Adverse changes in tendon–muscle physiology and physical function caused by an isolated acute rheumatoid knee effusion: A case study



Patients with rheumatoid arthritis (RA) have reduced muscle mass, probably due to the catabolic effects of systemic inflammation and disuse (1). This loss leads to reduced strength and physical function (1, 2). In addition, the presence of effusion in the knee joint results in reflex inhibition of the quadriceps as demonstrated by reduced electromyogram (EMG) activity and diminished strength (3, 4).

The wider characteristics of the tendon–muscle complex have not been investigated in acute RA or in joint effusions of other etiologies. In exercise science, electrophysiologic methods are used to investigate tendon–muscle properties in healthy populations. These include assessments of muscle-specific force (force/cross-sectional area), architecture (fiber fascicle length and pennation angle), voluntary activation capacity, contractile properties, and tendon stiffness. Application of these techniques in aging and disuse has demonstrated adverse changes in tendon and muscle properties that result in impaired function but respond well to exercise training (5, 6).

We have previously applied these methods in patients with well-controlled RA and found that tendon stiffness was reduced (Matschke V et al: unpublished observations), while muscle quality (specific force, voluntary activation capacity, contractile properties) was unaffected (7, 8). To our knowledge, these electrophysiologic methods have not been applied in acute RA, probably due to difficulties arising from confounding factors such as pain, fatigue, and the acute effects of inflammation in other joints.

Case description

A fit 53-year-old female marine biologist presented to the rheumatology clinic with a 6-month history of a swollen stiff left knee. There were no other musculoskeletal symptoms; she was otherwise well and was taking no medication. Examination showed a large left knee effusion. The right knee and all other joints were normal. Investigations showed mild inflammation as indicated by the erythrocyte sedimentation rate (ESR) of 39 mm/hour (normal value <10); positive anti–cyclic citrullinated peptide (anti-CCP) antibody, a marker specific for RA; and no erosions or loss of joint space on radiographs of the left knee. The Rheumatoid Arthritis Disease Activity Index (RADAI-5) is a patient questionnaire that assesses global disease activity over the past 6 months, and current disease activity in terms of swollen and tender joints, arthritis pain, general health, and duration of morning stiffness. Our participant scored 4.4 in the RADAI-5, indicating moderate activity (0 = no activity, 10 = active disease). A diagnosis of CCP-positive RA was made. This provided the opportunity to investigate the local and systemic effects of an acute inflammatory knee effusion on physical function and the tendon–muscle complex. She was a suitable participant because in spite of the large effusion, she had no pain or fatigue and had kept up regular exercise in the gym and as part of her vigorous outdoor occupation. Therefore, the effects of disuse and pain had minimal or no influence on tendon–muscle testing, which requires full cooperation and maximal effort from the participant.

After the initial assessment of the tendon–muscle properties the left knee was aspirated, yielding 30 ml of synovial fluid, and a long-acting corticosteroid (depomedrone 80 mg) was injected intraarticularly. The effusion settled without recurrence and the participant maintained her usual physical activity. One year later she remained well with no symptoms. Examination of the knees showed no effusions, the other joints were normal, ESR was 2 mm/hour, and the RADAI-5 score was 1. All baseline tendon, muscle, and function measures were repeated at 12 months.


All of the investigations were carried out in the Bangor University exercise physiology laboratory. The participant gave informed written consent.

Maximal isometric and concentric quadriceps strength was measured on both legs with separate unilateral contractions with the knee joints at 90° on an isokinetic dynamometer (CSMI Medical Solutions), with concurrent electromyography of the vastus lateralis and biceps femoris. Calculation of quadriceps force took into account moment arm and biceps femoris coactivation torque (9). Percutaneous electrical stimulation of the quadriceps at rest determined contractile properties (electrically evoked peak force, time to peak force, and half-relaxation time) and was superimposed on maximal voluntary quadriceps contraction (MVC) to determine voluntary activation capacity (10). Quadriceps anatomic cross-sectional area (ACSA) was measured using ultrasonography (US), which enabled determination of specific force (force/ACSA). Concentric force and velocity-specific power were determined from the strongest of 4 consecutive concentric quadriceps contractions at 50°/second (8).

Patellar tendon (PT) stiffness was determined using US as described by Onambele-Pearson and Pearson (9). The distance between the apex of the patella and the superior aspect of the tibial tuberosity was taken as resting PT length. PT excursion from its proximal and distal bone attachments was assessed during 3 ramped MVCs (i.e., building up to maximum force with increasing effort over 4–5 seconds) with the US probe held in the sagittal plane. Images were analyzed using digitizing software (Image J, National Institutes of Health). The tendon force–elongation relationship was assessed at intervals of 12.5% of the maximal force, and fitted with second-order polynomial functions forced through zero. Tendon stiffness was calculated from the slope of the tangent (first derivative of the polynomial function) at the level of maximum force. US images taken in the axial plane at 25%, 50%, and 75% of the PT length were used to determine PT CSA. Three measurements were taken at each level and averaged. Young's modulus (YM) was calculated as follows: YM = tendon stiffness × (resting tendon length/tendon CSA) (5).

