Review of Duloxetine in the Management of Diabetic Peripheral Neuropathic Pain
ABSTRACT
Objective. To appraise the effectiveness of duloxetine, compared with placebo, on patient-reported wellness outcomes over a 12-week period, in the management of diabetic peripheral neuropathic pain (DPNP).
Methods. The results were pooled from three 12-week multicenter, double-bullheaded studies. In study 1 (Northward = 457), patients with DPNP were randomly assigned to treatment with duloxetine 20 mg once daily (QD), 60 mg QD, 60 mg twice daily (BID), or placebo. In studies 2 (N = 334) and three (Due north = 348), patients with DPNP were randomly assigned to handling with duloxetine 60 mg QD, lx mg BID, or placebo. Patient-reported functional outcomes were measured by the Curt Grade 36 (SF-36), the interference portion of the Cursory Pain Inventory (BPI), and EuroQol 5D Health Questionnaire (EQ-5D). Results for all functional outcomes from the intent-to-treat and completer populations are discussed.
Results. In the SF-36 health survey and the BPI interference, duloxetine lx mg QD and lx mg BID were significantly superior to placebo in all the domains (P ≤ 0.03). In the analysis of the EQ-5D, duloxetine 60 mg QD (P = 0.004) and 60 mg BID (P < 0.001) were significantly amend than placebo on all items.
Conclusions. Acute treatment with duloxetine was associated with meaning improvement in functional outcomes in persons with DPNP.
Introduction
Diabetes mellitus affects approximately 20.eight million people in the U.s.a. [1], and it has been predicted that 220 million people worldwide will be afflicted past 2010 [2]. Diabetic neuropathy, which is often painful and disabling, is one of the nearly common complications of diabetes [3], and epidemiologic studies suggest that painful diabetic neuropathy occurs in 7.5% to 24% of patients with diabetes [4–6]. Symptoms of diabetic peripheral neuropathy include tingling, burning, prickling, precipitous pains, cramps, extreme sensitivity to impact, numbness, and loss of balance and coordination [vii], and there is a considerable negative bear upon on quality of life (QOL) [8,9]. Peripheral neuropathy tin can pb to foot ulcers and amputations, which have a major bear on on QOL in people with diabetes [10–13].
Diabetic neuropathic pain was reported to interfere with general activity, mood, mobility, work, social relations, sleep, leisure activities, walking ability, and enjoyment of life [three,six,14]. Increasing levels of pain severity stand for to increasing feet, depression, sleep problems, and decreases in concrete and mental functioning [xiv]. In a prospective survey of 105 patients with diabetic neuropathic hurting, 53% reported pain on a constant, daily basis [eight]. When quantifying the impact of diabetic neuropathic pain on wellness-related QOL, Sullivan and colleagues [15] reported a decrease in patient function with increasing disease severity. Historically, diabetic peripheral neuropathic pain (DPNP) has been managed with analgesics, anticonvulsants, antidepressants, and topical capsaicin, either alone or in any combination [5,xvi].
Duloxetine hydrochloride (Cymbalta®, Eli Lilly, Indianapolis, IN), future referred to every bit duloxetine, is a selective serotonin (5-hydroxytryptamine [v-HT]) and norepinephrine (NE) reuptake inhibitor that is relatively balanced in its affinity for both 5-HT and NE reuptake inhibition [17]. Duloxetine was the beginning Food and Drug Administration (FDA)-approved prescription drug for the direction of DPNP. In iii randomized, placebo-controlled, 12-week trials, duloxetine at doses of 60 mg once daily (QD) and 60 mg twice daily (BID) was found to exist constructive and rubber in the management of patients with DPNP [18–20]. In these three studies, the efficacy of duloxetine sixty mg QD and 60 mg BID was demonstrated in various measures, including 24-60 minutes boilerplate pain severity, 24-hour worst pain score and night hurting score, McGill Pain Index score, Clinical Global Impression of Severity, Patient Global Impression of Improvement, and Brief Pain Inventory (BPI) severity [18–20]. Treatment-emergent adverse events that occurred in duloxetine-treated patients at a rate of ≥v% and twice the charge per unit for placebo were nausea, somnolence, dizziness, constipation, fatigue, hyperhidrosis, dry mouth, decreased ambition, asthenia, and anorexia. These studies have independently provided testify that duloxetine is efficacious, safe, and tolerable in the direction of DPNP. In clinical studies, duloxetine has also been shown to be prophylactic and effective in the treatment of depression [21–27], and can significantly reduce painful physical symptoms associated with major depressive disorder (MDD) [28].
This report presents a pooled analysis of patient-reported wellness outcomes data from the three randomized, double-blind, placebo-controlled clinical studies designed and conducted for registration purposes for DPNP. Health outcomes are assessed using measurements of the Medical Outcomes Study Short Form-36 (SF-36) Health Survey Questionnaire, the interference portion of the Brief Pain Inventory (BPI), and the EuroQol 5D Health Questionnaire (EQ-5D).
Methods
Overview
Information from 3 randomized, double-blind, multicenter, parallel-group, placebo-controlled studies of duloxetine for the management of DPNP were included in the assay. The ethical review boards provided approval of the report protocols in accordance with the principles of the Annunciation of Helsinki. All patients provided written informed consent after the study was explained and prior to the performance of any protocol procedures and assistants of the report drug.
Entry Criteria
As detailed in Raskin et al. [19], patients were eligible for these studies if they were ≥18 years, and presented with pain due to bilateral peripheral neuropathy caused by type I or type Two diabetes mellitus. The pain had to begin in the feet with relatively symmetrical onset. Daily pain had to be present for at least vi months, and the diagnosis confirmed past a score of ≥3 on the Michigan Neuropathy Screening Musical instrument (MNSI). Patients were required to have a hateful score of ≥four for 24-hour average pain severity on an 11-signal Likert scale (from the patient diary prior to randomization), and stable glycemic command. Patients were excluded if they were pregnant or breastfeeding, had prior renal transplant or current renal dialysis, and had a serious or unstable disease, symptomatic peripheral vascular illness, or other medical or psychological condition that might compromise participation in the report. Patients were also excluded if they had a electric current (≤1 yr) Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-Four) [29] Axis I diagnosis of MDD, dysthymia, generalized anxiety disorder, alcohol, or eating disorders as adamant by the Mini International Neuropsychiatric Interview (MINI) [thirty], or if they had a DSM-Four diagnosis or a previous diagnosis of mania, bipolar disorder, or psychosis. Other exclusion criteria included historical exposure to drugs known to crusade neuropathy, history of substance abuse or dependence within the previous yr, a positive urine drug screen for whatever substances of corruption, or excluded medication. Concomitant medication exclusions included most centrally active drugs. Acetaminophen upwards to 4 k/day was the only analgesic immune, and aspirin upwards to 325 mg/day was immune for cardiac prophylaxis.
