Efficacy of hydrotherapy in fibromyalgia syndrome--a meta-analysis of randomized controlled clinical trials
Keywords
Article abstract
OBJECTIVE:
To systematically review the efficacy of hydrotherapy in FM syndrome (FMS).
METHODS:
We screened MEDLINE, PsychInfo, EMBASE, CAMBASE and CENTRAL (through December 2008) and the reference sections of original studies and systematic reviews on hydrotherapy in FMS. Randomized controlled trials (RCTs) on the treatment of FMS with hydrotherapy (spa-, balneo- and thalassotherapy, hydrotherapy and packing and compresses) were analysed. Methodological quality was assessed by the van Tulder score. Effects were summarized using standardized mean differences (SMDs).
RESULTS:
Ten out of 13 RCTs with 446 subjects, with a median sample size of 41 (range 24-80) and a median treatment time of 240 (range 200-300) min, were included into the meta-analysis. Only three studies had a moderate quality score. There was moderate evidence for reduction of pain (SMD -0.78; 95% CI -1.42, -0.13; P < 0.0001) and improved health-related quality of life (HRQOL) (SMD -1.67; 95% CI -2.91, -0.43; P = 0.008) at the end of therapy. There was moderate evidence that the reduction of pain (SMD -1.27; 95% CI -2.15, -0.38; P = 0.005) and improvement of HRQOL (SMD -1.16; 95% CI -1.96, -0.36; P = 0.005) could be maintained at follow-up (median 14 weeks).
CONCLUSIONS:
There is moderate evidence that hydrotherapy has short-term beneficial effects on pain and HRQOL in FMS patients. There is a risk to over-estimate the effects of hydrotherapy due to methodological weaknesses of the studies and to small trials included in meta-analysis.
Article content
Introduction
Hydrotherapy is one non-pharmacological therapy of FM syndrome (FMS) used by up to 75% of the patients [1, 2]. The use of water for medical therapy dates back to ancient cultures from China, Japan and Europe. Balneotherapy (drinking of and/or bathing in medicinal water, bathing in warm or cold water or mud) and spa therapy (drinking of and/or bathing in thermal or mineral water) are different forms of hydrotherapy.
Two qualitative systematic reviews were conducted on the efficacy of hydrotherapy in FMS, which searched the literature until July 2006 [3] and December 2006 [4], respectively. One review included only trials published in English language [3]. In the meantime, further studies on hydrotherapy in FMS have been published, which were not included in systematic reviews so far. To our knowledge, a meta-analysis providing effects sizes of hydrotherapy was not published yet. The aim of our review therefore was to determine the efficacy of hydrotherapy in FMS by updating the search without language restrictions and by a quantitative analysis of data.
Methods
Meta-analysis was performed according to the QUORUM (quality of reporting meta-analyses) guidelines [5].
Data sources and searches
The electronic bibliographic databases screened included MEDLINE, PsychInfo, SCOPUS, the Cochrane Central Register of Controlled Trials (CENTRAL) and CAMBASE (through December 2008). The search strategy for MEDLINE is detailed in supplementary Table 1 (available as supplementary data at Rheumatology Online). The search strategy was adapted for each database if necessary. In addition, reference sections of original studies, qualitative systematic reviews on hydrotherapy in FMS [3, 4] and evidence-based guidelines on the management of FMS [6–8] were screened manually. No language restrictions were made.
Study selection
Studies were required to meet the following criteria: (i) any kind of hydrotherapy without exercise; (ii) diagnosis of FMS based on recognized criteria; (iii) randomized controlled trials (RCTs) comparing hydrotherapy with any other intervention or with no intervention; (iv) at least one symptom-specific outcome of the ‘key symptoms’ of FMS such as pain, fatigue, sleep disturbances, depressed mood and health-related quality of life (HRQOL) [9]; and (v) publication of the study in full paper form.
Data extraction
Two authors screened the titles and abstracts of potentially eligible studies identified by the search strategy detailed above independently. The full text articles were then examined independently by two authors to determine if they met the selection criteria. For the preparation of the meta-analysis, two of the four authors independently extracted data (study characteristics and study results) using standard extraction forms.
Assessment of external validity
The external validity (representativeness of study samples for the FMS population in clinical practice and safety of treatment) was checked by analysing the inclusion and exclusion criteria, the socio-demographic and medical data of the study samples, the settings and referrals of the RCTs and the side effects reported.
