Dia Care-2003--770-6 (1)

8/10/2019 Dia Care-2003--770-6 (1)

1/7

The Sensory Symptoms of DiabeticPolyneuropathy Are Improved With

-Lipoic AcidThe SYDNEY TrialTHE SYDNEY T RIAL AUTHORS , FOR THE

SYDNEY T RIAL STUDY G ROUP :ALEXANDER S. AMETOV , MD 1

ALEXEI BARINOV , MD 2

PETER J. D YCK , MD 3

ROBERT HERMANN , MD 4

NATALIA K OZLOVA , MD 5

W ILLIAM J. LITCHY , MD 6

PHILLIP A. LOW , MD 3

DETLEF NEHRDICH , DIPL STAT 4

MARIA NOVOSADOVA , MD 5

PETER C. OBRIEN , PHD 3

MIROSLAV RELJANOVIC , MD 6

RUSTEM SAMIGULLIN , MD 4

K LEMENS SCHUETTE , BSC 3

IGOR STROKOV , MD 1

HANS J. TRITSCHLER , PHD 4

K LAUS W ESSEL , PHD 4

N IKOLAI YAKHNO , MD 2

DAN ZIEGLER , MD 7

OBJECTIVE Because -lipoic acid (ALA), a potent antioxidant, prevents or improvesnerve conduction attributes, endoneurial blood ow, and nerve (Na( ) K( ) ATPase activity inexperimental diabetes and in humans and may improve positive neuropathicsensory symptoms,in this report we further assess the safety and efcacy of ALA on the Total Symptom Score (TSS),a measure of positive neuropathic sensory symptoms.

RESEARCH DESIGN AND METHODS Metabolically stable diabetic patients withsymptomatic (stage 2) diabetic sensorimotor polyneuropathy (DSPN) were randomized to aparallel, double-blind study of ALA (600 mg) (n 60) or placebo (n 60) infused dailyintravenously for 5 days/week for 14 treatments. The primary end point was change of the sumscore of daily assessments of severity andduration of TSS. Secondary endpointswere sumscoresof neuropathy signs (NIS), symptoms (NSC), attributes of nerve conduction, quantitative sen-sation tests (QSTs), and an autonomic test.

RESULTS At randomization, the groups were not signicantly different by the criteria of metabolic control or neuropathic end points. After 14 treatments, the TSS of the ALA group hadimproved from baseline by an average of 5.7 points and the placebo group by an average of 1.8points ( P 0.001). Statistically signicant improvement from baseline of the ALA, as comparedwith the placebo group, was also found for each item of the TSS (lancinating and burning pain,asleep numbness and prickling), NIS, one attribute of nerve conduction, and global assessmentof efcacy.

CONCLUSIONS Intravenous racemic ALA, a potent antioxidant, rapidly and to a signif-icant andmeaningful degree, improved such positive neuropathic sensory symptoms as pain and

several other neuropathic end points. This im-provement of symptoms was attributed toimproved nerve pathophysiology, not to in-creased nerve ber degeneration. Because of its safety prole and its effect on positive neu-ropathic sensory symptoms and other neuro-pathic end points, this drug appears to be auseful ancillary treatment for the symptoms of diabetic polyneuropathy.

Diabetes Care 26:770776, 2003

T he neuropathies associated with di-abetes are heterogeneous (1). Per-haps the most common variety isdiabetic sensorimotor polyneuropathy(DSPN), the disorder studied here. Thisvariety appears to be caused by chronichyperglycemia and associated metabolicderangements, damaging neurons (ax-ons) or Schwann cells (or myelin) directlyor indirectly by functional and structural

alterations of microvessels or the bloodnerve barrier. Inammation, perhapsfrom immune mechanisms, may also beimplicated in some cases (2).

Recognizing that total hyperglycemicexposure may be the most importantmodiable risk covariate for diabeticcomplications such as DSPN (38), whysearch for ancillary treatments in additionto rigorous glucose control? First, despiteconsiderable effort to achieve near eugly-cemia (by frequent monitoring of plasmaglucose, use of multiple injections of in-

sulin, or use of insulin pumps), many di-abetic patients cannot, or will not, achievethe desired level of metabolic control(3,8). Such patients might benet fromancillary treatments. Second, the mecha-nisms underlying positive neuropathicsensory symptoms may be different oroverlapping from ones relating to neuro-pathic impairment (9). Among the ancil-lary therapies considered are aldosereductase inhibitors (10), myo-inositol(11), essential fatty acids (12,13), vita-mins, protein kinase C inhibitors, vasodi-

From the 1

Russian Medical Academy for Advanced Studies, Moscow, Russia; 2

Neurology Clinic, MoscowMedical Academy, Russia; 3 Mayo Clinic, Rochester, Minnesota; 4 Viatris GmbH, Frankfurt, Germany;5 Health Partners Medical Group, Minneapolis, Minnesota; 6 Ergomed, Zagreb, Croatia; and the 7 GermanDiabetes Research Institute, University of Du sseldorf, Germany.

