INDIVIDUALS WHO HAVE SUFFERED from poliomyelitis. Dynamic Water Exercise in Individuals With Late Poliomyelitis

Transcription

1 66 Dynamic Water Exercise in Individuals With Late Poliomyelitis Carin Willén RPT, MSc, Katharina Stibrant Sunnerhagen, MD, PhD, Gunnar Grimby, MD, PhD ABSTRACT. Willén C, Stibrant Sunnerhagen K, Grimby G. Dynamic water exercise in individuals with late poliomyelitis. Arch Phys Med Rehabil 2001;82: Objective: To evaluate the specific effects of general dynamic water exercise in individuals with late effects of poliomyelitis. Design: Before-after tests. Setting: A university hospital department. Participants: Twenty-eight individuals with late effects of polio, 15 assigned to the training group (TG) and 13 to the control group (CG). Intervention: The TG completed a 40-minute general fitness training session in warm water twice weekly. Assessment instruments included the bicycle ergometer test, isokinetic muscle strength, a 30-meter walk indoors, Berg balance scale, a pain drawing, a visual analog scale, the Physical Activity Scale for the Elderly, and the Nottingham Health Profile (NHP). Main Outcome Measures: Peak load, peak work load, peak oxygen uptake, peak heart rate (HR), muscle function in knee extensors and flexors, and pain dimension of the NHP. Results: The average training period was 5 months; compliance was 75% (range, 55 98). No negative effects were seen. The exercise did not influence the peak work load, peak oxygen uptake, or muscle function in knee extensors compared with the controls. However, a decreased HR at the same individual work load was seen, as well as a significantly lower distress in the dimension pain of the NHP. Qualitative aspects such as increased well-being, pain relief, and increased physical fitness were reported. Conclusions: A program of nonswimming dynamic exercises in heated water has a positive impact on individuals with late effects of polio, with a decreased HR at exercise, less pain, and a subjective positive experience. The program was well tolerated (no adverse effects were reported) and can be recommended for this group of individuals. Key Words: Poliomyelitis; Exercise test; Water; Rehabilitation by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation INDIVIDUALS WHO HAVE SUFFERED from poliomyelitis earlier in life often later complain of new symptoms related to their illness. 1-4 The most common new symptoms are From the Departments of Rehabilitation Medicine, Göteborg University, Göteborg, Sweden. Accepted in revised form February 27, Supported by the Swedish Medical Research Council (project no. 0388), the Swedish Association for Traffic and Polio Disabled, and the Greta and Einar Asker Foundation. The authors have chosen not to select a disclosure statement. Reprint requests to Carin Willén, Dept of Rehabilitation Medicine, Sahlgrenska University Hospital, S Göteborg, Sweden, carin.willen@fhs.gu.se /01/ $35.00/0 doi: /apmr fatigue, muscle and joint pain, and weakness leading to difficulties in activities of daily living, including physical mobility. These new symptoms, when fulfilling defined criteria, are termed postpoliomyelitis syndrome (PPS). 5 Evidence of deconditioning in polio survivals has been reported by Owen, 6 and Stanghelle et al, 7 and recently by Willén et al. 8 Thus the potential to improve physical performance by training should be explored. A few studies have demonstrated positive effects on muscle strength and general physical capacity from physical exercise in individuals with late effects of polio. Einarsson 9 reported increased muscle strength of the quadriceps muscle after a short-term resistance training program. In the study by Fillyaw et al, 10 isometric strength in exercised muscles increased gradually over a 2-year treatment period. The aerobic response to exercise was investigated by Jones et al. 11 They studied exercise at an intensity of 70% to 75% of maximum on cycle ergometer for 16 weeks, and found a conventional response to exercise as would be observed among the nondisabled. A 6-month study of a general training program with a focus on endurance but including also submaximal resistance training was done by Ernstoff et al, 12 who found performance improvement in an exercise test and strength improvement in some muscle groups. A modified program of treadmill walking has been found to reduce energy cost after the training period. 