Exposed group.

Workers are considered eligible for study participation if they meet the following inclusion criteria:

1. i. Aged between 18 and 67 years.
2. ii. Continuous employment in the current type of work for at least one year at the time of inclusion.
3. iii. Workplace where physical loads on hands and arms exceed the Action Limit of the Threshold Limit Values for Hand Activity Level [TLV for HAL; 19, 20].
4. iv. A daily repetitive strain on one or both hands on ≥ 4 days a week for a total of ≥ 20 hours a week.
5. v. Suffer from varying degrees of musculoskeletal pain in the elbow/forearm/hand-area (including no musculoskeletal pain) as assessed using the Cornell Musculoskeletal Discomfort Questionnaire [CMDQ; 21].
6. vi. Written informed consent to study participation and data protection.

Workers are considered ineligible for study participation if they meet one or more of the following exclusion criteria:

1. i. Suffering from musculoskeletal pain in body regions other than the elbow/forearm/hand-area, which is equal to or exceeds the severity reported for the elbow/forearm/hand-area.
2. ii. Suffering from chronic inflammatory conditions (e.g., rheumatism or lupus) or metabolic diseases (e.g., metabolic syndrome or hypercholesterinaemia), since these conditions are known to increase inflammatory markers systemically.
3. iii. Taking regular medications known to interact with inflammatory biomarkers [e.g., statins, ibuprofen; 22].
4. iv. Status after acute trauma that affects joints or muscles and/or is associated with acute inflammation.
5. v. Relative (e.g., fear of blood) and absolute (e.g., taking anticoagulants, few suitable veins that should only be used for therapeutic interventions) contraindications for repeated blood sampling.

Note that an occupational physician interpreted the differential blood analyses in the screening process – which apply to some of the above-mentioned eligibility criteria – and advised on study enrollment of the screened persons.

Unexposed group.

For workers exposed to lower physical loads, the same inclusion criteria apply as for the exposed group, with the exception of: iii) office workplaces should not exceed the TLV for HAL; v) not suffering from any musculoskeletal pain. The exclusion criteria for workers in the unexposed group are also similar to those of the exposed group, with the exception of: i) not suffering from musculoskeletal pain at all.

All study participants will receive an expense allowance of EUR 500 for complete participation and partial compensation for incomplete participation.

Physical workload assessment.

The workload on the upper limbs is assessed by using two different methods for the exposed group and the unexposed group. The TLV for HAL [19,20] is used at workplaces of participants in the exposed group, assessing the physical load on hands and arms. Only workplaces where this physical load exceeds the Action Limit of the TLV for HAL are considered for the study to ensure that there is a clear discrepancy with the unexposed group. The TLV for HAL showed fair (Cohen κ of 0.34) to strong (Spearman ρ of 0.65) inter-rater reliability [27].

Medical anamnesis, eligibility criteria and demographic data.

Taking the medical history follows a standardized written interview guide covering health status, general illnesses, suffering from musculoskeletal pain in body regions other than the elbow/forearm/hand-area, suffering from chronic inflammatory or metabolic diseases, acute trauma affecting joints or muscles, taking regular medications that interact with inflammatory biomarkers (e.g., statins, ibuprofen by case discussion) [22], and contraindications for repeated blood sampling (see also exclusion criteria). The specified medical examination is performed according to the guidelines for physicians as provided and developed by the German Social Accident Insurance [28]. The neutral-zero-method [23] is a known method among physicians to assess joint mobility. During the nine repeated measurements, the joint mobility of the hand-arm system only will be assessed using the neutral-zero-method. Specifically, the maximal active range of motion of the following joints is measured using a goniometer: elbow flexion and extension, forearm pronation and supination, wrist dorsal and palmar flexion, and wrist ulnar and radial abduction.

Eligibility criteria are checked by the physician. The demographic data of participants are collected, including: body weight, length, age, gender, leisure time activity level (i.e., sport in the past four weeks in h/week), smoking status, handedness and nationality. Both checklists are conducted in face-to-face interview format, where the physician reads the questions aloud and notes or ticks the answers.

