`
`Observations on Severe Ulnar Neuropathy
`in Diabetes
`
`W. Schady
`B. Abuaisha
`A. J. M. Boulton
`
`ABSTRACT
`
`We describe the clinical and neurophysiologic
`findings in a group of diabetic patients with a
`severe ulnar neuropathy. All patients attending a
`large inner-city diabetes center were
`prospectively screening for hand wasting and
`weakness due to ulnar nerve disease. Twenty
`diabetic patients fulfilling the clinical criteria
`underwent nerve conduction studies and
`electromyography. All but one patient with a motor
`ulnar neuropathy had systemic complications,
`mostly severe: ten were amputees, four had had a
`renal transplant, and two were blind. The onset
`of hand weakness was sudden in five. All patients
`had a classical "ulnar hand" (bilateral in five) but
`forearm muscles were little affected. Sensory loss
`was prominent in only one-half. Nerve conduction
`studies showed markedly reduced ulnar motor
`responses (mean, 1.2 m V versus 7.4 m V in
`
`controls) and ulnar/median motor ratios. Motor
`conduction was disproportionately slowed across
`the elbows, with or without conduction block, in
`only eight of 34 affected ulnar nerves. Five of these
`patients had a habit of leaning on their elbows and/
`or a Tinel's sign. Median sensory action potentials
`(SAPs) were recordable in 12 patients but ulnar
`SAPs were absent in 30 of 34 affected nerves.
`Electromyography revealed advanced denervation
`of ulnar supplied hand muscles. We conclude that
`motor ulnar neuropathy is not uncommon in
`patients with diabetes of long standing, especially
`in those with severe systemic complications. Nerve
`entrapment at the elbows occurs in some, but in
`many the lesion is axonal, and damage may occur
`through ischemia.
`(Journal of Diabetes and Its
`Complications 12; 3: 128-132, 1998.) © 1998 Elsevier
`Science Inc.
`
`INTRODUCTION
`
`It is well known that mononeuropathies can occur
`
`in diabetic patients, sometimes in the absence of
`a clinically significant background polyneuropa(cid:173)
`thy. There are descriptions of diabetic cranial,
`median, ulnar, femoral, peroneal, and truncal nerve
`lesions. Some occur as a result of external pressure or
`entrapment, such as the median nerve at the wrist or
`
`Departments of Neurology (W.S.) and Medicine (B.A., A.J.M.B.),
`Manchester Royal Infirmary, Manchester, United Kingdom
`Reprint requests to be sent to: Dr. W. Schady, Department of
`Neurology, Manchester Royal Infirmary, Oxford Road, Manchester
`M139WL, United Kingdom.
`
`Journal of Diabetes and Its Complications 1998; 12:128-132
`© 1998 Elsevier Science Inc. All rights reserved.
`655 Avenue of the Americas, New York, NY 10010
`
`the common peroneal nerve at the neck of the fibula.
`Though the ulnar nerve is often mentioned as poten(cid:173)
`tially vulnerable, there are surprisingly few detailed
`descriptions of ulnar neuropathy in diabetic patients.
`The prevailing view is that it occurs as a result of
`compression in the cubital tunnel. Marked slowing
`of motor conduction in the elbow segment has been
`observed in 15 diabetic patients with symptoms of an
`ulnar nerve lesion. 1 Seen from another perspective, 17%
`of 46 patients who had a surgical decompression for
`ulnar neuropathy suffered from diabetes.2
`While sensory disturbance in the little and ring fin(cid:173)
`gers from ulnar nerve compression at the elbow is
`probably quite common in diabetic patients, we were
`interested in studying those with more substantial dis-
`
`1056-8727 /98/$19.00
`PII S1056-8727(97)00094-9
`
`Sanofi Exhibit 2138.001
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`J Diab Comp 1998; 12:128-132
`
`ULNAR NEUROPATHY IN DIABETES 129
`
`ability, i.e., patients with predominantly motor ulnar
`neuropathies. This report deals with the clinical and
`electrophysiologic findings in 20 such patients and
`highlights the fact that entrapment may not be the
`main mechanism of damage in more disabling forms
`of ulnar diabetic neuropathy.
