https://orcid.org/0000-0002-4510-3324
ABSTRACT
The aim of this study was to investigate whether there is evidence of bilateral upper limb strength deficits in individuals with unilateral lateral elbow tendinopathy (LET).
The electronic databases Medline via Ovid, PubMed, and Scopus were searched from inception to March 2020. Included studies encompassed maximal strength outcomes of any upper limb and appendicular musculature in individuals with LET and an asymptomatic comparator. Study quality was rated using a modified version of the Epidemiological Appraisal Instrument. Hedges g effect sizes (ES) and 95% CIs were calculated for comparisons of maximal strength in the LET group and an asymptomatic control group. Meta-analysis using a random-effects model was performed when possible.
Fourteen studies were included. Quality appraisal resulted in a mean Epidemiological Appraisal Instrument score of 46% (SD = 10%). Meta-analysis revealed strength deficits in shoulder abduction (pooled ES = −0.37 [95% CI = −0.62 to −0.12]) and shoulder external rotation (pooled ES = −0.55 [95% CI = −0.83 to −0.28]) of the symptomatic limb compared with an asymptomatic control group. Meta-analysis also revealed maximal strength deficits in the upper trapezius (pooled ES = −0.26 [95% CI = −0.49 to −0.02]) of the asymptomatic limb compared with an asymptomatic control group. There was also consistent evidence for strength deficits in the serratus anterior, lower trapezius, and wrist extensor muscles and deficits in grip strength of the symptomatic limb as well as strength deficits in the wrist extensor muscles of the asymptomatic limb in individuals with unilateral LET.
In individuals with LET, there were maximal strength deficits in shoulder abduction, shoulder external rotation, serratus anterior and lower trapezius muscles, and wrist extension, as well as deficits in grip strength of the symptomatic limb compared with an asymptomatic control group. In addition, there appeared to be strength deficits in the upper trapezius muscle, wrist extension, and metacarpophalangeal joint flexion and extension, as well as deficits in grip strength of the asymptomatic limb in individuals with LET compared with an asymptomatic control group. These results suggest bilateral strength deficits.
These findings highlight the importance of a thorough physical examination and appropriate strengthening intervention for the upper limb with a focus on shoulder and scapular stabilizers, in addition to forearm muscles, in individuals with LET.
In people with tennis elbow, widespread strength deficits, including weakness of the shoulder, forearm, and wrist muscles, may exist. Interestingly, some of these weaknesses appear on both the affected and the unaffected sides in people with tennis elbow. A physical therapist can help strengthen these areas.
Introduction
Lateral elbow tendinopathy (LET) affects the common extensor tendon of the forearm1–4 and most commonly occurs in the dominant arm of individuals aged between 35 and 54 years who engage in manual labor.5,6 Despite a simple clinical diagnosis, the pathophysiology of LET is complex.2 Symptoms of LET can require between 6 months and 2 years to resolve,6,7 with recurrence rates as high as 62%.8,9 Prolonged rehabilitation time with high recurrence rates reflects the complex pathophysiology of LET,10 which may include previously unrecognized motor system impairments of muscles proximal and distal to the site of pathology.11
Most studies of LET have focused on motor system impairments during gripping12,13 and wrist extension.12,14 Although local strength deficits are relevant for prognosis and treatment,10 such isolated measures may not provide a complete assessment of the upper limb and could contribute to incomplete rehabilitation. Given that the upper limb is a linked system with muscular forces generated proximally shown to influence the kinematics and kinetics of the distal segment(s),15–17 there is the potential that proximal muscle dysfunction may contribute to the development of distal overuse conditions, such as LET. The primary aim of this systematic review was therefore to investigate whether maximal strength deficits exist in the upper limbs of individuals with unilateral LET compared with an asymptomatic control group or a contralateral limb.
Methods
This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines18 and was prospectively registered on PROSPERO (reference number: CRD42020189799).
Data Sources and Searches
Three electronic databases (Medline via Ovid, PubMed, and Scopus) were searched to identify all English-language studies published prior to March 2020. The search strategy encompassed the keywords “muscular strength OR motor system” AND “lateral epicondyl*” OR “tennis elbow OR tend*” (ie, where the asterisk is used to identify all words starting with the prefix). To identify grey literature (eg, theses or book chapters) or studies that may have been missed from the database search, an additional grey literature search was undertaken using any keywords within the Google search engine. Two reviewers (N.B. and T.P.) independently screened the titles and abstracts and then reviewed the eligible full-text versions using an a priori inclusion criterion, with discrepancies being mediated by a third party (L.J.H.). Manual reference list screening was used to identify additional studies.
Study Selection
Studies were included if they met the following 3 criteria: individuals with a clinical diagnosis of LET; any measure of maximal strength of any upper limb, appendicular, or cervical/thoracic musculature; and comparison with an asymptomatic control group and/or an asymptomatic contralateral limb. Reviews, case studies, abstracts only, and letters to the editor were excluded.
Data Extraction and Quality Appraisal
Two reviewers (B.D. and N.B.) completed data extraction, with all questions being resolved by a third party (L.J.H.). Descriptive statistics for maximal strength were extracted for the symptomatic limb (and, when possible, for the asymptomatic limb) of individuals with LET and for the dominant limb (and, when possible, for the nondominant limb) of individuals in an asymptomatic control group. We also extracted data on participant characteristics and key aspects of the maximal strength testing protocols.
The quality of included studies was assessed by 2 independent reviewers (B.D. and N.B.) using the Epidemiology Appraisal Instrument (EAI),19 with discrepancies being mediated by a third party (L.J.H.) (see Suppl. Appendix A for questions of the EAI). The EAI was modified to include only the 35 items deemed relevant. Items removed related to randomization/allocation of participants, concealment, and follow-up. One study was critiqued as a group to ensure appropriate interpretation of the EAI items. Each item was scored as follows: ``yes” (score = 1), ``partial” (score = 0.5), ``no” (score = 0), or ``unable to determine” (score = 0); items deemed ``not applicable” were removed. An overall percentage was calculated by dividing the sum of scores by the number of questions and multiplying the result by 100, with a high score indicating better methodological quality. Kappa (κ) statistics calculated using SPSS V.26 software (SPSS Inc., Chicago, IL, USA) were used to report the interrater agreement, which was considered poor (<0.00), slight (0.00–0.2), fair (0.21–0.4), moderate (0.41–0.6), substantial (0.61–0.8), or almost perfect (0.81–1.0).20
Data Synthesis and Analysis
We computed the Hedges g effect size (ES) and corresponding 95% CIs as a summary measure suitable for independent and paired-group study designs,21,22 the latter of which uses a conservative correlation value (r) of .7 to compute the ES.23 A negative ES indicated less strength of the symptomatic limb than of an asymptomatic control group or asymptomatic contralateral limb. The magnitude of the ES was interpreted as small (<0.5), medium (0.5–0.8), or large (>0.8).24
To investigate strength differences between individuals with LET and individuals in an asymptomatic control group, strength measures of the symptomatic limb were compared with the dominant arm of an asymptomatic control group and/or the asymptomatic contralateral limb. To investigate bilateral strength differences in individuals with unilateral LET, strength measures of the asymptomatic contralateral limb were compared with the nondominant limb of an asymptomatic control group. When possible, the included studies were pooled for meta-analysis using a random-effects model. Meta-analysis was conducted according to the muscle(s) or the movement. Only studies including an asymptomatic control group were included in the meta-analysis. The percentage change for all comparisons were calculated and included in each forest plot.
Heterogeneity between studies was assessed using the I2 statistic, which summarizes the percentage of total variation across studies due to differences between studies rather than chance.25 An I2 ≤ 30% indicates low heterogeneity; values >30% and 50% are indicative of moderate heterogeneity and substantial heterogeneity, respectively.25 When heterogeneity was considered substantial (ie, ≥50%), an a priori decision was made not to pool the data. In these situations, the results are reported using Hedges g (± 95% CIs). For all meta-analyses and forest plots, the Excel (version 2016; Microsoft Corp., Redmond, WA, USA) spreadsheets of Neyeloff et al26 were used.
Results
Search Strategy
A total of 5779 studies were identified via the electronic database and grey literature searches (Fig. 1). Following removal of duplicates and screening of titles and abstracts, 61 studies were selected for full-text review. Five studies were added from the manual reference list search. Of these 66 studies, 18 met the inclusion criteria. Three authors were contacted for additional data14,27,28 and clarification of their methods.27 Only 1 author replied14; thus, 2 studies were excluded because of the lack of available information. Two studies were also excluded because of replication of data from a previous study.29,30 Fourteen studies were included in the final analysis (Fig. 1).