Treatments
In written report 1, during the 12-calendar week, double-blind astute therapy phase, patients (N = 457) were randomly assigned to receive duloxetine 20 mg QD, duloxetine 60 mg QD, duloxetine 60 mg BID, or placebo. Patients assigned duloxetine 60 mg BID initiated therapy with duloxetine forty mg BID for 3 days before titrating up to the dose of threescore mg BID. The dose of 20 mg/day was included in written report one to validate dose–response for registration purposes. This dose was subsequently found to be subtherapeutic and is below the recommended dose range for both the canonical FDA (60 and 120 mg/day) and the European Medicines Evaluation Agency (lx mg/mean solar day) labels; hence, this treatment group was non included in the functional outcome analyses presented here.
In studies 2 (Due north = 334) and iii (N = 348), patients were treated in a double-blind manner for 13 weeks. The get-go 12 weeks of the study comprised the acute therapy stage with the one additional week included for recommended drug tapering. Patients who were randomized to duloxetine 60 mg BID were started at 60 mg QD for 3 days and so increased to 60 mg BID.
Wellness Outcome Measures
To assess the effectiveness of duloxetine, compared with placebo, on patient-reported health outcomes, the post-obit measures were used: the SF-36 [31,32], the interference portion of the BPI, and the EQ-5D. The SF-36 is the nigh widely used generic instrument for rating patients' self-perceived health-related QOL [31]. The original questionnaire was adult by Ware [33] and assesses both the physical and mental dimensions of QOL. The SF-36 questionnaire measures 8 categories of health [34]. These categories are: physical functioning, limitations in usual office activities due to physical wellness problems, actual pain, general health perceptions, vitality, social functioning, limitations in usual role activities because of emotional bug, and mental wellness. The replies to individual items in each domain are summed to derive a score, which is converted using an algorithm to a scale from 0 (poor health) to 100 (good health).
The BPI is a widely used and validated numeric rating calibration that measures severity of pain and its interference with daily function [35], and has been used extensively in the United States and in many clinical studies for neuropathic pain [36–39]. A primary benefit of the BPI is its ordinal rating format, which has been recommended for inquiry on neuropathic pain. The interference portion of the BPI has vii items on which patients rate how their pain interferes with enjoyment of life, activeness, walking, mood, slumber, work, and relations with others. These items are bounded by 0 = does not interfere and x = interferes completely, and the seven items are used to derive an average BPI interference score.
The EQ-5D cocky-report questionnaire is a standardized instrument for utilise as a generic measure of health-related QOL [40,41]. In this measure out, wellness condition is defined in terms of five dimensions: mobility, self-care, usual activities, hurting/discomfort, and feet/depression. Each dimension comprises three possible levels of response (i.e., one, 2, or 3), representing "no problems,""some/moderate problems," and "extreme bug." Respondents are asked to choose one level that reflects their "own wellness state today" for each of the five dimensions. An overall alphabetize measuring QOL, adjusted for country-specific social preferences, is derived from the responses to the five dimensions.
Statistical Analyses
For each health outcome measure out, least-squares mean changes from baseline to endpoint were compared between the treatment groups. The least-squares hateful change represents the adjusted hateful change for each treatment group subsequently controlling for the variability associated with all other factors included in the analysis of covariance (ancova) models. ancova models included main effects of treatment and report, with baseline values included every bit covariates. Handling-by-written report interactions were included at the 0.10 significance level. No adjustments were made for multiple comparisons, and significance, equally used throughout this article, refers to statistical significance as indicated by P ≤ 0.05.
Analyses were performed on all randomized patients as well as the subset of patients who completed the full 12 weeks of acute treatment, to place any affect with the findings that may be associated with patient withdrawal, particularly early in the study.
Results
There were no meaning differences among treatment groups in any of the patient demographics or disease characteristics, including origin, age, gender, weight, type of diabetes mellitus, elapsing of diabetes, duration of diabetic neuropathy, and MNSI (Tabular array 1). The majority of patients were male (55.9%) and Caucasian (85.five%). The mean patient age was 59.8 years. The mean duration of diabetes in all patients was 11.7 years, with blazon II diabetes being the most prevalent (88.3%). The hateful elapsing of diabetic peripheral neuropathy was 4 years. The mean MNSI score was five.two out of a possible maximum score of 10, with a higher score indicating a greater degree of neuropathy.
