You are here

  • Home
  • Archive
  • Volume 69, Issue 11
  • Ulnar neuropathy and ulnar neuropathy-like symptoms in relation to biomechanical exposures assessed by a job exposure matrix: a triple case–referent study

Article Text

Article menu

Download PDFPDF

Original article
Workplace
Ulnar neuropathy and ulnar neuropathy-like symptoms in relation to biomechanical exposures assessed by a job exposure matrix: a triple case–referent study
  1. Susanne Wulff Svendsen1,
  2. Birger Johnsen2,
  3. Anders Fuglsang-Frederiksen2,
  4. Poul Frost3
  1. 1Danish Ramazzini Centre, Department of Occupational Medicine, Herning Hospital, Herning, Denmark
  2. 2Department of Clinical Neurophysiology, Aarhus University Hospital, Aarhus, Denmark
  3. 3Danish Ramazzini Centre, Department of Occupational Medicine, Aarhus University Hospital, Aarhus
  1. Correspondence to Dr Susanne Wulff Svendsen, Danish Ramazzini Centre, Department of Occupational Medicine, Herning Regional Hospital, Gl. Landevej 61, Herning DK-7400, Denmark; susasven{at}rm.dk

Abstract

Objectives We aimed to evaluate relations between occupational biomechanical exposures and (1) ulnar neuropathy confirmed by electroneurography (ENG) and (2) ulnar neuropathy-like symptoms with normal ENG.

Methods In this triple case–referent study, we identified all patients aged 18–65 years, examined with ENG at a neurophysiological department on suspicion of ulnar neuropathy, 2001–2007. We mailed a questionnaire to 546 patients with ulnar neuropathy, 633 patients with ulnar neuropathy-like symptoms and two separate groups of community referents, matched on sex, age and primary care centre (risk set sampling). The two patient groups were also compared to each other directly. We constructed a Job Exposure Matrix to provide estimates of exposure to non-neutral postures, repetitive movements, hand–arm vibrations and forceful work. Conditional and unconditional logistic regressions were used.

Results The proportion who responded was 59%. Ulnar neuropathy was related to forceful work with an exposure–response pattern reaching an OR of 3.85 (95% CI 2.04 to 7.24); non-neutral postures strengthened effects of forceful work. No relation was observed with repetitive movements. Ulnar neuropathy-like symptoms were related to repetitive movements with an OR of 1.89 (95% CI 1.01 to 3.52) in the highest-exposure category (≥2.5 h/day); forceful work was unrelated to the outcome.

Conclusions Ulnar neuropathy and ulnar neuropathy-like symptoms differed with respect to associations with occupational biomechanical exposures. Findings suggested specific effects of forceful work on the ulnar nerve. Thus, results corroborated the importance of an electrophysiological diagnosis when evaluating risk factors for ulnar neuropathy. Preventive effects may be achieved by reducing biomechanical exposures at work.

https://doi.org/10.1136/oemed-2011-100499

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    After carpal tunnel syndrome (CTS), ulnar neuropathy at the elbow is the most common form of nerve entrapment.1 The disorder may be diagnosed based on symptoms and signs,2 ,3 and the clinical diagnosis may be confirmed by electroneurography (ENG).3 For clinically diagnosed ulnar neuropathy at the elbow, a prevalence of 0.6–0.8% was found in a French working population.4 Incidence rates of 20–25 per 100 000 person-years were reported for general population samples in Italy5 and the UK6 using a case definition requiring ENG or surgical evidence of ulnar neuropathy at the elbow5 or diagnoses registered in primary care.6

    Risk factors include trauma, systemic disorders (eg, diabetes, hypothyroidism and renal disease)7–9 and potentially alcohol consumption, smoking and high or low body mass index (BMI).9–12 A male preponderance has been reported.5 ,6 ,10 Substantial evidence exists that CTS is related to occupational biomechanical exposures,13 while epidemiological studies are rare regarding occupational risk factors for ulnar neuropathy. In cross-sectional or longitudinal studies of selected occupational groups, clinical case criteria for ulnar neuropathy have been applied, but because of small study sizes, too few cases have been observed to explore relations to occupational risk factors.14–19 A three-year follow-up study used clinical case criteria and identified 15 new cases in a cohort (n=598) with repetitive work.20 Owing to limited exposure contrast, effects of repetitive movements could not be evaluated, but self-reported holding a tool in position was identified as a risk factor.20 Two case–referent studies comprised a total of 10021 or 10111 persons referred for ENG on suspicion of ulnar neuropathy, who were divided into case and referent groups based on ENG findings. These studies found no associations between ulnar neuropathy and self-reported routine flexion of the elbow,11 ,21 gripping of objects11 ,21 and elbow pressure.21 Both studies probably underestimated any effects because all participants had symptoms that could be work-related. A third case–referent study compared patients (n=96) who had surgery at the elbow for ulnar neuropathy confirmed by ENG with patients (n=142) who had surgery for cervical or lumbar herniated disc.22 Exposure assessment was based on a ranking of nine job groups with respect to the physical exposure level. Ulnar neuropathy was associated with average physical exposure level for the entire working life and with recent heavy manual labour. Associations with occupational risk factors have also been described in case reports, including the report that initiated the present study.23

