ARTICLE OPEN ACCESS CLASS OF EVIDENCE

Randomized controlled crossover study of IVIgfor demyelinating polyneuropathy and diabetesAri Breiner MD MSc FRCPC Carolina Barnett Tapia MD PhD Leif Erik Lovblom MSc

Bruce A Perkins MD MPH FRCPC Hans D Katzberg MD MSc FRCPC and Vera Bril MD FRCPC

Neurol Neuroimmunol Neuroinflamm 20196e586 doi101212NXI0000000000000586

Correspondence

Dr Breiner

abreinertohca

AbstractObjectiveTo determine whether IV immunoglobulin (IVIg) is more effective than placebo at reducingdisability in patients with diabetes and demyelinating polyneuropathy features

MethodsThis is a double-blinded single-center randomized controlled crossover trial of IVIg treatmentvs placebo The primary outcome measure was the mean change in Overall NeuropathyLimitation Scale (ONLS) scores during the IVIg phasecompared with the placebo phaseSecondary outcomes include changes in the Rasch-built Overall Disability Scale MedicalResearch Council sum scores grip strength electrophysiologic measurements quality of lifeand adverse effects

ResultsTwenty-five subjects were recruited between March 2015 and April 2017 The mean change inONLS scores was minus02 points during the IVIg phase and 00 points during the placebo phase (p= 023) Secondary outcomes did not show significant differences between IVIg and placebo

ConclusionsIVIg did not reduce disability improve strength or quality of life in patients with demyelinatingpolyneuropathy features and diabetes after 3 months of treatment in comparison with placeboTherefore careful consideration of the primary diagnosis is required before immunomodula-tory therapy

Classification of evidenceThis study provides Class I evidence that for patients with diabetes and demyelinating poly-neuropathy features IVIg did not significantly reduce disability

MORE ONLINE

Class of EvidenceCriteria for ratingtherapeutic and diagnosticstudies

NPuborgcoe

From the Division of Neurology (AB) Department of Medicine the Ottawa Hospital Ottawa Hospital Research Institute (AB) Division of Neurology (CB HDK VB) Departmentof Medicine Ellen and Martin Prosserman Centre for Neuromuscular Diseases University Health Network University of Toronto Division of Endocrinology and Metabolism (LELBAP) Department of Medicine Mount Sinai Hospital and Lunenfeld Tanenbaum Research Institute University of Toronto Canada and Institute for Research and MedicalConsultations (VB) Imam Abdulrahman Bin Faisal University Dammam Saudi Arabia

Go to NeurologyorgNN for full disclosures Funding information is provided at the end of the article

The Article Processing Charge was funded by the authors

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License 40 (CC BY-NC-ND) which permits downloadingand sharing the work provided it is properly cited The work cannot be changed in any way or used commercially without permission from the journal

Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the American Academy of Neurology 1

There is a broad spectrum of polyneuropathy in patients withdiabetesmdashincluding distal symmetrical sensorimotor poly-neuropathy (DSP) proximal neuropathiesplexopathies anddemyelinating peripheral neuropathies The recognition ofchronic inflammatory demyelinating polyneuropathy (CIDP)in patients with diabetes is important as CIDP is a treatabledemyelinating peripheral neuropathy with a suspected auto-immune etiology1 The hallmark clinical findings are sym-metrical proximal and distal limb weakness sensory loss andabsent deep tendon reflexes1ndash3

IV immunoglobulin (IVIg) is a safe and effective treatment forCIDP both in short- and medium-term time horizons4ndash10

However in patients with diabetes mellitus (DM) the opti-mal treatment regimen for CIDP remains unknown aspatients with diabetes have been excluded from randomizedcontrolled clinical trials5ndash811 due to concerns that theirneuropathy may be secondary to diabetes rather than anautoimmune or inflammatory process In addition to the ab-sence of prospective data regarding treatment response thereare significant challenges to diagnosing CIDP in patients withdiabetes These challenges stem from overlap between elec-trophysiologic changes due to DSP and those due to CIDPAlthough diabetic polyneuropathy is considered a primarilyaxonal process many authors have reported electrophysiologicevidence of conduction velocity slowing and other de-myelinating findings1213 out of proportion to what would beexpected in a purely axonal polyneuropathy1415 These nerveconduction study (NCS) changes have been linked to localizedfoci of demyelination and remyelination16 and some authorshave reported inflammatory features (ie perivascular in-flammatory cell infiltrates) in nerve biopsies of patients withdiabetes17 The result is demyelinating-range NCS changes inpatients with diabetes making a distinction between those withCIDP +DSP and those with isolated DSP increasingly difficultThe coexistence of diabetes and CIDP in the same patient isa relatively common occurrence (4ndash65 of patients withCIDP have diabetes18ndash23) and is expected to only increase withrising diabetes prevalence worldwide In the medical literaturethere is conflicting opinion whether CIDP is more prevalent inpatients with diabetes than patients without diabetes23ndash26 butno convincing epidemiologic data to support an associationHowever the overall uncertainty underscores the need forfurther research to distinguish between DSP and CIDP27

Because of the lack of prospective studies there is a knowl-edge gap with regard to the appropriate method of diagnosing

and treating demyelinating polyneuropathies in patients withcoexistent diabetes Previous work from our group has shownthat NCSs may be confounded by abnormalities due to di-abetic polyneuropathy132829 Moreover the effect of IVIgtreatment is uncertain in this population as it has never beenstudied prospectively Therefore we performed a crossoverstudy to examine IVIg treatment responses in patients withdiabetes and demyelinating NCS changes with the goal ofdetermining whether IVIg could reduce disability in thispopulation

