Systematic Review

Medial Soft-Tissue Realignment Versus Medial Patellofemoral Ligament Reconstruction for Recurrent Patellar Dislocation: Systematic Review Jae-Gwang Song, M.D., Seung-Baik Kang, M.D., So-Hee Oh, Ph.D., Jae-Hwi Han, M.D., Daivesh Shah, M.S.Ortho., Hyung-Joon Park, M.D., Umid T. Kholmurodov, M.D., and Kyung-Wook Nha, M.D.

Purpose: To compare the clinical outcomes between medial soft-tissue surgery and medial patellofemoral ligament (MPFL) reconstruction for recurrent patellar dislocation without any evident predisposing factors. Methods: A literature search was performed on the established medical databases MEDLINE, EMBASE, and the Cochrane register. The inclusion criteria were as follows: English-language papers for recurrent patellar dislocation without any evident predisposing factors, clinical trial(s) with clear description of surgical technique, adult subjects, medial soft-tissue surgery or MPFL reconstruction without combined surgery, and a follow-up longer than 2 years. The methodological quality of all articles was assessed by 2 authors according to the Coleman methodology score. Results: Thirteen studies (mean Coleman methodology score value, 74.1; standard deviation, 11.5) were included in the analysis. Five studies reported the outcomes of patients undergoing medial soft-tissue surgery, compared with 7 studies reporting MPFL reconstruction. Overall, 109 patients underwent medial soft-tissue surgery with a minimum 2-years follow-up, compared with 308 patients of MPFL reconstruction. There was one direct comparative study between medial soft-tissue surgery and MPFL reconstruction. Of the patients who received medial soft-tissue surgery, 0 to 9.7% experienced redislocation, compared with 0 to 10.7% of the MPFL reconstruction group. The ranges of differences in Kujala scores were 23.6 to 31.7 points in patients who underwent medial soft-tissue surgery and 23.11 to 38.8 points in patients who underwent MPFL reconstruction. The ranges of postoperative congruence angles were
14.4 to 8.2 for medial soft-tissue surgery and
7.7 to
5.2 for MPFL reconstruction. The ranges of postoperative lateral patellofemoral angles were 7.9 to 9.4 for medial soft-tissue surgery and 5 to 5.3 for MPFL reconstruction. Conclusions: All studies on medial soft-tissue surgery and MPFL reconstruction for recurrent patellar dislocation without predisposing factors showed satisfactory outcomes despite the use of numerous surgical techniques, graft types, and follow-up periods. Level of Evidence: Level IV, Systematic Review.

From the Department of Orthopedic Surgery, Aerospace Medical Center, Republic of Korea Air Force (J-G.S.), Cheongju-si; Department of Orthopedic Surgery (S-B.K.) and Department of Biostatistics (S-H.O.), Seoul National University Boramae Hospital, Seoul; Department of Orthopedic Surgery, Inje University, Ilsan Paik Hospital (J-H.H., H-J.P., K-W.N.), Ilsan, Republic of Korea; Department of Orthopaedic Surgery, B. J. Medical College and New Civil Hospital (D.S.), Ahmedabad, India; and Department of Orthopaedic Surgery, Tashkent Medical Academy (U.T.K.), Tashkent, Uzbekistan. The authors report that they have no conflicts of interest in the authorship and publication of this article. J-G.S. and S-B.K. contributed equally to this work as co-first authors. Received December 10, 2014; accepted August 7, 2015. Address correspondence to Kyung-Wook Nha, M.D., 2240, Daehwa-dong, Ilsan-Segu, Koyang-Si, Ilsan 411-706, Korea. E-mail: kwnhamj@hotmail. com Ó 2015 by the Arthroscopy Association of North America 0749-8063/141035/$36.00 http://dx.doi.org/10.1016/j.arthro.2015.08.012

P

atellar instability is often multifactorial, with osseous and soft-tissue abnormalities leading to recurrent lateral dislocations. According to the causes of dislocation, different surgeries are adopted. In cases of severe trochlear dysplasia, trochleoplasty may be indicated.1-3 If extensor mechanism disorders such as increased distance from the tibial tubercle to the trochlear groove (TT-TG) or patellar alta are present, proximal realignment, distal realignment, or combined surgery follow. For lower-limb malalignment (high Qangle) or patella alta, distal realignment surgery is indicated. Ideal candidates for proximal soft-tissue procedures have a normal Q-angle, insufficient medial structures, and initial instability resulting from a traumatic event.2,3 Proximal soft-tissue procedures seek to balance the medial and lateral soft tissues. The rationale behind these procedures is that recurrent patellar

Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol

-,

No

-

(Month), 2015: pp 1-10

1

2

J-G. SONG ET AL.

dislocation is due to anatomical vulnerability of the medial structures including the medial patellofemoral ligament (MPFL), medial retinaculum, vastus medialis obliquus (VMO), and tightness of the lateral structures such as the retinaculum or the iliotibial band.4-6 Although various proximal soft-tissue procedures have been proposed, there is no consensus on the most effective procedure. Recently, MPFL reconstruction has become a popular treatment option for recurrent lateral patellar dislocations, as this ligament is the primary passive restraint to lateral patellar translation of early knee flexion. But medial soft-tissue surgery such as medial imbrication, reefing, and medial retinaculum plasty (MRP) has also shown good results, with numerous reports citing its simplicity and relatively easy manipulation of tension.7-9 The purpose of this study was to compare the clinical outcomes between medial soft-tissue surgery and MPFL reconstruction for recurrent patellar dislocation without any evident predisposing factors. Our initial hypothesis was that MPFL reconstruction would effect superior outcomes.