Table 1
Demographics and clinical characteristics at baseline
| Variable | Placebo (N = 339) | Duloxetine | P value | |
| threescore mg QD (Northward = 344) | 60 mg BID (Due north = 341) | |||
| Mean age, years (SD) | 60.1 (ten.iii) | 59.1 (11.2) | 60.3 (10.2) | 0.250 ‡ |
| Gender, Northward (%) | ||||
| Female | 158 (46.6) | 143 (41.6) | 151 (44.three) | 0.420 † |
| Male | 181 (53.4) | 201 (58.4) | 190 (55.7) | |
| Race (origin), North (%) | ||||
| Caucasian | 291 (85.8) | 293 (85.2) | 292 (85.6) | 0.475 † |
| Hispanic | 29 (8.6) | 29 (viii.four) | 35 (10.3) | |
| African descent | 16 (4.7) | eleven (3.2) | 9 (2.6) | |
| Other | three (0.9) | 11 (3.ii) | five (1.five) | |
| Mean weight, kg (SD) | 94.ix (22.4) | 94.0 (23.1) | 94.0 (22.ii) | 0.777 ‡ |
| Type of diabetes mellitus, N (%) | ||||
| Type I | 36 (10.vi) | 47 (13.7) | 37 (10.9) | 0.376 † |
| Blazon II | 303 (89.4) | 297 (86.3) | 304 (89.1) | |
| Mean duration of diabetes, years (SD) | 11.8 (9.7) | 12.0 (nine.1) | 11.iii (9.seven) | 0.654 ‡ |
| Mean duration of diabetic neuropathy, years (SD) | 3.9 (iii.6) | four.0 (4.1) | 4.i (iv.six) | 0.740 ‡ |
| Michigan Neuropathy Screening Instrument score (SD) | v.iv (1.vi) | 5.2 (one.5) | 5.2 (i.5) | 0.094 ‡ |
| Mean SF-36 full general health total score (SD) | 50.4 (20.half dozen) | 49.0 (21.6) | 49.iv (20.9) | 0.625 ‡ |
| Mean BPI boilerplate interference score | 4.one (two.iii) | 4.5 (2.3) | 4.5 (2.2) | 0.045 ‡ |
| Mean EQ-5D index | 0.59 (0.3) | 0.58 (0.three) | 0.57 (0.3) | 0.624 ‡ |
| Variable | Placebo (N = 339) | Duloxetine | P value | |
| 60 mg QD (Due north = 344) | 60 mg BID (North = 341) | |||
| Mean historic period, years (SD) | 60.1 (10.3) | 59.1 (11.2) | 60.three (10.ii) | 0.250 ‡ |
| Gender, N (%) | ||||
| Female | 158 (46.half-dozen) | 143 (41.6) | 151 (44.three) | 0.420 † |
| Male | 181 (53.4) | 201 (58.iv) | 190 (55.7) | |
| Race (origin), N (%) | ||||
| Caucasian | 291 (85.eight) | 293 (85.2) | 292 (85.6) | 0.475 † |
| Hispanic | 29 (eight.6) | 29 (eight.4) | 35 (10.3) | |
| African descent | 16 (4.vii) | 11 (3.ii) | ix (2.6) | |
| Other | 3 (0.nine) | eleven (3.2) | 5 (1.5) | |
| Mean weight, kg (SD) | 94.9 (22.four) | 94.0 (23.1) | 94.0 (22.ii) | 0.777 ‡ |
| Blazon of diabetes mellitus, N (%) | ||||
| Type I | 36 (10.6) | 47 (thirteen.seven) | 37 (10.9) | 0.376 † |
| Blazon II | 303 (89.4) | 297 (86.3) | 304 (89.1) | |
| Hateful duration of diabetes, years (SD) | xi.8 (9.7) | 12.0 (9.one) | xi.3 (nine.vii) | 0.654 ‡ |
| Mean duration of diabetic neuropathy, years (SD) | 3.9 (3.half dozen) | iv.0 (four.ane) | iv.1 (four.vi) | 0.740 ‡ |
| Michigan Neuropathy Screening Musical instrument score (SD) | 5.four (1.6) | v.2 (1.5) | 5.2 (i.five) | 0.094 ‡ |
| Mean SF-36 general health full score (SD) | 50.4 (20.6) | 49.0 (21.half-dozen) | 49.4 (20.9) | 0.625 ‡ |
| Mean BPI average interference score | 4.1 (two.3) | four.five (ii.3) | 4.5 (2.two) | 0.045 ‡ |
| Mean EQ-5D index | 0.59 (0.3) | 0.58 (0.iii) | 0.57 (0.3) | 0.624 ‡ |
† Cochran-Mantel-Haenszel examination for general association, decision-making for written report.
‡ Ways were analyzed using a master-furnishings analysis of variance (anova) model including treatment group and study.
SD = standard difference; QD = one time daily; BID = twice daily; SF-36 = Medical Outcomes Written report Short Course-36 Health Survey Questionnaire; BPI = Cursory Hurting Inventory; EQ-5D = EuroQol 5D Health Questionnaire.
Tabular array 1
Demographics and clinical characteristics at baseline
| Variable | Placebo (N = 339) | Duloxetine | P value | |
| 60 mg QD (N = 344) | threescore mg BID (Northward = 341) | |||
| Mean age, years (SD) | 60.1 (10.3) | 59.ane (eleven.2) | 60.3 (x.two) | 0.250 ‡ |
| Gender, N (%) | ||||
| Female | 158 (46.6) | 143 (41.6) | 151 (44.3) | 0.420 † |
| Male | 181 (53.iv) | 201 (58.iv) | 190 (55.7) | |
| Race (origin), N (%) | ||||
| Caucasian | 291 (85.8) | 293 (85.2) | 292 (85.6) | 0.475 † |
| Hispanic | 29 (8.half dozen) | 29 (eight.four) | 35 (10.three) | |
| African descent | 16 (four.7) | 11 (3.ii) | ix (2.vi) | |
| Other | 3 (0.9) | 11 (three.2) | 5 (ane.5) | |
| Hateful weight, kg (SD) | 94.9 (22.4) | 94.0 (23.1) | 94.0 (22.2) | 0.777 ‡ |
| Type of diabetes mellitus, Northward (%) | ||||
| Type I | 36 (10.half dozen) | 47 (xiii.vii) | 37 (ten.9) | 0.376 † |
| Blazon II | 303 (89.4) | 297 (86.3) | 304 (89.1) | |
| Mean duration of diabetes, years (SD) | 11.eight (9.7) | 12.0 (9.1) | xi.3 (9.7) | 0.654 ‡ |
| Mean duration of diabetic neuropathy, years (SD) | three.9 (3.6) | 4.0 (iv.one) | 4.ane (four.6) | 0.740 ‡ |
| Michigan Neuropathy Screening Instrument score (SD) | five.4 (1.six) | 5.2 (i.v) | 5.2 (1.5) | 0.094 ‡ |
| Mean SF-36 general health full score (SD) | 50.4 (20.6) | 49.0 (21.6) | 49.4 (20.9) | 0.625 ‡ |
| Mean BPI average interference score | iv.ane (2.3) | four.5 (ii.3) | 4.5 (2.2) | 0.045 ‡ |
| Mean EQ-5D index | 0.59 (0.3) | 0.58 (0.iii) | 0.57 (0.three) | 0.624 ‡ |
| Variable | Placebo (Northward = 339) | Duloxetine | P value | |
| threescore mg QD (Northward = 344) | threescore mg BID (Northward = 341) | |||
| Mean age, years (SD) | 60.1 (10.three) | 59.1 (11.2) | 60.three (10.2) | 0.250 ‡ |
| Gender, N (%) | ||||
| Female | 158 (46.6) | 143 (41.6) | 151 (44.iii) | 0.420 † |
| Male | 181 (53.iv) | 201 (58.4) | 190 (55.vii) | |
| Race (origin), Due north (%) | ||||
| Caucasian | 291 (85.8) | 293 (85.2) | 292 (85.vi) | 0.475 † |
| Hispanic | 29 (8.6) | 29 (8.iv) | 35 (x.3) | |
| African descent | sixteen (4.vii) | 11 (three.2) | nine (2.6) | |
| Other | 3 (0.nine) | eleven (three.2) | 5 (1.5) | |
| Mean weight, kg (SD) | 94.9 (22.4) | 94.0 (23.one) | 94.0 (22.2) | 0.777 ‡ |
| Type of diabetes mellitus, N (%) | ||||
| Type I | 36 (ten.6) | 47 (thirteen.7) | 37 (10.9) | 0.376 † |
| Blazon II | 303 (89.4) | 297 (86.3) | 304 (89.ane) | |
| Mean duration of diabetes, years (SD) | 11.viii (9.seven) | 12.0 (9.1) | xi.3 (9.7) | 0.654 ‡ |
| Mean duration of diabetic neuropathy, years (SD) | 3.nine (3.6) | 4.0 (4.ane) | 4.1 (4.6) | 0.740 ‡ |
| Michigan Neuropathy Screening Instrument score (SD) | 5.4 (1.6) | five.2 (ane.five) | 5.2 (1.five) | 0.094 ‡ |
| Hateful SF-36 general wellness total score (SD) | fifty.4 (20.half dozen) | 49.0 (21.6) | 49.4 (twenty.9) | 0.625 ‡ |
| Mean BPI average interference score | 4.1 (2.3) | 4.v (2.3) | four.5 (2.two) | 0.045 ‡ |
| Hateful EQ-5D index | 0.59 (0.3) | 0.58 (0.iii) | 0.57 (0.iii) | 0.624 ‡ |
† Cochran-Mantel-Haenszel exam for general association, controlling for study.