MethodsStudy design and participantsWe performed a randomized double-blind placebo-controlledcrossover study to test superiority of IVIg vs placebo (09NaCl in water) in patients with diabetes and demyelinatingpolyneuropathy features Study recruitment occurred at a singlecenter (Toronto General and Western Hospital sites of theUniversity Health Network) in Toronto Canada

Patients were eligible to participate if they were aged ge18years and had both DM (as per the American Diabetes As-sociation Criteria30) and clinical evidence of polyneuropathyIn addition at least 2 separate motor nerves (median ulnartibial or peroneal) had to meet the following electro-diagnostic criteria for demyelination (1) conduction velocitylt90 of the lower limit of normal distal latency gt110 upperlimit of normal (ULN) or minimal F-wave latency gt110ULN and (2) electrodiagnostic changes not exclusively due tomedian neuropathy at the wrist ulnar neuropathy at the el-bow or peroneal neuropathy at the fibular head Finally to beconsidered eligible there had to be clinical suspicion of pos-sible demyelinating polyneuropathy (CIDP) on the part ofthe treating neuromuscular specialist (eg proximal limbweakness or significant ataxia out of proportion to the degreeof axonal loss)

Potential subjects were excluded if they had a history of hered-itary polyneuropathy (Charcot-Marie-Tooth disease) otherpotential confounding causes of polyneuropathy (includingheavy alcohol consumption uremia and gammopathy-associated polyneuropathy) contraindication to IVIg treat-ment or a seriousunstable medical condition precluding its useThe full list of inclusion and exclusion criteria is found in theappendix (table e-1 linkslwwcomNXIA126)

GlossaryAE = adverse event ANCOVA = analysis of covariance CIDP = chronic inflammatory demyelinating polyneuropathy DM =diabetes mellitus DSP = diabetic sensorimotor polyneuropathy EFNS = European Federation of Neurological SocietiesIQR = interquartile range IVIg = IV immunoglobulin LLN = lower limit of normalMRC =Medical Research CouncilNCS =nerve conduction study ONLS = Overall Neuropathy Limitation Scale R-ODS = Rasch-based Overall Disability Scale SF =Short Form 36-item ULN = upper limit of normal

2 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

Standard protocol approvals registrationsand patient consentsEthics approval was obtained through the research ethicsboard of the University Health Network Hospital (Universityof Toronto) The study was registered on clinicaltrialsgov(Registration NCT02372149) Written informed consentwas obtained from all study participants As this is a clinicaltrial there are no photographs videos or other recognizableinformation being published

Randomization and maskingAllocation was performed in a 11 ratio for the initial infusiontype (IVIg or placebo) with block randomization in groups of5 Sequence generation was performed by the unblinded studyco-coordinator who was also responsible for documenting andmaintaining the master list of allocation assignments

IVIg was obtained through the blood bank and prepared inthe pharmacy before administration It was transferred to clearpolypropylene bags which were draped with bronze trans-lucent covers Placebo (09 NaCl) infusions consisted of theidentical volume of fluid in an identical polypropylene bagcovered by an identical bronze translucent shield All bagswere labeled only with the patientrsquos unique study ID numberThe evaluating physicians study infusion nurse and patientsdid not have access to the list of allocation assignments Onlythe pharmacists and one study coordinator were not blindedthese unblinded individuals were not allowed to contact theinvestigators or patients

ProceduresStudy infusions and procedures have been summarized in figure1 Gamunex IVIg (Grifols Inc) was administered as an initialloading dose of 20 gkg divided in 2 days with amaximumdailydose of 80 g (month 0) This was followed by a monthlymaintenance dose of 10 gkg at months 1 2 and 3 This dosingregimen for IVIg was derived from previous CIDP trials679

IVIg infusions were started at a rate of 001 mLkgmin andincreased as per patient tolerance to amaximumof 008mLkgmin Subjects who began treatment with placebo followed thesame infusion schedule Following the 3rd maintenance in-fusion there was a 3-month washout period Subjects ran-domized to IVIg in the second treatment phase received their20 gkg loading dose at month 6 followed by 10 gkg main-tenance dosing at months 7 8 and 9 The length of the washoutperiod was selected based on the half-life of IVIg and theexpected duration of benefit after monthly administration

The primary research question was whether IVIg could re-duce disability in patients with diabetes and demyelinatingpolyneuropathy features in comparison with placebo (Class Ilevel of evidence) We assessed the efficacy of IVIg using theOverall Neuropathy Limitation Scale (ONLSmdashprimary out-come measure) the Rasch-based Overall Disability Scale (R-ODS) theMedical Research Council (MRC) sum score GripStrength NCSs and a generic quality of life score Short Form36-item (SF-36)

The ONLS scale is a disability questionnaire with a score thatranges from 0 (no disability) to 12 (severe disability) It hasbeen validated in 100 patients with inflammatory poly-neuropathy and has strong correlation (r = 097) with theOverall Disability Sum Score31 Therefore it is the referencestandard for inflammatory polyneuropathy trials7910 TheR-ODS score is a disability questionnaire with a score thatranges from 0 (severe disability) to 48 (no disability) Com-pared with the ONLS score it may provide a better overallmeasure of disability a wider range of item difficulties anda rational method for item weighting32 The MRC sum scoreranges from 0 (total paralysis) to 80 (normal strength) and isthe sum of the MRC score of 8 muscles (4 upper and 4 lowerlimb) each graded from 0 to 5 Grip strength is measuredusing the Martin Vigorimeter in kilopascals Three trials weremeasured in each hand and the average recorded This isa relevant outcome because it relates to hand dexterity andstrength which are important for daily tasks The SF-36 isa self-administered questionnaire that measures physicalfunctioning role limitations due to physical problems socialfunctioning bodily pain mental health role limitations due toemotional problems vitality and general health The assess-ments were performed at screening (within 3 weeks of thefirst infusion) and at the month 3 6 and 9 study visits TheONLS and R-ODS scores were also repeated at the initialinfusion visit (month 0) to ensure that there had been nodecline or spontaneous improvement since the screening visitA telephone follow-up occurred at months 4 and 10 whichconsisted of the ONLS and R-ODS scores and adverse event(AE) monitoring For the purpose of data analysis the month3 and 9 ONLS scores were used

