Precise arthroscopic mini-trochleoplasty combined with medial patellofemoral ligament reconstruction yields comparable complications to isolated medial patellofemoral ligament reconstruction

Abstract

BACKGROUND

Trochlear dysplasia (TD) is the most prevalent risk factor of recurrent patellar dislocation (RPD). Surgeons remain hesitant to perform open trochleoplasty even in severe TD, due to high complication rates. A goal to reduce complications underlies the development of arthroscopic trochleoplasty. Our method, precise arthroscopic mini-trochleoplasty (PAM trochleoplasty), involves minimal osteotomy and damage to patellofemoral joint. Medial patellofemoral ligament reconstruction (MPFLR) is the most frequently selected safe procedure for RPD treatment. MPFLR is associated with low complication rates. No prior study has compared outcomes and complications between combined arthroscopic trochleoplasty with MPFLR and isolated MPFLR in RPD treatment.

AIM

To compare outcomes and complications, at 2-year follow-up in patients who underwent either combined PAM trochleoplasty with MPFLR or isolated MPFLR.

METHODS

Thirty patients who underwent PAM + MPFLR, and 20 patients isolated MPFLR were included. Clinical outcomes assessment consisted of patient-reported outcome measure scores (PROMs), including visual analog scale (VAS), International Knee Documentation Committee (IKDC) score, Knee Injury and Osteoarthritis Outcome Score (KOOS), Lysholm score, and Tegner activity score, and complications. Subgroups of cases with type B TD of Dejour classification were assessed. Radiological evaluation included tibial tubercle-trochlear groove (TT-TG) distance, lateral trochlear inclination (LTI), and lateral patellar tilt (LPT).

RESULTS

PAM + MPFLR group demonstrated more severe multiple dysplasia compared to MPFLR group (P < 0.001). At 2-year follow-up, all PROMs significantly improved in both groups (P < 0.001-0.05). No significant differences were observed between groups in VAS score, KOOS subscales for pain, symptoms, activities of daily living, and quality of life, or Lysholm score (P > 0.05). However, MPFLR group demonstrated significantly higher IKDC and Tegner scores (P < 0.05). In type B TD subgroups, no significant differences were found in most PROMs, except for IKDC score. Postoperative radiological outcomes showed that PAM + MPFLR group had significantly greater improvements in TT-TG distance, LTI, and LPT (P < 0.001), whereas MPFLR group showed significant improvement only in LPT. There were no significant differences in complication rates between two groups or within subgroups (P > 0.05).

CONCLUSION

Both PAM + MPFLR and isolated MPFLR result in favorable clinical outcomes and comparable complication rates. These findings suggest PAM + MPFLR may be considered as an option for treating RPD with severe TD.

Key Words: Recurrent patellar dislocation; Precise arthroscopic mini-trochleoplasty; Medial patellofemoral ligament reconstruction; Trochlear dysplasia; Patient-reported outcome measure scores

Core Tip: Open trochleoplasty for recurrent patellar instability with trochlear dysplasia (TD) demonstrates high complication rates. Our technique combines medial patellofemoral ligament reconstruction (MPFLR) with precise arthroscopic mini-trochleoplasty, yielding good clinical outcomes, and allowing for minimally invasive correction of severe TD with complication rates similar to those of conventional MPFLR. This provides a comprehensive arthroscopic solution for recurrent patellar dislocation associated with severe TD.

INTRODUCTION

Patellar instability is a common condition that primarily affects adolescents[1], leading to pain, loss of knee function, osteochondral lesions, and potential long-term development of patellofemoral arthritis. Approximately 30% to 70% of patients who experience a patellar dislocation develop recurrent patellar instability. Trochlear dysplasia (TD) is the most prevalent anatomic risk factor and the strongest predictor of recurrent patellar instability[2,3]. Several studies have reported favorable clinical outcomes using open trochleoplasty combined with medial patellofemoral ligament reconstruction (MPFLR) for the treatment of recurrent patellar dislocation (RPD) in patients with severe TD[4-7]. Although severe TD is a clear indication for trochleoplasty, many surgeons remain hesitant to perform this procedure, even in cases of high - grade dysplasia[8-10]. High complication rates, such as knee arthrofibrosis, increased patellofemoral pain, need for reoperation, and development of patellofemoral osteoarthritis-have been reported[11-13], which may deter the use of open trochleoplasty in clinical practice.