Lower body physical function was assessed by 30-second sit-to-stand, 8-foot up and go, 50-foot walk, and one-leg standing balance tests (11).

At baseline, results of the affected leg were compared with those of the unaffected leg. At followup, results were compared with baseline data for the same leg. US images were blinded for analysis.

Baseline results

Results at baseline and 1-year followup are shown in Table 1. The effusion caused considerable physical impairment. The sit-to-stand result of 10 corresponds to the 15th percentile of a healthy 60–64-year-old female reference population (11).

Table 1. Quality of the tendon–muscle complex in the affected and unaffected legs, physical function, and disease activity at baseline and at 1-year followup in the 53-year-old female rheumatoid arthritis participant*
 At baseline1-year followup
Both legsUnaffected legAffected legBoth legsUnaffected legAffected leg
  • *

    ACSA = anatomic cross-sectional area.

  • Indicates missing values for contractile properties.

Sit-to-stand, no. per 30 seconds10  13  
8-foot up and go, seconds5.22  4.81  
50-foot walk, seconds7.22  6.06  
Balance, seconds101  120  
Quadriceps isometric force, N 5,0682,863 4,8884,530
Quadriceps ACSA, cm2 53.937.3 51.241.0
Muscle-specific force, N/cm2 94.076.7 95.6110.5
Concentric torque at 50°/second, Nm 129.459.3 136.6103.6
Velocity-specific power, W 112.951.8 119.390.4
Voluntary activation capacity, % 90.693.0 92.998.7
Patellar tendon stiffness, N/mm 7,2424,088 4,9443,647
Patellar tendon CSA, cm2 1.031.01 1.051.06
Young's modulus, GPa 3.992.27 2.001.61
Resting twitch peak torque, Nm 17.9 21.017.3
Time to peak torque, seconds 0.0850.083 0.1040.084
Half-relaxation time, seconds 0.0420.047 

At baseline, the quadriceps ACSA of the affected leg was 31% less than that of the unaffected leg. Voluntary activation capacity of both quadriceps was more than 90%, demonstrating that maximal muscle contraction was not inhibited. Despite this, muscle quality was considerably impaired on the affected side, with isometric and concentric forces reduced by 44% and 55%, respectively. This force reduction could not be explained by diminished quadriceps ACSA alone, as specific force (force/ACSA) was reduced by 18% (Figure 1). Contractile properties were equal bilaterally, making it unlikely that the reduction in specific force was due to fiber type differences. PT stiffness (Figure 2) and YM were reduced in the affected leg by 48% and 43%, respectively, relative to the unaffected leg. There was no difference in PT CSA.

Figure 1.

Quadriceps isometric force, anatomic cross-sectional area (ACSA), and muscle-specific force in the affected and unaffected legs at baseline and at 1-year followup in the 53-year-old female rheumatoid arthritis participant.

Figure 2.

Patellar tendon stiffness in the affected and unaffected legs at baseline and at 1-year followup in the 53-year-old female rheumatoid arthritis participant.

Reassessment at 1 year

Physical function had improved substantially (sit-to-stand by 30%, 8-foot up and go by 8%, 50-foot walk by 16%, and one-leg standing balance by 18%). For example, she now performed 13 sit-to-stands equivalent to the 35th percentile of a healthy 60–64-year-old female reference population (11).

In the affected leg, isometric force had improved by 36.8%, concentric torque by 42.7%, and ACSA by 9%, while PT stiffness (Figure 2) and YM had declined a further 10% and 29%, respectively.

In the unaffected leg, PT stiffness (Figure 2) had deteriorated by 27% and YM by 50%, and there were minor reductions in isometric force (−3.6%) and quadriceps ACSA (−5.1%) relative to baseline (Table 1).


This case demonstrates the dramatic consequences of an acute inflammatory knee effusion on the physiology of the local tendon–muscle complex, and confirms the known adverse effect of a joint effusion on physical function. In particular, the presence of the knee effusion greatly diminished the participant's sit-to-stand test performance, in spite of her very active lifestyle. Her function improved accordingly with resolution of the effusion a year later. Specific electrophysiologic examination showed an adverse effect on PT stiffness. Acutely this was only seen in the affected leg, but at 1 year was evident in both legs. The acute changes could be a local effect of the joint effusion, whereas the more long-term bilateral changes suggest a systemic effect of the inflammatory process. Our results also confirm that the presence of effusion results in quadriceps wasting, a dramatic loss of force production that is not due to pain, and impaired muscle quality as shown by reduced muscle-specific force.