AE monitoring occurred during each clinical visit and eachIVIgplacebo infusion Patients with serious or intolerableAEs had all treatments stopped and were withdrawn from thestudy after discussion with the principal investigator Patientswith ongoing or catastrophic deterioration (ge3-point increasein the ONLS score with ge2-point increase in the lower ex-tremities) had their case assessed by the principal investigatorto determine whether they should be unblinded and treatedwith either the active treatment or plasmapheresis

Electrodiagnostic studiesNCSs were performed using Sierra Wave equipment (Cad-well Laboratories Inc Kennewick WA) using surfacestimulating and recording electrodes according to stan-dardized protocols3334 Skin temperature was maintained atge320degC in the hands and ge310degC in the feet We studied theresponses of the median ulnar tibial and peroneal motornerves on the more affected side and the sensory responsesof the median ulnar and sural nerves on the more affectedside

OutcomesThe primary efficacy outcome was comparison of the meanONLS change score in the experimental and placebo phasesSecondary outcomes were comparison of mean change

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 3

scores during experimental and placebo phases for the fol-lowing end points (1) R-ODS score (2)MRC sum score (3)grip strength (4) NCS parameters (distal latency conductionvelocity and compound motor action potential [CMAP]amplitude) and (5) SF-36 (total score) We also comparedAE rates during the experimental and placebo phases

Statistical analysisStatistical analyses were performed using SAS version 94 forWindows (SAS Institute Cary NC) Summary statistics arepresented as mean plusmn SD median and interquartile range(IQR) or as frequency () The changes in primary andsecondary outcomes were compared between IVIg and pla-cebo phases using 2 methods (1) an unadjusted analysisusing the paired Student t test and (2) an adjusted analysisusing repeated-measures analysis of covariance (ANCOVA)with the change score as the dependent variable and studyphase (IVIg or placebo) sequence and baseline value as theindependent variables The adjusted means and 95 confi-dence intervals of the change scores are presented We madeall comparisons on the intention-to-treat data set and weconsidered an α-level of 005 (2-sided) for all tests of statisticalsignificance

We assessed the carryover effects using the Student t test Toaccount for carryover effects an exploratory analysis of se-quence effect compared participants randomized to placebofirst vs those randomized to IVIg first This was done using(1) an unadjusted Student t test analysis and (2) an adjusted(ANCOVA) analysis with the change score as the dependent

variable and sequence and baseline value as the independentvariables Comparison of responders was made using theMcNemar test To compare the rates of AEs we used theMcNemar test and conditional maximum likelihood estimatesof the rate ratios

We calculated the sample size using data from a large ran-domized trial of IVIg in patients with CIDP7 In that studythe mean baseline ONLS score was 41 with a within-groupSD of 14 However calculation of the sample size fora crossover study35 requires within-patient SD which is notavailable for patients with inflammatory neuropathy36

Therefore rather than using within-group SD (14) whichwould provide a very conservative estimate for the samplesize we estimated a more realistic within-patient SD of 10ONLS pointmdashalso reflecting the change in the ONLS scorereported to be clinically significant on comparison withquality of life measures37 Based on these data the calculatedsample size was 18 patients with 80 power and α = 005Therefore enrollment was planned for 25 patients to ac-count for potential dropouts

Data availabilityIndividual participant data that underlie the results reportedin this article (including text tables figures and appendices)will be available after deidentification It will be available for 5years after publication to researchers who provide a meth-odologically sound proposal for the purposes of conductinga prespecified analysis Data will be accessed by sending anemail to the corresponding author

Figure 1 Study procedures

IVIG = IV immunoglobulin

4 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

ResultsBetween March 2015 and April 2017 74 patients wereapproached for participation Of these 33 underwent screen-ing and 25 met the criteria for enrollment Fourteen patientswere assigned to IVIg and 11 to placebo for the first in-tervention In the placebo-first group 2 patients dropped outafter the month 0 visit In the IVIg-first group 1 patientdropped out after the month 3 visit and 1 patient after themonth 6 visit Reasons for dropout included adverse effects ofaching and swelling (1) limited reimbursement for study visits(1) withdrew consent (1) and expired secondary to pancre-atitis (1) The CONSORT diagram is displayed in figure 2

Table 1 displays baseline demographic information for the 25enrolled subjects Themean age was 575 plusmn 92 years and 68were male In those patients who underwent lumbar puncture(n = 12) the mean CSF total protein was 097 plusmn 055 gLCSF total protein was above 045 gL in 1112 patients andwas above 060 gL in 1012 patients At baseline 84 ofpatients had symptoms of weakness or neuropathic pain in thelower extremities 60 had upper limb neuropathic pain 68had upper limb weakness and 79 of subjects had imbalanceProximal weakness of the upper or lower limbs was present in

68 of patients This was defined as muscle power of 45 orless on the MRC sum scoring in muscle groups proximal tothe elbow (ie deltoid or biceps) or the knee (quadriceps oriliopsoas) At baseline the mean ONLS score was 42 plusmn 17the mean R-ODS score was 308 plusmn 78 and the mean gripstrength was 584 plusmn 174 kPA (table 1)