This report was prepared by P.J.D. Address correspondence and reprint requests to Peter J. Dyck, MD, Mayo Clinic, 200 First St. SW,

Rochester, MN 55905. E-mail: [email protected] for publication 1 August 2002 and accepted in revised form 10 December 2002.P.J.D., W.J.L., P.A.L., and D.Z. have received honoraria from Vitaris GmbH and/or have served on an

advisory panel for Vitaris GmbH, the manufacturer of -lipoic acid.Abbreviations: ALA, -lipoic acid; DSPN, diabetic sensorimotor polyneuropathy; HP-DB, heart pulse

deep breathing; ITT, intent-to-treat; NIS, neuropathy impairment score; NIS(LL), NIS of lower limbs; NSC,neuropathy symptoms and change; NSC(LL), neuropathy symptoms and change of lower limbs; PP, perprotocol; QST, quantitative sensation test; TSS, Total Symptom Score.

A table elsewhere in this issue shows conventional and Systeme International (SI) units and conversionfactors for many substances.

E m e r g i n g T r e a t m e n t s a n d T e c h n o l o g i e sO R I G I N A L A R T I C L E

770 D IABETES C ARE , VOLUME 26, NUMBER 3, M ARCH 2003

8/10/2019 Dia Care-2003--770-6 (1)

2/7

lators (14),antiprostaglandins (15),nervegrowth factors (16), ACE inhibitors, lip-id-lowering agents, advanced glycationend product inhibitors (17), acetyl- L-carnitine (18), and -lipoic acid (ALA)(19)the drug studied here.

Prevention or amelioration of positiveneuropathic sensory symptoms deserveattention because symptoms are trouble-some; create anxiety and depression; in-terfere with work, activities of daily living,meeting family and social responsibilities,andattainingadequate rest andsleep;andbring patients to see their medical pro-vider (20).

Because there already are drugs thatrelieve positive neuropathic sensorysymptoms (e.g., analgesics, antidepres-sants, tranquilizers, anti-epileptic agents,sedatives, or opiates), putative ancillary

treatments should relieve symptoms byimproving the pathophysiology of nerveand not by causing nerve injury. Withprolonged use they might also be ex-pected to prevent or improve neuropathy(20). Here we test whether positive neu-ropathic sensory symptoms can be ame-liorated by the potent antioxidant ALAthat has additional favorable metabolicactions (21 25). Racemic ALA, 6,8-dithioctane (the active ingredient studiedhere) is widely distributed in biologicaltissues and has a low toxicity pro le (26 30). In dosages ranging from 100to 1,800mg/day (given orally or intravenously), ALA has been used extensively in medicalpractice in Germany since 1959, and it isconsidered to be safe and ef cacious forthe treatment of diabetic polyneuropathysymptoms (29 32). Using the improvedapproaches learned fromthe performanceof earlier therapeutic trials we here retestwhether intravenous ALA is ef caciousfor treatment of the positive neuropathicsensory symptoms of pain, paresthesia,and asleep numbness. Further, assumingthat the drug is ef cacious, we examined

whether this effect is attributable to im-proved pathophysiology of nerve or to adifferentmechanism(i.e., degeneration of sensory nerve bers).

Study design and demographiccharacteristics of patientsThis is a mono-center, randomized, dou-blymasked, parallel-designed study of in-travenous ALAversus placebo given in 14doses over 3 weeks on the positive neuro-pathic sensory symptoms of burning andlancinating pain, asleep numbness, and

prickling (Total Symptoms Score [TSS])of feet or legs in 120 patients (60 patientsper group, 18 74 years old) with symp-tomatic (stage 2) diabetic polyneurop-

athy. Patients had to have type 1 or 2diabetes (according to American Diabetes Association criteria); HbA1c 12%; a TSS(Table 1) 7.5 points (of a possible max-imum of 14.64 points); neurologic signs(neuropathy impairment score [NIS] 2points, described below); nerve conduc-tion or heart pulse deep breathing (HP-DB) abnormality; and during the run-inperiod (the rst week), the TSS must notimprove by 3 points. Patients were ex-cluded if they had confounding neuro-logic disease or neuropathy; symptomaticperipheral vascular disease; or clinicallycomplicating cardiac, pulmonary, gastro-intestinal, hematologic or endocrine dis-ease, or malignancy.

Initially, one of us (R.S.) presentedthe outline of the study to Moscow endo-crinologists. Of the names provided, 790were interviewed by telephone, 497 wereprescreened, 206 were screened, and 120patients were admitted to the hospitalstudy unit for the 4-week study. Theysigned informed consent to receive eitherthedrug or placebo with doublemasking;however, for the rst week, all patients

received placebo. After randomization,on Monday of week 2 and thereafter for14 treatments, as shown in Fig. 1, patientsreceived either racemic ALA (600 mg) orplacebo (0.04 mg of ribo avin, to give thesolution a straw color to match ALA),each made-up to a nal volume of 225 mlwith physiologic saline and given intrave-nously over 30 min. The solutions wereprepared by the sponsoring company(ASTA Medica, Inc., now Viatris Inc.),shipped in a concentrated form to thestudy site identi ed only by coded num-

ber, and diluted to the nal volume. Inweeks2 and3, infusions were given daily,Mondaythrough Friday; andin thefourthweek, only four infusions were given(Monday through Thursday).Patients fol-lowed their usual diabetic surveillanceand treatment regimens during the con-duct of the study. Physicians were en-couraged to reduce HbA 1c to levels below8 or 9%, but patients were not excluded if this degree of control had not beenachieved. Vital signs, weight, and meta-bolic alterations (especially diabetic,hepatic, and renal alterations) were mon-itoredbefore, during, andafter treatment.