Methods Literature Search Two of the authors (J-G.S., J-H.H.) independently performed comprehensive online literature searches of the MEDLINE, EMBASE, and Cochrane Library databases between August 3, 2014, and August 10, 2014. The following search protocol (modified for each of the other databases) was employed (Table 1). The same 2 authors independently screened the title and abstract of each search-returned article and then reviewed the full text of each article that had been selected on the basis of Table 1. Search Protocol 1 2

3 4

5 6 7 8 9 10 11 12 13 14

Search Terms Patella dislocation [Mesh] “Patella[tiab] AND (dislocation [tiab] OR subluxation[tiab] OR instability[tiab]) 1 OR 2 (“Medial patellofemoral ligament” [tiab] OR MPFL[tiab]) AND reconstruction[tiab] (“Medial patellofemoral ligament” [tiab] OR MPFL[tiab]) “Medial reefing” [tiab] “Medial augmentation” [tiab] “Medial plication” [tiab] “Medial imbrication” [tiab] “Quadricepsplasty” [tiab] “Proximal realignment” [tiab] 6 or 7 or 8 or 9 or 10 or 11 4 OR 5 OR 12 3 and 13

tiab, title/abstract.

Results 535 1,507

1,780 253

369

the inclusion and exclusion criteria (Table 2). If wholepatient data were obtainable from the articles, those patients who also satisfied the inclusion criteria were selected for reanalysis. In cases of two or more studies by the same author, we determined whether the patients were duplicated or not. If duplicated, we included only the latest study. Quality Assessment The methodological quality of each of the studies included in the analysis was evaluated by two of the authors (S-B.K., D.S.) individually according to the Coleman methodology score.10 Each study was assessed for each of the methodology’s 10 criteria, resulting in a final score ranging anywhere from 0 to 100. A perfect score of 100 indicated a study design that largely avoids the influence of chance, various biases, and confounding factors. Each author scored the methodological quality of the studies twice, with a 10-day interval between assessments. In cases of disagreement, the 2 authors debated the controversial score until reaching a consensus. Data Abstraction The studies were evaluated by 2 independent review authors for methodological quality. To extract data from the papers, we used a standardized form including the following: first author, publication year, publishing journal, study type, demographic factors, sample sizes, and results of research. Data were then extracted and cross-checked for accuracy. Subjects within the studies were divided into 2 treatment groups: those undergoing medial soft-tissue realignment and those undergoing MPFL reconstruction, respectively. Study data including (1) mean age of patients, (2) average dislocation before surgery, (3) surgical technique (medial soft-tissue surgery, MPFL reconstruction), and (4) follow-up are summarized in Table 3. The radiological and clinical outcome data extracted from studies included (1) overall radiological outcomes, (2) overall clinical results, (3) redislocation rates, and (4) complications, as summarized in Tables 4 and 5. The clinical and radiological outcome measures as a range of all included studies specifically recorded for the systematic review included (1) redislocation rates (at final follow-up), (2) Kujala score (mean difference of preoperative and postoperative score), (3) patellar tilt (at final follow-up), and (4) congruence angle (at final follow-up).

Results

182 542 292

Literature Search The electronic search initially identified 673 articles. Critical application of the inclusion and exclusion criteria subsequently reduced that number to 13 (Fig 1). The articles referenced 5 studies about medial softtissue surgery, 7 about MPFL reconstruction, and one

MEDIAL REEFING VS MPFL RECONSTRUCTION

3

Table 2. Inclusion and Exclusion Criteria Inclusion Criteria Studies with patients who received isolated medial soft-tissue surgery/MPFL reconstruction Studies in which patients showed minimum 2 patellar dislocations

Studies reporting a minimum 2-year follow-up data on clinical, functional, and imaging outcomes Level I, II, III, or IV evidence Articles written in English Human subjects

Exclusion Criteria Studies with patients who received medial soft-tissue surgery/MPFL reconstruction with combined procedure Studies with patients with: (1) Q-angle greater than 20, (2) patella alta (I/S ratio > 1.2), (3) patients who had surgery immediately after trauma due to a large cartilage lesion, (4) a history of previous surgery for recurrent patellar dislocation, (5) severe trochlear dysplasia (Dejour’s classification B, C, D), (6) a congenital patellar dislocation, or (7) skeletally immature patients Studies reporting less than 2-year follow-up data on clinical, functional and imaging outcomes Level V evidence (case report, technique note, letter to editor), biomechanical reports, and review articles Article written in language other than English Nonhuman subjects

MPFL, medial patellofemoral ligament.

