Effect of Monochromatic Infrared Energy on Quality of Life and Intraepidermal Nerve Fiber Density in Painful Diabetic Neuropathy: A Randomized, Sham Control Study
Keywords: Diabetic neuropathy, intraepidermal nerve-fiber density, neuropathic pain, norfolk quality of life, visual analog scale
Distal symmetrical polyneuropathy (DSPN) is by far the most common type of diabetic neuropathy, which affects more than 90% of the patients.,, DSPN is usually confirmed by evaluating large-fiber sensations with vibration perception or nerve conduction studies. However, the presence of small-fiber neuropathy portends more significance, as structural and functional small fiber changes precinct large-fiber changes and are causative for neuropathic pain, loss of protective sensation, foot ulceration, charcot foot, and consequent higher mortality.,,,,, Small-fiber neuropathy may manifest as painful diabetic peripheral neuropathy (DPN) that is characterized by symptoms of burning, tingling, lancinating, sharp, and shooting pain, generally worse at night causing insomnia.,, The pain can be persistent and influences the activities of daily living that may affect the quality of life (QoL). As a result, people affected with neuropathic pain have depressed mood, refrain from social and recreational activities that exacerbates their disability.,
A wide variety of drugs, used alone or in combination, have been shown to reduce neuropathic pain compared with placebo in randomized controlled trials.,,, However, the pharmacological treatment for diabetic neuropathy is predominantly targeted at symptomatic relief that is inadequate and not focused on the pathophysiological mechanism, limited by side effects, marked by the development of tolerance and significantly add to economic burden of diabetes treatment.,,,, Therefore, various adjunctive therapies including monochromatic infrared energy (MIRE) have been used for DPN., Few initial studies demonstrated a significant decrease in neuropathic pain, and improvement in tactile sensitivity with MIRE therapy,,, while others have not shown change in symptom scores, or plantar sensitivity. However, most studies with MIRE therapy had methodological limitations as they were retrospective, lacked a control group, treatment was not supervised, and were of short duration (2-4 weeks).
Moreover, intraepidermal nerve-fiber density (IENFD) which is an objective and earliest marker of small fiber neuropathy that is implicated for neuropathic pain and loss of protective sensation in people with diabetes has not been studied earlier. Therefore, we assessed the effect of MIRE therapy on IENFD and also on pain symptom and QoL in patients with painful DPN.
Patients with self-reported diabetes with history and clinical features of painful DPN, of either gender, age 18 years and above, on stable oral hypoglycemic drugs and/or insulin for the preceding 3 month, were included in the study after an explained and written informed consent. The study was incited after the approval of the Institutional Ethics Committee. The diagnosis of painful DPN was based on symptoms and signs of small fibre neuropathy as defined by the “Toronto Diabetic Neuropathy Expert Group” recommendations that include diabetic neuropathy symptom (DNS) score of more than 2, and/or (b) neuropathy disability score (NDS) of more than 6. DNS score is a symptom score that consists of four questions related to unsteadiness on walking, burning sensation, pricking sensation, and numbness on foot. The presence of each symptom is scored as 1 and absence as 0 to have a maximum score of 4. DNS has been validated with a high predictive value when screening for DPN. NDS is a clinical scoring system for neuropathy signs consisting of four items of vibration, temperature, pin prick sensation, and ankle reflex. A NDS score of more than 6 out of a maximum of 10 is associated with increased risk of neuropathic foot ulcer.
Participants with uncontrolled diabetes defined as HbA1c >8%, peripheral vascular disease (Ankle-Brachial Index < 0.9 and/or claudication), chronic kidney disease (eGFR <30 ml/min/m2), depression, malignancy, knee and back surgeries, dermatological diseases like leprosy and connective disuse disorders, familial neuropathies, on medications associated with peripheral neuropathy and pregnant women were excluded from the protocol.
All patients were engaged in an initial 2-week run-in period in order to ensure their compliance for follow-up visits after the withdrawal of pre-existing medication for painful DPN. Paracetamol was used as a rescue medication, if needed. After the run-in period of two weeks, patients were allocated MIRE (active) or sham (placebo) therapy in 1:1 simple randomization design. The groups were coded A (MIRE) and B (sham) in two envelopes and the participants were asked to pull out the envelope for group allocation. The investigator, the patient, and the outcome assessor were blinded to treatment allocation.
Active and sham cohorts were screened with DNS and NDS and further evaluated with Michigan neuropathy screening instrument (MNSI) that consists of symptom questionnaire (MNSI-Q) and clinical examination (MNSI-P) scores. The MNSI questionnaire provides a graded response of neuropathic symptoms, so that a higher score represents more neuropathic symptoms with a maximum score of 25 (15 – history, 10 – physical examination).