    Potential pathogenetic mechanisms have been described that link biomechanical exposures with ulnar nerve pathology, that is, direct compression,24 increased tension due to stretching,25 pressure due to forceful contraction of the flexor carpi ulnaris muscle especially when the elbow is flexed,26 friction during repetitive excursions27 and exposure to vibration.28 Conceivably, a different set of risk factors and/or other mechanisms may explain ulnar neuropathy-like symptoms with normal ENG, for example, tingling and numbness may be related to tender points in the infraspinatus and other muscles.29

    Our overall hypothesis was that ulnar neuropathy confirmed by ENG is a distinct disorder (nerve damage) that differs from ulnar neuropathy-like symptoms with normal ENG (no nerve damage) with respect to profiles of risk factors and pathogenetic mechanisms. In this study, we aimed to evaluate the following hypotheses:

    1. Ulnar neuropathy confirmed by ENG is associated with occupational biomechanical exposures (forceful work, non-neutral postures, repetitive movements and hand–arm vibrations (HAVs)), and the relations conform to exposure–response patterns.

    2. The profile of exposure–response patterns for ulnar neuropathy confirmed by ENG differs from that observed for ulnar neuropathy-like symptoms with normal ENG; that is, occupational biomechanical exposures have specific effects on the ulnar nerve.

    Results on individual risk factors will be reported in a separate paper.

    Methods

    Design and study population

    We chose a triple case–referent design. To select cases, we obtained data from the Danish National Patient Register30 on all contacts to the Department of Clinical Neurophysiology, Aarhus University Hospital, 1 March 2001–30 June 2007, with a referral diagnosis of mononeuropathy of upper limb (ICD-10 group G56) and a discharge diagnosis of ulnar neuropathy (ICD-10 code G56.2) or no neuropathy (ICD-10 codes Z) (>1 discharge diagnosis was possible). Data were delivered on 21 August 2007. We extracted patients who were ≥18–<65-years-old at the date of their first contact, the index date. We then identified patients who were examined by ENG across the elbow. According to the examination strategy of the department, this was performed only in patients who were clinically suspected of ulnar neuropathy by one of the two (in part of the period three) supervising consultants, based on referral information, history and/or signs. Further selection was based on neurophysiological data. Patients with abnormal ENG of the ulnar nerve were classified as cases of ulnar neuropathy, and patients with normal ENG of the ulnar nerve were classified as cases of ulnar neuropathy-like symptoms (see below). We excluded patients with a traumatic lesion or ulnar neuropathy as part of a polyneuropathy. Owing to registration errors, a number of patients with a Z-diagnosis actually had CTS. We left patients who (also) had CTS in both case groups, except in subanalyses.

    To provide referents, we used the Danish National Health Service Register to identify all persons who had the same primary healthcare provider as the cases at the index date. After exclusion because of death, emigration and name and/or address protection, three referents per case were drawn from these base populations, individually matched on sex, age (±2.5 years) and primary healthcare provider at the index date. Cases were eligible as referents up to the index date (risk-set sampling). The study population was updated in the Danish Civil Registration System 28 November 2007, with exclusion because of death, emigration, address in Greenland, name and/or address protection and protection against enquiries in connection with scientific studies. The triple case–referent study was composed as follows:

    • Study 1: Cases with ulnar neuropathy confirmed by ENG versus their matched referents.

    • Study 2: Cases with ulnar neuropathy-like symptoms with normal ENG versus their matched referents.

    • Study 3: Cases with ulnar neuropathy confirmed by ENG versus cases with ulnar neuropathy-like symptoms with normal ENG used as referents.

    The hypothesis of specific effects of occupational biomechanical exposures on the ulnar nerve would be corroborated if study 1 yielded larger effect measures than study 2 and/or different patterns of associations across exposures. The hypothesis would be further supported by positive findings in study 3 that benefited from ENG examination of both cases and referents. In study 3, the statistical power would be limited by the unmatched design and effects would be underestimated to the extent that the two conditions shared biomechanical risk factors, but significant associations would be difficult to explain by chance or bias.

    The study was authorised by the Danish Data Protection Agency. The Danish administrative regions approved the use of data from the Danish National Health Service Register, and the Danish National Board of Health permitted the use of neurophysiological data. In Denmark, register-based and questionnaire-based studies do not require approval by committees on biomedical research, nor informed consent.

