ELS EVIE R
`
`Clinical europhysiology I 13 (2002) 1429-1 434
`
`www.elsev ier.com/locate/cl inph
`
`An evaluation of gender, obesity, age and diabetes mellitus as risk factors
`for carpal tunnel syndrome
`
`Jefferson Beckera,b,*, Daniel B. Norab,c, Irenio Gomesa,d, Fernanda F. Stringari\
`Rafael Seitensusa, Juliana S. Panossoa, Joao Arthur C. Ehlersc,d
`
`"Deparun.enl of Clinical Neurophysiology, Hospital Lurerano, Universidade Lurerana do Brasil, Porto Alegre, RS, Brazil
`bUnit of Electromyography and Evoked Potenlials, Departme111 of Neurology, Hospital de Cl[nicas de Porto Alegre, Universidade Federal do Rio Grande do
`Sul, Porlo Alegre, RS, Brazil
`' Unit of Clinical Neurophysiology, Deparlmenl of Neurology, Hospital Seto Lucas, Ponlijfcia Uni versidade Cat6/ica do Rio Grande do Sul,
`Porto Alegre, RS, Brazil
`"Deparlmem of Clinical Neurophysiology, Hospilal Mae de Deus, Porto Alegre, RS, Brazil
`
`Accepted 24 June 2002
`
`Abstract
`
`Objectives: The aim of this study is to identify gender, high body mass index (BMI), age and diabetes mellitus (DM) as independent ri sk
`factors (RF) for carpal tunnel syndrome (CTS) and to analyse the strength of association of these factors, both globally and in individual
`subgroups.
`Methods: We performed a case- control study with 79 1 CTS cases and 98 1 controls. Patients were selected from those referred to nerve
`conduction studies and electromyography in 3 university hospitals and two private services. We calculated the odds ratio between the two
`groups to analyse the RF. Possible sources of bias were studied using stratified and multivariate analyses.
`Results: The mean BMI and age were greater in the case group than in the control. Female gender, BMI > 30, age of 4 1-60 years and OM
`were signjficantly more freq uent in the case group. Males tend to have a more severe CTS and DM was a significant RF fo r bilateral lesions.
`Stratified analysis showed female gender, obesity and age of 41-60 years as independent RF. OM, when stratified by BMI category, was not
`significantly associated with CTS.
`Conclusions: Our study confim1s that female gender, obesity and age are independent RF fo r CTS. OM may be a weak RF, especially
`among women. © 2002 Elsev ier Science Ireland Ltd. All rights reserved.
`
`Keywords: Carpal tunnel syndrome; Risk factors; Gender; Body mass index; Age; Diabetes mellitus
`
`1. Introduction
`
`Carpal tunnel syndrome (CTS) is the most commonly
`observed neuropathy in the general population (Dumitru
`and Zwarts, 200 1; Kouyoumdjian et al. , 2002), resulting
`from a compression of the median nerve at the level of
`the carpal tunnel (Tanaka et al., I 997; Leclerc et al.,
`1998; Lam and Thurston, 1998; Kouyoumdjian et al.,
`2000a,b; Dumitru and Zwart~, 2001). The estim ated preva(cid:173)
`lence of this condition, calculated in a Dutch population, is
`about 6.8% in women and 0.6% in men (De Krom et al .,
`1992).
`The most typical symptoms are pain and paresthesia,
`occurring especially at night, at the territory of the median
`nerve in the hands, although they can extend to the area
`
`* Correspondi ng author. Tel.: + 55-51-33782959 .
`E-mail address: jeffersonbecker@hotmail.com (J. Becker).
`
`innervated by the ulnar nerve, or eventually to all of the
`upper limb. It is also very common for patients to wake
`up with hypoesthesia in their hands, to feel paresthesia
`while reading or driving, to drop objects, and to obtain
`temporary relief of their symptoms by shaking hands (Stal(cid:173)
`lings et al., I 997; Lam and Thurston, 1998; Stevens et al.,
`1999; Bland, 2000; Kouyoumdjian et al., 2000a,b; Dumitru
`and Zwarts, 2001). According to Nathan et al. (1998) and
`Salemo et al. (1999), the most reliable method to obtain
`objective diagnosis of CTS is electrodiagnostic studies.