Since 2010, several reports have described various techniques for arthroscopic trochleoplasty[14-18], all of which utilize a thin osteochondral flap in the sulcus-deepening procedure. Compared with the technique introduced by Blønd and Barfod[14], who pioneered arthroscopic trochleoplasty, our method-referred to as precise arthroscopic mini-trochleoplasty (PAM trochleoplasty)[15]-involves a smaller osteotomy. PAM trochleoplasty is specifically designed to minimize potential damage or disruption to the patellofemoral joint and adjacent structures. Previous studies[19,20] on arthroscopic trochleoplasty, with follow-up periods ranging from 1 years to 5 years, have reported favorable clinical outcomes and improvements in certain patellar instability risk factors, including lateral trochlear inclination (LTI) and lateral patellar tilt (LPT), though not in tibial tubercle-trochlear groove (TT-TG) distance[19].

Survey-based studies have shown that MPFLR is the most frequently selected procedure for the surgical treatment of patellar instability, whereas trochleoplasty is far less commonly considered[8,9]. Unlike open trochleoplasty, MPFLR is associated with relatively low complication rates[21]; in particular, knee arthrofibrosis, increased patellofemoral pain, and reoperation are rarely observed within a 2-year follow-up period[22]. To our knowledge, no prior study has compared the clinical outcomes and complication rates between combined arthroscopic trochleoplasty with MPFLR and isolated MPFLR in the treatment of RPD. Therefore, the purpose of this study was to compare clinical and radiological outcomes, as well as complication rates, at a 2-year follow-up in patients who underwent either combined PAM trochleoplasty with MPFLR or isolated MPFLR. Clinical outcomes were assessed by recording the recurrence of patellar dislocation or subluxation, stiffness, reoperation, and various complications, along with patient-reported outcome measures (PROMs), including the visual analog scale for usual pain (VAS)[23], International Knee Documentation Committee (IKDC) score[24], Knee Injury and Osteoarthritis Outcome Score (KOOS)[25], Lysholm score[26], and Tegner activity score[26]. Radiological evaluation included three key risk factors for patellar instability: TT-TG distance, LTI, and LPT. It was hypothesized that combining PAM trochleoplasty with MPFLR would yield non-inferior KOOS - knee scores and VAS pain scores at 24 months compared with MPFLR alone, while maintaining comparable major complication rates.

MATERIALS AND METHODS

Patient selection

This retrospective cohort study obtained institutional review board approval from the Ethics Committee of the First Affiliated Hospital of the Air Force Medical University (No. KY20222161-C-1). Medical records of patient with patellar dislocation who underwent MPFLR, with or without PAM trochleoplasty, between June 2019 and February 2023 were reviewed.

The inclusion criteria were: (1) Two or more episodes of patellar dislocation; or (2) One episode of patellar dislocation accompanied by one or more episodes of patellar subluxation. The exclusion criteria included: (1) Previous knee surgery or revision surgery; (2) Open epiphyseal growth plates; (3) Habitual or fixed patellar dislocation; (4) Concomitant ligament reconstruction or other bony procedures; (5) Refusal to participate in the study; and (6) Loss to follow-up.

Based on the aforementioned criteria, 50 patients were included from a total of 86 screened MPFLR cases (Figure 1). Eligible patients were divided into two groups: The concomitant PAM trochleoplasty with MPFLR group (PAM + MPFLR) and the isolated MPFLR group. Patients with a normal trochlea or mild TD (type A TD according to the Dejour classification[27]), as well as those with severe TD (type B, C, or D) who declined the trochleoplasty procedure, were assigned to the MPFLR group (n = 20). Patients who refused to undergo trochleoplasty may introduce treatment allocation bias. In accordance with ethical principles, only MPFLR was performed on these cases. The remaining eligible patients with severe TD (type B, C, or D) were included in the PAM + MPFLR group (n = 30). All surgical procedures were performed by a single senior author to ensure consistency in surgical technique. Nevertheless, it introduced single - surgeon bias, which may limit external validity. Basic demographic data were collected, including age, sex, surgical side, and duration of follow-up.

Figure 1 Flowchart of the patient selection. Precise arthroscopic mini-trochleoplasty precise arthroscopic mini-trochleoplasty, medial patellofemoral ligament reconstruction medial patellofemoral ligament reconstruction. PAM: Precise arthroscopic mini-trochleoplasty; MPFLR: Medial patellofemoral ligament reconstruction.

