Article Text

Patients with different patellofemoral disorders display a distinct ligament stiffness pattern under instrumented stress testing
  1. Ana Leal1,2,
  2. Renato Andrade2,3,4,
  3. Betina Hinckel5,
  4. Marc Tompkins6,7,
  5. Ricardo Bastos2,3,8,
  6. Paulo Flores1,
  7. Filipe Samuel1,
  8. Joao Espregueira-Mendes2,3,9,10,11,
  9. Elizabeth Arendt6
  1. 1 CMEMS - Center for Micro-ElectroMechanical Systems, University of Minho, Campus Azurém, Guimarães, Portugal
  2. 2 Dom Henrique Research Centre, Porto, Portugal
  3. 3 Clínica do Dragão, Espregueira-Mendes Sports Centre - FIFA Medical Centre of Excellence, Porto, Portugal
  4. 4 Faculty of Sports, University of Porto, Porto, Portugal
  5. 5 Department of Orthopaedic Surgery, William Beaumont Hospital - Royal Oak, Royal Oak, Michigan, USA
  6. 6 Department of Orthopaedic Surgery, University of Minnesota, Minneapolis, Minnesota, USA
  7. 7 TRIA Orthopaedic Center, Bloomington, Minnesota, USA
  8. 8 The Biomechanics Group, Department of Mechanical Engineering, Imperial College London, London, UK
  9. 9 ICVS/3B’s–PT Government Associate Laboratory, Braga/Guimarães, Portugal
  10. 10 3B’s Research Group – Biomaterials, Biodegradables and Biomimetics, University of Minho,Headquarters of the European Institute of Excellence on Tissue Engineeringand Regenerative Medicine, Barco, Guimarães, Portugal
  11. 11 Escola de Medicina, Universidade do Minho, Braga, Portugal
  1. Correspondence to Dr Joao Espregueira-Mendes, Clínica do Dragão, Espregueira-Mendes Sports Centre, Porto 4350-415, Portugal; espregueira{at}


Objective Investigate the patellar force-displacement profile (ligament stiffness) of patellofemoral disorders.

Methods Fifty-two knees from 34 consecutive patients (mean 31.6 years and 53% male) were analysed including 24 knees with patellofemoral pain (PFP), 19 with potential patellofemoral instability (PPI) and 9 with objective patellofemoral instability (OPI). Physical examination, patient-reported outcome measures (Kujala and Lysholm Scores), standard radiography and MRI or CT were performed in all patients. Instrumented stress testing (Porto Patella testing device) concomitantly with imaging (MRI or CT) was performed to calculate ligament stiffness.

Results The force-displacement curves in patients with PPI and OPI displayed a similar pattern, which was different from that of the PFP group. Patients with PPI showed higher ligament stiffness (a higher force was required to displace the patella) than the patients in the OPI group. Patients with OPI had a statistically significant shallower trochlear groove and increased lateral tilt. More than half of the PPI and OPI population presented with at least one classic risk factor (patella alta, trochlear dysplasia, increased quadriceps vector, lateral tilt). In the PPI group, at least two risk factors were found in 37% of patients, whereas at least 33% of patients in the OPI group had three risk factors present. None of the patients presented with all four anatomical risk factors.

Conclusion Patients presenting with patellofemoral instability (PPI and OPI) display similar ligament stiffness patterns (force-displacement curve). Patients with PFP and PPI showed higher ligament stiffness as compared with patients with OPI.

Level of evidence Level V, case series.

  • knee
  • instability
  • CT-scan
  • MRI
  • ligament

Statistics from

What are the new findings?

  • Ligament stiffness patterns under instrumented stress testing can discriminate and characterise patients with patellofemoral disorders.

  • Patients with patellofemoral pain (PFP) and potential patellofemoral instability (PPI) have greater ligament stiffness than patients with objective patellofemoral instability (OPI).

  • Patients with PPI or OPI display a similar force-displacement curves pattern, but different from those of the patients with PFP, which may be related to the anatomical instability risk factors.


