RELATIONSHIPS BETWEEN TENDON STRUCTURE AND CLINICAL IMPAIRMENTS IN PATIENTS WITH PATELLAR TENDINOPATHY

Abstract

The clinical relevance of altered tendon structure in patellar tendinopathy is contested since structural change persists after symptom resolution. The purpose of this study was to explore the relationships between tendon structure and clinical impairments in patellar tendinopathy. In this retrospective, secondary analysis of individuals with patellar tendinopathy (n = 41), tendon structure (thickness, cross-sectional area [CSA], shear modulus, and viscosity), symptom severity, lower extremity function (counter-movement jump [CMJ] height), and quadriceps muscle performance (knee extension force and central activation ratio [CAR]) were recorded for the symptomatic limb. Relationships among structure, symptom severity, lower extremity function, and quadriceps muscle performance were examined using sequential regression models. Adjusting for age, sex, BMI, and pain levels, there were significant positive relationships for thickness (p<0.001, β=0.718) and viscosity (p=0.006, β=0.496) with CMJ height. There were significant negative relationships between CSA with both CMJ height (p=0.001, β= −0.538) and CAR (p=0.04, β= −0.517). This is the first study to demonstrate relationships between tendon structure and lower extremity function or quadriceps muscle performance in patients with patellar tendinopathy.

INTRODUCTION

The cardinal symptom of patellar tendinopathy is load-dependent patellar tendon pain^1,2. However, patellar tendinopathy is also accompanied by numerous other changes and impairments. Individuals with patellar tendinopathy have decreased self-reported lower extremity function^3–5, impaired muscle performance^6–8, and altered tendon structure^9–14. Tendon structure can be subdivided into two categories, morphology and mechanical properties. Altered morphology, or tendinosis, is typically assessed using B-mode ultrasound imaging and is characterized by localized tendon thickening, irregular fiber alignment and hypoechoic regions⁹. The degree and direction of change in mechanical properties in patellar tendinopathy is unclear¹⁴. However, most current evidence suggests that mechanical properties are altered in patellar tendinopathy^10–14, either as a direct result of the injury, reduced physical activity, or a combination of both¹⁵.

The clinical relevance of structural changes in patients with patellar tendinopathy has been extensively debated^16–18. Those that argue that structural changes are of little clinical importance do so based on several reasons. First, morphological changes are common in active individuals¹⁹ and up to 79% of those with abnormalities do not develop symptoms at short- or long-term follow-ups¹⁶. Second, symptoms typically resolve prior to changes in morphology or mechanical properties²⁰. Third, there is not a clear relationship between clinical improvements and normalization of tendon structure¹⁸. However, altered tendon structure is a risk factor for developing symptoms¹⁶. Additionally, the treatment with the highest level of evidence for patellar tendinopathy is exercise therapy²¹. The mechanism of action for exercises therapy is mechanotransduction, or the process by which forces applied to the tendon trigger cellular responses that result in structural remodeling²². So, those that argue against the clinical importance of structural changes typically utilize treatment protocols that are designed to address alterations in tendon structure.

Several studies have attempted to examine whether relationship exist between patellar tendon structure and clinical impairments^{6,10,16,23–32}. However, the presence of such relationships, their strength and/or direction varies between studies, so the importance of alterations in patellar tendon structure remains unclear. An improved understanding of the relationship between patellar tendon structure and clinical impairments may help explain the presence of symptoms, decreased lower extremity function, and altered quadriceps muscle performance. Additionally, this may result in more targeted treatments, which aim to resolve these impairments by addressing the contribution of altered tendon structure. Therefore, the purpose of this study was to explore the relationships between patellar tendon structure and clinical impairments (symptom severity, lower extremity function, and quadriceps muscle performance) in the symptomatic limb in individuals with patellar tendinopathy.

