Outcomes of MPFL Reconstruction with Grammont in Pediatric Patients with Patellar Instability and Increased TT-TG Distance: A Comparison Study

Abstract

Background:

Patellar instability in pediatric patients is commonly treated with medial patellofemoral ligament reconstruction (MPFLR). MPFLR alone does not directly address other underlying anatomic risk factors for patellar instability, such as lateralized patella tendon insertion. The Grammont procedure is a technique for medialization of the patella tendon insertion that can correct increased TT-TG distance in skeletally immature patients. This technique has not been studied in acute and recurrent traumatic patellar instability, nor has it been compared as an adjunct procedure to MPFLR.

Hypothesis/Purpose:

The purpose of this study was to compare medial patellofemoral ligament reconstruction (MPFLR) alone versus MPFLR with a Grammont procedure in pediatric patients. It was hypothesized that combining MPFLR with the Grammont procedure, compared with MPFLR alone, would reduce recurrent patellar instability, achieve greater correction of the tibial tubercle–trochlear groove (TT-TG) (MPFLR alone) or patellar tendon insertion–trochlear groove (PTI-TG) (MPFLR + Grammont) distance, and maintain a comparable complication rate.

Study Design:

Cohort study; Level of evidence, 3.

Methods:

A retrospective review was performed including all pediatric patients (<19 years) treated surgically for patellar instability at 1 institution between 2010 and 2023. A total of 45 patients underwent MPFLR and 36 patients underwent MPFLR plus Grammont. Rates of recurrent instability and revision surgery were compared. Pre- and postoperative imaging (radiography and magnetic resonance imaging) was used to evaluate changes in Caton-Deschamps ratio and TT-TG (MPFLR alone) or PTI-TG (MPFLR + Grammont). Additionally, the investigators compared the rate of modified Clavien-Dindo grade II and III complications between surgical groups.

Results:

In the MPFLR + Grammont group vs the MPFLR-only group, patients were younger at the time of surgery (median age 13.4 vs 15.3 years, respectively; P = .013), the percentage of female patients was higher (80.6% vs 55.8%; P = .016), and the mean preoperative TT-TG was larger (18.8 vs 15.7 mm; P = .001). Twenty (25.3%) patients had grade II/III complications, with no significant differences between surgical groups. Twelve patients (14.8%) experienced recurrent instability, with rates of 20.9% in the MPFLR-only group and 8.3% in the MPFLR + Grammont group (P = .120). Revision surgery for recurrent patellar instability was performed in 4 patients in the MPFLR-only group (9.3%) and 1 patient in the MPFLR + Grammont group (2.8%). In the MPFLR + Grammont group, 1 patient had a growth arrest of the tibial tubercle apophysis resulting in recurvatum deformity.

Conclusion:

In pediatric patients with patellar instability and increased TT-TG distance, adding a Grammont procedure to MPFLR resulted in similar overall complication rates compared with isolated MPFLR. The MPFLR + Grammont group was younger and had more severe anatomic risk factors; the investigators observed a possibly lower rate of recurrent instability and fewer revision surgeries.

Keywords

pediatric patellar instability MPFL reconstruction Grammont procedure recurrent instability and complications

Patellar instability is a complex condition with a wide array of surgical treatment options, each tailored to patient-specific factors. The tibial tubercle–trochlear groove (TT-TG) distance was popularized by Dejour et al¹¹ in 1994 and has been used to quantify the amount of laterization of the tibial tubercle. The TT-TG distance is the distance between the tibial tubercle and the trochlear groove, using axial imaging. Lateralization of the tibial tubercle is defined by an increase in the TT-TG distance and is a known risk factor for first-time and recurrent patellar dislocations.⁹ In skeletally mature patients, tibial tubercle osteotomies (TTOs) and medial patellofemoral ligament reconstruction (MPFLR) have demonstrated high success rates in addressing patellar instability in patients with an elevated TT-TG distance.

TTOs have emerged as an effective surgical intervention to correct an elevated TT-TG distance, with reported success rates of 72.5% to 86% over follow-up periods of 6.2 to 8.2 years.^4,30 Additionally, patients undergoing MPFLR in the presence of an elevated TT-TG distance have demonstrated improved functional outcomes and kinematics when combined with TTO.¹⁷ For skeletally immature patients with a lateralized tibial tubercle, traditional TTOs are contraindicated because they would damage the tibial tubercle apophysis. This has led to the development of physeal-sparing techniques, such as the Roux-Goldthwait procedure and the Galeazzi technique.^5,8,33 These approaches aim to address the lateralizing force on the patella by rerouting half of the patellar tendon medially (Roux-Goldthwait) or by anchoring medial hamstrings to the inferomedial patella. Grammont et al¹⁸ described a technique in 1985 in which the patellar tendon was detached from the tibial tubercle, preserving only the distal-most tendon-periosteal connection. This approach allowed the tendon to be resecured medially with periosteal fixation and included a lateral release. Krause et al²² later reported a modified Grammont technique in which no lateral release was performed and the medial side of the tendon was not affixed to the periosteum. Using this modified technique alone, Krause et al observed a 19% (n = 65) redislocation rate with a mean follow-up of 8 years.²²

To our knowledge no study has compared MPFLR alone versus MPFLR with a Grammont procedure in pediatric patients. We hypothesized that combining MPFLR with the Grammont procedure, compared with MPFLR alone, would reduce recurrent patellar instability, achieve medial shift of the tendon insertion, and maintain a comparable complication rate.

Methods

After institutional review board approval was granted, a retrospective review was conducted of pediatric patients (age ≤19 years at the time of surgery) at a single, Level 1 pediatric trauma center, who underwent MPFLR alone or MPFLR + Grammont procedure (Figures 1 -3) for patellar instability between January 2010 and March 2023. Patients with <3 months of postoperative follow-up, congenital or fixed dislocations, simultaneous trochleoplasty/TTO procedure, and revision cases were excluded. Patients were not excluded for incomplete imaging in the postoperative course.

