Knee Surg Sports Traumatol Arthrosc DOI 10.1007/s00167-013-2751-1
KNEE
Medial patellotibial ligament (MPTL) reconstruction for patellar instability Stefano Zaffagnini • Alberto Grassi • Giulio Maria Marcheggiani Muccioli • William F. Luetzow • Vittorio Vaccari • Andrea Benzi • Maurilio Marcacci
Received: 24 September 2013 / Accepted: 26 October 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To evaluate mid-term clinical and radiographic outcomes after an original medial patellotibial ligament reconstruction in patients with patellar dislocation. Methods Twenty-nine knees (27 patients, 8 males and 19 females) treated for patellar dislocation with medialization of the patellar tendon medial third combined with medial and lateral release were evaluated clinically and radiographically at a mean follow-up of 6.1 ± 2.5 years. Trochleoplasty was performed in case of severe flat trochlea (6 knees, 21 %). Aetiology of patellofemoral instability was traumatic in 6 (21 %) and atraumatic in 23 (79 %) knees. The mean age at first dislocation was 19.2 ± 10.1 years. WOMAC, subjective and objective IKDC, Kujala, VAS for pain, Tegner activity and EQ-5D scores were used. Anteroposterior, lateral and 30° axial views were performed for radiographic monitoring. Results There was a significant improvement of all clinical scores and significant reduction in knee pain. Twenty-four knees (83 %) were normal or nearly normal by objective IKDC score at final follow-up. Radiographs showed a higher incidence of patella alta and flat trochlea in the atraumatic group. Severe signs of patellar osteoarthritis were found in 1 knee (3 %). A higher body mass index (BMI) was correlated
S. Zaffagnini (&) A. Grassi G. M. Marcheggiani Muccioli V. Vaccari A. Benzi M. Marcacci Clinica Ortopedica e Traumatologica II, Laboratorio di Biomeccanica, ed Innovazione Tecnologica, Istituto Ortopedico Rizzoli, Via di Barbiano, 1/10, 40136 Bologna, Italy e-mail: [email protected]; [email protected] W. F. Luetzow Southern California Permanente Medical Group, San Diego, CA, USA
with worse pre-operative scores. Four knees (14 %) were considered failures (2 further dislocations, 2 revision surgeries). The overall survival rate at 6 years was 0.811. Conclusions The presented techniques produced good clinical and radiographic results at mean 6.1 years followup, with 14 % failures. Signs of patellofemoral dysplasia were found in patients with atraumatic patellar dislocation. BMI was related to worse pre-operative clinical status. Level of evidence Retrospective study, Level IV. Keywords
Knee Patella Instability Dislocation
Introduction A large number of surgical procedures have been described in attempt to treat patellofemoral instability [2, 22, 23, 24], but controversy still exists regarding the best type of procedure, indications and timing [6, 17, 18, 30, 33]. This is due to the fact that several anatomic factors such as trochlear dysplasia, patella alta, patellar tilt, increased Q angle, torsional deformities of both the femur and the tibia, generalized ligamentous laxity and weakness of vastus medialis have been related to patellofemoral dislocation [3, 5, 7, 10, 12, 13, 26, 29]. In the world of surgical options, Rillmann et al., Myer et al. and Marcacci et al. [19, 21, 31] have proposed similar techniques characterized by the medialization of the medial third of the patellar tendon, reproducing the course of the medial patellotibial ligament (MPTL). This ligament is actually a medial capsular reinforcement that goes from the inferior aspect of the medial patellar edge going medial on the tibia anteriorly to the medial collateral ligament, close to the articular rim [1]. This ligament has been demonstrated to represent the secondary main restraint to lateral patellar displacement
123
Knee Surg Sports Traumatol Arthrosc
after the MPFL [27, 28]. Thus, this kind of procedure can be practically considered a kind of reconstructions of the MPTL. The aim of this study was to evaluate mid-term clinical and radiographic outcomes after an original MPTL reconstruction in patients with patellar dislocation.
Materials and methods All the patients that underwent MPTL reconstruction, between 2003 and 2008, were included in this retrospective study. The main indication for this surgical procedure was patellar instability, either traumatic or atraumatic, with no resolution of symptoms after at least 3 months of conservative treatment. All patients were clinically evaluated before surgery. The main complaint was patellar instability with multiple dislocation episodes and reduced knee function. Also, patients with less than three dislocation episodes were treated surgically if they experienced continuous apprehension during common activities (like climbing stairs), subluxation episodes and severe limitation with daily and sport activities. Objectively all the patients presented positive apprehension test or patellar subluxation during the knee ROM. The decision to perform concomitant deepening trochleoplasty was made intraoperatively, after direct evaluation of patellofemoral tracking, patellar tilt and trochlear groove depth. Surgical technique After application of the tourniquet, a midline incision with the knee at 20° was performed. An extensive lateral release was performed from the tibial tubercle to the level of insertion of the vastus lateralis tendon on the proximal patella, until obtaining patellar eversion at 90° laterally. Also, a dissection of the vastus medialis oblique was performed, in order to assess the medial facet. Then, the tibial tubercle was isolated, and the medial third of the patellar tendon was harvested with a corresponding 7-mm-long and 7-mm-wide bone plug using an osteotome. This medial third was detached distally, maintaining its insertion on the inferior medial side of the patella, medialized and put under tension trying to find a medial insertion location close to the medial collateral ligament. Tourniquet was released to avoid any influence on this functional evaluation of patella tracking. Repeated dynamic analysis of patella tracking was carried out in order to find a reinsertion point that ensured stability without creating excessive tension on the ligament band. In this position, the patella stability was tested using
123
Fig. 1 Intraoperative step of MPTL reconstruction. The medial third of the patellar tendon was medialized and fixed to the medial tibial cortex with a cancellous screw
a finger pulling the patella laterally in the first degrees of flexion. When the patella was stable throughout the full ROM without any joint limitation and stretching of the ligament structure, a square was created in the cortical bone at the previously localized insertion point, where the bone plug was fixed to the tibia by a cancellous screw or a metal staple (Fig. 1). Avoidance of excessive medialization of the patella was fundamental to prevent clinical failure; therefore, the vastus medialis was never plicated. Trochleoplasty was performed when the Dejour grade D [9] trochlear dysplasia was present, confirming radiographic evaluation. The procedure was similar to the one described by Masse [21] and modified by Dejour et al. [8]. A rectangular trough 0.5 cm deep was created just proximal to the trochlear cartilage and going deep proximally and less deep distally in the trochlea. Then, the cartilage layer was deepened with a blunt tamp and fixed with two small fragment or resorbable screws. A knee brace was applied and locked in full-extension immediately postoperatively. Post-operative protocol The leg was placed in an extension brace. Weight-bearing with the brace was allowed as tolerated after 3 days. Passive motion was started on the third day as well as isometric exercises. At 1 week after surgery, patients began a closed kinetic chain strengthening programme. The
Knee Surg Sports Traumatol Arthrosc
extensor and flexor muscles were contracted in extension and at 30° of flexion, without any applied resistance. At 1 month, patients began a functional programme which included water exercises and progressive strengthening of muscles with return to unrestricted activity once the pain was resolved, by 12 weeks post-operatively. Follow-up evaluation Information regarding the aetiology of the patellar instability was collected in the office. If the patient reported the beginning of knee symptom after a knee trauma, the aetiology was considered traumatic; otherwise, it was considered atraumatic. The number of frank patellar dislocation episodes was collected as well, and the patients were classified accordingly into three groups: 1–3 episodes, more than 3 episodes, and habitual lateral patellar dislocation. The age at first dislocation episode was collected and the time to surgical treatment calculated. Age, sex, side, BMI and surgical history were obtained from medical records. At the final follow-up, the patients were clinically evaluated using EQ-5D, 100-points VAS for pain, WOMAC, subjective IKDC, Kujala and Tegner activity scores. All patients were asked to retrospectively complete the same scores recalling pre-operative status as well. Objective evaluation was performed using the objective IKDC form. The pre-operative objective IKDC values were obtained from medical records. Radiographic evaluation was assessed at final follow-up with AP and lateral 30° radiographs, Rosenberg’s view and axial view at 30° of flexion. Patellofemoral dysplasia was evaluated using the Dejour classification [9], while the patellar height was evaluated using the Insall-Salvati ratio [15]. The sulcus angle, patellar tilt and congruence angle [11] were measured on axial view. Patellar shape according to Wiberg [35] and patellofemoral osteoarthritis according to Iwano score [16] were evaluated on axial view as well. The patients that experienced further dislocations or that underwent revision surgeries due to patellofemoral problems during the follow-up were considered failures. In case of revision surgery, the follow-up and the clinical scores were referred to the time point immediately prior to the revision procedure. Ethics Approval of the study was obtained from the Institutional Review Board (IRB) of the Rizzoli Orthopaedic Institute (protocol number 0030637). Informed consent complied with European Union laws and was signed by the patient before enrolment.