‡ Means were analyzed using a main-furnishings analysis of variance (anova) model including treatment grouping and study.
SD = standard departure; QD = once daily; BID = twice daily; SF-36 = Medical Outcomes Study Short Form-36 Health Survey Questionnaire; BPI = Brief Pain Inventory; EQ-5D = EuroQol 5D Health Questionnaire.
The astute treatment phase completion rates were similar across all three groups (duloxetine 60 mg QD and 60 mg BID [77.half dozen% each grouping], placebo [79.four%]). In that location were significant differences in the discontinuation rate between treatment groups due to adverse events (duloxetine 60 mg QD and 60 mg BID [10.8% each group], placebo [5.3%], P < 0.001) and lack of efficacy (duloxetine 60 mg QD and threescore mg BID [0.half dozen% each group], placebo [2.ix%], P = 0.047). There were no treatment group differences in the discontinuation rates due to protocol violations, loss to follow-up, decease, or patient, medico, or sponsor decision.
At that place were no significant differences amidst handling groups in baseline SF-36 general health total score (P = 0.625) or in the baseline EQ-5D total score (P = 0.624). There was a statistically significant, but non clinically relevant, difference between the three groups in the baseline BPI boilerplate interference score (P = 0.045).
The baseline SF-36 scores of the DPNP patients investigated were consistently lower on all the domains compared with the normative data from adults anile 65–74 years, and from persons with type II diabetes with a mean age of 60.2 years [42] (Figure 1). Table 2 summarizes the results of the treatment group comparisons for each domain of the SF-36 for those patients completing 12 weeks of acute therapy. DPNP patients treated with duloxetine 60 mg QD or threescore mg BID had greater improvement, compared with placebo, in all SF-36 domains of physical performance, function-physical, actual pain, general health, vitality, social functioning, role-emotional, and mental health. Within-treatment group changes amidst the domain scores ranged from 0.9 points (mental health, placebo) to 23.5 (part-physical, duloxetine 60 mg BID). Duloxetine threescore mg BID showed some advantage over duloxetine 60 mg QD on full general health (P = 0.02) and mental wellness (P = 0.04) status. Consistent results were seen in the intent-to-treat population with the exception that the above-indicated advantages of duloxetine 60 mg BID over sixty mg QD in the domains of general and mental health were not statistically meaning.
Figure 1
Baseline characteristics of studied accomplice vs U.S. normal adult and diabetes norms* (*Ware et al. [42]). #A college score indicates a better functional status. DPNP = diabetic peripheral neuropathic pain; SF-36 = Medical Outcomes Written report Brusk Course-36 Wellness Survey Questionnaire.
Figure 1
Baseline characteristics of studied cohort vs U.Southward. normal adult and diabetes norms* (*Ware et al. [42]). #A higher score indicates a better functional status. DPNP = diabetic peripheral neuropathic pain; SF-36 = Medical Outcomes Study Short Class-36 Health Survey Questionnaire.
