Medial patellofemoral ligament reconstruction using a digital tensiometer to determine graft tension: Surgical technique and mid-term follow-up

Abstract

Purpose

The optimal method for achieving proper graft tension during patellofemoral ligament reconstruction is a topic of debate. In the past, a digital tensiometer was used to simulate the knee structure, and a tension of approximately 2N was identified as suitable for restoring the patellofemoral track. However, it is unclear whether this tension level is sufficient during the actual surgery. The objective of this study was to verify the efficacy of graft tension using a digital tensiometer for medial patellofemoral ligament (MPFL) reconstruction and to conduct a mid-term follow-up.

Methods

The study enrolled 39 patients who had experienced recurrent patellar dislocation. Preoperative computed tomography scans and X-rays confirmed patellar instability, patellar tilt angle patellar congruence angle and the history of dislocation and patellar apprehension test. Knee function was evaluated using preoperative and postoperative Lysholm and Kujala scores.

Results

The study included 39 knees, comprising 22 females and 17 males, with an average age of 21.10 ± 7.26. The patients were followed up for at least 24 months through telephone or face-to-face questionnaires. All patients had a preoperative history of ≥2 patellar dislocations, none of which were surgically treated. During surgery, all patients underwent isolated MPFL reconstruction and lateral retinacula release. The mean Kujala and Lysholm scores were 91.28 ± 4.90 and 90.67 ± 5.15, respectively. The mean PTA and PCA were 11.5 ± 2.63 and 2.38 ± 3.58, respectively. The study found that a tension of approximately 27.39 ± 5.57N (14.3–33.5N) was required to restore the patellofemoral track in patients with recurrent patellar dislocation. No patients required reoperation during the follow-up period. Overall, 36 out of 39 patients (92.31%) reported no pain when completing daily activities at the last follow-up.

Conclusion

In conclusion, a tension level of approximately 27.39 ± 5.57N is necessary to restore normal patellofemoral relationships during clinical practice, which indicates that using a tension of 2N is too low. The use of a tensiometer during patellofemoral ligament reconstruction is a more accurate and reliable surgical procedure for treating recurrent patellar dislocation.

Keywords

Medial patellofemoral ligament reconstruction surgery patellofemoral instability tensiometer patellofemoral track

Introduction

Patellar dislocation is a common orthopedic condition affecting mainly adolescents, where the medial articular surface of the patella loses contact with the femoral sulcus.¹ The medial patellofemoral ligament (MPFL) provides up to 60% of knee stability by restraining lateral dislocation when the knee is flexed from 0° to 60°.^1,2 The stability of the patella also depends on limb alignment, osseous morphometry, and the integrity of soft tissue constraints. Medial patellofemoral ligament tears occur in 96%^2,3 of patellar dislocations, and conservative treatment has a re-dislocation rate of up to 44%, making anatomical MPFL reconstruction the gold standard technique for recurrent patellar dislocation/instability.^3–6

Determining graft tension in clinical work involves direct arthroscopic visualization of the patellofemoral relationship or self-adjustment of the graft by flexion and extension of the knee. ^7–10 However, there is no specific figure to quantify graft tension, leading to subjectivity and experience-based decisions by surgeons, increasing uncertainty in surgery. In 2009, Beck¹¹ et al. simulated knee biomechanics using fresh frozen knee specimens fixed to a device, where a digital tensiometer was attached to the lateral femoral tunnel to determine graft tension, concluding that a pulling force no greater than 2N was necessary. However, this technique has not been applied to surgical procedures in the past decade, and it remains unclear whether a pulling force of 2N restores patellofemoral alignment in clinical work.

Therefore, we performed a follow-up study of patients with recurrent patellar dislocation who had undergone MPFL reconstruction using a digital tensiometer for a minimum of 2 years. We hypothesized that the tension of 2N recommended by the cadaveric study may not be sufficient to restore joint stability to the patellofemoral reconstruction during MPFL reconstruction and that greater tension may be necessary to achieve stability restoration.