Table 2 demonstrates the mean change scores for each ofprimary and secondary end points As shown the p values forprimary and secondary outcomes are not significantly differ-ent The mean change score for ONLS (primary outcome)was minus02 points during the IVIg phase and 00 points duringthe placebo phase (unadjusted p = 023) Adjusted p valueswere computed using ANCOVA analysismdashwith adjustmentfor sequence order and baseline valuemdashhowever again dif-ferences did not reach statistical significance

AE rates did not differ significantly between placebo and IVIg Intotal there were 26 AEs during the placebo phase and 22 in theIVIg phase (p = 045) The proportion of patients with ge1 AEwas 924 (38) in the placebo phase and 923 (39) in theIVIg phase (p = 032) Most AEs were mild or moderate inseveritymdashincluding fatiguedizzinesslightheadedness (128)swelling (128) body pain (106) nauseavomitinggastric

Figure 2 Consort diagram

IVIg = IV immunoglobulin

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 5

pain (85) headache (64) rashitching (64) dyspneachest tightness (64) and neck stiffnessjoint pain (64)Only 1 serious AE occurred during the trial at themonth 8markA patient receiving IVIg developed pancreatitis and subsequentlyexpired this event was not considered to be related to studytreatment but rather to a previous history of cholelithiasis

Post hoc exploratory analyses were performed to determinewhether the lack of treatment effect could be attributed to (1)order of administration (2) timing of end point measurementor (3) intention-to-treatmdashrather than per-protocolmdashanalysisThe findings consistently showed no significant differencesbetween groups with one exception (appendix and table e-2linkslwwcomNXIA126) When patients were stratifiedbased on treatment with placebo first or IVIg first those ran-domized to IVIg first showed a reduction in the mean ONLSscore (minus10 vs +03 points p = 0018) even after adjustment forsequence order and baseline value (adjusted p = 0020) In thisexploratory subgroup analysis the secondary outcomes did notshow significant change although there was a positive trend forthe R-ODS and SF-36 There was also evidence of a carryovereffect on the ONLS score (p = 0017) whereas no carryovereffect was observed for R-ODS grip strength SF-36 or MRCscore This carryover effect was most probably driven by theldquoIVIg-firstrdquo effect described above

Given the possible clinical benefit in the IVIg-first subgroup wereanalyzed the baseline data after stratifying by order of ad-ministration The only significant difference was in the durationof neuropathy (appendix and table e-3 linkslwwcomNXIA126) which had a median (IQR) of 10 (6ndash10) years in theplacebo-first group and 3 (2ndash6) years in the IVIg-first group (p =0033) Additional exploratory analyses were also performed inthe subgroup of patients who met the definite European Fed-eration of Neurological Societies (EFNS) electrodiagnostic cri-teria for patients with CIDP (1825)38 Within this subgrouptherewas no significant benefit to IVIg administration (change inthe ONLS score of minus02 in the IVIg phase and +01 in theplacebo phase p = 019 and change in the R-ODS score of +11in the IVIg phase and +02 in the placebo phase p = 029) In thesubgroup who did not meet the EFNS electrodiagnostic criteriathere was 1 patient who improved during the IVIg phase (fromONLS 4 to 2) and 1 patient who improved with placebo Finallyan exploratory responder analysis showed improvement of ge1ONLS point in 923 (39) of IVIg phases and 521 (24) ofplacebo phases although the result did not reach statistical sig-nificance (p = 032) Improvement of ge2 ONLS points wasobserved in only 3 patientsmdashall during the IVIg phase

DiscussionThe diagnosis and treatment of demyelinating poly-neuropathy in the setting of diabetes is a unique challengeFirst patients with diabetes have been excluded from CIDPtrials because of concerns about confounding effects of the 2

Table 1 General clinical characteristics of the 25 patientswho were randomized

CharacteristicMean plusmn SD median (IQR) orfrequency ()

Age (y) 575 plusmn 92

Height (m) 173 plusmn 011

Weight (kg) 915 plusmn 239

BMI (kgm2) 303 plusmn 62

Female sexmale sex 8 (32)17 (68)

Smoking history

Yes 2 (8)

Quit 8 (33)

Never 14 (58)

Diabetes duration (y) 11 [4 25]

Polyneuropathy duration (y) 55 [2 10]

CSF total protein (n = 12) (gL) 097 plusmn 055

Hemoglobin A1C () 79 plusmn 19

Previous IVIg exposure (singledose)

2 (8)

Clinical symptoms

Numbness lower limbs 19 (76)

Numbness upper limbs 15 (60)

Paresthesia lower limbs 16 (64)

Paresthesia upper limbs 12 (48)

Weakness lower limbs 21 (84)

Weakness upper limbs 17 (68)

Proximal weakness 24 (95)

Neuropathic pain lowerlimbs

21 (84)

Neuropathic pain upperlimbs

15 (60)

Imbalance 19 (79)

Falls 11 (46)

Baseline outcome measures

ONLS 42 plusmn 17

R-ODS 308 plusmn 78

Grip strength 584 plusmn 174

SF-36 368 plusmn 197

MRC grade 715 plusmn 64

Abbreviations BMI = body mass index CV = conduction velocity IQR =interquartile range IVIg = IV immunoglobulin MRC = Medical ResearchCouncil ONLS = Overall Neuropathy Limitation Score R-ODS = Rasch-basedOverall Disability Score SF-36 = 36-Item Short Form Health Survey

6 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

neuropathies Second the diagnosis of CIDP in a patient withdiabetes may prove especially challenging as diabetic neu-ropathy alone may produce signs of demyelination on nerveconduction testing Third there is often considerable overlapof clinical features and CSF profile Fourth there is no specificbiomarker for either disease