Characteristics of Included Studiesa
| Study | Population | Movement(s) and Muscle(s) Investigated | Equipment | Position(s) Tested | Contraction Type | EAI Score (%) |
|---|---|---|---|---|---|---|
| Alizadehkhaiyat et al12 | LET: 16 (sex: 8 M; age: 49 [range = 40–66] y; BMI: 25.7; DoS: NR; outcomes: NR) Con: 16 (sex: 9 M; age: 40 [range: 26–59] y; BMI: 23.9; outcomes: NR) | Shoulder abduction, external/internal rotation Wrist flexion/extension Metacarpophalangeal flexion/extension Grip strength | Handheld dynamometer Handgrip dynamometer | Shoulder abduction: shoulder abducted (30°) in scapular plane, elbow flexed (90°) Shoulder external/internal rotation: shoulder abducted (0°–30°) internally and externally, respectively, elbow flexed (90°) Wrist flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral | NR (isometric) | 33 |
| Bhalara and Sheth38 | LET: 66 (sex: 26 M; age: 42.4 [SD = 8.3] y; BMI: 25.6; DoS: NR; outcomes: NR) Con: 66 (sex: 25 M; age: 45.0 [SD = 10.8] y; BMI: 26.6; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior | Handheld dynamometer | NR | NR (isometric) | 43 |
| Bhargava et al32 | LET: 8 athletes (sex: 8 M; age: 34.8 [SD = 7.3] y; BMI: NR; DoS: 8.4 [SD = 9.5] mo; outcomes: NR) (contralateral comparison) LET: 22 nonathletes (sex: 5 M; age: 40.5 [SD = 6.8] y; BMI: NR; DoS: 7.6 [SD = 18.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral, wrist fixed at either 15° or 35° by external brace | NR (isometric) | 47 |
| Bisset et al34 | LET: 40 (sex: 24 M; age: 49.5 [range: 32–66] y; BMI: NR; DoS: 7.7 [SD = 10] mo; outcomes: NR) Con: 40 (sex: 24 M; age: 48.4 [range: 33–64] y; BMI: NR; outcomes: NR) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder flexed (90°), elbow extended (0°), forearm pronated | NR (isometric) | 43 |
| Blanchette and Normand33 | LET: 24 (sex: 11 M; age: 46 [SD = 10] y; BMI: NR; DoS: 29 [SD = 38] mo; outcomes: PRTEE score = 35 [SD = 19]) (contralateral comparison) | Wrist flexion/extension | Load cell | Wrist flexion: seated, elbow flexed (90°), forearm supinated Wrist extension: seated, elbow flexed (90°), forearm pronated | Isometric | 51 |
| Calder et al13 | LET: 11 (sex: 6 M; age: 46.6 [SD = 10.7] y; BMI: 26.6; DoS: NR; outcomes: DASH disability score = 32.39 [SD = 15.1]) At risk: 8 (sex: 2 M; age: 44.8 [SD = 13.5] y; BMI: 23.4; outcomes: DASH disability score = 0.66 [SD = 1.3]) Con: 37 (sex: 15 M; age: 27.1 [SD = 5.0] y; BMI: 23.0; outcomes: DASH disability score = NR | Wrist extension Grip strength | Load cell | Wrist extension: specifics NR Picture: shoulder abducted (~45°), elbow flexed (90°), forearm pronated, wrist neutral Grip: elbow flexed (90°), wrist neutral | NR (isometric) | 43 |
| Coombes et al31 | LET: 150 (sex: 94 M; age: 49.7 [SD = 8.2] y; BMI: 26.8 [SD = 26.8]; DoS: 5.9 [SD = 7.2] mo; outcomes: PRTEE score = 37.9 [SD = 12.8]) Con: 54 (sex: 34 M; age: 50.1 [SD = 10.1] y; BMI: 25.6 [SD = 3.6]; outcomes: NR) | Elbow flexion/extension | Handheld dynamometer | Elbow flexion/extension: standing, elbow flexed (90°), forearm neutral | Isometric | 59 |
| Day et al42 | LET: 28 (sex: 13 M; age: 46.8 [SD = 8.8] y; BMI: 29.0; DoS: NR; outcomes: QuickDASH score = 40.6% [SD = 16.3%]) Con: 28 (sex: 13 M; age: 46.1 [SD = 9.2] y; BMI: 25.1; outcomes: QuickDASH score = 2.6% [SD = 3.5%]) | Appendicular muscles: middle and lower trapezius, serratus anterior | Handheld dynamometer | Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) | NR (isometric) | 56 |
| De Smet and Fabry40 | LET: 55 (sex: 30 M; age: 45 [SD = 9.6] y; BMI: NR; DoS: NR; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: elbow flexed and elbow extended, specifics NR | NR (isometric) | 30 |
| Lucado et al35 | LET: 21 (sex: 21 F; age: 44.9 [SD = 5.2] y; BMI: 23.7 [SD = 3.0]; DoS: 6.1 [range: 1–24] mo; outcomes: NR) Con (nonplayers): 21 (sex: 21 F; age: 43.0 [SD = 8.4] y; BMI: 23.2 [SD = 3.6]; outcomes: NR) Con (tennis players): 21 (sex: 21 F; age: 46.8 [SD = 9.9] y; BMI: 23.9 [SD = 2.9]; outcomes: NR) | Appendicular muscles: upper and lower trapezius Shoulder abduction, external/internal rotation Elbow flexion/extension Wrist flexion/extension | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral, elbow at 90° Lower trapezius: prone, specifics NR Picture: shoulder abducted (~130°), elbow extended Shoulder abduction: seated, shoulder neutral Shoulder external/internal rotation: seated, shoulder abducted (30°) in scapular plane, elbow flexed (90°) Elbow flexion/extension: supine, shoulder neutral, elbow flexed (90°), forearm and wrist neutral Wrist flexion/extension: seated, shoulder abducted (20°), elbow flexed (90°), forearm neutral | Isometric | 47 |
| Nabil et al39 | LET: 10 (sex: 10 M; age: 25.2 [SD = 2.9] y; BMI: 24.4; DoS: NR; outcomes: NR) Con: 10 (sex: 10 M; age: 24.6 [SD = 2.4] y; BMI: 25.7; outcomes: NR) | Shoulder abduction, external rotation | Biodex – isokentic dynamometer | Shoulder abduction: seated, shoulder abducted (120°), elbow extended (range: 120° to neutral, total: 120°) Shoulder external rotation: seated, shoulder abducted (60°–75°), elbow flexed (90°) (range: 70° internal to 70° external, total: 140°) | Eccentric: 60°/s and 120°/s | 44 |
| Stratford et al36 | LET: 35 (sex: 17 M; age: 44.5 [SD = 8.6] y; BMI: NR; DoS: 3 [SD = 3.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder neutral, elbow extended | NR (isometric) | 44 |
| Ucurum et al37 | LET: 51 (sex: 12 M; age: 44.9 [SD = 9.7] y; BMI: 27.5 [SD = 7.4]; DoS: 12 [SD = 13.7] mo; outcomes: PRTEE score = 54.9 [SD = 18.2]) Con: 51 (sex: 9 M; age: 42.7 [SD = 9.7] y; BMI: 25.4 [SD = 4.2]; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior Shoulder abduction, external/internal rotation Grip strength | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) Shoulder abduction: seated, shoulder abducted (90°) in scapular plane Shoulder external/internal rotation: seated, shoulder neutral, elbow flexed (90°) Grip: seated, shoulder neutral, elbow flexed | NR (isometric) | 67 |
| Vigouroux et al14 | LET: 6 (sex: 6 M; age: 27.5 [SD = 8.4] y, BMI: 22.2; DoS: NR; outcomes: NR) Con (nonplayers): 10 (sex: 10 M; age: 26.8 [SD = 3.6] y; BMI: 21.8; outcomes: NR) Con (tennis players): 20 (sex: 20 M; age: 22.3 [SD = 9.5] y; BMI: 21.4; outcomes: NR) | Wrist flexion/extension Metacarpophalangeal flexion/extension | Custom-built apparatus | Wrist and metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow flexed (90°), forearm and wrist neutral | Isometric | 40 |
| Study | Population | Movement(s) and Muscle(s) Investigated | Equipment | Position(s) Tested | Contraction Type | EAI Score (%) |
|---|---|---|---|---|---|---|