Statistical analysis The statistical analysis was carried out using Analyse-it2.00 (Analyse-it Software Ltd, Leeds, UK). A priori sample size definition was prospectively estimated for a paired Student’s t test with a power of 90 %, starting from the hypothesis of a pre-operative Kujala score 52.4 ± 12.7 and final follow-up value of 93.8 ± 14.2 (referring to Oliva et al. [35]). Considering a high dropout rate in the long term, 32 patients were enrolled in this study. Statistical comparison between the pre-operative and follow-up parametric scores and angles was performed using paired Student’s t test. Differences between preoperative and follow-up of Tegner score were evaluated with nonparametric Wilcoxon test. For differences in the objective IKDC score, the Pearson’s X2 test was used. Differences between traumatic and atraumatic subgroups were performed using independent Student’s t test for parametric values and Mann–Whitney U test for ordinal variables. The population studied was tested for normal distribution before t test was applied. The life-table survival analysis with the log-rank test was used to compare cumulative rates of age at first dislocation between atraumatic and traumatic subgroups. A Kaplan–Meier survival analysis was conducted to obtain survival rates over time. Correlation analysis was carried out using Spearman’s q test. The level of significance was set at P \ 0.05. Results are expressed using mean values ± standard deviation (SD) for parametric values and median ± interquartile range (IQR) for nonparametric values.
Results Patients’ data Out of 32 patients considered eligible for the study, 27 (84 %) patients (29 knees), mostly with atraumatic aetiology, were available at 6.1 ± 2.5 years mean follow-up (Table 1). Two patients, with both bilateral atraumatic patellar instability and flat trochlea, underwent surgery before complete skeletal maturity; therefore, trochleoplasty was not performed and delayed afterwards in case of persistence of symptoms. Considering the aetiology, the mean age of first dislocation was 17.6 ± 8.5 and 26.3 ± 9.4 years for atraumatic and traumatic groups, respectively, presenting a statistically significant difference (P \ 0.0010); a similar finding was reported for trend of first dislocation episodes (P = 0.0350) (Fig. 2). Nine knees (38 %) underwent at least one surgical procedure before the index surgery, while
123
Knee Surg Sports Traumatol Arthrosc Table 1 Demographic, surgical and follow-up details
Mean Total patients
27
Total knees
29
Sex (male/female)
8 (30 %)/19 (70 %)
Knee involved (right/left)
15 (52 %)/14 (48 %)
Cause of luxation (atraumatic\traumatic)
23 (79 %)/6 (21 %)
Age at first luxation (years)
19.2
SD
Range
±10.1
(8.2–44.6)
Number of luxation 1–3
12 (41 %)
[3
12 (41 %)
Continuous
5 (18 %)
Age at surgery (years)
26.9
±10.6
(12.2–47.5)
Age at final follow-up (years)
33.0
±10.1
(17.2–55.1)
Final follow-up (years)
6.1
±2.5
(4.1–10.5)
Time from first luxation to surgery (years)
6.6
±5.9
(0.3–21.0)
BMI at surgery (kg/cm2) BMI at final follow-up (kg/cm2)
24.0 24.6
±5.1 ±6.1
(18.1–33.3) (18.1–36.2)
Table 2 Previous and concurrent surgcal procedures Procedure
Number
Previous surgery Lateral release
3 (10 %)
Distal realignment
3 (10 %)
Distal femoral osteotomy
1 (3 %)
Lateral meniscectomy
1 (3 %)
Medial meniscectomy
3 (10 %)
Hardware removal No surgery
2 (7 %) 18 (62 %)
Concurrent surgery
Fig. 2 Comparison of the age at first dislocation between the traumatic and atraumatic groups. The log-rank test showed a trend of early dislocation in the atraumatic group
17 knees (58 %) underwent at least one concomitant procedure (Table 2). Clinical and radiographic evaluation All the clinical scores were significantly improved from pre-operative status to final follow-up (Table 3). At the final objective evaluation, 24 knees (83 %) were normal or nearly normal, compared to 17 (58 %) before surgery. Regarding radiographic evaluation almost a quarter of the case series (6 knees, 21 %) had the highest degree of
123
Trochleoplasty
6 (21 %)
Medial facetectomy
1 (3 %)
Lateral facetectomy
1 (3 %)
dysplasia, while almost half patients still presented a patella alta (Fig. 3). When the aetiology of patellar instability was considered, a significant difference was found between the subgroups regarding Insall–Salvati ratio and sulcus angle, with lower mean values in the traumatic subgroups (Table 4). The evaluation of patellofemoral osteoarthritis showed none or mild signs in 21 knees (73 %). Correlation analysis The pre-operative BMI resulted correlated with the clinical scores at the pre-operative status (Table 5), indicating worse pre-operative clinical scores in patients with higher BMI, while the post-operative BMI was positively
Knee Surg Sports Traumatol Arthrosc Table 3 Clinical outcomes scores Score
Evaluation
P value
Preoperative
Final follow-up
Improvement
34.7 ± 19.3
71.0 ± 23.7
?36.3
0.0212
Kujala
41.0 ± 24.0
78.1 ± 20.3
?37.1
0.0116
WOMAC
55.9 ± 27.2
88.0 ± 15.0
?32.1
0.0255
VAS
60.8 ± 30.5
19.6 ± 22.0
-41.2
0.0016
Tegner activity level*
1 (1–9)
4 (2–9)
?3
0.0273
EQ-5D Objective IKDC
0.44 ± 0.35 3A, 14B, 12C, 0D
0.80 ± 0.24 14A, 10B, 5C, 0D
?0.36
0.0015 0.0224
Subjective IKDC
* Pre-injury Tegner score 6 (4–9). Significant difference between preinjury and pre-operative values (P = 0.0078)
correlated with the tilt angle (rs = 0.61; P = 0.0090) and negatively to the sulcus angle (rs = -0.51; P = 0.0386), indicating higher BMI in knees with radiographic evidence of patellofemoral dysplasia. A negative correlation between the age at first dislocation and both sulcus angle (rs = -0.56; P = 0.0227) and Insall-Salvati ratio (rs = -0.72; P = 0.0015) was reported as well. Patients with higher degree of dysplasia (C and D) compared to those with lower degree (A and B) presented a wider mean sulcus angle (150.6° ± 7.7° vs. 139.8° ± 8.1°; P = 0.0104) and a lower median Tegner activity score (5 vs. 3; P = 0.0353). The patellar shape and the patellar tilt did not show any significant correlations. Failures and subsequent surgery Four failures (14 %) were reported during the follow-up. One patient with bilateral knee involvement experienced 1 and 3 patellar dislocation episodes of left and right knee, respectively. Had she been older at the time of surgery (11 years for right knee and 13 years for left knee), a trochleoplasty would have performed based on her severe trochlear dysplasia (Dejour grade D and sulcus angle 160° for right knee; Dejour grade D and sulcus angle 158° for left knee). Despite this, the patient declined to undergo revision with trochleoplasty because she was satisfied by the results of the treatment and her quality of life. One patient of the traumatic dislocation group underwent total knee arthroplasty 2.6 years after MPTL reconstruction, as a consequence of knee trauma and dramatic worsening of knee pain. However, this patient was the oldest of the series and presented a 13° valgus knee with moderate patellofemoral and lateral compartment osteoarthritis at time of index
Fig. 3 Pre-operative (a, c, e) and 5.8-year follow-up (b, d, e) of a female patient that underwent MPTL reconstruction at 11.4 years old. Tendon and bone plug fixation was obtained with a metal staple in order to avoid growth disturbance
surgery. One patient underwent a MPFL reconstruction after 3 years due to persistent apprehension and subluxation episodes, but without frank dislocation. This patient presented hyperlaxity, obesity both at index surgery (BMI 30.8) and at revision surgery (BMI 34.2), and a scarce compliance. The overall survival rate at 6-year follow-up was 0.811 (Fig. 4). Fifteen knees (52 %) did not undergo surgery during the follow-up period, 10 knees (34 %) underwent removal of tibial screw, 3 knees (10 %) underwent arthroscopic removal of trochlear screws, and 1 knee (3 %) underwent arthroscopic removal of a loose body.
