Study design

It is a randomized controlled clinical trial, of pa rallel groups, with ratio 1:1 and simple blind, that was approved by the Scientific Ethical Committee of the Sociedad Pro Ayuda del Niño Lisiado - Teletón (Certi ficate N° 7/2012).

Participants

The patient population was determined by sear ching the database and reviewing clinical records, of which patients diagnosed with JIA at the Teleton Insti tute (TI) Santiago were included in the study between 2012 and 2015, aged 8 to 18, in the acute or subacute phase (given by sensation of pain, joint swelling, and functional deficit during the last three months befo re the intervention), who had a medical indication for hydrotherapy and who agreed to participate in the study by signing a consent by the child and informed consent by the legal guardian.

Patients who presented the following symptoms before the start of the study were excluded: sensation of pain, swelling or functionality limitation due to ex ternal causes to JIA (surgeries or trauma), intercurrent diseases (respiratory diseases, inflammatory disease of the ear, skin lesions, and infections), urinary and/or in testinal incontinence, and fear of water.

Interventions

This study considered two parallel groups, pre viously randomized, where the study group received a Watsu therapy protocol and the control group recei ved a conventional hydrotherapy protocol. Both pro tocols had a duration of 10 sessions, once a week, with 45-minute sessions.

a) Watsu: The patients received a protocol of passive movement sequences, stretches, and massages, be longing to the “Transition Flow” of Watsu thera py¹⁹ (Table 1).

Table 1 Group treatment Watsu and conventional hydrotherapy protocols. 

b) Hydrotherapy: The patients received a hydrothe rapy protocol for arthritis, adapted from the proposal of Epps et al.¹⁸, consisting of stretching, strengthening exercises, and swimming (Table 1).

In both groups, the treatment was carried out by physiotherapist with experience in rehabilitation of physically disabled people at TI Santiago. In the case of Watsu therapy, the therapist had 163 hours of practical training, certified by the Institut fur Aquatische Kor- perarbeit (IAKA) and the Chilean Aquatic Bodywork Federation.

60% or more adherence to the total scheduled treatment in each patient was considered as accom plished treatment. In addition, the continuity of their routine treatments was accepted, such as scheduled admissions to regular physiotherapy, occupational thera py, and changes in their pharmacological dosage due to ethical considerations.

Outcome measures

HRQoL, functional health status, sensation of pain, and joint ranges of passive motion were evaluated in both groups in the previous, subsequent and 3-month follow-up periods. The professionals in charge of carrying out the measurements were a psychologist and two physiotherapists trained in the use of each evalua tion instrument.

Main outcome

Health-related quality of life: The Pediatric Qua lity of Life Inventory Scale (PedsQL 4.0 Generic Core Scale) was used, which measures HRQoL in people aged 2 to 18, healthy or with a chronic health condi tion, and it is validated for the Chilean population²⁴. This scale evaluates the child’s physical, emotional, so cial, and school performance domains²⁵, giving a 0 to 100% HRQoL score.

Secondary outcomes

a) Functional health status: Defined as the effect of JIA on the ability to perform common tasks⁹ and measured through the Childhood Health As sessment Questionnaire (CHAQ). This question naire is used in patients with JIA between 1 and 19 years of age²⁶, validated in Chile in 2001²⁷. It has a disability index (which contains questions of functionality in daily activities), discomfort, and health status. The results were turned into a score from 0 to 3 which is translated into absence of di sability or maximum functional disability respectively^{(26, 27)}.

b) Sensation of pain: Pain was measured by the dis comfort index present in CHAQ²⁶. This initially used a numeral scale from 0 to 100 (from no pain to maximum pain) to represent the perceived sen sation of pain, which was then changed into a score from 0 to 3 of the global questionnaire.

c) Joint ranges of motion: The 10-joints Global Ran ge of Motion Scale (GROMS) was used, which evaluates 10 passive movements considered essen tial for functionality, such as elbow flexion, wrist extension, general mobility of proximal metacar pophalangeal and interphalangeal joints (from the second to the fifth finger), thumb flexion and ab duction, hip flexion-extension, and knee flexion- extension. It uses a score from 0 to 1, where 1 is normal joint range in all 10 joints^{(28, 29)}. However, this scale has not been used previously in clinical studies, but its creators state that its results would be related to those obtained in CHAQ²⁸.

Sample size

It was not determined since we used the total po pulation of patients with JIA from the TI Santiago who met the inclusion criteria and agreed to participate in the study.

Randomization

a) Sequence generation: A simple randomization method was performed in groups of 10 patients, with ratio 1:1, after accepting to participate in the study.

b) Assignment concealment mechanism: The patient’s personal information was hidden using an identi fication code (ID), then assigned to the Watsu and conventional hydrotherapy groups using a ran dom number generator (nosetup.org). This infor mation was put into sealed envelopes, which were given to the patient and/or legal guardian.

Implementation

The methodological team of the Research and De velopment Administration Teletón performed the se quence of random assignment and interventions, and the principal investigator carried out the registration of participants.

Blinding

The evaluators who performed each outcome mea surement (pre-intervention, post-intervention, and follow-up) were blind to randomization and assign ment of participants to study groups. However, given the evident differences in interventions (Watsu and conventional hydrotherapy), patients and physiothe rapists who administered the therapies were not blin ded.