`Some epidemiological studies have been performed to
`identify risk factors for CTS. Although they present some
`contradicting data, the most consistent risk factors in these
`studies have been female gender, obesity, a high body mass
`index (BMI), age above 30 years, repetitive motor acti vity
`and a number of systemic diseases, such as diabetes mellitus
`(DM), rheumatoid arthritis and hypothyroidism (Werner et
`
`1388-2457/02/$ - see front matter © 2002 Elsevier Science Ireland Ltd. All rights reserved .
`PII: S 1388 -2457(02)0020 1-8
`
`CU PH 2002572
`
`Sanofi Exhibit 217 4.001
`Mylan v. Sanofi
`IPR2018-01675
`
`

`

`1430
`
`J. Becker el al. I Clinical Neurophysiology 113 (2002) 1429-1434
`
`al., 1994; Nordstrom et al., 1997; Stallings et al., 1997;
`Tanaka et al., 1997; Lam and Thurston, 1998; Leclerc et
`al., 1998; Buschbacher, 1999; Stevens et al., 1999; Sungpet
`et al., 1999; Bland, 2000; Kouyoumdjian et al., 2000a,b;
`Dumitru and Zwarts, 2001; Kouyoumdjian et al., 2002).
`The possibility of a confusion bias among these variables,
`as well as the strength of each of these associations in
`various subgroups, however, have not been well studied.
`After performing both stratified and multivariate analyses
`of their data, Werner et al. (1994) found that BMI > 29,
`gender and age were all independent RF for CTS. Stallings
`et al. (I 997) also identified a high BMI as an independent
`risk factor. However, even though OM is associated both
`with obesity and CTS, no study has analysed this possible
`source of confusion bias.
`In order to identify independent risk factors for CTS and
`to analyse the strength of association of these factors, both
`globally and in individual subgroups, we performed an
`epidemiological study in a population of patients under(cid:173)
`going nerve conduction studies and electromyography in
`southern Brazil.
`
`2. Methods
`
`We studied all patients who were referred to nerve
`conduction studies and electromyography with sensory or
`motor complaints in the upper limbs in the 7-month period
`between February and August of 2001. Data were gathered
`by 4 neurologists specializing in clinical neurophysiology.
`The study was performed in the neurophysiology units of 3
`university hospitals and two private services in Rio Grande
`do Sul state, Brazil.
`After informed consent was obtained for the study,
`patients were asked about their symptoms, age, gender,
`weight, height and associated diseases. The electrodiagnos(cid:173)
`tic studies (Uncini et al., 1993; Campion, 1996; Stevens,
`1997; Dumitru and Zwarts, 2001) included the following
`techniques: (a) sensory orthodromic conduction of the
`median and ulnar nerves, registered at the wrist, with stimu(cid:173)
`lation at the third and fifth digits, respectively, and at the
`palm of the hand (8 cm distally to the recording (E-1) elec(cid:173)
`trode) in both nerves ' territories; (b) motor conduction of
`the median nerve, registered at the abductor pollicis brevis
`muscle (APB), with stimulation at the wrist (8 cm to E-1)
`and antecubital fossa; (c) sensory antidromic conduction of
`the median and radial nerves, registered at the thumb, with
`stimulation (10 cm to E-1 ) at the nerves' trajectory, if there
`was a conduction abnormality in the ulnar nerve; (d) distal
`motor latencies (DML) from the median nerve to the second
`lumbrical and from the ulnar nerve to the second inteross(cid:173)
`eous muscle with equal distances, in the presence of border(cid:173)
`line differences in the 'a' portion of the study or in the
`absence of response at the APB; and (e) electromyography
`and other neurophysiological studies required for individual
`cases. Surface electrodes were used for registering the data
`
`and extremities were warmed up if skin temperature was
`below 32°C. Studies were performed in one or both of the
`upper limbs, depending on the referring physician's request.
`All studies were performed using either Medelec-Oxford
`Synergy or Nihon-Kohden Neuropack 4 equipment.
`The study had a case- control design. The case group
`included patients with sensory deficits in the affected
`hand, sometimes with symptoms involving the whole
`upper limb or tenar atrophy. All indi viduals in this group
`had a neurophysiologic diagnosis ofCTS . The control group
`included patients with or without upper limb symptoms, but
`not fulfilling neurophysiologic criteria for CTS. Diagnosis
`of CTS was made through the presence of any one of the
`following criteria: (a) a difference greater than 10 mis
`between the conduction velocities of the ulnar and median
`nerves in the palm-wrist segment; (b) a difference greater
`than 0.5 ms between the peak latencies of the palm-wrist
`segments of the median and ulnar nerves; (c) a difference
`greater than 0.5 ms between the sensory peak latencies of
`the median and radial nerves; (d) absence of a sensory or
`mixed response of the median nerve, when a diagnosis of
`polyneuropathy, brachia! plexus injury and median nerve
`injury proximally to the wrist could be excluded; and (e) a
`difference greater than 0.4 ms between the DML from the
`median and ulnar nerves to the second lumbrical and inter(cid:173)
`osseous muscles, respectively.