A 10 g (5.07) Semmes Weinstein monofilament test was employed for testing loss of protective sensation as a part of MNSI clinical examination and vibration perception threshold (VPT) using biothesiometry [Vibrometer-VPT® (Diabetic Foot Care, Madras Engineering Service, India)]. VPT was assessed at the distal part of the plantar surface of great toe, first, third, and fifth metatarsal head, midsole, and heel. The voltage was gradually increased at the rate of 1 mV/s and the VPT score was denoted as the level of voltage at which the patient indicates that he or she first perceived the sense of vibration. The mean of six records was obtained and neuropathy was considered if VPT was ≥25 mV (at any site).
Monofilament was tested at five standard sites as per prior recommendations of plantar aspect of great toe and the first, third, and fifth metatarsal head and the heal.
A 10-point visual analogue scale (VAS) was used to assess the discomfort due to painful diabetic neuropathy. We used Norfolk QOL-DN instrument for QoL as it is a validated nerve fiber-specific questionnaire for evaluating the overall QoL and impact on the functional status and activities of daily living of patients with peripheral neuropathy.
Skin biopsy for IENFD
A 3 mm skin punch biopsy was performed 10 cm proximal to lateral malleolus in the limb which scored worst on former evaluation tools. The skin biopsy specimen was divided into two halves for Haematoxylin and Eosin (H and E) stain and immunohistochemistry (IHC)/immunofluorescence (IF). Tissue sections were fixed in 10% buffered formalin processed and embedded in paraffin, following a standard protocol for H and E stain. The other half was stored at -20°C for IHC/IF. Sections were cut at 50 μm sections on precoated slides with poly-L-lysine. The paraffin/frozen sections were deparaffinized in xylene and then hydrated gradually by washing in 100% ethanol twice for 10 min each, then in 95% ethanol twice for 10 min each, and lastly in deionized H2O for 1 min with stirring. The slides were incubated with primary antibody–protein gene product 9.5 (PGP 9.5) (Abcam, UK) a neuronal ubiquitin carboxy terminal hydrolase, for 30 min at room temperature or overnight at 4°C. Quantification of IENFD was done by counting the density of nerve fibers that cross the dermo-epidermal junction per 1 mm in at least two to three nonadjacent sections procured from different sites within the specimen as a standardized protocol.
A single MIRE unit was used in this study – Anodyne Model 480 – Infrared Therapy System (Medassist, Tampa, FL). The device has a main power unit with four flexible therapy pads; each pad measures 3.0 × 7.5 cm and contains 60 super luminous gallium–aluminium arsenide diodes that emit light energy in the near-infrared spectrum (890-nm wavelength). The active treatment unit was set to deliver 1.95 J/cm2/min when activated.
The four therapy pads were placed over the following sites: 1) distal posterior leg, 2) distal anterior leg, 3) plantar foot over metatarsal heads, and 4) plantar arch of foot. For sham therapy, the pads of the MIRE device were applied to the foot similarly, but the switch was not activated and the patient was blinded to this information. The illuminating bars on the display were covered with black cellophane tape beforehand. Therapy was given for 30 min a session, three sessions per week for a total of 36 sessions over 12 weeks. Patient was deemed compliant if he/she receives 80% or above of 36 sessions.
Patients in both cohorts were subsequently evaluated for change in VAS score, Norfolk QOL-DN questionnaire score, VPT, MNSI score, and IENFD after intervention. Patients who attended 80% of the total allotted sessions of MIRE were included in the analysis. The primary outcome measure was the change in VAS, and change in the IENFD was coprimary outcome. Secondary outcomes were the change in VAS, DNS, NDS, MNSI, and Norfolk QoL-DN score after 3 months. Intergroup a difference of >20% in the change in VAS score between two groups was sought to be the primary outcome measure for sample size calculation as per IMPACT recommendations. Intragroup >50% decrease in VAS was considered as substantial and clinically significant improvement, lest statistical significance.
Considering the prevalence of peripheral neuropathy of 40% in people with diabetes and the that of painful diabetic neuropathy accounting for 8–10% of all patients of DPN, demonstrating at least 20% difference in the primary outcome measure, between the intervention and sham cohort at 80% power and with a two-tailed alpha error of 0.05, a sample size of 24 participants was obtained. Considering 20% drop out, a total sample size of 30 participants was considered. A modified intention-to-treat analysis was performed. Mean ± SD was used for parametric data and median with interquartile range for nonparametric data. Baseline parameters were compared by paired Student's t-test, Chi-square test, or Fisher exact test. The patient's VAS, Norfolk QOL-DN questionnaire scores, MNSI-questionnaire and physical exam scores, VPT score, NDS, and DNS scores were compared between the groups using the Mann–Whitney and Wilcoxon matched pair test to evaluate the changes in cohort scores before and after therapy. A P value < 0.05 was taken as significant. SPSS version 22 was employed to perform statistical analysis.