Table ii
Summary of changes in health condition as measured by the SF-36 and the EQ-5D: completer analyses
| Variable* | Therapeutic Categories | ||||
| Placebo (N = 270–272) Mean (SD) † LS Mean (SE) ‡ | Duloxetine threescore mg QD (N = 275–277) | Duloxetine sixty mg BID (N = 253–256) | |||
| Mean (SD) † LS Mean (SE) ‡ | P value vs Placebo** | Hateful (SD) † LS Mean (SE) ‡ | P value vs Placebo** | ||
| SF-36 physical operation | |||||
| Baseline | 49.8 (24.2) | 51.ii (26.0) | 48.1 (25.5) | ||
| Modify from baseline | 5.7 (1.1) | 10.0 (one.ane) | 0.007 | nine.8 (i.2) | 0.014 |
| SF-36 part-physical | |||||
| Baseline | 38.ane (forty.5) | 39.5 (40.viii) | 36.7 (39.0) | ||
| Change from baseline | 12.9 (2.1) | 20.4 (two.1) | 0.013 | 23.5 (2.ii) | <0.001 |
| SF-36 bodily pain | |||||
| Baseline | 40.5 (18.7) | 40.2 (19.0) | 39.0 (18.1) | ||
| Change from baseline | 12.nine (1.ii) | eighteen.5 (i.2) | <0.001 | 19.8 (i.2) | <0.001 |
| SF-36 general health | |||||
| Baseline | 49.8 (21.0) | 48.half dozen (21.5) | 48.7 (20.9) | ||
| Modify from baseline | ii.nine (0.ane) | 7.0 (0.9) | 0.001 | ten.0 (0.9) | <0.001 |
| SF-36 vitality | |||||
| Baseline | 47.viii (17.iv) | 47.three (19.4) | 47.9 (19.6) | ||
| Modify from baseline | iv.4 (i.0) | nine.0 (1.0) | 0.001 | 9.0 (1.0) | 0.002 |
| SF-36 social functioning | |||||
| Baseline | 72.9 (23.8) | 72.0 (23.7) | 70.9 (24.8) | ||
| Change from baseline | 5.4 (one.i) | 8.9 (1.1) | 0.030 | ix.8 (ane.2) | 0.008 |
| SF-36 part-emotional | |||||
| Baseline | 70.ane (40.half dozen) | 68.4 (41.0) | 68.0 (39.viii) | ||
| Change from baseline | 4.iv (1.9) | 10.3 (1.nine) | 0.032 | 10.v (2.0) | 0.030 |
| SF-36 mental wellness | |||||
| Baseline | 72.4 (17.5) | 71.4 (xviii.viii) | 71.1 (18.7) | ||
| Modify from baseline | 0.9 (0.ix) | 4.ix (0.viii) | 0.001 | seven.3 (0.9) | <0.001 |
| Variable* | Therapeutic Categories | ||||
| Placebo (Northward = 270–272) Mean (SD) † LS Hateful (SE) ‡ | Duloxetine 60 mg QD (N = 275–277) | Duloxetine 60 mg BID (N = 253–256) | |||
| Hateful (SD) † LS Mean (SE) ‡ | P value vs Placebo** | Hateful (SD) † LS Hateful (SE) ‡ | P value vs Placebo** | ||
| SF-36 physical functioning | |||||
| Baseline | 49.8 (24.ii) | 51.2 (26.0) | 48.1 (25.5) | ||
| Alter from baseline | 5.7 (one.ane) | ten.0 (one.1) | 0.007 | 9.eight (1.two) | 0.014 |
| SF-36 office-physical | |||||
| Baseline | 38.1 (xl.5) | 39.5 (forty.8) | 36.7 (39.0) | ||
| Change from baseline | 12.ix (2.ane) | 20.4 (2.ane) | 0.013 | 23.five (2.2) | <0.001 |
| SF-36 bodily pain | |||||
| Baseline | xl.5 (eighteen.vii) | 40.2 (19.0) | 39.0 (18.1) | ||
| Change from baseline | 12.9 (ane.2) | 18.5 (1.2) | <0.001 | 19.8 (1.two) | <0.001 |
| SF-36 general health | |||||
| Baseline | 49.8 (21.0) | 48.6 (21.v) | 48.vii (20.nine) | ||
| Change from baseline | 2.9 (0.1) | 7.0 (0.nine) | 0.001 | 10.0 (0.ix) | <0.001 |
| SF-36 vitality | |||||
| Baseline | 47.8 (17.iv) | 47.3 (19.iv) | 47.nine (19.6) | ||
| Change from baseline | iv.iv (one.0) | 9.0 (i.0) | 0.001 | ix.0 (i.0) | 0.002 |
| SF-36 social performance | |||||
| Baseline | 72.9 (23.viii) | 72.0 (23.7) | 70.9 (24.8) | ||
| Alter from baseline | 5.iv (1.1) | viii.9 (1.1) | 0.030 | 9.eight (i.2) | 0.008 |
| SF-36 function-emotional | |||||
| Baseline | 70.1 (40.6) | 68.4 (41.0) | 68.0 (39.eight) | ||
| Change from baseline | 4.4 (1.ix) | 10.three (1.9) | 0.032 | 10.5 (2.0) | 0.030 |
| SF-36 mental wellness | |||||
| Baseline | 72.4 (17.5) | 71.four (18.8) | 71.1 (eighteen.seven) | ||
| Alter from baseline | 0.9 (0.9) | iv.ix (0.viii) | 0.001 | seven.3 (0.9) | <0.001 |
Table 2
Summary of changes in health condition as measured by the SF-36 and the EQ-5D: completer analyses
| Variable* | Therapeutic Categories | ||||
| Placebo (N = 270–272) Mean (SD) † LS Mean (SE) ‡ | Duloxetine 60 mg QD (N = 275–277) | Duloxetine 60 mg BID (N = 253–256) | |||
| Mean (SD) † LS Mean (SE) ‡ | P value vs Placebo** | Hateful (SD) † LS Mean (SE) ‡ | P value vs Placebo** | ||
| SF-36 physical performance | |||||
| Baseline | 49.viii (24.ii) | 51.2 (26.0) | 48.ane (25.5) | ||
| Change from baseline | 5.7 (ane.i) | x.0 (1.ane) | 0.007 | 9.8 (ane.ii) | 0.014 |
| SF-36 role-concrete | |||||
| Baseline | 38.1 (twoscore.5) | 39.5 (twoscore.eight) | 36.7 (39.0) | ||
| Change from baseline | 12.ix (2.1) | 20.4 (2.one) | 0.013 | 23.5 (2.2) | <0.001 |
| SF-36 bodily hurting | |||||
| Baseline | 40.5 (18.7) | twoscore.2 (19.0) | 39.0 (eighteen.1) | ||
| Change from baseline | 12.9 (1.2) | 18.v (one.2) | <0.001 | nineteen.eight (i.2) | <0.001 |
| SF-36 full general health | |||||
| Baseline | 49.8 (21.0) | 48.6 (21.5) | 48.seven (20.9) | ||
| Alter from baseline | 2.9 (0.1) | 7.0 (0.9) | 0.001 | 10.0 (0.9) | <0.001 |
| SF-36 vitality | |||||
| Baseline | 47.8 (17.4) | 47.3 (19.iv) | 47.ix (19.6) | ||
| Change from baseline | 4.4 (1.0) | 9.0 (ane.0) | 0.001 | ix.0 (1.0) | 0.002 |
| SF-36 social functioning | |||||
| Baseline | 72.9 (23.viii) | 72.0 (23.7) | 70.nine (24.8) | ||