`Cases which met the criteria described above were
`divided into 3 groups of severity according to the following
`criteria: mild CTS, with sensory amplitude after digital
`stimulation greater in the median than in the ulnar nerve,
`and with a DML from the median nerve to the APB of
`4.5 ms or less; moderate CTS, with sensory amplitude
`after digital stimulation greater in the ulnar than in the
`median nerve (or with an amplitude either lower than
`8 µ V or lower than half of the amplitude in the contralateral
`median nerve, in cases where comparison with the ulnar
`nerve was not possible), and with a DML from the median
`nerve to the APB of 4.5 ms or less; and severe CTS, with a
`DML from the median nerve to the APB greater than
`4.5 ms, reduced amplitude (below 5.0 mV baseline-to(cid:173)
`peak) or absent response. Patients with bilateral lesions
`were classified according to their most severe lesion.
`The control group comprised all patients in which criteria
`for a diagnosis of CTS were not met, excluding those with
`age inferior to 18 years (as there were no cases found in this
`age group), and those in which nerve conduction studies
`were performed only on one side (due to the possibility
`that they might have undiagnosed CTS on the opposite
`side). Patients with previous median nerve surgery were
`excluded in both groups.
`BMI was calculated as weight/height2
`. Patients were clas(cid:173)
`
`(BMI ::=; 21 kg/m 2), medium
`sified as being slender
`(BMI > 21 and <25 kg/m2
`), overweight (BMI 2: 25 and
`
`<30 kg/m 2), obese (BMI 2: 30 and <35 kg/m 2
`) or morbidly
`obese (BMI 2: 35 kg/m 2
`) (Stallings et al., 1997).
`Demographic, clinical and neurophysiological informa-
`
`Sanofi Exhibit 2174.002
`Mylan v. Sanofi
`IPR2018-01675
`
`

`

`J. Becker el al. I Clinical Neurophysiology l 13 (2002 ) 1429- 1434
`
`1431
`
`tion were inserted into a database constructed for this study
`using Microsoft Access ® 97 and data were analysed using
`SPSS ® version 6. For comparisons between mean age and
`BMJ between the two groups, we used Student's t test for
`different variances, with these being previously compared
`by the F test. A P val ue of 0.05 or less was considered
`statistically significant. For the analysis of risk factors for
`CTS , we calculated the odds ratio (OR) between the two
`groups, with a confidence interval of 95%. For the evalua(cid:173)
`tion of possible confounding variables, we used both strati(cid:173)
`fied and multivariate analysis. In stratified analysis, the OR
`was calculated by the Mantel-Haenszel method (ORMH) to
`verify interactions among independent variables. Multivari(cid:173)
`ate analysis was carried out through logistic regression,
`using second- and third-order interactions in the model.
`
`3. Results
`
`The study lasted 7 months, and in this period we
`performed 255 5 nerve conduction studies and electromyo(cid:173)
`graphy in volving at least one of the upper limbs, with a total
`of 415 3 hands examined. Of these, 791 (3 1 %) studies (1339
`hands) filled diagnostic criteria for CTS. Among the I 764
`patients who did not have a diagnosis of CTS, 1050 had a
`bilateral study, and 98 l were older than 18 years, and were
`therefore included in the study.
`
`3.1. Case population
`
`Of those patients with a di agnosis of CTS, nerve conduc(cid:173)
`tion studies and electromyography were performed in both
`upper limbs in 548 (69.2%). Of these, 420 (76.6%) had
`bilateral CTS, 94 (17. I%) had CTS only on the right side,
`and 34 (6.2%) on the left side. Of the remaining patients
`with CTS, in 174 the study was performed only on the right
`side, and in 69 only on the left side. Among all cases, we
`found a female:male ratio of7.4: I . Age varied from 18 to 87
`years, with a mean of 49.1 ± 11.9 and a median of 49. The
`majority of cases (62%) were found in the age group
`between 41 and 60 years. Of the patients with CTS, 7.3%
`mentioned having a systemic disease, most commonly DM
`(4.6%), rheumatic disease (1 .6%) and thyroid dysfunction
`(0.5%). In 5.3 % of cases, a second neurophysiologic diag(cid:173)
`nosis was found, with ulnar neuropathy being the most
`common one.