`
`METHODS
`
`Patients. Diabetic patients attending the Manchester
`Diabetes Centre were screened for small muscle wast(cid:173)
`ing in one or both hands. Those in whom such wasting
`was identified were referred for examination by a neu(cid:173)
`rologist who, in turn, selected patients with an ulnar
`neuropathy. The criteria for this diagnosis were: (1)
`moderate/severe wasting and weakness of interossei
`and abductor digit minimi, (2) sparing of the thenar
`muscle mass, and (3) absence of any other pathology
`likely to cause muscle wasting. Patients were excluded
`if they had symptoms of cervical spondylotic radiculo(cid:173)
`pathy, namely pain and/ or paresthesia radiating down
`the affected limb, if there were features of a thoracic
`outlet syndrome, or if they suffered from any other
`disease liable to cause a neuropathy.
`Twenty patients fulfilling the above criteria were
`identified over an 18-month period. They underwent
`detailed clinical and neurophysiologic assessment.
`
`Clinical Evaluation. The neurological symptom
`score (NSS)3 was obtained by means of a standardized
`questionnaire. Complaints of weakness at various sites,
`sensory disturbance and autonomic dysfunction were
`recorded as either present or absent and the number
`of symptoms present constituted the score (maximum
`total, 17). In addition, a neurological disability score
`(NDS) was calculated by assessing limb weakness, re(cid:173)
`flexes, and sensation on both sides and scoring the
`severity of the deficit from O (absent) to4 (very severe).3
`The NSS and NDS were used as measures of the sever(cid:173)
`ity of the patient's background diabetic polyneuropa(cid:173)
`thy. After a detailed neurological examination, the
`palmar and dorsal aspect of the hands were photo(cid:173)
`graphed.
`
`Nerve Conduction Studies. Nerve conduction stud(cid:173)
`ies were performed with standard techniques, using a
`Medelec MS6 electrodiagnostic system (Medelec, Old
`Woking, UK). Surface recording and stimulating elec(cid:173)
`trodes were employed throughout. Motor conduction
`studies were undertaken on the median, ulnar, com(cid:173)
`mon peroneal, and tibial nerves and sensory action
`potentials (SAPs) were obtained from the median, ul(cid:173)
`nar, and sural nerves (both upper limbs and right lower
`extremity). Limb surface temperature was maintained
`above 32°C. Upper limb SAPs were recorded ortho(cid:173)
`dromically, sural SAPs antidromically. For the ulnar
`nerve, motor responses were recorded at the hypothe(cid:173)
`nar eminence and conduction velocity was measured
`
`across the elbow as well as in the forearm segment,
`with the forearm flexed at 120°.
`The electrophysiological data were compared with
`those from 60 healthy controls, aged 37-73 years. Para(cid:173)
`metric statistics were used for most comparisons and
`the limits of normality were defined as mean :::'::: 2.5
`SD. For some ratios that were non-normally distributed
`nonparametric statistics were employed.
`
`Electromyography. Concentric needle electromyog(cid:173)
`raphy was carried out in the following muscles in the
`affected limbs: first dorsal interosseous, abductor digiti
`minimi, abductor pollicis brevis, and flexor carpi ul(cid:173)
`naris or flexor digitorum profundus. Selected lower
`limb muscles were also sampled.
`
`RESULTS
`
`Clinical Assessment. Results are shown in Table 1.
`The patients comprised 16 men and 4 women. Their
`ages ranged from 39 to 77 years (mean, 60.3). Seven had
`type I (insulin-dependent) and 13 type II (non-insulin(cid:173)
`dependent) diabetes. The mean duration of diabetes
`was 22.3 years (range, 4-38 years). Seven of the type
`II patients had been taking insulin for 2-10 years prior
`to their assessment for the purpose of this study. Con(cid:173)
`trol of the diabetes was variable: mean glycosylated
`hemoglobin values for the previous 4 years ranged
`from 5.3% to 12.7% (normal range for nondiabetic sub(cid:173)
`jects, 5%-7%). None of the patients drank alcohol to
`excess, and none was on potentially neurotoxic medi(cid:173)
`cation.