| Change from baseline | 5.4 (ane.1) | 8.9 (ane.one) | 0.030 | 9.eight (1.2) | 0.008 |
| SF-36 role-emotional | |||||
| Baseline | 70.1 (forty.6) | 68.4 (41.0) | 68.0 (39.8) | ||
| Change from baseline | 4.4 (1.nine) | ten.3 (ane.9) | 0.032 | x.5 (2.0) | 0.030 |
| SF-36 mental health | |||||
| Baseline | 72.4 (17.5) | 71.four (xviii.8) | 71.i (18.seven) | ||
| Change from baseline | 0.nine (0.9) | 4.9 (0.8) | 0.001 | 7.3 (0.9) | <0.001 |
| Variable* | Therapeutic Categories | ||||
| Placebo (Due north = 270–272) Mean (SD) † LS Mean (SE) ‡ | Duloxetine sixty mg QD (Northward = 275–277) | Duloxetine 60 mg BID (N = 253–256) | |||
| Mean (SD) † LS Hateful (SE) ‡ | P value vs Placebo** | Mean (SD) † LS Mean (SE) ‡ | P value vs Placebo** | ||
| SF-36 physical operation | |||||
| Baseline | 49.viii (24.2) | 51.2 (26.0) | 48.1 (25.five) | ||
| Alter from baseline | 5.7 (i.1) | ten.0 (one.ane) | 0.007 | 9.8 (1.2) | 0.014 |
| SF-36 role-physical | |||||
| Baseline | 38.ane (40.5) | 39.5 (twoscore.8) | 36.seven (39.0) | ||
| Modify from baseline | 12.9 (2.1) | 20.four (2.1) | 0.013 | 23.5 (ii.2) | <0.001 |
| SF-36 actual pain | |||||
| Baseline | xl.v (18.seven) | twoscore.2 (xix.0) | 39.0 (18.1) | ||
| Change from baseline | 12.9 (i.2) | xviii.5 (i.two) | <0.001 | 19.viii (1.two) | <0.001 |
| SF-36 general health | |||||
| Baseline | 49.8 (21.0) | 48.6 (21.5) | 48.vii (20.nine) | ||
| Change from baseline | 2.9 (0.1) | seven.0 (0.9) | 0.001 | 10.0 (0.ix) | <0.001 |
| SF-36 vitality | |||||
| Baseline | 47.eight (17.4) | 47.3 (19.4) | 47.9 (19.6) | ||
| Modify from baseline | iv.4 (1.0) | 9.0 (1.0) | 0.001 | 9.0 (1.0) | 0.002 |
| SF-36 social functioning | |||||
| Baseline | 72.9 (23.8) | 72.0 (23.7) | 70.nine (24.8) | ||
| Alter from baseline | five.iv (ane.ane) | eight.9 (1.1) | 0.030 | 9.eight (1.2) | 0.008 |
| SF-36 role-emotional | |||||
| Baseline | lxx.ane (40.6) | 68.4 (41.0) | 68.0 (39.viii) | ||
| Change from baseline | 4.4 (i.ix) | 10.iii (1.9) | 0.032 | 10.5 (2.0) | 0.030 |
| SF-36 mental wellness | |||||
| Baseline | 72.four (17.5) | 71.4 (18.8) | 71.1 (eighteen.7) | ||
| Alter from baseline | 0.ix (0.9) | 4.9 (0.8) | 0.001 | seven.3 (0.9) | <0.001 |
| Placebo (N = 272) Hateful (SD) | Duloxetine 60 mg QD (North = 273) | Duloxetine 60 mg BID (N = 255) | |||
| Mean (SD) | P value vs Placebo* | Hateful (SD) | P value vs Placebo* | ||
| EQ-5D alphabetize | |||||
| Baseline | 0.59 (0.25) | 0.57 (0.26) | 0.56 (0.26) | ||
| Change from baseline | 0.1 (0.01) | 0.2 (0.01) | 0.004 | 0.2 (0.01) | <0.001 |
| Placebo (North = 272) Mean (SD) | Duloxetine lx mg QD (Due north = 273) | Duloxetine 60 mg BID (N = 255) | |||
| Hateful (SD) | P value vs Placebo* | Mean (SD) | P value vs Placebo* | ||
| EQ-5D index | |||||
| Baseline | 0.59 (0.25) | 0.57 (0.26) | 0.56 (0.26) | ||
| Change from baseline | 0.ane (0.01) | 0.2 (0.01) | 0.004 | 0.2 (0.01) | <0.001 |
* Modify from baseline refers to change from baseline to endpoint; mean and SD are reported for baseline; LS Mean and SE are reported for change from baseline.
** P values are from pairwise comparisons of LS Ways.
† Mean and SD are shown at baseline.
‡ LS Mean and SE are shown for change from baseline.
LS Means = least-squares means; SD = standard deviation; SE = standard fault.
| Placebo (N = 272) Mean (SD) | Duloxetine 60 mg QD (North = 273) | Duloxetine 60 mg BID (Due north = 255) | |||
| Hateful (SD) | P value vs Placebo* | Mean (SD) | P value vs Placebo* | ||
| EQ-5D index | |||||
| Baseline | 0.59 (0.25) | 0.57 (0.26) | 0.56 (0.26) | ||
| Change from baseline | 0.1 (0.01) | 0.two (0.01) | 0.004 | 0.2 (0.01) | <0.001 |
| Placebo (North = 272) Hateful (SD) | Duloxetine threescore mg QD (N = 273) | Duloxetine 60 mg BID (North = 255) | |||
| Mean (SD) | P value vs Placebo* | Mean (SD) | P value vs Placebo* | ||
| EQ-5D index | |||||
| Baseline | 0.59 (0.25) | 0.57 (0.26) | 0.56 (0.26) | ||
| Change from baseline | 0.1 (0.01) | 0.2 (0.01) | 0.004 | 0.2 (0.01) | <0.001 |
* Change from baseline refers to change from baseline to endpoint; hateful and SD are reported for baseline; LS Mean and SE are reported for change from baseline.
** P values are from pairwise comparisons of LS Means.
† Mean and SD are shown at baseline.
‡ LS Mean and SE are shown for change from baseline.
LS Means = least-squares ways; SD = standard deviation; SE = standard fault.
Duloxetine 60 mg QD and sixty mg BID were significantly superior to placebo at reducing scores in all BPI interference items (Figure ii), thereby indicating improvements in all seven items, with similar results demonstrated for the intent-to-treat population. In the analysis of the EQ-5D, patients on duloxetine 60 mg QD (P = 0.004) and 60 mg BID (P < 0.001) were both significantly amend compared with placebo for study completers (Table 2). Results for the intent-to-treat analysis were consequent, thus demonstrating the superiority of duloxetine 60 mg QD and sixty mg BID compared with placebo with regard to changes in all included function and QOL measures.