`
`3.2. Control population
`
`Of the 98 1 patients in the control group, 777 (79 .2%) had
`a normal examination. Among the others, the most frequent
`neurophysiologic fi ndings were polyneuropathy (6.9% ),
`cervical
`radiculopathy
`(5.2%) and ulnar neuropathy
`(1.7%). Approximately two-thirds (67 %) of controls were
`female, and age vari ed between 18 and 90 years, with a
`mean of 45 .9 (± 13.5) and a median of 45. In this group,
`6.8% of people mentioned a systemic disease, with DM
`(2.4%), neoplasia (1.4%) and rheumatic disease (1.3%)
`being the most frequent.
`
`3.3. Risk factors studied
`
`The mean BMJ was greater in the case group (27.5 ± 4.8)
`than in the control group (25.2 ± 4.1 ) (P < 0.0001) . Mean
`age was also greater among cases (P < 0.0001). Age group
`distribution showed that the proportion of patients in age
`between 41 and 60 years was greater in the case group, with
`this age group being later analysed as a risk facto r. With
`respect to BMI, an increase in the frequency of CTS was
`seen w ith a BMI greater than 25; however, thi s difference
`was significant only when BMI was above 30. Among the
`associated diseases studied, only DM was prevalent enough
`to be studied as a risk factor or as a possible confusion
`factor.
`The conditions studied as possible risk factors for CTS,
`therefore, were female gender, age between 41 and 60 years,
`obesity (BMl > 30) and DM. All of these were shown to be
`statistically significant when analysed separately, with
`gender and BMI having the strongest association (Table
`1). When analysing if any of the factors predisposed to
`more severe CTS, we found that only gender was signi fi(cid:173)
`cantly correlated with severity, as the fe male predominance
`was less marked in the patients with severe lesions (Table
`2). Only DM was a significant ri sk factor for the presence of
`bilateral lesions (OR = 9.42; CI 95% = 1.27-69.84).
`
`3.4. Stratified and multivariate analyses
`
`Stratified analysis of the data showed that the independent
`risk factors for CTS were female gender (ORMH = 3.66-
`3.76), obesity (ORMi-, = 2.73- 2.87) and age between 41
`and 60 years (ORM11 = 1.81- 1.92). The presence of DM,
`when stratified by BMI category, was not significantly asso(cid:173)
`ciated with CTS (Table 3), although a non-significant trend
`
`Tabl e I
`Prevalence of ri sk factors in the studied groups, with the respective odds ratio and its 95% confidence interval presented for each association'
`
`Risk factor in study
`
`Cases (79 1) (%)
`
`Controls (98 1) (%)
`
`Female gender
`BM! > 30
`Age between 4 1 and 60 years
`Diabetes mellitus
`
`88.1
`26.8
`62.1
`4.6
`
`67 .0
`11.2
`46.2
`2.5
`
`" OR, odds ratio; Cl 95%, 95% confidence interval ; BM I, body mass index.
`
`OR
`
`3.66
`2.90
`1.91
`1.82
`
`Cl95%
`
`2.84--4.71
`2.25- 3.73
`1.58- 2.3 1
`l.08- 3.06
`
`Sanofi Exhibit 2174.003
`Mylan v. Sanofi
`IPR2018-01675
`
`

`

`1432
`
`J. Becker el al. I Clinical Neurophysiology 113 (2002) 1429-1434
`
`Table 2
`Prevalence of studied factors among patients with severe and mild/moderate CTS, with the respective odds ratio and its 95% confidence interval presented for
`each associationa
`
`Risk factor in study
`
`Severe (400)b %
`
`Mild/moderate (364) b %
`
`Female gender
`BMl > 30
`Age between 41 and 60 yea.rs
`Diabetes mellitus
`
`85.5
`28.3
`62.0
`5.5
`
`91.8
`25.5
`62.4
`3.6
`
`" Twenty-seven patients were not classified wi th respect to CTS severity.
`b OR, odds ratio; Cl 95%, 95% confidence interval; BM!, body mass index.
`
`OR
`
`0.53
`1.15
`0.98
`1.57
`
`Cl 95 %
`
`0.33-0.84
`0.83-1.58
`0.73-1.32
`0.78-3.17
`
`this stratified analysis
`in
`of associat.10n was found
`(ORM 11 = 1.66, CI 95% = 0.94-2.95). The strength of the
`association between female gender and CTS was similar
`among all ages, but it was greater among non-obese patients
`(OR = 4.26) and among diabetic patients (OR = 7 .37).