Figure 1.

STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) flow chart. G, Grammont; MPFLR, medial patellofemoral ligament reconstruction.

Figure 2.

Grammont technique on a left knee. (A) The incision is marked slightly lateral to midline, extending from the distal pole of the patella (white star) along the patellar tendon (PT, white rectangle) to the midportion of the tibial tubercle (TT, white triangle). Note the lateralized attachment of the PT on the TT. (B) The incision is carried down to the peritenon, and flaps are elevated medially and laterally. (C) The peritenon is incised along the lateral border of the patellar tendon (white rectangle) and elevated off the tendon. (D) The tendon (white rectangle) is reflected off the tibial tubercle (white triangle), with care taken not to damage the apophysis. The dissection is carried down distally to the periosteum, and a periosteal flap is elevated. The entire flap is allowed to translate medially. (E) In all cases, the tendon (white rectangle) was repositioned and secured medially using a suture anchor (white arrow, 2 sutures used in this case). The distal extensor mechanism attachment was medialized approximately 1.5 cm, leaving the tibial tubercle exposed (white triangle).

Figure 3.

(A) Schematic of a skeletally immature right knee with a lateralized tibial tubercle. The vector of force between the quadriceps tendon and patellar tendon is not colinear, creating a Q-angle. The area marked with a double “x” represents a more medialized position, which could be achieved with a modified-Grammont procedure. (B) The dotted lines represent where the sharp dissection should occur. The patellar tendon is detached off the tibial tubercle apophysis. This dissection continues distally as the tendon blends with the periosteum. Note that the distal-most portion of the periosteum is left attached. A curved apex, lateral sharp dissection as shown allows the patellar tendon–periosteal complex to translate as 1 unit medially over the double “x.” With knee flexion and extension, the tendon is allowed to find a new medialized position before fixation. (C) The new position of the medialized patellar tendon over the double “x” position. The quadriceps tendon vector of force and patellar tendon are now colinear. The dotted line represents the still-attached distal periosteum. Fixation at the double “x” position can be performed with bone anchors or direct suture repair to the periosteum.

Patients were separated into 2 groups: MPFLR only and MPFLR + Grammont. Group assignment was based on the procedure performed as determined by the treating surgeon, reflecting clinical and radiographic decision-making. A total of 79 patients were included in this study, with 43 undergoing MPFLR only and 36 undergoing MPFLR + Grammont procedure. Data collection was done through retrospective chart review of the electronic medical record without any prospective patient contact or standardized follow-up protocol. This included patient demographic characteristics (age at surgery, sex, and body mass index [BMI]), clinical characteristics (operative side, time from initial instability to surgery, history of multiple dislocations), contralateral instability/surgery, alignment/rotational correction, simultaneous osteochondral surgical treatment, and postoperative course. History of multiple dislocations was recorded as a binary variable (yes/no), based on whether the patient reported >1 prior dislocation at the time of the initial visit. Alignment or rotational correction was captured as a binary variable (yes/no), defined as the performance of any adjunctive procedure intended to modify anatomic or rotational alignment, including TTO, distal femoral osteotomy, or rotational osteotomy. Patient follow-up was conducted according to individual surgeon preference and clinical indications, without a standardized protocol for the study; during these routine follow-up appointments, advanced imaging or additional evaluations were obtained only as deemed necessary by the treating surgeon. Postoperative complications were graded by the modified Clavien-Dindo classification after retrospective chart review.⁴³

The Clavien-Dindo classification is a standardized grading system that grades surgical complications based on the type of treatment required, stratifying complications into 5 main grades (I through V).⁴³ Sink et al⁴³ adapted this classification for orthopaedic surgery by preserving the original 5-grade structure while incorporating orthopaedic-specific definitions. This modified version was specifically tested and validated in the context of hip surgery and has since been applied to multiple orthopaedic specialties.

Preoperative and postoperative imaging data were obtained using radiographs and magnetic resonance imaging (MRI). Imaging variables collected include preoperative Caton-Deschamps ratio, preoperative TT-TG distance, postoperative Caton-Deschamps ratio, correction TT-TG distance, and final Caton-Deschamps ratio. Caton-Deschamps ratio was evaluated on lateral knee radiographs in 30° of knee flexion. Preoperative and postoperative coronal plane alignment was assessed using full-length hip-to-ankle radiographs. MRI measurements (preoperative TT-TG distance, patellar tilt, trochlear inclination, Dejour classification, and postoperative patellar tendon insertion–trochlear groove [PTI-TG] distance for MPFLR + Grammont and postoperative TT-TG distance for MPFLR only) were made on axial cuts on T2-weighted MRI sequences. Figure 4 provides details on how TT-TG and PTI-TG measurements were performed. Postoperative TT-TG distance and correction TT-TG distance were collected for 17 patients who had a postoperative MRI scan. Reviewers were not blinded to whether measurements were preoperative or postoperative. Reviewers were blinded to the specific surgical procedure performed (MPFLR alone or MPFLR + Grammont).

Figure 4.

Pre- and postoperative tibial tubercle–trochlear groove (TT-TG) and patellar tendon insertion–trochlear groove (PTI-TG) distance measurements on axial magnetic resonance imaging scan in a patient who underwent medial patellofemoral ligament reconstruction + Grammont. (A) TT-TG distance measurement: A line is drawn through the tibial tubercle at the center of the patellar tendon insertion perpendicular to the axis of the posterior femoral condyles. (B) A line is drawn through the deepest point of the trochlear groove perpendicular to the posterior femoral condyles. The horizontal distance between these parallel lines represents the preoperative TT-TG distance. In this patient, TT-TG distance is 20.28 mm. (C) Postoperative PTI-TG distance measurement: A line is drawn through the tibial tubercle at the center of the medialized patellar tendon insertion perpendicular to the axis of the posterior femoral condyles. (D) A line is drawn through the deepest point of the trochlear groove perpendicular to the posterior femoral condyles. The horizontal distance between these parallel lines represents the PTI-TG distance. In this patient, PTI-TG distance is 7.60 mm.