Surgical technique

The PAM trochleoplasty was performed according to the technique reported by Xu et al[15] in 2020. Preoperative three-dimensional computed tomography (CT) imaging was used to meticulously design the new trochlear sulcus, ensuring precise alignment and integration with the patient’s individual anatomy, including the repositioned patella following MPFLR (a1 in Figure 2A). The area of bone resection and the osteotomy volume were substantially smaller than those in standard open sulcus-deepening trochleoplasty. The osteotomy was limited to the supratrochlear spur and the proximal portion of the trochlear bump (Figure 2). Using additional superolateral and superomedial portals, the supratrochlear spur and proximal trochlear bump were accurately targeted and resected, and a tunnel-shaped subchondral groove was created. The new groove was created 3-5 mm laterally to the original groove; as a result, the postoperative TT-TG distance may be reduced.

Figure 2 Preoperative and postoperative three-dimensional computed tomography images of a case of left knee recurrent patellar dislocation with type D Dejour trochlear dysplasia. A: Before surgery, a spur on the proximal side of the trochlear groove and subluxation of the patella could be seen. The black-dashed line was the plan for the new groove (a1); two days after precise arthroscopic mini-trochleoplasty (PAM trochleoplasty), the spur was removed, and two pieces of subchondral bone attached to the osteochondral flap could be seen on the new groove (black arrow). Reduction of the patella to a normal position fit for a new groove showed good patellofemoral congruence (a2); subchondral bone healing resulted in a smooth groove at 7 months after surgery (a3); bone remodeling of the groove and better patellofemoral congruence at 2 years after surgery (a4); B: Before surgery, a sagittal image illustrated the spur (white arrow) on the proximal side of the groove (b1); two days after PAM trochleoplasty, the spur and part of the trochlear bump were removed (white arrow), and subchondral bone pieces attached to the osteochondral flap could be seen (b2); bone healing of the subchondral bone formed a smooth groove at 7 months after surgery (b3); bone remodeling of the new groove at 2 years after surgery (b4). *: Lateral row anchors of suture bridge for fixation of osteochondral flap.

Maximal preservation of the original condylar ridge height was essential to maintain patellar stability. A 2-mm subchondral bone layer was preserved on the osteochondral flap. The borders of the new groove were trimmed, and the flap was molded to be sufficiently malleable for bending and conformation to the new sulcus. The flap was then firmly pressed onto the underlying bone using a suture bridge fixation technique to facilitate rapid postoperative bone healing. Both the joint capsule and the cartilage of the groove remained intact, and no additional incisions on the osteochondral flap were necessary to accommodate the new position.

MPFLR was performed using the semitendinosus tendon. Two 4.5-mm TWINFIX PEEK anchors (Smith & Nephew, MA, United States) were inserted into the superior half of the medial patellar facet, just adjacent to the patellar cartilage. The prepared graft was then secured to the bone using sutures from the anchors. Intraoperative C-arm fluoroscopy was utilized to identify Schöttle et al’s point[28] on a standard lateral knee view. A guide pin was drilled through Schöttle et al's point[28] in a proximal direction to avoid interference with the previously placed three anchors [one 4.5-mm TWINFIX PEEK anchor (Smith & Nephew, MA, United States)] and two 4.9-mm Versalok anchors (DePuy Synthes, Johnson & Johnson, NJ, United States), which were used for osteochondral flap fixation in the PAM + MPFLR group. The graft was passed through a soft-tissue tunnel between layers 2 and 3 of the medial patellar region, tensioned to approximately 0.5 pounds of force, and fixed in the femoral bone tunnel using an absorbable interference screw with the knee flexed at 90° to prevent overtightening.

Postoperative rehabilitation

Postoperatively, a knee brace was applied for six weeks in both groups. The brace was adjusted to allow 0° to 60° of mobilization beginning on postoperative day 3, increased to 0° to 90° from day 4 through week 4, and then progressed to an unrestricted range of motion after week 4. Quadriceps strengthening exercises were initiated immediately following surgery, while straight-leg raises commenced on postoperative day 3.

In the MPFLR group, partial weight bearing was permitted during weeks 1 to 2, followed by full weight bearing after week 2. In the PAM + MPFLR group, both the partial and full weight-bearing phases were delayed by one additional week.

