ginsberg JS. The post-thrombotic syndrome: current knowledge, controversies, and directions for future research. Blood Rev 2002; 16: Pierson S, Pierson D, Swallow R, Johnson GJ. Efficacy of graded elastic compression in the lower leg. JAMA 1983; 249: Jones NA, Webb PJ, Rees RI, Kakkar VV. A physiological study of elastic compression s in venous disorders of the leg. Br J Surg 1980; 67: Ginsberg J, Hirsh J, Julian J, Vander LaandeVries M, Magier D, MacKinnon B, Gent M. Prevention and treatment of postphlebitic syndrome: results of a 3-part study. Arch Intern Med 2001; 161: Villalta S, Bagatella P, Piccioli A, Lensing A, Prins M, Prandoni P. Assessment of validity and reproducibility of a clinical scale for the post-thrombotic syndrome (Abstract). Haemostasis 1994; 24: 158a. 8 Mair SD, Seaber AV, Glisson RR, Garrett WE Jr. The role of fatigue in susceptibility to acute muscle strain injury. Am J Sports Med 1996; 24: Berard A, Kurz X, Zuccarelli F, Ducros JJ, Abenhaim L. Reliability study of the leg-o-meter, an improved tape measure device, in patients with chronic venous insufficiency of the leg. Angiology 1998; 49: Boone DC, Azen SP, Lin CM, Spence C, Baron C, Lee L. Reliability of goniometric measurements. Phys Ther 1978; 58: Gajdosik RL, Bohannon RW. Clinical measurement of range of motion. Review of goniometry emphasizing reliability and validity. Phys Ther 1987; 67: NIH Consensus Statement. In: Leon, AS, eds. Physical Activity and Cardiovascular Health a National Consensus Champaign III. Human Kinetics 1997; Guyatt GH, Juniper EF, Walter SD, Griffith LE, Goldstein RS. Interpreting treatment effects in randomised trials. Br Med J 1998; 316: Hyers TM, Agnelli G, Hull RD, Morris TA, Samama M, Tapson V, Weg J. Antithrombotic therapy for venous thromboembolic disease. Chest 2001; 119: 176S 93S. 15 Geerts WH, Heit JA, Clagett GP, Pineo GF, Colwell CW, Anderson FA Jr, Wheeler HB. Prevention of venous thromboembolism. Chest 2001; 119: 132S 75S. 16 O Donnell TFJ, Rosenthal DA, Callow AD, Ledig BL. Effect of elastic compression on venous hemodynamics in postphlebitic limbs. JAMA 1979; 242: