Effects of group kinesiotherapy on primary and secondary symptoms of osteoarthrosis

. The aim of this study was to investigate the influence of a specific, kinesiotherapy-based rehabilitation program on the various symptoms of osteoarthrosis (OA), following group treatment. Thirty-one individuals, of both sexes, aged over 50 years and with medical diagnosis of OA, underwent 16 sessions, twice a week, totaling eight weeks, of a specific rehabilitation protocol based on group kinesiotherapy. Primary OA symptoms were assessed (directly related to the disease: OA symptoms, trunk flexibility, balance and pain), and so were secondary ones (indirectly related to the disease: signs of depression and anxiety, and quality of life). Data were tested through Student's t test or Wilcoxon's test, and contingencies of categorical data were analyzed using McNemar's test. There was an improvement in all primary symptoms of OA after the kinesiotherapy protocol was applied. Signs of anxiety and depression improved only in contingency, when risk stratification was taken into account. In addition, physical components of quality of life also showed improvement, which did not occur with mental components though. Therefore, the kinesiotherapy-based rehabilitation program was capable of positively influencing all primary symptoms, and only some aspects of secondary OA symptoms


Introduction
Osteoarthrosis (OA) is a chronic-degenerative rheumatic disease of the cartilage, characterized by wear due to an imbalance between formation and destruction of its main components, as a consequence of trauma, overuse and genetic predisposition (Hayashi, Roemer, & Guermazi, 2018;Pereira, Ramos, & Branco, 2015).OA is estimated to affect 10% of the world's population aged over 60 years, being the most common musculoskeletal complaint nowadays (Bhatia, Bejarano, & Novo, 2013).In Brazil, there are few studies addressing the incidence and prevalence of this disease, but results appear to be quite similar to the worldwide scenario (Cunha-Miranda, Faustino, Alves, Vicente, & Barbosa, 2015;Silva, Menezes, & Noal, 2009).
Individuals with OA usually seek medical attention due to arthralgia.However, in addition to it, joint stiffness and limitation also occur, with progressive loss of repair of the cartilage and of the subchondral bone (Sakalauskienė & Jauniškienė, 2010;Sandell, 2012).Pain is the main cause of restriction to physical activity, leading to muscle weakness and hypotrophy, which culminate in decreased physical fitness.The latter, in its turn, leads to an increase in pain, generating incapacity (Runhaar, Luijsterburg, Dekker, & Bierma-Zeinstra, 2015;Sakalauskiene et al., 2010;Sandell, 2012;Silva et al., 2009).
Physiotherapeutic treatment is one of the main forms of rehabilitation for individuals with OA and aims to decrease joint load, improving the alignment between segments and range of motion, which restores the physiological neuromuscular function.Currently, there are no guidelines that indicate which physiotherapeutic procedures are most effective in treating OA.Most likely, this factor is closely linked to variations in the onset of the condition, to the specificity of the treatment and to each patient's individual responses (Page, Hinman, & Bennell, 2011).
In addition to other techniques, physiotherapy includes kinesiotherapy, which can contribute to reducing pain and to the functional improvement of this population.Some current pieces of evidence show that physical activity has been efficient in the prevention and adjunct treatment of OA, especially in the limits specific knee exercises (Bennell et al., 2010).In addition to these gains, decrease in the extent of impairments, and improvements in proprioception are identified as possible mediators in a positive association between physical exercise and OA symptoms (Runhaar et al., 2015).
In this regard, it appears that the multiple physical and psychological factors in people with OA and chronic pain are associated with the development of incapacities and worsened pain in the long run.In this way, pain self-efficacy, negatively charged emotion, and expectations related to the evolution of the sensation of pain are extremely important factors in the approach of people with knee OA (Edwards, Cahalan, Mensing, Smith, & Haythornthwaite, 2011;Helminen, Sinikallio, Valjakka, Väisänen-Rouvali, & Arokoski, 2016;Sinikallio, Helminen, Valjakka, Väisänen-Rouvali, & Arokoski, 2014).Preventing and treating the emergence of these factors may be the key for the chronification condition not to become widespread (Pergolizzi Jr., Raffa, & Taylor Jr., 2014), and for the success of rehabilitation to be reached more effectively.
Taking into account the aforementioned factors, using group therapy may have a beneficial impact on this population, since it has potential to promote improvements in psychological aspects related to motivation, social interaction, interpersonal relationships, exchange of individual experiences, and others (Courtois, Cools, & Calsius, 2015).Studies with rehabilitation of rheumatic diseases have shown, a while ago, that group therapy can lead to better results in the primary and secondary symptoms of diseases (Hidding et al., 1993;Hidding et al., 1994).More recent evidence does not point to significant differences in OA symptoms when physical therapy is performed individually as opposed to in group (Allen et al., 2016).Therefore, there seems to be no loss in the efficacy of physical therapy performed collectively.
In addition to the factors already mentioned, perhaps the divergence about the real benefits of exercising to OA symptoms is a consequence of the lack of specificity of some protocols that do not meet the current literature.With respect to this, there is need for filling in some remaining gaps about the actual relationship between physical exercise and OA symptoms (Runhaar et al., 2015).Besides, the prescription and dosage of exercise for this purpose has not been fully elucidated either (Bennell, Buchbinder, & Hinman, 2015;Runhaar et al., 2015).
Therefore, the importance of physiotherapeutic treatment, with emphasis on physical exercises, in individuals with OA, is worth noting.However, specific aspects related to how physical therapy influences each OA symptom are still being discussed in the scientific literature.Thus, this study proposes to investigate the influence of a specific, kinesiotherapy-based rehabilitation program on OA symptoms, following group treatment.

Participants
The sample consisted of 31 individuals, of both sexes, aged over 50 years, with medical diagnosis of hip or knee OA and physiotherapy prescription.The study excluded patients with other associated musculoskeletal or neuromotor diseases and who missed up to two sessions during the intervention period.All participants read and signed a Free and Informed Consent Form agreeing to participate in the study.The Ethics Criteria on Research Involving Human Beings were complied with, in accordance with Resolution 466/12 of the Brazilian National Health Council.The study was approved by the local Research Ethics Committee under CAAE protocol No. 69457717.1.0000.5402.

OA Symptoms
The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), which aims to assess pain, stiffness and physical function in individuals with knee or hip OA, was used for tracking symptoms.Said instrument consists of 24 questions distributed into three dimensions: A (pain), B (stiffness) and C (physical function).Its total score may vary between zero and 96, or being obtained by domain.The assessed dimension becomes worse the higher the score obtained in it (Bellamy, Buchanan, Goldsmith, Campbell, & Stitt, 1988).

Trunk Flexibility
The sit and reach test was used for assessing trunk flexibility.To do so, the participant, sitting with their knees extended, was asked to flex their trunk anteriorly, with the head between the arms, also extended.Data were recorded in centimeters, using a Wells bench (Instant Flex, Sanny ® , Brazil).Three attempts were made, and the best result was considered (Wells & Dillon, 1952).

Balance
To assess balance, the Berg Balance Scale was employed, which considers postures and transitions, such as standing, getting up, walking, leaning forward, transferring and turning.The total score is 56, and each item has an ordinal scale of five alternatives, ranging from zero to four, according to the level of difficulty.The higher the score, the better the balance levels (Berg, Wood-Dauphine, Wiliams, & Gayton, 1989).It was duly translated into Portuguese and showed reproducibility in the Brazilian culture (Miyamoto, Lombardi Junior, Berg, Ramos, & Natour, 2004).For stratification of fall risk, it has been proposed that scores between 53-46 suggest low to moderate risk, and scores below 46 suggest high risk for falls (Shumway-Cook, Baldwin, Polissar, & Gruber, 1997).

Pain
To assess pain level, the Visual Analogue Scale (VAS) was used -a continuous scale for pain magnitude estimates, being considered one of the best methods for this purpose.The VAS is one-dimensional and composed of a 10-cm line.Score zero -beginning of the line -indicates absence of pain, whereas ten -end of the line -denotes unbearable pain.(Carlsson, 1983;Joos, Peretz, Beguin, & Famaey, 1991).For application, the patient was asked to indicate what level of pain they were feeling at rest at the time of the assessment.