Musculoskeletal pain.

The German CMDQ is used to map in which body areas (main) musculoskeletal pain is perceived. This questionnaire showed a good validity (Cohen κ 0.38–1.00; Spearman ρ 0.40–1.00), a satisfactory internal consistency (Cronbach α 0.75–0.82) and medium to substantial test-retest reliability (Spearman ρ 0.56–0.72) [21]. The CMDQ is conducted in a face-to-face interview format, where the researcher reads the questions aloud and notes or ticks the answers.

The 11-point NRS is used to rate perceived musculoskeletal pain (i.e., RPP) in the elbow-forearm-hand region. The 11-point NRS handles a categorical answer format with endpoints 0 (no pain) and 10 (strongest imaginable pain) and is characterized by a low error rate, high acceptance, easy handling and high sensitivity [29,30]. The NRS-11 has good to excellent validity (Pearson r 0.94 with the 100-mm visual analogue scale) and excellent reliability (ICC 0.95) when assessing musculoskeletal pain [26]. The RPP is collected at four time-points during each of the nine measurements. The first RPP (i.e., participant's current RPP) is collected immediately before pain provocation (i.e., pressure pain thresholds and functionally triggered pain). On the Friday or Saturday of the work week, the other three values are collected reflecting the past work week (i.e., minimum RPP, maximum RPP and RPP by night), for which the participant is called by one of the researchers involved in the project.

Blood sampling during screening.

By the end of the screening, blood is collected for performing differential blood and laboratory chemical analyses to verify that participants are free from acute infection or metabolic diseases. Three blood samples are collected through a single venous puncture (Safety-Multifly® needle, 21G x 3/4”, green, 120 mm tube length; Sarstedt AG & Co. KG, Nümbrecht, Germany) preferably from the median cubital vein at the bend of the participant’s preferred arm: lithium-heparin tube (S-Monovette® LH, 5.5 ml, cap orange, 75 x 15 mm; Sarstedt AG & Co. KG, Nümbrecht, Germany), EDTA-K3E tube (S-Monovette® EDTA K3E, 2.6 ml, cap red, 65 x 13 mm; Sarstedt AG & Co. KG, Nümbrecht, Germany), Fluoride/EDTA tube (S-Monovette® FE, 2.7 ml, cap yellow, 75 x 13 mm; Sarstedt AG & Co. KG, Nümbrecht, Germany). All other standard operating procedures for hygienic blood sampling are adhered to. Various metabolic and inflammatory blood parameters are gathered, including: gamma-glutamyl-transferase (GGT), HbA1c (HPLC and glycated haemoglobin), high- and low-density lipoprotein cholesterol (i.e., HDL and LDL, respectively), triglyceride, enzymatic creatinine and large blood count including thrombocytes. The blood values are weighted against standard values and based on this a decision is made to include or exclude the screened person.

Pressure pain thresholds.

Pain is provoked using a digital pressure algometer (CE-certificated; ATORN ZD2) on five bilateral anatomical sites of interest to map so-called pressure pain thresholds (PPT), a methodology also used in clinical practice [31]. The PPT procedure showed to have a very high intra-rater reliability [Cronbach α 0.94–0.98; 24]. The five sites are selected based on an extensive literature search, including: (1) carpal tunnel [32], (2) lateral and (3) medial epicondyles [33], (4) extensor carpi radialis brevis and (5) flexor carpi radialis [34]. The measuring tip (1 cm²) of the algometer is placed at a 90-degree-angle to the skin’s surface – for exact locations, see Table 1 – while gradually applying increased pressure by ~3 N/s until the participant experiences the slightest pain sensation. Without seeing the pressure value, the participant is instructed to express pain by either saying it or raising their hand of the non-involved arm when the first slight pain sensation is felt. This procedure is repeated on each bilateral anatomical site for three times, after which the average value (N) for each bilateral anatomical site will be used for statistical analyses.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Location of the algometer’s 1–cm² measurement tip to determine PPT at the five bilateral anatomical sites.

https://doi.org/10.1371/journal.pone.0334244.t001

Functionally triggered pain.