Assessment of methodological quality
The methodological quality was assessed by the van Tulder score using 11 items. We arbitrarily classified methodology as high (score 8–11), moderate (score 5–7) or low quality (score 1–4) [10]. We used the following modified levels of evidence descriptors to classify the results of the meta-analysis: strong: consistent findings in at least three moderate quality RCTs; moderate: consistent findings in at least three RCTs with at least one moderate RCT; limited: consistent findings in two low-quality RCTs; conflicting: inconsistent findings among multiple RCTs; no evidence: one or no RCTs [10].
Dealing with missing data
We contacted authors of studies in case of missing data in the publication. If the s.d. (post) was not reported and not provided on request, the missing s.d. (post) was substituted by the mean of the s.d. (post) of the other studies if the outcome was reported by at least three studies on the same scale.
Data analysis
For the comparison of proportions the chi-squared test was applied. Non-parametric tests (Mann–Whitney U-test) were used for the comparison of continuous variables. A two-sided P-value of ⩽0.05 was considered significant. Meta-analyses were conducted using RevMan Analyses software (RevMan 5.0.17) of the Cochrane Collaboration [11].
Standardized mean difference (SMD) as effect measure was used by calculating SMD by means and s.d. or change scores for each intervention. For the calculation of SMDs, the data of at least two studies should be available. Examination of the combined results was performed by a random effects model, because this model is more conservative than the fixed effects model and incorporates both within-study and between-study variance [12]. SMD used in Cochrane reviews is the effect size known as Hedges (adjusted) g. We used Cohen's categories to evaluate the magnitude of the effect size, calculated by SMD, with g > 0.2–0.5, small effect size; g > 0.5–0.8, medium effect size; and g > 0.8, large effect size [13].
Assessment of publication bias
Potential publication bias was intended to investigate by visual assessment of the funnel plot (plots of effect estimates against sample size) [14] calculated by RevMan Analyses software. Furthermore, we tested the sensitivity of our results to potential unpublished studies using a file drawer test for meta-analysis. This test determines how many negative studies with an effect size of d = 0.01 would be needed to negate our findings (fail-safe-N) [15]. If fail-safe-N > file-drawer N (5k + 10; k, number of studies meta-analysed), the results of the meta-analysis can be regarded as robust against potential reporting bias [16].
Assessment of heterogeneity
Heterogeneity was tested using the chi-squared test with a P-value conservatively set at 0.1 and the I2-statistic with I2-values >50% indicating strong heterogeneity [17].
Subgroup analyses
Where at least two studies were available, subgroup analyses were performed for type (thermal bath vs other types) and intensity of hydrotherapy (200 vs > 200 min), co-therapies (allowed or not), control group (active therapy vs no therapy or treatment as usual), setting (outpatients vs inpatients) and sex ratios (only women vs mixed sample). These subgroup analyses were also used to examine potential sources of clinical heterogeneity.
Sensitivity analysis
Sensitivity analyses were planned by removing studies based on the following methodological quality criteria: inadequate randomization, no allocation concealment, drop out rate >20% in treatment group or not reported, low-quality score and missing values substituted for the calculation of effect sizes. These sensitivity analyses were also used to examine potential sources of methodological heterogeneity.
Results
Study selection
The literature search produced 96 citations involving FMS, hydrotherapy and RCTs, 13 of which met initial inclusion criteria (Fig. 1). On more detailed review of these 13 initially selected articles, further three papers were excluded for the following reasons: one study, because means and/or s.d. of pre-test and post-test data were not included in the publication, and were not provided by the authors on request and could not be calculated [18]; one study because the outcomes assessed did not meet the inclusion criteria [19]; and one study because of double publication of the data [20]. Finally 10 studies met our selection criteria and were included for meta-analysis [21–30].
QUORUM flow diagram.
External validity
Some characteristics of the studies and the patients are presented in Table 1. Details are presented in supplementary material 1 (available as supplementary data at Rheumatology Online). Side effects or adverse events were explicitely mentioned by four studies: three reported no side effects [26, 27, 30] and one study reported slight flashes in 10% of the patients [23].
Main characteristics of studies with hydrotherapy
Methodological quality
Three studies had a moderate quality (van Tulder score 5–7) [24, 25, 27], and the other ones had had a low quality (van Tulder score <5). Only two studies [24, 25] reported adequate methods of randomization. No study performed an adequate concealment of treatment allocation or an intention-to-treat analysis.