    Neurophysiological examinations and determination of case status

    ENG with needle or surface electrodes was performed by physicians or technicians, respectively. Needle electrodes were preferred, but depending on available staff resources surface electrodes were also used. The needle electrode technique included placement of three needle electrodes near the ulnar nerve at the wrist, 5 cm distal to the medial epicondyle, and 5 cm proximal to the medial epicondyle.31 Placement of the needles was accepted when a compound motor action potential of the abductor digiti minimi muscle could be recorded with a stimulation intensity of <1 mA. With the needles in place, digit five was stimulated with ring electrodes, and averaged sensory responses were recorded at the wrist, below the medial epicondyle and above the medial epicondyle. Motor nerve conduction was recorded with surface electrodes from the abductor digiti minimi and the adductor pollicis muscles by supramaximal stimulation with the needles. The surface electrode technique comprised antidromic sensory nerve conduction studies with stimulation at the wrist and recording from ring electrodes on digit five, and motor nerve conduction studies with bar electrode stimulation at the wrist, 3 cm distal to the medial epicondyle and 7 cm proximal to the medial epicondyle with recordings from the abductor digiti minimi muscle and often also from the first dorsal interosseus muscle. With both techniques, skin temperature was kept above 34°.

    Reference values of the department were used. Distal latencies, motor and sensory amplitudes and conduction velocities were expressed in Z-scores, that is, SD from normal mean values. The Z-score for distal latency was considered abnormal if it was above 2, and Z-scores for motor and sensory amplitude as well as conduction velocity were considered abnormal if they were below −2.32 If at least one Z-score was abnormal or there was a partial motor conduction block,33 the patient was classified as having ulnar neuropathy, otherwise as having ulnar neuropathy-like symptoms. The cut-off values were in accordance with the department's clinical practice.

    To evaluate consequences of altering the cut-off values and to examine the existence of exposure–effect relations between occupational exposures and graded neuropathological severity, we expressed the electrophysiological severity of ulnar nerve affection in two indices, assumed to reflect different neuropathological aspects. We collapsed the two case groups for this purpose. The first index ‘localised demyelination’ was calculated as a combination of the amplitude ratio of the compound muscle action potentials from the abductor digiti minimi muscle obtained by stimulation at the wrist and above the medial epicondyle and the motor nerve conduction velocity to the abductor digiti minimi muscle measured around the medial epicondyle. The second index ‘axonal degeneration’ was calculated as a combination of the amplitude of the compound muscle action potential from the abductor digiti minimi muscle recorded after stimulation at the wrist and the amplitude of the distal sensory response.

    We categorised nerve affections as localised to the wrist, probably localised to the elbow, definitely localised to the elbow or impossible to localise using criteria based on (1) distal latencies, (2) absolute, (3) relative motor and sensory nerve conduction velocities in the forearm and around the elbow and (4) the presence of motor nerve conduction block.32 The criteria for localisation to the elbow and for demyelination partly overlapped, which is why we chose the term localised demyelination. For patients with bilateral ulnar neuropathy, severity indices for the left side were used (results did not change when right-sided data were applied).

    Assessment of occupational biomechanical exposures

    We constructed a Job Exposure Matrix (JEM) based on experts’ ratings. Exposure estimates for each participant were obtained by combining self-reported job histories with quantitative job exposures from the JEM. Questionnaire data were collected from 7 January–30 April 2008. We asked for the main job title in the year before the index year and up to five preceding job titles. Each job title was recoded into 1 of 2227 occupational titles in the Danish version of the International Standard Classification of Occupations (D-ISCO 88). A total of 806 occupational titles occurred. We divided these occupational titles into 169 groups of jobs that were expected to have similar exposure profiles with respect to computer use and the exposures mentioned below. We then selected 15 job groups that covered a wide range of exposures. Five experienced occupational health physicians consensus rated these 15 job groups so that they could serve as benchmarks during the remaining rating process. In this way, the experts could calibrate their estimates to a common scale.34 For each of the remaining 154 job groups, the experts independently rated.

    • Forceful work in terms of the mean force exerted with the hand and arm across a full working day relative to the maximal strength of a ‘standard person’ (a healthy 35-year-old man), using a five-point scale (0=light, 1=somewhat hard, 2=hard, 3=very hard and 4=near maximal).35

    The experts also rated the mean number of hours per day (in half-hour intervals) with

    • repetitive movements of elbow or wrist (≥4 movements per minute),36 excluding computer use,

    • non-neutral postures of elbow (flexion >100°, or ≥near maximal pronation/supination) or wrist (>5° radial deviation, >10° ulnar deviation or >15° palmar/dorsal flexion)36 ,37 and

    • exposure to HAVs with accelerations ≥3 m/s2.

    The benchmarks and the mean values of the experts’ ratings were included in the JEM.

    Each participant received (1) estimates of recent exposures, that is, daily exposures in the year before the index year (force-score, repetition-time, non-neutral-posture-time and HAV-time) and (2) estimates of cumulative exposures for a 5-year period prior to the index year. Initial analyses showed that recent and cumulative exposures were highly correlated (correlation coefficients 0.93–0.95), so we only used recent exposures in the analyses.