Patellofemoral disorders are often present in young, active patients. There is a spectrum of the presentation, and may include patellofemoral pain (PFP), potential patellar instability (PPI) and objective patellar instability (OPI). While anatomical risk factors have been identified,1–8 little is known on how biomechanical factors influence these diverse presentations.9–13

A variety of predisposing risk factors has been shown to contribute to the various patellofemoral disorders. These include bony or soft tissue anatomy (trochlea dysplasia, patella alta, increased quadriceps vector and/or increased lateral tilt) and soft tissue imbalance (ligamentous laxity of the medial patellar stabilisers and tightness of the lateral retinaculum).8 14–16 The patellofemoral medial stabilisers include the medial patellofemoral ligament (MPFL), the medial quadriceps tendon femoral ligament, the medial patellotibial ligament and the medial patellomeniscal ligament.17 Importantly, the abnormal behaviour of these ligaments (due to injury or patient’s anatomy) may lead to abnormal patellofemoral tracking, consequently altering the joint contact forces which can result in patellofemoral instability and/or joint degeneration.18 19

Aiming to optimise non-operative management and surgical outcomes, there is a need to better understand the complex anatomy and pathokinematics of the patellofemoral joint. Physical examination, while playing a pivotal role in assessment of the patellofemoral joint, is limited by intraobserver and interobserver variability between examiners.20 21 Several cadaveric studies have studied the biomechanical properties of the medial ligamentous stabilisers of the patellofemoral joint.13 17 22 23 In vitro studies are limited by a lack of the dynamic elements of the patellofemoral joint, which are difficult to duplicate in the laboratory.

In vivo biomechanical testing in patients may help to better understand the dynamic role of the soft tissues and the active interplay between the various patellofemoral disorders. The use of instrumented examination can objectify, quantify and standardise patellofemoral assessment, while assessing the correlation between the mechanical behaviour of soft tissue restraints with their morphological status.

The aim of this study was to compare the force-displacement profile of three main PF disorders: PFP, PPI and OPI, and observe if there are differences in the stiffness profile between patients with instability (OPI and PPI) and patients with pain (PFP). It is hypothesised that patients with objective (OPI) or potential instability (PPI) will present with lower ligament stiffness than the ones with no instability (PFP).

Materials and methods

All consecutive patients presenting with patellofemoral joint-related complaints, between early January and late March 2016, were evaluated clinically, either by MRI or CT combined with stress instrument testing—the Porto Patella testing device (PPTD). All patients signed informed consent.

Patient selection

Patients were recruited from Clínica do Dragão. The inclusion criteria were patients diagnosed, by history and physical exam, with PFP or patellofemoral instability. The exclusion criteria were: history of knee fracture (including patella), and patients with metabolic or inflammatory disease. Patients were divided into three groups: PFP (n=24; 12 M, 12 F; 32.1 years; 24.6 kg/m2), PPI (n=19; 8 M, 11 F; 30.3 years; 24.9 kg/m2) and OPI (n=9; 5 M, 4 F; 28.3 years; 23.5 kg/m2). OPI was diagnosed by clinical history of patellar dislocation, indicative physical examination (joint swelling or pain at the region of the MPFL, medial patella, or retinaculum) or radiological findings.24 PPI was defined as, when in the absence of dislocation events, the patient presents with classic instability risk factors: patella alta (Caton-Deschamps Index >1.2), trochlear dysplasia (trochlear sulcus angle >140°), increased trochlear tuberosity–trochlear groove (TT-TG) distance (>20 mm) or increased patellar tilt (>20°).3 25 26 PFP was defined by a clinical diagnosis of PFP without the presence of any of the previously indicated risk factors. Clinical diagnosis of PFP was based on history (pain located anterior, peripatellar or retropatellar, worse with stairs, squats and sitting for long periods of time) and physical exam (tenderness in the palpation of patella facets, positive Smillie and Zohlen tests). Diagnosis of PFP was made when most criteria were present, but without the classic risk factors.