METHODS

Study Design

This study is a retrospective, secondary analysis of cross-sectional data from individuals that participated in two previous studies of patellar tendinopathy (level III evidence). These studies were approved by the University of Delaware Institutional Review Board and participants provided written informed consent prior to study enrollment. All participants included had a clinical diagnosis of patellar tendinopathy defined as 1) pain and stiffness localized to the patellar tendon, 2) recreation of symptoms with palpation to the patellar tendon, and 3) load-dependent symptoms, which increased with demands placed on the patellar tendon². The baseline evaluation was used for each participant, where age, sex, BMI, physical activity level, sport participation, tendon structure, symptom severity, lower extremity function and quadriceps muscle performance were recorded for the symptomatic limb. If participants had bilateral symptoms, the most symptomatic limb, based on self-report, was used. Individuals’ values for tendon structure were excluded if the participant had an invasive procedure to the patellar tendon that may alter tendon structure, such as a graft harvest. Furthermore, values for lower extremity function and quadriceps muscle performance were excluded if participants had another injury that may influence their test results.

Measures

Physical Activity Level

Physical activity level prior to injury was recorded using the Physical Activity Scale (PAS), which is a 6-point ordinal scale³³. Scores range from 1 to 6, with higher scores indicating greater physical activity.

Morphology

Patellar tendon morphology was assessed by B-mode ultrasound imaging using a LOGIC e Ultrasound (GE Healthcare, Chicago, IL) system with a wide-band linear array probe (5.0 – 13.0 MHz) by an unblinded examiner. The participant was positioned in supine with knees flexed to 30° and supported by a bolster. Three extended-field-of-view long axis images were taken at the midline of the tendon from the tibial tuberosity to the inferior pole of the patella and three short axis images were taken 1 cm distal to the inferior pole of the patella³³. A custom MATLAB program was used to obtain maximal tendon thickness from long axis images. (Figure 1). Short axis images were used to measure cross-sectional area (CSA) using Osirix MD imaging software (Pixmeo, Geneva, Switzerland) (Figure 1). The mean value of three images was used for data analysis.

Figure 1.

a) Extended-field-of-view image of maximal patellar tendon thickness and b) short-axis image patellar tendon cross-sectional area measures. Yellow line = patellar tendon borders; Red dashed line = maximal tendon thickness; Tib = tibial tuberosity; Pat = patella; FP = Hoffa’s fat pad.

Mechanical Properties

Tendon mechanical properties were evaluated using continuous shear wave elastography (cSWE) to obtain values of static shear modulus and viscosity. cSWE was performed using a SonixMDP Q+ (Ultrasonix, Vancouver, BC, Canada) ultrasound scanner with an L14-5/38 probe³⁴. Participants were seated on an adjustable plinth at 90° of hip and knee flexion, with the back supported and lower legs stabilized. Skin markings were placed 1 cm distal to the inferior pole of the patella, along the imaginary line connecting the inferior pole and the tibial tuberosity. The ultrasound transducer was positioned over the 1 cm mark, aligned with the long axis of the tendon and clamped in an adjustable 3-prong clamp (Figure 2). Shear waves were generated at 11 ascending frequencies (322, 339, 358, 379, 402, 420, 460, 495, 536, 585, 643 Hz) using a Minishaker Type 4810 (Bruel and Kjaer, Norcross, GA, USA) placed on the quadriceps tendon. The resultant linear displacement of the tendon was recorded by the ultrasound probe at a framerate of 6438 frames/sec. Three trials were performed and a custom MATLAB code was used in post-processing to obtain mechanical properties, as described by Cortes et al and Corrigan et al^34,35. The average of three trials was used for data analysis.

Figure 2.

Set-up for cSWE.

Symptom Severity

Symptom severity was assessed using the Victorian Institute of Sport Assessment – Patellar Tendon (VISA-P) questionnaire³³. The VISA-P is a patient reported outcome measure designed to assess symptom severity in individuals with patellar tendinopathy. Scores ranges from 0 to 100 with lower scores indicating more severe symptoms.

Lower Extremity Function

Lower extremity function was evaluated using the counter-movement jump (CMJ) test³³. Participants began on flat ground, standing on their symptomatic leg with their arms behind their back (Figure 3). They were instructed to quickly bend their leg and then jump as high as possible, landing on the same leg. Three trials were performed and participants were asked to rate their pain during the activity on the numeric pain rating scale (NPRS) (0 = no pain, 10 = worst pain imaginable). Flight time was recorded for each trial using a light mat (MuscleLab®, Ergotest Innovations, Stathelle, Norway) and was used to estimate jump height. The average of three trials was used for data analysis.