    Other potential risk factors

    Questionnaire information was collected on height, weight, handedness, smoking, alcohol consumption, co-morbidity (diabetes, thyroid or renal disorders), fractures of the upper extremities at or below elbow level (ever: no/yes) and hand–arm intensive sports >1 h/week in the year before the index year (no, moderately intensive, highly intensive—classified by the researchers). For a period of 5 years up to the index year, questionnaire information was collected on full anaesthesia (no/yes), use of crutches (no/yes) and loss of >10 kg of weight within half a year (no/yes). Only fractures on the ENG-side of the cases were included in the analyses (same side for matched referents). BMI was calculated as weight (kg)/height (m)2. Pack-years of smoking were calculated as ‘number of years with regular smoking’בnumber of cigarette packs per day ((ie, number of cigarettes+grams of pipe tobacco+4×number of cigars)/20 cigarettes per pack)’.

    As an indicator for socioeconomic status, we used education level as judged from unemployment insurance fund membership in the year before the index year according to the Danish National Register on Public Transfer Payments. This information was categorised into higher-level or medium-level education, vocational education and training, low education level and cash benefit and missing.38

    Sample size

    According to power calculations for matched case–referent studies,39 we could show an OR of 1.5 with a probability of 80% (β=0.20) and a significance level of 0.05 (α=0.05) if we included 300 cases and three referents per case, assuming that 50% of the population was exposed. Assuming that 25% of the population was exposed, we would need 341 cases for the same power.

    Statistical methods

    Proportions were compared by Pearson χ2 test. In studies 1 and 2, we applied conditional logistic regression to analyse the risk of each outcome in relation to categorised occupational exposures, yielding ORs for each exposure category. These analyses were supplemented by tests for trend using exposure category as a continuous variable. Patients with bilateral ulnar neuropathy counted as one case. A priori, we decided not to include education level in the main analyses to avoid overadjustment due to correlations with occupational biomechanical exposures. Risk was analysed for one occupational exposure at a time: (1) unadjusted except for the matching factors, (2) partly adjusted, that is, adjusted for other potential risk factors except for education level (see the subheading ‘other potential risk factors’ above) and (3) in full models with mutual adjustment for all occupational exposures and all other potential risk factors that we assessed, except for education level. We constructed variables that combined the exposures two-by-two as well as corresponding interaction terms. Likelihood-ratio test was used to demonstrate interactions. To explore whether forceful work was related to electrophysiological severity, we applied interaction terms between force-score (0, >0) and four levels—mainly based on quartiles—of severity of localised demyelination and axonal degeneration, respectively. In these analyses, matched controls followed their index case to one of the four levels of severity. Therefore, the OR for the interaction term between force and severity represents the effect of force within each level of severity, which corresponds to analyses stratified by severity. Demonstration of a stronger effect of force in more severe cases would support the existence of an exposure–effect relation. In study 3, logistic regression was used.

    If no job title was reported for the year before the index year, but at least one previous job title was listed, the exposure estimates were set at zero. If no job titles were reported at all, the person was excluded from the analyses. Persons with missing information on other potential risk factors were allocated to the reference levels, except that persons with missing information on BMI received the mean of non-missing values for the entire study population. For studies 1 and 2, we repeated the analyses excluding participants with missing values. Analyses were carried out using STATA V.11.2.

    Results

    The proportion who returned the questionnaire was 59% (n=2513). Respondents were on average 2–3 years older than non-respondents and a little more likely to be females, table 1. Among respondents, persons with low education level and persons on cash benefit were underrepresented—this was true both for cases and for referents (table 1). A male preponderance was observed in the group with ulnar neuropathy and the opposite in the group with ulnar neuropathy-like symptoms.

    View this table:
    Table 1

    Collection of questionnaire data for studies 1 and 2 and characterisation of non-respondent and respondent cases and referents with respect to age, sex and education level

    The ratio between dominant-sided and non-dominant-sided cases was 1:1.5 for ulnar neuropathy and 1:0.8 for ulnar neuropathy-like symptoms. Ulnar neuropathy occurred more frequently on the left than on the right side, both among left-handed (27/38; 71%) and among right-handed (172/265; 65%) participants. Bilateral cases (n=10) and patients who were ambidextrous (n=9) or had missing information on handedness (n=2) were left out of these analyses. ENG with needle electrodes was performed in 80.6% of cases with ulnar neuropathy (82% of men and 78% of women) and in 50% of cases with ulnar neuropathy-like symptoms (44% of men and 53% of women); the proportions (80.6% and 50%) differed significantly (χ2=72, p=0.00). The affection of the ulnar nerve was definitely localised to the elbow in 35.5%, probably localised to the elbow in 21%, localised to the wrist in 1.2% and impossible to localise in 42.3%. Among patients with ulnar neuropathy, 5.3% were diagnosed with CTS and 29.3% were examined for this disorder. These percentages were 8.3% and 62.9% for patients with ulnar neuropathy-like symptoms. Supplementary table S2 shows characteristics of the participants. Correlation coefficients were 0.54 between force-score and non-neutral-posture-time, 0.75 between force-score and repetition-time and 0.69 between non-neutral-posture-time and repetition-time. Among participants with higher-level or medium-level education, 16.7% received a force-score ≥1; this percentage was 36.0% for vocational education and training and 63.7% for low education level.