| Alizadehkhaiyat et al12 | LET: 16 (sex: 8 M; age: 49 [range = 40–66] y; BMI: 25.7; DoS: NR; outcomes: NR) Con: 16 (sex: 9 M; age: 40 [range: 26–59] y; BMI: 23.9; outcomes: NR) | Shoulder abduction, external/internal rotation Wrist flexion/extension Metacarpophalangeal flexion/extension Grip strength | Handheld dynamometer Handgrip dynamometer | Shoulder abduction: shoulder abducted (30°) in scapular plane, elbow flexed (90°) Shoulder external/internal rotation: shoulder abducted (0°–30°) internally and externally, respectively, elbow flexed (90°) Wrist flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral | NR (isometric) | 33 |
| Bhalara and Sheth38 | LET: 66 (sex: 26 M; age: 42.4 [SD = 8.3] y; BMI: 25.6; DoS: NR; outcomes: NR) Con: 66 (sex: 25 M; age: 45.0 [SD = 10.8] y; BMI: 26.6; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior | Handheld dynamometer | NR | NR (isometric) | 43 |
| Bhargava et al32 | LET: 8 athletes (sex: 8 M; age: 34.8 [SD = 7.3] y; BMI: NR; DoS: 8.4 [SD = 9.5] mo; outcomes: NR) (contralateral comparison) LET: 22 nonathletes (sex: 5 M; age: 40.5 [SD = 6.8] y; BMI: NR; DoS: 7.6 [SD = 18.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral, wrist fixed at either 15° or 35° by external brace | NR (isometric) | 47 |
| Bisset et al34 | LET: 40 (sex: 24 M; age: 49.5 [range: 32–66] y; BMI: NR; DoS: 7.7 [SD = 10] mo; outcomes: NR) Con: 40 (sex: 24 M; age: 48.4 [range: 33–64] y; BMI: NR; outcomes: NR) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder flexed (90°), elbow extended (0°), forearm pronated | NR (isometric) | 43 |
| Blanchette and Normand33 | LET: 24 (sex: 11 M; age: 46 [SD = 10] y; BMI: NR; DoS: 29 [SD = 38] mo; outcomes: PRTEE score = 35 [SD = 19]) (contralateral comparison) | Wrist flexion/extension | Load cell | Wrist flexion: seated, elbow flexed (90°), forearm supinated Wrist extension: seated, elbow flexed (90°), forearm pronated | Isometric | 51 |
| Calder et al13 | LET: 11 (sex: 6 M; age: 46.6 [SD = 10.7] y; BMI: 26.6; DoS: NR; outcomes: DASH disability score = 32.39 [SD = 15.1]) At risk: 8 (sex: 2 M; age: 44.8 [SD = 13.5] y; BMI: 23.4; outcomes: DASH disability score = 0.66 [SD = 1.3]) Con: 37 (sex: 15 M; age: 27.1 [SD = 5.0] y; BMI: 23.0; outcomes: DASH disability score = NR | Wrist extension Grip strength | Load cell | Wrist extension: specifics NR Picture: shoulder abducted (~45°), elbow flexed (90°), forearm pronated, wrist neutral Grip: elbow flexed (90°), wrist neutral | NR (isometric) | 43 |
| Coombes et al31 | LET: 150 (sex: 94 M; age: 49.7 [SD = 8.2] y; BMI: 26.8 [SD = 26.8]; DoS: 5.9 [SD = 7.2] mo; outcomes: PRTEE score = 37.9 [SD = 12.8]) Con: 54 (sex: 34 M; age: 50.1 [SD = 10.1] y; BMI: 25.6 [SD = 3.6]; outcomes: NR) | Elbow flexion/extension | Handheld dynamometer | Elbow flexion/extension: standing, elbow flexed (90°), forearm neutral | Isometric | 59 |
| Day et al42 | LET: 28 (sex: 13 M; age: 46.8 [SD = 8.8] y; BMI: 29.0; DoS: NR; outcomes: QuickDASH score = 40.6% [SD = 16.3%]) Con: 28 (sex: 13 M; age: 46.1 [SD = 9.2] y; BMI: 25.1; outcomes: QuickDASH score = 2.6% [SD = 3.5%]) | Appendicular muscles: middle and lower trapezius, serratus anterior | Handheld dynamometer | Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) | NR (isometric) | 56 |
| De Smet and Fabry40 | LET: 55 (sex: 30 M; age: 45 [SD = 9.6] y; BMI: NR; DoS: NR; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: elbow flexed and elbow extended, specifics NR | NR (isometric) | 30 |
| Lucado et al35 | LET: 21 (sex: 21 F; age: 44.9 [SD = 5.2] y; BMI: 23.7 [SD = 3.0]; DoS: 6.1 [range: 1–24] mo; outcomes: NR) Con (nonplayers): 21 (sex: 21 F; age: 43.0 [SD = 8.4] y; BMI: 23.2 [SD = 3.6]; outcomes: NR) Con (tennis players): 21 (sex: 21 F; age: 46.8 [SD = 9.9] y; BMI: 23.9 [SD = 2.9]; outcomes: NR) | Appendicular muscles: upper and lower trapezius Shoulder abduction, external/internal rotation Elbow flexion/extension Wrist flexion/extension | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral, elbow at 90° Lower trapezius: prone, specifics NR Picture: shoulder abducted (~130°), elbow extended Shoulder abduction: seated, shoulder neutral Shoulder external/internal rotation: seated, shoulder abducted (30°) in scapular plane, elbow flexed (90°) Elbow flexion/extension: supine, shoulder neutral, elbow flexed (90°), forearm and wrist neutral Wrist flexion/extension: seated, shoulder abducted (20°), elbow flexed (90°), forearm neutral | Isometric | 47 |
| Nabil et al39 | LET: 10 (sex: 10 M; age: 25.2 [SD = 2.9] y; BMI: 24.4; DoS: NR; outcomes: NR) Con: 10 (sex: 10 M; age: 24.6 [SD = 2.4] y; BMI: 25.7; outcomes: NR) | Shoulder abduction, external rotation | Biodex – isokentic dynamometer | Shoulder abduction: seated, shoulder abducted (120°), elbow extended (range: 120° to neutral, total: 120°) Shoulder external rotation: seated, shoulder abducted (60°–75°), elbow flexed (90°) (range: 70° internal to 70° external, total: 140°) | Eccentric: 60°/s and 120°/s | 44 |
| Stratford et al36 | LET: 35 (sex: 17 M; age: 44.5 [SD = 8.6] y; BMI: NR; DoS: 3 [SD = 3.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder neutral, elbow extended | NR (isometric) | 44 |
| Ucurum et al37 | LET: 51 (sex: 12 M; age: 44.9 [SD = 9.7] y; BMI: 27.5 [SD = 7.4]; DoS: 12 [SD = 13.7] mo; outcomes: PRTEE score = 54.9 [SD = 18.2]) Con: 51 (sex: 9 M; age: 42.7 [SD = 9.7] y; BMI: 25.4 [SD = 4.2]; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior Shoulder abduction, external/internal rotation Grip strength | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) Shoulder abduction: seated, shoulder abducted (90°) in scapular plane Shoulder external/internal rotation: seated, shoulder neutral, elbow flexed (90°) Grip: seated, shoulder neutral, elbow flexed | NR (isometric) | 67 |
| Vigouroux et al14 | LET: 6 (sex: 6 M; age: 27.5 [SD = 8.4] y, BMI: 22.2; DoS: NR; outcomes: NR) Con (nonplayers): 10 (sex: 10 M; age: 26.8 [SD = 3.6] y; BMI: 21.8; outcomes: NR) Con (tennis players): 20 (sex: 20 M; age: 22.3 [SD = 9.5] y; BMI: 21.4; outcomes: NR) | Wrist flexion/extension Metacarpophalangeal flexion/extension | Custom-built apparatus | Wrist and metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow flexed (90°), forearm and wrist neutral | Isometric | 40 |
aItalic type indicates that the position or contraction type was inferred. BMI = body mass index (kg/m2); Con = controls; DASH = Disabilities of the Arm, Shoulder and Hand; DoS = duration of symptoms; EAI = Epidemiology Appraisal Instrument; F = female; LET = lateral elbow tendinopathy; M = male; NR = not reported; PRTEE = Patient-Rated Tennis Elbow Evaluation; QuickDASH = Quick version of the Disabilities of the Arm, Shoulder and Hand.