Materials and method

The local ethics committee approved this study, and each patient provided informed consent. We included 40 knees from patients with recurrent patellar dislocation between January 2016 and January 2019 who underwent isolated MPFL reconstruction and excluded one patient who was lost during follow-up. Before surgery, patients were informed of the procedure and obtained consent. Inclusion criteria consisted of a history of two preoperative dislocations, MRI and CT scan evidence of patellofemoral ligament tears, and failure of conservative treatment Exclusion criteria:high patella (Insall-Salvati index>1.2)¹², severe trochlear dysplasia (Dejour score>B)¹³, TT-TG distance (tubercle ‐trochlear Grove) >20 mm¹⁴, Q-angle >20°¹⁵, ACL (Anterior Cruciate Ligament) injury, meniscal injury.¹⁶ The study recorded and observed pre-operative and post-operative CT and MRI scans at 30° of knee flexion, as well as the Insall-Salvati index, PCA, PTA, Q angle, and femoral Sulcus angle. Kujala scores and Lysholm scores were collected through face-to-face questionnaires or telephone follow-ups, both pre-and post-operation. Basic patient information can be found in Table 1.

Table 1.

Patient information.

Variable	Value
Sex (M/F), N	17:22
Side (L/R), N	12:27
Age for surgery (year), mean ± SD	21.10 ± 7.26
Height (cm), mean ± SD	171.28 ± 8.89
Weight (kg), mean ± SD	70.95 ± 15.9
BMI, mean ± SD	23.99 ± 4.04
Insall-Salvati index, mean ± SD	1.11 ± 0.10
Sulcus angle (degrees), mean ± SD	131.78 ± 6.99
TT-TG distance (mm), mean ± SD	15.16 ± 2.54
Follow-up,month, mean ± SD	31.54 ± 6.46
Tension (N), mean ± SD	27.39 ± 5.57 N

Surgical procedure

The technique for MPFL reconstruction using a digital tensiometer is shown in Figures 1(a) and 1(b), and it was performed by the same surgical team. Firstly, the patellofemoral joint was examined arthroscopically, and any associated lesions were treated to determine patellofemoral stability. Next, a semitendinosus tendon measuring approximately 12 cm was harvested, and the two free ends were sutured using No. 2 Ethibond suture. The grafts were then covered with 0.9% saline gauze and set aside. An approximately 3 cm incision was made at the mid-superior pole of the medial border of the patella to expose the MPFL patellar checkpoint. Two 2.5 mm guide wires were used to drill parallel bone tunnels along the midpoint and mid-superior 1/3 of the upper and lower poles of the patella, respectively. Using a 4.5 mm hollow drill, a 2 cm deep tunnel was drilled along the guide wire, and the prepared tendon ends were introduced into the hole at the end of the guide pin. The graft caudal line was drawn inwards and outwards, and it was drawn close to the lateral aspect of the patella. Finally, the graft was fixed at the medial edge of the patellar tunnel using a pair of non-resorbable suture anchors (Figure 2(a)).

Figure 1.

(a, b). A schematic diagram of the anatomic reconstruction of the medial patellofemoral ligament with a digital thetensiometer (figures created by the author, no other references used).

Figure 2.

Surgical procedure. (a) Lateral fixation of the patella of the graft using a non-absorbable suture anchor; (b, c, d) variation in graft length observed at 0°, 30° and 60° angles; (e) measurement of graft tension and observation of patella trajectory using a digital tensiometer (f, g) preoperative maximum knee flexion in ortholateral position, (h, i) 4 months post-operative maximal knee flexion in ortholateral position.

To determine the Schottle point, bony landmarks such as the medial epicondyle of the femur were touched, and a 4 cm incision was made to expose the medial condyle and the adductor tuberosity. A 2.0 mm guide wire was used to drill the bone tunnel, and standard lateral fluoroscopy was used to acquire the Schottle point.⁷ The graft was then sent along the gap from the patellar end to the femoral end, and the graft was rounded onto the guide wire. The knee joint was flexed at 0°, 30°, and 60°, and the graft was observed to change in length as the knee joint was flexed and extended, respectively. An isometric reconstruction was defined as a femoral endpoint positioned by the Schottle point, with a change in graft length of less than 2 mm during 0–60° of knee flexion (Figure 2(b)–(d)). A 7 mm hollow drill was used to drill the femoral tunnel in the direction of the guide wire, and the suture was pulled through the hole at the end of the guide wire. The suture was fed to the lateral surface of the distal femur, and the end of the suture was knotted. The appropriate tension was given by pulling the graft with a digital tensiometer from the lateral side when the lateral patellar edge was positioned in line with the lateral trochlear border in 30–45° of flexion under the arthroscope (Figure 2(e)). The digital tensiometer values were recorded, and the arthroscope was withdrawn from the joint cavity. The assistant ensured that the patellofemoral joint was restored to its proper relationship by pulling and maintaining the tensiometer values through the recorded values. The surgeon then used bioresorbable interference screws screwed into the femoral tunnel on the inside of the femur in 45° flexion. Finally, a drainage tube was placed, and the wound was closed in layers.