    Table 2 presents the main results. Study 1 showed clear exposure–response relations between force-score and ulnar neuropathy. This remained the case when the full model was expanded with education level (results not shown). In unadjusted and/or partly adjusted models, associations were found between ulnar neuropathy and the highest exposure categories of repetition-time, non-neutral-posture-time and HAV-time, but these associations disappeared in the full model. Fully adjusted analyses of the variable that combined forceful work and non-neutral postures pointed to a potential interaction (results not shown); a full model analysis with inclusion of the corresponding interaction term supported this by an OR of 3.16 (95% CI 1.10 to 9.05, p=0.03) suggesting a synergistic effect of non-neutral postures and forceful work. Findings indicated no other interactions (results not shown). Study 2 showed that ulnar neuropathy-like symptoms were associated with repetition-time ≥2.5 h/day and also with non-neutral-posture-time, but without an exposure–response pattern; no association was observed with force-score in the full model. Results for repetition-time remained statistically significant when the full model was expanded with education level (results not shown). We repeated the fully adjusted analyses (1) after restriction to match groups where cases were examined with needle electrodes, (2) after restriction to cases without CTS and (3) after restriction to participants without missing covariates. Results were similar to those presented in table 2 (results not shown).

    View this table:
    Table 2

    Risk of ulnar neuropathy, study 1, and ulnar neuropathy-like symptoms, study 2, in relation to occupational biomechanical exposures

    Table 3 presents results of study 3 that compared cases from study 1 with cases from study 2 used as referents. The association between force-score and ulnar neuropathy retained its significance. This was also the case after entry of education level (results not shown). Patients with ulnar neuropathy and ulnar neuropathy-like symptoms did not differ with respect to other occupational exposures (results not shown).

    View this table:
    Table 3

    Risk of ulnar neuropathy confirmed by electroneurography (ENG) in relation to forceful work, according to analyses using patients with ulnar neuropathy-like symptoms with normal ENG as referents, study 3

    As shown in table 4, results suggested an exposure–effect relation between force-score and the severity index of localised demyelination, although the findings did not reach statistical significance. For this index, it is seen that the OR increased steeply between severity grades 1 and 2; it is also seen that the majority of cases with ulnar neuropathy had a severity grade ≥2 while the majority of cases with ulnar neuropathy-like symptoms had a severity grade ≤1. Results did not suggest an exposure–effect relation between forceful work and the severity index of axonal degeneration.

    View this table:
    Table 4

    Graded severity of localised demyelination and axonal degeneration as judged by electroneurography (ENG) of the ulnar nerve (four grades, 0–3) in relation to forceful work (force-score, dichotomised)

    Discussion

    This study compared patients with ulnar neuropathy confirmed by ENG and patients with ulnar neuropathy-like symptoms with normal ENG to two separate referent groups and to each other directly. For ulnar neuropathy, we found an exposure–response relationship with forceful work, and forceful work and non-neutral postures interacted; we observed no relation with repetitive movements. For ulnar neuropathy-like symptoms, we found associations with repetitive movements and non-neutral postures, not forceful work. The relationship between forceful work and ulnar neuropathy remained when we compared the two patient groups directly. Forceful work was related to a severity index of localised demyelination in an exposure–effect manner.

    So far, this is the largest epidemiological study focussing on the work-relatedness of ulnar neuropathy. The study benefited from a triple case–referent design that enabled us to refine our interpretation of results when compared with a single case–referent design. Hence, we were able to judge whether observed associations reflected specific effects of occupational biomechanical exposures on the ulnar nerve.

    We took care to ensure that the matched referents represented the distribution of occupational biomechanical exposures in the base populations. Non-participation could have caused inflation bias if exposed cases were more likely to respond than exposed referents. The proportion who responded did not differ between case and referent groups, but participation was related to education level that correlated with the studied exposures. Reassuringly, low education level was equally underrepresented among cases and referents.

    The main weakness of the study was that we were unable to explore if clinical differences distinguished between the two case groups. The proportion that was examined by surface electrodes was relatively high in the group with ulnar neuropathy-like symptoms, which may reflect a tendency for use of surface electrodes in less obvious cases of ulnar neuropathy. ENG signs of CTS were found in some patients in both case groups. However, all patients were clinically suspected of ulnar neuropathy by the supervising consultant according to the examination strategy of the neurophysiological department.

