This was a single-blind (treatment concealed to the investigators) randomized controlled trial with two parallel groups. Patients attended a spa treatment course in addition to continuing their usual medical care. The tested hypotheses were that patients with advanced chronic venous insufficiency (CVD with CEAP C classes C4-C5) would have a decreased 1-year incidence of leg ulcers (primary outcome criterion) and a long-standing improvement (remaining significant at month 12) in their signs, symptoms, and quality of life (secondary outcome criteria) compared with a control group with no spa treatment.

Inclusion criteria were primary or post-thrombotic CVD with skin changes but no active ulcer (C4a, C4b, or C5) and evidence of venous incompetence demonstrated by duplex ultrasound examination with at least one significant reflux (of more than 1-second duration in a standing position) in the superficial, deep and/or perforator veins. Patients had to be at least 18 years old and willing to participate (written informed consent) in the study (ie, to attend a 3-week course of spa treatment in one of the 12 French spa resorts treating patients with CVD) (ie, Argeles-Gazost, Barbotan-les-Thermes, Bagnoles de l'Orne, Dax, Evaux-les-Bains, Jonzac, La Léchère-les-Bains, Luxeuil, Luz-Saint-Sauveur, Rochefort-sur-Mer, Saint Paul-les-Dax, and Saubusse) and to accept a follow-up of 18 months' duration including a monthly self-administered questionnaire and a medical examination every 6 months.

Patients were not included if a surgical or endovascular treatment of the venous disease was planned at any time during the entire study period (18 months) or had been performed less than 6 months prior to the inclusion visit. Patients with spa treatment in the previous 6 months or a contraindication to spa treatment (life-threatening disease, cardiac or renal failure, immunodeficiency, psychiatric disorders, severe difficulty walking) were also not included, as well as those with edema of nonvenous origin (clinical lymphedema, cardiac failure, hypoalbuminemia), symptomatic neurologic diseases of the lower limbs (neurogenic pain or abnormal neurologic examination of the lower limbs) or significant peripheral arterial disease (ankle brachial index <0.70).

Centralized randomization was performed immediately after the inclusion visit, and its result was kept hidden from the investigator, with a stratification by center and CEAP C class (C4 vs C5) to ensure better comparability. The treatment took place in the spa resort closest to the patient's home.

Patients in the treatment group attended a 3-week spa treatment course in the spa resort closest to their home, soon after randomization. The control group also attended a spa treatment course, but after the comparison period (ie, starting soon after day 365; Fig 1).

Fig 1

Pseudo cross-over design of the study. The comparison concerns the first year (ie, 1 year after treatment for the treatment group, vs 1 year after randomization but before spa treatment for the control group). CIVIQ2, Chronic Venous Insufficiency Questionnaire 2; EQ5D, EuroQol 5D; VCSS, Venous Clinical Severity Score.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

The spa treatment course was performed according to the rules of the French health authority, and study patients were cared for in the same way as any other CVD patient, with their participation in the study concealed from resort staff. The treatment regimen consisted of four balneotherapy sessions per day, 6 days a week during 3 weeks, and educational activities that were differently organized in each resort. The balneotherapy sessions included¹⁵ a 15-minute walking session in a specially designed pool with tracks in semideep (80 cm) cool (28°C) water (training of muscle pump function under water compression); a 20-minute whirlpool bath session with automatic air and water massage cycles (aimed at relaxation and mobilization of the superficial skin volume flow); a 10-minute bath session with customized underwater strong massaging jets (mobilization and softening of the sclerotic subcutaneous tissues); and a 10-minute massage session of the leg and ankle skin areas by a registered physiotherapist under a light spray shower (softening of the sclerotic subcutaneous tissues) or a 15-minute joint mobilization session in a deep (150 cm) warm (34°C) pool under the supervision of a physiotherapist (improvement of ankle, and also knee and hip joint mobility for better ambulation and muscle-pump functioning).