Figure 3.

Starting position for the CMJ test.

Quadriceps Muscle Performance

Knee extension strength and quadriceps muscle activation were assessed using the burst-superimposition technique on a KinCom dynamometer (Model 50 H, Isokinetic International, Chattanooga, TN, USA)³³. Participants were positioned in 90° of hip flexion and 60° of knee flexion. Self-adhesive electrodes were placed over the vastus medialis and vastus lateralis muscle bellies (Figure 4). After a standardized warm-up and familiarization with procedures, the participants performed a 5-second maximum voluntary isometric contraction (MVIC). During the MVIC, a supramaximal, 10-pulse (600 μs, 130 V, 100 pulses per second) train of electrical stimulation was delivered to the muscle using an electrical stimulator (Grass Technologies, Champaign, IL). If the participant was unable to activate the quadriceps fully, testing was repeated up to 4 times, with 3 minutes rest between trials. The maximal voluntary force and the force attributable to the electrical stimulation was recorded (Figure 4). After each trial, participants were asked to rate their pain on the NPRS. The best trial, based on force production and visual inspection of the force production graph, was selected to calculate quadriceps central activation ratios (CAR = [MVIC force/burst augmented force] × 100%) (Figure 4). The CAR is a measure of quadriceps inhibition, where lower values indicate more quadriceps inhibition.

Figure 4.

a) Set-up for knee extension strength and activation testing. b) Force production trace with increase in force at onset of burst superimposition, characteristic of quadriceps inhibition.

Statistical Analysis

Statistical analysis was performed using RStudio version 1.2.5033. Descriptive statistics were calculated for participant demographics and clinical impairments for the full sample, male and females, and those with unilateral and bilateral symptoms. Mann-Whitney U tests were used to examine differences between males and females and for comparisons between those with unilateral and bilateral symptoms. The relationships between tendon structure and symptom severity, lower extremity function and quadriceps muscle performance were examined using regression.³⁶ Two sequential regression models were performed for each of our primary outcomes, VISA-P, CMJ height, MVIC force, and CAR. All models included covariates to adjust for, age, sex, and BMI, in the first block. For measures of lower extremity function and muscle performance, the respective pain levels recorded during testing were also entered as a covariate. The first model included measures of tendon morphology (thickness and CSA) and the second model included measures of tendon mechanical properties (shear modulus and viscosity) in the second blocks. The ΔR² between blocks was tested to see if tendon morphology or mechanical properties were significantly related to our outcomes after adjusting for covariates. Assumptions were tested, using Shapiro-Wilk’s test for normality, data was screened for outliers and extreme cases using visual inspection, Cook’s distance, and leverage plots. Alpha level was set at 0.05 for all analysis.

RESULTS

Participants

Forty-one individuals were included in the analysis. Descriptive statistics are displayed in Table 1 for the full sample, males and females, and unilateral and bilateral symptoms. Other than males jumping significantly higher than females in the CMJ there were no significant differences between sexes or those with unilateral and bilateral symptoms. Twenty of the participants (49%) reported participation in a jumping sport.

Table 1:

Descriptive statistics for the full sample, males and females, and unilateral and bilateral symptoms.