`All but one patient had long-term complications of
`diabetes. The serum creatinine was mildly elevated
`(less than 150 µmol/L) in eight, moderately elevated
`(150-400 µmol/L) in nine, and markedly elevated in
`three (pre-dialysis, 518,824, and 1000 µmol/L, respec(cid:173)
`tively). One patient was on continuous ambulatory
`peritoneal dialysis, and four others had received a renal
`transplant. All but two patients had retinopathy requir(cid:173)
`ing laser therapy and two were registered blind. Most
`(15 of 20) had symptoms of intermittent claudication
`or other manifestations of peripheral vascular disease.
`One-half of the patients had had amputations of toes or
`one or both legs. Seven had symptomatic heart disease.
`Four patients had mild/moderate intermittent neck
`pain with radiological changes of early cervical spon(cid:173)
`dylosis and one had had a cervical laminectomy 5 years
`earlier. None of them had symptoms suggestive of
`ongoing cervical nerve root disease.
`Neurological symptom scores were low (median, 5;
`range, 3-8). They were compared to those of 12 control
`subjects attending hospital for non-neurological condi(cid:173)
`tions (mean age, 56.5 years; range, 39-78 years). The
`control scores were 0-2, reflecting non-motor com(cid:173)
`plaints, such as postural fainting, numbness, and impo(cid:173)
`tence. The median neurological disability score (cor-
`
`Sanofi Exhibit 2138.002
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`130 SCHADY ET AL.
`
`J Diab Comp 1998; 12:128-132
`
`TABLE 1. CLINICAL AND ELECTROPHYSIOLOGIC DATA IN OUR 20 PATIENTS WITH A SEVERE
`ULNAR NEUROPATHY
`
`Patient
`Number
`
`Type of
`Age
`(years) Diabetes
`
`R
`
`L
`
`R
`
`Duration of Duration of CMAP (mV) MCV Forearm (m/sec) MCV Elbow (m/sec)
`Diabetes
`Symptoms
`(years)
`(years)
`
`L
`
`R
`
`L
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`
`70
`59
`39
`58
`68
`65
`49
`52
`56
`50
`71
`77
`59
`62
`65
`71
`69
`56
`54
`75
`
`II
`I
`I
`II
`II
`II
`I
`II
`I
`I
`II
`II
`I
`II
`II
`II
`II
`I
`II
`II
`
`20
`35
`23
`16
`20
`17
`35
`27
`28
`26
`17
`27
`29
`30
`8
`6
`18
`38
`23
`4
`
`4
`2
`1
`8/2
`3
`7
`15
`9
`3
`5
`1/,
`2
`2
`2
`1
`½
`2
`6
`7
`2
`
`0
`2.0
`0.4
`0
`1.4
`0.2
`0.8
`
`0.2
`0.2
`
`1.2
`2.4
`1.6
`0.7
`2.8
`1.7
`0.8
`
`2.9
`
`1.6
`0.6
`
`1.0
`0.1
`0.7
`2.6
`0.2
`0.4
`0.2
`
`3.2
`1.5
`0
`4.2
`1.6
`0.9
`1.5
`2.0
`
`50
`46
`
`46
`25
`26
`
`31
`24
`
`31
`50
`41
`37
`59
`47
`24
`
`53
`
`47
`42
`
`51
`
`34
`44
`33
`38
`42
`
`58
`44
`
`55
`51
`47
`38
`51
`
`63
`40
`
`49
`37
`30
`
`28
`32
`
`35
`39
`24
`30
`43
`40
`40
`
`45
`
`49
`32
`
`33
`
`32
`42
`30
`39
`13
`
`32
`32
`
`37
`37
`39
`30
`57
`
`Age and duration are given in years. The first five patients presented acutely. All electrophysiologic data are for affected ulnar nerves.