Physical examination

All patients were clinically evaluated by a senior orthopaedic surgeon (JE-M) and a physiotherapist blinded to the patient’s condition (RA). The examination included clinical history and patellofemoral physical exam. Physical examination comprised common patellofemoral testing for patellar pain, tracking and mobilisation that are routinely performed at our institution.27 28 All patients filled out the Kujala and Lysholm Scores.29 30

Imaging and PPTD evaluation protocol

Two independent experienced musculoskeletal radiology technicians (blinded to the patient’s condition) performed all imaging procedures, including bilateral standard radiographs, standard MRI or CT, and combined them with instrumented assessment of stress using PPTD. Radiological and imaging protocols were performed as previously described.27 28 The PPTD (figure 1) is a stress instrumented device that is safe and compatible for use within MRI/CT systems. This device is valid and reliable to objectively quantify the patellar position (PP) and displacement after stress testing, providing enhanced accuracy and precision, and less variability than physical examination.27 28

Figure 1

Porto Patella testing device (PPTD) set-up for stress testing within imaging equipment.

The PPTD procedures were performed as detailed in previous studies.27 28 Patients were positioned (with no stress) and a first scan was performed to record the initial patellar position (figure 2A). To obtain the stress images, the patella was stressed on the medial facet using a lateral vector (at 30°) (figure 2B). To compute the stiffness pattern, we applied an incremental force sequence using five different levels of force: 21 N [F1], 42 N [F2], 62 N [F3], 83 N [F4] and 104 N [F5]. Values were previously calculated through pilot test of an incremental force with reference to a safety factor (SF) equivalent to the reported MPFL tensile strength (208 N)31 divided by each of the force values (SF=208/F[0, 1, 2, 3, 4, 5]), which resulted in an SF between 10 and 2, respectively, from F0 to F5, that is, the maximum force applied was still approximately half of normal MPFL tensile strength. We applied the stress gradually and stopped if the patient pain/apprehension threshold was achieved (individual’s subjective SF).23 31

Figure 2

Measurements of the patellar position (PP) with Porto Patella testing device (PPTD) within MRI: (A) At rest (PP_NS). (B) After lateral translation (PP_LT). LT, lateral translation; NS, non-stress.

Measurements protocol

We used axial (20°) and lateral radiographs to measure the trochlear sulcus angle (°) and Caton-Deschamps Index, respectively;14 26 32 and MRI/CT images to measure the TT-TG distance (mm), patellar subluxation distance (mm) and lateral tilt angle (°). The patellar subluxation and tilt were measured during rest and with quadriceps contraction.14 26 32

In the PPTD evaluation, we measured the PP without stress (PP_NS) and after lateral stress (PP_LT) (figure 2). The PP is the distance between two parallel lines that are perpendicular to a line that is tangential to the posterior femoral condyles. One line crosses the deepest point of the TG and the other intersects the most posterior point of the patellar central ridge. The lateral patellar displacement (PP_LT-NS_diff) is measured by calculating the difference between the PP after lateral stress (PP_LT) and patella at rest (PP_NS).27 28

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics V.24.0, considering a significance level of 0.05 for all statistic inference situations. Absolute (n) and relative (%) frequencies were computed for the categorical variables. Continuous variables were tested for outliers and normality (Kolmogorov-Smirnov test). These were described using mean and SD (if normally distributed) or median and IQR otherwise. Since outcome variables followed a non-parametrical distribution, the Wilcoxon signed-rank test was used to test statistical difference between two groups and the Friedman χ2 test between three groups. Spearman’s rank correlation coefficient was applied to calculate the correlations between the PPTD measurements and the variables measured by MRI/CT/radiographs as well as the number of risk factors expressed for each patellofemoral disorder (PFP, PPI, OPI); absolute values for correlations were classified as strong if r>0.7.


Fifty-six knees from 34 consecutive patients were available for analysis in this study (18 male, 16 female; 31.6 years, 24.8 kg/m2), divided into three groups: PFP (n=24; 12 male, 12 female; 32.1 years, range 15–58 years; 24.6 kg/m2), PPI (n=19; 8 male, 11 female; 30.3 years, range 17–44 years; 24.9 kg/m2) and OPI (n=9; 5 male, 4 female; 28.3 years, range 15–36 years; 23.5 kg/m2). Four patients presented with an asymptomatic knee (contralateral knee) and these four asymptomatic knees were excluded from the analysis.