    We chose not to exclude patients with CTS because this could cause underestimation of associations to the extent that CTS and ulnar neuropathy or ulnar neuropathy-like symptoms shared risk factors. On the contrary, if the conditions under study and CTS tended to coexist, the results might reflect well-known associations between occupational exposures and CTS. However, results did not change after exclusion of patients with CTS. In study 3, we compared two groups of patients, who were referred on suspicion of mononeuropathy of the upper limb, and thereby minimised the possibility of referral bias.

    Case–referent studies are subject to recall bias if they rely on retrospective exposure assessment based on self-report. We used an expert-based JEM to obtain exposure estimates independently of the symptom status of the participants. Since the exposure estimates were not validated by, for example, technical measurements, it may be questioned if the job groups were ranked correctly and if they were equally homogenous with respect to all exposures assessed. If exposure variation within job groups was larger relative to between job groups for some exposure variables (eg, repetition-time) than for others (eg, force-score), the JEM-based approach would be more sensitive to effects of the last-mentioned type of variables. This could have caused the force-score to seem more important than repetition-time. Correlations between the occupational exposure variables could also make it difficult to disentangle their effects. However, we do not see how these potential errors might explain the fact that we found different profiles of associations in studies 1 and 2.

    The relationship between ulnar neuropathy (predominantly localised to the elbow) and forceful work—especially when combined with nonneutral postures—agreed with the potential pathogenetic mechanism that involves pressure on the nerve due to forceful contraction of the flexor carpi ulnaris muscle in combination with flexion of elbow.23 ,26

    In accordance with previous findings, we observed a male preponderance among patients with ulnar neuropathy,5 ,6 ,10 and left-sided cases prevailed irrespective of handedness.21 The last-mentioned finding may have to do with work practices where the non-dominant hand is more often used to exert force, for example, in order to hold objects in position while working on them with the dominant hand. Another possibility is compression of the nerve while resting the left elbow against the table during computer work.9 We asked participants if they were inclined to rest their elbows against the table, and intend to return to this issue.

    Our ENG criteria for ulnar neuropathy identified a group of patients with a specific occupational risk profile. We distinguished between normal and abnormal ENG-findings applying the cut point used in clinical practice, which seemed reasonable also judged from the relation between forceful work and graded severity of localised demyelination, where the steepest increase in OR occurred between severity grades 1 and 2—this cut point discriminated quite well between the original case groups. Moreover, the indication of an exposure–effect relationship suggested that our findings would be robust to slight alterations of the definition of abnormal.

    On the basis of the considerations above and the fact that our main findings were internally consistent and statistically significant, we think that the differences observed between the occupational risk profiles for ulnar neuropathy and ulnar neuropathy-like symptoms could not be explained by selection bias, information bias or chance. We do not think that important potential confounders were left out. This leaves a specific causal effect of exertion of force on the ulnar nerve as the most likely explanation of our findings.40 A causal interpretation was supported by the biological link suggested by the indication of an exposure–effect relation between forceful work and severity of localised demyelination. Thus, the results corroborated our hypothesis that ulnar neuropathy confirmed by ENG is a distinct neural disorder that is clearly demarcated from ulnar neuropathy-like symptoms with respect to the profile of risk factors and pathogenetic mechanisms.

    In conclusion, ulnar neuropathy confirmed by ENG and ulnar neuropathy-like symptoms with normal ENG differed with respect to associations with occupational biomechanical exposures. Ulnar neuropathy was related to forceful work in an exposure–response manner, whereas ulnar neuropathy-like symptoms were associated with repetitive movements. Thus, the results corroborated the importance of an electrophysiological diagnosis when evaluating risk factors for ulnar neuropathy. For both conditions, the results suggested that preventive effects may be achieved by reducing occupational biomechanical exposures. The results also have implications for workers’ compensations.

    What this study adds

    Studies of occupational risk factors for ulnar neuropathy are rare, and the diagnostic weight of electroneurography (ENG) has been a matter of discussion.

    • Ulnar neuropathy confirmed by ENG and ulnar neuropathy-like symptoms with normal ENG differed with respect to associations with occupational biomechanical exposures.

    • In aetiological research based on clinical case criteria for ulnar neuropathy without confirmation by ENG, specific relations to risk factors may be overlooked.

    • Findings suggested specific effects of forceful work on the ulnar nerve, which has implications for prevention and workers’ compensations.

    Acknowledgments

    We would like to express our gratitude to Johan Hviid Andersen, Jens Peder Lind Haahr and Lone Donbæk Jensen who provided expert ratings for the Job Exposure Matrix. We also wish to thank data manager Michael Victor Christensen for his preparation of data from the Danish National Register on Public Transfer Payments for analysis.

