Infrared-Mediated Hyperthermia Is Effective in the Treatment of Scleroderma-Associated Raynaud's Phenomenon

confidence interval infrared A modified Rodnan skin scores Raynaud's phenomenon standard deviation visual analog scale To the Editor: Scleroderma is a systemic autoimmune disease featuring variable organ involvement as well as Raynaud's phenomenon (RP). To date, no studies exist on the effect of systemic body temperature elevation on the severity of RP. Water-filtered near-infrared (infrared A (IRA)) irradiation is particularly effective in transdermal heat delivery (Meffert and Sonnichsen, 1974Meffert H. Sonnichsen N. Rewarming, heat conduction ability and blood circulation of the skin in healthy subjects and scleroderma patients.Dermatol Monatsschr. 1974; 160: 282-290PubMed Google Scholar; Meffert et al., 1989Meffert H. Scherf H.P. Baumler H. et al.Systemic effects of ultraviolet, visible and infrared radiation in serial whole body irradiation. I. Oxygen utilization, flow properties of blood, hemodynamics, blood components and phagocytosis.Dermatol Monatsschr. 1989; 175: 609-622PubMed Google Scholar; Meffert and Meffert, 2000Meffert B. Meffert H. Optical radiation and its effect on the skin.Biomed Tech (Berlin). 2000; 45: 98-104Crossref PubMed Scopus (8) Google Scholar). Prompted by preliminary findings (Meffert et al., 1990Meffert H. Buchholtz I. Brenke A. Mild infrared a hyperthermia in treatment of systemic scleroderma.Dermatol Monatsschr. 1990; 176: 683-686PubMed Google Scholar), we here examined the effect of IRA treatment on RP. We employed fingertip rewarming in response to cold challenge (Wise et al., 2004Wise R.A. Wigley F.M. White B. et al.Efficacy and tolerability of a selective alpha(2C)-adrenergic receptor blocker in recovery from cold-induced vasospasm in scleroderma patients: A single-center, double-blind, placebo-controlled, randomized crossover study.Arthritis Rheum. 2004; 50: 3994-4001Crossref PubMed Scopus (60) Google Scholar; Foerster et al, in press) as well as a clinical activity score (Merkel et al., 2002Merkel P.A. Herlyn K. Martin R.W. et al.Measuring disease activity and functional status in patients with scleroderma and Raynaud's phenomenon.Arthritis Rheum. 2002; 46: 2410-2420Crossref PubMed Scopus (247) Google Scholar) as outcome variables. In addition, we explored the effect of IRA treatment on skin thickness and scleroderma-associated joint pain. Methods, a study flow chart, as well as immediate effects of IRA treatment are detailed in the online supplement. Because of the specified inclusion criteria (see Tables S1–S3), patients were in stable-disease phase. Download .jpg (.03 MB) Help with files Table S1Study flow chart. Fingertip rewarming after a defined cold challenge was determined using a recently described device (Foerster et al, in press). The variable τ denotes the time elapsed until 63% of pre-cooling fingertip temperature has been regained. Figure 1 shows that the median cold response, though varying between individuals, is significantly improved after only one IRA treatment and continues to improve until the tenth treatment. After the end of treatment, this effect is steadily reversed but still detectable 6 wk after the last treatment. Importantly, there was no correlation between the τ-values obtained and ambient temperature (not shown). Table I summarizes the corresponding mean τ-values for the entire cohort as well as the subgroup presenting with digital ulcers. The latter subgroup exhibited slight, but not statistically significant increased mean τ-values. Thus, IRA-mediated hyperthermia has a significant effect on fingertip rewarming after cold challenge. Moreover, we also found that the improvement to be expected from a series of IRA treatments for a given individual can be predicted based on a single treatment (Figure S1, Figure S2).Table IResponse to cold challengeaFingertip rewarming after cold challenge was determined as detailed in Methods. The τ-value represents the time elapsed before approximately 63% of pre-cooling temperature is regained. Data shown represents n=56 individuals.Before IRAAt first IRAAfter 5 × IRAAfter 10 × IRA1 wk post3 wk post6 wk postAll patients (n=58)Mean±SD8.39±4.546.00±3.455.07±4.124.53±4.185.21±3.345.58±4.135.36±3.1595% CI7.24–9.545.01–6.554.01–6.133.44–6.014.25–6.184.52–7.054.44–6.27p-valuebOne-sided