`
`rected for amputees) was 54 (range, 20-80) in the
`patients and O in the controls (range, 0-4). All patients
`had absent ankle jerks and distal sensory impairment
`to at least two modalities in the lower limbs.
`Muscle thinning, weakness, or clumsiness in one or
`both hands had been noticeably present for between 4
`months and 15 years (median, 2.5 years). Three patients
`denied any functional loss in the affected hand, despite
`obvious wasting. The onset of motor symptoms was
`insidious in 15 patients and acute in five (four bilateral
`and one unilateral). Seven patients admitted to leaning
`on their elbows frequently. One-half of the patients
`complained of numbness in the hands, often with par(cid:173)
`esthesia on waking, though not necessarily restricted
`to the little and ring fingers.
`The examination findings were characteristic of a
`predominantly motor ulnar neuropathy, affecting one
`hand in six patients (three dominant, three nondomi(cid:173)
`nant) and both sides in the remaining 14 patients ( often
`asymmetric). Severe wasting and weakness were ob(cid:173)
`served in dorsal and palmar interossei and in adductor
`digiti minimi in all, with obvious clawing of the little
`and ring fingers in about one-half. The thenar muscles
`were either completely spared or marginally flattened
`but normal in strength. Flexor carpi ulnaris and/ or
`flexor digitorum profundus III/IV were clinically weak
`in only a quarter of cases. Dupuytren's contracture
`was present in seven. One or more upper limb tendon
`
`reflexes were depressed in most patients, but they were
`absent in just one (both triceps jerks). Sensation was
`distinctly impaired in an ulnar nerve distribution in
`nine patients and more diffusely in the hand in a fur(cid:173)
`ther four. Tinel's sign at the elbows was present in six
`patients.
`
`Nerve Conduction Studies. The amplitude of the
`compound muscle action potential (CMAP) in 34 af(cid:173)
`fected ulnar nerves (distal stimulation) was 1.2 :::'::: 1.0
`m V (mean :::'::: 1 SD) compared to 7.4 :::'::: 1.3 m V in con(cid:173)
`trols. The ulnar CMAP was unrecordable in three pa(cid:173)
`tients and it fell below the minimum normal (mean -
`2.5 SD) in all but one of the remaining patients. Because
`even in the absence of a focal ulnar neuropathy some
`reduction in the ulnar CMAP would be expected due
`to background diabetic polyneuropathy, we compared
`the median and ulnar CMAP amplitudes for each af(cid:173)
`fected limb. The median ratio for median versus ulnar
`CMAP in the patients was 4.7 (range, 0.8-87), as op(cid:173)
`posed to 1.1 (range, 0.5-2.0) in controls (p < 0.001).
`Distal motor latencies were longer in the ulnar than
`the median nerve in six cases but always modestly so
`(difference range, 0.2-1.7 ms).
`The mean (SD) motor conduction velocity in the
`median nerves of the diabetic patients was 43.8 m/ sec
`(5.5). Three had a carpal tunnel syndrome. The mean
`MCV in the forearm segment of affected ulnar nerves
`
`Sanofi Exhibit 2138.003
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`J Diab Comp 1998; 12:128-132
`
`ULNAR NEUROPATHY IN DIABETES 131
`
`was 42.1 m/sec (10.1). Motor conduction velocity was
`disproportionately slowed across the elbow (i.e., by
`more than 27% of the forearm value, the maximum
`normal) in eight affected nerves. Motor conduction
`block in the elbow segment was suspected (a drop in
`amplitude of more than 30% in the absence of temporal
`dispersion) in three of these nerves and in three others.