13 Studies on the effects of nonswimming exercise in water are sparse. In a study by Ruoti et al, 14 the effects of water exercises in older healthy adults were investigated. They found that exercising in water for 12 weeks had a positive effect on cardiorespiratory function and on muscle endurance in the shoulder muscles compared with controls. Studies among patient groups are reported in the rheumatology literature. Danneskiold-Samsoe et al 15 showed, in 7 patients with rheumatoid arthritis who trained in heated water for 2 months, that muscle strength in the knee extensors increased in isometric performance and in low isokinetic velocities (30 /s, 60 /s). An improvement in aerobic capacity was also reported. The exercise program, which was combined with individual resistance training, focused mainly on leg exercises; no adverse side-effects were observed. In a controlled study by Stenström et al, patients with rheumatoid arthritis trained for 4 years in heated water. The training group improved their hand grip strength and showed a higher activity level after the training period compared with a control group. However, there were no differences between the groups in different functional tests or in pain ratings. Exercising in water is commonly recommended for individuals with late effects of polio. The water provides resistance but minimizes biomechanical stress on muscles and joints, which is important for these individuals when exercising. However, no study has reported what training response can be expected and how this type of exercise benefits the individual. Accordingly, the present study attempts to identify specific effects of general dynamic water exercises in individuals with late polio. The training was offered as a complement to ordinary physiotherapy routines at the polio clinic, which provided a 1-hour consultation with a physiotherapist from the polio

2 DYNAMIC EXERCISE IN POLIO, Willén 67 team. This consultation was aimed at providing information and advice on physical activity and to determine the need of further physical therapy interventions. METHODS Subjects Thirty individuals with late effects of polio were included in the study, which was approved by the Ethics Committee of the Göteborg University Faculty of Medicine. Subjects were asked whether they wanted to participate and were investigated by a physician at the polio clinic to confirm that there were no contraindications from other disorders such as severe cardiovascular diseases. The presence of polio-affected muscles was determined according to each patient s own history and by electromyogram (EMG) recordings showing typical neurogenic changes in at least 3 to 5 relevant muscle groups. The limbs and trunk of each individual were then classified using the National Rehabilitation Hospital post-polio limb classification (fig 1), where class I is no clinical polio, class II is subclinical polio, class III is clinically stable polio (no new or increased muscle weakness), class IV is clinically unstable polio (increased muscle weakness), and class V is severely atrophic muscles. 5 All Fig 1. Polio classification of the training group and the control group using the National Rehabilitation Hospital post-polio limb classification. 5 participants were community ambulators with or without devices. Of this group, 15 individuals (7 men, 8 women) became the training group (TG) and 15 (9 men, 6 women) the control group (CG). The individuals assigned to the CG were those who, for practical reasons, were unable to participate the required 2 times weekly and at the time set for the exercise sessions. Members of the CG were assured the opportunity to join the TG at the end of the study period. No one in either group received individual physical therapy during the study. Two of the individuals in the CG did not complete the second test measurements; 1 man underwent a planned operation and 1 woman was prevented by family problems. One individual in the TG was on beta-blocking agents for moderate hypertension. The mean age was 51 years (range, yrs) for the TG and 49 years (range, yr) for the CG. The mean age standard deviation at the onset of disease was 4 3 years for both the TG and CG, and the average stable phase lasted for approximately 35 8 years in the TG and 34 7 years in the CG. Thirteen individuals in the TG and 12 in the CG fulfilled the criteria for PPS: a prior history of polio; typical findings on EMG; neurologic recovery followed by an extended period of stability of at least 20 years; onset of new or increased muscle weakness and also other symptoms as fatigue and muscle and joint pain; and exclusion of other conditions that might cause the new health problems. 5 The remaining subjects also had late effects of polio (verified diagnosis with remaining symptoms), but not fulfilling the above criteria of PPS due to no progressive muscle weakness. Assessment Instruments All tests before and after the training period (exercises in water) were supervised by an unbiased investigator. Oxygen uptake and anaerobic threshold. The peak oxygen uptake (V O2 ) and anaerobic threshold were tested on a bicycle ergometer. a The exercise started at 0 and continued with a 10W stepless increase every minute. Before each test, the individual became familiarized with the equipment and was asked to breathe through the mask for 3 minutes. This was followed by a 3-minute period of pedaling without friction resistance before the exercise test started at 0W. The test was supervised by a physiotherapist, and a physician was present in the laboratory. Each individual performed a maximum test, ie, was verbally encouraged to continue as far as he/she could. Perceived effort was rated on the Borg category scale (Borg CR scale), with exponential increments from 0 to In some cases, the individual had to wear a brace or be strapped around the foot to be able to perform the exercise test. Gas exchange was recorded during bicycle exercise by measuring the expired gas flow and expiratory oxygen and carbon dioxide concentrations with gas meters, which were calibrated against gas mixtures of known concentrations before each test. Oxygen consumption and carbon dioxide production were measured continuously at standard temperature and pressure. The anaerobic threshold was defined as the point at which the rise in carbon dioxide production was higher than the rise in oxygen consumption. 18 This was determined by a specially designed computer. All variables from the bicycle ergometer test were compared with reference values from a normal population. 19 Heart rate. Heart rate (HR) was recorded from a 3-lead echocardiogram tracing, and blood pressure was measured with a sphygmomanometer once every minute during the test. HR at the watt level prior to the peak work level in the first test (before the exercise program) was compared with the same

3 68 DYNAMIC EXERCISE IN POLIO, Willén level in the second test. If exactly the same watt level was not used in the second test, the HR was extrapolated between the value at the levels above and below that level. Muscle strength, walking, and balance. Muscle strength for knee-extension and flexion was measured using the KIN- COM dynamometer. The individual was seated, back against the seat. A seatbelt was strapped around the waist and the thigh to avoid undesired movements. Before the test, the patient warmed up for 5 minutes on a bicycle ergometer. Peak isometric strength was measured at a 60 knee angle in extension and flexion. Maximal isokinetic strength was measured at a velocity of 60 /s and 180 /s for concentric muscle actions. Isometric endurance was measured as the time the individual could keep 40% of his/her isometric peak torque in an extension at a 60 knee angle. All measurements were performed by both legs if possible. Values are presented as the individual s strong leg (the stronger of the 2 legs) and weak leg, and compared with values from a normal population (unpublished data). Spontaneously chosen and maximal walking speeds were measured for 30 meter indoors. The test started with walking at the spontaneously chosen speed. The values were compared with those from a normal population (unpublished data). Balance was assessed with Berg balance scale. 20 This instrument consists of 14 items that require the subject to maintain positions of varying difficulty. Each item is graded 0 to 4, and the maximum score is 56. Quality of life. Pain characteristics and location of pain were assessed by pain-drawing in the form of a human figure. A fixed set of symbols was selected that described types of pain such as aching, burning, numbing, cutting/throbbing, shooting, and cramping pain. The subject was asked to draw these symbols on the figure in all the areas where pain was felt. The intensity of experienced pain when present was measured by means of a horizontal visual analog scale (VAS) ranging from 0 to 100, with no pain on the left and the worst possible pain on the right. 21 Physical activity level was measured by means of the Physical Activity Scale for the Elderly (PASE). 22 This instrument is comprised of self-reported occupational, household, and leisure activity items over the most recent 1-week period. The PASE score is calculated from weights and frequency values, and the maximal score is 360 points. The individuals filled in the Nottingham Health Profile (NHP), a self-administered questionnaire. 23 The instrument consists of 2 parts, but only the first part was used in this study. This part has 6 dimensions assessing distress in emotional reaction, sleep, energy, pain, physical mobility, and social isolation with the answers yes or no, for a total of 38 different statements. Each answer is multiplied by a specific weight, and the weighted means for each dimension are calculated. Zero indicates no problem in a dimension and 100 the highest level of distress. Training The training period lasted for 8 months. A 3-week break was taken at Christmas time. The subjects were introduced into the group in turns. Training sessions were held twice weekly in warm water ( 33 C) and subjects paid the ordinary fee for group training in the clinic. The 40-minute training program was led by a physiotherapist and was accompanied by music. It was designed to train general physical fitness including resistance and endurance activities, balance, stretching, and relaxation (table 1). The participants were told to pace the exercises at the intensity level where muscle fatigue was not present during or after the training session, including the night after. To form an idea of the intensity of the program, a HR monitor with an electrode worn around the wrist was used once on 5 randomly chosen participants after a few weeks of training. When the training period was over, the subjects in the TG were asked to write in their own words what the training period had meant to them. Statistical Methods Conventional formula were used for calculations of mean, median, and standard deviations. Nonparametric tests were used for analyses. Wilcoxon s signed-rank test was used to test differences within the groups and the Mann-Whitney U test was used to test differences between groups. A significance level of p.05 was used throughout the study. RESULTS The training program was tolerated well, and all participants completed the training period. The average training time was 5 months, and the mean value for compliance expressed as a percentage of possible training opportunities during this period was 75% (range, 55 98). The responses of the measurements before and after training in the TG and the CG are presented in table 2. Except for anaerobic threshold in relation to predicted max V O2 where TG had a higher mean value than CG, no significant differences between the 2 groups were found in the pretraining measurements (table 2). Oxygen uptake and anaerobic threshold. Three examples of the variation in HR during the exercise session are shown in figure 2. The peak values during the training are compared with each individual s peak value from the exercise test on the bicycle ergometer before training. The mean HR during the session for these 3 individuals was around 75% of their peak values at the bicycle test, and in some exercises they exerted themselves rather close to or over the peak HR value at the test. There were no differences after training in peak load, peak V O2, and peak HR between the 2 groups. There was, however, a significantly larger reduction in HR at the same watt level in the TG compared with the CG (p.05) (fig 3). The average value was beats/min before the training period and beats/min after. Corresponding values for the CG were and beats/min, respectively. Seven individuals in the TG had a reduction in HR of 10 beats, as compared with 2 in the CG. A significant difference (p.05) was found in the anaerobic threshold in the CG, where the relative value was 46% 15% after training, compared with 35% 13% before training. No change was seen in the TG. The peak respiratory exchange ratio had a higher mean value in the TG after the training period as compared with the CG (p.01). Muscle strength, walking, and balance. No significant differences between the groups were found after training in any of the measured variables of muscle function (table 2). The small reduction in knee extensor muscle strength in the TG was mainly due to 1 subject who had a relatively large and unexplained reduction in strength; were this subject excluded, no significant difference would remain in the TG. The values for extension in the stronger leg in both groups were around 65% of normal values, and for flexion 75% of normal values. In the weaker leg, values for extension were 40% of normal values and 50% for flexion in the TG. Corresponding values for the CG were 50% for both extension and flexion. The 30-meter walking speed showed no significant difference at spontaneous or maximal velocity between the TG and the CG after the training period (table 2). There were no significant changes with training in the results of the balance test nor for the TG or the CG. The median value

4 DYNAMIC EXERCISE IN POLIO, Willén 69 Table 1: The Exercise Program* 1. Standing: Alternating unilateral shoulder lifting followed by walking in place. Arms stretched out at surface in front of the body bilateral arm movements forward and backward followed by walking in place. Dog swim as fast as possible. 2. Standing: Jog in place with arm movements in alternating directions. Change tempo. 3. Standing: Arms stretched out at surface in abduction, inward and outward rotations, arm crossing in front of the body in wide circles. Arms stretched out at surface in front of the body, float up and down, change between long and short lever. Bilateral armswing backward and forward. Arms stretched out at surface in abduction, float up and down, change between long and short lever. Arms crossing in front of the body in wide circles. 4. Standing (hold on to the bar if needed): Alternating left and right, lift 1 leg with knee bent, stretch the knee bend back and lower the leg. Small quick abduction movements of the leg, alternating left and right. Step forward feet together step backward, alternating right and left leg. 5. Bicycling: In supine position, alternating with paddling arm movements in front of the body. 6. Standing: Body twisting with arms stretched out in abduction at surface. Jump up and down in the water. Jog in place. Repeat the above until the music stops. 7. Standing: Breast stroke exercises with the arms. Alternating unilateral arm movements in front of the body draw an lying 8. Arms stretched out at surface in abduction, rotate the trunk to right and left respectively. Breast stroke exercises. Shake the shoulders. Hands on the hips, unilateral by putting the hand at the back of the neck and down at the hip again, end with the same exercise but bilateral. 8. Standing (hold on to the bar if necessary): Swing the straight leg in front of the body and back in abduction alternating right and left. Draw a circle with the foot alternating with the left and right leg. Walk sideways, first to the right and then to the left. 9. Standing (holding a ball in the hands): Stretch forward pump the ball up and down, changing different directions. Pump the ball around the body, change direction. Pump the ball with 1 hand from left to right, change hand in front of the body, use big sweeping movements around the body. 10. Walking: Walk forward and backward in the water as fast as possible. Jog in place with arm movements. Walk sideways, first to one side then to the other, as fast as possible. Body twist with arm movements. Tramp water. Arms stretched above head, sway from side to side. Jog in place. End with a vertical jump. 11. Slow walking: Slow arm movements unilaterally and bilaterally. Stretch 1 arm upward and the other downward, change. Stretch by bending the trunk to the side. Walk slowly. Slow arm movements. Stretch the back by huddling. Stretch arms forward and backward on the surface. 12. Stretching exercises: Stretch back extensors, hip flexors, knee flexors, extensors, and plantar flexors. End with breathing deeply, exhaling through pursed lips. * Performed for 5 months, twice weekly. Move around without having the feet on the bottom of the pool. was the same in the CG: 56 points before and after training; and in the TG: 55 and 56, respectively (table 2). Quality of life. The physical activity level measured by PASE showed a great variation among the individuals in both groups, but there was no significant difference between the 2 groups. The pain drawings demonstrated few variations in the 2 groups, indicating a relatively similar picture for the individuals before and after the training period. Three individuals in the CG marked a cramping pain as a new sensation in the second test; and 1 individual in the TG marked no cramping pain after the training period, as compared with before the training. No change in the intensity of pain measured by VAS was observed after training compared to before in either of the groups (table 2). The median values in the TG and CG in the 6 different dimensions in NHP before and after training are shown in figure 4. In the dimension of pain, the TG showed significantly lower distress after the training period compared with the CG (p.05). The median value decreased from 37 (range, 0 100) to 18 (range, 0 100) in the TG; in the CG, there was a small increase from 41 (range, 0 82) to 46 (range, 0 100). Three different qualitative aspects could be identified from the individuals own description of what the training had meant to them: feelings of increased well-being, pain relief, and increased physical fitness. DISCUSSION A program of nonswimming dynamic exercises in heated water can have a positive impact among individuals with late effects of polio as shown in this study. A decreased HR at the same watt level after training was seen, as well as lower distress in the dimension of pain, measured by the NHP. The design of our study, in which those participants who at the time of the study could fit the training program into their schedules became the study group (TG), may have influenced the results. However, the controls were also interested in training but chose for different practical reasons not to participate. The polio classification showed that the TG were more affected in the trunk and in the upper extremities compared with the CG. This may have affected the ability to train and influenced the outcomes of the training. The classification also showed that the affect of polio varied considerably among participants, hence, individuals intensity of exercise probably varied during the training sessions and may also have affected the training response differently. There were no indications that the training program had been too hard or had caused any harm to participants. The individuals own experience of the training was positive, compliance with the regimen was good, and no one discontinued the training period. The pain drawings made after the training period gave no indication that the individuals experienced new pain sensations as a result of overtraining. No one showed any substantial reduction in performance on the muscle or bicycle tests. The HR at a submaximal load in the bicycle exercise was reduced after training. The extrapolating of the HR was based on the fact that there is a linear relationship between HR and intense of work. 24 The lowered HR indicates an effect on