paired Student's t test.0.001<0.001<0.001<0.0010.003<0.001Patients with digital ulcers (n=15)Mean±SD9.01±4.076.35±3.085.41±3.235.31±2.315.53±2.245.23±2.276.03±3.0895% CI6.56–11.065.00–8.103.58–7.244.12–6.504.37–7.084.06–6.414.24–7.41p-valuebOne-sided paired Student's t test.0.0790.0220.0110.0190.0080.038IRA, infrared A; CI, confidence interval; SD, standard deviation.a Fingertip rewarming after cold challenge was determined as detailed in Methods. The τ-value represents the time elapsed before approximately 63% of pre-cooling temperature is regained. Data shown represents n=56 individuals.b One-sided paired Student's t test. Open table in a new tab Download .jpg (.03 MB) Help with files Figure S1Effect of near-infrared (IRA) hyperthermia on fingertip cold-challenge. Download .jpg (.03 MB) Help with files Figure S2Time course of scleroderma-associated variables improved by IRA-mediated hyperthermia.. IRA, infrared A; CI, confidence interval; SD, standard deviation. We assessed patient self-assessment of RP severity by a previously validated visual analog scale (VAS) (details in the online supplement). A statistically significant reduction of subjectively felt RP severity is observed after five IRA treatments (Table II). The maximum VAS reduction is observed at the end of treatment, remaining detectable 6 wk after the last treatment, thus closely paralleling the cold challenge response. The magnitude of VAS reduction (29.1% in the unstratified cohort; 44% in patients with ulcers), as well as the reversal noted in the observational phase suggest that the observed VAS changes are unlikely because of placebo effect alone.Table IIVisual analogue scale (VAS)—Raynaud's phenomenonaThe VAS was administered as detailed in Methods at the time points indicated in the table. Data shown represent mean±SD for n=58 patients.At first treatmentAfter 5 × IRAAfter 10 × IRA1 wk post3 wk post6 wk postAll patients (n=58)Mean±SD1.17±0.710.94±0.680.83±0.580.94±0.720.90±0.710.94±0.6295% CI0.98–1.350.76–1.120.68–0.990.75–1.130.71–1.080.77–1.10% reduction19.729.119.723.119.7p-valuebCompared with pre-treatment value in a one-sided paired Student's t test.0.002<0.0010.0040.0030.005Patients with digital ulcers (n=15)Mean±SD1.28±0.660.88±0.680.71±0.590.76±0.70.79±0.660.73±0.5595% CI0.95–1.620.53–1.220.4–1.020.39–1.130.45–1.140.44–1.01% reduction31.344.540.638.343.0p-valuebCompared with pre-treatment value in a one-sided paired Student's t test.0.0020.0010.0110.0110.001IRA, infrared A; CI, confidence interval; SD, standard deviation.a The VAS was administered as detailed in Methods at the time points indicated in the table. Data shown represent mean±SD for n=58 patients.b Compared with pre-treatment value in a one-sided paired Student's t test. Open table in a new tab IRA, infrared A; CI, confidence interval; SD, standard deviation. Modified Rodnan skin scores (MRSS) were obtained in order to detect changes in skin thickness. The mean MRSS decreased by 25.4% of pre-treatment values at the end of IRA treatment and remained at this level throughout the observational phase (Table III). In patients with diffuse scleroderma a maximum mean MRSS reduction of 28% was observed at 3 wk post-treatment. As expected, MRSS scores were significantly lower in patients with limited scleroderma. Nevertheless, a similar magnitude of MRSS reduction upon IRA treatment was found in this subgroup as well. Of note, MRSS scores did not decrease further during the observational phase, suggesting that the score changes did not represent spontaneous improvements in the trial setting. Thus, these data indicate that IRA-mediated hyperthermia may exert a transient beneficial effect on skin thickness.Table IIISkin thickness before and after IRA treatmentaThe modified Rodnan skin thickness score was obtained as detailed in Methods.BeforeAfter 5 × IRAAfter 10 × IRA1 wk post3 wk post6 wk postAll patients (n=58)Mean±SD12.9±9.311.3±8.49.6±7.59.7±7.99.3±6.89.8±7.395% CI10.5–15.39.1–13.47.6–11.67.6–11.87.5–11.17.9–11.8% reduction12.425.624.827.924.0p-valuebAs compared with pre-treatment value by a paired two-sided Student's t test.