`In all, eight patients had at least one of these two
`neurophysiologic pointers to ulnar nerve entrapment
`at the elbow. Five of them admitted to a habit of leaning
`on the elbows and/or they had Tinel's sign over the
`ulnar nerve at the elbow. All but one had type II diabe(cid:173)
`tes. The ulnar nerve lesion developed acutely in two
`of these patients. In other respects their patient profile
`was the same as for those who did not have electro(cid:173)
`physiologic evidence of a focal lesion at the elbow.
`A median nerve sensory action potential was re(cid:173)
`cordable, at least on one side, in 13 patients (amplitude
`range, 0.5-7 µ V). By contrast, digital ulnar SAPs could
`be obtained in only two of 34 affected nerves.
`Electromyography (EMG) revealed advanced dener(cid:173)
`vation (fibrillation and marked reduction in the num(cid:173)
`ber of recordable motor units) in interossei and/ or
`hypothenar muscles in 17 patients and less severe de(cid:173)
`nervation at these sites in two. By contrast, EMG was
`normal or only mildly deranged in abductor pollicis
`brevis, flexor carpi ulnaris, and flexor digitorum pro(cid:173)
`fundus.
`Turning to the lower limbs, common peroneal and/
`or tibial motor action potentials could be recorded in
`4/15 patients. They were diminished in two (10-500
`µ V) and normal in two (both with unilateral ulnar
`nerve lesions). In all cases where the common peroneal
`motor conduction velocity could be measured, it was
`reduced (29-32 m/sec). Sural sensory action potentials
`were absent in all but one patient.
`
`DISCUSSION
`
`The 20 patients in this study all had clear-cut motor
`signs of an ulnar nerve lesion: selective and profound
`wasting and weakness of the interossei and hypothe(cid:173)
`nar muscles was the hallmark. We took care to exclude
`patients with thenar involvement or symptoms and
`signs of cervical nerve root or brachial plexus disease.
`Sensory disturbance in the ring and little fingers was
`not a required criterion and, in fact, was present in only
`one-half of the patients. This was somewhat surprising,
`but, from the neurophysiologic findings, there can be
`no doubt that sensory fibers were also affected. Several
`of the patients were uncomplaining as regards their
`hands. This contrasts with patients suffering from dia(cid:173)
`betic retinopathy, in whom sensory symptoms pre(cid:173)
`dominate, with few patients complaining of muscle
`weakness.4 It is noteworthy that a recent study of mus(cid:173)
`cle power in longstanding type I diabetic patients dem-
`
`onstrated significant weakness in the lower limb, but
`not upper limb, muscle strength.5
`Community based studies suggest that the preva(cid:173)
`lence of ulnar neuropathy in diabetic patients may be
`of the order of 2%.6 Our figure of 1 % (20 patients from
`about 2000 attending a diabetes center) is lower despite
`the selected population we screened, presumably be(cid:173)
`cause we did not include patients with principally sen(cid:173)
`sory involvement. In any case, it is clear that the prob(cid:173)
`lem is not rare and that the ulnar nerve should be
`high on the list of limb mononeuropathies in diabetes. 7
`While this may seem obvious to those involved in the
`diagnosis and management of diabetes, it has not been
`carefully documented in the literature. Small muscle
`wasting in the hands is easy to observe and should alert
`the observer to the possibility that a diabetic patient has
`other major long-term complications. Moreover, as the
`majority of patients in the current series had type II
`diabetes, which often presents with long-term compli(cid:173)
`cations, the presence of unexplained muscle wasting
`should suggest the possible diagnosis of diabetes.
`It is worth highlighting some aspects of the patient
`profile in our cases. The male:female ratio was 4:1. Male
`predominance has previously been noted in diabetic
`ulnar nerve lesions,1 by contrast with the much com(cid:173)
`moner occurrence of the carpal tunnel syndrome in
`female diabetic patients. Dominance, age, and type of
`diabetes were not relevant. All patients had signs of a
`distal symmetrical sensorimotor polyneuropathy in
`the lower limbs. Their high neurological disability
`scores6 and very abnormal lower limb nerve conduc(cid:173)
`tion studies indicate that the background polyneuropa(cid:173)
`thy was usually severe.