directly comparing clinical outcomes between medial soft-tissue surgery and MPFL reconstruction. Quality Assessment The mean modified Coleman methodology score value of the included studies was 74.1  11.5 (range, 54 to 94). The mean Coleman methodology score for each criterion are shown in Table 6, respectively. Data Abstraction Medial Soft-Tissue Surgery Five studies8,11-14 reported the outcomes of patients undergoing medial soft-tissue surgery. Overall, 109 patients underwent medial soft-tissue surgery with a minimum 2-years of follow up. Operations performed included all-arthroscopic imbrication, arthroscopic assisted medial capsular or retinacular plication, medial reefing, and MRP. Lateral retinacular release was performed in 3 studies. Miller et al.12 reported that lateral release might not be necessary in patients without excessive patellar tilt. Nam and Karzel8 reviewed 22 patients (23 knees) who underwent medial reefing with a minimum 2 years of follow-up. Among 23 knees, 13 knees were excluded: 11 knees with I/S ratio >1.2 and 2 knees with trochlear dysplasia. Thus, 10 patients satisfied our inclusion criteria, and we formed a subgroup for the analysis. One study reported the outcomes of medial soft-tissue surgery with a long-term follow-up (mean follow up, 11.6 years).13 MPFL Reconstruction Seven studies satisfied our inclusion criteria for MPFL reconstruction group.15-21 Two were reported by the same investigator (H. Kang).16,21 Some of the patients were duplicated, and so we included only those from

the more recent study. Overall, 308 patients underwent MPFL reconstruction with a minimum-2 years followup. Several MPFL reconstruction techniques were followed, so we formed a subgroup by surgical technique according to single/double bundle and patellar fixation technique (transpatellar tunnel/nontranspatellar tunnel fixation). There was no long-term follow-up study. Single Bundle: Transpatellar Tunnel Fixation. Becher et al.17 compared static and dynamic MPFL reconstruction. In the dynamic MPFL reconstruction group, the distal part of the gracilis tendon is detached from its tibial insertion, transferred to the medial margin of the patella, and fixed by passing the graft through the transpatellar tunnel. Single Bundle: Nontranspatellar Tunnel Fixation. Wang et al.18 evaluated double-bundle anatomical reconstruction and a single-bundle isometric procedure. In single-bundle isometric reconstruction, the midpoint between adductor tubercle and the highest point of the medial femoral condyle was selected as the initial insertion point on the femur, and the supermedial patellar medial edge was selected as the insertion point on the patella. The graft was sutured to the anchors in the patella. Witonski et al.15 reported MPFL reconstruction using patellar tendon. Then the medial third part of patellar tendon was separated from the tuberosity of the tibia, first distally and then proximally. The graft was left attached to uppermedial quadrant of the patella. The free end of the graft was sewn by Krackow’s stitch. Double Bundle: Transpatellar Tunnel Fixation. Ronga et al.20 and Panni et al.19 reported anatomic reconstruction of the MPFL with the hamstring tendon passed through a double patellar transverse bony tunnel.

4

Table 3. Patients’ Demographic Information and Surgical Procedures Study Medial Soft-Tissue Surgery Halbrecht11

No. of Patients

Mean Age

Average Dislocation

26 (29 knees)

30 (17-50)

NA

Miller et al.12 Lee et al.13

24 (25 knees) 31

28.1 (13-54) 23.9 (14-32)

NA 9.5

Barkatali et al.14

14

25 (16-65)

NA

Nam and Karzel8 (subgroup)

9 (10 knees)

19.5 (6.35)

3 (0.8)

Surgical Procedure All arthroscopic medial capsular plication with lateral release Arthroscopic assisted medial capsular plication Arthroscopic assisted medial capsular plication with lateral release Arthroscopic medial capsular plication using suture anchor Miniopen medial reefing

Graft type

Panni et al.19 Kang et al.21 Wang et al.18

Witonski et al.15 Becher et al.17 Kang et al.16 Comparative Study Ma et al.22

28

32.5 (19-40)

6.3 (3)

48 (51 knees) Group Y, 40; group C, 42 Group SB, 21 (26 knees); group DB: 37 (44 knees) 10

28 (16-60) 28.3 (4.8); 29.4 (5.6) 23 (10); 26 (7)

5 (3-10) NA; NA

Autogenous hamstring tendon Semitendinosus tendon Semitendinosus tendon

NA

Semitendinosus tendon

27.2 (18-42)

>5 in 9 patients; 2

Gracilis tendon Semitendinosus tendon

Semitendinosus tendon; medial retinaculum plasty

Minimum 2 years 61 months (24-120 months) 11.6 (2.4) years 52 months (24-72) 4.6 (4.25) years

Patellar fixation 2 transverse patellar tunnel

3.1 (2.5-4) years

Divergent 2 patellar tuneel Sutured to the prepatellar fascia and retinaculum Suture anchor

33 months (24-54 months) 32 months (7.1 months); 33.9 months (6.9 months) 48 months; 48 months

Graft was left attached to upper medial quadrant of patella Group D: transpatellar tunnel Group S: Two suture anchors Sutured to the prepatellar fascia and retinaculum

3 years 7months

Inferior bundle: suture anchor; superior bundle: suture

26 months (0.6) 33.7 months (8.4)

40 months (24-55 months)

NOTE. Data presented as mean (range) or mean (standard deviation). e, not available; C, C-shaped graft; D, dynamic reconstruction; DB, double bundle; MPFLR, medial patellofemoral ligament reconstruction; MRP, medial retinaculum plasty; NA, not available; S, static reconstruction; SB, single bundle; Y, Y-shaped graft.