<0.001<0.001<0.001<0.001<0.001Limited SSc (n=31)Mean±SD7.7±5.46.5±4.65.8±4.45.7±4.55.7±4.46.1±4.795% CI5.8–9.64.8–8.14.3–7.44.1–7.34.2–7.34.5–7.8% reduction15.624.726.026.020.8p-valuebAs compared with pre-treatment value by a paired two-sided Student's t test.0.001<0.001<0.001<0.001<0.001Diffuse SSc (n=27)Mean±SD18.7±9.416.6±8.414.2±8.114.5±8.513.6±6.614.3±7.395% CI15.2–22.313.4–19.811–17.311.1–17.811.1–16.211.4–17.2% reduction11.224.122.527.323.5p-valuebAs compared with pre-treatment value by a paired two-sided Student's t test.<0.001<0.001<0.001<0.0010.003IRA, infrared A; CI, confidence interval; SD, standard deviation.a The modified Rodnan skin thickness score was obtained as detailed in Methods.b As compared with pre-treatment value by a paired two-sided Student's t test. Open table in a new tab IRA, infrared A; CI, confidence interval; SD, standard deviation. We quantified joint complaints in scleroderma patients using the DAS28 score (see online Methods). DAS28 scores decreased slightly, but statistically significant after five treatments (Table IV). As the unstratified patient cohort contains patients with no or minimal joint complaints at baseline we also analyzed the subgroup positive for rheumatoid factor. The reduction in arthralgia score is even more pronounced in this subgroup, exhibiting a reversal beginning 3 wk post-treatment. The observed reduction of arthralgia severity was not because of an effect on systemic inflammatory status, as the erythrocyte sedimentation rate, which is included in the DAS28, did not change during IRA treatment (not shown). The magnitude of DAS28 reduction seen in the present cohort by far exceeds changes attributable to placebo effects based on several placebo-controlled studies (e.g., (Gerlag et al., 2004Gerlag D.M. Haringman J.J. Smeets T.J. et al.Effects of oral prednisolone on biomarkers in synovial tissue and clinical improvement in rheumatoid arthritis.Arthritis Rheum. 2004; 50: 3783-3791Crossref PubMed Scopus (123) Google Scholar; McCarey et al., 2004McCarey D.W. McInnes I.B. Madhok R. et al.Trial of atorvastatin in rheumatoid arthritis (TARA): Double-blind, randomised placebo-controlled trial.Lancet. 2004; 363: 2015-2021Abstract Full Text Full Text PDF PubMed Scopus (707) Google Scholar; Choy et al., 2005Choy E.H. Kingsley G.H. Khoshaba B. Pipitone N. Scott D.L. A two year randomised controlled trial of IM depot steroids in patients with established rheumatoid arthritis who have shown an incomplete response to disease modifying anti-rheumatic drugs.Ann Rheum Dis. 2005; 64: 1288-1293Crossref PubMed Scopus (51) Google Scholar), especially given the 11-wk study period. Thus, IRA-mediated hyperthermia transiently improves arthralgia in scleroderma patients harboring joint complaints.Table IVJoint involvement before and after IRA treatmentaThe arthralgia score DAS28 was employed to quantify joint complaints (see Methods).BeforeAfterAfter 5 × IRAAfter 10 × IRA1 wk post3 wk post6 wk postAll patients (n=58)Mean±SD4.2±1.54.1±1.53.8±1.43.8±1.33.9±1.33.8±1.33.9±1.595% CI3.8–4.63.7–4.53.4–4.23.5–4.13.5–4.23.5–4.23.5–4.3p-valuebAs compared with pre-treatment value by a paired two-sided Student's t test.NS0.0190.0210.0260.0080.021Rheumatoid factor-positive patients (n=21)Mean±SD4.8±1.34.9±1.34.5±1.24.2±1.04.2±0.94.5±1.04.5±1.295% CI4.3–5.44.3–5.43.9–5.03.8–4.73.8–4.64.0–4.94.0–5.0p-valuebAs compared with pre-treatment value by a paired two-sided Student's t test.NS0.0170.0030.0050.0600.019IRA, infrared A; CI, confidence interval; SD, standard deviation.a The arthralgia score DAS28 was employed to quantify joint complaints (see Methods).b As compared with pre-treatment value by a paired two-sided Student's t test. Open table in a new tab IRA, infrared A; CI, confidence interval; SD, standard deviation. Changes in the health assessment questionnaire paralleled the results described above (see online supplement). Furthermore, we did not observe IRA-mediated hyperthermia on systemic inflammatory status. Unexpectedly, however, lung diffusion capacity improved in those patients presenting with ≤75% of predicted baseline DLCO. The mean % DLCO in this subgroup increased from 61.3±11.3 to 66±15.9 (p=0.005; two-sided paired t test) at 6-wk post-treatment. This study demonstrates that IRA-mediated hyperthermia reduces the severity of RP. Two independent outcome variables (fingertip cold response and