123
Knee Surg Sports Traumatol Arthrosc Table 4 Radiographic measurement
Item Insall-Salvati
Total
Atraumatic (n = 23)
Traumatic (n = 6)
P value 0.0032
1.21 ± 0.20
1.30 ± 0.18
1.00 ± 0.11
\0.80
0 (0 %)
0 (0 %)
0 (0 %)
[0.80; \1.20
15 (52 %)
9 (39 %)
5 (100 %)
[1.20
14 (48 %)
14 (61 %)
0 (0 %)
Sulcus angle
145.8° ± 9.2°
148.0° ± 8.3°
139.1° ± 9.8°
0.0482
Lateral PF angle
9.15° ± 4.2°
8.7° ± 4.3°
10.6° ± 3.1°
n.s.
Congruence angle
-0.5° ± 14.3°
-2.2° ± 13.5°
?4.4° ± 17.1°
Dejour classification A
4 (14 %)
3 (13 %)
B
12 (41 %)
9 (39 %)
1 (17 %) 3 (50 %)
C D
7 (24 %) 6 (21 %)
6 (26 %) 5 (22 %)
1 (17 %) 1 (17 %)
Patellar shape
n.s.
I
10 (35 %)
8 (35 %)
2 (33 %)
II
16 (55 %)
12 (52 %)
4 (67 %)
III
3 (10 %)
3 (13 %)
0 (0 %)
0
15 (52 %)
13 (57 %)
2 (32 %)
I
6 (21 %)
5 (22 %)
1 (17 %)
II
3 (10 %)
2 (8 %)
1 (17 %)
III
4 (14 %)
3 (13 %)
1 (17 %)
IV
1 (3 %)
0 (0 %)
1 (17 %)
PF osteoarthritis
n.s.
Table 5 Correlation between pre-operative BMI and clinical scores Score
Correlation coefficient
P value
Subjective IKDC
-0.74
0.0091
Kujala
-0.73
0.0108
WOMAC VAS
-0.61 ?0.57
0.0453 n.s.
EQ-5D
-0.71
0.0145
Fig. 4 Kaplan–Meier survival curve
123
n.s. n.s.
Discussion The most important finding of this study is that the presented technique for MPTL reconstruction improved knee function in patients with both traumatic and atraumatic patellar instability in 86 % of cases. It is, however, noteworthy how 2 failures involving the same patient were caused most likely by the impossibility to correct a flat trochlea through a trochleoplasty, due to skeletal immaturity, while the other two failures (MPFL reconstruction and TKA) did not experience any frank redislocation episodes. Several techniques similar to the one presented in this study have been proposed in the past. Rillmann et al. [31] described a modified Elmslie-Trillat procedure, involving a patellar tendon split and the medialization of the medial third of the anterior tibial tubercle, reporting no redislocations but symptoms of instability in 11 % of patients. Furthermore, they reported a patient needing further lateral release and another one needing medial release due to patella infera. Myers et al. [21] added a lateral release and a vastus medialis obliquus muscle advancement to the Rillman’s technique, reporting 76.2 % good or excellent results and 9.5 % redislocation rate, while Oliva et al. [25], using the same technique but only in skeletally immature children
Knee Surg Sports Traumatol Arthrosc
without evident skeletal abnormalities, reported only one case (4.2 %) of recurrence of dislocation. Marcacci et al. [19] used the presented technique combined with Elmslie-Trillat procedure, showing satisfactory results in 88 % of patients with no redislocations, but with one case of poor results due to overcorrection. The technique described in this study differs from the above-presented procedures because does not require complete osteotomy of the anterior tibial tubercle and vastus medialis obliquus muscle reefing, allowing distal realignment without extreme medialization and excessive morbidity. This avoids the risk of premature physeal closure, allowing us to treat also \14 year old patients with open tibial tubercle apophyses. The results of this study are similar to what was reported by Rillman et al. and Marcacci et al., however, without complications due to overcorrection or requiring of subsequent procedures on the soft tissue. The quite poorer results reported by Myers et al. could be possibly due to their execution of medial reefing of vastus medialis obliquus muscle, while the higher outcomes reported by Oliva et al. are most likely due to the exclusion of skeletally abnormal knees. The high rate of concurrent trochleoplasty of the present study (21 %) could have biased the results, as demonstrated by Marcacci et al. [19], who reported worse results in patients with concomitant tibial tubercle medialization and trochleoplasty. However, the patients with patellar instability are often complex cases, with a mixture of soft tissue and bone abnormalities. Therefore, obtaining a homogeneous population for study purpose that need only one isolated procedure is difficult. An interesting finding of this study is the difference of the mean age at first dislocation between the patients with traumatic and atraumatic aetiology. The early dislocation in the atraumatic group suggests skeletal abnormalities that increased the risk of dislocation. The correlation of high BMI to worse symptoms at the pre-operative status confirms what reported in the literature, as high values of BMI have been reported to be associated with higher strains on the articular cartilage of medial and lateral compartments [34], increased prevalence of patella cartilage defects [14] and increased rates of patellofemoral arthritis [20, 21]. Moreover, worse results in term of post-operative sport participation were obtained in patients with severely dysplastic knees. Although an important limitation of this study is the lack of availability of pre-operative radiographs, interesting conclusions can be drawn on the radiographic findings at the final follow-up. First of all, skeletal abnormalities were present especially in the patients with dislocation not caused by trauma, compared to the patients of the traumatic group where normal patellar height was reported. Also, the
Dejour grade of patellofemoral dysplasia [9] presented higher degrees in knees with a flat trochlea, as its shape on lateral radiographs represents an important element of this grading system. Despite a trochleoplasty having been performed to correct flat trochlea, no special treatment was applied to correct patella alta, as demonstrated by the high number of knees (48 %) still presenting abnormal Insall-Salvati ratio at final follow-up. However, to avoid deleterious complication such as patella infera, lowering of the tibial tuberosity was never performed, as excellent results can be obtained also in patients with patella alta [4, 32]. Lastly, the evidence of no gross signs of patellofemoral osteoarthritis at post-operative radiographs in most of knees, combined with 17 % of knees with quite severe degeneration could be likely due to the presence of a wide range of age at final follow-up. Thus, in the day-by-day clinical work, the present technique represents a valid surgical option for the treatment of patellar dislocation, especially in combination with trochleoplasty, ensuring patellar stability also in young patients. In addition to the previously mentioned limitations, the retrospective design of the study represents without doubts a major weakness. Furthermore, the relatively small number of patients reduces the power of correlations and differences between the subgroups, while the mid-term follow-up may not have been able to detect the evolution towards osteoarthritis of this procedure and the possible cases of recurrence of instability.