J-G. SONG ET AL.

Reconstruction Ronga et al.20

Follow-up

Table 4. Radiological Outcomes in Studies Congruence Angle Postoperative

Preoperative

Postoperative

8.2
2.5 (2)
2.8 (2.7) d
14.4 (7.5)


3 5.5 (4.8)
4.2 (1.9) d
0.3 (6)

9.4 8.7 (3.9) 8.2 (2.5) d 7.9 (2.1)

d d

d d

d d


7.7 (6.5)


6.6 (5.5)

d d d

Lateral Patellar Displacement Postoperative

Preoperative

Postoperative

1.2 mm 0.2 (1.5) mm d d d

d d d d d

d d d d d

d d

d d

d 11.1

d 8.9

5.0 (3.4)

13.2 (6.7) mm

2.9 (2.4) mm

24.5 (5.6)

13.6 (4.2)

d d d

d d d

18.8 (5.0) mm 20.0 (4.4) mm d

11.3 (2.1) mm 10.2 (1.5) mm d

12.1 (2.8) 12.8 (3.6) d

9.2 (1.6) 7.8 (1.4) d

d d

d d

d d

d d

d d

d d

d d


5.2 (5.1)


7.5 (5.4)

5.3 (2.6)

13.8 (6.2) mm

2.5 (1.4) mm

24.7 (6)

d d

d d

21 (2.7) mm 20.7 (2.9) mm

9.4 (1.9) mm 8.4 (1.7) mm

15 (3) 15 (3)

4 (3) 4 (2)



Preoperative

Patellar Tilt

8 mm 2 (5.3) mm d d d

14.4 (4.5)

MEDIAL REEFING VS MPFL RECONSTRUCTION

Study Preoperative Medial Soft-Tissue Surgery: 30.7 Halbrecht11 12 Miller et al. 4.4 (15) Lee et al.13 16.5 (3) d Barkatali et al.14 Nam and Karzel8 18.6 (10.6) Reconstruction: Ronga et al.20 d d Panni et al.19 Kang et al.21 C-shaped graft 18.2 (5.3) Wang et al.18 Single bundle d Double bundle d Witonski et al.15 d Becher et al.17 Static d Dynamic d Kang et al.16 Y-shaped graft 20 (7.1) Comparative Study (MRP v MPFL Reconstruction) Ma et al.22 MRP 20 (3) MPFLR 20 (2)

Lateral Patellofemoral Angle

9 (3) 8 (3)

NOTE. Data presented as mean (standard deviation). e, not available; MPFLR, medial patellofemoral ligament reconstruction; MRP, medial retinaculum plasty.

5

6

Table 5. Overall Clinical Outcomes and Complications in Studies Study Medial Soft-Tissue Surgery: Halbrecht11

Redislocation 0

Miller et al.12

0

Lee et al.13

3

Barkatali et al.14

1

Nam and Karzel8

0

3

Panni et al.19

0

Kang et al.21

0

Wang et al.18

3 (SB)

Witonski et al.15

Becher et al.17

Kang et al.16 Comparative study (MRP v MPFL reconstruction): Ma et al.22

0

1 (group D)

0

0

1. 2. 1. 2.

Ninety-three percent of patients reported significant subjective improvement. Lysholm score improved from 41.5 to 79.3 (P < .05). Ninety-six percent (24/25) were satisfied with their results Lysholm knee scores improved from an average of 54 to 91 (P < .001). Tegner activity level improved from an average of 3.3 to 6.2. The mean Kujala score improved from 57.5  13.2 preoperatively to 89.2  8.7 points at the final follow-up (P < .0001). The median Tegner activity score improved from 3 to 7 at the final follow-up (P < .0001). The mean preoperative Kujala score was 54.6 (SD, 24.4). The mean postoperative Kujala score was 78.2 (SD, 20.5). The mean increase in the Kujala score was 23.6 The mean Kujala score improved to 91.9  11.8 points at the final follow-up (P < .0001). The median Tegner activity score improved from 3 to 7 at the final follow-up (P < .0001).