Characteristics of Included Studiesa
| Study | Population | Movement(s) and Muscle(s) Investigated | Equipment | Position(s) Tested | Contraction Type | EAI Score (%) |
|---|---|---|---|---|---|---|
| Alizadehkhaiyat et al12 | LET: 16 (sex: 8 M; age: 49 [range = 40–66] y; BMI: 25.7; DoS: NR; outcomes: NR) Con: 16 (sex: 9 M; age: 40 [range: 26–59] y; BMI: 23.9; outcomes: NR) | Shoulder abduction, external/internal rotation Wrist flexion/extension Metacarpophalangeal flexion/extension Grip strength | Handheld dynamometer Handgrip dynamometer | Shoulder abduction: shoulder abducted (30°) in scapular plane, elbow flexed (90°) Shoulder external/internal rotation: shoulder abducted (0°–30°) internally and externally, respectively, elbow flexed (90°) Wrist flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral | NR (isometric) | 33 |
| Bhalara and Sheth38 | LET: 66 (sex: 26 M; age: 42.4 [SD = 8.3] y; BMI: 25.6; DoS: NR; outcomes: NR) Con: 66 (sex: 25 M; age: 45.0 [SD = 10.8] y; BMI: 26.6; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior | Handheld dynamometer | NR | NR (isometric) | 43 |
| Bhargava et al32 | LET: 8 athletes (sex: 8 M; age: 34.8 [SD = 7.3] y; BMI: NR; DoS: 8.4 [SD = 9.5] mo; outcomes: NR) (contralateral comparison) LET: 22 nonathletes (sex: 5 M; age: 40.5 [SD = 6.8] y; BMI: NR; DoS: 7.6 [SD = 18.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral, wrist fixed at either 15° or 35° by external brace | NR (isometric) | 47 |
| Bisset et al34 | LET: 40 (sex: 24 M; age: 49.5 [range: 32–66] y; BMI: NR; DoS: 7.7 [SD = 10] mo; outcomes: NR) Con: 40 (sex: 24 M; age: 48.4 [range: 33–64] y; BMI: NR; outcomes: NR) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder flexed (90°), elbow extended (0°), forearm pronated | NR (isometric) | 43 |
| Blanchette and Normand33 | LET: 24 (sex: 11 M; age: 46 [SD = 10] y; BMI: NR; DoS: 29 [SD = 38] mo; outcomes: PRTEE score = 35 [SD = 19]) (contralateral comparison) | Wrist flexion/extension | Load cell | Wrist flexion: seated, elbow flexed (90°), forearm supinated Wrist extension: seated, elbow flexed (90°), forearm pronated | Isometric | 51 |
| Calder et al13 | LET: 11 (sex: 6 M; age: 46.6 [SD = 10.7] y; BMI: 26.6; DoS: NR; outcomes: DASH disability score = 32.39 [SD = 15.1]) At risk: 8 (sex: 2 M; age: 44.8 [SD = 13.5] y; BMI: 23.4; outcomes: DASH disability score = 0.66 [SD = 1.3]) Con: 37 (sex: 15 M; age: 27.1 [SD = 5.0] y; BMI: 23.0; outcomes: DASH disability score = NR | Wrist extension Grip strength | Load cell | Wrist extension: specifics NR Picture: shoulder abducted (~45°), elbow flexed (90°), forearm pronated, wrist neutral Grip: elbow flexed (90°), wrist neutral | NR (isometric) | 43 |
| Coombes et al31 | LET: 150 (sex: 94 M; age: 49.7 [SD = 8.2] y; BMI: 26.8 [SD = 26.8]; DoS: 5.9 [SD = 7.2] mo; outcomes: PRTEE score = 37.9 [SD = 12.8]) Con: 54 (sex: 34 M; age: 50.1 [SD = 10.1] y; BMI: 25.6 [SD = 3.6]; outcomes: NR) | Elbow flexion/extension | Handheld dynamometer | Elbow flexion/extension: standing, elbow flexed (90°), forearm neutral | Isometric | 59 |
| Day et al42 | LET: 28 (sex: 13 M; age: 46.8 [SD = 8.8] y; BMI: 29.0; DoS: NR; outcomes: QuickDASH score = 40.6% [SD = 16.3%]) Con: 28 (sex: 13 M; age: 46.1 [SD = 9.2] y; BMI: 25.1; outcomes: QuickDASH score = 2.6% [SD = 3.5%]) | Appendicular muscles: middle and lower trapezius, serratus anterior | Handheld dynamometer | Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) | NR (isometric) | 56 |
| De Smet and Fabry40 | LET: 55 (sex: 30 M; age: 45 [SD = 9.6] y; BMI: NR; DoS: NR; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: elbow flexed and elbow extended, specifics NR | NR (isometric) | 30 |
| Lucado et al35 | LET: 21 (sex: 21 F; age: 44.9 [SD = 5.2] y; BMI: 23.7 [SD = 3.0]; DoS: 6.1 [range: 1–24] mo; outcomes: NR) Con (nonplayers): 21 (sex: 21 F; age: 43.0 [SD = 8.4] y; BMI: 23.2 [SD = 3.6]; outcomes: NR) Con (tennis players): 21 (sex: 21 F; age: 46.8 [SD = 9.9] y; BMI: 23.9 [SD = 2.9]; outcomes: NR) | Appendicular muscles: upper and lower trapezius Shoulder abduction, external/internal rotation Elbow flexion/extension Wrist flexion/extension | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral, elbow at 90° Lower trapezius: prone, specifics NR Picture: shoulder abducted (~130°), elbow extended Shoulder abduction: seated, shoulder neutral Shoulder external/internal rotation: seated, shoulder abducted (30°) in scapular plane, elbow flexed (90°) Elbow flexion/extension: supine, shoulder neutral, elbow flexed (90°), forearm and wrist neutral Wrist flexion/extension: seated, shoulder abducted (20°), elbow flexed (90°), forearm neutral | Isometric | 47 |
| Nabil et al39 | LET: 10 (sex: 10 M; age: 25.2 [SD = 2.9] y; BMI: 24.4; DoS: NR; outcomes: NR) Con: 10 (sex: 10 M; age: 24.6 [SD = 2.4] y; BMI: 25.7; outcomes: NR) | Shoulder abduction, external rotation | Biodex – isokentic dynamometer | Shoulder abduction: seated, shoulder abducted (120°), elbow extended (range: 120° to neutral, total: 120°) Shoulder external rotation: seated, shoulder abducted (60°–75°), elbow flexed (90°) (range: 70° internal to 70° external, total: 140°) | Eccentric: 60°/s and 120°/s | 44 |
| Stratford et al36 | LET: 35 (sex: 17 M; age: 44.5 [SD = 8.6] y; BMI: NR; DoS: 3 [SD = 3.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder neutral, elbow extended | NR (isometric) | 44 |
| Ucurum et al37 | LET: 51 (sex: 12 M; age: 44.9 [SD = 9.7] y; BMI: 27.5 [SD = 7.4]; DoS: 12 [SD = 13.7] mo; outcomes: PRTEE score = 54.9 [SD = 18.2]) Con: 51 (sex: 9 M; age: 42.7 [SD = 9.7] y; BMI: 25.4 [SD = 4.2]; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior Shoulder abduction, external/internal rotation Grip strength | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) Shoulder abduction: seated, shoulder abducted (90°) in scapular plane Shoulder external/internal rotation: seated, shoulder neutral, elbow flexed (90°) Grip: seated, shoulder neutral, elbow flexed | NR (isometric) | 67 |
| Vigouroux et al14 | LET: 6 (sex: 6 M; age: 27.5 [SD = 8.4] y, BMI: 22.2; DoS: NR; outcomes: NR) Con (nonplayers): 10 (sex: 10 M; age: 26.8 [SD = 3.6] y; BMI: 21.8; outcomes: NR) Con (tennis players): 20 (sex: 20 M; age: 22.3 [SD = 9.5] y; BMI: 21.4; outcomes: NR) | Wrist flexion/extension Metacarpophalangeal flexion/extension | Custom-built apparatus | Wrist and metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow flexed (90°), forearm and wrist neutral | Isometric | 40 |
| Study | Population | Movement(s) and Muscle(s) Investigated | Equipment | Position(s) Tested | Contraction Type | EAI Score (%) |
|---|---|---|---|---|---|---|