Effects and heterogeneity
The effects of hydrotherapy at the end of therapy and at latest follow-up are shown in supplementary Figs 2–5 (available as supplementary data at Rheumatology Online).
There was moderate evidence for a reduction of pain (SMD −0.78; 95% CI −1.42, −0.13; P < 0.0001; I2 = 83%) (nine study arms) and improved HRQOL (SMD −1.67; 95% CI −2.91, −0.43; P = 0.008; I2 = 90%) (four studies) at the end of therapy. The test for overall effect on depressed mood (SMD −0.55; 95% CI −0.55, 0.02; P = 0.06; I2 = 0%) (two studies) was not significant.
There was moderate evidence for a reduction of pain (SMD −1.27; 95% CI −2.15, −0.38; P = 0.005; I2 = 84%) and improved HRQOL (SMD −1.16; 95% CI −1.96, −0.36; P = 0.005; I2 = 84%) (four studies each) at latest follow-up.
Based on Cohen's categories for evaluating the magnitude of effect sizes, the effects of hydrotherapy were large for pain and HRQOL at the end of treatment and at follow-up.
Publication bias
The small number of studies included in the funnel plot limited the meaningful interpretation. The fail-safe-N's calculations indicated that a publication bias was not evident for the data of this meta-analysis (for details see supplementary material 2, available as supplementary data at Rheumatology Online).
Subgroup analyses
The comparisons of subgroups showed overlapping CIs of the outcome pain at the end of therapy (details not presented), except the comparison of spa therapy (SMD −1.63; 95% CI −2.31, −0.96; P < 0.0001; I2 = 73%) (five studies) vs other types (SMD 0.01; 95% CI −0.45, 0.47; P = 0.98; I2 = 12) (two studies). I2 was >50% for all subgroups except hydrotherapies with a dosage of 200 min with I2 = 23%.
Sensitivity analyses
The sensitivity analyses did not change the results (for details see supplementary material 3, available as supplementary data at Rheumatology Online).
Discussion
The aim of this systematic review was to determine the efficacy of hydrotherapy in FMS. We found moderate evidence for the efficacy of spa therapy in reducing pain at the end of treatment and at follow-up. There is no evidence of the efficacy of medical, Stanger and mud baths. We conclude from the low frequency of side effects reported and the low drop out rates in the treatment groups that hydrotherapy is a safe treatment option with a high acceptance by the patients. The fact that spa therapy reduced pain in out-patients, who visited spa resorts in their surroundings and continued their normal life, gives support to the hypothesis that the benefits of spa therapy cannot be attributed to a ‘holiday effect’, but by physical and chemical factors inherent in the thermal water used [33] as well as psychological factors (promotion of psychophysiological well-being).
Our results are in line with a recent systematic qualitative review which concluded that there is moderate evidence for the efficacy of hydrotherapy in FMS [3].
The methodological quality of the RCTs analysed was limited for the following reasons: only three studies had sample sizes of at least 25 per group, which had been identified as appropriate for the detection of clinically important differences between two active treatments [34]. The methodological quality of most trials was low. No study performed an intention-to-treat analysis but analysed the completers. Even if the drop out rates were low, this procedure might have favoured the results of hydrotherapy. Most studies did not report the method of randomization used, all trials did not ensure that the treatment allocation was concealed. Therefore, it is not possible to assess the extent to which selection bias may have occurred in these studies. Furthermore, the studies which allowed co-therapies did not control their effects for dosage or changes in concomitant therapies.
The external validity of the RCTs analysed was limited, because non-Caucasians, patients >65 years and <18 years old and with inflammatory arthritic diseases were not included.
This review has limitations. Some study outcomes, mainly the FIQ subscales, were incompletely reported by most studies and only provided by one author on request. Therefore not all outcomes could be meta-analysed. We found a high heterogeneity and wide CIs of most effect sizes. The small number of trials did not allow to conduct all projected analyses of heterogeneity. Because the meta-analysis included only small trials leading to a large sampling variability, there is a risk to over-estimate the effects of hydrotherapy [35].
In conclusion, spa therapy is a first-line non-pharmacological treatment option of pain in FMS patients living near spa resorts. There is a need for high-quality studies with larger sample sizes to confirm this recommendation.