Assessment of neuropathic endpointsThe primary end point was the TSS, whichis a summation of the presence, severity,

and duration of lancinating pain, burningpain, prickling,and asleep numbness (Ta-ble 1). TSS was evaluated daily (just be-fore infusion and on treatment days) bythe same neurologist (M.N.) in order toreduce interobserver variability. Standardquestioning and scoring was used. Sec-ondary end points, evaluated before andduring the rst week andat the endof thefourth week, were the NIS, the NIS of lower limbs [NIS(LL)]), neuropathicsymptoms of lower limbs (neuropathysymptoms and change of lower limbs(NSC[LL]), nerve conductions, and HP-DB; each of these end points was assessedtwice, at onset and at end of the study.

The NIS is the sum score of a standardgroup of examinations of muscle weak-ness (0 normalto 4 paralyzed); reexloss (0 normal and 2 absent withreinforcement); and touch-pressure, vi-bration, joint position and motion, andpinprick (0 normal and 2 absent andfor each modality) of index nger andgreat toe and is scoredfor bothsides of thebody. Age, sex, physical tness, and an-thropomorphic features are to be taken

into account in making judgements of ab-normality. Typically, persons withoutneuropathy would have a score of 0. Aperson with mild weakness of toe exten-sors (1 and 1) and ankle dorsi exor mus-cles (1and 1), absentankle re exes (2and2), and decrease of all four sensory mo-dalities of the great toes (4 and 4) wouldhave a score of 16 points. The NSC scores(number, severity, and change) are de-rived from answers to 38 questions (mus-cle weakness, Q119; sensation, Q20 29; and autonomic symptoms, Q30 38)

Table 1 Symptoms* and scoring of theTSS (31)

Frequency

Intensity

Notpresent Mild Moderate Severe

Occasionally 0 1.00 2.00 3.00Often 0 1.33 2.33 3.33Continuous 0 1.66 2.66 3.66*The symptomsscored: sticking or lancinating pain,burning, prickling, or asleep numbness. Each of these is a positive neuropathic sensory symptom.Scoring: total score 014.64.

The SYDNEY Trial Authors

D IABETES C ARE , VOLUME 26, NUMBER 3, M ARCH 2003 771

8/10/2019 Dia Care-2003--770-6 (1)

3/7

(20). The questions are also subdividedby anatomical site (head and neck, chest,upper limbs and lower limbs); by largeber sensory function; by small ber sen-sory function; by positive sensory symp-toms (Q23 29 and separable by vesites); by negative sensory symptoms(Q20 22 for ve sites); and by pain(Q2529 for ve sites). Number equalsnumberof symptoms (of 38), andseverityequals number severity (1 mild, 2moderate, and 3 severe). Change innumber and severity is obtained by sub-tracting the mean of the two end valuesfrom the mean of the onset values. Thechange score is the patients comparisonof the symptoms at last evaluation to thesymptoms at onset (unchanged 0, im-proved [1 slightly, 2 moderately, or3 much], or worsened [ 1, 2, or

3]). The NSC (number, severity, andchange) are independent measures of symptoms, whereas the remainder of the

scores are subscores. Experienced andcertied (by P.J.D. and colleagues) neu-rologists (A.B. and M.N.) evaluated theNIS and NSC.

The nerve conduction, QSTs, and au-tonomic tests were performed by trainedand certi ed personnel (by W.J.L., P.J.D.,P.A.L., and colleagues). All neurologic,nerve conduction, and QST results wereinteractively evaluated by the Readingand Quality Assurance Centers (at MayoClinic and Health Partners). Eligibility,baseline conditions, waveforms, stimulus

response patterns, and test values werealso assessed.

Analysis planBased on previously reported results(33,34), the study was designed for 120patients (60 in each group) to achieve80%power to detect a difference betweentreatment groups of 1.0 point of the pri-mary end point, assuming a SD of 1.8points. Theprimary endpointwaschangein TSS from the rst day of treatment tothe termination visit. The correspondingprimary analysis for comparing the treat-ment groupswasa two-sample t test at the0.05 level. All results are reported as two-sided tests.

Regression analysis was also used, in-cluding treatment group, sex, and dura-tion of diabetes as independent variablesto account for imbalance (albeit small and

not signicant for duration) betweentreatmentgroupswith respect to the lattertwo variables. Wilcoxon s rank-sum testswere used to compare groupswith respectto continuous secondary end points be-cause many of these were highly skewed.