| Alizadehkhaiyat et al12 | LET: 16 (sex: 8 M; age: 49 [range = 40–66] y; BMI: 25.7; DoS: NR; outcomes: NR) Con: 16 (sex: 9 M; age: 40 [range: 26–59] y; BMI: 23.9; outcomes: NR) | Shoulder abduction, external/internal rotation Wrist flexion/extension Metacarpophalangeal flexion/extension Grip strength | Handheld dynamometer Handgrip dynamometer | Shoulder abduction: shoulder abducted (30°) in scapular plane, elbow flexed (90°) Shoulder external/internal rotation: shoulder abducted (0°–30°) internally and externally, respectively, elbow flexed (90°) Wrist flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow NR, forearm pronated, wrist neutral Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral | NR (isometric) | 33 |
| Bhalara and Sheth38 | LET: 66 (sex: 26 M; age: 42.4 [SD = 8.3] y; BMI: 25.6; DoS: NR; outcomes: NR) Con: 66 (sex: 25 M; age: 45.0 [SD = 10.8] y; BMI: 26.6; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior | Handheld dynamometer | NR | NR (isometric) | 43 |
| Bhargava et al32 | LET: 8 athletes (sex: 8 M; age: 34.8 [SD = 7.3] y; BMI: NR; DoS: 8.4 [SD = 9.5] mo; outcomes: NR) (contralateral comparison) LET: 22 nonathletes (sex: 5 M; age: 40.5 [SD = 6.8] y; BMI: NR; DoS: 7.6 [SD = 18.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: seated, shoulder neutral, elbow flexed (90°), forearm neutral, wrist fixed at either 15° or 35° by external brace | NR (isometric) | 47 |
| Bisset et al34 | LET: 40 (sex: 24 M; age: 49.5 [range: 32–66] y; BMI: NR; DoS: 7.7 [SD = 10] mo; outcomes: NR) Con: 40 (sex: 24 M; age: 48.4 [range: 33–64] y; BMI: NR; outcomes: NR) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder flexed (90°), elbow extended (0°), forearm pronated | NR (isometric) | 43 |
| Blanchette and Normand33 | LET: 24 (sex: 11 M; age: 46 [SD = 10] y; BMI: NR; DoS: 29 [SD = 38] mo; outcomes: PRTEE score = 35 [SD = 19]) (contralateral comparison) | Wrist flexion/extension | Load cell | Wrist flexion: seated, elbow flexed (90°), forearm supinated Wrist extension: seated, elbow flexed (90°), forearm pronated | Isometric | 51 |
| Calder et al13 | LET: 11 (sex: 6 M; age: 46.6 [SD = 10.7] y; BMI: 26.6; DoS: NR; outcomes: DASH disability score = 32.39 [SD = 15.1]) At risk: 8 (sex: 2 M; age: 44.8 [SD = 13.5] y; BMI: 23.4; outcomes: DASH disability score = 0.66 [SD = 1.3]) Con: 37 (sex: 15 M; age: 27.1 [SD = 5.0] y; BMI: 23.0; outcomes: DASH disability score = NR | Wrist extension Grip strength | Load cell | Wrist extension: specifics NR Picture: shoulder abducted (~45°), elbow flexed (90°), forearm pronated, wrist neutral Grip: elbow flexed (90°), wrist neutral | NR (isometric) | 43 |
| Coombes et al31 | LET: 150 (sex: 94 M; age: 49.7 [SD = 8.2] y; BMI: 26.8 [SD = 26.8]; DoS: 5.9 [SD = 7.2] mo; outcomes: PRTEE score = 37.9 [SD = 12.8]) Con: 54 (sex: 34 M; age: 50.1 [SD = 10.1] y; BMI: 25.6 [SD = 3.6]; outcomes: NR) | Elbow flexion/extension | Handheld dynamometer | Elbow flexion/extension: standing, elbow flexed (90°), forearm neutral | Isometric | 59 |
| Day et al42 | LET: 28 (sex: 13 M; age: 46.8 [SD = 8.8] y; BMI: 29.0; DoS: NR; outcomes: QuickDASH score = 40.6% [SD = 16.3%]) Con: 28 (sex: 13 M; age: 46.1 [SD = 9.2] y; BMI: 25.1; outcomes: QuickDASH score = 2.6% [SD = 3.5%]) | Appendicular muscles: middle and lower trapezius, serratus anterior | Handheld dynamometer | Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) | NR (isometric) | 56 |
| De Smet and Fabry40 | LET: 55 (sex: 30 M; age: 45 [SD = 9.6] y; BMI: NR; DoS: NR; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: elbow flexed and elbow extended, specifics NR | NR (isometric) | 30 |
| Lucado et al35 | LET: 21 (sex: 21 F; age: 44.9 [SD = 5.2] y; BMI: 23.7 [SD = 3.0]; DoS: 6.1 [range: 1–24] mo; outcomes: NR) Con (nonplayers): 21 (sex: 21 F; age: 43.0 [SD = 8.4] y; BMI: 23.2 [SD = 3.6]; outcomes: NR) Con (tennis players): 21 (sex: 21 F; age: 46.8 [SD = 9.9] y; BMI: 23.9 [SD = 2.9]; outcomes: NR) | Appendicular muscles: upper and lower trapezius Shoulder abduction, external/internal rotation Elbow flexion/extension Wrist flexion/extension | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral, elbow at 90° Lower trapezius: prone, specifics NR Picture: shoulder abducted (~130°), elbow extended Shoulder abduction: seated, shoulder neutral Shoulder external/internal rotation: seated, shoulder abducted (30°) in scapular plane, elbow flexed (90°) Elbow flexion/extension: supine, shoulder neutral, elbow flexed (90°), forearm and wrist neutral Wrist flexion/extension: seated, shoulder abducted (20°), elbow flexed (90°), forearm neutral | Isometric | 47 |
| Nabil et al39 | LET: 10 (sex: 10 M; age: 25.2 [SD = 2.9] y; BMI: 24.4; DoS: NR; outcomes: NR) Con: 10 (sex: 10 M; age: 24.6 [SD = 2.4] y; BMI: 25.7; outcomes: NR) | Shoulder abduction, external rotation | Biodex – isokentic dynamometer | Shoulder abduction: seated, shoulder abducted (120°), elbow extended (range: 120° to neutral, total: 120°) Shoulder external rotation: seated, shoulder abducted (60°–75°), elbow flexed (90°) (range: 70° internal to 70° external, total: 140°) | Eccentric: 60°/s and 120°/s | 44 |
| Stratford et al36 | LET: 35 (sex: 17 M; age: 44.5 [SD = 8.6] y; BMI: NR; DoS: 3 [SD = 3.5] mo; outcomes: NR) (contralateral comparison) | Grip strength | Handgrip dynamometer | Grip: standing, shoulder neutral, elbow extended | NR (isometric) | 44 |
| Ucurum et al37 | LET: 51 (sex: 12 M; age: 44.9 [SD = 9.7] y; BMI: 27.5 [SD = 7.4]; DoS: 12 [SD = 13.7] mo; outcomes: PRTEE score = 54.9 [SD = 18.2]) Con: 51 (sex: 9 M; age: 42.7 [SD = 9.7] y; BMI: 25.4 [SD = 4.2]; outcomes: NR) | Appendicular muscles: upper, middle, and lower trapezius, serratus anterior Shoulder abduction, external/internal rotation Grip strength | Handheld dynamometer | Upper trapezius: seated, shoulder and head neutral Middle trapezius: prone, shoulder abducted (90°), elbow flexed Lower trapezius: prone, shoulder abducted (135°), elbow extended Serratus anterior: supine, shoulder and elbow flexed (90°) Shoulder abduction: seated, shoulder abducted (90°) in scapular plane Shoulder external/internal rotation: seated, shoulder neutral, elbow flexed (90°) Grip: seated, shoulder neutral, elbow flexed | NR (isometric) | 67 |
| Vigouroux et al14 | LET: 6 (sex: 6 M; age: 27.5 [SD = 8.4] y, BMI: 22.2; DoS: NR; outcomes: NR) Con (nonplayers): 10 (sex: 10 M; age: 26.8 [SD = 3.6] y; BMI: 21.8; outcomes: NR) Con (tennis players): 20 (sex: 20 M; age: 22.3 [SD = 9.5] y; BMI: 21.4; outcomes: NR) | Wrist flexion/extension Metacarpophalangeal flexion/extension | Custom-built apparatus | Wrist and metacarpophalangeal flexion/extension: shoulder abducted (0°–15°), elbow flexed (90°), forearm and wrist neutral | Isometric | 40 |
aItalic type indicates that the position or contraction type was inferred. BMI = body mass index (kg/m2); Con = controls; DASH = Disabilities of the Arm, Shoulder and Hand; DoS = duration of symptoms; EAI = Epidemiology Appraisal Instrument; F = female; LET = lateral elbow tendinopathy; M = male; NR = not reported; PRTEE = Patient-Rated Tennis Elbow Evaluation; QuickDASH = Quick version of the Disabilities of the Arm, Shoulder and Hand.