Conclusions The presented technique of MPTL reconstruction improved knee function in patients with both traumatic and atraumatic patellar instability in 86 % of patients. The procedure did not correct patella alta; however, neither patella infera nor medial instability was produced. Early age at first dislocation correlated with the severity of skeletal abnormalities.
References 1. Arendt EA (2010) Medial side patellofemoral anatomy: surgical implications in patellofemoral instability. In: Zaffagnini S, Dejour D, Arendt EA (ed) Patellofemoral Pain, Instability, and Arthritis. Springer, Heidelberg, pp 149–152 2. Arendt EA, Dejour D (2013) Patella instability: building bridges across the ocean a historic review. Knee Surg Sports Traumatol Arthrosc 21(2):279–293 3. Atkin DM, Fithian DC, Marangi KS, Stone ML, Dobson BE, Mendelsohn C (2000) Characteristics of patients with primary acute lateral patellar dislocation and their recovery within the first 6 months of injury. Am J Sports Med 28(4):472–479
123
Knee Surg Sports Traumatol Arthrosc 4. Brown E, Alexander AH, Lichtman DM (1984) The Elmslie Trillat procedure: evaluation in patellar dislocation and subluxation. Am J Sports Med 12(2):104–109 5. Buchner M, Baudendistel B, Sabo D, Schmitt H (2005) Acute traumatic primary patellar dislocation: long-term results comparing conservative and surgical treatment. Clin J Sport Med 15:62–66 6. Cash JD, Hughston JC (1988) Treatment of acute patellar dislocation. Am J Sports Med 16(3):244–249 7. Dejour H et al (1990) Dysplasia of the femoral trochlea. Rev Chir Orthop Reparatrice Appar Mot 76(1):45–54 8. Dejour H, Walch G, Neyret P, Adeline P (1990) La dysplasie de la trochlee femorale. Rev Chir Orthop 76:45–54 9. Dejour D, Reynaud P, Lecoultre B (1998) Douleurs et instabilite´ rotulienne. Essai de Classification. Me´decine et Hygie`ne 1466–1471 10. Dejour D, Le Coultre B (2007) Osteotomies in patello-femoral instabilities. Sports Med Arthrosc Rev 15(1):39–46 11. Dejour D, Saggin PR, Meyer X, Tavernier T (2010) Standard X-ray examination: patellofemoral disorders. In: Zaffagnini S, Dejour D, Arendt EA (ed) Patellofemoral Pain, Instability, and Arthritis. Springer, Heidelberg, pp 51–60 12. Fithian DC, Paxton EW, Stone ML et al (2004) Epidemiology and natural history of acute patellar dislocation. Am J Sports Med 32(5):1114–1121 13. Gent JJ, Fithian DC (2010) Natural history of patellofemoral dislocations. In: Zaffagnini S, Dejour D, Arendt EA (ed) Patellofemoral Pain, Instability, and Arthritis. Springer, Heidelberg, pp 29–34 14. Gunardi AJ, Brennan SL, Wang Y, Cicuttini FM, Pasco JA, Kotowicz MA, Nicholson GC, Wluka AE (2013) Associations between measures of adiposity over 10 years and patella cartilage in population-based asymptomatic women. Int J Obes (Lond). doi:10.1038/ijo.2013.42 15. Insall JES (1971) Patella position in the normal knee joint. Radiology 101:101–104 16. Iwano T, Kurosawa H, Tokuyama H et al (1990) Roentgenographic and clinical findings of patellofemoral osteoarthrosis. With special reference to its relationship to femorotibial osteoarthrosis and etiologic factors. Clin Orthop Relat Res 252:190–197 17. Larsen E, Lauridsen F (1982) Conservative treatment of patellar dislocations—influence of evident factors on the tendency to redislocation and the therapeutic result. Clin Orthop Relat Res 171:131–136 18. Maenpaa H, Lehto MU (1995) Surgery in acute patellar dislocation–evaluation of the effect of injury mechanism and family occurrence on the outcome of treatment. Br J Sports Med 29(4):239–241 19. Marcacci M, Zaffagnini S, Lo Presti M, Vascellari A, Iacono F, Russo A (2004) Treatment of chronic patellar dislocation with a modified Elmslie-Trillat procedure. Arch Orthop Trauma Surg 124(4):250–257 20. McAlindon T, Zhang Y, Hannan M et al (1996) Are risk factors for patellofemoral and tibiofemoral knee osteoarthritis different? J Rheumatol 23:332–337
123
21. Myers P, Williams A, Dodds R, Bu¨low J (1999) The three-in-one proximal and distal soft tissue patellar realignment procedure. Results, and its place in the management of patellofemoral instability. Am J Sports Med 27(5):575–579 22. Nietosvaara Y, Aalto K, Kallio PE (1994) Acute patellar dislocation in children: incidence and associated osteochondral fractures. J Pediatr Orthop 14:513–515 23. Nikku R, Nietosvaara Y, Aalto K, Kallio PE (2005) Operative treatment of primary patellar dislocation does not improve medium-term outcome: a 7-year follow-up report and risk analysis of 127 randomized patients. Acta Orthop 76:699–704 24. Masse Y (1978) La trochlepastie. Restauration de lagouttierre trochleenne dans les subluxation et luxations de la rotule. Rev Chir Orthop 64:3–17 25. Oliva F, Ronga M, Longo UG, Testa V, Capasso G, Maffulli N (2009) The 3-in-1 procedure for recurrent dislocation of the patella in skeletally immature children and adolescents. Am J Sports Med 37(9):1814–1820 26. Palmu S, Kallio PE, Donell ST, Helenius I, Nietosvaara Y (2008) Acute patellar dislocation in children and adolescents: a randomized clinical trial. J Bone Joint Surg Am 90:463–470 27. Panagiotopoulos E, Strzelczyk P, Herrmann M, Scuderi G (2006) Cadaveric study on static medial patellar stabilizers: the dynamizing role of the vastus medialis obliquus on medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc 14(1):7–12 28. Philippot R, Boyer B, Testa R, Farizon F, Moyen B (2012) The role of the medial ligamentous structures on patellar tracking during knee flexion. Knee Surg Sports Traumatol Arthrosc 20(2):331–336 29. Post WR, Teitge R, Amis A (2002) Patellofemoral malalignment: looking beyond the view box. Clin Sports Med 21:521–546 30. Reider B, Marshall JL, Warren RF (1981) Clinical characteristics of patellar disorders in young athletes. Am J Sports Med 9(4):270–274 31. Rillmann P, Dutly A, Kieser C, Berbig R (1998) Modified Elmslie-Trillat procedure for instability of the patella. Knee Surg Sports Traumatol Arthrosc 6(1):31–35 32. Shelbourne KD, Porter DA, Rozzi W (1994) Use of modified Elmslie Trillat procedure to improve abnormal patellar congruence angle. Am J Sports Med 22:318–323 33. Stefancin JJ, Parker RD (2007) First-time traumatic patellar dislocation: a systematic review. Clin Orthop Relat Res 455:93–101 34. Widmyer MR, Utturkar GM, Leddy HA, Coleman JL, Spritzer CE, Moorman CT 3rd, Defrate LE, Guilak F (2013) High body mass index is associated with increased diurnal strains in the articular cartilage of the knee. Arthritis Rheum 65(10): 2615–2622 35. Wiberg G (1941) Roentgenographic and anatomic studies on the femoropatellar joint. Acta Orthop Scand 12:319–410
KNEE
Medial patellotibial ligament (MPTL) reconstruction for patellar instability Stefano Zaffagnini • Alberto Grassi • Giulio Maria Marcheggiani Muccioli • William F. Luetzow • Vittorio Vaccari • Andrea Benzi • Maurilio Marcacci
Received: 24 September 2013 / Accepted: 26 October 2013 Ó Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To evaluate mid-term clinical and radiographic outcomes after an original medial patellotibial ligament reconstruction in patients with patellar dislocation. Methods Twenty-nine knees (27 patients, 8 males and 19 females) treated for patellar dislocation with medialization of the patellar tendon medial third combined with medial and lateral release were evaluated clinically and radiographically at a mean follow-up of 6.1 ± 2.5 years. Trochleoplasty was performed in case of severe flat trochlea (6 knees, 21 %). Aetiology of patellofemoral instability was traumatic in 6 (21 %) and atraumatic in 23 (79 %) knees. The mean age at first dislocation was 19.2 ± 10.1 years. WOMAC, subjective and objective IKDC, Kujala, VAS for pain, Tegner activity and EQ-5D scores were used. Anteroposterior, lateral and 30° axial views were performed for radiographic monitoring. Results There was a significant improvement of all clinical scores and significant reduction in knee pain. Twenty-four knees (83 %) were normal or nearly normal by objective IKDC score at final follow-up. Radiographs showed a higher incidence of patella alta and flat trochlea in the atraumatic group. Severe signs of patellar osteoarthritis were found in 1 knee (3 %). A higher body mass index (BMI) was correlated
S. Zaffagnini (&) A. Grassi G. M. Marcheggiani Muccioli V. Vaccari A. Benzi M. Marcacci Clinica Ortopedica e Traumatologica II, Laboratorio di Biomeccanica, ed Innovazione Tecnologica, Istituto Ortopedico Rizzoli, Via di Barbiano, 1/10, 40136 Bologna, Italy e-mail: [email protected]; [email protected] W. F. Luetzow Southern California Permanente Medical Group, San Diego, CA, USA