In this study we enrolled 25 patients with diabetes and fea-tures of demyelination on NCSs aiming to determinewhether IVIg is an effective treatment to reduce disabilityincrease muscle strength and improve quality of life Wefound no significant benefit to treating with IVIg As noted in

table 2 there was no significant difference in mean changescores for the primary or secondary outcome measures after 3months of treatment even with adjusted p values (using anANCOVA model) to account for baseline values

The primary findings in this study suggest that patients withdiabetes and demyelinating polyneuropathy features should notbe empirically treated with IVIg unless the clinical picture isconvincing for CIDP Some suggestions might include restrictingto those cases with (1) significant proximal weakness (2) definiteconduction block (3) shorter neuropathy duration or (4) thosepatients meeting more conservative CIDP criteria (American

Table 2 Outcomes according to study arm

Outcome

Placebo n = 24 IVIg n = 23 p Value for drug effectb

Baseline D (95 CI)a Baseline D (95 CI)a Unadjusted p value Adjusted p value

Primary

ONLS 40 plusmn 17 00 (minus04 to 04) 43 plusmn 16 minus02 (minus06 to 01) 023 038

Secondary

R-ODS 302 plusmn 89 00 (minus31 to 31) 303 plusmn 73 +10 (minus20-40) 027 037

Grip strength 565 plusmn 165 minus01 (minus48 to 45) 580 plusmn 163 minus40 (minus84 to 04) 016 022

SF-36 409 plusmn 221 +34 (minus20-88) 384 plusmn 203 +21 (minus31-73) 088 073

MRC grade 710 plusmn 65 minus01 (minus16 to 13) 714 plusmn 68 +05 (minus09 to 19) 070 054

Motor nerve conduction

Median nerve

Latency wrist 51 plusmn 18 00 (minus03 to 02) 52 plusmn 23 00 (minus02 to 03) 095 073

Amplitude wrist 64 plusmn 34 05 (00 to 10) 67 plusmn 38 03 (minus02 to 08) 051 038

CV forearm 412 plusmn 59 +06 (minus02 to 14) 414 plusmn 56 00 (minus08 to 08) 021 021

Ulnar nerve

Latency wrist 34 plusmn 11 +04 (00 to 08) 37 plusmn 14 00 (minus04 to 04) 016 014

Amplitude wrist 50 plusmn 31 +02 (minus02 to 06) 51 plusmn 36 +02 (minus02 to 06) 058 092

CV below elbow 409 plusmn 80 minus11 (minus44 to 23) 41 plusmn 78 minus10 (minus44 to 24) 092 094

Peroneal nerve

Latency ankle 105 plusmn 60 +07 (minus20-34) 110 plusmn 70 +06 (minus21-33) 055 079

Amplitude ankle 09 plusmn 15 00 (minus02 to 01) 09 plusmn 15 +01 (minus01 to 02) 018 024

CV lower leg 290 plusmn 53 minus05 (minus17 to 06) 283 plusmn 48 +05 (minus06 to 17) 018 014

Tibial nerve

Latency ankle 94 plusmn 51 minus01 (minus05 to 02) 99 plusmn 51 minus05 (minus08 to 01) 0093 014

Amplitude ankle 13 plusmn 22 minus02 (minus03 to 00) 12 plusmn 21 minus01 (minus02 to 01) 012 030

CV lower leg 299 plusmn 45 +05 (minus02 to 12) 291 plusmn 42 +01 (minus06 to 08) 048 036

Abbreviations CV = conduction velocity IVIg = IV immunoglobulin MRC = Medical Research Council ONLS = Overall Neuropathy Limitation Score R-ODS =Rasch-based Overall Disability Score SF-36 = 36-Item Short Form Health SurveyBaseline values displayed as mean plusmn SDNerve conduction latencies displayed in ms amplitudes in mV and conduction velocities in msa Intention-to-treat analysisb Unadjusted p value from the paired Student t test adjusted p value from ANCOVA adjusted for sequence order and baseline value

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 7

Academy of Neurology or Inflammatory Neuropathy Cause andTreatment criteria) Similar suggestions for distinguishing CIDPfrom proximal forms of diabetic polyneuropathy have beenproposed by other authors1939 The results of this study do notexclude the possibility of superimposed CIDP in a patient withdiabetes nor should they be used to deny IVIg treatment topatients with CIDP and diabetes Rather the findings in thisnegative study highlight the importance of a cautious approach todiagnosing CIDP particularly in the diabetic population

Post hoc exploratory analyses of our data resulted in severalinteresting findings and raised some further questions Firsta responder analysis showed improvement of ge1 ONLS pointin 923 (39) of IVIg phases and 521 (24) of placebophases and improvements of ge2 ONLS points in 3 patients allduring the IVIg phase This finding suggested that a subset ofpatients responded to IVIg therapy However the effect mayhave been diluted by other patients who either (1) did not havea treatment-responsive neuropathy (ie diabetic poly-neuropathy) or (2) had irreversible axonal loss possibly relatedto a longer duration of neuropathy Second we noted a lack ofsignificant effect in the subgroup of 1825 patients whomet theEFNS electrophysiologic criteria for CIDP38 possibly sug-gesting that these criteria are suboptimal in the setting of di-abetes In a previous publication from our group we reportedthat in patients with CIDPwith diabetes increasing numbers ofdemyelinating features could not predict treatment re-sponsiveness unlike in patients with CIDP but without DM40

Despite the rarity of CIDP modified diagnostic criteria in thesetting of diabetes might be required This concept has alreadybeen explored by the Lotan et al41 group who proposeda novel scoring system incorporating specific clinical features(namelymdashprogressiverelapsing motor weakness of 2ndash6months of duration proximal involvement upper limb weak-ness large gt small-fiber sensory impairment and recent onsetrelatively well-controlled diabetes) electrophysiological fea-tures and ancillary studies Similarly the Laughlin et al25

publication used specific clinical criteria (symmetrical poly-radiculoneuropathy lasting gt8 weeks with proximal and distalinvolvement reduced reflexes and large gt small-fiber sensoryimpairment) to preselect patients with suspected CIDP