Kujala Scores increased from a mean 45 to 89 (P ¼ .03). Muscle volume of the thigh of the operated limb increased with time, but significant isokinetic strength difference was found between the operated and the contralateral limb (.05 < P < .006) 1. Eighty-nine percent of patients were either satisfied or very satisfied with the pain relief achieved; 64% of patients returned to the same type of sports at the same level, and 16% reduced the level or type of sport. 2. Mean Kujala score improved significantly from 56.7  17.7 to 86.8  14.4 (P  .01). Modified Lysholm score improved significantly from 57.6  19.6 to 88.1  16.2. (P < .01) Lysholm scores for Y-graft group and C-graft group were 92.3  3.9 and 88.4  6.8, at final follow-up (P ¼ .001). The mean postoperative Kujala scores were 95.9  4.7 and 91.3  9.7 for the Y-graft group and C-graft group, respectively (P ¼ .001) 1. Subjective outcome was better in group DB than in group SB, especially in the long term. 2. Kujala scores were improved from 57.35  7.14 to 80.46  3.59 in group SB and 61  5.17 to 92.86  2.47 in group DB at 48 months of follow-up. 1. SF-36 score revealed a significant improvement in bodily pain, general health, physical role functioning, social role functioning, and physical functional domain. 2. Mean Kujala score improved significantly from 59.7 to 84.4. Mean Kujala scores were 82  17 and 84  12; Lysholm scores were 79  18 and 78  17; Tegener scores were 4.4  1.8 and 5.2  1.9 in group S and group D, respectively, at final follow-up with no significant between-group differences found. Mean Lysholm score improved from 51.8  6.2 to 91.7  4.1, and mean Kujala score was from 53.4  5.3 to 90.9  6.6 (P < .01), at last follow-up.

Complications No No

1. Hematoma, 2 knees 2. Superficial infection, one knee 3. Loss of flexion of less than 135 , 3 knees No

One hyposthesia medial to patella; 13 anterior knee pain; 2 knee stiffness One Transverse patella fracture

The median Kujala score had significantly improved after surgery. However, no statistically significant difference was found between the 2 groups (MRP v MPFLR) at the latest follow-up (P > .05)

D, dynamic reconstruction; DB, double bundle; MPFLR, medial patellofemoral ligament reconstruction; MRP, medial retinaculum plasty; S, static reconstruction; SB, single bundle; SD, standard deviation; SF-36, Short Form Health Survey 36.

J-G. SONG ET AL.

Reconstruction: Ronga et al.20

Clinical Results

MEDIAL REEFING VS MPFL RECONSTRUCTION

7

Comparative Study Ma et al.22 evaluated the clinical results of MRP and MPFL reconstruction with concomitant lateral retinacular release. (MRP is a surgical technique that strengthens and tightens the VMO and medial retinaculum by advancement distally for the VMO and proximally for the medial retinaculum at the medial patellar border.) In the MPFL reconstruction group, a double-bundle graft was used and fixed by nontranspatellar tunnel fixation (suture anchor).

Fig 1. Flowchart of articles during selection process.

Double Bundle: Nontranspatellar Tunnel Fixation. Kang et al.21 reported the Y-graft technique (Y-shaped graft; femoral fixation with separate tensioning of 2 bundles) and the C-graft technique (C-shaped graft; patellar fixation with simultaneous tensioning of both bundles). Both techniques were double-bundle MPFL reconstruction, and the grafts were sutured to the prepatellar fascia and retinaculum at the patellar side. Becher et al.17 reported the double-bundle reconstruction with suture anchor fixation at the patellar side in the static MPFL reconstruction group. In addition, Wang et al.18 used suture anchors for fixation of the graft at the patellar side in the doublebundle reconstruction group.

Table 6. Overall Coleman Methodology Score for Each Criterion Criteria (Maximum Score) Part A: Study size (10) Mean follow-up (5) No. of procedures (10) Type of study (15): Diagnostic certainty (5) Surgery description (5) Rehabilitation description (10) Part B: Outcome criteria (10) Procedure for outcomes (15) Selection process (15) Total score

Mean

Standard Deviation

Range

5.2 4.8 10

3.4 0.8 0

0-10 2-5 10

5 5 8.5

0 0 3.7

5 5 0-10

7 7 15 74.4

0 3.2 0 11.5

7 3-12 15 54-94

Clinical and Radiological Outcomes Redislocation Rates. The rates of redislocation at final follow-up were analyzed. The ranges of rates were as follows: Of the patients who received medial soft-tissue surgery, 0 to 9.7% experienced redislocation, compared with 0 to 10.7% of the MPFL reconstruction group. In the long-term follow-up study, 9.7% of patients who underwent medial soft-tissue surgery experienced redislocation. In the MPFL reconstruction group, the ranges of rates were as follows: 7% of single-bundle reconstruction with transpatellar fixation, 0% of singlebundle reconstruction with nontranspatellar fixation, 0 to 10.7% of double-bundle reconstruction with transpatellar fixation, and 0 to 7% of double-bundle reconstruction with nontranspatellar fixation. In the comparative study, there was no redislocation in either group. Kujala Scores. Kujala scores were used in 9 studies for subjective outcomes (2 studies on medial soft-tissue surgery, 6 on MPFL reconstruction, one comparative study). The mean differences between the preoperative and postoperative scores at final follow-up were analyzed. One study reported the postoperative score, so just 9 studies were analyzed. The ranges of differences were 23.6 to 31.7 points in patients who underwent medial soft-tissue surgery and 23.11 to 38.8 points in patients who underwent MPFL reconstruction. The mean differences of long-term follow-up study were 31.7 points in the medial softtissue surgery group. In the MPFL reconstruction group, the ranges of differences were as follows: 23.11 to 24.7 points for single-bundle reconstruction with nontranspatellar fixation, 30.1 to 38 points for doublebundle reconstruction with transpatellar fixation, and 31.86 to 38.8 points for double-bundle reconstruction with nontranspatellar fixation. In the comparative study, there was an improvement of 29 points in the medial soft-tissue surgery group and 33 points in the MPFL reconstruction group. Congruence Angle. The congruence angle at final follow-up was measured in 4 studies on medial softtissue surgery, 2 on MPFL reconstruction, and one comparative study. At the final follow-up, the range of