Postoperative protocol

The drainage tube was removed 24 h after surgery and the patient began quadriceps muscle training. An adjustable knee brace was used to maintain a full extension position. At the 2-weeks mark, the patient was advised to walk with a walking aid. After 4 weeks, walking with crutches was permitted, and the adjustable brace was maintained at 0° for up to 4 weeks. Knee flexion began at 0°–30° after 4 weeks and increased by 30° per week until full flexion. After 6 weeks, the patient could walk on the ground without aids. At 8 weeks, the patient could practice standing on the affected limb alone at home. After 12 weeks, patients were cleared to return to daily life and work, and resume normal exercise, such as running, after 6 months post-surgery. Knee flexion function before and after surgery was recorded and radiographic assessments were conducted before and after surgery, as shown in Figure 2(f)–(i), and Figure 3(a)–(h), respectively.

Figure 3.

3a-b shows the patient’s preoperative MRI scan in front and side flexion, 3c-d shows the patient’s preoperative X-rays, 3E shows the patient’s preoperative CT scan at 30° flexion showing patellofemoral relationship, 3f-g shows the postoperative MRI scan in front and side flexion, 3H shows the patient’s postoperative CT scan at 30° flexion showing patellofemoral relationship.

Clinical assessments

The study protocol involved pre-operative, 3-months post-operative, and final follow-up assessments of knee function using the Kujala and Lysholm scores. The stability of the patellofemoral joint was assessed pre- and post-operatively using the apprehension test. Range of motion and any complications were also evaluated both before and after surgery. Pre-operative and 1-week post-operative imaging, including X-ray, CT or MRI, were conducted to examine the tibial tubercle-trochlear groove distance, PTA, PCA, and the success of the patellofemoral ligament reconstruction. Data collection and collation were carried out by a dedicated technician not involved in the procedure or treatment.

Statistical analysis

The data were analysed using SPSS statistical software (SPSS, IL, USA). Sample data were expressed as mean ± standard deviation ( $\bar{X}$ ± S). A paired samples t-test was performed to analyze pre- and post-operative imaging data and knee function scores that followed a normal distribution. A difference was considered statistically significant if the p-value was less than .01.

Results

All 39 patients completed at least a 2-years follow-up, and no cases of post-operative re-dislocation were observed in any of the patients. Both the angle of concordance and the angle of patellar tilt were significantly reduced before and after surgery (p < .001). At the last follow-up, significant improvements were observed in knee Function Score, Kujala scores, which increased from 37.18 ± 6.68 to 91.28 ± 4.90 (p < .001), and Lysholm scores, which increased from 36.23 ± 6.34 to 90.67 ± 5.15 (p < .001) (Table 2). At postoperative follow-up, 36 patients (92.31%) expressed satisfaction with the results, while three patients (7.6%) reported slight pain at the femoral or patellar endpoint during colder weather. One patient experienced partial prolapse of the lateral patellar transosseous sutures, likely due to early and intense exercise, but the patient did not require special treatment as there was no significant discomfort. Four patients had limited knee flexion and extension of approximately 15°–25° less than the healthy knee 1 year after surgery. The graft tension required to restore a normal patellofemoral relationship was approximately 27.39 ± 5.57N (range 14.3–33.5N). Post-operative complication rates are detailed in Table 3.

Table 2.

Comparison of preoperative and postoperative outcome parameters.

Measurement	Preoperative	Follow-up	p
Kujala score	37.17 ± 6.68	91.28 ± 4.91	<.001
Lysholm score	36.23 ± 6.35	90.67 ± 5.15	<.001
Patellar congruence angle	25.03 ± 5.07	11.50 ± 2.63	<.001
Patellar tilt angle	24.14 ± 5.73	2.38 ± 3.58	<.001

Table 3.

Surgical failure and complications during follow-up.

