EMLA cream as a topical anesthetic for the repeated mechanical debridement of venous leg ulcers A double-blind, placebo-controll

EMLA cream as a topical anesthetic for therepeated mechanical debridement of venous legulcers: A double-blind, placebo-controlled studyCatherine Lok, MD,a Carle Paul, MD,b Pierre Amblard, MD,c Didier Bessis, MD,dCllia Debure, MD,i Brigitte Faivre, MD,e Bernard Guillot, MD,f Jean Paul Ortonne, MD,gGunilla Huledal, MSc(Pharm),j and Bernard Kalis, MDhBackground: A granulating surface is important for skin grafting and healing of leg ulcers.Mechanical debridement to remove necrotic tissue often must be stopped before comple-tion because of pain.Objective: Our purpose was to assess the effect of EMLA cream on the number of debride-ments required to obtain a clean ulcer and on pain during debridement and to determine itssafety after repeated doses.Methods: In this randomized double-blind, placebo-controlled study, 69 patients withvenous leg ulcers received cream before debridement until a clean ulcer was obtained (ora maximum of 15 debridements).Results: EMLA decreased the median number of debridements required for a clean ulcer(EMLA 11.5, placebo 15; P = .019) and decreased pain by 50% (P = .003). Plasma lev-els of lidocaine, prilocaine, and their main metabolites were low without any apparentaccumulation.Conclusion: EMLA produces effective pain relief for the debridement of leg ulcers andshortens the time to a clean ulcer.(J Am Acad Dermatol 1999;40:208-13.)Most leg ulcers (90%) are of venous origin.Debridement of leg ulcers is an accepted andrecommended practice to remove dead tissue,slough, and fibrinous plaques. A clean ulcerpromotes the growth of granulation tissue,increases the overall healing rate,1,2 andimproves the results of skin grafting.3Mechanical debridement often involves the useFrom the Departments of Dermatology, Sud Hospital, Amiens (JeanPaul Denoux, MD, Chief)a; Saint Louis Hospital, Paris (LouisDubertret, MD, Chief)b; Centre Hospitalier Universitaire,Grenoblec; Saint Charles Hospital, Montpellier (Jean JacquesGuilhou, MD, Chief)d; Saint Jacques Hospital, Besanon (PhilippeHumbert, MD, Chief)e; Caremeau Hospital, Nmesf; PasteurHospital, Niceg; Robert Debr Hospital, Reimsh; the Departmentof Vascular Re-education, Broussais Hospital, Parisi; and ClinicalResearch and Development, Astra Pain Control AB, Sdertlje.jSupported by a grant from Laboratoires Astra France.Reprint requests: Catherine Lok, MD, Department of Dermatology ofJean Paul Denoeux, MD, Sud Hospital, Amiens, France.Copyright 1999 by the American Academy of Dermatology, Inc.0190-9622/99/$8.00 + 0 16/1/94744208of a sharp curette or scissors to remove necrosisjust above the level of viable tissue. Because ofthe pain, mechanical debridement is oftenstopped before its completion.EMLA cream 5%, a eutectic mixture of the twolocal anesthetics lidocaine (25 mg/g) and prilo-caine (25 mg/g), is effective for anesthesia of theskin. The analgesic effect of EMLA cream 5% forthe debridement of leg ulcers has been shown inone double-blind, single-dose study.4 No localreactions were observed in this study. In an openstudy by Hansson et al5 of 43 patients, the anal-gesic effect and local tolerability were confirmedfor 8 consecutive debridement procedures. Painfrom ulcer cleansing and the frequency of post-cleansing pain were found to be significantlylower in the EMLA-treated patients.The primary objective of this randomized, dou-ble-blind, placebo-controlled study was to assessthe effect of EMLA cream on the number ofmechanical debridement sessions required toobtain a clean ulcer. Blood samples were collectedJournal of the American Academy of DermatologyVolume 40, Number 2, Part 1to determine the plasma levels of lidocaine andLok et al 209Table I. Demographics and treatmentprilocaine and their main metabolites.PATIENTS AND METHODSPatientsAfter Ethical Committee approval and writteninformed consent, 69 patients from 9 Departments ofDermatology or Phlebology, scheduled for mechani-cal debridement of a venous leg ulcer, with or withouta slight arterial component (ankle/arm index 0.8),were included in the study. Further inclusion criteriawere an