Tendons are extensible structures that reversibly deform when a mechanical load is applied. The content and organization of the extracellular matrix (collagen, elastin, water) enable the tendons to store elastic strain energy and return it on recoil, a unique mechanical property that is essential for locomotion and complex functional tasks (9). The extent of elongation of a tendon to loading, i.e., the tendon stiffness, influences the performance of the attached muscle, and thereby determines the magnitude and speed with which force is transmitted from muscle to bone. Therefore, stiffer tendons result in increased and faster force production, whereas the opposite effect is seen in more compliant tendons since increased elongation of the tendon requires further shortening of the muscle fibers causing electromechanical delay (5). Outside a characteristic elastic range, tendons become less efficient for muscle output and motor control, and more vulnerable to injury (9). Tendon stiffness is reduced in aging and following immobilization due to a concomitant decrease in collagen and an increase in the more extensible elastin content (5). In contrast, increases in tendon stiffness follow exercise training (5).

Our case is, we believe, the first to show reduction of tendon stiffness in acute inflammatory joint disease. In patients with stable established RA in a previous study, we also observed reduced tendon stiffness compared with controls of similar physical fitness (Matschke V et al: unpublished observations). RA is known to cause synovial inflammation and hypertrophy of tendon sheaths and on occasion, synovial tissue infiltration of the tendon itself (12). However, the PT does not have a tendon sheath. Whereas US features of PT enthesitis such as PT thickening and increased vascularity have been described in RA patients even in the absence of clinical symptoms of pain, stiffness, or clinically detectable swelling of the tendon (13), in our study comparing stable RA patients with healthy controls we did not observe PT thickening (Matschke V et al: unpublished observations). Similarly, in this case study, no PT thickening was observed. The reduced PT stiffness in the affected leg at baseline could be a result of mechanical effects of the knee effusion or a detraining effect with the attenuated muscle force leading to reduced tendon loading. Local diffusion of inflammatory cells and cytokines from knee joint synovitis may have contributed to the adverse effects in the tendon. Interestingly, 1 year later, despite recovery of the muscle properties and continual physical activity, there was no recovery of tendon stiffness. Furthermore, the tendon stiffness in the unaffected leg was also reduced. This strongly suggests a systemic effect of RA on the tendon.

Our findings of reduced muscle force and size in the affected leg with a dramatic reduction in physical function are in accord with other studies on the effects of pathologic and simulated knee effusions on quadriceps and EMG activity. In most reports, the presence of an effusion resulted in reduced strength and EMG activity when compared to the same joint with no effusion. The EMG changes indicate that quadriceps muscle reflex inhibition causes the reduction in strength. In chronic effusions, strength is also affected by muscle wasting (14). In our participant, force was diminished more than muscle size, resulting in reduced muscle-specific force. This phenomenon is also seen with immobilization and aging, where changes in muscle architecture, activation, and fiber type are featured (6, 10). Exercise training initially increases strength, followed by muscle hypertrophy (5). Therefore, muscle-specific force rises more quickly during the early stages of exercise training. In our case, after treatment of the effusion and continued physical activity, the quadriceps improved more in strength than in size. We have previously shown that muscle-specific force is unimpaired in patients with stable established RA (7, 8), but it might have been reduced in the acute stage of the disease and subsequently recovered. In the literature, extremity dominance does not appear to play a role in differences of lower leg strength in healthy individuals (15).

Presumably, the difference in muscle size and force between the affected and unaffected leg in our participant is a result of excess synovial fluid causing mechanical distension of the knee joint. Rheumatoid cachexia might also play a role, but cannot be commented upon in this case in the absence of information about body composition before disease onset. The corticosteroid injection soon after the baseline tests resolved the joint effusion in our participant, and thereby improved the mechanical conditions for recovery of the tendon–muscle complex. However, there are also reports on adverse effects of local corticosteroid injections on tendon strength, but this may be transient (16).

The findings of this case study highlight several important points. First, a knee joint effusion results in dramatic physical impairment. Second, early local and subsequent systemic effects of RA reduce PT stiffness. Third, in addition to loss of muscle quantity, there is impaired muscle quality in acute RA, a characteristic not evident in stable RA (7, 8). Fourth, resolution of disease activity and continued physical activity led to partial recovery of muscle and physical function, but did not resolve the tendon abnormalities. Further areas of research suggested by this study include clarification of the role tendon abnormalities play in the impaired physical function of RA, the effect of tendon-focused exercise training, and the consequences of medication on rheumatoid tendons. Finally, this case highlights the adverse effect of an inflammatory joint effusion on the tendon–muscle complex and the subsequent need for early intervention, including joint aspiration, steroid injection, and continuing physical activity, to minimize disability in these patients.


All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Matschke had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Matschke, Thom, Lemmey, Maddison, Jones.

Acquisition of data. Matschke, Thom.

Analysis and interpretation of data. Matschke, Thom, Lemmey, Maddison, Jones.