Clinical assessment

Clinical assessments were conducted preoperatively and at the two-year postoperative follow-up. Postoperative evaluations included documentation of patellar dislocation or subluxation recurrence, positive apprehension test, reduction in range of motion (> 15°), reoperation, and other complications such as increased pain, stiffness, infection, and deep vein thrombosis.

PROMs were collected preoperatively and at two years postoperatively. These included the VAS for pain[23], the IKDC[24], KOOS[25], Lysholm[26], and Tegner activity scores[26].

Radiological evaluation

All patients included in the study underwent CT scans preoperatively and at two years postoperatively. Several well-established risk factors for patellar instability, including TT-TG distance[29], LTI[30,31], and LPT[32], were evaluated using CT imaging (Figure 3). The literatures reported the threshold of the TT-TG distance measured by CT was 20 mm[29], or 16.4 mm in Chinese population[30], the threshold of LTI measured by magnetic resonance imaging (MRI) was 11°[31], or 16.2° by CT in Chinese population[30], and the threshold of LPT measured by CT was 20°[29], or 17.2° in Chinese population[30]. In this study, the figures for the thresholds measured by CT in the Chinese population[30] were used as reference. Trochlear dysplasia is widely recognized as the most critical predisposing factor for patellar instability[33,34]. In this study, trochlear morphology was evaluated according to Dejour classification[27] as follows: Normal, type A, B, C, or D of TD. The subgroups were composed of cases with type B TD in the PAM + MPFLR or MPFLR group, respectively. The use of CT instead of MRI imaging limited cartilage assessment. However, CT’s superior bone detail, shorter acquisition time, and wider availability in our institution facilitated the inclusion.

Figure 3 Measurement of patellar instability risk factors, lateral trochlear inclination and lateral patellar tilt. Lateral trochlear inclination was the angle between lateral trochlear facet line (blue line b) and posterior condyles line (green line a). Lateral patellar tilt was the angle between patellar axis line (orange line c) and posterior condyles line (green line a). Green dash line a’ is the parallel line of line a.

All measurements were independently conducted using Adobe Photoshop CC2018 by an orthopedic consultant and a senior resident, both of whom had substantial experience in recurrent patellar dislocation and related imaging measurements. Interobserver and intraobserver reliabilities were assessed. For intraobserver reliability, radiological measurements were repeated by the senior resident after a four-week interval. Intraclass correlation coefficient (ICC) values with 95% confidence intervals were calculated; an ICC value greater than 0.8 was considered indicative of excellent reliability.

Statistical analysis

All statistical analyses were conducted using SPSS software, version 22.0 (SPSS Inc., Chicago, IL, United States). The normality of data distribution was assessed using the Kolmogorov-Smirnov test. Continuous variables were analyzed using the two-tailed Student’s t test for normally distributed data or the Mann-Whitney U test for nonparametric data. Categorical variables were compared using Pearson’s χ² test or Fisher’s exact test, as appropriate. A P value < 0.05 was considered statistically significant.

Sample size calculation was performed using G*Power software version 3.1.9.7 (Heinrich Heine, Universität Düsseldorf, Düsseldorf, Germany). A priori power analysis, with a prespecified α level < 0.05 and an effect size of 0.4, indicated that a minimum sample size of 50 patients was required to detect significant differences with a statistical power of 0.86.

RESULTS

The basic demographic characteristics are presented in Table 1. No significant differences were observed between the PAM + MPFLR and MPFLR groups in terms of age, sex, side of involvement, or duration of follow-up. However, the severity of trochlear dysplasia differed significantly between groups, with the PAM + MPFLR group demonstrating more severe dysplasia compared to the MPFLR group (P < 0.001).

Table 1 Patient demographics, n (%)/mean ± SD.

	PAM + MPFLR group (n = 30)	MPFLR group (n = 20)	P value	PAM + MPFLR cases with type B TD (n = 14)	MPFLR cases with type B TD (n = 10)	P value
Age (year)	25.47 ± 8.27	24.25 ± 7.28	0.5961	25.36 ± 8.39	28.30 ± 7.30	0.3821
Sex (male/female) (n)	6/24	7/13	0.2362	2/12	5/5	0.0853
Side (L/R) (n)	17/13	9/11	0.4192	10/4	5/5	0.2852
Follow-up (month)	27.27 ± 5.19	27.55 ± 4.47	0.8431	27.64 ± 6.67	27.40 ± 4.38	0.9211
Dejour type			0.0003
Normal	0 (0)	0 (0)
Type A	0 (0)	8 (40.0)
Type B	14 (46.7)	10 (50.0)
Type C	3 (10.0)	1 (5.0)
Type D	13 (43.3)	1 (5.0)

¹Mann-Whitney U test.