Redislocation/subluxation, n (%)	0
Apprehension sign, n (%)	0
Flexion deficit and MUA, n (%)	10.3
Pain, n (%)	7.6
Suture anchors breakout, n (%)	2.5

Discussion

The most important finding of this study was that the average tension of the graft during surgery to restore the patellofemoral relationship was 27.39 ± 5.57N (14.3–33.5N), which is significantly higher than the 2N graft tension reported in earlier laboratory cadaveric investigations. The tension applied during surgery was found to effectively restore the normal patellofemoral relationship.

The application of patellofemoral ligament reconstruction as a treatment for patellar instability has emerged as an effective approach.^{2,8,10,17–19,20} However, the management of graft tension remains a contentious issue. Overall, promising clinical outcomes have been reported for MPFL reconstruction utilizing diverse methods for controlling graft tension (as demonstrated in Table 4).

Table 4.

Literature Review about tension control type.


Lead author	No. of patients	Tension control type	Graft	Follow-up, mo	Redislocation Rate, %	Kujala score
Zhang²⁰	29	Arthroscopy-view	GT	37.5	0	89.5
Giuli0³	17	Arthroscopy-view	FLA	64.3	6	86.7
Lippacher⁸	72	Patellar mobility	GT	24.8	2.94	87.5
Kang¹⁰	23	Self-balance	ST	32.7	0	91.4
Niu¹⁸	30	Hemostats fixation	ST	55.1	0	92
Erickson¹⁷	90	Patellar mobility	ST	14.4	4	91.6
Li¹⁹	91	Arthroscopy-view	ST	41.1	0	85.5

GT, gracilis tendon; ST, semitendinosus tendon; FLA, fascia lata allograft.

Achieving appropriate graft tension can be a difficult task that requires the surgeon’s experience and skill in evaluating the tension during the procedure. Previous studies have reported different recommended levels of graft tension for successful MPFL reconstruction, with some suggesting 2N ¹¹ and others suggesting up to 10N.^6,21 However, these studies used cadavers with normal patellofemoral relationships, and we believe that a force of 2N is inadequate for patients with recurrent patellar dislocation and abnormal patellofemoral relationships. In our study, we used a digital tensiometer to measure graft tension and found that an average force of approximately 27N was required to restore the correct patellofemoral track. However, it is important to note that this value is only the average, and we have observed tension ranging from 14.3N to 35.5N in different patients.

Unlike frozen cadaver studies, surgical tension can be affected by various factors, including the patient’s depth of anesthesia, quadriceps muscle strength, lateral retinacular strength, skin and fascia tension, and graft-bone tunnel friction. However, the most significant factor is the tension of the lateral retinacular and the graft-bone tunnel friction. We suggest that if the Medial Patellar Glide (MPG) test indicates less than one quadrant of movement, it indicates lateral retinacular tightness and may require lateral retinacular release. If the MPG test indicates four quadrants of movement, it indicates medial patellar subluxation.²² The diameter of the graft-bone tunnel is kept constant at 1 mm to ensure consistent graft-bone tunnel friction in all patients.

Excessive tensioning of a graft may be more common than currently believed. Surgeons may inadvertently increase graft tension in their efforts to achieve a smaller patellar tilt angle and patellar congruence angle after surgery. However, this can lead to increased pressure on the medial patellofemoral joint, potentially causing the later development of patellofemoral arthritis or chondromalacia patellae in patients. It is important to note that the native MPFL primarily acts as a checkrein and not a tensioner. Clinically, Thaunat and Erasmus²³ have suggested using similar medial and lateral movements to the contralateral patella as a reference for tensioning the graft.

Zumbansen²⁴ et al. utilized extracortical clamping for femoral hybrid fixation to eliminate the influence of graft slippage on measurements. Erickson¹⁷ et al. determined graft tension using medial and lateral patellar movement and range of motion. The others fixed the femoral end, then the patellar end, and arthroscopically adjusted the tension before fixing the patellar end with an extracortical clamp to reduce graft slippage and tension fluctuations.^18,25 However, in clinical practice, the most common method is to observe the patellofemoral joint relationship via arthroscopic knee flexion at 30°, while the lateral femoral assistant pulls the graft tail and adjusts the tension accordingly.^8,9

We advocate the use of a tensiometer to pull the graft, as opposed to manual pulling by an assistant. This approach leads to greater stability and accuracy in graft tension, reduces manual fluctuations, and minimizes the risk of over-loosening or over-tightening the graft during surgery. Tightening the MPFL graft excessively may overload the patellofemoral cartilage, restrict the range of motion, and result in a postoperative loss of knee flexion.^11,12 Conversely, patellar instability may persist, and even lead to recurrent patellar dislocation.⁶