`Obesity was shown to be a stronger risk factor among
`male patients (OR = 5.64). No association was seen
`between age and CTS among obese and diabetic patient
`subgroups, while DM was a significant risk factor only
`among females (OR = 3.13).
`Among patients with CTS, female gender (ORM 11 = 0.53;
`CI 95% = 0.33- 0.87), was shown in stratified analysis to be
`an independent protection factor for severe disease (i.e.
`there was a strong inverse association with female gender
`and cases with severe CTS). The presence of DM, mean(cid:173)
`while, was an independent risk factor for bilateral lesions in
`
`Table 3
`Stratified analysis of risk factors for CTS"
`
`Risk factor
`
`Stratifying factor
`
`n (RF/no RF)
`
`Female gender
`
`BMI > 30
`
`Age 4 1-60 years
`
`Diabetes mell itus
`
`BM! > 30
`BM! s30
`Age 41-60 years
`Other age groups
`With diabetes
`Without diabetes
`
`Female gender
`Male gender
`Age 41-60 years
`Otl1er age groups
`With diabetes
`Without diabetes
`
`Female gender
`Male gender
`BM! > 30
`BMIS30
`With diabetes
`Without diabetes
`
`Female gender
`Male gender
`BMl > 30
`BMI S 30
`Age 41-60 years
`Other age groups
`
`265/57
`1089/361
`735/209
`619/209
`38/23
`1316/395
`
`265/1089
`57/36 1
`203/74 1
`119/709
`18/43
`304/1407
`
`735/619
`209/209
`203/1 19
`741/709
`3 1/30
`913/798
`
`38/13 16
`23/395
`18/304
`43/1407
`3 1/913
`30/798
`
`stratified analysis (ORMit = 9.32- 9.36; CI 95% = 1.52-
`57.47).
`The final logistic regression model (Table 4) confirmed
`that obesity, female gender and age between 41 and 60 years
`were independent and highly significant risk factors for
`CTS. The association between obesity and male gender
`was also identified as a risk factor, but not that between
`female sex and DM. In thi s analysis, differently from what
`was found through stratified analysis, DM was found to be a
`statistically significant independent risk factor.
`
`4. Discussion
`
`The findings of our study confirm that female gender,
`obesity and age are independent risk factors for CTS, as
`
`OR
`
`1.81
`4.26
`3.84
`3.43
`7.37
`3.65
`
`2.39
`5.64
`2.33
`3.51
`3.34
`2.85
`
`1.95
`1.74
`1.29
`1.94
`2.10
`1.91
`
`3.1 3
`1.55
`1.87
`1.60
`1.98
`1.80
`
`Cl 95 %
`
`ORMH
`
`Cl 95%
`
`0.97- 3.38
`3.21 - 5.66
`2.77- 5.34
`2.35- 5.02
`2.22- 24.43
`2.83-4.70
`
`1.80-3.18
`3.22- 9.89
`1.68-3.24
`2.37-5 .2 1
`0.79-1 4.09
`2.21-3.68
`
`1.57-2.43
`I .07-2.84
`0.76-2.20
`1.57-2.41
`0.59-7.45
`1.57- 2.32
`
`1.46-6.68
`0.44--5.44
`0.44--8.06
`0.82- 3.13
`0.86--4.57
`079-4.09
`
`3.66
`
`3.65
`
`3.76
`
`2.75
`
`2.73
`
`2.87
`
`1.91
`
`1.81
`
`1.92
`
`2.45
`
`1.66
`
`1.89
`
`2.84--4.71
`
`2.85-4.68
`
`2.94--4.81
`
`2. 14--3.53
`
`2.13- 3.50
`
`2.24--3.68
`
`1.57-2.33
`
`1.49-2.21
`
`1.58-2.32
`
`1.36--4.41
`
`0.94--2.95
`
`1.09-3.28
`
`" RF, risk fac tor; OR, odds ratio; ORMH, Mantel- Haenszel odds ratio; Cl 95%, 95% confid ence interval; BM!, body mass index.