	Total (n = 41) Median (IQR)	Males (n = 26) Median (IQR)	Females (n = 15) Median (IQR)	Unilateral (n = 28) Median (IQR)	Bilateral (n = 13) Median (IQR)
Age (years)	27.0 (13.0)	29.0 (10.3)	24.0 (8.0)	27.5 (11.0)	25.0 (11.0)
BMI (kg/m 2 )	24.7 (5.1)	25.8 (4.6)	23.0 (3.7)	24.9 (4.7)	23.0 (6.4)
PAS (points)	6.0 (1.0)	6.0 (1.0)	6.0 (1.0)	6.0 (1.0)	6.0 (1.0)
Symptom Duration (mo)	7.8 (26.3)	9.8 (34.5)	7.8 (20.3)	6.7 (23.9)	15.8 (54.4)
VISA-P (points)	60.0 (24.0)	60.5 (24.0)	55.0 (16.0)	61.5 (18.8)	55.0 (29.0)
CMJ Height (cm)	12.0 (6.6) n = 35	13.2 (4.6) n = 24	7.7 (3.8)* n = 11	12.1 (7.0) n = 24	11.4 (6.1) n = 11
CMJ Pain (NPRS)	2.0 (4.0) n = 35	2.0 (4.0) n = 24	2.5 (2.8) n = 11	2.0 (4.0) n = 24	3.0 (2.3) n = 11
MVIC Force (N)	839.6 (358.1) n = 22	799.6 (394.8) n = 16	869.8 (238.7) n = 6	758.2 (386) n = 15	841.0 (319.8) n = 7
CAR (%)	84.6 (24.5) n = 22	80.3 (29.5) n = 16	89.4 (8.4) n = 6	84.0 (33.7) n = 15	86.6 (10.9) n = 7
Burst Pain (NPRS)	0.0 (2.3) n = 22	2.0 (3.0) n = 16	0.0 (0.0) n = 6	0.0 (0.0) n = 15	0.0 (5.0) n = 7
Tendon Thickness (mm)	6.3 (3.3) n = 39	6.9 (2.6) n = 25	5.3 (3.2) n = 14	6.9 (3.3) n = 26	5.8 (3.3) n = 13
Tendon CSA (mm 2 )	106.0 (51.3) n = 38	111.9 (39.1) n = 24	94.0 (70.0) n = 14	103.2 (58.1) n = 25	120.0 (34.7) n = 13
Shear Modulus (kPa)	68.3 (23.2) n = 35	70.2 (21.4) n = 24	59.2 (22.6) n = 11	70.2 (17.0) n = 24	55.5 (22.4) n = 11
Viscosity (Pa*s)	28.2 (12.8) n = 35	28.9 (12.9) n = 24	24.3 (9.5) n = 11	28.9 (12.0) n = 24	22.7 (12.4) n = 11

Sample size provided (n) when values were not available for the entire group. BMI = body mass index; CMJ = counter movement jump; NPRS = numeric pain rating scale; MVIC = maximal voluntary isometric contraction; CSA = cross-sectional area.

^*= significantly different (p<0.05) between males and females.

Symptom Severity and Tendon Properties

In the initial analysis, two outliers were identified for the relationship between symptom severity and tendon mechanical properties. After removing these cases, assumptions of normality were met for all analyses. No significant relationship was found between VISA-P scores and tendon morphology (p = 0.81, ΔR² change = 0.012) or mechanical properties (p = 0.19, ΔR² = 0.099) after adjusting for age, sex and BMI (Supplementary materials).

Lower Extremity Function and Tendon Properties

Two outliers were identified for the relationship between CMJ height and tendon morphology. After removing these cases, assumptions of normality were met for all analyses. There was a significant positive relationship between CMJ height and tendon thickness (p < 0.001, β = 0.718) and a significant negative relationship between CMJ height and CSA (p = 0.001, β = −0.538), after adjusting for age, sex, BMI, and pain levels (Table 2). There was a significant positive relationship between CMJ height and viscosity (p = 0.006, β = 0.496) and no significant relationship between CMJ height and shear modulus (p = 0.09, β = −0.278) (Table 3).

Table 2:

Regression analysis for counter-movement jump height with morphology as predictor variables after removal of outliers.

	Model 1				Model 2
Variable	b	SE B	β	p-value	b	SE B	β	p-value
Age	−0.147	0.057	−0.369	0.016	−0.093	0.044	−0.234	0.044
Sex	−5.215	1.138	−0.595	<0.001	−3.813	0.881	−0.435	<0.001
BMI	−0.257	0.136	−0.277	0.070	−0.276	0.102	−0.297	0.012
Pain	0.538	0.277	0.246	0.062	0.874	0.231	0.400	0.001
Thickness					14.273	2.847	0.718	<0.001
CSA					−5.360	1.477	−0.538	0.001
R 2	0.575				0.788
F for change in R 2	9.137				12.563

BMI = body mass index; CSA = cross-sectional area.