Statistical Analysis

Continuous variables were assessed using intraclass correlation coefficient (ICC), whereas kappa statistics were used for categorical variables. Categorical variables were assessed by the senior author (E.L.N.) to minimize variability in interpretation.

Unpaired t test and nonparametric Mann-Whitney U tests were used to compare normally and nonnormally distributed continuous data, respectively, between the 2 groups determined by a Shapiro-Wilk test. Chi-square test and/or Fisher exact test (if any of the expected cell sizes were <5) was used to compare binary or categorical data between 2 groups.

Each patient's measurements were evaluated by 4 raters. The 4 raters included 3 attending orthopaedic surgeons (E.L.N., E.S., S.I.A.) and 1 medical student (A.D.). The ICC was calculated to assess interrater reliability between raters. Interobserver reliability was assessed among the 4 raters using a subset of the first 26 patients in the data set to ensure measurement consistency before dividing the remaining cases among the raters. The measurements duplicated for reliability assessment included preoperative radiographic Caton-Deschamps ratio, full-length angle, preoperative TT-TG distance, lateral trochlear inclination angle, and patellar tilt. For continuous variables, ICCs were calculated using a 2-way random-effects model based on single ratings and absolute agreement. The mean ICC values, along with 95% CI, were reported to determine the level of agreement, with ICC values categorized as poor (<0.40), fair (0.40-0.59), good (0.60-0.74), and excellent (0.75-1.0).⁷

All statistical tests were 2-sided and performed at the .05 level of significance. Statistical analysis was performed using R statistical software, Version 4.3.2 (R Core Team, 2020).

Results

Patients in the MPFLR + Grammont group were younger at time of surgery (median age 13.4 vs 15.3 years; P = .013) and more patients in that group were female (80.6% vs 55.8%; P = .016). No significant differences in BMI or operative side were found between the groups (Table 1). The time from initial instability event to surgery was comparable across both groups (median 26.4 months; P = .548). The rate of preoperative patella alta was 57.1% in the MPFLR + Grammont group and 48.4% in the MPFLR-only group (P = .478). The mean preoperative TT-TG was higher in the MPFLR + Grammont group (18.8 vs 15.7 mm; P = .001).

Table 1

Demographic and Clinical Characteristics Between MPFLR-Only and MPFLR + Grammont Groups ^a

Characteristic	Overall (N = 79)	MPFLR Only (n = 43)	MPFLR + Grammont (n = 36)	P
Age at surgery, y	14.3 (8.2 to 18.3)	15.3 (9.7 to 18.3)	13.4 (8.2 to 17.6)	.013
Sex
Male	26 (33.3)	19 (45.2)	7 (19.4)	.016
Female	53 (67.1)	24 (55.8)	29 (80.6)
Body mass index	22.5 (16.2 to 41.6)	22.5 (16.9 to 41.6)	22.4 (16.2 to 40.6)	.753
Operative side
Left	48 (60.8)	24 (55.8)	24 (66.7)	.325
Right	31 (39.2)	19 (44.2)	12 (33.3)
Time from initial instability to surgery, mo	26.4 (0.5 to 122.2)	26.6 (0.5 to 115.4)	26.1 (2.0 to 122.2)	.548
History of multiple dislocations
No	57 (72.2)	29 (67.4)	28 (77.8)	.307
Yes	22 (27.8)	14 (32.6)	8 (22.2)
Alignment correction
No	69 (87.3)	39 (90.7)	30 (83.3)	.327
Yes	10 (12.7)	4 (9.3)	6 (16.7)
Rotational correction
No	77 (97.5)	41 (95.3)	36 (100.0)	.190
Yes	2 (2.5)	2 (4.7)	0 (0.0)
Osteochondral surgery
No	56 (70.9)	30 (69.8)	26 (72.2)	.811
Yes	23 (29.1)	13 (30.2)	10 (27.8)
Time from surgery to final postoperative follow-up, mo	18.9 (3.4 to 67.0)	14.2 (3.9 to 67.0)	21.9 (3.4 to 63.7)	.028
Time to return to activity, mo	7.5 (–6.1 to 26.9)	7.7 (–6.1 to 17.9)	7.5 (2.8 to 26.9)	.899

Data are expressed as median (range) or n (%). P values were obtained from the Mann-Whitney U test for continuous variables and the chi-square test for categorical variables. Boldface indicates statistical significance. MPFLR, medial patellofemoral ligament reconstruction.

The median follow-up time for the entire cohort was 18.9 months (range, 3.4-67.0 months), and a significant difference was observed in the MPFLR + Grammont group (21.9 months) compared with the MPFLR-only group (14.2 months). Grade II or III complications occurred in 25.3% of patients (MPFLR, 27.9%; MPFLR + Grammont, 22.2%; P = .563) (Table 2). Recurrent instability was observed in 12 (15.2%) patients, with rates of 20.9% in the MPFLR-only group and 8.3% in the MPFLR + Grammont group (P = .120). Revision surgery for recurrent patellar instability was performed in 4 patients (9.3%) in the MPFLR-only group and 1 patient (2.8) in the MPFLR + Grammont group. Extension and flexion deficits occurred at similar rates in the MPFLR-only and MPFLR + Grammont groups (Table 2). In the MPFLR + Grammont group, 1 patient had a growth arrest of the tibial tubercle apophysis resulting in recurvatum deformity. Acquired valgus deformity occurred in 1 patient in the MPFLR + Grammont group.