Database search and inclusion of studies.
Study Characteristics
The number of participants with symptoms within a study ranged from 614 to 150.31 Symptom duration was reported in 7 of the 14 studies31–37 and ranged from 336 to 2933 months. Six studies had comparisons with an asymptomatic control group,13,14,34,35,38,39 3 had comparisons with the asymptomatic contralateral limb,33,36,40 and 5 included both comparisons.31,32,37,41,42 The methods of strength assessment varied between studies and included the position of testing, equipment, and measurement units, with only 2 studies normalizing strength measures to body mass.37,39 Four studies reported upper, middle, and lower trapezius and serratus anterior strength35,37,38,42; 3 reported shoulder abduction, internal rotation, and external rotation strength35,37,41; 2 reported elbow flexion and extension strength31,35; 4 reported wrist flexion and extension strength14,33,35,41; 2 reported finger flexion and extension strength14,41; and 5 reported grip strength.13,34,36,37,41 One study39 assessed eccentric strength, and the remaining studies stated or inferred maximal isometric strength.13,14,31–38,40–42 Three of the included studies reported patient-reported outcome measures (Table). All studies excluded individuals with current upper limb musculoskeletal conditions other than clinically diagnosed LET.
Quality Assessment
The interrater agreement for the quality assessment was almost perfect (κ = 0.95, P ≤ .001), with 19 disagreements from a total of 490 decisions. The mean quality score was 46% (range = 30%–67%) (Table). Four studies32,37,38,42 described the source of the participant population, with participation rates being reported in 2 studies.13,33 Participants lost to follow-up were described in 3 studies,31,33,42 with only 1 study42 accounting for the loss in the analysis. Four studies described participant characteristics, including age, sex, duration of symptoms, functional disability score, hand dominance, and body mass index.31,35,37,42 Two studies reported sample size calculations.32,37 The reliability of the primary outcomes was reported in 7 studies,31,33,35,36,38,39,42 whereas only 2 reported validity.36,39 Adjustments for covariates and cofounders were reported in 3 studies,31,39,42 with assessor masking being reported in only 1 study.32
Meta-Analysis of Symptomatic Limb Versus Asymptomatic Control
Appendicular Strength
For the upper trapezius, substantial heterogeneity prevented meta-analysis (I2 = 53%). Three of the 4 comparisons had a negative ES (range = −0.60 to −0.23), equating to a 6% to 36% decrease in maximal strength in the symptomatic limb relative to the results for an asymptomatic control group35,37; however, only 1 comparison was statistically significant (ES = −0.60 [95% CI = −0.95 to −0.26]) (Fig. 2).38 Two studies revealed no significant difference in strength between the symptomatic limb and the asymptomatic contralateral limb (Fig. 2).37,38
Effect sizes and associated 95% CIs for maximal strength variables of the appendicular muscles: upper trapezius, middle trapezius, lower trapezius, and serratus anterior. aSymptomatic limb vs asymptomatic control group. bSymptomatic limb vs asymptomatic contralateral limb. *Individuals with lateral elbow tendinopathy vs tennis players who were not symptomatic.
For the middle trapezius, substantial heterogeneity prevented meta-analysis (I2 = 77%). Two of the 3 comparisons had a negative ES (range = −0.31 and − 0.60), equating to a 7% to 38% decrease in maximal strength in the symptomatic limb relative to the results for an asymptomatic control group;37,38 however, only 1 comparison was statistically significant (ES = −0.60 [95% CI = −0.95 to −0.25]) (Fig. 2).38 There were no significant differences in strength between the symptomatic limb and the asymptomatic contralateral limb (0%–8% decrease) (Fig. 2).37,38,42
For the lower trapezius, substantial heterogeneity prevented meta-analysis (I2 = 65%). Four of the 5 comparisons had a negative ES (range = −0.23 to −0.60), equating to a 7% to 29% decrease in maximal strength in the symptomatic limb relative to the results for an asymptomatic control group35,37,38,42; 2 comparisons were statistically significant (ES = −1.27 to −0.55) (Fig. 2).35,38 Two studies revealed significantly lower maximal strength in the symptomatic limb than in the asymptomatic contralateral limb (ES range = −0.43 to −0.29; 11%–16% decrease)38,42; however, 1 study showed no significant difference (Fig. 2).37
For the serratus anterior, substantial heterogeneity prevented meta-analysis (I2 = 69%). Three studies showed a negative ES (range = −0.17 to −1.00), equating to a 5% to 37% decrease in maximal strength in the symptomatic limb relative to the results for an asymptomatic control group37,38,42; 2 comparisons were significant (ES = −1.00 to −0.63) (Fig. 2).38,42 One study revealed significantly less strength in the symptomatic limb than in the asymptomatic contralateral limb (ES = −0.33 [95% CI = −0.65 to −0.03])42; however, 2 studies found no significant difference (Fig. 2).37,38
Shoulder Strength
For shoulder abduction, meta-analysis revealed a small but statistically significant pooled ES (−0.37 [95% CI = −0.62 to −0.12]; I2 = 0%), suggesting that shoulder abduction strength was reduced in the symptomatic limb relative to that in an asymptomatic control group. In contrast, 2 studies revealed no significant strength differences between the symptomatic limb and the asymptomatic contralateral limb (Fig. 3).37,41 One study (not shown in Fig. 3) reported a statistically significant deficit in maximal eccentric shoulder abduction strength (from 120° to 0°) in the symptomatic limb relative to the results for an asymptomatic control group at both 60 degrees/s (ES = −6.92 [95% CI = −9.70 to −4.71]; 40% decrease) and 120 degrees/s (ES = −3.50 [95% CI = −5.10 to −2.16]; 38% decrease).39
Effect sizes and associated 95% CIs for maximal strength variables of shoulder movements: abduction, external rotation, and internal rotation. aSymptomatic limb vs asymptomatic control group. bSymptomatic limb vs asymptomatic contralateral limb. *Individuals with lateral elbow tendinopathy vs tennis players who were not symptomatic.
For shoulder external rotation, meta-analysis revealed a significant pooled ES (= −0.55 [95% CI = −0.83 to −0.28]; I2 = 0%), equating to a 10% to 26% decrease in shoulder external rotation strength in the symptomatic limb relative to the results for an asymptomatic control group. Of the 2 studies investigating maximal strength differences between the symptomatic and asymptomatic limbs, 1 revealed less strength in the symptomatic limb,37 whereas the other found no significant difference.12 One study (not shown in Fig. 3) compared maximal eccentric external rotation strength values (140-degree angular displacement) and reported a statistically significant reduction in the symptomatic limb relative to the results for an asymptomatic control group at both 60 degrees/s (ES = −2.38 [95% CI = −3.65 to −1.27]; 30% decrease) and 120 degrees/s (ES = −1.95 [95% CI = −3.11 to −0.92]; 28% decrease).39
For shoulder internal rotation, substantial heterogeneity prevented meta-analysis (I2 = 63%). Two of 4 comparisons showed a negative ES (−1.07 and − 0.25), equating to a 7% to 36% decrease in maximal strength in the symptomatic limb relative to the results for an asymptomatic control group; only the results of 1 study were significant (−1.07 ES [95% CI = −1.84 to −0. 35]) (Fig. 3).37 One study revealed significantly less strength in the symptomatic limb than in the asymptomatic control limb (ES = −0.25 [95% CI = −0.47 to −0.03])37; however, another study found no significant difference (Fig. 3).12
Elbow Strength
For elbow extension, meta-analysis revealed no significant difference in elbow extension strength between the symptomatic limb and an asymptomatic control group (pooled ES = −0.02 [95% CI = −0.55 to 0.59]; I2 = 38%) (Suppl. Fig. 1). In addition, there were no significant differences in elbow extension strength between the symptomatic limb and the asymptomatic contralateral limb.31
For elbow flexion, meta-analysis revealed no significant difference in elbow flexion between the symptomatic limb and an asymptomatic control group (pooled ES = −0.27 [95% CI = −0.60 to 0.05]; I2 = 28%) (Suppl. Fig. 1). One study reported no significant difference in elbow flexion strength between the symptomatic limb and the asymptomatic contralateral limb.31
Wrist Strength
For wrist extension, substantial heterogeneity prevented meta-analysis (I2 = 79%). Five of the 7 comparisons revealed a negative ES (range = −1.06 to −0.43), equating to an 8% to 33% decrease in maximal wrist extension strength in the symptomatic limb relative to the results for an asymptomatic control group; 4 comparisons were statistically significant (Fig. 4).12,35 Two comparisons revealed a positive ES (range = 0.56–0.95), indicating 31% greater strength in the symptomatic limb than in an asymptomatic control group.13,14 There were no significant differences in wrist extension strength between the symptomatic limb and the asymptomatic contralateral limb.12,37
Effect sizes and associated 95% CIs for maximal strength variables of wrist flexion and extension. aSymptomatic limb vs asymptomatic control group. bInjured side vs asymptomatic contralateral limb. †Individuals with lateral elbow tendinopathy (LET) vs population at-risk. *Individuals with LET vs tennis players who were not symptomatic.