2 tests were used to compare treatmentgroups for dichotomous variables.

The analysis followed the intent-to-treat (ITT) principle. A per protocol anal-ysis was also conducted for the primaryend point.

RESULTS

Patient characteristicsThese are shown in Table 2. No patientsneeded to be excluded after the rst weekbecause they were hyper-responders.During the trial, one patient was with-drawn from study because of fever (later

Figure 1 ALA (intravenous ra-cemic ALA, 600 mg/day, 5 days/ week) was g iven on the 14treatment days as shown. Placebo(intravenous riboavin in physio-logic saline) was also given on the14 treatment days. The color, rateof infusion, and total volume wasessentially similar between pla-cebo and ALA. Most patients re-

ported that they thought they hadreceived ALA, although only one-half had.

Table 2 Patient characteristics

ALA Placebo P

n 60 60 NS Age (years) 56.8 9.65 55.4 8.66 NSSex (M/F) 14/46 24/36 0.06Type of diabetes (1/2) 15/45 15/45 NSDuration of diabetes (years) 15.1 8.8 14.0 8.2 NSDuration of neuropathy (years) 3.7 6.0 3.4 3.9 NSHeight (cm) 163.9 8.01 167.3 9.74 NSBMI (kg/m2 ) 29.4 4.93 29.3 5.23 NSData are means SD.

Diabetic polyneuropathy: the SYDNEY Trial


8/10/2019 Dia Care-2003--770-6 (1)

4/7

attributed to an upper respiratory viral in-fection) and a second because of cardiacarrhythmia (later diagnosed as paroxys-mal supraventricular tachycardia). Bothhad been on placebo. Following comple-tion of the study, six patients were foundto have laboratory values just outside theacceptable range (ve had received pla-cebo andone ALA). Theresults, therefore,were evaluated as ITT, including personswith protocol violations and as per proto-col (PP), excluding the six patients). Be-cause the results of ITT and PP wereessentially alike, only the results of ITTare shown.

Adverse eventsOf the 120 patients, 8 experienced a totalof 11 adverse events: 4 with or immedi-ately following treatment events, duringthe single-blind run-in phase; 1 with ALA; and 3 with placebo. No adverseevent (AE) was judged to be causally re-

lated to the trial medication.

Efcacy At baseline, there were no signicant de-mographic, anthropomorphic, or diseasedifferences between treatment groups ex-ceptfor a higherratio of men towomeninthe placebo group (Table 2). After 14treatments over 3 weeks, the mean valueof the primary end point (TSS) had im-proved by a mean of 5.72 points ( 1.53)for ALA and a mean of 1.83 points( 1.97) for placebo (P 0.001) (Fig. 2).

The difference between treatment groupsremained statistically signi cant (P 0.001) in regression analysis taking intoaccount the greater preponderance of fe-males in the ALA group. A very small butstatistically signicant difference (P 0.021)favoring ALAwas rstobserved onthe fourth treatment day, with the degreeof the difference increasing and remain-ing signicant thereafter (Fig. 1). Each of the four component symptoms of the TSSwas assessed individually, and signi -cantly greater improvement was found,favoring ALA over placebo, for each of them.

The NSC score, a 38-question surveyof motor, sensory (positive and negative),andautonomicsymptoms,wasused as anindependent measure of change of neuro-pathic symptoms (20). The mean changeof motor or autonomic scores (number,severity, and change) was not signi -cantly different between ALA and pla-

cebo. By contrast, large and statisticallysignicant greater improvement of posi-tive pain (stabbing, deep aching, burning,and excessive tenderness to contactor pressure; prickling; and asleep-numbness) and negative (loss of tactile,thermal, and pain sensation) neuropathicsensory symptoms had occurred with useof ALA as compared with placebo (Table3). Also, patients judgement of symptomchange from baseline was signicantlylarger for ALA than for placebo (P 0.001). ALA also improved the following

symptom scores to a better degree thandid placebo at the P 0.001 level:NSC(LL)number, severity, and change;NSC(LL) Sensationnumber andchange; NSC(LL) Large-Fiber Sensa-tion change; NSC(LL) Small-Fiber Sen-sationnumber and change; NSC(LL)Negative Sensationnumber; NSC(LL)Positive Sensationnumber and change;NSC(LL) Painnumber and change;NSC(LL) Small-Fiber Autonomic number, severity, and change. Signi cantat the 0.05 level: NSC(LL) Large-FiberSensationnumber; NSC(LL) NegativeSensation change.

Another secondary measure, the NIS,improved by 2.7 points ( 3.37) in ALAand by 1.2 points ( 4.14) in placebo(P 0.001), but the NIS of lower limbsdid not quite reach statistical signi cance

(P 0.076). Signicant differences werenot found between ALA and placebo forchange in individual or composite scoresof attributes of nerve conduction exceptfor distal latency of the sural nerve, whichimproved more in the ALA (mean begin-ning minus end difference of 0.86 ms)than in placebo (mean difference of 0.353ms, P 0.017). No signicant changewas found for QST or for autonomic testresults.