8

J-G. SONG ET AL.

congruence angle were
14.4 to 8.2 for medial softtissue surgery and
7.7 to
5.2 for MPFL reconstruction. The congruence angle of the long-term follow-up study was
2.8 in the medial soft-tissue surgery group. In the MPFL reconstruction group, the congruence angle was measured only on doublebundle reconstruction in the nontranspatellar fixation group. In the comparative study, the congruence angle at final follow-up was 4 in both groups. Lateral Patellofemoral Angle. The lateral patellofemoral angle at final follow-up was measured in 4 studies on medial soft-tissue surgery and 2 on MPFL reconstruction. At the final follow-up, the range of lateral patellofemoral angle were 7.9 to 9.4 for medial soft-tissue surgery and 5 to 5.3 for MPFL reconstruction. The lateral patellofemoral angle of the long-term follow-up study was 8.2 in the medial soft-tissue surgery group. In the MPFL reconstruction group, the congruence angle was measured only on double-bundle reconstruction in the nontranspatellar fixation group.

Discussion Two important aspects emerged from this study. First, there was only one direct comparative study between medial soft-tissue surgery and MPFL reconstruction. Most articles were written with a focus on surgical technique and reported the outcomes of a retrospective series. Moreover, the included studies showed significant heterogeneity, with numerous surgical techniques, graft types, and follow-up periods, so we could not compare the 2 techniques using statistical methods or conclude which surgical procedures were better. Second, despite these limitations, all studies on medial softtissue surgery and MPFL reconstruction for recurrent patellar dislocation without predisposing factors showed satisfactory outcomes, despite using numerous surgical techniques, graft types, average dislocation before surgery, and follow-up periods. Patellar dislocation and recurrent instability are attributed to numerous predisposing factors, including trochlear dysplasia, patellar height, and rotational abnormalities of the tibia and femur. Because our purpose was to show the clinical outcomes of medial soft-tissue procedures and MPFL reconstruction, we intentionally removed these confounding variables from consideration by narrowing the inclusion criteria. Medial soft-tissue procedures addressing patellofemoral instability were introduced several decades ago. Several technical variations, as reflects the lack of a standardized surgical technique, have been reported. Most articles included in our study reported arthroscopic techniques and miniopen approaches. Halbrecht11 performed an all arthroscopic technique in 26 patients (29 knees) affected by patellar dislocation. At

an average follow-up of 20 months, 93% of patients felt their symptoms significantly improved, and none had experienced any further dislocation. Miller et al.12 reported prospective outcomes of arthroscopic techniques. They followed 24 patients (25 knees) with recurrent patellar dislocation or subluxation for an average period of 60 months. At the last follow-up, 96% of patients were satisfied with their results. The average follow-up of the study by Lee et al.13 is the longest in the literature. They reported the outcomes of medial capsule repair and lateral release in a series of 31 patients at 11.6 years of follow-up. Excellent or good results were achieved in 90% of cases; however, recurrence of dislocation was reported in 3 cases. Nam and Karzel8 reviewed the outcomes of 22 patients (23 knees) affected by patellar dislocation. At an average follow-up of 4.4 years, 91% of patients were rated as excellent or good; however, one redislocation and one recurrent subluxation were observed. For the purposes of the present study, we excluded patients who had other predisposing factors (patella alta, high Q-angle, trocholear dysplasia) and formed a subgroup that satisfied our criteria. The outcomes of this subgroup included no redislocation and no recurrent subluxation. Reconstruction of the MPFL has become a popular procedure for recurrent patellar instability. MPFL reconstruction attracts attention because it directly addresses the deficiencies of the medial restraining structure.23 In fact, MPFL provides approximately 60% of the total medial restraining force against lateral patellar displacement.4 This ligament is often damaged or torn during patellar dislocation.24,25 Various techniques for MPFL reconstruction have been reported, and to date no technique is considered to represent the gold standard. Besides surgical details such as graft choice, femoral tunnel position, and the patellar fixation method, the knee flexion angle for graft tension has been one of the most controversial issues. We categorized the surgical technique as single-/double-bundle and graft fixation methods (transpatellar tunnel/nontranspatellar tunnel techniques). Despite using numerous surgical techniques, MPFL reconstruction for recurrent patellar dislocation showed satisfactory outcomes. There were 3 comparative studies of MPFL reconstruction depending on surgical techniques. Becher et al.17 compared outcomes of static and dynamic MPFL reconstruction. The static technique group (15 patients) underwent rigid fixation of the gracilis tendon at the anatomic femoral MPFL insertion and the superomedial border of the patella; the dynamic technique group (15 patients) underwent detachment of the gracilis tendon at the pes anserinus with fixation to the proximal medial patellar margin via tunnel transfer obliquely through the patella. At last follow-up, one case of redislocation was observed in the dynamic group. No significant betweengroup differences were found in mean Kujala, Tegner,