Our study results would seem to conflict with previous retro-spective studies showing treatment responsiveness in patientswith CIDP with diabetes1942 However we believe that theseretrospective studiesmay have had selection bias whereby thosepatients reported to have CIDP and diabetic polyneuropathywere those who exhibited treatment response (hence confirm-ing the clinical suspicion) Conversely our current studyanswers the more practical clinical question of how to managepatients with diabetes presenting with demyelinating poly-neuropathy features In addition the current study was specif-ically looking for improvement in the ONLS and R-ODS scores(validated outcome measures in CIDP) whereas other retro-spective studies defined improvement based on MRC musclescoring Rankin scale and global clinical impressionmdashwhich areless specific to inflammatory neuropathies

There are some limitations to our study that require mentionFirst the allocation was performed in 5-subject blocks whichresulted in slight inequality in the initial phase (14 IVIg and 11placebo) Second there was an unblinded study coordinatorin our clinic who was responsible for allocation design and forretrieving the product to be administered (which originatedfrom the blood bank for IVIg and from the pharmacy forplacebo) To mitigate this potential risk of breaking the blindthe coordinator was forbidden from speaking with other studypersonnel during the trial However we did not perform anypost hoc surveys to determine whether this resulted in anyunblinding Third the trial design may have influenced theoutcome as there were a small number of patients and shorttreatment phases in a chronic disorder and it may be thatlonger treatment phases would have shown different resultsAlso the interval of every 4 weeks for maintenance therapymay have undertreated patients as some other trials have usedq3week dosing7 Fourth there was a relatively high level ofdropout (425 patients) although a sufficient numberremained in the trial to perform adequately powered statisticalcalculations Fifth the primary end point looked for im-provement in the ONLS score but it may miss those patientswho had subtle improvements or stopped deteriorating Sixththe duration of neuropathy at baseline may have been longerthan expected in typical CIDP favoring a diabetes-relatedetiology Seventh the clinical and electrodiagnostic inclusioncriteria were relatively liberal and did not conform to pub-lished CIDP diagnostic criteria thus risking the inclusion ofroutine diabetic polyneuropathy However this design wasdeliberate to simulate the clinical scenario of patients withdiabetes and some suggestiveborderline CIDP features Fi-nally there seems to have been a mild degree of carryovereffect which is a risk inherent in any crossover study

In conclusion this study provides Class I evidence that IVIgdoes not reduce disability in patients with demyelinatingpolyneuropathy features and diabetes It is imperative thata careful history clinical examination and electrodiagnostictesting be used to recognize those patients with diabetes whohave demyelinating abnormalities due to diabetes alone vsthose who have true CIDP This study does not exclude thepossibility of CIDP in a patient with diabetesmdashbut rathersuggests that demyelinating changes on electrophysiologymay be related to diabetes itself thus not responsive to im-munomodulatory treatment

AcknowledgmentThe authors acknowledge Ms Mylan Ngo and Dr EduardoNg for their assistance

Study fundingThis trial is funded by Grifols Therapeutics Inc

DislosureA Breiner reports pilot research grants from the GBS-CIDPFoundation and Grifols (the current study) and speaker orother honoraria from CSL Behring Pfizer Allergan and

8 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

Akcea Therapeutics C Barnett Tapia reports an investigator-initiated research grant from Octapharma consultant feesfrom UCB and CSL and potential royalties from the MGIIscore LE Lovblom reports no disclosures BA Perkinsreports grants from the CIHR NIH and JDRF speakerhonoraria from Medtronic Johnson amp Johnson InsuletAbbott NovoNordisk and Sanofi and research grant supportfrom Medtronic and Boehringer Ingelheim and serves asa consultant for Boehringer Ingelheim Insulet and NovoNordisk HD Katzberg reports clinical trial funding andinvestigator-initiated grants for CSL Behring Octapharmaand Grifols and speaker fees consulting travel support andor DSMB participation for CSL Behring Amazentis Flex-pharma Momenta Octapharma Terumo Grifols and Gen-zyme V Bril reports acting as a consultant for CSL BehringUCB Argenx Alnylam Alexion Grifols Octapharma Takeda(Shire) and Pfizer She has received research support fromCSL Behring UCB Argenx Grifols and Octapharma Go toNeurologyorgNN for full disclosures

Publication historyReceived by Neurology Neuroimmunology amp NeuroinflammationFebruary 13 2019 Accepted in final form May 28 2019

References1 Koller H Kieseier BC Jander S Hartung HP Chronic inflammatory demyelinating

polyneuropathy N Engl J Med 20053521343ndash13562 Brannagan TH III Current diagnosis of CIDP the need for biomarkers J Peripher

nervous Syst 201116(suppl 1)3ndash133 Van den Bergh PY Rajabally YA Chronic inflammatory demyelinating poly-

radiculoneuropathy Presse Med 201342e203ndashe2154 Dyck PJ Litchy WJ Kratz KM et al A plasma exchange versus immune globulin

infusion trial in chronic inflammatory demyelinating polyradiculoneuropathy AnnNeurol 199436838ndash845

5 Hahn AF Bolton CF Zochodne D Feasby TE Intravenous immunoglobulin treat-ment in chronic inflammatory demyelinating polyneuropathy A double-blindplacebo-controlled cross-over study Brain 1996119(pt 4)1067ndash1077