`One of the most striking findings in our patients was
`the high incidence of diabetic complications, including
`not just retinopathy and renal failure but vascular dis(cid:173)
`ease. This is in keeping with previous reports. 1 Angina
`was common and no fewer than ten patients had had
`an amputation of a toe, foot, or leg for gangrene. It is
`also worth pointing out that the onset of hand symp(cid:173)
`toms was acute in one-quarter of the patients. These
`observations raise the possibility that ischemia is at the
`root of the ulnar nerve lesion in some of our patients.
`Nerve conduction studies (NCS) showed absent or
`very reduced motor and sensory responses from the
`affected ulnar nerves, reflecting loss of axons. There
`was no comparable reduction in the response from the
`ipsilateral median nerves. The EMG finding of selective
`denervation in interossei and/ or hypothenar muscles
`also indicates a severe axonal ulnar neuropathy. This
`can be interpreted in one of two ways: either the lesion
`was primarily compressive but so severe that axon loss
`ensued, or it was axonal from the onset (i.e., probably
`ischemic). Much hinges on the electrophysiologic local(cid:173)
`ization of the lesion by either EMG or NCS.
`The fact that flexor carpi ulnaris (FCU) and flexor
`
`Sanofi Exhibit 2138.004
`Mylan v. Sanofi
`IPR2018-01675
`
`
`
`132 SCHADY ET AL.
`
`J Diab Comp 1998; 12:128-132
`
`digitorum profundus (FDP) muscles were weak in just
`five patients and denervated (as judged by EMG) in
`only three might be taken to indicate that in the remain(cid:173)
`der the lesion was distal to the elbow. Alternatively,
`it could be that the axons innervating the hand muscles
`are more vulnerable to compression because they tend
`to be located medially in the ulnar nerve at the elbow. 8
`Moreover, the motor branch to FCU sometimes leaves
`the ulnar nerve proximally to the medial epicondyle. In
`any event, ulnar neuropathies at the elbow frequently
`spare the forearm muscles.9 In one series of surgically
`proven cubital tunnel syndrome, only five of 36 pa(cid:173)
`tients showed signs of clinical involvement of FCU.2
`As regards nerve conduction studies, we took care
`to perform them with the arm always in the same
`position, namely flexed at 120° at the elbows.10 We were
`able to demonstrate focal entrapment in only one-third
`of the affected nerves (slowing of motor conduction
`and/ or motor conduction block across the elbow seg(cid:173)
`ment). Because NCS are not invariably abnormal in
`patients with the cubital tunnel syndrome,11 it could
`be argued that in our patients the lesion may have
`been at the elbow even though we failed to prove that
`this was so. Having said that, the proportion of false
`negative motor conduction studies at the elbow in non(cid:173)
`diabetic patients with a cubital tunnel syndrome is
`10%-20%,12 a far lower figure than the proportion of
`patients in our study without signs of entrapment at
`the elbow.
`The balance of the neurophysiologic evidence in our
`cases suggests that there is a subgroup of patients with
`longstanding diabetes and major complications who
`have motor manifestations of a severe ulnar neuropa(cid:173)
`thy, in whom the lesion is more than a simple compres(cid:173)
`sive event at the elbow. Ischemia is a possible explana(cid:173)
`tion in such cases, especially in unilateral lesions of
`acute onset. Ischemic mononeuropathies in cranial and
`peripheral nerves have been described before in associ(cid:173)
`ation with diabetes mellitus.13 Interestingly, in such
`cases the motor deficit often outweighs the sensory
`involvement, as in our patients. In instances where the
`ulnar nerve lesion is bilateral and concurrent it may
`be that both mechanisms are operating, namely, com(cid:173)
`pression acting on nerves verging on ischemia. Indeed,
`it is possible that the elbow is a locus minore resistentia
`for both entrapment and vascular damage to the ulnar
`nerve due to repeated subclinical trauma and fibrosis.