9

MEDIAL REEFING VS MPFL RECONSTRUCTION

Lysholm, or visual analogue scale scores or radiographic parameters. Kang et al.21 compared clinical outcomes of MPFL reconstruction between the Y-graft and C-graft techniques. Forty patients in the Y-graft group and 42 patients in the C-graft group were followed for 24 months. No recurrent dislocation or subluxation was reported in either group. The mean postoperative Lysholm scores for the Y-graft and C-graft groups were 92.3  3.9 and 88.4  6.8, respectively (P ¼ .001). The mean postoperative Kujala scores were 95.9  4.7 and 91.3  9.7 for the Y-graft and C-graft groups, respectively (P ¼ .001), with a good or excellent rate of 97.5% in the Y-graft group compared with 83.3% in the C-graft group (P ¼ .031). Wang et al.18 compared the clinical outcome of double-bundle anatomical versus single-bundle isometric reconstruction of the MPFL. Single-bundle isometric technique was performed in 21 patients (26 knees), and double-bundle anatomical reconstruction was performed in 37 patients (44 knees). At last followup, redislocation occurred in 3 patients in the singlebundle technique, while no redislocation occurred in the double-bundle technique. They concluded that single- and double-bundle reconstruction of the MPFL can both effectively restore patella stability and improve knee function. However, outcomes in the follow-up period showed that the double-bundle surgery procedure was much better than the single-bundle surgery. A few studies directly comparing medial reefing and MPFL reconstruction have been conducted.22,26 The outcomes of medial reefing versus MPFL reconstruction for recurrent patellar instability in adults are controversial. Ma et al.22 reported that MRP yields results similar to those of MPFL reconstruction in terms of the maintenance of the corrected patellar position and rotation. The measurement results for the congruence angle, patellar tilt angle, and patellar lateral shift decreased significantly, from the pretreatment values to the normal range at the final follow-up, without a statistically significant difference between the 2 groups (P > .05). However, Zhao et al.26 reported that MPFL reconstruction results in better static patellar position and functional outcome than MRP in the treatment of recurrent patellar dislocation in adults. They followed 43 patients in an MRP group and 45 in an MPFL reconstruction group for 5 years, receiving complete evaluations. Four patients (9.3%) in the MRP group and one (2.2%) in the MPFL reconstruction group experienced episodes of redislocation, and 7 (16.3%) in the MRP group and 3 (6.7%) in the MPFL reconstruction group experienced multiple episodes of patellar instability (P ¼ .037). The discrepancy between the recurrence rate in their study and that in ours might be due to patient selection, that is, Zhao et al.’s patients might have had more severe predisposing conditions. Furthermore, for most of their patients, combined procedures (anteromedial TT transfer, TT distalization) were performed. On these bases, we excluded their study.

Limitations There are several limitations to our study. First, this review is based on low-quality studies. Most of the studies included in the analysis, a total of 10, have level IV evidence, 2 have level II evidence, and one has level III evidence; there were no level I studies. This factor might have affected the conclusions. Second, there was only one available direct comparative study between the 2 techniques, and hence no direct conclusions could be made therefrom. Third, there were 5 studies of medial soft-tissue surgery compared with 7 studies with MPFL reconstruction. Medial soft-tissue surgery was performed in 108 patients compared with MPFL reconstruction in 309 patients. Fourth, there was significant heterogeneity among MPFL reconstructive techniques. To address the heterogeneity, we categorized the techniques. Fifth, lateral tightness in recurrent patellar dislocation was not addressed in this study. Lateral tightness also contributed to patellar instability. Sixth, even patients with traumatic patellar dislocations were probably predisposed to patellar dislocations with some amount of trochlea dysplasia, alta, or increased TT-TG, otherwise they would not have dislocated at all. But we have attempted to exclude all potential osseous confounders (TT-TG, Q-angle, patellar height, mechanical axis), thus primarily generating a relatively homogenous population of medially deficient soft tissues after recurrent patellar dislocation in patients with recurrent patellar dislocation. Finally, significant sources of selection bias, including significant heterogeneity in patient populations, clinical outcome measures, and different follow-up periods were often present in the studies reviewed. Because of this significant heterogeneity, we could not compare the 2 techniques using statistical methods. Further research needs to be performed, particularly studies with uniform indications and direct comparisons between the 2 procedures, to allow for definitive conclusions.

Conclusions All studies on medial soft-tissue surgery and MPFL reconstruction for recurrent patellar dislocation without predisposing factors showed satisfactory outcomes, despite the use of numerous surgical techniques, graft types, and follow-up periods.

References 1. Dejour D, Le Coultre B. Osteotomies in patella-femoral instabilities. Sports Med Arthrosc 2007;15:39-46. 2. von Knoch F, Bohm T, Burgi ML, von Knoch M, Bereiter H. Trochleaplasty for recurrent patellar dislocation in association with trochlear dysplasia. A 4- to 14year follow-up study. J Bone Joint Surg Br 2006;88: 1331-1335.