Sessions were customized for each patient by the specialist spa physician according to the patient's needs and capabilities on arrival at the spa resort and at the two medical visits systematically performed during the spa treatment course as required by French health authority rules.

During the entire study period, patients in both groups remained in the care of their regular physician, who provided them with any treatment they considered necessary, including wearing compression stockings. No standardized basal treatment or counseling was provided by the investigators.

The follow-up examinations were performed by the same investigator for a given patient every 6 months during the study period, with a final visit at month 18. Before each visit, the patient was reminded by the research assistant that he had to conceal his status regarding the spa treatment from the investigator, to maintain blinding for the evaluation.

The primary outcome criterion was the 1-year incidence of leg ulcer, defined as a wound to the lower leg involving the dermis, which did not heal within 6 weeks if it was the first in a given skin area, or within 2 weeks for a relapse at the site of a previous ulcer. Photos of the wounds were systematically collected by the investigators, and a critical event committee, consisting of two experienced vascular physicians blinded to any treatment information, validated the diagnosis.

The severity of the venous disease was evaluated using the Venous Clinical Severity Score (VCSS), described by Rutherford et al¹⁸ modified as follows. Spa treatment is a composite treatment that includes patient education, with compliance to compression therapy as one of its aims; the latter was not taken into account in the score calculation but only the parameters related to clinical status, resulting in a modified VCSS ranging from 0 to 27. The score was calculated for both legs and the highest recorded.

Self-evaluation of the intensity of leg symptoms was performed each month by the patient and reported in a diary using two 10-cm visual analog scale (VAS). The intensity of symptoms was rated separately for each leg, from “no discomfort” at the bottom (0.0), to “unbearable” at the top (10.0), according to a method previously described and discussed in detail.¹⁹ The first measurement, taken as baseline, was at month 1; for the analysis, only the VAS of the most severely affected leg was taken into account.

Quality of life was measured every 6 months by the validated French versions of the generic EuroQol 5D instrument²⁰ and the vein-specific self-administered Chronic Venous Insufficiency Questionnaire 2 (CIVIQ2), a disease-specific quality-of-life instrument dedicated to CVDs and validated in French for this condition.²¹ In addition to the overall CIVIQ2 scale, its four components (pain, physical, psychological, and social components) were separately analyzed.

Data regarding the direct cost of medical and nursing care were recorded for subsequent medicoeconomic analysis, but are not reported in this article.

At each visit, any adverse events were recorded and any change in the patient's treatment. The diagnosis of erysipelas²² and venous thromboembolic events were blindly evaluated by the critical event committee in the same way as for the leg ulcers.

All data (including images) were recorded using an electronic case report form with automated controls (Clininfo) to minimize data capture errors and to allow immediate quality control.

The study flow chart is shown in Fig 2. Out of the 780 patients screened for inclusion, most (746) were in response to advertisements in the regional press, and 355 of these were found to be ineligible mainly because they did not show C4 or C5 CVD (n = 271); were not available for a spa treatment (n = 35); refused (n = 7); had a contraindication to spa treatment (n = 10); had an open (n = 8) or a recent leg ulcer (n = 5); had peripheral arterial disease, peripheral neuropathy, or recent erysipelas (n = 11); or had a recent (n = 6) or planned (n = 19) venous intervention or had recent spa treatment (n = 2).

Fig 2

Study flow chart. CVDs, Chronic venous disorders.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

The comparison of randomized subjects is shown in Table I. Similar distributions of demographic and prognostic as well as baseline criteria were found in both groups.

Nine patients in the “spa treatment” group did not attend their spa course; nevertheless, they are included in the analysis (intention-to-treat approach). The delay between randomization and spa therapy ranged from 3 to 74 days (median, 24 days). The most frequently attended spa therapy sessions were the whirlpool bath (88%), the in-water walking sessions (81%), the underwater massage (79%), and the light shower massage (28%). Sixty-six percent attended a therapeutic education program. Appendix I (online only) gives further details of the spa treatments.