`These considerations guided our approach to treat(cid:173)
`ment. We feared that ulnar nerves whose circulation
`was already compromised were vulnerable to further
`damage if subjected to surgical exploration, with little
`chance of recovery of function. Two of the patients
`whose NCS suggested entrapment of the ulnar nerve
`
`at the elbow underwent operative treatment. One had
`a simple decompression and the other an anterior
`transposition of the nerve, but neither improved. Sur(cid:173)
`gery may, of course, still be the preferred treatment
`option in patients with a milder, predominantly sen(cid:173)
`sory ulnar neuropathy, especially if the evolution is
`gradual and the lesion is picked up in its early stages.
`In conclusion, it appears that severe motor ulnar
`neuropathy is not uncommon in male patients with
`diabetes and multiple complications, in whom the eti(cid:173)
`ology may well be ischemic rather than simple com(cid:173)
`pression. Surgical decompression may exacerbate
`rather than alleviate the condition in such cases.
`
`REFERENCES
`1. Fraser DM, Campbell IW, Ewing DJ, Clarke BF: Mono(cid:173)
`neuropathy in diabetes mellitus. Diabetes 28:96-101,
`1979.
`2. LeRoux PD, Ensign TD, Burchiel KJ: Surgical decom(cid:173)
`pression without transposition for ulnar neuropathy:
`factors determining outcome. Neurosurgery 27:709-714,
`1990.
`3. Dyck PJ, Sherman WR, Hallcher LM, Service FJ, O'Brien
`PC, Grina LA, et al.: Human diabetic endoneurial sorbi(cid:173)
`tal, fructose and myo-inositol related to sural nerve mor(cid:173)
`phometry. Ann Neural 8:590-596, 1980.
`4. Boulton AJM: Diabetic neuropathies and pain. Clin En(cid:173)
`docrinol Metab 15:917-931, 1986.
`5. Andersen H, Pousen PL, Mogensen CE, Jakobsen J: Iso(cid:173)
`kinetic muscle strength in longterm IDDM patients in
`relation to diabetic complications. Diabetes 45:440-445,
`1996.
`6. Dyck PJ, Kratz KM, Karnes JL, Litchy WJ, Klein R, Pach
`JM, et al: The prevalence by staged severity of various
`types of diabetic neuropathy, retinopathy, and nephrop(cid:173)
`athy in a population based cohort: The Rochester dia(cid:173)
`betic neuropathy study. Neurology 43:817-824, 1993.
`7. Asbury AK: Focal and multifocal neuropathies of diabe(cid:173)
`tes, in Dick PJ, Thomas PK, Asbury AK, Winegrad AI,
`Porte D (eds), Diabetic Neuropathy. Philadelphia, Saun(cid:173)
`ders, 1987, pp. 45-55.
`8. Sunderland S: Nerves and Nerve Injuries, 2nd edition.
`London, Churchill Livingstone, 1978.
`9. Stewart JD: The variable clinical manifestations of ulnar
`neuropathies at the elbows. J Neural Neurosurg Psychia(cid:173)
`try 50:252-258, 1987.
`10. Kincaid JC, Phillips LH, Daube JR: The evaluation of
`suspected ulnar neuropathy at the elbow. Normal con(cid:173)
`duction study values. Arch Neural 43:44-47, 1986.
`11. Payan P: Electrophysiological localisation of ulnar nerve
`lesions. J Neural Neurosurg Psychiatry 32:208-220, 1969.
`12. Kincaid JC: The electrodiagnosis of ulnar neuropathy
`at the elbow. Muscle Nerve 11:1005-1015, 1988.
`13. Raff MC, Asbury AK: Ischemic mononeuropathy and
`mononeuropathy multiplex in diabetes mellitus. N Engl
`J Med 279:17-22, 1968.
`
`Sanofi Exhibit 2138.005
`Mylan v. Sanofi
`IPR2018-01675
`
`