6 Hughes R Bensa S Willison H et al Randomized controlled trial of intravenousimmunoglobulin versus oral prednisolone in chronic inflammatory demyelinatingpolyradiculoneuropathy Ann Neurol 200150195ndash201

7 Hughes RA Donofrio P Bril V et al Intravenous immune globulin (10 caprylate-chromatography purified) for the treatment of chronic inflammatory demyelinatingpolyradiculoneuropathy (ICE study) a randomised placebo-controlled trial LancetNeurol 20087136ndash144

8 Mendell JR Barohn RJ Freimer ML et al Randomized controlled trial of IVIg inuntreated chronic inflammatory demyelinating polyradiculoneuropathy Neurology200156445ndash449

9 Nobile-Orazio E Cocito D Jann S et al Intravenous immunoglobulin versus in-travenous methylprednisolone for chronic inflammatory demyelinating poly-radiculoneuropathy a randomised controlled trial Lancet Neurol 201211493ndash502

10 van Schaik IN Eftimov F van Doorn PA et al Pulsed high-dose dexamethasoneversus standard prednisolone treatment for chronic inflammatory demyelinatingpolyradiculoneuropathy (PREDICT study) a double-blind randomised controlledtrial Lancet Neurol 20109245ndash253

11 Thompson N Choudhary P Hughes RA Quinlivan RM A novel trial design to studythe effect of intravenous immunoglobulin in chronic inflammatory demyelinatingpolyradiculoneuropathy J Neurol 1996243280ndash285

12 De Sousa EA Chin RL Sander HW Latov N Brannagan TH III Demyelinatingfindings in typical and atypical chronic inflammatory demyelinating polyneuropathysensitivity and specificity J Clin Neuromuscul Dis 200910163ndash169

13 Dunnigan SK Ebadi H Breiner A et al Conduction slowing in diabetic sensorimotorpolyneuropathy Diabetes Care 2013363684ndash3690

14 Herrmann DN Ferguson ML Logigian EL Conduction slowing in diabetic distalpolyneuropathy Muscle Nerve 200226232ndash237

15 Wilson JR Stittsworth JD Jr Kadir A Fisher MA Conduction velocity versus am-plitude analysis evidence for demyelination in diabetic neuropathy Muscle Nerve1998211228ndash1230

16 Behse F Buchthal F Carlsen F Nerve biopsy and conduction studies in diabeticneuropathy J Neurol Neurosurg Psychiatry 1977401072ndash1082

17 Younger DS Rosoklija G Hays AP Peripheral nerve immunohistochemistry in di-abetic neuropathy Semin Neurol 199616139ndash142

18 Chio A Plano F Calvo A et al Comorbidity between CIDP and diabetes mellitusonly a matter of chance Eur J Neurol 200916752ndash754

19 Gorson KC Ropper AH Adelman LS Weinberg DH Influence of diabetes mellituson chronic inflammatory demyelinating polyneuropathy Muscle Nerve 20002337ndash43

20 Kalita J Misra UK Yadav RK A comparative study of chronic inflammatory de-myelinating polyradiculoneuropathy with and without diabetes mellitus Eur J Neurol200714638ndash643

21 Sharma KR Cross J Farronay O Ayyar DR Shebert RT Bradley WG Demyelinatingneuropathy in diabetes mellitus Arch Neurol 200259758ndash765

22 Uncini A De Angelis MV Di Muzio A et al Chronic inflammatory demyelinatingpolyneuropathy in diabetics motor conductions are important in the differentialdiagnosis with diabetic polyneuropathy Clin Neurophysiol 1999110705ndash711

23 Bril V Blanchette CM Noone JM Runken MC Gelinas D Russell JW The dilemmaof diabetes in chronic inflammatory demyelinating polyneuropathy J DiabetesComplications 2016301401ndash1407

24 Chio A Cocito D Bottacchi E et al Idiopathic chronic inflammatory demyelinatingpolyneuropathy an epidemiological study in Italy J Neurol Neurosurg Psychiatry2007781349ndash1353

25 Laughlin RS Dyck PJ Melton LJ III Leibson C Ransom J Dyck PJ Incidence andprevalence of CIDP and the association of diabetes mellitus Neurology 20097339ndash45

26 Lozeron P Nahum L Lacroix C Ropert A Guglielmi JM Said G Symptomaticdiabetic and non-diabetic neuropathies in a series of 100 diabetic patients J Neurol2002249569ndash575

27 Latov N Biomarkers of CIDP in patients with diabetes or CMT1 J Peripher Nerv Syst201116(suppl 1)14ndash17

28 Dunnigan SK Ebadi H Breiner A et al The characteristics of chronic inflammatorydemyelinating polyneuropathy in patients with and without diabetesmdashan observa-tional study PLoS One 20149e89344

29 Dunnigan SK Ebadi H Breiner A et al Comparison of diabetes patients with ldquode-myelinatingrdquo diabetic sensorimotor polyneuropathy to those diagnosed with CIDPBrain Behav 20133656ndash663

30 American Diabetes Association Diagnosis and classification of diabetes mellitusDiabetes care 201033(suppl 1)S62ndashS69

31 Graham RC Hughes RA A modified peripheral neuropathy scale the OverallNeuropathy Limitations Scale J Neurol Neurosurg Psychiatry 200677973ndash976

Appendix Authors

Name Location Role Contribution

Ari BreinerMD MScFRCPC

University ofOttawaOttawaCanada

Author Conceived the study wrotethe protocol recruitedparticipants collected dataand wrote the first draft ofthe manuscript

CarolinaBarnett MDPhD

University ofTorontoTorontoCanada

Author Advised on the protocolrecruited participants andmade substantialcontributions to thedraftingrevision of themanuscript for intellectualcontent