Eighty patients were screened, of which 45 patients satisfied the inclusion criteria. Seven patients did not provide the consent; hence, 38 patients went through the run-in period. Thirty patients were randomized as four patients did not report after run-in period, two withdrew consent and two patients could not tolerate the drug withdrawal phase in view of intolerable neuropathic symptoms. The detailed CONSORT flow diagram is shown in [Figure 1]. The baseline characteristics of the 30 enrolled participants are shown in [Table 1]. The mean duration of diabetes was 12.9 ± 3.1 years, and duration of painful DPN symptoms was 3.9 ± 3.7 years in the MIRE cohort. All patients were on metformin except one patient.
The IENFD was 0.90 ± 0.73 mm, 0.6 ± 0.89 mm (P = 0.43) at baseline; and 1.71 ± 1.11 and 2.17 ± 0.98 mm (P = 0.45) after 3 months of therapy in the MIRE and sham cohort, respectively. There was an increase in IENFD of 1 ± 1.26 and 2 ± 1.15 mm (P = 0.22) with MIRE and sham cohort, respectively, at the end of 3 months [Figure 2].
The median baseline VAS was 9.0 (7–9.6) and 6.9 (6.5–8.8), P = 0.106, in MIRE and sham cohorts, respectively. VAS scores at the end of the first, second, and third month of MIRE therapy were 5.3 (4.2–7.7), 4.5 (3.2–6.1), and 2.5 (0.9–4.8) points, respectively [Figure 3] with an overall 56.7% decrease in VAS score. There was a decrease in VAS score by 3.1 (2.6–5.8) points and 2.4 (0.6–2.7) points, (intergroup difference, P = 0.01) at 2 months and 5.1 (4–7.6) points and 3 (0.4–5.6) points (intergroup difference, P = 0.04) at the end of 3 months in the MIRE and sham cohort, respectively. The median Norfolk QOL-DN scores were 24 (17–27) and 23 (13.5–25), P = 0.48 at baseline and 8 (6–12) and 12 (8.2–21), P = 0.07 at the end of 3 months in MIRE and sham cohorts, respectively [Figure 4]. The Norfolk QOL-DN score increased by 15 (11–18) points after MIRE therapy and 4 (4–14.2) in sham cohort (intergroup difference, P = 0.021).
The median baseline DNS and NDS scores were 3.0 (2.0–3.0) and 8.0 (7.0–8.0) in the MIRE cohort. No change in DNS (intergroup difference P = 0.47) and NDS scores (intergroup difference P = 0.51) were observed after 3 months of MIRE or sham therapy. The median MNSI-Q and MNSI-P scores at baseline were 6.0 (5.0–8.0), 4.0 (3.0–4.0) and 5.0 (5.0–6.0) and 3.5 (2.2–5.0) in the MIRE cohort and sham cohort, respectively. Post therapy at 3 months, a decline in MNSI-Q of 3 (2–4) and 2 (-0.5 to 2.7) points (intergroup difference, P = 0.06) was observed in MIRE cohort and sham cohort, respectively. However, no change in MNSI-PE scores was observed with either treatment at the end of 3 month.
No adverse events specific to the interventions were observed. The MIRE therapy was well tolerated. However, four patients in the sham cohort and one patient in intervention arm required rescue medications for symptomatic neuropathic pain relief.
The results of the present sham-control study suggest that MIRE therapy do not cause a significant change in IENFD; but it is associated with a considerable decrease in the symptoms of neuropathic pain as assessed by VAS and improvement in QoL scales as assessed with Norfolk QOL scores. There was no significant change in the objective measures of neuropathy including VPT, DNS, NDS scores, MNSI questionnaire, and physical examination scores at the end of 12 weeks with intervention.