²Pearson’s χ² test.

³Fisher’s exact test.

PAM: Precise arthroscopic mini-trochleoplasty; MPFLR: Medial patellofemoral ligament reconstruction; TD: Trochlear dysplasia.

Type B TD was the most common subtype in both the PAM + MPFLR (46.7%) and MPFLR (50%) groups. These cases were included in the respective type B subgroups. The second most frequent subtype in the PAM + MPFLR group was type D (43.3%), whereas in the MPFLR group, it was type A (40%). Since type B TD was the largest subtype, there were no cases of type A in the PAM + MPFLR group, and very few cases of type C/D in the MPFLR group, the type B TD subgroups were selected for analysis.

Clinical assessment

In both groups, all PROMs improved significantly at two years postoperatively compared with preoperative values (P < 0.001-0.05, Table 2). Prior to surgery, the MPFLR group reported significantly greater pain than the PAM + MPFLR group, as indicated by the VAS score and the pain subscale of the KOOS (P < 0.001). Similar findings were observed in the subgroups with type B TD. No significant differences were found between the two groups or between the two subgroups in other PROMs before surgery (P > 0.05, Table 2). Postoperatively, the MPFLR group demonstrated significantly higher scores than the PAM + MPFLR group in the IKDC (P < 0.001), the sports and recreation subscale of the KOOS (P < 0.05), and the Tegner activity score (P < 0.05). No significant differences were observed between the groups in VAS score, KOOS subscales for pain, symptoms, activities of daily living, and quality of life, or in the Lysholm score (P > 0.05, Table 2).

Table 2 Comparison of patient-reported outcome measure scores between the precise arthroscopic mini-trochleoplasty + medial patellofemoral ligament reconstruction and medial patellofemoral ligament reconstruction groups and subgroups with type B trochlear dysplasia.

	PAM + MPFLR group (n = 30)	MPFLR group (n = 20)	P value1	PAM + MPFLR cases with type B TD (n = 14)	MPFLR cases with type B TD (n = 10)	P value1
VAS score
Preoperative	1.73 ± 1.48	5.45 ± 2.67	0.000	1.71 ± 1.27	5.00 ± 2.63	0.000
Postoperative	1.03 ± 0.85	0.95 ± 0.83	0.733	1.21 ± 0.89	0.80 ± 0.92	0.280
P value1	0.029	0.000		0.238	0.000
IKDC score
Preoperative	58.93 ± 22.71	57.64 ± 16.60	0.829	56.32 ± 24.01	54.14 ± 7.05	0.784
Postoperative	77.28 ± 9.58	87.13 ± 6.07	0.000	77.83 ± 7.62	86.44 ± 5.33	0.006
P value1	0.000	0.000		0.004	0.000
KOOS score
Pain
Preoperative	80.09 ± 15.56	65.83 ± 14.19	0.002	78.57 ± 16.43	64.72 ± 13.74	0.041
Postoperative	90.83 ± 6.93	92.64 ± 3.85	0.295	90.67 ± 6.94	93.06 ± 3.98	0.341
P value1	0.001	0.000		0.017	0.000
Symptoms
Preoperative	68.93 ± 20.29	67.86 ± 19.15	0.852	72.70 ± 18.46	70.71 ± 15.13	0.782
Postoperative	87.50 ± 9.37	87.32 ± 7.81	0.944	89.80 ± 7.66	84.64 ± 8.26	0.130
P value1	0.000	0.000		0.004	0.020
ADL
Preoperative	74.02 ± 21.90	67.06 ± 20.16	0.262	72.90 ± 22.53	68.68 ± 15.76	0.616
Postoperative	90.88 ± 6.69	92.35 ± 2.68	0.356	91.60 ± 7.15	91.62 ± 2.20	0.992
P value1	0.000	0.000		0.006	0.000
Sport/Rec
Preoperative	61.33 ± 27.70	53.75 ± 21.51	0.307	58.57 ± 26.71	54.50 ± 18.17	0.681
Postoperative	77.33 ± 17.06	86.00 ± 5.28	0.033	78.57 ± 17.91	86.50 ± 4.74	0.188
P value1	0.013	0.000		0.028	0.000
QOL
Preoperative	55.42 ± 21.88	55.94 ± 21.41	0.934	54.02 ± 24.95	58.13 ± 9.79	0.628
Postoperative	80.42 ± 8.00	82.50 ± 6.60	0.339	80.80 ± 5.73	81.25 ± 7.80	0.873
P value1	0.000	0.000		0.001	0.000
Lysholm score
Preoperative	51.40 ± 27.49	48.95 ± 7.84	0.700	49.29 ± 28.71	48.80 ± 8.00	0.959
Postoperative	87.73 ± 9.46	89.65 ± 3.39	0.390	87.00 ± 7.76	88.40 ± 2.88	0.593
P value1	0.000	0.000		0.000	0.000
Tegner activity score
Preoperative	2.87 ± 1.96	1.90 ± 1.25	0.057	2.71 ± 1.86	1.60 ± 1.07	0.104
Postoperative	5.03 ± 1.54	5.95 ± 1.10	0.026	4.93 ± 1.49	5.90 ± 1.20	0.103
P value1	0.000	0.000		0.002	0.000