Forty-six patients from the control group did not attend the spa treatment course, whereas 29 subjects from the treated group attended a second spa course at their own initiative (nine before the end of the first year), because they thought they had benefited from this treatment and did not want to wait for 2 years (due to the seasonal activity of the spa resorts) before repeating it. These deviations had little effect on the outcome criteria, since they mostly occurred after the planned 12-month comparison period; however, they could substantially influence the observations during the poststudy follow-up (months, 13-18).

No difference regarding the 1-year incidence of leg ulcers was found between the groups, or between their C4 and C5 subgroups (Table II).

The physical state of veins, as measured by the modified VCSS, was found to be significantly improved after 1 year in the spa treatment group (Spa: −1.2; 95% confidence interval [CI], −1.6 - −0.8) compared to controls (Ctr: −0.6; 95% CI, −1.0 - −0.2; P = .040; Fig 3).

Fig 3

Time course of variations in the Venous Clinical Severity Score (modified VCSS) (mean ± standard error of the mean) (solid lines = comparative period; dotted lines = follow-up) At *: P < .05 for the difference with M0. At the end of 1 year, the improvement in VCSS was significantly greater in treated patients than in controls. After 1 year, the control patients were given the spa treatment and their VCSS scores caught up with those of the other group.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Leg symptoms, as assessed by monthly self-evaluation using the VAS, were considerably improved, with a maximal effect between months 4 and 6 (Spa: M6-M1 = −0.88; 95% CI, −1.36 - −0.40; Ctr: M6-M1 = +0.14; 95% CI, −0.16 - +0.44; P < .001) and remained significantly improved 1 year after treatment (Spa: M12-M1 = −0.03; 95% CI, −0.57 - +0.51; Ctr: M12-M1 = +0.87; 95% CI, +0.46 - +1.26; P = .009; Fig 4).

Fig 4

Time course of variations in leg symptoms as assessed by monthly visual analog scale (VAS) (mean ± standard error of the mean) (solid lines = comparative period; dotted lines = follow-up). At **: P < .01 for the difference with M0. After spa therapy, the treated group soon experienced a reduction in leg symptoms, which reached a maximum at 4 to 6 months. The difference with the control group remained substantial and highly significant for the entire year. After month 12 and their own spa treatment, the control patients also showed a dramatic improvement in their symptoms.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Vein-related quality of life, as expressed by the CIVIQ2 scale, for which the score decreases when quality of life improves, also improved significantly at 6 months (Spa: M6-M0 = −4.77; 95% CI, −6.88 - −2.66; Ctr: M6-M0 = −0.09; 95% CI, −2.17 - +1.99; P = .002) and remained significantly improved at 12 months (Spa: M12-M0 = −2.0; 95% CI, −4.4 - +0.4; Ctr: M12-M0 = +2.4; 95% CI, +0.2 - +4.7; P = .008; Fig 5). All four components of the CIVIQ2 scale showed similar patterns but only the physical and the pain subscales showed significant differences at month 12 (Fig 6).

Fig 5

Time course of variations in the venous-related quality-of-life scale Chronic Venous Insufficiency Questionnaire 2 [CIVIQ2] (mean ± standard error of the mean). Higher scores are related to larger changes in quality of life (solid lines = comparative period; dotted lines = follow-up). At **: P < .01 for the difference with M0. In spite of seasonal variations (the spa treatment is performed mostly during summer months), the treated group showed a substantial and significant improvement vs the control group, with the difference remaining significant at month 12. Again, after the control patients were given spa treatment, their CIVIQ2 scale reached the same level as the other group.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Fig 6

Time course of variations in the four components of the Chronic Venous Insufficiency Questionnaire 2 (CIVIQ2) scale (solid lines = comparative period; dotted lines = follow-up). At *: P < .05; and at **: P < .01 for the differences with M0. All four subscales of the CIVIQ2 quality-of-life instrument, related to psychological (A), physical (B), pain (C), and social (D) components, were significantly improved in the treated group compared to the controls, with similar time course patterns.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

General health-related quality of life, as measured by the EuroQol 5D scale also showed a similar difference at month 6 (Spa: M6-M0 = +0.01; 95% CI, −0.02 - +0.05; Ctr: M6-M0 = −0.05; 95% CI, −0.09 - −0.02; P = .003), remaining significant after 1 year (Spa: M12-M0 = +0.01; 95% CI, −0.02 - +0.04; Ctr: M12-M0 = −0.07; 95% CI, −0.10 - −0.04; P < .001; Fig 7).