Figure ii
Mean change in interference items of the Brief Hurting Inventory (Note: These results are based on patients completing the acute treatment menses). *P < 0.05 vs placebo; **P < 0.01 vs placebo; ***P < 0.001 vs placebo. LS = least squares; QD = once daily; BID = twice daily.
Figure 2
Mean alter in interference items of the Cursory Pain Inventory (Annotation: These results are based on patients completing the acute treatment period). *P < 0.05 vs placebo; **P < 0.01 vs placebo; ***P < 0.001 vs placebo. LS = least squares; QD = once daily; BID = twice daily.
Word
The results of this pooled analysis demonstrate that, on all wellness consequence measures included in the three studies presented here, patients treated with duloxetine improved significantly compared with those treated with placebo. These findings were consistent betwixt the intent-to-care for population and the subset of patients who completed the 12-week astute treatment periods for the three studies, thus demonstrating robust improvements in measures of functional outcome, unaffected by either the time or incidence of patient withdrawal.
The diabetic population investigated here experienced greater impairments in baseline health status equally evidenced past consistently lower baseline SF-36 scores compared with normative data for persons with type Ii diabetes and the average 65- to 74-yr-erstwhile adult. When treated with duloxetine, patients with DPNP experienced significant improvement in all eight domains of the SF-36.
Previous studies using patient-rated measures of QOL take reported that diabetic patients score lower than individuals without the affliction [43], and that diabetes is associated with deficits in wellness-related QOL [44–46]. Galer and colleagues [8] reported interference of pain associated with diabetic neuropathy on daily activities every bit assessed past responses to the modified BPI. On boilerplate, patients in this study [eight] reported pain that caused substantial interference (score >five on a scale of 0–10) in sleep and enjoyment of life; and moderate interference (score iii–v) in normal piece of work, mobility, recreational activities, full general activity, social activities, and mood. Duloxetine demonstrated significant improvement on all the BPI interference items, thus suggesting that duloxetine can significantly improve QOL in DPNP patients.
Limitations
Generalizability of these results may be express by the fact that patients in these studies were advisedly selected to exclude psychiatric and medical comorbidities, and could be less severely ill than patients in the full general diabetic population with DPNP. All measures used in the study, although collected using validated instruments, were self-reported. The results are based on acute treatment duration of 12 weeks and may not generalize to longer periods of treatment. Boosted studies are needed to demonstrate a maintenance of event in functional and QOL improvements associated with long-term treatment with duloxetine.
Although the pooling of data beyond studies, when advisable, may provide more precise estimates of population treatment effects, the larger sample sizes associated with pooled data will increase the statistical power associated with comparisons. Consequently, statistical significance with pooled information may not e'er be consequent with that observed in the respective private studies. The magnitude of treatment grouping differences, and its associated clinical relevance, should exist considered in conjunction with statistical significance when reviewing the findings of analyses from pooled data.
The wellness outcome measures used were general rather than status-specific, and therefore did not inquire almost functional bug specific to diabetic peripheral neuropathy, such as disturbances in balance or symptoms associated with reduced sensation in the feet. Foot ulcers that affect 2–iii% of the diabetic population each year [47] tin also have a profound effect on the QOL of persons with diabetes [48–50]. The NeuroQol, a recently developed and validated QOL instrument, includes a symptom checklist and may exist used as an effect measure in future clinical studies [51], as information technology differentiates among specific symptomatic expressions of neuropathy and provides the clinician with specific points for interventions.
Decision
Hurting due to diabetic peripheral neuropathy has a meaning negative impact on QOL. Improvement in QOL has been viewed to be i of the goals of treatment of patients with DPNP. The present analysis of pooled results from three placebo-controlled studies suggests that duloxetine tin significantly improve functional status and QOL in patients with DPNP.
Acknowledgments
The inquiry was supported past Eli Lilly and Company and Boehringer Ingelheim. Equally Chappell, TK Le, DK Kajdasz, DN D'Souza, and JM Russell—employees and shareholders of Eli Lilly and Company. DG Armstrong and One thousand Backonja have no financial relationships (advisor, consultant, speaker, stockholder, etc.), with a cumulative value of more than $10 000 per year in the concluding ii years, with any company whose products may be related to the topic of this manuscript.
References
2
.
The ascension global burden of diabetes and its complications: Estimates and projections to the twelvemonth 2010
.
Diabet Med
1997
;
14
(
suppl 5
):
S1
–
85
.
3
et al.
Diabetic neuropathies: A statement by the American Diabetes Association
.
Diabetes Intendance
2005
;
28
:
956
–
62
.
four
.
Potential mechanisms of neuropathic hurting in diabetes
.
Int Rev Neurobiol
2002
;
fifty
:
205
–
28
.
v
.
Treatment options in painful diabetic neuropathy
.
Acta Neurol Scand Suppl
1999
;
173
:
36
–
42
.
6
.
Epidemiology and affect on quality of life of postherpetic neuralgia and painful diabetic neuropathy
.
Clin J Pain
2002
;
xviii
:
350
–
4
.
eight
.
Painful diabetic polyneuropathy: Epidemiology, pain description, and quality of life
.
Diabetes Res Clin Pract
2000
;
47
:
123
–
8
.
9
.
Understanding the impact of painful diabetic neuropathy
.
Diabetes Metab Res Rev
2003
;
19
(
suppl one
):
S2
–
8
.
x
.
Preventive human foot intendance in people with diabetes
.
Diabetes Care
2003
;
26
(
suppl 1
):
S78
–
9
.
11
.
Psychological aspects of diabetic neuropathic human foot complications: An overview
.
Diabetes Metab Res Rev
2004
;
20
(
suppl 1
):
S13
–
eight
.
12
.
Preventing foot ulcers in patients with diabetes
.
JAMA
2005
;
293
:
217
–
28
.
13
.
Diabetic lower extremity infection: Influence of physical, psychological, and social factors
.
J Diabetes Complications
2005
;
19
:
107
–
12
.
xiv
, et al.
Hurting severity in diabetic peripheral neuropathy (DPN) impacts patient functioning, symptom levels of feet and depression, and slumber
.
23rd APS Annual Scientific Meeting
,
Vancouver, BC, Canada
,
May 6–9, 2004
.
xv
et al.
Health state preference cess in DPN
.
Pharmacoeconomics
2002
;
20
:
1079
–
89
.
16
.