Functionally triggered pain is examined by adopting the protocol of Garg, Hegmann [25], including three examinations. First, the participant executes maximal force during closing of the fist until a pain trigger follows (if any). Second, spontaneous pain is assessed (if any) when moving the hand against resistance in the six directions supination, pronation, lateral flexion radial, lateral flexion ulnar, palmar flexion, and dorsal extension. Third, spontaneous pain is assessed (if any) when moving the forearm against resistant in the two directions flexion and extension. The researcher provides sufficient resistance for the participant to develop maximum force for every movement, except for closing the fist. The researcher notes whether the participant experienced pain (yes) or not (no).

Blood sampling during repeated measures.

Blood samples are collected through a single venous puncture preferably from the median cubital vein at the elbow bend of the participant’s preferred arm. All other standard operating procedures for hygienic blood sampling are adhered to. Three different procedures are followed to treat the different blood samples. All laboratory analyses are performed blinded with regard to the group (exposed vs. non-exposed) the study participant’s blood belongs to. First, a lithium-heparin tube (S-Monovette® LH, 5.5 ml, cap orange, 75 x 15 mm; Sarstedt AG & Co. KG, Nümbrecht, Germany) is filled and stored at room temperature, after which it is analysed by enzymatic-photometric analysis by the Central Laboratory of the University Hospital of Tübingen (Hoppe Seyler Straße 3, 72076 Tübingen, Germany) to determine CRP and IL-6 values.

Second, TruCulture-Tubes (HOT Screen GmbH, Aspenhaustraße 25, 72770 Reutlingen, Germany) are prepared (S-Monovette® neutral Z, 2.7 ml, cap white, 66 x 11 mm; Sarstedt AG & Co. KG, Nümbrecht, Germany) before study start in batch with a stimulating environment (i.e., for inducing the production of cytokines, chemokines and growth factors) in a volume of 2 ml buffered media, and maintained (i.e., stored) at −20°C at the Institute of Occupational and Social Medicine and Health Services Research of the University Hospital of Tübingen (Wilhelmstraße 27, 72074 Tübingen, Germany) until time of use. The stimulant that was used, included LPS + TStim; LPS is a lipopolysaccharide from the bacterial endotoxin Escherichia coli (E. coli, O55:B5), and TStim is an antibody-based mimic of a superantigen initiating a T cell response by cross linking the T cell receptor and major histocompatibility complex (MHC) on monocytes/ macrophages. The exact time-point (hh:mm) of blood sample collection is noted on a label specifically designed for this study that also contains information on participant’s sex, age, date and measurement repetition. Directly after labelling, the TruCulture-Tubes (non-stimulated and stimulated) are filled with 1 ml of whole blood, placed in a dry metal block thermostat (i.e., with heating trays; Type V.69107.61220; VLM GmbH, Bielefeld, Germany pre-heated to 37.0°C) and incubated for exactly 24 h. At the end of the 24-hour incubation period, both the un-stimulated and stimulated tubes are opened to insert a Seraplas® valve filter (white, for separation of the TruCulture supernatants; Sarstedt AG & Co. KG, Nümbrecht, Germany) from the top until 1–2 mm above the sedimented cells to separate them from the different cellular layers of the supernatant and stop the stimulation reaction. With the valve installed, the tubes will be frozen at –20°C until final analysis is scheduled after data collection is completed. An exploratory analysis of the un-stimulated and stimulated TruCulture tubes is performed by HOT Screen GmbH applying Luminex® xMAP® technology (DiaSorin S.p.A., Saluggia, Italy). Since we currently do not have information available regarding the direction in which the target parameters might change, the goal of these exploratory analyses will be to identify patterns related to pathogenesis.