Generally, in clinical trials with pharmacological agents for DPN, the treatment is considered successful if patient would obtain 50% of reduction in the pain level. Previous studies have demonstrated a 37.1–45.2% decrease in pain on 10-point or 100-point VAS scale with 4–12 weeks of MIRE therapy., We observed a 56.7% decrease in pain scores on 10-point VAS scale with MIRE therapy, along with a significant improvement in QoL. A better symptomatic benefit in our study could be because of the longer duration of MIRE therapy (12 weeks) and a higher baseline VAS score (median VAS of 9.0) as compared to the previous studies, that used MIRE therapy for 4–6 weeks and had lower baseline VAS. Contrary to these results, Lavery et al. did not observe any effect of MIRE therapy on neuropathic pain; but the study had methodological limitations, as it was based on domiciliary MIRE therapy and not a supervised intervention. A recent meta-analysis for randomized control studies with MIRE therapy showed no difference between MIRE and comparison groups [mean difference 0.80 (95% CI: 0.30–0.68)].
MIRE therapy has also been previously evaluated for its effects on the restoration of protective sensations with the assessment of SWS monofilament perception before and after therapy.,, The results were heterogeneous as few authors observed a significant (66–80.5%) improvement in foot sensation as assessed by the number of sites sensate to SWM monofilament,,, while others could not., We also observed no apparent change in SWM sensation, VPT score, and overall NDS and DNS scores with MIRE therapy. A recent meta-analysis has demonstrated no increase in plantar regions sensitive to monofilament compared to the sham cohort. However, a subgroup analysis showed improved sensations with 2 weeks of MIRE therapy, but the effect were not sustained for the next 2 weeks. Possibly, the duration of therapy is too short to alter the vasculature, blood flow, and nerve regeneration. Also, predominantly large fibers are tested with VPT, which are unlikely to be affected by photoemissions.
The quantitative estimation of small nerve fiber damage by evaluating IENF density is the recommended sensitive modality for the diagnosis and assessment of the efficacy of interventions for peripheral neuropathy.,, Supervised exercise has been shown to improve cutaneous nerve fibee branching in DPN;, however, IENFD has not been previously studied with MIRE therapy. MIRE therapy is expected to improve the cutaneous nerve fiber density by an increase in blood flow through the generation of nitric oxide in vasa-nervosum. We studied the density of unmyelinated nerve fibres crossing the dermo-epidermal junction as a robust marker of small fiber diabetic neuropathy. IENFD at baseline in our cohort was similar to that reported previously, suggesting an adequate sampling and procedure of IENFD assessment. We observed a nonsignificant increase in IENFD in MIRE cohort at the end of 3 months with no intergroup difference. The plausible explanation for the lack of efficacy of intervention could be a longer duration of diabetes, as it is observed that the duration of diabetes is an important determinant of IENFD loss. A 1 mm loss of nerve fiber per year is observed in patients with diabetic neuropathy contributing to severe loss, which increases with increasing duration of diabetes that may be irreversible with short-term interventions. Also, short duration of MIRE therapy in the present study may have precluded the sensitivity of IENFD estimation. Peculiarly, there was a strong placebo effect in our study and most of the other previous randomized controlled studies.,, “Hawthorne effect” may explain the findings in uncontrolled studies, which showed unequivocal effectiveness of MIRE therapy.
The strengths of the study include that it is the first randomized, sham controlled study to evaluate IENFD in patients with painful DPN with close follow-up and good adherence to intervention. Further, multiple established rating scales were used to assess symptomatic changes along with objective measures of neuropathy assessment. We perceived certain limitations. A number of patients could not tolerate the withdrawal of pharmacological therapy for painful neuropathy in the placebo group and required rescue medications. The technical difficulties in processing skin biopsy for IENFD may have led to nonuniform estimation of IENFD. The duration of therapy in the present study was 12 weeks that might not be appropriate to demonstrate any perceptible changes in quantitative neuropathy assessment index of IENFD. Symptoms of autonomic dysfunction (sudomotor, vasomotor, gastrointestinal, and cardiac) were enquired but an objective assessment for autonomic function could not be performed. Though we tried to exclude other causes of painful neuropathy, serum B12 levels were not performed, despite most of the patients on chronic metformin therapy.
In conclusion, MIRE therapy alleviates pain symptoms and improves QoL in patients with painful DPN. However, MIRE therapy does not result in a significant increase in the cutaneous reinnervation as assessed with IENFD.
We thank Research Society of Study of Diabetes in India (RSSDI) for partially funding the study. AR takes full responsibility for the work as a whole, including the study design, access to data, and the decision to submit and publish the manuscript.
PU was involved in clinical care of the participants, did biopsy for IENFD, and wrote the initial draft of the manuscript; AR was involved in the design of the study, recruitment, clinical care, and follow up of the participants, writing and editing the manuscript; AB was involved in the design of the study and editing the manuscript; UNS performed the histopathology of tissue specimen and edited the manuscript.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]