    References

    1. ↵
      1. Caliandro P,
      2. La TG,
      3. Padua R,
      4. et al
      . Treatment for ulnar neuropathy at the elbow. Cochrane Database Syst Rev 2011;(2):CD006839.
    2. ↵
      1. Ohlsson K,
      2. Attewell RG,
      3. Johnsson B,
      4. et al
      . An assessment of neck and upper extremity disorders by questionnaire and clinical examination. Ergonomics 1994;37:891–7.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Sluiter JK,
      2. Rest KM,
      3. Frings-Dresen MH
      . Criteria document for evaluating the work-relatedness of upper-extremity musculoskeletal disorders. Scand J Work Environ Health 2001;27(Suppl 1):1–102.
      OpenUrlPubMedWeb of Science
    4. ↵
      1. Roquelaure Y,
      2. Ha C,
      3. Leclerc A,
      4. et al
      . Epidemiologic surveillance of upper-extremity musculoskeletal disorders in the working population. Arthritis Rheum 2006;55:765–78.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Mondelli M,
      2. Giannini F,
      3. Ballerini M,
      4. et al
      . Incidence of ulnar neuropathy at the elbow in the province of Siena (Italy). J Neurol Sci 2005;234:5–10.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Latinovic R,
      2. Gulliford MC,
      3. Hughes RA
      . Incidence of common compressive neuropathies in primary care. J Neurol Neurosurg Psychiatry 2006;77:263–5.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. El-Salem K,
      2. Ammari F
      . Neurophysiological changes in neurologically asymptomatic hypothyroid patients: a prospective cohort study. J Clin Neurophysiol 2006;23:568–72.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Fincham RW,
      2. Cape CA
      . Neuropathy in myxedema. A study of sensory nerve conduction in the upper extremities. Arch Neurol 1968;19:464–6.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. McPherson SA,
      2. Meals RA
      . Cubital tunnel syndrome. Orthop Clin North Am 1992;23:111–23.
      OpenUrlPubMedWeb of Science
    10. ↵
      1. Richardson JK,
      2. Green DF,
      3. Jamieson SC,
      4. et al
      . Gender, body mass and age as risk factors for ulnar mononeuropathy at the elbow. Muscle Nerve 2001;24:551–4.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Richardson JK,
      2. Jamieson SC
      . Cigarette smoking and ulnar mononeuropathy at the elbow. Am J Phys Med Rehabil 2004;83:730–4.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Landau ME,
      2. Barner KC,
      3. Campbell WW
      . Effect of body mass index on ulnar nerve conduction velocity, ulnar neuropathy at the elbow, and carpal tunnel syndrome. Muscle Nerve 2005;32:360–3.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Palmer KT,
      2. Harris EC,
      3. Coggon D
      . Carpal tunnel syndrome and its relation to occupation: a systematic literature review. Occup Med (Lond) 2007 Jan;57(1): 57–66. Epub 2006 Nov 2. Review. PubMed PMID: 17082517.
    14. ↵
      1. Hansson G-Å,
      2. Balogh I,
      3. Ohlsson K,
      4. et al
      . Impact of physical exposure on neck and upper limb disorders in female workers. Appl Ergon 2000;31:301–10.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Lipscomb H,
      2. Kucera K,
      3. Epling C,
      4. et al
      . Upper extremity musculoskeletal symptoms and disorders among a cohort of women employed in poultry processing. Am J Ind Med 2008;51:24–36.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Ohlsson K,
      2. Hansson G-Å,
      3. Balogh I,
      4. et al
      . Disorders of the neck and upper limbs in women in the fish processing industry. Occup Environ Med 1994;51:826–32.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Ohlsson K,
      2. Attewell RG,
      3. Pålsson B,
      4. et al
      . Repetitive industrial work and neck and upper limb disorders in females. Am J Ind Med 1995;27:731–47.
      OpenUrlPubMedWeb of Science
    18. ↵
      1. Palmer K,
      2. Crane G,
      3. Inskip H
      . Symptoms of hand-arm vibration syndrome in gas distribution operatives. Occup Environ Med 1998;55:716–21.
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Ranney D,
      2. Wells R,
      3. Moore A
      . Upper limb musculoskeletal disorders in highly repetitive industries: precise anatomical physical findings. Ergonomics 1995;38:1408–23.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Descatha A,
      2. Leclerc A,
      3. Chastang JF,
      4. et al
      . Incidence of ulnar nerve entrapment at the elbow in repetitive work. Scand J Work Environ Health 2004;30:234–40.
      OpenUrlPubMedWeb of Science
    21. ↵
      1. Richardson JK,
      2. Ho S,
      3. Wolf J,
      4. et al
      . The nature of the relationship between smoking and ulnar neuropathy at the elbow. Am J Phys Med Rehabil 2009;88:711–18.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Bartels RH,