This study offers several advantages. Firstly, we conducted a comprehensive review of surgical modalities, utilized a digital tensiometer for patellofemoral ligament reconstruction during surgery for the first time, and provided a more accurate figure to quantify graft tension. This figure can also help determine whether there is excessive medial patellofemoral pressure by reflecting the magnitude of graft tension laterally. Secondly, the use of a digital tensiometer reduces the risk of human-induced changes in graft tension during surgery, a detail that has been overlooked in previous procedures and can affect outcomes and complications. However, there are some limitations to this study. Firstly, the sample size was small, and the follow-up period was only medium-term. We did not have a reference group because most patients with patellar dislocation have bony structural abnormalities, making it difficult to find suitable participants who meet the exclusion criteria. Therefore, a larger sample size and long-term follow-up are necessary to verify the results. Secondly, due to the COVID-19 pandemic, some patients were followed up via telephone, which may have affected the accuracy of their post-operative knee function score. Thirdly, we excluded patients with congenital developmental abnormalities as they are an important factor affecting prognosis. This exclusion criterion may have limited the generalizability of our findings.

Conclusions

Based on our findings, we propose that the average graft tension required to restore a normal patellofemoral relationship in clinical practice is approximately 27.39 ± 5.57N and that the use of 2N is insufficient. Our use of a tensiometer during patellofemoral ligament reconstruction has proven to be a more accurate and stable surgical procedure for the treatment of recurrent patellar dislocation. We have not encountered a single case of post-operative re-dislocation in at least 2 years of follow-up, indicating the efficacy of this approach. Therefore, we believe that this method should be widely adopted.

Footnotes

Acknowledgements

The authors would like to thank all of the patients for participating in this study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the National Natural Science Foundation of China (No. 8176150186); Key Program of Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2021B03006); Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2022D01C170).

ORCID iD

Wenhao Zhang

References

Yeung

Leblanc

Ayeni

, et al. Indications for medial patellofemoral ligament reconstruction: a systematic review. J Knee Surg 2016; 29(7): 543–554. DOI: 10.1055/s-0035-1564730

Golant

Quach

Rosen

. Medial patellofemoral ligament reconstruction with a looped semitendinosus tendon, using knotless anchor fixation on the patella and hybrid fixation on the femur. Arthrosc Tech 2014; 3(2): e211–e216. DOI: 10.1016/j.eats.2013.10.001

Marcheggiani Muccioli

Lullini

Grassi

, et al. Good results are reported at 60-month follow-up after medial patella-femoral ligament reconstruction with fascia lata allograft for recurrent patellar dislocation. Knee Surg Sports Traumatol Arthrosc 2021; 29(4): 1191–1196. DOI: 10.1007/s00167-020-06142-x

Lorbach

Zumbansen

Kieb

, et al. Medial patellofemoral ligament reconstruction: impact of knee flexion angle during graft fixation on dynamic patellofemoral contact pressure-a biomechanical study. Arthroscopy 2018; 34(4): 1072–1082. DOI: 10.1016/j.arthro.2017.09.047.

Sappey-Marinier

Sonnery-Cottet

O’Loughlin

, et al. Clinical outcomes and predictive factors for failure with isolated mpfl reconstruction for recurrent patellar instability: a series of 211 reconstructions with a minimum follow-up of 3 years. Am J Sports Med 2019; 47(6): 1323–1330. DOI: 10.1177/0363546519838405

Stephen

Kittl

Williams

, et al. Effect of medial patellofemoral ligament reconstruction method on patellofemoral contact pressures and kinematics. Am J Sports Med 2016; 44(5): 1186–1194. DOI: 10.1177/0363546516631736

Schöttle

Schmeling

Rosenstiel

, et al. Radiographic landmarks for femoral tunnel placement in medial patellofemoral ligament reconstruction. Am J Sports Med 2007; 35(5): 801–804. DOI: 10.1177/0363546506296415

Lippacher

Dreyhaupt

Williams

SRM

, et al. Reconstruction of the medial patellofemoral ligament. Am J Sports Med 2014; 42(7): 1661–1668. DOI: 10.1016/j.eats.2017.06.039

Tscholl

Wanivenhaus

Centmaier-Molnar

, et al. Clinical and radiological results after one hundred fifteen MPFL reconstructions with or without tibial tubercle transfer in patients with recurrent patellar dislocation-a mean follow-up of 5.4 years. Int Orthop 2020; 44(2): 301–308. DOI: 10.1007/s00264-019-04441-8

10.