Third, an EDTA-K3 tube (S-Monovette® EDTA K3, 9.0 ml, cap red, 92 x 16 mm; Sarstedt AG & Co. KG, Nümbrecht, Germany) is filled and placed in a water-ice bath to cool to 3–4°C. Within two hours after blood sampling, the tube is centrifuged at 1500G at 4°C for 10 min. Plasma is stored at –70°C in aliquots at the laboratory of the Department of Sports Medicine of the University Hospital of Tübingen (Hoppe-Seylerstraße 3, 72076 Tübingen) until the final analysis is scheduled after the study’s data collection is completed. The thawed aliquots with plasma will be analysed for CRP, TNF-α, IL-6, and IL-1β using commercial enzyme-linked immunosorbent assay (ELISA) kits from eBioscience (Thermo Fisher Scientific, San Diego, CA, USA) according to the manufacturer’s instructions by the same Department of Sports Medicine. Residual blood plasma will be kept stored at –70°C in aliquots at the laboratory of the Department of Sports Medicine of the University Hospital of Tübingen (Hoppe-Seylerstraße 3, 72076 Tübingen) for possible future use such as the exploratory analysis of further molecular markers, such as miRNA [35].

The blood analyses performed by HOT Screen GmbH (Reutlingen, Germany) and the Department of Sports Medicine of the University Hospital of Tübingen (Germany) will be performed blinded, i.e., the persons performing the laboratory analyses are blinded with regard to the group to which the study participants belong to (exposed vs. unexposed) and with regard to the repeated measurements. The blood analyses performed by the Central Laboratory of the University Hospital of Tübingen (Germany) will be performed by a group of people who are only blinded to results of previously analysed samples.

Primary objective.

For assessing the reliability of systemic biomarkers of inflammation (i.e., CRP, TNF-α, IL-6, IL-1β) over time, we will create Bland-Altman plots for providing visual representations of the difference between two repeated measures and the average of two repeated measures for both the exposed and unexposed group. The Bland-Altman plots will be used to identify any systematic bias in the data. For assessing absolute test-retest reliability, we will apply a two-way mixed-effects model with a single rater for both consistency and agreement to calculate the intra-class correlation coefficient for the exposed and unexposed group [ICC; 36, 37]. For assessing relative test-retest reliability, we will use the standard error of measurement [SEM; 37].

We will also exploratory build different ICCs to obtain information about how many repeated measurements result in a stable ICC. In other words, from how many repeated measurements the ICC is above an acceptable limit.

Secondary objectives.

We will use the same procedures for determining reliability as described above, only with different parameters as repeated measurements, i.e., rating of perceived pain, pressure pain thresholds and functionally triggered pain.

For assessing the correlation between single systemic biomarkers of inflammation and rating of perceived pain, we will use the Spearman ρ correlation coefficient. For assessing the correlation between single systemic biomarkers of inflammation and pressure pain thresholds, we will use the Pearson r correlation coefficient. For assessing the correlation between single systemic biomarkers of inflammation and functionally triggered pain thresholds, we will use the point-biserial correlation coefficient. For assessing the correlation between the systemic biomarkers of inflammation IL-6 assessed by different laboratory procedures (i.e., enzymatic-photometric analysis versus ELISA), we will use the Pearson r correlation coefficient.

For comparing the (a) level of systemic biomarkers of inflammation, (b) rating of perceived pain, (c) pressure pain thresholds and (d) functionally triggered pain between the two groups of higher and lower physical workload, we will use mixed model analyses.

Missing values.

In case of missing values, we will analyse only available data, which means that we include the number of repetitions corresponding to the subjects with the smallest _ lowest number of repetitions (i.e., per definition at least six repeated measures; see Outcomes), i.e., ignoring missing data. However, ignoring missing values may create a biased estimate, whereas imputing missing values will produce a more unbiased result [38]. Thus, in addition to complete data analysis, we will impute missing data on all data (i.e., biomarkers, pain intensity, pain thresholds, triggered pain) using replacement values according to the best fitting imputation method [39].