5 70 DYNAMIC EXERCISE IN POLIO, Willén Table 2: Values From Before and After Training* Training Group (n 15) Control Group (n 13) Before After Difference Before After Difference Peak load (W) Peak load (% of pred) Peak V O2 (ml/kg/min) Peak V O2 (% of pred max) Peak HR (beats/min) AT (% of pred max V O2 ) Respiratory exchange ratio Perceived exertion, Borg CR scale (0 10) Stronger Leg Knee extension 60 (Nm) Knee extension 60 /s (Nm) Isometric endurance 40% max (sec) Knee flexion 60 (Nm) Knee flexion 60 /s (Nm) Weaker Leg Knee extension 60 (Nm) Knee extension 60 /s (Nm) Isometric endurance 40% max (s) Knee flexion 60 (Nm) Knee flexion 60 /s (Nm) Self-chosen walking speed (m/s) Maximal walking speed (m/s) Berg balance scale, points (max 56) 55 (24 56) 56 (37 56) 0 ( 1 to 13) 56 (43 56) 56 (43 56) 0 ( 5 to8) PASE, points (max 360) 130 (30 297) 122 (30 340) 4 ( 171 to 57) 135 (27 289) 130 (50 302) 12 ( 206 to 165) Pain assessment, VAS (0 100) 66 (9 93) 53 (7 91) 3.5 ( 47 to 51) 54 (30 90) 58 (27 86) 2 ( 55 to 21) * Mean values SD are presented; medians are presented for the Berg balance scale and PASE; pain assessment for the VAS. Denotes a difference between TG and CG in the before training measurements, p.05. Denotes a difference between TG and CG in the before and after training measurements, p.05. Denotes a difference between the before and after training measures, p.05. Abbreviations: % of pred percentage of predicted; max maximum. circulatory regulatory mechanism; it was not reflected in the variables of peak load and peak V O2, which remained unchanged. Perhaps for many of the participants, exercise program design had an impact mainly on muscle factors, with a better capacity to use oxygen in the exercising muscles thereby influencing the central circulatory system. 25 This so-called peripheral training model has been shown to decrease HR during exercise in healthy subjects. 26 It may be difficult to stress central circulatory demand for many individuals with impaired muscle function. However, as seen in the recorded Fig 2. The variation in HR, mean HR, and range during the training session in 3 individuals with late effects of polio. The peak HR in each individual at the bicycle test is marked as a reference. Fig 3. The HR at the same work level in the TG and the CG before and after the training period.

6 DYNAMIC EXERCISE IN POLIO, Willén 71 HR response to the training program, the relative work was rather high. The significantly lower anaerobic threshold in the CG at the first measurement with an increase at the second measurement may reflect variation in use of the polio-affected muscles in the bicycle exercise test, because no other variable from the exercise test was altered. The present results demonstrate limitations in anaerobic threshold value as a measure of cardiorespiratory function in subjects with impaired muscle function. There was no significant difference between the 2 groups in the change in muscle strength or muscle endurance between the 2 measurements. Training twice a week may not be frequent enough to achieve better muscle performance in the thigh muscles among individuals with late effects of polio. The studies in which improvements in muscle function and aerobic capacity were shown were designed for training 3 or more times a week This was not possible in our study for practical reasons, and because of the need for greater recovery time after muscle exercises in postpolio subjects with new muscle weakness, twice weekly may be an optimal level. 27 Moreover, in studies in which muscle strength is improved after a training period, the exercises have been designed more directly toward this goal. This was the case in the studies by Einarsson 9 and Fillyaw et al, 10 in which exercises were directed toward the knee extensors. The exercise program in our study had more a general body conditioning design, and thus failure to show improvements in muscle performance may not be surprising. Improvements in balance could not be shown with the Berg balance scale. We know that individuals with late effects of polio have problems with postural control primarily as a result of muscle weakness. The instrument used was not sensitive enough for this group of individuals because most scored almost maximal points when tested. Increasing capacity on the disability level is important for individuals with impairments. This study could not show a significant difference in improvement in walking 30 meters when the TG and the CG were compared, but many individuals in the TG increased maximal walking speed as compared with the CG (fig 5). A greater difference between the self-chosen and maximal walking speed means that the potential to experience fatigue in daily activities involving walking may decrease, because the individual can act at a relatively lower level. Fig 4. Median values and 25th and 