Table 3:

Regression analysis for counter-movement jump height with mechanical properties as predictor variables.

	Model 1				Model 2
Variable	b	SE B	β	p-value	b	SE B	β	p-value
Age	−0.131	0.082	−0.292	0.123	−0.097	0.074	−0.217	0.204
Sex	−5.824	1.741	−0.571	0.003	−6.150	1.550	−0.603	0.001
BMI	−0.33895	0.204	−0.336	0.109	−0.477	0.186	−0.473	0.017
Pain	0.343	0.355	0.155	0.343	0.301	0.313	0.136	0.347
Shear Modulus					−0.070	0.040	−0.278	0.094
Viscosity					0.253	0.083	0.496	0.006
R 2	0.382				0.556
F for change in R 2	4.018				4.701

BMI = body mass index.

Quadriceps Muscle Performance and Tendon Properties

Three outliers were identified for the relationship between MVIC force and tendon morphology, two for the relationship between MVIC force and mechanical properties, and four for the relationship between CAR and tendon morphology. After removing these cases, assumptions were met for all analyses. There was no relationship between MVIC force and tendon morphology (p = 0.27, ΔR² = 0.11) or mechanical properties (p = 0.24, ΔR² = 0.12), after adjusting for age, sex, BMI and pain levels (Supplementary materials). There was a significant negative relationship between CAR and tendon CSA (p = 0.04, β = −0.517) but no significant relationship between CAR and tendon thickness (p = 0.38, β = 0.214) (Table 4). No significant relationship was found between CAR and mechanical properties (p = 0.83, ΔR² = 0.01) (Supplementary materials).

Table 4:

Regression analysis for CAR with morphology as predictor variables after removal of outliers.

	Model 1				Model 2
Variable	b	SE B	β	p-value	b	SE B	β	p-value
Age	−0.559	0.372	−0.250	0.159	−0.208	0.358	−0.093	0.573
Sex	11.806	5.879	0.369	0.068	18.289	6.079	0.571	0.013
BMI	−2.547	1.218	−0.385	0.059	−0.908	1.546	−0.137	0.570
Pain	−4.416	1.054	−0.705	0.001	−5.157	1.065	−0.823	0.001
Thickness					20.687	22.429	0.214	0.378
CSA					−23.570	9.853	−0.517	0.038
R 2	0.683				0.798
F for change in R 2	6.455				2.863

CAR = central activation ratio; BMI = body mass index; CSA = cross-sectional area.

Summary tables for regression analyses prior to outlier removal are provided in the supplementary materials.

DISCUSSION

In this study, measures of tendon structure were found to relate to lower extremity function and muscle performance, but not symptom severity, in the symptomatic limb after adjusting for age, sex, BMI and activity provoked pain-levels. This is first study to examine relationships between tendon structure and lower extremity function or quadriceps muscle activation in patients with patellar tendinopathy. There was a positive relationship between viscosity and CMJ height, indicating that individuals with lower viscosity had worse jumping performance. There were also significant relationships between maximum tendon thickness and CSA and CMJ height, although the direction of these relationships was opposite. Individuals with thinner tendons and larger CSA had worse jump performance. Furthermore, there was a negative relationship between CSA and quadriceps CAR, suggesting that individuals with larger CSA have decreased quadriceps activation. All other relationships did not achieve significance.

Symptom Severity

We found no relationship between tendon structure and symptom severity. Prior studies utilizing elastography to assess measures of tendon stiffness report that stiffness may positively^10,37 or negatively³⁸ relate to symptom severity. In these previous studies, various methods to measure symptom severity were used including VISA-P scores, pain-pressure threshold, and pain during loading. In addition, the authors did not control for potential confounding factors, such as age, sex or BMI, which might explain the varied results. Although prior studies are inconclusive regarding the relationship between tendon structure and symptom severity, our findings support those that found no relationship.