with worse pre-operative scores. Four knees (14 %) were considered failures (2 further dislocations, 2 revision surgeries). The overall survival rate at 6 years was 0.811. Conclusions The presented techniques produced good clinical and radiographic results at mean 6.1 years followup, with 14 % failures. Signs of patellofemoral dysplasia were found in patients with atraumatic patellar dislocation. BMI was related to worse pre-operative clinical status. Level of evidence Retrospective study, Level IV. Keywords
Knee Patella Instability Dislocation
Introduction A large number of surgical procedures have been described in attempt to treat patellofemoral instability [2, 22, 23, 24], but controversy still exists regarding the best type of procedure, indications and timing [6, 17, 18, 30, 33]. This is due to the fact that several anatomic factors such as trochlear dysplasia, patella alta, patellar tilt, increased Q angle, torsional deformities of both the femur and the tibia, generalized ligamentous laxity and weakness of vastus medialis have been related to patellofemoral dislocation [3, 5, 7, 10, 12, 13, 26, 29]. In the world of surgical options, Rillmann et al., Myer et al. and Marcacci et al. [19, 21, 31] have proposed similar techniques characterized by the medialization of the medial third of the patellar tendon, reproducing the course of the medial patellotibial ligament (MPTL). This ligament is actually a medial capsular reinforcement that goes from the inferior aspect of the medial patellar edge going medial on the tibia anteriorly to the medial collateral ligament, close to the articular rim [1]. This ligament has been demonstrated to represent the secondary main restraint to lateral patellar displacement
123
Knee Surg Sports Traumatol Arthrosc
after the MPFL [27, 28]. Thus, this kind of procedure can be practically considered a kind of reconstructions of the MPTL. The aim of this study was to evaluate mid-term clinical and radiographic outcomes after an original MPTL reconstruction in patients with patellar dislocation.
Materials and methods All the patients that underwent MPTL reconstruction, between 2003 and 2008, were included in this retrospective study. The main indication for this surgical procedure was patellar instability, either traumatic or atraumatic, with no resolution of symptoms after at least 3 months of conservative treatment. All patients were clinically evaluated before surgery. The main complaint was patellar instability with multiple dislocation episodes and reduced knee function. Also, patients with less than three dislocation episodes were treated surgically if they experienced continuous apprehension during common activities (like climbing stairs), subluxation episodes and severe limitation with daily and sport activities. Objectively all the patients presented positive apprehension test or patellar subluxation during the knee ROM. The decision to perform concomitant deepening trochleoplasty was made intraoperatively, after direct evaluation of patellofemoral tracking, patellar tilt and trochlear groove depth. Surgical technique After application of the tourniquet, a midline incision with the knee at 20° was performed. An extensive lateral release was performed from the tibial tubercle to the level of insertion of the vastus lateralis tendon on the proximal patella, until obtaining patellar eversion at 90° laterally. Also, a dissection of the vastus medialis oblique was performed, in order to assess the medial facet. Then, the tibial tubercle was isolated, and the medial third of the patellar tendon was harvested with a corresponding 7-mm-long and 7-mm-wide bone plug using an osteotome. This medial third was detached distally, maintaining its insertion on the inferior medial side of the patella, medialized and put under tension trying to find a medial insertion location close to the medial collateral ligament. Tourniquet was released to avoid any influence on this functional evaluation of patella tracking. Repeated dynamic analysis of patella tracking was carried out in order to find a reinsertion point that ensured stability without creating excessive tension on the ligament band. In this position, the patella stability was tested using
123
Fig. 1 Intraoperative step of MPTL reconstruction. The medial third of the patellar tendon was medialized and fixed to the medial tibial cortex with a cancellous screw
a finger pulling the patella laterally in the first degrees of flexion. When the patella was stable throughout the full ROM without any joint limitation and stretching of the ligament structure, a square was created in the cortical bone at the previously localized insertion point, where the bone plug was fixed to the tibia by a cancellous screw or a metal staple (Fig. 1). Avoidance of excessive medialization of the patella was fundamental to prevent clinical failure; therefore, the vastus medialis was never plicated. Trochleoplasty was performed when the Dejour grade D [9] trochlear dysplasia was present, confirming radiographic evaluation. The procedure was similar to the one described by Masse [21] and modified by Dejour et al. [8]. A rectangular trough 0.5 cm deep was created just proximal to the trochlear cartilage and going deep proximally and less deep distally in the trochlea. Then, the cartilage layer was deepened with a blunt tamp and fixed with two small fragment or resorbable screws. A knee brace was applied and locked in full-extension immediately postoperatively. Post-operative protocol The leg was placed in an extension brace. Weight-bearing with the brace was allowed as tolerated after 3 days. Passive motion was started on the third day as well as isometric exercises. At 1 week after surgery, patients began a closed kinetic chain strengthening programme. The
Knee Surg Sports Traumatol Arthrosc
extensor and flexor muscles were contracted in extension and at 30° of flexion, without any applied resistance. At 1 month, patients began a functional programme which included water exercises and progressive strengthening of muscles with return to unrestricted activity once the pain was resolved, by 12 weeks post-operatively. Follow-up evaluation Information regarding the aetiology of the patellar instability was collected in the office. If the patient reported the beginning of knee symptom after a knee trauma, the aetiology was considered traumatic; otherwise, it was considered atraumatic. The number of frank patellar dislocation episodes was collected as well, and the patients were classified accordingly into three groups: 1–3 episodes, more than 3 episodes, and habitual lateral patellar dislocation. The age at first dislocation episode was collected and the time to surgical treatment calculated. Age, sex, side, BMI and surgical history were obtained from medical records. At the final follow-up, the patients were clinically evaluated using EQ-5D, 100-points VAS for pain, WOMAC, subjective IKDC, Kujala and Tegner activity scores. All patients were asked to retrospectively complete the same scores recalling pre-operative status as well. Objective evaluation was performed using the objective IKDC form. The pre-operative objective IKDC values were obtained from medical records. Radiographic evaluation was assessed at final follow-up with AP and lateral 30° radiographs, Rosenberg’s view and axial view at 30° of flexion. Patellofemoral dysplasia was evaluated using the Dejour classification [9], while the patellar height was evaluated using the Insall-Salvati ratio [15]. The sulcus angle, patellar tilt and congruence angle [11] were measured on axial view. Patellar shape according to Wiberg [35] and patellofemoral osteoarthritis according to Iwano score [16] were evaluated on axial view as well. The patients that experienced further dislocations or that underwent revision surgeries due to patellofemoral problems during the follow-up were considered failures. In case of revision surgery, the follow-up and the clinical scores were referred to the time point immediately prior to the revision procedure. Ethics Approval of the study was obtained from the Institutional Review Board (IRB) of the Rizzoli Orthopaedic Institute (protocol number 0030637). Informed consent complied with European Union laws and was signed by the patient before enrolment.