For wrist flexion, substantial heterogeneity prevented pooled analysis (I2 = 83%). Three of the 5 comparisons showed a significant negative ES (range = −0.97 to −0.95), equating to a 10% to 25% decrease in maximal wrist flexion strength in the symptomatic limb relative to the results for an asymptomatic control group (Fig. 4).12,33 In contrast, 1 study reported significantly greater (32%) wrist flexion strength in the symptomatic limb than in an asymptomatic control group.14 Two studies revealed no significant difference in wrist flexion strength between the symptomatic limb and the asymptomatic contralateral limb.12,33
Metacarpophalangeal Strength
For metacarpophalangeal (MCP) extension, meta-analysis revealed no significant difference between the symptomatic limb and an asymptomatic control group (pooled ES = −0.30 [95% CI = −0.78 to 0.19]; I2 = 0%) (Suppl. Fig. 2). One study revealed no significant difference in MCP extension strength between the symptomatic limb and the asymptomatic contralateral limb.12
For MCP flexion, substantial heterogeneity prevented meta-analysis (I2 = 83%). One comparison revealed a negative ES that was significant (ES = −1.21 [95% CI = −1.99 to −0.47]; 37% decrease), whereas 2 comparisons demonstrated a positive ES (0.49 and 0.54; 14% increase) (Suppl. Fig. 2). No significant differences were observed between the symptomatic and asymptomatic limbs (ES = −0.13; 8% decrease) (Fig. 2).37,38
Maximal Grip Strength
Maximal grip strength data are presented separately for testing positions of elbow flexion and extension. Substantial heterogeneity prevented meta-analysis for grip with the elbow flexed (I2 = 53%). Two studies revealed a significant negative ES (−0.94 and − 0.78), equating to a 25% to 29% decrease in maximal grip strength in the symptomatic limb relative to the results for an asymptomatic control group12,13; however, 1 study showed no significant difference (ES = 0.12; 4% increase).37 All 7 comparisons from 4 studies showed a negative ES (range = −1.47 to −0.02), equating to a 1% to 24% decrease in the maximal grip strength with the elbow flexed in the symptomatic limb relative to the results for the asymptomatic contralateral limb (Fig. 5).12,32,37,40
Effect sizes and associated 95% CIs for maximal strength variables of grip strength with a flexed or an extended elbow position. aSymptomatic limb vs asymptomatic control group. bSymptomatic limb vs asymptomatic contralateral limb. †Individuals with lateral elbow tendinopathy (LET) vs population at-risk. #Nonathletes. ‡Athletes.
One study revealed a nonsignificant negative ES (−0.39 [95% CI = −0.84 to 0.05]), equating to a 14% decrease in maximal grip strength with the elbow extended in the symptomatic limb relative to the results for an asymptomatic control group (Fig. 5).34 Three studies revealed a significant negative ES (range = −1.39 to −0.36), equating to a 12% to 50% decrease in maximal grip strength in the symptomatic limb relative to the results for the asymptomatic contralateral limb.34,36,40
Meta-analysis of Asymptomatic Limb Versus Asymptomatic Control Group
Appendicular Strength
For the upper trapezius, meta-analysis showed a small but statistically significant pooled ES (−0.26 [95% CI = −0.49 to −0.02], I2 = 0%), equating to a 4% to 24% decrease in maximal upper trapezius strength in the asymptomatic limb relative to the results for an asymptomatic control group (Suppl. Fig. 3).
For the middle trapezius, substantial heterogeneity prevented meta-analysis (I2 = 85%). One study had a significant negative ES (−0.49 [95% CI = −0.84 to −0.15]), equating to a 32% decrease in the maximal strength of the middle trapezius in the asymptomatic limb relative to the results for an asymptomatic control group38; another study found no significant differences (Suppl. Fig. 3).37
For the lower trapezius, meta-analysis revealed no significant difference between the asymptomatic limb and an asymptomatic control group (pooled ES = −0.02 [95% CI = −0.29 to 0.34]; I2 = 34%) (Suppl. Fig. 3).
For the serratus anterior, substantial heterogeneity prevented meta-analysis (I2 = 71%). One study had a significant negative ES (−0.50 [95% CI = −0.85 to −0.15]),38 equating to a 29% decrease in the maximal strength of the serratus anterior in the asymptomatic limb relative to the results for an asymptomatic control group; another study found no significant difference (ES = 0.00, 0% difference) (Suppl. Fig. 3).37
Shoulder Strength
For shoulder abduction, meta-analysis revealed no strength differences between the asymptomatic limb and an asymptomatic control group (pooled ES = −0.25 [95% CI = −0.59 to 0.09]; I2 = 0%) (Suppl. Fig. 3).
For shoulder external rotation, meta-analysis revealed no strength differences between the asymptomatic limb and an asymptomatic control group (pooled ES = −0.15 [95% CI = −0.33 to 0.03]; I2 = 0%) (Suppl. Fig. 3).
For shoulder internal rotation, meta-analysis revealed no strength differences between the asymptomatic limb and an asymptomatic control group (−0.18 [95% CI = −0.68 to 0.31]; I2 = 40%) (Suppl. Fig. 3).
Elbow Strength
For elbow extension, 1 study revealed no significant differences in elbow extension strength between the asymptomatic limb and an asymptomatic control group (Suppl. Fig. 4).31
For elbow flexion, 1 study revealed no significant differences in elbow flexion strength between the asymptomatic limb and an asymptomatic control group (Suppl. Fig. 4).31
Wrist Strength
For wrist extension, 1 study revealed a significant negative ES (−0.74 [95% CI = −1.48 to −0.03]), equating to a 28% decrease in wrist extension strength in the asymptomatic limb relative to the results for an asymptomatic control group (Suppl. Fig. 4).12
For wrist flexion, 1 study revealed no significant differences in wrist flexion strength between the asymptomatic limb and an asymptomatic control group (Suppl. Fig. 4).12
Metacarpophalangeal Joint Strength
For MCP joint extension, 1 study revealed a significant negative ES (−1.04 [95% CI = −1.48 to −0.03]), equating to a 29% decrease in MCP joint extension strength in the asymptomatic limb relative to the results for an asymptomatic control group (Suppl. Fig. 4).12
For MCP joint flexion, 1 study revealed a significant negative ES (−0.80 [95% CI = −1.54 to −0.09]), equating to a 26% decrease in MCP joint flexion strength in the asymptomatic limb relative to the results for an asymptomatic control group (Suppl. Fig. 4).12
Maximal Grip Strength
Meta-analysis revealed a moderate statistically significant pooled ES (−0.57 [95% CI = −0.91 to −0.22]; I2 = 0%), equating to a 16% to 17% decrease in grip strength with the elbow flexed in the asymptomatic limb relative to the results for an asymptomatic control group (Suppl. Fig. 4). In contrast, 1 study revealed a significant moderate ES (0.59 [95% CI = 0.14 to 1.04]; 21% increase), showing greater maximal grip strength with the elbow extended in the asymptomatic limb than in an asymptomatic control group.34
Discussion
Our meta-analyses provide evidence of maximal strength deficits in shoulder abduction and external rotation, but not elbow flexion and extension, in the symptomatic limb of individuals with LET relative to the results for an asymptomatic control group. Although substantial heterogeneity prevented additional meta-analyses, there was also evidence for maximal strength deficits in the lower trapezius and serratus anterior muscles, wrist extension, and grip in the symptomatic limb relative to the results for an asymptomatic control group. Similarly, there was consistent evidence for reduced strength of the upper trapezius, wrist extensors, finger flexors and extensors, and maximal grip in the asymptomatic limb relative to the results for an asymptomatic control group. Our findings suggest that local and global deficits in maximal strength are evident in both limbs in individuals with unilateral LET relative to the results for an asymptomatic control group, highlighting the complex presentation of LET.