Lower Extremity Function

Our study is the first study to examine the relationship between lower extremity function and tendon structure in individuals with patellar tendinopathy. Our results indicate that viscosity is positively related to CMJ height. The CMJ is a ballistic movement, requiring rapid force transmission from muscle to bone. Thus, a positive relationship between viscosity and CMJ height is expected, as a more viscous tendon will deform less and provide more efficient force transfer³⁹. Tendon pathology is accompanied by an increase in proteoglycans, hypervascularization, and increased water content²², which may reduce viscosity and result in less efficient force transfer and decreased lower extremity function in patients with patellar tendinopathy.

Interestingly, we observed that tendon thickness and CSA had opposing relationships with CMJ height. Tendon thickness was positively related while CSA was negatively related. A potential explanation for this discrepancy is that tendinopathy induced changes in tendon size are not always uniform,⁴⁰ while changes due to loading typically are. Therefore, CSA may be better able to capture tendinopathy induced changes in tendon morphology, such as increased water content and collagen fiber disorganization, while patellar tendon thickness may be more of a reflection of loading history. Figure 5 provides a representative image of a tendon in a participant with patellar tendinopathy. This participant had focal thickening in the medial portion of the tendon, which did not align with the location of thickness measures. In our study, the cross-sectional area was 22% larger on average in the symptomatic limb compared to the asymptomatic limb while tendon thickness was 17% larger. However, it is unclear to what degree these structural changes were present prior to injury. Had tendon thickness been obtained at the point of maximal thickness, rather than using a standardized, central location, the relationships between CMJ height and thickness and CSA may have been more similar. Therefore, further investigation is needed to determine whether the magnitude of morphological changes due to patellar tendinopathy are similar for thickness and CSA.

Figure 5.

Short axis images of an (a) uninjured and (b) injured patellar tendon in a collegiate athlete. Medial focal thickening is present in the injured tendon. The yellow line represents the CSA measurement and the red dashed line represents the approximate location where thickness measures are taken on long axis images.

One prior study has examined relationships between tendon structure and jumping performance in a healthy cohort of competitive soccer players.²⁶ In this study, lower tendon stiffness was associated with better horizontal, but not vertical, jumping ability, whereas a larger CSA was associated with better horizontal and vertical jumping ability. These relationships are the inverse of those seen in our study. However, our results are not directly comparable since the prior used uninjured individuals and tendon structure is altered by patellar tendinopathy.^9–14Relationships between tendon structure and drop CMJ height were also tested but not significant, so they were removed from the manuscript (supplementary materials). It is unclear why relationships like those of the CMJ were not observed. Anecdotally, participants have reported more apprehension when performing the drop CMJ, since it involves dropping off a box. This may influence their performance and presence of such relationships.

Prior studies of jumping performance have found that individuals with patellar tendinopathy have better jumping performance than asymptomatic peers, which has been deemed the “Jumper’s Knee Paradox”⁴¹. This calls into question whether lower extremity function is impaired in patellar tendinopathy. However, these studies utilized bilateral jumping tasks, which may not capture unilateral impairments⁴¹. Additionally, comparison to an asymptomatic cohort may not be appropriate to determine if lower extremity function is impaired, since jumping ability is a risk factor for patellar tendinopathy^41,42.

Quadriceps Muscle Performance

We found a negative relationship between patellar tendon CSA and quadriceps CAR, indicating that greater CSA is associated with decreased quadriceps activation. This is the first study to examine relationships between patellar tendon structure and quadriceps activation but not the first study to observe altered quadriceps activity in the presence of patellar tendinopathy. Rio et al. found that isometric quadriceps contractions resulted in an immediate reduction in pain, decreased inhibitory neural drive to the quadriceps, and increased knee extension MVIC⁷. In a follow-up study, the same research group found that individuals with patellar tendinopathy have elevated corticospinal excitability to the quadriceps, which may impair their ability to match quadriceps activation to task requirements⁶. Additionally, this altered excitability was not present in patients with anterior knee, or patellofemoral pain. Therefore, pain is unlikely to be the sole explanation for these changes in quadriceps activity. Furthermore, similar associations between altered tendon structure and muscle activity have been found in other forms of tendinopathy. Chang and Kulig investigated the impact of altered Achilles tendon structure without pain, or Achilles tendinosis, on triceps surae activity⁴³. They found that these individuals had decreased activity in the triceps surae of the involved limb compared to the uninvolved limb during a hopping task.