Statistical analysis The statistical analysis was carried out using Analyse-it2.00 (Analyse-it Software Ltd, Leeds, UK). A priori sample size definition was prospectively estimated for a paired Student’s t test with a power of 90 %, starting from the hypothesis of a pre-operative Kujala score 52.4 ± 12.7 and final follow-up value of 93.8 ± 14.2 (referring to Oliva et al. [35]). Considering a high dropout rate in the long term, 32 patients were enrolled in this study. Statistical comparison between the pre-operative and follow-up parametric scores and angles was performed using paired Student’s t test. Differences between preoperative and follow-up of Tegner score were evaluated with nonparametric Wilcoxon test. For differences in the objective IKDC score, the Pearson’s X2 test was used. Differences between traumatic and atraumatic subgroups were performed using independent Student’s t test for parametric values and Mann–Whitney U test for ordinal variables. The population studied was tested for normal distribution before t test was applied. The life-table survival analysis with the log-rank test was used to compare cumulative rates of age at first dislocation between atraumatic and traumatic subgroups. A Kaplan–Meier survival analysis was conducted to obtain survival rates over time. Correlation analysis was carried out using Spearman’s q test. The level of significance was set at P \ 0.05. Results are expressed using mean values ± standard deviation (SD) for parametric values and median ± interquartile range (IQR) for nonparametric values.
Results Patients’ data Out of 32 patients considered eligible for the study, 27 (84 %) patients (29 knees), mostly with atraumatic aetiology, were available at 6.1 ± 2.5 years mean follow-up (Table 1). Two patients, with both bilateral atraumatic patellar instability and flat trochlea, underwent surgery before complete skeletal maturity; therefore, trochleoplasty was not performed and delayed afterwards in case of persistence of symptoms. Considering the aetiology, the mean age of first dislocation was 17.6 ± 8.5 and 26.3 ± 9.4 years for atraumatic and traumatic groups, respectively, presenting a statistically significant difference (P \ 0.0010); a similar finding was reported for trend of first dislocation episodes (P = 0.0350) (Fig. 2). Nine knees (38 %) underwent at least one surgical procedure before the index surgery, while
123
Knee Surg Sports Traumatol Arthrosc Table 1 Demographic, surgical and follow-up details
Mean Total patients
27
Total knees
29
Sex (male/female)
8 (30 %)/19 (70 %)
Knee involved (right/left)
15 (52 %)/14 (48 %)
Cause of luxation (atraumatic\traumatic)
23 (79 %)/6 (21 %)
Age at first luxation (years)
19.2
SD
Range
±10.1
(8.2–44.6)
Number of luxation 1–3
12 (41 %)
[3
12 (41 %)
Continuous
5 (18 %)
Age at surgery (years)
26.9
±10.6
(12.2–47.5)
Age at final follow-up (years)
33.0
±10.1
(17.2–55.1)
Final follow-up (years)
6.1
±2.5
(4.1–10.5)
Time from first luxation to surgery (years)
6.6
±5.9
(0.3–21.0)
BMI at surgery (kg/cm2) BMI at final follow-up (kg/cm2)
24.0 24.6
±5.1 ±6.1
(18.1–33.3) (18.1–36.2)
Table 2 Previous and concurrent surgcal procedures Procedure
Number
Previous surgery Lateral release
3 (10 %)
Distal realignment
3 (10 %)
Distal femoral osteotomy
1 (3 %)
Lateral meniscectomy
1 (3 %)
Medial meniscectomy
3 (10 %)
Hardware removal No surgery
2 (7 %) 18 (62 %)
Concurrent surgery
Fig. 2 Comparison of the age at first dislocation between the traumatic and atraumatic groups. The log-rank test showed a trend of early dislocation in the atraumatic group
17 knees (58 %) underwent at least one concomitant procedure (Table 2). Clinical and radiographic evaluation All the clinical scores were significantly improved from pre-operative status to final follow-up (Table 3). At the final objective evaluation, 24 knees (83 %) were normal or nearly normal, compared to 17 (58 %) before surgery. Regarding radiographic evaluation almost a quarter of the case series (6 knees, 21 %) had the highest degree of
123
Trochleoplasty
6 (21 %)
Medial facetectomy
1 (3 %)
Lateral facetectomy
1 (3 %)
dysplasia, while almost half patients still presented a patella alta (Fig. 3). When the aetiology of patellar instability was considered, a significant difference was found between the subgroups regarding Insall–Salvati ratio and sulcus angle, with lower mean values in the traumatic subgroups (Table 4). The evaluation of patellofemoral osteoarthritis showed none or mild signs in 21 knees (73 %). Correlation analysis The pre-operative BMI resulted correlated with the clinical scores at the pre-operative status (Table 5), indicating worse pre-operative clinical scores in patients with higher BMI, while the post-operative BMI was positively
Knee Surg Sports Traumatol Arthrosc Table 3 Clinical outcomes scores Score
Evaluation
P value
Preoperative
Final follow-up
Improvement
34.7 ± 19.3
71.0 ± 23.7
?36.3
0.0212
Kujala
41.0 ± 24.0
78.1 ± 20.3
?37.1
0.0116
WOMAC
55.9 ± 27.2
88.0 ± 15.0
?32.1
0.0255
VAS
60.8 ± 30.5
19.6 ± 22.0
-41.2
0.0016
Tegner activity level*
1 (1–9)
4 (2–9)
?3
0.0273
EQ-5D Objective IKDC
0.44 ± 0.35 3A, 14B, 12C, 0D
0.80 ± 0.24 14A, 10B, 5C, 0D
?0.36
0.0015 0.0224
Subjective IKDC
* Pre-injury Tegner score 6 (4–9). Significant difference between preinjury and pre-operative values (P = 0.0078)
correlated with the tilt angle (rs = 0.61; P = 0.0090) and negatively to the sulcus angle (rs = -0.51; P = 0.0386), indicating higher BMI in knees with radiographic evidence of patellofemoral dysplasia. A negative correlation between the age at first dislocation and both sulcus angle (rs = -0.56; P = 0.0227) and Insall-Salvati ratio (rs = -0.72; P = 0.0015) was reported as well. Patients with higher degree of dysplasia (C and D) compared to those with lower degree (A and B) presented a wider mean sulcus angle (150.6° ± 7.7° vs. 139.8° ± 8.1°; P = 0.0104) and a lower median Tegner activity score (5 vs. 3; P = 0.0353). The patellar shape and the patellar tilt did not show any significant correlations. Failures and subsequent surgery Four failures (14 %) were reported during the follow-up. One patient with bilateral knee involvement experienced 1 and 3 patellar dislocation episodes of left and right knee, respectively. Had she been older at the time of surgery (11 years for right knee and 13 years for left knee), a trochleoplasty would have performed based on her severe trochlear dysplasia (Dejour grade D and sulcus angle 160° for right knee; Dejour grade D and sulcus angle 158° for left knee). Despite this, the patient declined to undergo revision with trochleoplasty because she was satisfied by the results of the treatment and her quality of life. One patient of the traumatic dislocation group underwent total knee arthroplasty 2.6 years after MPTL reconstruction, as a consequence of knee trauma and dramatic worsening of knee pain. However, this patient was the oldest of the series and presented a 13° valgus knee with moderate patellofemoral and lateral compartment osteoarthritis at time of index
Fig. 3 Pre-operative (a, c, e) and 5.8-year follow-up (b, d, e) of a female patient that underwent MPTL reconstruction at 11.4 years old. Tendon and bone plug fixation was obtained with a metal staple in order to avoid growth disturbance
surgery. One patient underwent a MPFL reconstruction after 3 years due to persistent apprehension and subluxation episodes, but without frank dislocation. This patient presented hyperlaxity, obesity both at index surgery (BMI 30.8) and at revision surgery (BMI 34.2), and a scarce compliance. The overall survival rate at 6-year follow-up was 0.811 (Fig. 4). Fifteen knees (52 %) did not undergo surgery during the follow-up period, 10 knees (34 %) underwent removal of tibial screw, 3 knees (10 %) underwent arthroscopic removal of trochlear screws, and 1 knee (3 %) underwent arthroscopic removal of a loose body.