ulcer area of 5 to 50 cm2, debris and necrosison 50% or more of the ulcer area, ulcer judged torequire debridement during the first week at least 3times a week, and no previous EMLA treatment of aleg ulcer.TreatmentA regimen of oral dextropropoxyphene (30 mg) andacetaminophen (400 mg) was given to all patients 1hour before debridement. A thick layer of cream (1-2g/10 cm2, max 10 g) was applied to the ulcer and cov-EMLA(n = 36)Age (y) (mean SD) 71 13Sex (Male/Female) 10/26Type of ulcerVaricose 16Postthrombosis 6Varicose + Postthrombosis 6Slight arterial component 8Ulcer area on admission 14 (5-62)(cm2); median (range)Ulcer area at study 11.8 (2-62)termination (cm2);median (range)Dose (g/10 cm2); 3 (1-7.1)mean (range)Mean time between 2 1.4 (1-4)debridements (days);mean (range)Placebo(n = 33)73 1110/2311103916 (4-127)13 (2-127)2.7 (0.3-6)1.4 (1-5)ered with an occlusive dressing (Saran Wrap) for 30 to45 minutes. Within 10 minutes of removal of creamand dressing, the mechanical debridement was started.The procedure consisted of curettage of fibrinous tis-sue and the removal of necrotic tissue and crusts usinga curette, scissors, and/or a bistoury. After debride-ment, the ulcer was covered with a sterile petrolatum(Vaseline) dressing. After the first week, the frequencyof the procedures was as judged necessary by thephysicians. A maximum of 15 debridements wasallowed. If further debridements were necessary afterthe fifteenth procedure, the patient discontinued thestudy. Elastic compression bandages was standardtherapy at 6 of the centers and were also applied inselected patients at two other centers. Bandages anddressings were removed daily to assess the need fordebridement.AssessmentsAn ulcer was defined as clean when 75% or more ofthe ulcer area was free from necrotic and fibrinous tis-sue and crusts. The number of debridement sessionsrequired for a clean ulcer was assessed. The size ofulcer was assessed by tracing the ulcer area on paperand then weighing this area. Debridement pain wasrated by the patient after each procedure on a 10-cmvisual analog scale (VAS) graduated from 0 (“no pain”)to 10 (“worst imaginable pain”).6 Premature interrup-tion of debridement because of pain was recorded. Thequality of debridement was assessed on a 4-point scale.Local reactions were assessed and adverse events wererecorded.Bioanalytic methodBlood samples were collected either 1 and 6 or 3 and8 to 24 hours after the start of application of the first andthe last dose of the cream. The plasma levels of lido-caine, monoethylglycinexylidide (MEGX), glycinexyli-dide (GX), 2,6-xylidine, prilocaine, and o-toluidinewere determined by a gradient LC/MS method. Thelimits of quantification were 1 ng/ml for lidocaine,prilocaine and o-toluidine, 2ng/ml for MEGX, 9 ng/mlfor GX, and 6 ng/ml for 2,6-xylidine. The precision washigh for all analyses, with interassay (n = 10) coeffi-cient of variation values lower than 20% at the limit ofquantification. The accuracy was close to 100% of thetheoretical concentrations.Statistical methodsThe number of debridements required to obtain aclean ulcer was compared between treatments usingnonparametric survival analysis; observations werecensored when an ulcer was not clean after a maximumallowable number of debridements. Explanatory vari-ables (ulcer area on admission, type of ulcer) weretested by means of regression analysis on the rank testsusing the log rank scores.To provide an overall comparison of analgesic effect,the mean response VAS over time was calculated foreach patient. To obtain a measure of the change of VASover time, the regression coefficients were calculatedfor each patient. Analysis of variance was used to testthe difference between the two treatment groups and astepwise regression was used to identify variables(ulcer area on admission, type of ulcer, treatment) capa-210 Lok et al Journal of the American Academy of DermatologyFebruary 1999Fig 1. Kaplan-Meier graph of the number of debridements until clean ulcer (P = .019).Straight line, EMLA; dotted line, placebo.Table II. Quality of debridementEMLA PlaceboVery satisfactorySatisfactoryFairUnsatisfactory21.9%61.9%15.4%0.8%9.1%52.5%31.6%6.8%Fig 2. Number