Table 2

Postoperative Complications ^a

Complications	Overall(N = 79)	MPFLR Only(n = 43)	MPFLR + Grammont(n = 36)	P
Complication grade II or III	20 (25.3)	12 (27.9)	8 (22.2)	.563
Recurrent instability	12 (15.2)	9 (20.9)	3 (8.3)	.120
Revision surgery for recurrent instability	5 (6.3)	4 (9.3)	1 (2.8)	.236
Postoperative extension deficit	1 (1.3)	1 (2.3)	0 (0.0)	.357
Postoperative flexion deficit	3 (3.8)	2 (4.7)	1 (2.8)	.664
Pain	6 (7.6)	3 (7.0)	3 (8.3)	.821
Infection	1 (1.3)	1 (2.3)	0 (0.0)	.271
Recurvatum deformity	1 (1.3)	0 (0.0)	1 (2.8)	.271
Acquired valgus deformity	1 (1.3)	0 (0.0)	1 (2.8)	.271

Values are expressed as n (%). P values were obtained from the chi-square test for categorical variables. Extension deficit was defined as >5° extension loss compared with the contralateral knee. Flexion deficit was defined as >15° flexion loss compared with the contralateral knee.

Preoperative TT-TG distance was significantly greater in the MPFLR + Grammont group (18.8 mm) compared with the MPFLR-only group (Table 3). Postoperative PTI-TG distance correction was greater in the MPFLR + Grammont group compared with the MPFLR-only group, but the difference was not statistically significant (Table 3). Final Caton-Deschamps ratio was significantly greater in the MPFLR + Grammont group (Table 3). Among patients stratified by Dejour classification, the rates of grade II/III complications were 38.9% for type A, 13.3% for type B, 14.3% for type C, and 75.0% for type D (P = .01) (Table 4). Interrater reliability of these measurements is shown in Table 5.

Table 3

Preoperative and Postoperative Imaging Outcomes ^a

Imaging Outcome	MPFLR Only	MPFLR + Grammont	P
Preoperative Caton-Deschamps ratio	1.3 (0.9 to 2.1)	1.3 (0.7 to 1.9)	.337
Preoperative Caton-Deschamps class
Normal	15 (48.4)	15 (42.9)	.478
Patella alta	15 (48.4)	20 (57.1)
Patella baja	1 (3.2)	0 (0.0)
Preoperative TT-TG, mm	15.7 (9.8 to 26.1)	18.8 (12.8 to 27.0)	.001
Postoperative Caton-Deschamps ratio	1.2 (0.8 to 2.0)	1.2 (0.7 to 1.7)	.102
Postoperative Caton-Deschamps class
Normal	29 (82.9)	23 (65.7)	.101
Patella alta	6 (17.1)	12 (34.3)
Postoperative correction distance ^b			.697
TT-TG distance	1.6 (–8.8 to 24.0)
PTI-TG distance		6.1 (–3.1 to 11.4)
Final Caton-Deschamps ratio	1.2 (0.9 to 1.7)	1.3 (1.0 to 1.8)	.015
Final Caton-Deschamps class
Normal	20 (76.9)	17 (53.1)	.061
Patella alta	6 (23.1)	15 (46.9)

In the MPFLR + Grammont group, PTI-TG distance was used instead of TT-TG distance because medialization of the patellar tendon insertion renders TT-TG an unreliable measure of extensor mechanism alignment.

Table 4

Complications vs Preoperative Dejour Classification ^a

Complications	Preoperative Dejour Classification				P
Complications	A (n = 18)	B (n = 45)	C (n = 7)	D (n = 4)	P
Complication grade II or III	7 (38.9)	6 (13.3)	1 (14.3)	3 (75.0)	.01
Recurrent instability	4 (22.2)	5 (11.1)	0 (0.0)	0 (0.0)	.351
Growth arrest or deformity	1 (5.6)	0 (0.0)	0 (0.0)	0 (0.0)	.369
Infection	1 (5.6)	0 (0.0)	0 (0.0)	0 (0.0)	.369
Pain	2 (11.1)	0 (0.0)	1 (14.3)	1 (25.0)	.05
Other complications	4 (22.2)	6 (13.3)	0 (0.0)	2 (50.0)	.141

Values are expressed as n (%). P values were obtained from the chi-square test for categorical variables. Values are expressed as n (%). Boldface indicates statistical significance.

Table 5

Interrater Reliability Between 4 Raters Using Intraclass Correlation Coefficient ^a

Measurement	ICC (95% CI)	P	Agreement Level
Preoperative radiographic Caton-Deschamps index	0.72 (0.51-0.86)	<.001	Good
Full-length angle (valgus +/varus –)	0.82 (0.56-0.93)	<.001	Excellent
Preoperative TT-TG	0.77 (0.59-0.89)	<.001	Excellent
Lateral trochlear inclination angle	0.74 (0.57-0.86)	<.001	Good
Patellar tilt (MRI)	0.78 (0.63-0.89)	<.001	Excellent

ICC, intraclass correlation coefficient; MRI, magnetic resonance imaging; TT-TG, tibial tubercle–trochlear groove. Boldface indicates statistical significance.

Discussion

The treatment of patellar instability in pediatric patients is challenging due to an increased risk of recurrence with nonoperative treatment combined with the surgical considerations when treating a pediatric population.^{9,21,22,32,48} We compared the outcomes of MPFLR alone versus MPFLR + Grammont in pediatric patients and found that the latter group had comparably favorable outcomes in terms of recurrent instability rates. Additionally, we found similar rates of overall complications and specifically a low rate of growth disturbance in both groups, albeit in a more complex patient population seen in the MPFLR + Grammont group with younger age and greater TT-TG distance preoperatively.