The results from physical examination and functional scores are presented in online supplementary 1. Pain-related variables such as tender facet palpation and Zholen were positive in more than half of the study population across all groups. Functional score medians are similar in intrapathological and interpathological groups, particularly considering the SD margin.

Supplemental material

There were no differences in conventional imaging evaluation between the PFP and PPI groups (table 1). The OPI group had significantly shallower TG and increased lateral tilt. Although subluxation with quadriceps contraction was not statistically different, the value in the OPI group was several times the values in the PFP and PPI groups. The most prevalent risk factor was trochlear dysplasia. More than half of OPI and PPI population presented with one risk factor, while in the PPI there were 37% with two risk factors, and in the OPI group there were 33% with three risk factors. No patients had all four risk factors.

Table 1

Conventional imaging evaluation

Instrumented stress testing measurements are summarised in table 2. Patients with PFP and PPI were able to reach the highest force level [F5] (29% and 42%, respectively), but in the PFP sample the force level [F4] was the most representative (54%). Patients with OPI did not reach F5 and the majority did not exceed F3 (78%). Overall, patients with PPI and OPI had greater PP after LT (PP_LT) than the patients with PFP, in direct proportion to the instability level and for all the different levels of force reached. However, this was not observed for the total lateral patellar displacement (PP_LT-NS_diff), given the fact that patients with OPI have a more lateralised position of the patella at rest which was also reflected on the stiffness values. Statistically significant differences between groups were found for the displacement variables (PP_LT-NS_diff), from F0 to F3. There were some outliers in all three groups.

Table 2

PPTD evaluation—instrumented stress-testing measurements

Figure 3 illustrates the most representative force-displacement curves between patients from each of the pathological groups. The figure shows that patients with PPI and OPI display similar patterns in the force-displacement curves, which are different from the PFP group. However, a higher force was required to displace the patella in the PPI than in the OPI group.

Figure 3

Representative force-displacement curves for PFP (blue squares), PPI (orange dots) and OPI (red triangles) groups. The bold line and dashed lines represent different patients. LT, lateral translation; OPI, optional patellofemoral instability; PFP, patellofemoral pain; PPI, potential patellofemoral instability.

Tilt and tilt with contraction showed the highest number of correlations with PPTD PP and displacements, particularly in the initial stages of lateral displacement (F0-F1). The full report on significant correlations between PPTD measurements and anatomical imaging parameters is given in online supplementary 2.


The most important finding of this study was that patients with PPI and OPI display similar ligament stiffness patterns, and both were different from that of the patients with PFP. Comparing the PPI and OPI groups, a higher force was required to displace the patella in the PPI than in the OPI group.

The tolerance to higher forces was dependent on the severity of symptoms. During the instrumented stress testing only patients with PFP and PPI were able to reach the highest force level (F5=104 n). In the case of patients with OPI, F5 was not tolerated and may be related to pain-derived or fear-derived muscle guarding. F3 (62 n) was clearly predominant as the maximum tolerable force in the OPI group (78%), which is consistent with the trend for the abrupt curve to the slope in the force-displacement assessment. A more tilted and lateralised position of the patella at rest (PT_NS and PP_NS) and a more lateralised patella after stress (PP_LT), for patients with OPI compared with patients with PPI and for patients with PPI compared with patients with PFP, may be related to laxity of the medial restraints. Due to a more lateralised position of the patella at rest (4.8 mm for OPI vs 1.3 mm for PPI), the mean lateral patellar displacement (PP_LT-NS_diff) obtained was not always higher in patients with OPI than in the patients with PPI, which was also reflected in the stiffness values; however, the few patients with OPI that reached F4 obtained the highest maximum displacement (20 mm) across all the groups.