of patients with clean ulcers at the endof the study (P = .008). Open bar, Clean; solid bar, notclean.ble of influencing VAS regression coefficients. Ulcerarea was compared between treatments using an analy-sis of covariance with ulcer area on admission ascovariate. A weighted regression analysis was used tofind potential explanatory variables (ulcer area onadmission, type of ulcer, treatment). The differencesbetween treatment groups for categorical variables weretested in accordance with Mantel-Haenszel stratified forcenter.The influence of leg ulcer area on the plasma con-centrations of lidocaine, prilocaine, o-toluidine andMEGX were investigated for the first dose given usinglinear regression models.In all the statistical tests performed, a two-tailedvalue of P less than .05 was considered to be statistical-ly significant.RESULTSThe demographics were similar in EMLA- andplacebo-treated patients. The mean doses of creamadministered were 3 and 2.7 g/10 cm2 in the twogroups, respectively (Table I).Number of debridements. There was a signif-icant difference in the number of debridementsuntil clean ulcer between the two groups: median11.5 in the EMLA group and more than 15 in theplacebo group (Fig 1) (P = .019). None of theexplanatory variables tested (ulcer area on admis-sion, type of ulcer) were retained in the regressionJournal of the American Academy of DermatologyVolume 40, Number 2, Part 1 Lok et al 211Fig 3. VAS pain scores during debridement over time (P = .003).Table III. Plasma levels of lidocaine, MEGX, prilocaine, and o-toluidineTime after first dose (hr)1368-24Time after last dose (hr)1368-24Lidocaine(ng/mL)55 (5-203)13 (3-37)14 (2-48)0.5 (1-3)86 (7-410)44 (3-337)20 (2-128)5 (1-32)MEGX(ng/mL)2 (2-7)2 (2-6)2 (2-7)24 (2-17)4 (2-12)4 (2-29)1 (2-5)Prilocaine(ng/mL)22 (3-77)5 (1-18)3 (1-11)1 (1-3)24 (4-70)6 (1-18)4 (1-22)1 (1-12)o-Toluidine(ng/mL)2 (1-11)1 (1-3)1 (1-3)12.7 (1-11)1.8 (1-7)1 (1-5)0.4 (1-3.5)Data are expressed as mean value with range given in parentheses.model. The number of patients with a clean ulcerat the end of the study was significantly higher inthe EMLA group (24 patients, 66.7%) comparedwith the placebo group (11 patients, 33.3%) (P =.008) (Fig 2).Pain. EMLA cream significantly decreased VASpain scores for debridement by approximately 50%compared with placebo (P = .003). The median painscores for all the procedures in both the EMLA- andplacebo-treated groups were 2.3 and 5, respectively.This difference appeared at the first debridementand remained constant throughout the study (Fig 3).In the regression model, ulcer area on admissionwas the only explanatory variable that was shown toinfluence VAS pain scores. The number of patientswho prematurely interrupted at least one debride-ment because of pain was significantly higher (P .001) in the placebo group (75.8%) than in theEMLA group (41.7%).Duration and quality of debridement. Themedian duration of debridement was 4 minutes inthe EMLA group and 3 minutes in the placebogroup (P = .253). EMLA significantly improvedthe mean quality of debridement (P = .02, TableII). This improvement was observed at the firstdebridement and continued throughout the study.Ulcer area at study termination. Ulcer area onadmission (P .001) and number of debridements(P = .047) were explanatory variables for ulcerarea at study termination.212 Lok et alLocal reactions. At the time of removal of thecream, redness was observed in 36% and 39% andpallor was observed in 14% and 12% of theEMLA- and placebo-treated patients, respectively.Burning, itching, and pain were experienced by44%, 28%, and 39% of the EMLA-treatedpatients, respectively; the corresponding figures inthe placebo-treated patients were 31%, 15%, and39%. No signs of sensitization were observed.Plasma levels. The plasma levels of lidocaine,MEGX, prilocaine, and o-toluidine are summarizedin Table III. In general, the individual plasma levelswere highest 1 or 3 hours after the start of applica-tion of EMLA cream and nonquantifiable 8 to 24hours after the start of application of EMLA cream.Lidocaine and its metabolites: MEGX, GX,and 2,6-xylidine. The plasma levels of lidocainewere 410 ng/mL (1750 nM) or less and the