Signs of Anxiety and Depression
To assess levels of anxiety and depression, the Hospital Anxiety and Depression Scale (HADS), translated into and validated for Portuguese, was used, which aims to detect mild levels of affective disorders in nonpsychiatric environments.It is made up of 14 items in total.Seven items correspond to the assessment of anxiety (HADS-A), and the other seven items correspond to the assessment of depression (HADS-D).The score for each item ranges from zero to three, with the maximum score being 21 for both HADS-A and HADS-D.The authors describe the score greater than or equal to nine, with a cutoff for symptom categorization.Thus, for both subscales, scores below nine are classified as absence of symptoms, whereas scores greater than or equal to nine are classified as presence of symptoms (Botega, Bio, Zomignani, Garcia Jr., & Pereira, 1995;Zigmond & Snaith, 1983).

Quality of life
The SF-36 questionnaire was used for assessing the participants' quality of life.This generic instrument is easy to apply and understand; it is multidimensional and composed of 36 items distributed into eight domains: functional capacity; physical aspects; pain; general state of health; vitality; social aspects; emotional aspects, and mental health.It was duly translated and validated in Portuguese.The total score ranges from zero to 100, where zero corresponds to the worst general state of health, and 100 to the best state of health (Ciconelli, Ferraz, Santos, Meinão, & Quaresma, 1999;McHorney, Ware Jr., Lu, & Sherbourne, 1994 ).

Procedures
Two weekly intervention sessions were carried out, lasting 55 minutes, for eight weeks, totaling 16 sessions.The kinesiotherapy protocol was composed of four stages: 1. Global stretches; 2. Global flexibility exercises with stick and ball; 3. Balance and proprioception training, and 4. Lumbar stabilization exercises.The length of the first exercise stage was 10 minutes, and the other ones lasted 15 minutes.
Table 1 describes the muscle stretches performed at stage 1 (all positions were performed in orthostatism and held for 15 seconds).The second stage included four free active exercises with stick and ball, two in orthostatism, and two in sitting position.A detailed description is displayed in Table 2 (two sets of 15 repetitions were executed for each exercise).Lean the trunk to both sides.
Trunk/Multi-joint Sitting.Ball between hands, arms above head, legs extended and abducted.
Put the arms down, resting the ball on the floor, between the legs.Roll the ball forward while progressing into the trunk flexion range.
The third stage included tasks with standing balance training.Each task lasted one minute, with a rest interval of 20 seconds between them.Detailed description below:

•
With two small rubber balls under the heels, forefoot resting on the floor, make successive movements, with small range, of knee flexion and extension, while the heels 'bounces' on the balls.

•
Release the weight on a rubber ball positioned at three different areas of the soles of the feet (rearfoot, midfoot and forefoot).

•
Walk on a mat, on a physiological and reduced support base.• Stay on one-leg support and with the sole of the other foot against the inner thigh.
• Keep balance on the latero-lateral proprioception board; • Keep balance on the anteroposterior proprioception board.The fourth stage included isometric postures held with or without association of static, upper limb positions, with activation of lumbar-spine-stabilizing muscle groups and deep abdominal muscles.The postures used were orthostatism, sitting, supine and four supports, and can be seen in Figure 1.
Initially, the individuals were only requested to keep their posture (each posture was repeated five times and lasted 10 seconds); afterwards, static, upper limb positions were associated while they were holding positions (example, Figure 1E).First, they were asked to raise an upper limb, then the other; subsequently, they were requested to raise a lower limb, then the other (five times for each limb).Finally, they were instructed to move their limbs alternately, maintaining the position (10 times).
In addition, at that stage, they also executed the following isometric exercises while recruiting lumbarstabilizing and deep abdominal muscles:

•
Side plank held with support on the forearm and on the knees flexed on the mat.
• Hip abduction held in supine position.
• Squat held with ball between the back and the wall, reaching a maximum knee flexion of 90°.
• Sitting position held, on Swiss ball (unstable surface), with the feet resting on the floor.
• Bridge (pelvic elevation) held, with the feet on proprioception board.In all exercises at that stage, the postures were repeated five times, each lasting 10 seconds.