Upper Limb Strength Deficits in LET
The novelty of our review is the observation of strength deficits in the appendicular and shoulder muscles in individuals with unilateral LET relative to the results for an asymptomatic control group and an asymptomatic contralateral limb. Except for 3,35,37 all comparisons for scapular muscle strength were negative, with small to large ES estimates for lower trapezius35,38,42 and serratus anterior37,38,42 during isometric testing; however, many of the 95% CIs crossed the zero line, limiting our ability to draw robust conclusions. Similar but smaller differences were also observed between the symptomatic and asymptomatic limbs.37,38,42 In addition, there was a small to medium negative ES for shoulder abduction and external rotation strength between the symptomatic limb and an asymptomatic control group. These differences were more pronounced under eccentric conditions normalized to body mass.39 Except for Alizadehkhaiyat et al,12 who reported a 36% decrease in internal rotation strength, there was no evidence for a deficit in internal rotation strength. Consistent with lower limb tendinopathies (eg, Achilles and patellar)43–45 it may be that concomitant weakness of the proximal stabilizers represents either a cause or a consequence of the more distal tendinopathy.
From the perspective of proximal strength deficits as a cause of LET, there is evidence that the proximal shoulder muscles play a critical role in scapular stabilization and positioning of the glenohumeral joint during distal extremity movements.15 For instance, in healthy controls, fatigue of the scapular stabilizers results in altered elbow kinematics during overhead throwing.16 These findings suggest that altered muscle control of the appendicular musculature can directly influence control of the distal limb segments, which may lead to increased loading of the wrist-forearm extensors during grip-related upper limb tasks. Although it is difficult to extrapolate such findings to LET, alterations in scapular control and shoulder position may contribute to increased extensor tendon loading during repetitive upper limb activities involving gripping37 and increase the likelihood of overuse. Although prospective studies would be required to establish whether deficits in proximal muscle strength increase the risk of LET, this review does support that such deficits are evident in individuals with LET and may represent a modifiable risk factor for the prevention of LET.
From the perspective of proximal strength deficits as a consequence of LET, animal models provide evidence of local and systemic muscle and tendon changes following the development of an overuse injury.46 More specifically, Barbe et al46 reported an increase in inflammatory mediators (ie, immunoreactive macrophages) of both the reaching and nonreaching limbs, as well as the shoulder and hind limb, in rats performing a repetitive unilateral reaching task. This finding suggests a systemic mediator in repetitive overuse injuries that could result in both local and global effects on muscle and tendon function. In addition, pain and/or the fear of pain has been shown to decrease maximum force output in individuals with chronic musculoskeletal pain,47 whereby altered motor patterns may contribute to limb disuse and atrophy of the proximal shoulder musculature. However, if upper limb strength deficits were attributed to disuse, one might argue that patients’ asymptomatic limb would display greater strength, which was not consistently observed in our review. In general, systemic inflammatory mediators or pain inhibition might explain the widespread strength deficits observed in our study; however, future high-quality prospective studies are required to investigate whether proximal muscle weakness is related to the cause or effect of LET.
Consistent with the current understanding of LET,2 we identified medium to large effects for maximal strength deficits in the wrist extensors, wrist flexors, and gripping in individuals with LET relative to the results for an asymptomatic control group. Although there was consistent evidence for reduced wrist extension and flexion strength, 1 study showed significantly greater strength in the symptomatic limb than in an asymptomatic control group for wrist flexion.14 This difference is likely explained by sport-specific adaptations given the inclusion of experienced tennis players (3–4 years of training) compared with individuals not involved in any activity requiring repetitive use of the finger or wrist muscles.14 Overall, there is consistent evidence of reduced strength of the wrist extensors and gripping, but not of the finger extensors and flexors, in the symptomatic limb relative to the results for an asymptomatic control group. Given the study design of the included studies, it is not possible to identify whether these changes are related to the cause or effect of LET.
Bilateral Deficits in Unilateral LET
Our review identified consistent evidence for maximal strength deficits in the asymptomatic upper limb of individuals with unilateral LET relative to the results for asymptomatic controls. More specifically, deficits in strength of the upper trapezius, wrist extension, MCP joint flexion and extension, and maximal grip strength with the elbow flexed were identified. These findings suggest widespread strength deficits in individuals with LET, which may be mediated by central nervous system adaptations.11 In contrast, maximal grip with the elbow extended was significantly greater in the asymptomatic limb than in an asymptomatic control group. This latter finding may support the theory of compensation due to disuse of the symptomatic limb.
Consistent with our findings, bilateral strength deficits have been identified in individuals with carpal tunnel syndrome,48 de Quervain disease,49 and nonspecific hand and wrist pain.50 More specifically, individuals with LET have shown bilateral deficits during reaction time tasks, speed of movement tasks, and electromechanical delay measures relative to the results for an asymptomatic control.11 Bilateral strength deficits in individuals with unilateral LET suggest the contralateral side may not be ``unaffected”; thus, comparison with an asymptomatic control group is recommended. Although it is not clear whether bilateral strength deficits are related to the cause or effect of LET, these findings suggest a more widespread mechanism than local tendon pathology.
Clinical Implications and Future Research
The existence of bilateral strength deficits of forearm, shoulder, and appendicular muscles in individuals with unilateral LET has direct implications for assessment and rehabilitation. We recommend that a thorough assessment of an individual with LET include tests of shoulder muscle strength, focusing on the external rotators and scapular stabilizers. Clinicians can then tailor rehabilitation programs based on any identified deficits. Although the inclusion of proximal strength training in individuals has received some attention in the literature,51,52 further studies are required, with larger sample sizes and longer follow-up to quantify the clinical benefits of proximal strength training in LET. Furthermore, given the evidence of persistent bilateral strength deficits in unilateral LET,12 rehabilitation of muscle-tendon function of the unaffected limb should occur to ensure full functional recovery of the upper limb kinetic chain bilaterally; however, this theory requires confirmation with scientific research. Finally, use of the asymptomatic limb as a comparison may not be optimal given the bilateral strength deficits identified. Where possible, comparison with age- and sex-matched normative data may lead to full recovery of muscle-tendon function, which may be a more appropriate indication of complete recovery of LET.
Limitations
There are several methodological considerations that should be mentioned. First, given all the included studies employed cross-sectional designs, it remains unclear if the strength deficits observed in our review are related to the cause or an effect of LET. Second, our search only included studies published in English, which may have resulted in missed studies. Third, despite attempts to contact authors, we were unable to acquire missing data for 2 eligible studies,27,28 which may have subtly altered our findings. Fourth, although the majority of our ES are negative, signifying less maximal strength in the affected limb of individuals with LET, the 95% CIs crossed 0 for a number of the individual studies, limiting our ability to draw strong conclusions. Fifth, our review identified differences in the methods used to assess strength (eg, handheld dynamometer, custom-built, load-cell, isokinetic dynamometer) and the positions tested in the included studies. These differences may impact the pooled ES observed within our review. Sixth, the use of a fixed correlation coefficient (r = .7) for the within-subject ES calculation was conservative23 but likely had little impact on our overall findings. Seventh, the meta-analyses included only a small number of studies with small sample sizes, limiting the ability to draw robust conclusions. Finally, because of substantial heterogeneity, we were unable to run meta-analysis for several strength measurements, limiting the scope of our findings.
In conclusion, this review highlights the presence of bilateral strength deficits, including shoulder and appendicular musculature, in individuals with unilateral LET relative to the results for asymptomatic controls. These findings could reflect involvement of the central nervous system in systemic motor system deficits, which may contribute to the expression of pain and disability, as well as high recurrence rates in LET. Our results demonstrate that the contralateral side should not be used for comparison during physical assessments, either in clinical practice or research, and that assessment and rehabilitation targeting muscles other than local wrist extensor muscles are likely to be required to address the widespread strength deficits.
Author Contributions
Concept/idea/research design: L.J. Heales, N. Bout, B. Dines, T. Parker, K. Reddiex, C.O. Kean, S.J. Obst
Writing: L.J. Heales, N. Bout, B. Dines, T. Parker, K. Reddiex, C.O. Kean, S.J. Obst
Data collection: L.J. Heales, N. Bout, B. Dines, T. Parker, K. Reddiex
Data analysis: L.J. Heales, N. Bout, B. Dines, T. Parker, K. Reddiex, C.O. Kean, S.J. Obst
Project management: L.J. Heales
Consultation (including review of manuscript before submitting): L.J. Heales, C.O. Kean, S.J. Obst, N. Bout, B. Dines, T. Parker, K. Reddiex
Systematic Review Registration
This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines and was prospectively registered on PROSPERO (CRD42020189799).
Funding
There are no funders to report for this study.
Disclosures
The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.
References
Comments