Kang

Wang

Cao

, et al. A prospective randomized trial evaluating two different tensioning techniques for medial patellofemoral ligament reconstruction. The Knee 2016; 23(5): 826–829. DOI: 10.1016/j.knee.2016.02.008

11.

Beck

BrownBrown

NAT

Greis

, et al. Patellofemoral contact pressures and lateral patellar translation after medial patellofemoral ligament reconstruction. Am J Sports Med 2007; 35(9): 1557–1563. DOI: 10.1177/0363546507300872

12.

Insall

Salvati

. Patella position in the normal knee joint. Radiology 1971; 101(1): 101–104. DOI: 10.1148/101.1.101

13.

Dejour

Le Coultre

. Osteotomies in patella-femoral instabilities. Sports Med Arthrosc Rev 2007; 15(1): 39–46. DOI: 10.1097/JSA.0b013e31803035ae

14.

Moezzi

. Editorial commentary: elevated tibial tubercle-trochlear groove distance and patellar instability. Arthroscopy 2016; 32(1): 62. DOI: 10.1016/j.arthro.2015.11.007

15.

Hvid

Andersen

Schmidt

. Chondromalacia patellae: the relation to abnormal patellofemoral joint mechanics. Acta Orthop Scand 1981; 52(6): 661–666. DOI: 10.3109/17453678108992164

16.

Vavalle

Capozzi

. Isolated reconstruction of the medial patellofemoral ligament with autologous quadriceps tendon. J Orthop Traumatol 2016; 17(2): 155–162. DOI: 10.1007/s10195-015-0375-6

17.

Erickson

Nguyen

Gasik

, et al. Isolated medial patellofemoral ligament reconstruction for patellar instability regardless of tibial tubercle-trochlear groove distance and patellar height: outcomes at 1 and 2 years. Am J Sports Med 2019; 47(6): 1331–1337. DOI: 10.1177/0363546519835800

18.

Niu

, et al. Medial patellofemoral ligament reconstruction with semi-patellar tunnel fixation: surgical technique and mid-term follow-up. Med Sci Monit 2017; 23: 5870–5875. DOI: 10.12659/msm.905583

19.

Wang

, et al. Medial patellofemoral ligament reconstruction: a comparison of single-bundle transpatellar tunnel and double-anchor anatomic techniques for the treatment of recurrent lateral patellar dislocation in adults. Arthroscopy 2019; 35(3): 845–854. DOI: 10.1016/j.arthro.2018.08.050

20.

Zhang

Liang

Clinical outcomes after medial patellofemoral ligament reconstruction with suture fixation of the gracilis tendon via transosseous tunnels. Orthop J Sports Med 2020; 8(2): 2325967119900373. DOI: 10.1177/2325967119900373

21.

Philippot

Boyer

Testa

, et al. Study of patellar kinematics after reconstruction of the medial patellofemoral ligament. Clin Biomech 2012; 27(1): 22–26. DOI: 10.1016/j.clinbiomech.2011.08.001

22.

Liu

Duan

Niu

, et al. Lateral retinaculum plasty instead of lateral retinacular release with concomitant medial patellofemoral ligament reconstruction can achieve better results for patellar dislocation. Knee Surg Sports Traumatol Arthrosc 2018; 26(10): 2899–2905. DOI: 10.1007/s00167-017-4798-x

23.

Thaunat

Erasmus

. Management of overtight medial patellofemoral ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2009; 17(5): 480–483. DOI: 10.1007/s00167-008-0702-z

24.

Zumbansen

Haupert

Kohn

, et al. Selective bundle tensioning in double-bundle MPFL reconstruction to improve restoration of dynamic patellofemoral contact pressure. Knee Surg Sports Traumatol Arthrosc 2020; 28(4): 1144–1153. DOI: 10.1007/s00167-019-05789-5

25.

Raoulis

Zibis

Fyllos

, et al. Reconstruction of the medial patellofemoral ligament using two blind transverse semi-patella tunnels and an implant-free technique for patellar fixation: a technical note. J Orthop Surg Res 2021; 16(1):25. DOI: 10.1186/s13018-020-02161-z