`
`Sanofi Exhibit 2174.004
`Mylan v. Sanofi
`IPR2018-01675
`
`

`

`J. Becker el al. I Clinical Neurophysiology l 13 (2002 ) 1429-1434
`
`1433
`
`Table 4
`Final multiple logistic regression model"
`
`Ri sk factor in study
`
`B
`
`SE
`
`R
`
`p
`
`Female gender
`BM! > 30
`Age between 4 1 and 60 years
`Diabetes mel litus
`Male gender and BM! > 30
`Female gender and diabetes
`mellitus
`
`0.63
`0.62
`0.30
`0.40
`0.22
`0.14
`
`0.1 7
`0.08
`0.05
`0.16
`0.08
`0.16
`
`0.07
`0. 15
`0.12
`0.04
`0.04
`0.00
`
`< 0.001
`< 0.00 1
`< 0.00 1
`0.01 2
`0.011
`0.48 1
`
`OR
`
`1.87
`1.85
`1.35
`1.49
`1.24
`1. 12
`
`" B , B value of the logistic regression; SE, standard devi ation of the B
`value; R, R value of the logistic regression; P, P value of the logistic
`regression; OR, odds rat io; BM!, body mass index.
`
`has been described by other authors (Werner et al., 1994;
`Stallings et al. , 1997; Tanaka et al., 1997; Lam and Thur(cid:173)
`ston, 1998; Kouyoumdjian et al., 2002). Literature also
`suggests that DM is a possible ri sk factor for CTS. However,
`its infl uence as an independent factor is not precisely
`defined, as this association could also represent a confusion
`bias due to the association between obesity and type 11 DM
`(Werner et al., 1994). In our analysis we fo und that obesity
`and female gender were the strongest risk factors, even after
`stratification by the presence of DM. Other systemic
`diseases were not prevalent enough to be analysed as risk
`factors or to have an impact as confusion factors.
`One should keep in mind, however, that a possible limita(cid:173)
`tion of this study is the fact that it was undertaken in a
`population of patients referred to neurophysiologic evalua(cid:173)
`tion. Although this might not represent the general popula(cid:173)
`tion, however, both groups were probably similar in most
`aspects except for the outcome in study, which makes the
`study of risk fac tors a valid enterprise in this case. Another
`possible bias is the inclusion, in the control group, of
`patients with incipient CTS and borderline values in neuro(cid:173)
`physiologic analysis, as most control patients complained of
`sensory symptoms, which were sometimes similar to those
`of the syndrome. This, however, would have led us to under(cid:173)
`estimate the strength of any association in study, strength(cid:173)
`ening our belief that the risk factors we found are indeed
`real, and that the associations we fo und could be even stron(cid:173)
`ger.
`The association between obesity and CTS can be
`explained either by the accumulation of fat tissue inside
`the carpal tunnel or by an increase in hydrostatic pressure
`through this canal , exerting a compressive effect on the
`median nerve. Stratified analysis of obesity as a risk factor
`showed a similar association strength among different age
`groups. However, in agreement with what was found by
`Werner et al. (1994), obesity was shown to be a stronger
`risk factor among males. This could be explained by the fact
`that female sex is by itself a strong risk factor, making the
`association with obesity numerically more evident in men.
`Another hypothesis is that there could be a real and inde(cid:173)
`pendent influence of the association of obesity and male
`gender on the risk of CTS through unknown pathophysio-
`
`logic mechanisms, as was suggested by multivariate analy(cid:173)
`sis of our data.
`Although female gender was an independent risk factor
`for CTS, this association was not shown to be significant
`among obese patients. This was probably due to the fact that
`obesity is specially strong as a risk factor among males, as
`was discussed above, making it more difficult to establish
`female gender as a risk factor in this subgroup. On the other
`hand, we found that the association between female gender
`and CTS was greater in the subgroup of diabetic patients,
`although this was not confirmed by multivariate analysis.
`More studies must be done to confirm this finding , as in our
`stu dy the prevalence of diabetic women with CTS was
`small, leading to a wide confidence interval.
`DM is mentioned as a risk fac tor for CTS (Dumitrn and
`Zwarts, 2001), although De Krom et al. (1990) did not find
`this in their study of a Dutch population. In our analysis, the
`presence of DM was found to be a statistically significant
`risk factor by logistic regression; however, when stratified
`analysis was perfonn ed with obesity as the stratifying
`factor, this finding was not confirmed to be signjficant,
`although a trend of association was found. It is feasible
`that DM might not be a real risk factor, being associated
`with CTS only due to its strong relation to a defin ite risk
`factor (obesity). However, considering that the association
`between DM and CTS was sign ificant in multivariate analy(cid:173)
`sis, that a trend of association was found even when analysis
`was stratified by BMI category, and that our sample of
`diabetic patients was relatively small, it is quite possible
`that the lack of significance in stratified analysis was merely
`due to an insufficient sample size. In any case, even if DM is
`indeed a ri sk factor for CTS , it is probably not as strongly
`associated with the disease Like the other risk factors we
`found, as if this association was strong it should have
`been proved significant by the Mantel- Haenszel analysis
`even in our sample. This association could also be present
`only in some subgroups, as stratified analysis showed an
`association of DM and CTS only in the female population;
`however,
`the association of these variables was not
`confirmed to be significant by the logistic regression model.