Despite some interpatient variability, the most representative ligament stiffness patterns (force-displacement curves) for each of the patellofemoral disorder groups—represented in figure 3—corroborate the results obtained for level of force applied, lateral patellar displacement obtained (PP_LT-NS_diff) and stiffness values. As would be expected, in the low linear region, the patella is gradually and proportionally displaced laterally as the force is applied (lower stiffness values are predominant in these earlier stages), whereas after surpassing a certain force level (between F2=21 N or F3=42 N, depending on each group), the lateral patellar displacement slows down as the soft tissues get stiffer, especially in the PPI and OPI groups.22 33 Analysing in more detail, both patients with OPI and PPI display the same curve pattern (steep increase close to the final displacement), but with patients with PPI showing higher stiffness than patients with OPI as would be expected if their medial soft tissue stabilisers are in better condition than the OPI group where the medial restraints have presumably been injured. For maximum lateral displacement, values for patients with PPI are closer to the values for patients with PFP because both presumably have intact medial soft tissue stabilisers and therefore can tolerate greater force application than the patients with OPI. These results suggest that the force-displacement curve pattern is directed by the anatomy and the presence of risk factors while the amount of displacement is related to the integrity of the medial patellar restraints.

The most predominant and strongest correlations found were between the initial patellar lateral displacement (F0-F1) and the patellar-tilted position at rest (with or without contraction). It is important to underline that tilt was the most representative risk factor within the studied OPI population. Although none of the previous literature covering patellofemoral instrumented stress testing have studied or report this correlation,34 this corroborates the findings of a more recent study correlating three-dimensional patellar shift and tilt.35 This study showed that, independent of other predisposing factors, there is a natural association between patellar lateral displacement and tilt as a result of the knee extensor mechanism function where the patella naturally inclines laterally (tilt) when simultaneously translating laterally (shift).

Further in vivo studies with the instrumented testing device are warranted to refine the understanding of the role of medial soft tissue stabilisers in patellofemoral biomechanics,9 12 36 to fine-tune diagnostic assessment and to improve the treatment algorithm.37 38

The are some limitations in the study. We acknowledge that this study comprises a small sample size that may increase the risk of type II error. Distal risk factors for PFP (spine, hip and ankle-related) were not assessed and these might have led to heterogeneity in characteristics of patients with PFP and influenced the stiffness pattern in these patients. There were some outliers in all the groups, where none or very slight lateral patellar displacement variation was observed. This could be related to uncontrollable variables such as patient inability to control muscle activation (regardless of the device position holding and immobilisation as well as examiner instructions) or patient pain tolerance. Another limitation was that although the PPTD was adjusted to fit each patient biometric characteristics, there still may be variability in terms of device function for the patellofemoral morphology of individual knees. Pain and apprehension threshold is subjective and varies among patients which can interfere with patient tolerance and the amount of force that could be applied. This limitation could be overcome by applying local anaesthesia, which was not possible in this study.

For further research, in future studies with larger samples, F3 could be set as the maximum target force as it was the most suitable force/load applied during patellofemoral stress testing (reinforcing the result trend in previous studies).27 28 Thus, considering a unique target force level standardises the procedure, reduces study variables, thus reducing variability and testing time. This would be more cost-effective and attractive for research purposes in other research centres, and become practical for daily clinical routine in the future.


Patients presenting with patellofemoral instability (PPI and OPI) display similar ligament stiffness patterns (force-displacement curve). Patients with PFP and PPI showed higher ligament stiffness as compared with patients with OPI.


The authors thank the Clinical Director of SMIC Dragão (Dr Rui Aguiar) and his team for the technical support in the imaging study.


Supplementary material

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors AL, RA, MT, EAA, JE-M designed the study. AL, PF, FS, JE-M developed the PPTD device; AL, JE-M developed the PPTD protocol; AL carried out the PPTD procedure and measurements; BH, MT, EAA, JE-M validated and interpreted the PPTD measurements. AL, RA performed the statistical analysis and drafted the manuscript. BH, RB, MT, EAA, JE-M validated and critically reviewed the scientific content of the manuscript. All authors read and approved the final manuscript.

  • Funding This study was funded by Fundação para a Ciência e a Tecnologia (FCT) and Fundação Luso-Americana para o Desenvolvimento (FLAD), in the form of support given, respectively, through the PhD Studentship in Industry (SFRH/BDE/51821/2012) and internship grant (Proj.16/2015).

  • Competing interests None declared.

  • Patient consent for publication Obtained.

  • Ethics approval The study was approved by the Institutional Review Board (protocol number #013/0015) and informed consent form was obtained from all patients.

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

  • Data availability statement Additional data are available upon reasonable request.

Request Permissions

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