10

J-G. SONG ET AL.

3. Donell ST, Joseph G, Hing CB, Marshall TJ. Modified Dejour trochleoplasty for severe dysplasia: Operative technique and early clinical results. Knee 2006;13:266-273. 4. Desio SM, Burks RT, Bachus KN. Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med 1998;26:59-65. 5. Vainionpaa S, Laasonen E, Silvennoinen T, Vasenius J, Rokkanen P. Acute dislocation of the patella. A prospective review of operative treatment. J Bone Joint Surg Br 1990;72:366-369. 6. Conlan T, Garth WP Jr, Lemons JE. Evaluation of the medial soft-tissue restraints of the extensor mechanism of the knee. J Bone Joint Surg Am 1993;75:682-693. 7. Ali S, Bhatti A. Arthroscopic proximal realignment of the patella for recurrent instability: Report of a new surgical technique with 1 to 7 years of follow-up. Arthroscopy 2007;23:305-311. 8. Nam EK, Karzel RP. Mini-open medial reefing and arthroscopic lateral release for the treatment of recurrent patellar dislocation: A medium-term follow-up. Am J Sports Med 2005;33:220-230. 9. Small NC, Glogau AI, Berezin MA. Arthroscopically assisted proximal extensor mechanism realignment of the knee. Arthroscopy 1993;9:63-67. 10. Coleman BD, Khan KM, Maffulli N, Cook JL, Wark JD. Studies of surgical outcome after patellar tendinopathy: Clinical significance of methodological deficiencies and guidelines for future studies. Victorian Institute of Sport Tendon Study Group. Scand J Med Sci Sports 2000;10:2-11. 11. Halbrecht JL. Arthroscopic patella realignment: An allinside technique. Arthroscopy 2001;17:940-945. 12. Miller JR, Adamson GJ, Pink MM, Fraipont MJ, Durand P Jr. Arthroscopically assisted medial reefing without routine lateral release for patellar instability. Am J Sports Med 2007;35:622-629. 13. Lee JJ, Lee SJ, Won YG, Choi CH. Lateral release and medial plication for recurrent patella dislocation. Knee Surg Sports Traumatol Arthrosc 2012;20:2438-2444. 14. Barkatali BM, Lea M, Aster A, Marynissen H. Arthroscopic medial capsular plication using the suture anchor technique. Knee Surg Sports Traumatol Arthrosc 2014;22:2513-2517. 15. Witonski D, Keska R, Synder M, Sibinski M. An isolated medial patellofemoral ligament reconstruction with patellar tendon autograft. BioMed Res Int 2013;2013:637678.

16. Kang HJ, Cao JH, Pan S, Wang XJ, Yu DH, Zheng ZM. The horizontal Y-shaped graft with respective graft tension angles in anatomical two-bundle s reconstruction. Knee Surg Sports Traumatol Arthrosc 2014;22:2445-2451. 17. Becher C, Kley K, Lobenhoffer P, Ezechieli M, Smith T, Ostermeier S. Dynamic versus static reconstruction of the medial patellofemoral ligament for recurrent lateral patellar dislocation. Knee Surg Sports Traumatol Arthrosc 2014;22:2452-2457. 18. Wang CH, Ma LF, Zhou JW, et al. Double-bundle anatomical versus single-bundle isometric medial patellofemoral ligament reconstruction for patellar dislocation. Int Orthop 2013;37:617-624. 19. Panni AS, Alam M, Cerciello S, Vasso M, Maffulli N. Medial patellofemoral ligament reconstruction with a divergent patellar transverse 2-tunnel technique. Am J Sports Med 2011;39:2647-2655. 20. Ronga M, Oliva F, Longo UG, Testa V, Capasso G, Maffulli N. Isolated medial patellofemoral ligament reconstruction for recurrent patellar dislocation. Am J Sports Med 2009;37:1735-1742. 21. Kang H, Cao J, Yu D, Zheng Z, Wang F. Comparison of 2 different techniques for anatomic reconstruction of the medial patellofemoral ligament: A prospective randomized study. Am J Sports Med 2013;41:1013-1021. 22. Ma LF, Wang F, Chen BC, Wang CH, Zhou JW, Wang HY. Medial retinaculum plasty versus medial patellofemoral ligament reconstruction for recurrent patellar instability in adults: a randomized controlled trial. Arthroscopy 2013;29:891-897. 23. Bicos J, Fulkerson JP, Amis A. Current concepts review: The medial patellofemoral ligament. Am J Sports Med 2007;35:484-492. 24. Garth WP Jr, DiChristina DG, Holt G. Delayed proximal repair and distal realignment after patellar dislocation. Clin Orthop Relat Res 2000:132-144. 25. Nomura E, Horiuchi Y, Inoue M. Correlation of MR imaging findings and open exploration of medial patellofemoral ligament injuries in acute patellar dislocations. Knee 2002;9:139-143. 26. Zhao J, Huangfu X, He Y. The role of medial retinaculum plication versus medial patellofemoral ligament reconstruction in combined procedures for recurrent patellar instability in adults. Am J Sports Med 2012:1355-1364.