The treating physician graded the re-sponse to treatment in the ALA group asvery good or good in 66.7% and as satis-factory in 33.3%;whereas forplacebo, the judgements were very good or good in1.7%, satisfactory in 68.3%, unsatisfac-tory in 26.7%, and not assessable in 3.3%(P 0.001).

Patients judged overall ef cacy fol-lowing ALA as very good or good in 80%and as satisfactory in 20%. For placebo,the ratios and percents were very good orgood in 8.3%, satisfactory in 73.3%, andunsatisfactory in 15%; two patients werenot assessed. The difference in percent of very good or good versus other was sig-

nicantly better (P 0.001) for the ALAgroup.The benecial effect of ALA on neu-

ropathic symptoms could not be attrib-uted to a greater improvement inmetabolic control. The HbA 1c of the ALAgroup had improved (from onset to end)by a mean value of 0.22%, whereas theplacebo group had improved by a meanvalue of 0.12% (P 0.05).

CONCLUSIONS Var i ou s e n dpoints have been used to assess ef cacy

Table 3 Neuropathy symptom and change (NSC) scores

Score

Mean atonset

(points)Mean change

(points) WilcoxonP value ALA* PLA ALA PLA

NSCnumber 9.4 10.1 4.2 2.2 0.001NSCseverity 18.4 20.5 11.1 7.2 0.001NSCchange 0.2 0.0 14.5 9.2 0.001NSC Weaknessseverity 0.0 0.0 0.0 0.0 NSNSC Sensorynumber 7.2 7.6 3.3 1.4 0.001NSC Sensoryseverity 14.9 16.4 9.4 5.7 0.001NSC Sensorychange 0.2 0.0 12.4 7.4 0.001NSC(LL) Sensationseverity 14.9 5.5 9.4 5.7 0.001NSC(LL) Large-Fiber Sensationseverity 2.8 3.1 1.2 0.6 0.001NSC(LL) Small-Fiber Sensationseverity 12.1 13.3 8.3 5.1 0.001NSC(LL) Negative Sensationseverity 2.7 3.5 1.2 0.7 0.043NSC(LL) Positive Sensationseverity 12.2 12.9 8.3 5.0 0.001NSC(LL) Painseverity 10.0 10.6 7.3 4.6 0.001NSC Autonomicseverity 3.5 4.2 1.7 1.5 NS*n 60; PLA, placebo, n 58.



8/10/2019 Dia Care-2003--770-6 (1)

5/7

in treatment trials of diabetic polyneur-opathy: sum scores of neurologic abnor-malities (e.g., the NIS), QSTs, nerveconduction, autonomic tests, combina-tions of these, andmorphometryof nerve.In the present study, we use positive neu-ropathicsensory symptoms using TSS. An Ad Hoc Panel on End points for DiabeticNeuropathy Trials (20) had previouslydened positive neuropathic sensorysymptoms,discussed their useas trial endpoints, cited approaches for their assess-ment, and de ned the degree of changethat might be considered clinically mean-ingful and able to meet standards for be-ing ef cacious in controlled trials. Assuming that in such a trial the studywas rigorously performed and doublymasked, it would be necessary to rstshow a statistically signicant degree of

symptom change for the trial drugabove that found for the placebo; the de-gree of change should be of suf cientmagnitude to be considered meaningfuland should not have been due to worsen-ing of neuropathy.

In using the TSS, a four-item measureof positive neuropathic sensory symp-toms, as the primary outcome, as donehere, we imply that these symptoms alonemay be serious and debilitating healthoutcomes irrespective of the degree of neuropathic ndings or impairments. It isincreasingly recognized that positive neu-ropathy sensory symptoms are separatephenomena from negative symptoms(loss of tactile, thermal, pain, or other sen-sations) or from severity of neuropathicsigns or impairments. Thus, for some pa-tients with onset of polyneuropathy, pos-itivesymptoms may predominate withoutor with only a few neuropathic ndings.Later, as positive neuropathic sensorysymptoms abate, negative symptoms andimpairments may become evident andworsen. Also, we note that these positivesensory symptoms may be more bother-

some than negative neuropathic sensorysymptoms, may inspire patients seekingrelief from pain to visit physicians, andmay be moredebilitating (interfering withmeeting work, family, and social respon-sibilities) than negative symptoms or im-pairments. These positive neuropathicsymptoms are thought to be due to small-ber sensory nerve ber involvement. Al-though it is known that analgesics andanti-epileptic and tranquilizing medica-tions may relieve positive neuropathicsensory symptoms, we here test the idea

that symptoms may be improved by an-other mechanism, e.g., by improving thepathophysiology of nerves.