¹Mann-Whitney U test.

Data are expressed as n or mean ± SD unless otherwise indicated. PAM: Precise arthroscopic mini-trochleoplasty; MPFLR: Medial patellofemoral ligament reconstruction; TD: Trochlear dysplasia; VAS: Visual analog scale; IKDC: International Knee Documentation Committee; KOOS: Knee Injury and Osteoarthritis Outcome Score; ADL: Activities of daily living; QOL: Quality of life.

In the comparison of postoperative outcomes between subgroups with type B TD, no significant differences were found in most PROMs, with the exception of the IKDC score.

Radiological evaluation

The ICC values for radiological measurements, including TT-TG distance, LTI, and LPT, indicated excellent reliability. Interobserver ICCs ranged from 0.826 to 0.929, and intraobserver ICCs ranged from 0.912 to 0.971, demonstrating strong intra- and interobserver consistency.

Compared with preoperative values, the PAM + MPFLR group showed significant improvements in TT-TG distance, LTI, and LPT at two years postoperatively (P < 0.001; Table 3). In contrast, the MPFLR group demonstrated significant improvement only in LPT (P < 0.001), with no significant changes observed in TT-TG distance or LTI (P > 0.05).

Table 3 Comparison of patellar instability risk factors between the two groups.

	PAM + MPFLR group (n = 30)	MPFLR group (n = 20)	P value1
TT-TG distance (mm; threshold: 16.4 mm[26])
Preoperative	18.30 ± 2.86	15.47 ± 3.24	0.002
Postoperative	13.90 ± 2.32	14.59 ± 2.22	0.306
P value1	0.000	0.324
LTI (°; threshold: 16.2°[26])
Preoperative	-4.91 ± 11.00	10.52 ± 7.07	0.000
Postoperative	15.67 ± 9.81	10.98 ± 6.96	0.071
P value1	0.000	0.839
LPT (°; threshold: 17.2°[26])
Preoperative	29.33 ± 9.05	26.05 ± 7.12	0.179
Postoperative	11.58 ± 8.13	14.13 ± 6.87	0.254
P value1	0.000	0.000

¹Mann-Whitney U test.

Data are expressed as n or mean ± SD unless otherwise indicated. PAM: Precise arthroscopic mini-trochleoplasty; MPFLR: Medial patellofemoral ligament reconstruction; LTI: Lateral trochlear inclination; TT-TG: Tibial tubercle-trochlear groove; LPT: Lateral patellar tilt.

Preoperatively, the PAM + MPFLR group exhibited a significantly larger TT-TG distance (P = 0.002) and a significantly lower LTI (P < 0.001) compared to the MPFLR group (Table 3). Before the operation, the PAM + MPFLR group had an abnormal TT-TG distance of 18.3 mm, whereas the MPFLR group had a normal one (15.47 mm). The results showed that, in addition to trochlear dysplasia, the PAM + MPFLR group simultaneously had a more significant malalignment of the extensor than the MPFLR group. After the operation, the TT-TG distance in the PAM + MPFLR group decreased to a normal value of 13.9 mm, which was smaller than in the MPFLR group (14.57 mm), but the difference was not statistically significant. Before the operation, both the PAM + MPFLR and MPFLR groups had abnormal LTI (-4.91° and 10.52°, respectively). After the operation, the LTI of the PAM + MPFLR group was significantly increased to 15.67°, which was very close to the standard threshold[30] of 16.2°, whereas the MPFLR group stayed in an abnormal value of 10.98, which was nearly same as that of pre-operation.