      2. Verbeek AL
      . Risk factors for ulnar nerve compression at the elbow: a case control study. Acta Neurochir (Wien) 2007;149:669–74.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Svendsen SW,
      2. Juhl AH
      . [Ulnar neuropathy in a poultry worker.] Ulnarisneuropati hos fjerkræslagteriarbejder. Ugeskr Laeger 2008;170:3131–2.
      OpenUrlPubMed
    24. ↵
      1. Yayama T,
      2. Kobayashi S,
      3. Nakanishi Y,
      4. et al
      . Effects of graded mechanical compression of rabbit sciatic nerve on nerve blood flow and electrophysiological properties. J Clin Neurosci 2010;17:501–5.
      OpenUrlCrossRefPubMed
    25. ↵
      1. Wright TW,
      2. Glowczewskie F Jr.,
      3. Cowin D,
      4. et al
      . Ulnar nerve excursion and strain at the elbow and wrist associated with upper extremity motion. J Hand Surg Am 2001;26:655–62.
      OpenUrlPubMed
    26. ↵
      1. Werner CO,
      2. Ohlin P,
      3. Elmqvist D
      . Pressures recorded in ulnar neuropathy. Acta Orthop Scand 1985;56:404–6.
      OpenUrlPubMedWeb of Science
    27. ↵
      1. Millesi H,
      2. Zoch G,
      3. Rath T
      . The gliding apparatus of peripheral nerve and its clinical significance. Ann Chir Main Memb Super 1990;9:87–97.
      OpenUrlCrossRefPubMed
    28. ↵
      1. Strömberg T,
      2. Dahlin LB,
      3. Brun A,
      4. et al
      . Structural nerve changes at wrist level in workers exposed to vibration. Occup Environ Med 1997;54:307–11.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Travell D,
      2. Simons DG
      . Myofascial pain and dysfunction. The trigger point manual. Volume 1. Upper half of the body. 2nd edn. Baltimore: Williams & Wilkins, 2011.
    30. ↵
      1. Lynge E,
      2. Sandegaard JL,
      3. Rebolj M
      . The Danish National Patient Register. Scand J Public Health 2011;39:30–3.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Payan J
      . Electrophysiological localization of ulnar nerve lesions. J Neurol Neurosurg Psychiatry 1969;32:208–20.
      OpenUrlFREE Full Text
    32. ↵
      1. Fuglsang-Frederiksen A,
      2. Pugdahl K
      . Current status on electrodiagnostic standards and guidelines in neuromuscular disorders. Clin Neurophysiol 2011; 122:440–55.
      OpenUrlCrossRefPubMed
    33. ↵
      1. Olney RK
      . Guidelines in electrodiagnostic medicine. Consensus criteria for the diagnosis of partial conduction block. Muscle Nerve Suppl 1999;8:S225–9.
      OpenUrlPubMed
    34. ↵
      1. Teschke K,
      2. Olshan AF,
      3. Daniels JL,
      4. et al
      . Occupational exposure assessment in case-control studies: opportunities for improvement. Occup Environ Med 2002;59:575–93.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Moore JS,
      2. Garg A
      . The Strain Index: a proposed method to analyze jobs for risk of distal upper extremity disorders. Am Ind Hyg Assoc J 1995;56:443–58.
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      1. McAtamney L,
      2. Nigel CE
      . RULA: a survey method for the investigation of work-related upper limb disorders. Appl Ergon 1993;24:91–9.
      OpenUrlCrossRefPubMedWeb of Science
    37. ↵
      1. Thomsen JF,
      2. Mikkelsen S,
      3. Andersen JH,
      4. et al
      . Risk factors for hand-wrist disorders in repetitive work. Occup Environ Med 2007;64:527–33.
      OpenUrlAbstract/FREE Full Text
    38. ↵
      1. Svendsen SW,
      2. Frost P,
      3. Jensen LD
      . Time trends in surgery for non-traumatic shoulder disorders and postoperative risk of permanent work disability: a nationwide cohort study. Scand J Rheumatol 2012;41:59–65.
      OpenUrlCrossRefPubMed
    39. ↵
      1. Dupont WD
      . Power calculations for matched case-control studies. Biometrics 1988;44:1157–68.
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Miettinen OS
      . Etiologic research: needed revisions of concepts and principles. Scand J Work Environ Health 1999;25:484–90.
      OpenUrlPubMedWeb of Science

    Footnotes

    • Contributors SWS and PF conceptualised and designed the study and analysed the data. All authors contributed to data collection and interpretation of data. SWS drafted the article and all coauthors revised it critically for important intellectual content. All authors have approved the version submitted for publication.

    • Funding The project was funded by the Danish Working Environment Research Fund (grant no. 13-2007-03/20070014773).

    • Competing interests None.

    • Ethics approval The Danish Data Protection Agency, the five Danish administrative regions and the Danish National Board of Health.

    • Provenance and peer review Not commissioned; externally peer reviewed.