The present study appears to show anunequivocal and large bene cial effect of intravenous racemic ALA on the fre-quency and severity of the positive neu-ropathic sensory symptoms due todiabetic polyneuropathy, the effect of which cannot be attributed to placebo orto worsening of polyneuropathy. Theconclusion that the improvement is indic-ative of pharmacological ef cacy comesalso from the observations that the effectwas statistically signicant, was found foreach component of the TSS, and was con-rmed by similar results from indepen-dent assessments of symptoms using aseparate clinical instrument (NSC) andthat the improvement was suf ciently

large to be considered meaningful bystandards set by an ad hoc consensuspanel (20).Additionally, ef cacy wassug-gestedby thetime courseof improvement(it was delayed and then increasingly im-proved over the treatment period, notrapidly, andthen staticor fadingas wouldbe expected with analgesics and opiates)(35), negative sensory symptoms andneuropathic impairment also improved,and the effect could not be explained bybias (patients had remained masked). Ad-ditionally, further analysis indicated thatthe treatment effect could not be ex-plained by the unequal distribution bygender.

This benecial effect from ALA hadpreviously been reported (see referencesin ref. 22). By comparison with earlierstudies, the present study was more rig-orous in the following respects: a placeborun-in phase, pretraining and certi ca-tion of investigators, interobserver vari-ability was reduced by use of a singleinvestigator performing critical end pointevaluations, retentionof essentially all pa-tients throughout the study, use of read-

ing and quality assurance of neuropathicend points, independent analysis of re-sults, and evidence that masking wasmaintained.

Since it is generally thought that ab-normalities of attributes (or compositescores) of nerve conduction or of HP-DBareamong themost sensitive indicators of diabetic polyneuropathy, why did theseend points not show statistically signi -cant improvement with use of ALA? Whydid QST not show improvement? To be-gin with, one attribute (sural nerve la-

tency) did show statistically signicantimprovement. Also, recognition of wors-ening by nerve conduction assessmentmaybe more sensitive than recognition of improvement, and the latter may takemore time. Thus, in the Diabetes ControlandComplications Trial (DCCT), a statis-tically signicant difference in nerve con-duction was not recognized until severalyears had elapsed (6,7). In human neu-ropathies, it is common clinical experi-ence that improvement or recovery of nerve conduction may lag well behindclinical improvement.

The mechanisms underlying the im-provement of positive neuropathic sen-sory symptoms was not studied here, butantioxidant activity appears to be thelikely mechanism. There is accumulatingevidence from experimental animal and

tissue culture studies that full radical-mediated oxidative stress is implicated inthe pathogenesis of diabetic polyneurop-athy by inducing neurovascular defectsthat result in endoneurial hypoxia andsubsequent nerve dysfunction (36 39). Administration of physio logical anti-oxidants, including ALA, a potent li-pophilic free radical scavenger (40,41),provides a basis fora potentialtherapeuticeffect. Diabetic peripheral nerves demon-stratefootprints of oxidativestressandre-spond to treatment with lipoic acid (42).More recently, these ndings have beensupplanted by immunocytochemical evi-dence of DNA damage and cellular local-ization (43). Reduced oxygen speciescause irreversibleDNA damageto speci cproteins. In recent years, antibodies havebeen generated against modi ed struc-tures specic for reactive oxidative spe-ciesinduced damage (43). The epitopesinclude 8-hydroxy-2 -deoxyguanosineand 4-hydroxy-2-nonenal-modi edprotein (44). Urinary 8-hydroxy-2 -deoxyguanosine (8-OHdG) has been re-ported to be increased in human diabetes

(45,46).Finally, the lack of toxicity to ALA inour trial is perhaps not surprising sinceseveral million dosages of the drug havealready been given by German physi-cians, and toxicity appears to be ex-tremely low (29,30). No toxicity wasrecognized in the present trial.

Acknowledgments The study was spon-sored and paid for by ASTA Medica, Inc.,Frankfurt,Germany. P.J.D. receives grant sup-



8/10/2019 Dia Care-2003--770-6 (1)

6/7

port from a National Institutes of NeurologicDisease and Stroke grant (NS36797).

References1. Dyck PJ, Giannini C: Pathologic alter-

ations in the diabetic neuropathies of hu-

mans: a review. J Neuropathol Exp Neurol55:1181 1193, 19962. Younger DS, Rosoklija G, Hays AP, Troja-

borg W, Latov N: Diabetic peripheralneuropathy: a clinicopathologic andimmunohistochemical analysis of suralnervebiopsies. Muscle Nerve19:722 727,1996

3. Dyck PJ, Davies JL, Wilson DM, ServiceFJ, Melton LJ, OBrien PC:Risk factors forseverity of diabetic polyneuropathy: in-tensive longitudinal assessment of theRochester Diabetic Neuropathy Study co-hort. Diabetes Care 22:1479 1486, 1999

4. Service FJ, Rizza RA, Daube JR, OBrien

PC, Dyck PJ: Near normoglycaemia im-proved nerve conduction and vibrationsensation in diabetic neuropathy. Diabe-tologia 28:722727, 1985

5. Amthor K-F, Dahl-Jorgensen K, Berg TJ,Skard Heier M, Sandvik L, Aagenaes O,Hanssen KF: The effect of 8 years of strictglycaemic control on peripheral nervefunction in IDDM patients: the OsloStudy. Diabetologia 37:579 584, 1994