Statistical analysis
The analyzed continuous data (total values obtained in quantitative measurements) are presented with the aid of descriptive statistics, through means and standard deviations, whereas categorical data (variables organized by categories or groups) are presented by their total values and prevalence (%).The Kolmogorov-Smirnov test was applied to verify data distribution and thus define the use of comparison tests (between pre-and post-intervention moments) for each variable.In the case of normal distribution (Gaussian), Student's t test was applied to paired samples (OA symptoms, trunk flexibility variables, and pain); otherwise, Wilcoxon's test (balance variables, signs of anxiety and depression, and quality of life) was applied.Categorical data (referring to the balance scale and the anxiety and depression scale) were compared using McNemar's chi-square test to find differences between contingencies before and after treatment.The analyses were run on software Statistical Package for the Social Sciences (IBM ® SPSS ® ), and the significance level was 0.05.

Results
The sample was composed of 31 participants, comprising 12 men and 19 women, aged on average 64.01±8.69-ranging from 51 to 81 years, with body mass index of 30.99±6.76 -ranging from 18.5 to 44.6 kg m -2 .
Table 3 displays mean values (standard deviations), medians (maximum and minimum), statistic values (Student's t or Wilcoxon's T) and respective p-values, for primary OA symptoms, after application of the indicated tests.For the Berg Balance Scale, McNemar's chi-square test showed differences between the contingencies of low and moderate risk scores and the contingencies of high risk scores for falls (χ 2 =0.008; p=0.021), after intervention.In addition, this difference occurred in greater number for the change from high to moderate and low risk (n=13; 41.93%), compared to the opposite (n=3; 9.67%).McNemar's chi-square test, for HADS-A, showed significant differences between the contingencies of scores for signs of anxiety and the contingencies of scores without signs of anxiety (χ 2 =2.839; p=0.003), after intervention.In addition, this difference occurred in greater numbers for the change from presence to absence of signs of anxiety (n=12; 38.70%), compared to the opposite (n=1; 3.22%).McNemar's chi-square test, for HADS-D, did not show significant differences between the contingencies of scores of signs of depression and the contingencies of scores without signs of depression, after intervention, despite a marginal p-value (χ 2 =12.519; p=0.063).