In pediatric patients undergoing surgical treatment, many techniques and modifications have been described to achieve proximal and distal patellar realignment and stabilization while protecting the distal femoral and proximal tibia physis and tibial tubercle apophysis.^29,36,48 As techniques have evolved, physeal-sparing or physeal-protecting MPFLR has become the favored technique for restoring the medial ligamentous checkrein to lateral patellar instability.^{2,20,23,37,40} However, despite significant advancements in our understanding of MPFL anatomy and ability to perform anatomic reconstructions in the pediatric population, the rate of recurrent instability after isolated MPFLR in pediatric patients is reported between 7.0% to 20.5%.^§ This is significantly higher than reported rates of failure in adult MPFLR of 0.1% to 4.7%.^35,49,51 Specifically, Husen et al²⁰ noted in their cohort of skeletally immature patients an overall recurrence rate after surgical stabilization of 13.9%. The rate of failure increased to 20.5% in patients who had isolated MPFLR. The authors concluded that failure of surgical treatment could be mitigated by combining MPFLR with TTO.

Lateralization of the attachment of the patellar tendon on the tibial tubercle, typically quantified by TT-TG, has been established as a biomechanical risk factor for patellar instability.^6,9,13,34 In the adult population, TTO is often used to address increased TT-TG.^13,25 However, in a skeletally immature patient, TTO carries the risk of growth arrest or disturbance of the proximal tibia, which can result in recurvatum, genu valgum, and/or patella baja.^27,38,47 The Grammont procedure was originally described for distal patellar tendon medialization in combination with a Langenskiöld procedure (reconstructive soft tissue procedure) in the setting of congenital or fixed patellar dislocation.^18,22,28,29 Recently, the Grammont procedure has been used as a physeal-protecting soft tissue realignment technique to address elevated TT-TG in idiopathic pediatric lateral patellar instability, with small series reporting on favorable outcomes and safety profile.^22,28,50 Kraus et al²² performed a modified Grammont without a concomitant medial stabilization (ie, MPFLR or medial plication) and reported an early redislocation rate of 12.3%. Wegmann et al⁵⁰ performed MPFLR with a modified Grammont on 7 knees with median TT-TG distance of 17 mm and reported no cases of recurrent dislocation at median 50 months postoperative. However, the efficacy and safety of MPFLR combined with Grammont procedure cannot be determined from these prior studies due to limitations of small sample sizes and lack of comparison groups.

From a technical standpoint, the Grammont procedure has a few key benefits over other distal realignment procedures. Compared with adult TTO, a soft tissue procedure poses less risk to the proximal tibial physis and apophysis. Additionally, the lack of osteotomy negates the risks of nonunion and hardware-related pain.^26,39 Postoperative pain and risk of compartment syndrome are similarly decreased. Compared with other soft tissue–only procedures such as the Roux Goldthwait, the Grammont procedure does not require splitting or complete detachment of the patellar tendon attachment.^8,16,25,33 Conversely, with the modified Grammont technique, actual medial translation of the patellar tendon insertion may be limited due to lack of medial fixation; with the Grammont technique used in the current study, the tendon was repositioned and secured medially using a suture anchor. It is also uncertain whether there is a risk of insufficiency of the distal patellar tendon attachment over time with this technique. Additionally, dissection of the patellar tendon off the tibial tuberosity still carries a risk of iatrogenic damage to the apophysis and proximal tibia growth arrest.

Our study aims to clarify some of these lingering questions regarding the safety and efficacy of the Grammont procedure by comparing isolated MPFLR versus MPFLR + Grammont. We found that the MPFLR + Grammont group was younger, with greater mean TT-TG distance compared with the isolated MPFLR group. This represents a high-risk group of patellar instability patients due to young age and combined biomechanical risk factors.²⁴ Overall, our cohort had a rate of postoperative instability (15.2%) comparable to what is reported in the literature. Although not statistically different, the rate was 8.3% in the MPFLR + Grammont group and 20.9% in the MPFLR-only group. Overall complications were similar between the 2 groups; however, the MPFLR + Grammont group had 1 case of recurvatum deformity that occurred due to arrest of the anterior proximal tibia physis. The patient was asymptomatic despite the deformity and was treated with observation. However, this finding highlights the need for close radiographic monitoring in these patients, with routine lateral knee or preferably full-length lateral extremity films. Caution should be exercised when applying this technique in younger patients with significant growth remaining and the highest risk of developing a clinically significant deformity in the setting of physeal injury. There was also 1 case of valgus deformity in the MPFLR + Grammont group. The patient was noted to have bilateral genu valgum deformity stemming from the distal femur, first noted approximately 2 years after MPFLR + Grammont surgery. Growth plates were noted to be open, and the patient underwent guided growth with complete resolution of valgus deformity. This is likely a case of acquired bilateral valgus rather than a case of growth disturbance secondary to surgery. However, given that we cannot completely exclude the latter possibility, the case was included as a complication. Given the response to guided growth, we can at least conclude that no growth arrest resulted from MPFLR surgery but we cannot rule out a growth disturbance or overgrowth occurring at the medial distal femur that led to the valgus deformity on the MPFLR-Grammont knee.

It remains unclear which patients may benefit most from the addition of the Grammont procedure. Unlike more established adult thresholds for surgical indications for TTO, elevated TT-TG distance in pediatric patients differs by age and sex.^31,32 Generally, the TT-TG distance has been shown to increase gradually during growth, and therefore younger patients may have a lower threshold for what can be considered “tolerated” TT-TG distance. Further, an increasing body of evidence shows that in adult patients, MPFLR with or without TTO yields similar outcomes and rates of recurrence.^{12,14,35,44,49} This emphasizes that we are still in the infancy of our understanding of the complex interplay of factors contributing to patellar instability, specifically in the case of pediatric patients. Future directions require prospective randomized studies with large numbers of skeletally immature patients, followed to skeletal maturity, to confirm our findings and further investigate long-term outcomes. This study is the first to directly compare MPFLR with and without distal soft tissue realignment in a pediatric population. Complications, including persistent postoperative pain and any episodes of recurrent instability (including mild subluxation), were systematically documented. Growth disturbances were also evaluated by comparing pre- and postoperative full-length radiographs.