Pathophysiology and treatment of painful diabetic neuropathy of the lower extremity
.
South Med J
1998
;
91
:
894
–
viii
.
17
.
Dual serotonin and noradrenaline uptake inhibitor class of anti-depressants—Potential for greater efficacy or just hype?
Prog Drug Res
2002
;
58
:
169
–
222
.
18
et al.
A randomized controlled trial of duloxetine in diabetic peripheral neuropathic pain
.
Neurology
2006
;
67
:
1411
–
20
.
nineteen
et al.
A double-blind, randomized multicenter trial comparing duloxetine with placebo in the management of diabetic peripheral neuropathic pain
.
Pain Med
2005
;
half dozen
(
6
):
346
–
56
.
20
.
Duloxetine vs. placebo in patients with painful diabetic neuropathy
.
Pain
2005
;
116
:
109
–
18
.
21
.
Duloxetine, 60 mg once daily, for major depressive disorder: A randomized double-bullheaded placebo-controlled trial
.
J Clin Psychiatry
2002
;
63
:
308
–
15
.
22
.
Duloxetine lx mg once daily dosing versus placebo in the acute treatment of major depression
.
J Psychiatr Res
2002
;
36
:
383
–
90
.
23
.
Duloxetine in the treatment of major depressive disorder: A double-blind clinical trial
.
J Clin Psychiatry
2002
;
63
:
225
–
31
.
24
et al.
Duloxetine in the treatment of low: A double-blind placebo-controlled comparing with paroxetine
.
J Clin Psychopharmacol
2004
;
24
:
389
–
99
.
25
et al.
Duloxetine for the handling of major depressive disorder
.
Psychopharmacol Bull
2002
;
36
:
106
–
32
.
26
et al.
Duloxetine: A new treatment for the emotional and physical symptoms of low
.
J Clin Psychiatry
2003
;
5
:
xix
–
28
.
27
et al.
Safety and tolerability of duloxetine in the treatment of major depressive disorder: Assay of pooled data from eight placebo-controlled clinical trials
.
Hum Psychopharmacol
2005
;
20
:
327
–
41
.
28
et al.
Effects of duloxetine on painful physical symptoms associated with low
.
Psychosomatics
2004
;
45
:
17
–
28
.
29
American Psychiatric Association (APA).
Diagnostic and Statistical Manual of Mental Disorders
, 4th edition.
Washington, DC
:
American Psychiatric Association
;
1994
.
xxx
et al.
The mini-international neuropsychiatric interview (MINI): The development and validation of a structured diagnostic psychiatric interview for DSM-4 and ICD-10
.
J Clin Psychiatry
1998
;
59
(
suppl twenty
):
22
–
33
.
31
.
SF-36 Health Survey Transmission and Interpretation Guide
.
Boston, MA
:
The Health Establish, New England Medical Heart
;
1993
.
32
.
The MOS 36-item short-form wellness survey (SF-36). I. Conceptual framework and item choice
.
Med Care
1992
;
30
:
473
–
83
.
33
.
SF-36 Health Survey Update
.
Spine
2000
;
25
:
3130
–
9
.
34
.
SF-36® Concrete and Mental Health Summary Scales: A User's Manual
.
Boston, MA
:
The Wellness Institute
;
1994
.
35
.
Pain cess: Global utilize of the Cursory Pain Inventory
.
Ann Acad Med Singapore
1994
;
23
:
129
–
38
.
36
.
A randomized controlled trial of high dose dextromethorphan in facial neuralgias
.
Neurology
2000
;
55
:
964
–
71
.
37
.
Doubleblind, randomized trial of bupropion SR for the treatment of neuropathic pain
.
Neurology
2001
;
57
:
1583
–
8
.
38
.
Controlled- release oxycodone for pain in diabetic neuropathy: A randomize controlled trial
.
Neurology
2003
;
60
:
927
–
34
.
39
et al.
Advances in neuropathic pain: Diagnosis, mechanisms, and treatment recommendations
.
Arch Neurol
2003
;
60
:
1524
–
34
.
40
.
The EuroQoL instrument: An index of health-related quality of life
. In: , ed.
Quality of Life and Pharmacoeconomics in Clinical Trials
, 2d edition.
Philadelphia, PA
:
Lippincott-Raven Publishers
;
1996
:
191
–
201
.
41
EuroQoL Group.
EuroQoL—A new facility for the measurement of health-related quality of life
.
Health Policy
1990
;
sixteen
:
199
–
208
.
42
.
SF-36 Health Survey. Manual and Interpretation Guide
.
Lincoln, RI
:
QualityMetric Incorporated
;
1993, 2000
.
43
et al.
Functional status and well-beingness of patients with chronic conditions: Results from the medical outcomes study
.
JAMA
1989
;
262
:
907
–
13
.
44
.
Levels and risks of depression and feet symptomatology amid diabetic adults
.
Diabetes Care
1997
;
20
:
585
–
90
.
45
.
Quality of life and associated characteristics in a large national sample of adults with diabetes
.
Diabetes Care
1997
;
20
:
562
–
seven
.
46
.
Cocky-rated wellness and diabetes of long duration: The Wisconsin epidemiologic study of diabetic retinopathy
.
Diabetes Care
1998
;
21
:
236
–
forty
.
47
.
Lower extremity foot ulcers and amputations in individuals with diabetes
. In: , eds.
Diabetes in America
, 2nd edition. National Institutes of Health Publication No. 95-1468.
Bethesda, Md
:
National Institutes of Health
;
1995
:
409
–
28
.
48
.
Psychological aspects of diabetic peripheral neuropathy
.
Diabetes Rev
1999
;
7
:
387
–
94
.
49
.
Psychological and behavioral problems in diabetic neuropathic foot ulceration
. In: , eds.
The Foot in Diabetes
, 3rd edition.
Chichester, UK
:
John Wiley & Sons
;
2000
:
123
–
32
.
50
.
Diabetic foot ulcers: A quality of life issue
.
Diabetes Metab Res Rev
2001
;
17
:
246
–
9
.
51
et al.
The development and validation of a neuropathy- and foot ulcer-specific quality of life musical instrument
.
Diabetes Care
2003
;
26
:
2549
–
55
.
Source: https://academic.oup.com/painmedicine/article/8/5/410/1847879
0 Response to "Review of Duloxetine in the Management of Diabetic Peripheral Neuropathic Pain"
Postar um comentário