Leif ErikLovblomMSc

University ofTorontoTorontoCanada

Author Performed data analysisand draftedrevisedportions of the manuscriptrelated to data analysis

Bruce APerkins MDMPH FRCPC

University ofTorontoTorontoCanada

Author Advised on the protocolrecruited participants andmade substantialcontributions to thedraftingrevision of themanuscript for intellectualcontent

Hans DKatzbergMD MScFRCPC

University ofTorontoTorontoCanada

Author Advised on the protocolrecruited participants andmade substantialcontributions to thedraftingrevision of themanuscript for intellectualcontent

Vera BrilMD FRCPC

University ofTorontoTorontoCanada

Author Conceived the study wrotethe protocol recruitedparticipants collected dataand wrote the first draft ofthe manuscript

NeurologyorgNN Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 9

32 van Nes SI Vanhoutte EK van Doorn PA et al Rasch-built Overall Disability Scale (R-ODS) for immune-mediated peripheral neuropathies Neurology 201176337ndash345

33 AANEM Recommended Policy for Electrodiagnostic Medicine Am Acad Neuro-muscul Electrodiagnostic Med 200411ndash16

34 Bolton CF Benstead TJ GrandrsquoMaison F Tardif GS Weston LE Minimum stand-ards for electromyography in Canada a statement of the Canadian Society of ClinicalNeurophysiologists Can J Neurol Sci 200027288ndash291

35 SchoenfeldDA Statistical considerations for clinical trials and scientific experiments [online]Available at hedwigmghharvardedusample_sizesizehtml Accessed January 29 2015

36 Merkies ISJ Evaluation of Scales and Measurement Instruments in Immune-Mediated Polyneuropathy Rotterdam Erasmus University 2001

37 Merkies IS van Nes SI Hanna K Hughes RA Deng C Confirming the efficacy ofintravenous immunoglobulin in CIDP through minimum clinically important dif-ferences shifting from statistical significance to clinical relevance J Neurol NeurosurgPsychiatry 2010811194ndash1199

38 Van den Bergh PY Hadden RD Bouche P et al European Federation of NeurologicalSocietiesPeripheral Nerve Society guideline on management of chronic

inflammatory demyelinating polyradiculoneuropathy report of a joint task force ofthe European Federation of Neurological Societies and the Peripheral NerveSocietymdashfirst revision Eur J Neurol 201017356ndash363

39 Stewart JD McKelvey R Durcan L Carpenter S Karpati G Chronic in-flammatory demyelinating polyneuropathy (CIDP) in diabetics J Neurol Sci199614259ndash64

40 Abraham A Breiner A Katzberg HD Lovblom LE Perkins BA Bril V Treatmentresponsiveness in CIDP patients with diabetes is associated with unique electro-physiological characteristics and not with common criteria for CIDP Expert Rev ClinImmunol 201511537ndash546

41 Lotan I Hellman MA Steiner I Diagnostic criteria of chronic inflammatory de-myelinating polyneuropathy in diabetes mellitus Acta Neurol Scand 2015132278ndash283

42 Haq RU Pendlebury WW Fries TJ Tandan R Chronic inflammatory de-myelinating polyradiculoneuropathy in diabetic patients Muscle Nerve 200327465ndash470

10 Neurology Neuroimmunology amp Neuroinflammation | Volume 6 Number 5 | September 2019 NeurologyorgNN

DOI 101212NXI000000000000058620196 Neurol Neuroimmunol Neuroinflamm

Ari Breiner Carolina Barnett Tapia Leif Erik Lovblom et al diabetes

Randomized controlled crossover study of IVIg for demyelinating polyneuropathy and

This information is current as of July 10 2019

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm

ServicesUpdated Information amp

httpnnneurologyorgcontent65e586fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent65e586fullhtmlref-list-1

This article cites 40 articles 6 of which you can access for free at

Subspecialty Collections

httpnnneurologyorgcgicollectionperipheral_neuropathyPeripheral neuropathy

rolled_consort_agreementhttpnnneurologyorgcgicollectionclinical_trials_randomized_contClinical trials Randomized controlled (CONSORT agreement)

httpnnneurologyorgcgicollectionclass_1Class I

nating_polyneuropathyhttpnnneurologyorgcgicollectionchronic_inflammatory_demyeliChronic inflammatory demyelinating polyneuropathyfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm

ServicesUpdated Information amp

httpnnneurologyorgcontent65e586fullhtmlincluding high resolution figures can be found at

References httpnnneurologyorgcontent65e586fullhtmlref-list-1

This article cites 40 articles 6 of which you can access for free at

Subspecialty Collections

httpnnneurologyorgcgicollectionperipheral_neuropathyPeripheral neuropathy

rolled_consort_agreementhttpnnneurologyorgcgicollectionclinical_trials_randomized_contClinical trials Randomized controlled (CONSORT agreement)

httpnnneurologyorgcgicollectionclass_1Class I

nating_polyneuropathyhttpnnneurologyorgcgicollectionchronic_inflammatory_demyeliChronic inflammatory demyelinating polyneuropathyfollowing collection(s) This article along with others on similar topics appears in the

Permissions amp Licensing

httpnnneurologyorgmiscaboutxhtmlpermissionsits entirety can be found online atInformation about reproducing this article in parts (figurestables) or in

Reprints

httpnnneurologyorgmiscaddirxhtmlreprintsusInformation about ordering reprints can be found online

Academy of Neurology All rights reserved Online ISSN 2332-7812Copyright copy 2019 The Author(s) Published by Wolters Kluwer Health Inc on behalf of the AmericanPublished since April 2014 it is an open-access online-only continuous publication journal Copyright

is an official journal of the American Academy of NeurologyNeurol Neuroimmunol Neuroinflamm