Fig 7

Time course of variations in the generic quality-of-life index EuroQol 5D (solid lines = comparative period; dotted lines = follow-up). At **: P < .01 for the difference with M0. The most important variations are seen in the control group, which deteriorates over time, showing a significant difference with the treated group at month 12, but improves after their own spa treatment.

 

View Large Image | View Hi-Res Image | Download PowerPoint Slide

The control group responded in a similar manner to the treatment group to its own postponed spa treatment course (immediately after month 12), with approximately the same magnitude of improvement at month 18 as in the treatment group at month 6, for physical status, symptoms, and quality of life (Fig 3, Fig 4, Fig 5, Fig 6, Fig 7).

Compression stockings were used by 76% of patients in both groups as part of their “usual treatment” (see Table I and Appendix II, online only, for details of use during follow-up).

We found spa treatment to have no effect on the incidence of leg ulcers. Several factors may explain this. (1) The annual incidence of leg ulcers in the control group (6.1%) was more than three times lower than expected (20%, ie, 30% in C5 and 10% in C4 patients) from the literature,²³ and this considerably reduced the power of the study for the primary outcome. (2) The patients enrolled in this study showed good compliance to compression therapy (76%) meeting one of the aims of the American Venous Forum program on the prevention of venous ulcers (75% compliance within 5 years).²⁴ Good compliance could explain both the low incidence of ulcers and the difficulty to demonstrate any additional therapeutic benefit on top of such well-conducted “usual treatment.” (3) Spa treatment is a multifaceted health intervention whose educational component may help prevent leg ulcers by improving compliance to usual therapies such as compression therapy. Nevertheless, in this study, spa treatment as currently practiced in France provided no improvement in the incidence of leg ulcers per se at 1 year.

However, for the secondary criteria this large-scale multicenter randomized controlled trial confirms our previous finding in a single-center trial that spa therapy is beneficial as adjunctive treatment to the usual medical care of patients with severe CVD. Our results clearly show a significant improvement in the CVD-related physical status (VCSS), symptoms (VAS), and quality of life (CIVIQ2 and EuroQoL 5D scales). These effects are sustained for at least 1 year.

This trial was not double-blinded. This may be feasible for a comparison of the effects of different mineral waters or for assessing a technical parameter of balneotherapy, but not for evaluating the overall effect of the spa treatment course, which was our aim. Nevertheless, we made all possible efforts to avoid any evaluation bias, with centralized randomization bypassing the investigator, and blinded data management, critical event evaluation and statistical analysis.

The unusual design of the study (Fig 1), with a spa treatment given to the control group following the 1-year comparison period, was adapted from studies of spa therapy for severe low back pain.²⁵ In our previous smaller study,¹⁷ we had found that this approach avoided dropouts in the control group, and we are, thus, able to reasonably rule out a disappointment bias in the control group. The treatment tested was the course of spa therapy taken as a whole, as defined and reimbursed by French health insurance. The results combine the effects of each component of the therapy and their potential synergy but do not allow us to differentiate between the effects of each component, including the possible effects of the nature of the mineral water,²⁶ the therapeutic education,¹⁶ and the psycho-sociological influence of a 3-week health-centered retreat, as well as the intensive active balneotherapy.

In the absences of contraindications,¹⁵ tolerance was very good. Thus, a spa treatment course could be proposed to all patients with advanced chronic venous insufficiencies who are able to invest their time in it, particularly in cases with post-thrombotic syndrome, when the restoration of satisfactory hemodynamic venous function may be difficult to achieve.