The modified Grammont procedure shares similarities with the historical Hauser procedure in that both aim to correct recurrent patellar instability by altering the vector of the patellar tendon.¹⁹ The Hauser procedure was developed in 1938; a bone block containing the tibial tubercle and patellar tendon insertion is osteotomized and repositioned through a window in the proximal tibia, resulting in a fixed medial and posterior transfer designed to counteract abnormal lateralizing forces on the patella.^10,19 Although long-term follow-up studies reported satisfactory outcomes in approximately 70% of patients, with some series extending beyond 30 years, the procedure was ultimately abandoned due to an unacceptably high complication profile.^3,5,10 Reported complications included higher overall complication rates compared with alternative techniques, persistent instability in a subset of patients, patella baja, loss of extensor mechanism efficiency related to a reverse Maquet effect, and high rates of patellofemoral arthrosis at long-term follow-up.^3,5,10 Additional concerns included genu recurvatum when performed before epiphyseal closure, particularly relevant in pediatric populations.^3,5,10 In contrast, the Grammont procedure avoids TTO and bone block fixation, relying instead on soft tissue mobilization of the patellar tendon, thereby preserving tendon length and avoiding rigid posterior or distal transfer. Additionally, this spares the tibial tubercle apophysis from potential injury (although as noted in this study, growth arrest is still possible). These differences may reduce the risk of patella baja, excessive patellofemoral contact pressures, arthritis rates, and long-term extensor mechanism dysfunction. Although the shared principle of modifying patellar tendon alignment warrants careful long-term surveillance, the absence of a fixed bony transfer and the more physiologic, dynamic nature of the Grammont procedure suggest a lower likelihood of reproducing the deleterious patellofemoral mechanics historically associated with the Hauser procedure. This is an area that warrants further study and long-term follow-up.

Limitations

There are several limitations to this study inherent to its retrospective design. The study had a modest sample size, with 43 patients in the MPFLR-only group and 36 in the MPFLR + Grammont group. Selection bias may have been present as patients were not randomized, and there were no standardized criteria for selecting MPFLR alone versus MPFLR + Grammont. The MPFLR-only and MPFLR + Grammont groups were not fully equivalent with respect to age, sex distribution, patella alta, and TT-TG distance, which may have introduced confounding factors. This was a single-center experience, and the findings may not be generalizable. Patients up to 19 years of age were included without assessment of skeletal maturity (eg, Risser stage), which may introduce heterogeneity and affect generalizability of our findings. The follow-up period in this study was relatively short, and follow-up was done retrospectively through chart review. This may have limited the ability to detect late complications or longer term outcomes after surgery. In addition, formal standardized postoperative follow-up evaluations were not performed, and outcomes were therefore based on available clinical and radiographic data. Blinding to pre- versus postoperative status was not feasible, which may have introduced measurement bias during radiographic data collection. Interrater reliability testing was limited to the preoperative variables. As a result, postoperative measurements were not included in the reliability analysis. No postoperative functional testing was conducted, including objective strength or extensor mechanism testing; as a result, potential changes in extensor torque generation after posterior transfer of the patellar tendon could not be directly assessed and should be considered when interpreting the findings. We were unable to reliably calculate potential posterior displacement of the patellar tendon insertion after the Grammont procedure. On a number of patients’ MRI scans, we noted remodeling and medialization of the tibial tubercle apophysis after the Grammont procedure. This finding would benefit from longer term MRI follow-up; further, the potential for patellofemoral compartment degeneration after Grammont was outside the scope of this particular study. A limited number of patients underwent postoperative MRI (17/79) and usually only in the setting of recurrent instability or persistent pain. Therefore, the true amount of patellar tendon medialization achieved with the Grammont technique is unclear.

Conclusion

In this retrospective cohort study, we observed similar rates of success and a comparable complication profile, with low risk of growth disturbance, when MPFLR was combined with a Grammont procedure in pediatric patients with patellar instability and an increased TT-TG distance. Although further investigation is needed to determine the biomechanical and anatomic thresholds and the ideal patient population for its application, this technique has the potential to enhance surgical outcomes and reduce the rates of recurrent postoperative patellar instability in carefully selected pediatric patients.

Footnotes

Final revision submitted January 29, 2026; accepted February 16, 2026.

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

IRB approval for this study was obtained from Children's National.

ORCID iDs

Anthony Dure

Emily L. Niu

Notes

References

Abouelsoud

Abdelhady

Elshazly

Anatomic physeal-sparing technique for medial patellofemoral ligament reconstruction in skeletally immature patients with ligamentous laxity. Eur J Orthop Surg Traumatol. 2015;25(5):921-926. doi:10.1007/s00590-015-1618-1

Accadbled

Kerdoncuff

Hidalgo Perea

Green

DW.

Failure of isolated medial patellofemoral ligament reconstruction in children: risk factors and management. J Child Orthop. 2023;17(1):34-39. doi:10.1177/18632521221149055

Barbari

Raugstad

Lichtenberg

Refvem

The Hauser operation for patellar dislocation: 3-32-year results in 63 knees. Acta Orthop Scand. 1990;61(1):32-35. doi:10.3109/17453679008993061

Buuck

Fuklerson

JP.

Anteromedialization of the tibial tubercle: a 4- to 12-year follow-up. Oper Tech Sports Med. 2000;8(2):131-137. doi:10.1053/otsm.2000.6581

Chrisman

Snook

Wilson

TC.

A long-term prospective study of the Hauser and Roux-Goldthwait procedures for recurrent patellar dislocation. Clin Orthop Relat Res. 1979;(144):27-30.

Christensen

Sanders

Pareek

Mohan

Dahm

Krych

AJ.

Risk factors and time to recurrent ipsilateral and contralateral patellar dislocations. Am J Sports Med. 2017;45(9):2105-2110. doi:10.1177/0363546517704178

Cicchetti

DV.

Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol Assess. 1994;6(4):284-290. doi:10.1037/1040-3590.6.4.284

Clark

Metcalfe

Wogan

Mandalia

Eldridge

Adolescent patellar instability: current concepts review. Bone Joint J. 2017;99-B(2):159-170. doi:10.1302/0301-620X.99B2.BJJ-2016-0256.R1

Danielsen

Poulsen

Eysturoy

Mortensen

Hölmich

Barfod

KW.

Trochlea dysplasia, increased TT-TG distance and patella alta are risk factors for developing first-time and recurrent patella dislocation: a systematic review. Knee Surg Sports Traumatol Arthrosc. 2023;31(9):3806-3846. doi:10.1007/s00167-022-07255-1

10.

DeCesare

WF.

Late results of Hauser procedure for recurrent dislocation of the patella. Clin Orthop Relat Res. 1979;(140):137-144.

11.

Dejour

Walch

Nove-Josserand

Guier

Factors of patellar instability: an anatomic radiographic study. Knee Surg Sports Traumatol Arthrosc. 1994;2(1):19-26. doi:10.1007/BF01552649

12.

Dennis

Marmor

Propp

, et al. Isolated medial patellofemoral ligament reconstruction for recurrent patellar instability regardless of tibial tubercle-trochlear groove distance and patellar height: minimum 5-year outcomes. Am J Sports Med. 2024;52(9):2196-2204. doi:10.1177/03635465241260039

13.

Drapeau-Zgoralski

Swift

Caines

Kerrigan

Carsen

Pickell

Lateral patellar instability. J Bone Joint Surg Am. 2023;105(5):397-409. doi:10.2106/JBJS.22.00756

14.

Erickson

Nguyen

Gasik

Gruber

Brady

Shubin Stein

BE.

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

15.

Farrow

Alentado

Abdulnabi

Gilmore

Liu

RW.

The relationship of the medial patellofemoral ligament attachment to the distal femoral physis. Am J Sports Med. 2014;42(9):2214-2218. doi:10.1177/0363546514539917

16.

Fondren

Goldner

Bassett

III . Recurrent dislocation of the patella treated by the modified Roux-Goldthwait procedure: a prospective study of forty-seven knees. J Bone Joint Surg Am. 1985;67(7):993-1005.

17.

Franciozi

Ambra

Albertoni

LJB

, et al. Anteromedial tibial tubercle osteotomy improves results of medial patellofemoral ligament reconstruction for recurrent patellar instability in patients with tibial tuberosity-trochlear groove distance of 17 to 20 mm. Arthroscopy. 2019;35(2):566-574. doi:10.1016/j.arthro.2018.10.109

18.

Grammont

Latune

Lammaire

IP.

Die Behandlung der Subluxation und Luxation der Kniescheibe beim Kind. Technik von Elmslie mit beweglichem Weichteilstiel (8-Jahres-Ubersicht) [Treatment of subluxation and dislocation of the patella in the child. Elmslie technic with movable soft tissue pedicle (8 year review)]. Orthopade. 1985;14(4):229-238.

19.

Hauser

The classic: total tendon transplant for slipping patella: a new operation for recurrent dislocation of the patella. Clin Orthop Relat Res. 2006;452:7-16. doi:10.1097/01.blo.0000238831.50186.87

20.

Husen

Milbrandt

Shah

Krych

Stuart

Saris

DBF

. Medial patellofemoral ligament reconstruction using allografts in skeletally immature patients. Am J Sports Med. 2023;51(6):1513-1524. doi:10.1177/03635465231164400

21.

Koh

Stewart

Patellar instability. Clin Sports Med. 2014;33(3):461-476. doi:10.1016/j.csm.2014.03.011

22.

Kraus

Lidder

Švehlík

, et al. Patella re-alignment in children with a modified Grammont technique. Acta Orthop. 2012;83(5):504-510. doi:10.3109/17453674.2012.736168

23.

Leite

CBG

Hinckel

Ribeiro

, et al. Medial patellofemoral ligament reconstruction in skeletally immature patients without correction of bony risk factors leads to acceptable outcomes but higher failure rates. J ISAKOS. 2023;8(3):189-196. doi:10.1016/j.jisako.2023.02.003

24.

Ling

Brady

Arendt

, et al. Development of a multivariable model based on individual risk factors for recurrent lateral patellar dislocation. J Bone Joint Surg Am. 2021;103(7):586-592. doi:10.2106/JBJS.20.00020

25.

Longo

Rizzello

Ciuffreda

, et al. Elmslie-Trillat, Maquet, Fulkerson, Roux Goldthwait, and other distal realignment procedures for the management of patellar dislocation: systematic review and quantitative synthesis of the literature. Arthroscopy. 2016;32(5):929-943. doi:10.1016/j.arthro.2015.10.019

26.

Lundeen

Macalena

Agel

Arendt

High incidence of complication following tibial tubercle surgery. J ISAKOS. 2023;8(2):81-85. doi:10.1016/j.jisako.2022.11.005

27.

McCall

Ratts

Soft-tissue realignment for adolescent patellar instability. J Pediatr Orthop. 1999;19(4):549-552. doi:10.1097/00004694-199907000-00026

28.

Musielak

Premakumaran

Janusz

Dziurda

Koch

Walczak

Good outcomes of modified Grammont and Langenskiöld technique in children with habitual patellar dislocation. Knee Surg Sports Traumatol Arthrosc. 2021;29(6):1983-1989. doi:10.1007/s00167-020-06284-y

29.

Pilone

Bonasia

Rosso

, et al. Medial patellofemoral ligament reconstruction and nonanatomic stabilization techniques in skeletally immature patients. Joints. 2019;7(3):98-106. doi:10.1055/s-0039-3400451

30.

Pritsch

Haim

Arbel

Snir

Shasha

Dekel

Tailored tibial tubercle transfer for patellofemoral malalignment: analysis of clinical outcomes. Knee Surg Sports Traumatol Arthrosc. 2007;15(8):994-1002. doi:10.1007/s00167-007-0325-9

31.