Discussion
The aim of this study was to verify the influence of group kinesiotherapy on OA symptoms.For some time now, kinesiotherapy and group therapy have been used for treating patients with OA, but the specific effects of this resource for this purpose are not yet clear in the scientific literature.A current literature review describes that aerobic resistance and performance exercises, neuromuscular exercises and muscle strengthening are not effective in reducing pain in knee OA.Besides, it describes that currently available evidence suggests a small to moderate effect of exercise (compared to 'non-exercise') on hip or knee OA.However, it emphasizes that exercise prescription and ideal dose are not fully elucidated with studies (Bennell et al., 2015).
Regarding the primary symptoms of OA, the proposed treatment program was effective in improving this population's general symptoms, since the WOMAC score was lower in the post-test.There is solid evidence, reported in previous studies, that most types of physical exercises at rehabilitation level are effective in reducing direct symptoms of OA, assessed by WOMAC (Allen et al., 2018;Allen et al., 2016;Cheung et al., 2016).The results of the present study do not differ from this trend, pointing to the patient's overall improvement as to these aspects.
It was hypothesized, in this study, that gains in some specific aspects of physical capacity in OA may be related to this improvement, since previous studies have been increasingly supporting this statement (Bennell et al., 2015;Bennell et al., 2010 ;Runhaar et al., 2015).Trunk flexibility, balance and quality of life with respect to the participants' physical capacity improved after training, as expected.Moreover, after stratification of fall risk, the contingency of individuals who went from high risk to moderate and low risk was significantly greater.Just as with all other aspects involving the complex systematism of the development of OA clinical condition, issues related to flexibility and body balance, as well as their impact on functional capacity, are interdependent and influenced by the specificity of the provided treatment (Page et al., 2011 ).Specific training for balance and muscle strength gain has great potential to reach these goals (Chaipinyo & Karoonsupcharoen, 2009;Lin, Lin, Lin, & Jan, 2009).
The kinesiotherapy protocol involved conducts directly linked to these aspects, such as muscle stretching, stimulus for postural reactions, active and proprioceptive exercises.It is known that OA may lead to a decrease in somatic and proprioceptive afferences due to the local pathological process and to restricted movement and generalized pain around the joint (Sandell, 2012).Furthermore, conditions related to the onset and worsening of pain, such as centralization, catastrophization, imbalance in the descending control of pain, and kinesiophobia must be considered, since they can influence the success of the treatment in an incisive manner (Arendt-Nielsen et al. , 2010;Domenech et al., 2014;Monticone et al., 2013).Recent studies that seek to understand the relationship between the most adequate types of exercise for each OA symptom describe that the association of muscle stretching exercises, flexibility exercises and aerobic exercises has been the best protocol for reducing pain and improving functional measures (Page et al., 2011;Uthman et al., 2013).The present intervention protocol did not include aerobic exercises, but covered exercises for body awareness, activation of deep stabilizing axial muscles and proprioception, and, similarly, found improvements in these aspects.
Recently, efforts have been made towards a standardized training dosage for individuals with OA.The American College of Sports Medicine (ACSM) provides recommendations for exercise prescription to adults with levels ranging from moderate to high, for virtually all types of exercise (Garber et al., 2011).The adoption of these recommendations in the training of individuals with OA has resulted in changes in pain, but not in changes in physical function reported by patients (Moseng et al., 2017).This paradigm can perhaps be better understood if we consider the multifactorial aspects of a chronic pain condition.Pain remission resulting from different management techniques (physical exercises, physical therapy activities, analgesic techniques, and drug treatment) may be related to the remission of symptoms linked to cognitive, motivational and emotional processes of the condition (Arendt-Nielsen et al., 2010).However, once again, it seems that improvements in the physical function of patients with OA can be attributed to the most adequate technique for each goal (Chaipinyo & Karoonsupcharoen, 2009;Lin et al., 2009;Page et al., 2011).Thus, one must be cautious when replicating the ACSM recommendations and prescribing any type of therapy or physical exercise to treat patients with OA.
In the present study, no analgesia physiotherapeutic resource was used in combination with kinesiotherapy but, even so, there was improvement in reported pain and in some aspects of quality of life, after the training program.It is known that exercise can contribute to lessening pain in OA (Jansen, Viechtbauer, Lenssen, Hendriks, & de Bie, 2011), but this is still a controversial subject, as there is an inconsistency in the literature that delimits a gap about how the mechanisms involved in improving pain and function correlate (Matsudo & Calmona, 2009), despite recent efforts to reach this understanding (Bennell et al., 2010;Domenech et al., 2014;Lluch Girbes et al., 2016;Monticone et al., 2013;Riddle et al., 2015;Thompson et al., 2010;Van Ginckel et al., 2016).A big portion of the studies that report little effectiveness of physical exercises on function and pain compare their results with a control group without any type of treatment (Bennell et al., 2015), which compromises the generalization of results reported by them.