It should also be noted that there was some evidence suggesting a relationship between tendon morphology and mechanical properties with MVIC force, although we were not adequately powered to reach significance. Based on the observed changes in R², the effect sizes were medium. In healthy individuals, increased tendon size and stiffness are positively related to quadriceps strength^27,44. Our findings are similar for shear modulus but the opposite for viscosity, thickness and CSA. However, further investigation with a larger sample size is needed.

Adjusting for Covariates

We chose to adjust for age, sex, and BMI, as well as activity provoked pain for select outcome measures, in our analysis due to their potential influence on measures of tendon structure and clinical impairments. Age is associated with a decline in muscle strength and recruitment, which may influence results of jump tests and quadriceps muscle performance^45,46. Males have greater knee extension strength⁴⁷, patellar tendon thickness⁴⁸ and CSA⁴⁹, and jump height compared to females⁵⁰, even when accounting for training history and muscle size. There is also conflicting evidence in uninjured tendons that males have greater patellar tendon stiffness than females^48,49. BMI has been shown to relate to measures of tendon morphology and patellar tendon stiffness²⁷. Finally, induced knee pain has been shown to reduce quadriceps strength and activation⁵¹. Since pain potentially influences knee extension strength and activation, it also likely influences measures of lower extremity function.

Limitations

One limitation of the study was that symptom duration was not used as a covariate, nor were the models run separately for acute (symptom duration < 3 months) and chronic tendinopathies (symptom duration ≥ 3 months). Tran et al. recently examined the early phases of patellar tendinopathy (symptom duration < 3 months) and found that morphology, but not mechanical properties, were altered during this acute phase.⁵² Therefore, the inclusion of acute patellar tendinopathies may alter the relationships observed for mechanical properties. In our study, 12 out of 41 participants (29%) had a symptom duration of less than 3 months. However, in a post-hoc analysis, there were no significant relationships between symptom duration and tendon structure (supplementary materials). Additionally, morphology and mechanical properties were not significantly different between participants with acute or chronic symptoms (supplementary materials). Thus, we believe it is unlikely that symptom duration had a large influence on the results of this study.

Additionally, there are other limitations to this study that should be considered when interpreting results. First, this study had a relatively small sample size. Therefore, we may not have been adequately powered to detect all relationships between tendon structure and the outcome variables of interest. However, in a post-hoc analysis we assessed the effect sizes of non-significant models to determine if a relationship may exist in a larger population. Except for knee extension MVIC, effect sizes were small or negligible. Second, there was a relatively large proportion of missing data, especially for measures of quadriceps muscle performance. Knee extension strength and quadriceps activation testing were added to the study after initiation so early participants did not complete these tests. For other measures, the missing data was due to a combination of equipment failure or exclusion criteria. Since some data was intentionally removed, it was not appropriate to utilize multiple imputation, which assumes that the data is missing at random³⁶. Third, participants were not limited to chronic tendinopathy (>3 months symptom duration) and changes in patellar tendon morphology and mechanical properties are marginal during the first few months after injury onset⁵². Fourth, imaging was not used to rule out the presence of a partial tendon tear, which may not have the same structural alterations as patellar tendinopathy. However, all participants met the clinical diagnostic criteria for patellar tendinopathy. Finally, this study lacked a healthy cohort for comparison and is cross-sectional, therefore we cannot determine whether the observed relationships were solely due to the presence of patellar tendinopathy.

CONCLUSIONS

In this exploratory study, we found that tendon structure influences lower extremity function and quadriceps muscle performance, even after adjusting for age, sex, BMI and activity-provoked pain, in the symptomatic limb. Since alterations in tendon structure persist after symptom resolution, monitoring and addressing these changes may play a role in optimizing outcomes of patellar tendinopathy treatment.

Statement of Clinical Significance:

Since structural changes persist after symptom resolution; addressing these changes may assist in restoring lower extremity function and quadriceps muscle performance.