`The cutoff point in which age begins and ceases to be a
`risk factor varies among the various studies, although most
`agree that the risk of CTS increases after the age of 30
`(Werner et al., 1994; Tanaka et al., 1997; Stallings et al.,
`1997; Lam and Thurston, 1998). ln our study, the age group
`between 41 and 60 years was shown to be an independent
`risk factor for the syndrome, al though the strength of thi s
`association was small. Analysis of increasing age as a risk
`factor is diffic ul t due to the fact that CTS is a chronic
`disease, becoming therefore more prevalent as people
`grow older. Our study confirmed, however, that CTS is
`very rare below the age of 20, as has been observed by
`other authors (Deymeer and Jones, 1994; Werner et al.,
`1994; Al-Qattan et al., 1996; Stallings et al., 1997; Tanaka
`et al. , I 997; Cruz-Martfnez and Arpa, 1998; Lam and Thur(cid:173)
`ston, 1998; Dumitru and Zwarts, 2001).
`
`Sanofi Exhibit 2174.005
`Mylan v. Sanofi
`IPR2018-01675
`
`

`

`1434
`
`J. Becker el al. I Clinical Neurophysiology 113 (2002) 1429- 1434
`
`With respect to risk factors for more severe CTS, our only
`finding was that in male patients with the disease there was a
`significantly greater proportion of severe cases. There
`appears to be no previous description of this finding in the
`literature. Since obesity was not significantly associated
`with a greater proportion of severe disease in our study, as
`was also found by Kouyoumdji an et al. (2002), we believe
`that there could be another physiopathologic explanation for
`the fact that, even though males are much less commonly
`afflicted with CTS than females, those who do develop the
`disease are more likely to have it in a severe form. This
`finding could also be due to females either seeking medical
`attention earlier or being referred for evaluation earlier in
`the course of the disease, with males being diagnosed only
`in later stages of the syndrome.
`The presence of bilateral lesions was more frequent
`among patients with DM. Differently from Sungpet et al.
`( 1999), we did not find an association between obesity and
`bilateral CTS. In our opinion, the findings of Sungpet et al.
`( 1999) were significant due to the fact that the authors
`compared the mean BMI of patients with unilateral and
`bilateral lesions, without analysing obesity as a categorical
`variable. However, in our study, the presence of obesity
`(defined as a BMI greater than 30), which in our opinion
`has greater clinical application, did not show a significant
`association.
`In brief, female gender and obesity were the strongest
`independent risk factors for CTS in our study. We also
`found that CTS is very rare among children and adolescents.
`We could not prove that DM is indeed a ri sk factor for CTS,
`although we believe that it would not be a strong one. We
`think that a prospective study with diabetic patients could be
`a better way to elucidate the real association between this
`disease and CTS .
`
`Acknowledgements
`
`We thank Olavo B. Amaral for editorial assistance.
`
`References
`
`AJ -Qattan MM, Thomson HG, C larke HM . Carpal tunnel syndrome in
`children and adolescents wi th no hi story of trauma. J Hand Surg
`I996;2 1B(l):1 08- I 1 1.
`Bland JDP. The value of the history in the diagnosis of carpal tunnel
`syndrome. J Hand Surg Br 2000;25 B:445-450.
`Buschbacher RM. M ixed nerve conduction studies of the median and ulnar
`nerves. Am J Phys Med Rehabi l 1999;78(Suppl):S69-S74.
`Campion D. Electrodiagnostic testing in ha nd surgery. J Hand Surg
`1996;2 1A947- 956.
`
`Cruz-Martinez A, Arpa J. Carpal tun nel syndrome in child hood: study of 6
`cases. Electroenceph clin Neurophysiol !998; 109 :304-308.
`De Krom MC, Kester AD, Knipschild PG, Spaans F. Risk factors for carpal
`tunnel syndrome. Am J Ep idemiol 1990; 132(6):1102-1110.