The key to understanding so much divergence among researchers in this regard can be found in two factors that are still little addressed in the scientific literature.The first one is the fact that pain, in OA, can have peripheral and central components.The traditional rehabilitation model is beneficial in improving the peripheral components of pain, but its effects on the processes involved with central sensitization are not yet known (Lluch Girbes, Nijs, Torres-Cueco, & Lopez Cubas, 2013).It is believed that removing the stimulus that causes its incitation can normalize the central processing of pain (Arendt-Nielsen et al., 2010).It is generically suggested that this can be achieved with gains in muscle strength, improvements in joint condition, and decrease in injury extent (Bennell et al., 2010;Runhaar et al., 2015).The second one, in its turn, consists of the fact that both pain and function are closely related to psychosomatic aspects.These aspects are extremely individual and interdependent, and vary according to a series of conditions that involve the patient as a whole, and not just their illness.Presence of depression and pain catastrophization are associated with the severity of reported pain, with physical incapacity and with treatment failure.Some measures seem to mediate these deleterious effects, and their handling should be an important therapeutic target in the multimodal management of these conditions (Edwards et al., 2011;Sinikallio et al., 2014).
For this reason, it was pertinent to include in the present study the assessment of some secondary symptoms in which a chronic condition, such as OA, can culminate.Post-treatment effects with physical exercises on the secondary symptoms of OA are not widespread in the literature.However, the assessment of these aspects is very important, as they configure conditions that impact the understanding and living with chronic pain (Cunha-Miranda et al., 2015).In the present study, there was no decrease in signs of anxiety and depression after the kinesiotherapy program.However, when risk stratification is considered, the contingency that ceased to present symptoms of anxiety was significantly greater, whereas for depression it was marginal.It is known that there is an inversely proportional correlation between physical function (WOMAC) and anxiety, and between lower limb function and depression (Scopaz, Piva, Wisniewski, & Fitzgerald, 2009), so these are factors that, if not factored in OA treatment, can interfere with the improvement in overall physical condition.
The effects of physical therapy on psychosocial factors caused by OA remain inconclusive (Cheung et al., 2016).However, engagement in physical exercise by individuals with OA provides almost simultaneous beneficial effects on the physical variables, so it is difficult to believe in a purely physical response.Physical therapy ends up covering several psychosocial components, such as acceptance of treatment that demystifies pain and prevents joint damage due to exercise, offering an active coping strategy and, ultimately, reducing incapacity and social isolation (Kruger-Jakins, Saw , Edries, & Parker, 2016).Additionally, it shows potential to promote a rebalancing between pain-regulating descending systems and cognitive and emotional processes.It has been reported that aerobic exercises and kinesiotherapy results in a generalized increase in pain tolerance during and immediately after their execution.The physiological mechanisms of such improvement are still not fully elucidated, although they tend to be attributed to the release of endogenous opioids and to the activation of nociceptive inhibitory mechanisms.However, one must bear in mind that aerobic exercises activate the facilitation of pain in some patients with chronic pain and central sensitization.Kinesiotherapy, in its turn, may lead to increased generalized sensitivity to pain (Nijs, Kosek, Van Oosterwijck, & Meeus, 2012).
For quality of life, as previously described, there was improvement in functional capacity and pain, after intervention.The SF-36 is a scale that can be generically divided into physical and mental components (Ware Jr., 2000); thus, it can be said that, among the mental components, there was no significant difference.A literature review aimed at describing the influence of physical exercises on quality of life, through SF-36, in individuals with OA, observed that most of the included studies found significant improvements both in the physical components and in the mental components of the scale (Tanaka et al., 2015).However, again, there was no standardization of the intervention protocols, and the types of physical exercises, as well as the chosen frequency and length differed greatly, which makes a direct comparison with the present results impossible.As already suggested, the understanding, acceptance and impact that a chronic disease has potential to cause are totally individualized, showing great heterogeneity among patients.Perhaps, if there was a risk stratification within each sub-item of SF-36, one could find information that is hidden by the variability of patient responses.
With the conduction of this study, it was possible to understand a little better about how a kinesiotherapy protocol can act on primary and secondary symptoms of OA.However, if this protocol had been compared to a protocol with another type of physical therapy and/or physical exercises, a better understanding would certainly be achieved, which is a limitation of the present study.In this sense, new researches are worth being developed with this experimental design and covering central sensitization activities, with task-oriented exercises, because, in addition to stimulating motivation, they recruit memory and learning areas that have a close relationship with limbic areas where some of the mental processes affected by chronic diseases occur (Lluch Girbes et al., 2013;Muratori, Lamberg, Quinn, & Duff, 2013).

Conclusion
The kinesiotherapy-based rehabilitation program was capable of beneficially influencing the primary symptoms of OA.Regarding secondary symptoms, there was a tendency of improved psychosomatic symptoms, only in relation to contingents within the risk stratification.As for quality of life, some physical components improved, but the program seems to have no effect on the mental components of this aspect.

•
Stay on one-leg support on the floor, with eyes open and closed; • Stay on one-leg support on the mat, with eyes open and closed; • Stay with one foot in front of the other on the floor, with eyes closed;

Table 1 .
Description of the muscle stretches performed at stage 1.

Table 2 .
Description of free active exercises with stick and ball performed at stage 2.

Table 3 .
Means (standard deviations) -for parametric tests, and medians (range) -for non-parametric tests; statistic values (Student's t, for parametric, or Wilcoxon's T, for non-parametric) and their respective p-values for primary OA symptoms.

Table 4
displays median values (maximum and minimum), Wilcoxon's T statistic values, and respective p-values, for secondary OA symptoms, after application of the indicated tests.

Table 4 .
Medians (range) -for non-parametric tests; statistic values and their respective p-values for secondary OA symptoms.