123
Knee Surg Sports Traumatol Arthrosc Table 4 Radiographic measurement
Item Insall-Salvati
Total
Atraumatic (n = 23)
Traumatic (n = 6)
P value 0.0032
1.21 ± 0.20
1.30 ± 0.18
1.00 ± 0.11
\0.80
0 (0 %)
0 (0 %)
0 (0 %)
[0.80; \1.20
15 (52 %)
9 (39 %)
5 (100 %)
[1.20
14 (48 %)
14 (61 %)
0 (0 %)
Sulcus angle
145.8° ± 9.2°
148.0° ± 8.3°
139.1° ± 9.8°
0.0482
Lateral PF angle
9.15° ± 4.2°
8.7° ± 4.3°
10.6° ± 3.1°
n.s.
Congruence angle
-0.5° ± 14.3°
-2.2° ± 13.5°
?4.4° ± 17.1°
Dejour classification A
4 (14 %)
3 (13 %)
B
12 (41 %)
9 (39 %)
1 (17 %) 3 (50 %)
C D
7 (24 %) 6 (21 %)
6 (26 %) 5 (22 %)
1 (17 %) 1 (17 %)
Patellar shape
n.s.
I
10 (35 %)
8 (35 %)
2 (33 %)
II
16 (55 %)
12 (52 %)
4 (67 %)
III
3 (10 %)
3 (13 %)
0 (0 %)
0
15 (52 %)
13 (57 %)
2 (32 %)
I
6 (21 %)
5 (22 %)
1 (17 %)
II
3 (10 %)
2 (8 %)
1 (17 %)
III
4 (14 %)
3 (13 %)
1 (17 %)
IV
1 (3 %)
0 (0 %)
1 (17 %)
PF osteoarthritis
n.s.
Table 5 Correlation between pre-operative BMI and clinical scores Score
Correlation coefficient
P value
Subjective IKDC
-0.74
0.0091
Kujala
-0.73
0.0108
WOMAC VAS
-0.61 ?0.57
0.0453 n.s.
EQ-5D
-0.71
0.0145
Fig. 4 Kaplan–Meier survival curve
123
n.s. n.s.
Discussion The most important finding of this study is that the presented technique for MPTL reconstruction improved knee function in patients with both traumatic and atraumatic patellar instability in 86 % of cases. It is, however, noteworthy how 2 failures involving the same patient were caused most likely by the impossibility to correct a flat trochlea through a trochleoplasty, due to skeletal immaturity, while the other two failures (MPFL reconstruction and TKA) did not experience any frank redislocation episodes. Several techniques similar to the one presented in this study have been proposed in the past. Rillmann et al. [31] described a modified Elmslie-Trillat procedure, involving a patellar tendon split and the medialization of the medial third of the anterior tibial tubercle, reporting no redislocations but symptoms of instability in 11 % of patients. Furthermore, they reported a patient needing further lateral release and another one needing medial release due to patella infera. Myers et al. [21] added a lateral release and a vastus medialis obliquus muscle advancement to the Rillman’s technique, reporting 76.2 % good or excellent results and 9.5 % redislocation rate, while Oliva et al. [25], using the same technique but only in skeletally immature children
Knee Surg Sports Traumatol Arthrosc
without evident skeletal abnormalities, reported only one case (4.2 %) of recurrence of dislocation. Marcacci et al. [19] used the presented technique combined with Elmslie-Trillat procedure, showing satisfactory results in 88 % of patients with no redislocations, but with one case of poor results due to overcorrection. The technique described in this study differs from the above-presented procedures because does not require complete osteotomy of the anterior tibial tubercle and vastus medialis obliquus muscle reefing, allowing distal realignment without extreme medialization and excessive morbidity. This avoids the risk of premature physeal closure, allowing us to treat also \14 year old patients with open tibial tubercle apophyses. The results of this study are similar to what was reported by Rillman et al. and Marcacci et al., however, without complications due to overcorrection or requiring of subsequent procedures on the soft tissue. The quite poorer results reported by Myers et al. could be possibly due to their execution of medial reefing of vastus medialis obliquus muscle, while the higher outcomes reported by Oliva et al. are most likely due to the exclusion of skeletally abnormal knees. The high rate of concurrent trochleoplasty of the present study (21 %) could have biased the results, as demonstrated by Marcacci et al. [19], who reported worse results in patients with concomitant tibial tubercle medialization and trochleoplasty. However, the patients with patellar instability are often complex cases, with a mixture of soft tissue and bone abnormalities. Therefore, obtaining a homogeneous population for study purpose that need only one isolated procedure is difficult. An interesting finding of this study is the difference of the mean age at first dislocation between the patients with traumatic and atraumatic aetiology. The early dislocation in the atraumatic group suggests skeletal abnormalities that increased the risk of dislocation. The correlation of high BMI to worse symptoms at the pre-operative status confirms what reported in the literature, as high values of BMI have been reported to be associated with higher strains on the articular cartilage of medial and lateral compartments [34], increased prevalence of patella cartilage defects [14] and increased rates of patellofemoral arthritis [20, 21]. Moreover, worse results in term of post-operative sport participation were obtained in patients with severely dysplastic knees. Although an important limitation of this study is the lack of availability of pre-operative radiographs, interesting conclusions can be drawn on the radiographic findings at the final follow-up. First of all, skeletal abnormalities were present especially in the patients with dislocation not caused by trauma, compared to the patients of the traumatic group where normal patellar height was reported. Also, the
Dejour grade of patellofemoral dysplasia [9] presented higher degrees in knees with a flat trochlea, as its shape on lateral radiographs represents an important element of this grading system. Despite a trochleoplasty having been performed to correct flat trochlea, no special treatment was applied to correct patella alta, as demonstrated by the high number of knees (48 %) still presenting abnormal Insall-Salvati ratio at final follow-up. However, to avoid deleterious complication such as patella infera, lowering of the tibial tuberosity was never performed, as excellent results can be obtained also in patients with patella alta [4, 32]. Lastly, the evidence of no gross signs of patellofemoral osteoarthritis at post-operative radiographs in most of knees, combined with 17 % of knees with quite severe degeneration could be likely due to the presence of a wide range of age at final follow-up. Thus, in the day-by-day clinical work, the present technique represents a valid surgical option for the treatment of patellar dislocation, especially in combination with trochleoplasty, ensuring patellar stability also in young patients. In addition to the previously mentioned limitations, the retrospective design of the study represents without doubts a major weakness. Furthermore, the relatively small number of patients reduces the power of correlations and differences between the subgroups, while the mid-term follow-up may not have been able to detect the evolution towards osteoarthritis of this procedure and the possible cases of recurrence of instability.