Pruneski

O’Mara

Perrone

Kiapour

AM.

Changes in anatomic risk factors for patellar instability during skeletal growth and maturation. Am J Sports Med. 2022;50(9):2424-2432. doi:10.1177/03635465221102917

32.

Randsborg

Banitalebi

Årøen

Straume-Næsheim

The tibial tuberosity-trochlear groove distance can either increase or decrease during adolescent growth. Children (Basel). 2024;11(5):504. doi:10.3390/children11050504

33.

Redler

Wright

ML.

Surgical management of patellofemoral instability in the skeletally immature patient. J Am Acad Orthop Surg. 2018;26(19):e405-e415. doi:10.5435/JAAOS-D-17-00255

34.

Sanders

Pareek

Hewett

Stuart

Dahm

Krych

AJ.

High rate of recurrent patellar dislocation in skeletally immature patients: a long-term population-based study. Knee Surg Sports Traumatol Arthrosc. 2018;26(4):1037-1043. doi:10.1007/s00167-017-4505-y

35.

Sappey-Marinier

Sonnery-Cottet

O’Loughlin

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

36.

Schlichte

Sidharthan

Green

Parikh

SN.

Pediatric management of recurrent patellar instability. Sports Med Arthrosc Rev. 2019;27(4):171-180. doi:10.1097/JSA.0000000000000256

37.

Schlumberger

Schuster

Hofmann

, et al. Midterm results after isolated medial patellofemoral ligament reconstruction as first-line surgical treatment in skeletally immature patients irrespective of patellar height and trochlear dysplasia. Am J Sports Med. 2021;49(14):3859-3866. doi:10.1177/03635465211050419

38.

Sezgin

Szerb

Tóth

Hangody

Treatment of a combined deformity following tibial tubercle ventromedialization with single proximal tibial closing wedge osteotomy. Eklem Hastalik Cerrahisi. 2019;30(3):316-321. doi:10.5606/ehc.2019.19002

39.

Shah

Uppstrom

Rizy

Gomoll

Strickland

SM.

Incidence of complications after tibial tubercle osteotomy and tibial tubercle osteotomy with distalization. Am J Sports Med. 2024;52(5):1274-1281. doi:10.1177/03635465241235883

40.

Shamrock

Day

Duchman

Glass

Westermann

RW.

Medial patellofemoral ligament reconstruction in skeletally immature patients: a systematic review and meta-analysis. Orthop J Sports Med. 2019;7(7):2325967119855023. doi:10.1177/2325967119855023

41.

Shea

Polousky

Jacobs

Jr , et al. The patellar insertion of the medial patellofemoral ligament in children: a cadaveric study. J Pediatr Orthop. 2015;35(4):e31-e35. doi:10.1097/BPO.0000000000000399

42.

Shea

Styhl

Jacobs

Jr , et al. The relationship of the femoral physis and the medial patellofemoral ligament in children: a cadaveric study. Am J Sports Med. 2016;44(11):2833-2837. doi:10.1177/0363546516656366

43.

Sink

Leunig

Zaltz

Gilbert

Clohisy

; Academic Network for Conservational Hip Outcomes Research Group. Reliability of a complication classification system for orthopaedic surgery. Clin Orthop Relat Res. 2012;470(8):2220-2226. doi:10.1007/s11999-012-2343-2

44.

Song

Wang

Yan

, et al. Tibial tubercle osteotomy may not provide additional benefit in treating patellar dislocation with increased tibial tuberosity-trochlear groove distance: a systematic review. Arthroscopy. 2021;37(5):1670-1679.e1. doi:10.1016/j.arthro.2020.12.210

45.

Tanaka

Chahla

Farr

II , et al. Recognition of evolving medial patellofemoral anatomy provides insight for reconstruction. Knee Surg Sports Traumatol Arthrosc. 2019;27(8):2537-2550. doi:10.1007/s00167-018-5266-y

46.

Uppstrom

Price

Black

Gausden

Haskel

Green

DW.

Medial patellofemoral ligament (MPFL) reconstruction technique using an epiphyseal femoral socket with fluoroscopic guidance helps avoid physeal injury in skeletally immature patients. Knee Surg Sports Traumatol Arthrosc. 2019;27(11):3536-3542. doi:10.1007/s00167-019-05412-7

47.

Vähäsarja

Kinnunen

Lanning

Serlo

Operative realignment of patellar malalignment in children. J Pediatr Orthop. 1995;15(3):281-285. doi:10.1097/01241398-199505000-00002

48.

Vavken

Wimmer

Camathias

Quidde

Valderrabano

Pagenstert

Treating patella instability in skeletally immature patients. Arthroscopy. 2013;29(8):1410-1422. doi:10.1016/j.arthro.2013.03.075

49.

Vivekanantha

Kahlon

Cohen

de Sa

. Isolated medial patellofemoral ligament reconstruction results in similar postoperative outcomes as medial patellofemoral ligament reconstruction and tibial-tubercle osteotomy: a systematic review and meta-analysis. Knee Surg Sports Traumatol Arthrosc. 2023;31(6):2433-2445. doi:10.1007/s00167-022-07186-x

50.

Wegmann

Würnschimmel

Kraus

, et al. Medial patellofemoral ligament (MPFL) reconstruction in combination with a modified Grammont technique leads to favorable mid-term results in adolescents with recurrent patellofemoral dislocations. Knee Surg Sports Traumatol Arthrosc. 2018;26(3):705-709. doi:10.1007/s00167-017-4550-6

51.

Wilkens

Hannink

van de Groes

SAW

. Recurrent patellofemoral instability rates after MPFL reconstruction techniques are in the range of instability rates after other soft tissue realignment techniques. Knee Surg Sports Traumatol Arthrosc. 2020;28(6):1919-1931. doi:10.1007/s00167-019-05656-3