`De Krom MC, Knipsch.ild PG, Kester AD, Thijs CT, Boekkooi PF, Spaans
`F. Carpal tunnel syndrome: prevale nce in the general popul ation. J Clin
`Epidemiol 1992;45(4):373- 376.
`Deymeer F, Jones Jr R. Ped iatric median monone uropathies: a clinical and
`electromyographic study. Muscle Nerve 1994; 17:755- 762.
`Dumitru D, Zwarts MJ. Focal peripheral neuropathies. In : Dumitru D,
`Amato AA, Zwa1ts Ml, editors. Electrodiagnostic medicine, Philadel(cid:173)
`phia, PA: Hanley and Belfos, 200 1. pp. !043-1 126.
`Kouyoumdjian JA, Morita MPA, Rocha PRF, Miranda RC, Gouveia GM.
`Body mass index and carpal tu nnel syndrome. Arq Neuropsi qu iatr
`2000a;58(2A):252- 256.
`Kouyoumdjian JA, Morita MPA, Rocha PRF, Miranda RC, Gouveia GM .
`Wrist a nd palm indexes in carpal tunnel syndrome. Arq Neuropsiquiatr
`2000b;58(3A):625--<i29.
`Kouyoumdj ian JA, Zanella OM, Morita MPA. Evaluation of age, body
`mass index, and wrist index as ri sk factors for carpal tunnel syndrome
`severity. Muscle Nerve 2002;25( 1 ):93- 97.
`Lam N, Thurston A. Association of obesity, gender, age and occupation
`wi th carpal tunnel sy ndrome. Aust NZ J Surg 1998;68: 19(}-193.
`Leclerc A, Franchi P, Cristofori B, Delemotte B, Mereau P, Teyssier-Cotte
`C, Touranchet A. Carpal tunnel syndrome and work organisati on in
`repetitive work: a cross-sectional study in France. Occup En viron
`Med 1998;55 :L80- 187.
`Nathan PA, Keniston RC, Myers LE, Meadows KO, Lockwood RS. Natural
`history of med ian nerve sensory conduction in industry: relationship to
`sympto ms and carpal tun nel syndrome in 558 hands over 11 years.
`Muscle Nerve l 998;2 l :711 - 721.
`Nordstrom DL, Vierkant RA, De Stefano F, Layde PM. Risk factors for
`carpal tunnel syndrome in a general population. Occ up En viron Med
`1997;54:734-740.
`Salerno OF, Werner RA, Albers JW, Becker MP, Armstrong TJ, Franzblau
`A. Reliability of nerve conduction stud ies among active workers.
`Muscle Nerve 1999;22: 1372-1 379.
`Stallings SP, Kasdan ML, Soergel TM , Corwin HM. A case-control study
`of obesity as a risk factor for carpal tunnel syndrome in a population of
`600 patients presenting for inde pendent medical exami nation. J Hand
`Surg 1997;22A:211 - 2 15.
`Stevens JC. AAEM min imonograph #26: the electrodiagnosis of carpal
`tunnel syndrome. A merican Association of Electrodiagnostic Medici ne.
`Muscle Nerve 1997;20: 1477-1 486.
`Stevens JC, Smith BE, Weaver AL, Bosch EP, Deen HG, Wil kens JA.
`Symptoms of 100 patients with electro myographically verified carpal
`tunnel syndrome. Muscle Nerve 1999;22: 1448- 1456.
`Sungpet A, Suphachatwong C, Kawinwonggowit V. The relationship
`between body mass index a nd the number of si des of carpal tunnel
`syndrome. J Med Assoc Thai 1999;82(2): 182- 185.
`Tanaka S, Wi ld D, Cameron LL, Freund E. Association of occu pational and
`non-occupational risk factors wi th the prevalence of self- reported
`carpal tunnel syndrome in a national survey of the working population.
`Am J Ind Med ! 997;32:550--556.
`Uncini A, Di Muzio A, Awad J, Manente G, Tafuro M, Gambi D. Sensi(cid:173)
`ti vity of three med ian-to-ulnar comparative tests in di agnosis of mild
`carpal tu nnel syndrome. Muscle erve 1993; 16: 1366-- 1373.
`Werner RA, Albers JW, Franzblau A, Armstrong TJ. The relationship
`between body mass index mid the d iagnosis of carpal tunnel syndrome.
`Muscle Nerve 1994; 17:632--036.
`
`Sanofi Exhibit 2174.006
`Mylan v. Sanofi
`IPR2018-01675
`
`