75th percentiles for the 6 dimensions in the NHP. * p <.05 between the TG and the CG after training. ** p <.01 denotes difference within the TG. Fig 5. Maximal walking speed (m/s) in the TG and the CG before and after the training period. Pain is a central symptom among individuals with late effects of polio. If the experience of pain can be affected, this may lead to less concern about performing daily activities. The training did not have an impact on the experienced intensity of pain when pain was present, but rather on the experience of pain in different situations as they are assessed with the items in the NHP. CONCLUSION The role of exercise in the management of individuals with late effects of polio has been discussed, and it has been questioned whether strength or aerobic exercises have a deteriorating effect on the individual. 28 The training in the present study seems to have had a positive functional impact, giving the individuals a subjective positive experience, less pain, lower HR at a submaximal work level, as well as no adverse effects. Acknowledgments: The authors thank Åsa Cider for supervising the bicycle tests and Marita Hedberg for supervising the muscle tests. References 1. Halstead LS, Rossi CD. New problems in old polio patients: results of a survey of 539 polio survivors. Orthopedics 1985;8: Agre JC, Rodriquez AA, Sperling KB. Symptoms and clinical impressions of patients seen in a postpolio clinic. Arch Phys Med Rehabil 1989;70: Westbrook MT. A survey of post-poliomyelitis sequelae: manifestations, effects on people s lives and responses to treatment. Aust J Physiother 1991;37: Stanghelle JK, Helseth R, Roaldsen KS. [42 patients with postpolio syndrome. A retrospective study from Sunnaas hospital] [Norwegian]. Tidsskr Nor Laegeforen 1991;111: Gawne CA, Halstead LS. Post-polio syndrome: pathophysiology and clinical management. Crit Rev Phys Rehabil Med 1995;7: Owen RR. Polio residuals clinic: conditioning exercise program. Orthopedics 1985;8: Stanghelle JK, Festvåg L, Aksnes A. Pulmonary function and symptom-limited exercise stress testing in subjects with late sequelae of poliomyelitis. Scand J Rehabil Med 1993;25: Willén C, Cider Å, Stibrant-Sunnerhagen K. Physical performance in individuals with late effects of polio. Scand J Rehabil Med 1999;31: Einarsson G. Muscle conditioning in late poliomyelitis. Arch Phys Med Rehabil 1991;72:11-4.

7 72 DYNAMIC EXERCISE IN POLIO, Willén 10. Fillyaw MJ, Badger GJ, Goodwin GD, Bradley WG, Fries TJ, Shukla A. The effects of long-term non-fatiguing resistance exercise in subjects with post-polio syndrome. Orthopedics 1991;14: Jones DR, Speier J, Canine K, Owen R, Stull A. Cardiorespiratory responses to aerobic training by patients with postpoliomyelitis sequelae. JAMA 1989;261: Ernstoff B, Wetterqvist H, Kvist H, Grimby G. The effects of endurance training on individuals with post-poliomyelitis. Arch Phys Med Rehabil 1996;77: Dean E, Ross J. Effect of modified aerobic training on movement energetics in polio survivors. Orthopedics 1991;14: Ruoti RG, Troup JT, Berger RA. The effects of nonswimming water exercises on older adults. J Orthop Sports Phys Ther 1994; 19: Danneskiold-Samsoe B, Lyngberg K, Risum T, Telling M. The effect of water exercise therapy given to patients with rheumatoid arthritis. Scand J Rehabil Med 1987;19: Stenström CH, Lindell B, Swanberg E, Swanberg P, Harms- Ringdahl K, Nordemar R. Intensive dynamic training in water for rheumatoid arthritis functional class II a long-term study of effects. Scand J Rheumatol 1991;20: Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 1982;14: Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol 1986; 60: Wasserman K, Hansen JE. Principles of exercise testing and interpretation. Philadelphia: Lea & Febiger; Berg K, Wood-Dauphinee S, Williams JI. The balance scale: reliability assessment with elderly residents and patients with an acute stroke. Scand J Rehabil Med 1995;27: Carlsson AM. Assessment of chronic pain. Aspects of the reliability and validity of visual analogue scale. Pain 1983;16: Washburn RA, Smith KW, Jette AM, Janney CA. The physical activity scale for the elderly (PASE): development and evaluation. J Clin Epidemiol 1993;45: Wiklund I. The Nottingham Health Profile a measure of healthrelated quality of life. Scand J Prim Health Care 1990;1: Åstrand P-O, Rodahl K. Textbook of work physiology. 3rd ed. Singapore: McGraw-Hill; Saltin B. Physiological adaptation to physical conditioning. Acta Med Scand 1986;711 Suppl: Gaffney AF, Grimby G, Danneskiold-Samsoe B, Halskov O. Adaptation to peripheral muscle training. Scand J Rehabil Med 1981;13: Agre JC, Rodriquez AA, Franke TM. Subjective recovery time after exhausting muscular activity in postpolio and control subjects. Am J Phys Med Rehabil 1998;77: Dean E. Clinical decision making in the management of the late sequealae of poliomyelitis. Phys Ther 1991;71: Supplier a. Chattanooga Group, Inc, 4717 Adams Rd, Hixson, TN