6. The Diabetes Control and Complica-tions Trial Research Group: The effect of intensive treatment of diabetes on the de-velopment and progression of long-termcomplications in insulin-dependent dia-

betesmellitus. NEnglJMed329:977 986,19937. DCCT Research Group: The Diabetes

Control and ComplicationsTrial (DCCT):design and methodologic considerationsfor the feasibility phase. Diabetes35:530 545, 1986

8. Greene DA,Stevens MJ,Feldman EL:Gly-cemic control. In Diabetic Neuropathy.2nd ed. Dyck PJ, Thomas PK, Eds. Phila-delphia, W.B. Saunders Company, 1999,p. 297315

9. Willner C, Low PA: Pharmacologic ap-proaches to neuropathic pain. In Periph-eral Neuropathy. 3rd ed. Dyck PJ, ThomasPK, Grif n JW, Low PA, Poduslo JF, Eds.Philadelphia, W.B. Saunders Company,1993, p. 1709 1720

10. Tomlinson DR: Role of aldose reductaseinhibitors in the treatment of diabeticpolyneuropathy. In Diabetic Neuropathy.2nd ed. Dyck PJ, Thomas PK, Eds. Phila-delphia, W. B. Saunders Company, 1999,p. 330 340

11. Gregersen G: Myo-inositol supplementa-tion. In Diabetic Neuropathy. 1st ed. DyckPJ, Thomas PK, Asbury AK, Winegrad AI,Porte D, Eds. Philadelphia, W.B. Saun-ders Company, 1987, p. 188 189

12. Cameron NE,Cotter MA:Role of linolenicacid in diabetic polyneuropathy. In Dia-betic Neuropathy. 2nd ed. Dyck PJ,Thomas PK, Eds. Philadelphia, W.B.Saunders Company, 1999, p. 359 367

13. Jamal GA, Carmichael H: The effect of gamma-linolenic acid on human diabetic

peripheral neuropathy: a double-blindplacebo-controlled trial. Diabet Med7:319 323, 1990

14. Cameron NE, Cotter MA, Low PA: Nerveblood ow in early experimental diabetesin rats: relation to conduction de cits. Am J Physiol 261:E1E8, 1991

15. Yasuda H, Kikkawa R: Role of antipros-taglandins in diabetic neuropathy. In Di-abetic Neuropathy. 2nd ed. Dyck PJ,Thomas PK, Eds. Philadelphia, W.B.Saunders Company, 1999, p. 368 376

16. Feldman EL, Windebank AJ: Growth fac-tors and peripheral neuropathy. In Dia-betic Neuropathy. 2nd ed. Dyck PJ,

Thomas PK, Eds. Philadelphia, W.B.Saunders Company, 1999, p. 377 38617. Brownlee M: Advanced glycation end

products and diabetic peripheral neurop-athy. In Diabetic Neuropathy. 2nd ed.Dyck PJ, Thomas PK, Eds. Philadelphia, W. B. Saunders Company, 1999, p. 353 358

18. Quatraro A, Roca P, Donzella C, Acam-pora R, Marfella R, GiuglianoD: Acetyl-L-carnitine for symptomatic diabeticneuropathy (Letter). Diabetologia 38:123,1995

19. Low PA, Nickander KK, Scionti L: Roleof hypoxia, oxidative stress, and excita-

tory neurotoxins in diabetic neuropa-thy. In Diabetic Neuropathy. 2nd ed.Dyck PJ, Thomas PK, Eds. Philadelphia, W.B. Saunders Company, 1999, p.317 329

20. Ad Hoc Panel on Endpoints for DiabeticNeuropathy Trials: Positive neuropathicsensory symptoms as endpoints in dia-betic neuropathy trials. J Neurol Sci 189:3 6, 2001

21. Packer L, Witt EH, Tritschler HJ: Alpha-lipoic acid as a biological antioxidant (Re-view Article).Free Radic Biol Med19:227 250, 1995

22. Biewanga G, Vriesmann MF, HaenenGRMM, Bast A: The identication of anew metabolite of lipoic acid in man:3-ketolipoic acid. In Lipoic Acid: a Phar-macochemical Study. Biewanga G, Ed. Am-sterdam, Vrije Universiteit Amsterdam,1997, p. 137 152

23. Bravenboer B, Kappelle AC, Hamers FPT,van Buren T, Erkelens DW, Gispen WH:Potential use of glutathione for the pre-vention and treatmentof diabetic neurop-athy in the streptozotocin-induced diabeticrat. Diabetologia 35:813 817, 1992

24. Packer L: Interactions among antioxi-dants in health anddisease: vitamin E and

its redox cycle. Proc Soc Exp Biol Med200:271276, 1992

25. Schmidt P, Grafen P, Goedde HW: Che-mie und biochemie der liponsa ure. Ange-wandte ChemieInternational Edin English77:900 911, 1965

26. Eger W: Thioctsaure bei Leberschadi-

gung. Angew Med 11:251 255, 195727. Fuke H, Ivanami K,Watanabe N,KumadaS: Acute, subacute and chronic toxicity of lipoic acid in rats. Folia Pharmacol Japan68:256 275, 1972