Conclusions The presented technique of MPTL reconstruction improved knee function in patients with both traumatic and atraumatic patellar instability in 86 % of patients. The procedure did not correct patella alta; however, neither patella infera nor medial instability was produced. Early age at first dislocation correlated with the severity of skeletal abnormalities.
References 1. Arendt EA (2010) Medial side patellofemoral anatomy: surgical implications in patellofemoral instability. In: Zaffagnini S, Dejour D, Arendt EA (ed) Patellofemoral Pain, Instability, and Arthritis. Springer, Heidelberg, pp 149–152 2. Arendt EA, Dejour D (2013) Patella instability: building bridges across the ocean a historic review. Knee Surg Sports Traumatol Arthrosc 21(2):279–293 3. Atkin DM, Fithian DC, Marangi KS, Stone ML, Dobson BE, Mendelsohn C (2000) Characteristics of patients with primary acute lateral patellar dislocation and their recovery within the first 6 months of injury. Am J Sports Med 28(4):472–479
123
Knee Surg Sports Traumatol Arthrosc 4. Brown E, Alexander AH, Lichtman DM (1984) The Elmslie Trillat procedure: evaluation in patellar dislocation and subluxation. Am J Sports Med 12(2):104–109 5. Buchner M, Baudendistel B, Sabo D, Schmitt H (2005) Acute traumatic primary patellar dislocation: long-term results comparing conservative and surgical treatment. Clin J Sport Med 15:62–66 6. Cash JD, Hughston JC (1988) Treatment of acute patellar dislocation. Am J Sports Med 16(3):244–249 7. Dejour H et al (1990) Dysplasia of the femoral trochlea. Rev Chir Orthop Reparatrice Appar Mot 76(1):45–54 8. Dejour H, Walch G, Neyret P, Adeline P (1990) La dysplasie de la trochlee femorale. Rev Chir Orthop 76:45–54 9. Dejour D, Reynaud P, Lecoultre B (1998) Douleurs et instabilite´ rotulienne. Essai de Classification. Me´decine et Hygie`ne 1466–1471 10. Dejour D, Le Coultre B (2007) Osteotomies in patello-femoral instabilities. Sports Med Arthrosc Rev 15(1):39–46 11. Dejour D, Saggin PR, Meyer X, Tavernier T (2010) Standard X-ray examination: patellofemoral disorders. In: Zaffagnini S, Dejour D, Arendt EA (ed) Patellofemoral Pain, Instability, and Arthritis. Springer, Heidelberg, pp 51–60 12. Fithian DC, Paxton EW, Stone ML et al (2004) Epidemiology and natural history of acute patellar dislocation. Am J Sports Med 32(5):1114–1121 13. Gent JJ, Fithian DC (2010) Natural history of patellofemoral dislocations. In: Zaffagnini S, Dejour D, Arendt EA (ed) Patellofemoral Pain, Instability, and Arthritis. Springer, Heidelberg, pp 29–34 14. Gunardi AJ, Brennan SL, Wang Y, Cicuttini FM, Pasco JA, Kotowicz MA, Nicholson GC, Wluka AE (2013) Associations between measures of adiposity over 10 years and patella cartilage in population-based asymptomatic women. Int J Obes (Lond). doi:10.1038/ijo.2013.42 15. Insall JES (1971) Patella position in the normal knee joint. Radiology 101:101–104 16. Iwano T, Kurosawa H, Tokuyama H et al (1990) Roentgenographic and clinical findings of patellofemoral osteoarthrosis. With special reference to its relationship to femorotibial osteoarthrosis and etiologic factors. Clin Orthop Relat Res 252:190–197 17. Larsen E, Lauridsen F (1982) Conservative treatment of patellar dislocations—influence of evident factors on the tendency to redislocation and the therapeutic result. Clin Orthop Relat Res 171:131–136 18. Maenpaa H, Lehto MU (1995) Surgery in acute patellar dislocation–evaluation of the effect of injury mechanism and family occurrence on the outcome of treatment. Br J Sports Med 29(4):239–241 19. Marcacci M, Zaffagnini S, Lo Presti M, Vascellari A, Iacono F, Russo A (2004) Treatment of chronic patellar dislocation with a modified Elmslie-Trillat procedure. Arch Orthop Trauma Surg 124(4):250–257 20. McAlindon T, Zhang Y, Hannan M et al (1996) Are risk factors for patellofemoral and tibiofemoral knee osteoarthritis different? J Rheumatol 23:332–337
123
21. Myers P, Williams A, Dodds R, Bu¨low J (1999) The three-in-one proximal and distal soft tissue patellar realignment procedure. Results, and its place in the management of patellofemoral instability. Am J Sports Med 27(5):575–579 22. Nietosvaara Y, Aalto K, Kallio PE (1994) Acute patellar dislocation in children: incidence and associated osteochondral fractures. J Pediatr Orthop 14:513–515 23. Nikku R, Nietosvaara Y, Aalto K, Kallio PE (2005) Operative treatment of primary patellar dislocation does not improve medium-term outcome: a 7-year follow-up report and risk analysis of 127 randomized patients. Acta Orthop 76:699–704 24. Masse Y (1978) La trochlepastie. Restauration de lagouttierre trochleenne dans les subluxation et luxations de la rotule. Rev Chir Orthop 64:3–17 25. Oliva F, Ronga M, Longo UG, Testa V, Capasso G, Maffulli N (2009) The 3-in-1 procedure for recurrent dislocation of the patella in skeletally immature children and adolescents. Am J Sports Med 37(9):1814–1820 26. Palmu S, Kallio PE, Donell ST, Helenius I, Nietosvaara Y (2008) Acute patellar dislocation in children and adolescents: a randomized clinical trial. J Bone Joint Surg Am 90:463–470 27. Panagiotopoulos E, Strzelczyk P, Herrmann M, Scuderi G (2006) Cadaveric study on static medial patellar stabilizers: the dynamizing role of the vastus medialis obliquus on medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc 14(1):7–12 28. Philippot R, Boyer B, Testa R, Farizon F, Moyen B (2012) The role of the medial ligamentous structures on patellar tracking during knee flexion. Knee Surg Sports Traumatol Arthrosc 20(2):331–336 29. Post WR, Teitge R, Amis A (2002) Patellofemoral malalignment: looking beyond the view box. Clin Sports Med 21:521–546 30. Reider B, Marshall JL, Warren RF (1981) Clinical characteristics of patellar disorders in young athletes. Am J Sports Med 9(4):270–274 31. Rillmann P, Dutly A, Kieser C, Berbig R (1998) Modified Elmslie-Trillat procedure for instability of the patella. Knee Surg Sports Traumatol Arthrosc 6(1):31–35 32. Shelbourne KD, Porter DA, Rozzi W (1994) Use of modified Elmslie Trillat procedure to improve abnormal patellar congruence angle. Am J Sports Med 22:318–323 33. Stefancin JJ, Parker RD (2007) First-time traumatic patellar dislocation: a systematic review. Clin Orthop Relat Res 455:93–101 34. Widmyer MR, Utturkar GM, Leddy HA, Coleman JL, Spritzer CE, Moorman CT 3rd, Defrate LE, Guilak F (2013) High body mass index is associated with increased diurnal strains in the articular cartilage of the knee. Arthritis Rheum 65(10): 2615–2622 35. Wiberg G (1941) Roentgenographic and anatomic studies on the femoropatellar joint. Acta Orthop Scand 12:319–410