VAS-RP) confirm this conclusion. Remarkably, the effect is detectable at least 6 wk after the end of IRA treatment, suggesting that this treatment may be of considerable practical value, especially in light of its favorable safety profile. Regarding the effect of IRA on skin sclerosis, documented levels of intra- and inter-observer variability (see online Methods) rule out artifacts. In a recent trial of relaxin, patients with a baseline MRSS≥20 exhibited a mean MRSS reduction from 27.5 to 20.0 within 12 wk (placebo: from 26.7 to 24.4;Seibold et al., 2000Seibold J.R. Korn J.H. Simms R. et al.Recombinant human relaxin in the treatment of scleroderma. A randomized, double-blind, placebo-controlled trial.Ann Intern Med. 2000; 132: 871-879Crossref PubMed Scopus (229) Google Scholar). When analyzing this particular subgroup (baseline MRSS≥20) in the present cohort, a mean MRSS reduction from 28.6 to 21.8 is found in this 11-wk study period, suggesting that mild hyperthermia may harbor a comparable therapeutic potential. As the MRSS score has been proposed as a marker for natural disease progression (Steen and Medsger, 2001Steen V.D. Medsger Jr, T.A. Improvement in skin thickening in systemic sclerosis associated with improved survival.Arthritis Rheum. 2001; 44: 2828-2835Crossref PubMed Scopus (182) Google Scholar), our data also raise the question as to whether long-term IRA-mediated hyperthermia may be able to modify natural disease progression. Although joint complaints are frequent in scleroderma patients (La Montagna et al., 2005La Montagna G. Sodano A. Capurro V. Malesci D. Valentini G. The arthropathy of systemic sclerosis: A 12 month prospective clinical and imaging study.Skeletal Radiol. 2005; 34: 35-41Crossref PubMed Scopus (71) Google Scholar), appropriate scores have not been validated. Therefore, we employed the DAS28 index, which has been proposed as outcome for joint involvement in scleroderma (Akesson et al., 2003Akesson A. Fiori G. Krieg T. van den Hoogen F.H. Seibold J.R. Assessment of skin, joint, tendon and muscle involvement.Clin Exp Rheumatol. 2003; 21: S5-S8PubMed Google Scholar). The rheumatoid-factor-positive subgroup that we analyzed may in fact include a significant subgroup of patients fulfilling diagnostical criteria for RA (Misra et al., 1995Misra R. Darton K. Jewkes R.F. Black C.M. Maini R.N. Arthritis in scleroderma.Br J Rheumatol. 1995; 34: 831-837Crossref PubMed Scopus (56) Google Scholar). The effect of IRA treatment on scleroderma-associated arthralgia was highly significant, especially considering the 11-wk study period. Therefore, IRA-mediated hyperthermia should be evaluated for use in RA. In conclusion, IRA-mediated mild hyperthermia is effective for the treatment of scleroderma-associated RP and may be therapeutically effective for other disease manifestations. Based on these data, further randomized-controlled studies are warranted to establish whether a long-term maintenance treatment may be able to sustain treatment responses. If so, mild hyperthermia may offer a valuable complementary treatment when drug side effects are limiting. All data obtained from human subjects in this study were subject to prior approval by the Charité institutional review board and were carried out in accordance to the Helsinki declaration. All patients gave written informed consent prior to enrolling in the study. These studies were supported by the Deutsches Netzwerk für Sklerodermie. The Iratherm 1000 infrared bench was kindly made available by the von Ardenne institute. The following material is available online for this article. Figure S1. Effect of near-infrared (IRA) hyperthermia on fingertip coldchallenge. Figure S2. Time course of scleroderma-associated variables improved by IRA-mediated hyperthermia. Table S1. Study flow chart Table S2. Patient cohort Table S3. General scleroderma activity before and after IRA treatment. Supplemental text Detailed Methods section, description of immediate effects of IRA treatment, data on the prediction of therapy response, as well as additional references. Download .doc (.11 MB) Help with doc files Supplemental textDetailed Methods section, description of immediate effects of IRA treatment, data on the prediction of therapy response, as well as additional references.