28. Kirnberger E, Braun W, Stille G, Wolf W:Beziehungen zwischen leberschultz undzucherstoffwechsel. Arznm Forsch/DrugRes 1:7276, 1958

29. Ziegler D, Reljanovic M, Mehnert H, GriesFA: Alpha-lipoic acid in the treatment of diabetic polyneuropathy in Germany:current evidence from clinical trials. ExpClin Endocr Diabetes 107:421 430, 1999

30. Ziegler D, Gries FA: -Lipoic acid in thetreatment of diabetic peripheral and car-diac autonomic neuropathy. Diabetes 46(Suppl. 2):S62S66, 1997

31. Ziegler D, Conrad F, Ulrich H: Effects of treatment with the antioxidant -lipoicacid on cardiac autonomic neuropathy innon-insulin-dependent diabetic patients:a 4-month randomized controlled multi-center trial (DEKAN-study). DiabetesCare20:369 373, 1997

32. Ziegler D, Hanefeld M, RuhnauKJ,Meiss-ner HP, Lobisch M, Schutte K, Gries FA:Treatment of symptomatic diabetic pe-ripheral neuropathywith the anti-oxidantalpha-lipoic acid: a 3-week multicenterrandomized controlled trial (ALADINStudy). Diabetologia38:1425 1433, 1995

33. Ziegler D, Hanefeld M, Ruhnau KJ,Hasche H, Lobisch M, Schutte K, KerumG, Malessa R: Treatment of symptomaticdiabetic polyneuropathywith the anti-ox-idant alpha-lipoic acid (ALADIN III studygroup). DiabetesCare22:1296 1301,1999

34. Strokov IA, Kozlova NA, Mozolevsky YV,Myasoedov SP, Yakhno NN: Ef cacy of ivadministered trometamol salt of thioctic(alpha-lipoic) acid in diabetic neuropa-thy. Zh-Nevrol-Psikhiatr-Im-S-S-Korsakova99:18 22, 1999

35. Max MB, Culname M, Schafer SC, LawlorBA, Murphy DL, Max MB.: Amitriptylinerelieves diabetic neuropathy pain in pa-tients with normal or depressed mood.Neurology 37:589 596, 1987

36. Malik RA, Williamson S, Abbott C, Car-rington AL, Iqbal J, Schady W, Boulton AJ: Effect of angiotensin-converting-en-zyme (ACE) inhibitor trandolapril on hu-man diabetic neuropathy: randomiseddouble-blind controlled trial. Lancet 352:1978 1981, 1998

37. Low PA, Nickander KK: Oxygen free rad-ical effectsin sciatic nerve in experimental



8/10/2019 Dia Care-2003--770-6 (1)

7/7

diabetes. Diabetes 40:873 877, 199138. Nickander KK, Schmelzer JD, Rohwer

DA, Low PA: Effect of -tocopherol de -ciency on indices of oxidative stress innormal and diabetic peripheral nerve. J Neurol Sci 126:6 14, 1994

39. Cameron NE, Cotter MA, Maxeld EK:

Anti-oxidant treatment prevents the de-velopment of peripheral nerve dysfunc-tion in streptozotocin-diabetic rats.Diabetologia 36:299 304, 1993

40. Nagamatsu M, Nickander KK, Schmelzer JD, Raya A, Wittrock DA, Tritschler H,Low PA: Lipoic acid improves nerveblood ow, reduces oxidative stress, andimproves distal nerve conduction in ex-perimental diabetic neuropathy. Diabetes

Care 18:1160 1167, 199541. Suzuki Y, Mizuno M, Tritschler HJ,

Packer L: Redox regulation of NF-kBDNA binding activity by dihydrolipoate.BBRC 36:241246, 1995

42. Kishi M, Tanabe J, SchmelzerJD, Low PA:Morphometry of dorsal root ganglion of

chronic experimental diabetic neuropa-thy. Diabetes 51:819 824, 200243. Toyokuni S: Reactive oxygen species-in-

duced molecular damage and its applica-tion in pathology. Pathol Int 49:91102,1999

44. Toyokuni S, Tanaka T, Hattori Y, Nish-iyama Y, Yoshida A, Uchida K, Hiai H,Ochi H, Osawa T: Quantitative immuni-histochemical determination of 8-hydroxy-

2 -deoxyguanosine by a monoclonalantibody N45.1: its application to ferricnitrilotriacetate-induced renal carcino-genesis model. Lab Invest 76:365374,1997

45. Leinonen J, Lehtimaki T, Toyokuni S,Okada K, Tanaka T, Hiai H, Ochi H,Laip-

pala P, Rantalaiho V, Wirta O, Pasternack A, Alho H: New biomarker evidence of oxidative DNA damage in patients withnon-insulin-dependent diabetes mellitus.FEBS Lett 417:150 152, 1997

46. Hinokio Y, Suzuki S, Hirai M, Chiba M,Hirai A, Toyota T: Oxidative DNA dam-age in diabetes mellitus: its associationwith diabetic complications. Diabetologia42:995 998, 1999



Documents

Dia Care-2003--770-6 (1)