plasmalevels of the lidocaine metabolite MEGX, in gen-eral lower than the lidocaine levels, were 29ng/mL (140nM) or less. The ratio of MEGX andlidocaine was comparable for the first and the lastdose; hence there was no apparent accumulation ofMEGX. The lidocaine metabolite GX was belowthe limit of quantification in all samples, whereasthe lidocaine metabolite 2,6-xylidine was quantifi-able in 6 of the samples, but did not exceed 12ng/mL (100 nM) in any of the samples.Prilocaine and its metabolite o-toluidine.Prilocaine plasma levels did not exceed 77ng/mL(352 nM) in any of the samples. The prilocainemetabolite o-toluidine was quantifiable in 62 ofthe 136 samples but did not exceed 11 ng/mL (103nM) in any of the patients.Although it was not possible to make any statis-tical comparison between the first and last dose,there was no apparent accumulation of any of thesubstances. There was a significant correlation (P .05) between the plasma levels of lidocaine,prilocaine, and o-toluidine and the dose of EMLAcream applied and the leg ulcer area. Up to 57% ofthe variation in plasma levels of lidocaine, prilo-caine, and o-toluidine could be explained by theleg ulcer area.DISCUSSIONIn this placebo-controlled study, the effect ofEMLA cream on the outcome of repeated mechan-ical debridements has been evaluated. In agree-ment with previous studies in patients with legulcers,4,5 EMLA cream produced satisfactoryJournal of the American Academy of DermatologyFebruary 1999analgesia for mechanical debridements of legulcers for up to 15 repeated sessions (Fig 3). Thequality of debridement was improved by EMLA(Table II), most probably as a result of a more thor-ough process with anesthesia than with placebo. Itis also likely that the more efficient cleansing, inturn, resulted in the faster appearance of cleanulcers observed in the EMLA-treated patients. Themedian number of debridements required to reacha clean ulcer was statistically significantly lower(P = .019) in the EMLA-treated patients (11.5)compared with the placebo-treated patients.Because only about 30% of placebo-treatedpatients had a clean ulcer after 15 treatments, therequired median number of debridement sessionswas higher than 15 in the placebo group.According to this, the potential savings are at least4 treatment sessions and, by extrapolation of thesurvival curve (Fig 1), probably 8 to 10 treatments,per patient. Consequently, according to our data,there will be considerable cost savings for healthcare and significant improvements in the medicaltreatment of patients with venous leg ulcers ifEMLA is used.We observed a considerable variation in plasmaconcentrations that could be explained in part bythe differences in ulcer areas and doses betweenpatients. Moreover, the variability in patient charac-teristics and the design of the study involving dif-ferent plasma sampling times between patientsshould be considered. The plasma levels of prilo-caine were mostly lower than the plasma levels oflidocaine, which is in accordance with the largervolume of distribution and higher clearance forprilocaine compared to lidocaine.7 Both lidocaineand prilocaine were expected to be completelyeliminated between each dose because the half-lives of these substances are in the order of 1 to 2hours. Although it was not possible to make any sta-tistical comparison between the first and last dose,there was no apparent accumulation of either lido-caine or prilocaine after up to 15 repeated applica-tions of EMLA cream on 3 to 7 days per week. Theplasma levels of lidocaine and prilocaine were farbelow the levels of lidocaine and prilocaine associ-ated with initial signs of central nervous systemtoxicity (5000-6000 ng/mL).8 There was no appar-ent accumulation of MEGX. This is in accordancewith the elimination half-life of this compound,which is reported to be slightly longer than that oflidocaine.8 GX was not quantifiable in any of theJournal of the American Academy of DermatologyVolume 40, Number 2, Part 1samples, which is in accord with the lower plasmalevel reported for this metabolite.8 The plasma lev-els of 2,6-xylidine were quantifiable in 6 of the 136plasma samples in the EMLA-treated patients.Previ