Statistical methods

The data were recorded in an Excel spreadsheet and processed using the statistical software SPSS v17.0. Summary measures were calculated for study variables, Student’s T-test was used for means, and Pearson’s chi-square to determine differences in age, gender, and type of JIA between groups.

Shapiro-Wilk normality test was performed. Va riables comparisons between groups were made using Kruskal-Wallis and Mann-Whitney tests, and varia bles differences within each study group at baseline, post-treatment and follow-up were analyzed with Friedman and Wilcoxon tests. P<0.05 was considered a statistically significant difference. Data were analy zed on an ‘intention-to-treat’ basis, so the dependent variables of all randomized patients were included in the analysis.

Participant flow

231 JIA patients aged 8-18 years were found in the TI Santiago database (2012-2015), and 59 of them met the inclusion criteria. Out of these, 46 agreed to parti cipate and signed informed consent and/or assent, and then were assigned by simple randomization in Watsu (n = 24) and hydrotherapy (n = 22).

During the intervention process, five cycles of para llel treatment were carried out, between June 2013 and December 2015, concluding the intervention phase at the end of 2015, due to the lack of new patients who met the inclusion criteria.

In the 2013-2015 period, 16 patients interrupted their treatment (34.78% of total loss) due to the pre sence of intercurrent diseases, scheduling problems, and family issues (Figure 1). This recorded loss in both the Watsu group (n = 8; 33% loss) and conven tional hydrotherapy (n = 8; 36% loss) did not pre sent statistically significant intergroup differences (p = 0.537).

Figure 1 Diagram of the flow of patients during the execution of the ran domized clinical trial (Ba sed on CONSORT, 2010). 

Clinical and demographic characteristics

The study participants presented homogeneity bet ween the Watsu and hydrotherapy groups, in terms of age and subtype of JIA. The mean age was 13.17 ± 3.02 years for the Watsu group and 12.68 ± 3.00 years for the conventional hydrotherapy group, without statis tically significant intergroup differences (p = 0.880). According to JIA subtype, there was a greater presen ce of polyarticular JIA (26.1%), undifferentiated JIA (26.1%), and oligoarticular JIA (19.6%) in the total number of participants and no significant differences were observed in the intergroup analysis (p = 0.454).

On the other hand, most of the study participants were female (76.1%), however, there were statisti cally significant differences (p = 0.038) in grouping patients by gender in the study groups, with a higher number of males concentrated in the Watsu group. (Table 2).

Table 2 Clinical and demographic baseline characteristics of the study groups. 

Main result

Regarding health-related quality of life measured through PedsQL4.0, Watsu therapy presented a sta tistically significant difference in the physical functio ning sub-dimension at basal level (p = 0.028), and a statistically significant post-intervention improvement (p = 0.041) compared to conventional hydrotherapy.

In the intra-group evaluation, Watsu therapy ob tained significant differences in improvement in the psychosocial health sub-dimension between baseline evaluation and follow-up (p = 0.021). Conventional hydrotherapy showed no significant intra-group di fferences in sub-dimensions or overall quality of life scores (Table 3).

Table 3 Health-related quality of life (HRQoL) in each study group measured with Pediatric Quality of Life Inventory Scale (Peds QL 4.0 Generic Core Scale). 

Secondary results

a) Functional health status and sensation of pain: Upon application of the Childhood Health Assessment Questionnaire (CHAQ), Watsu therapy obtained sta tistically significant improvements in the post-inter vention evaluation of the disability index (p = 0.015), discomfort index (p = 0.031), health status index (p = 0.013), and total CHAQ score (p = 0.003) compa red to conventional hydrotherapy.

In the intra-group evaluation of each CHAQ index, no statistically significant differences were observed in the disability index in the Watsu or hydrotherapy group (Table 4). In the discomfort index, Watsu thera py obtained a statistically significant improvement in post-intervention evaluation (p = 0.007), in contrast, in the health status index, Watsu therapy obtained a statistically significant deterioration between post-in tervention evaluation and follow-up (p = 0.012). On the other hand, the health status index of conventional hydrotherapy showed statistically significant impro vements between post-intervention evaluation and follow-up (p = 0.033), and deterioration between ba seline evaluation and follow-up (p = 0.017).

Table 4 Functional health status (CHAQ), its index (disability, discomfort and health status) and range of motion (GROMS) in each study group. 

Finally, in the CHAQ total value, Watsu thera py showed a statistically significant improvement in functional health status between baseline and post intervention evaluations (p = 0.004) and a statistica lly significant deterioration of the same variable bet ween post-intervention and follow-up evaluations (p = 0.039) (Table 4).

b) Range of motion: In the 10-joints Global Range of Motion Scale (GROMS) evaluation, there were no statistically significant differences between the Watsu and hydrotherapy groups (Pre-intervention p = 0.794; Post-intervention p = 0.190, and follow-up p = 0.383).

In the intragroup evaluation, Watsu therapy sig nificantly improved the range of motion between ba seline and post-intervention evaluation (p = 0.023). (Table 4)

Adverse effects

No adverse events related to therapeutic interven tions were reported by study participants or their legal guardians.