Complication rates

No significant differences in complication rates were observed between the PAM + MPFLR and MPFLR groups, including redislocation, postoperative subluxation, positive apprehension sign, reduced range of motion, reoperation, increased pain, stiffness, infection, and deep vein thrombosis (P > 0.05, Table 4). Similarly, no significant differences were found between the subgroups with type B TD (P > 0.05, Table 4). Residual instability of the PAM + MPFLR type B TD subgroup was 7.1%, whereas the MPFLR subgroup was 40%. However, no significant difference was found (P = 0.075).

Table 4 Comparison of postoperative complications and clinical data between the precise arthroscopic mini-trochleoplasty + medial patellofemoral ligament reconstruction and medial patellofemoral ligament reconstruction groups, n (%).

	PAM + MPFLR group (n = 30)	MPFLR group (n = 20)	P value1	PAM + MPFLR cases with type B TD (n = 14)	MPFLR cases with type B TD (n = 10)	P value1
Redislocation	1 (3.3)	0 (0)	1.000	0 (0)	0 (0)	1.000
Postoperative subluxation	0 (0)	0 (0)	1.000	0 (0)	0 (0)	1.000
Apprehension sign	2 (6.7)	5 (25.0)	0.100	1 (7.1)	4 (40.0)	0.075
Residual instability	3 (10.0)	5 (25.0)	0.240	1 (7.1)	4 (40.0)	0.075
Reduced range of motion	1 (3.3)	0 (0)	1.000	1 (7.1)	0 (0)	1.000
Reoperation	1 (3.3)	0 (0)	1.000	1 (7.1)	0 (0)	1.000
Increased pain	2 (6.7)	1 (5.0)	1.000	0 (0)	0 (0)	1.000
Stiffness	0 (0)	0 (0)	1.000	0 (0)	0 (0)	1.000
Infection	0 (0)	0 (0)	1.000	0 (0)	0 (0)	1.000
Deep vein thrombosis	1 (3.3)	0 (0)	1.000	0 (0)	0 (0)	1.000

¹Fisher’s exact test.

PAM: Precise arthroscopic mini-trochleoplasty; MPFLR: Medial patellofemoral ligament reconstruction; TD: Trochlear dysplasia.

DISCUSSION

The key findings of this study were as follows: (1) Both concomitant PAM trochleoplasty with MPFLR and isolated MPFLR yielded favorable clinical outcomes and similarly low complication rates at the two-year follow-up; and (2) Compared with preoperative values, the combined PAM trochleoplasty with MPFLR demonstrated superior postoperative radiological outcomes compared to MPFLR alone.

Trochleoplasty, as a surgical intervention for patellar instability, has been shown to significantly enhance joint stability and improve functional outcomes[4,6,13]. In the present study, all PROMs, including VAS, IKDC score, KOOS, Lysholm score, and Tegner activity score, improved significantly at two years post-operatively in both the PAM + MPFLR and MPFLR groups (P < 0.001; Table 2). Similar improvements were observed in the subgroups with type B TD. Preoperatively, the PAM + MPFLR group demonstrated more severe trochlear dysplasia compared to the MPFLR group (P < 0.001, Table 1), and the PAM + MPFLR group had an abnormal TT-TG distance of 18.3 mm[30], whereas the MPFLR group had a normal one (15.47 mm, P = 0.002, Table 3). The PAM + MPFLR group had multiple dysplasia concomitantly, including trochlear dysplasia and extensor malalignment, and other potential dysplasia, probably because of more severe condition for an integrative case. Even with severe TD and more complex surgical procedures, PAM + MPFLR group improved significantly in all PROMs postoperatively, and had comparable outcomes in VAS score, KOOS subscales for pain, symptoms, activities of daily living, and quality of life, and Lysholm score with MPFLR-only group (P > 0.05, Table 2). The MPFLR group exhibited higher scores than the PAM + MPFLR group in the sports and recreation sub - scale of the KOOS (P = 0.03) and the Tegner activity score (P = 0.03). These P values were just below the threshold (0.05) of statistical significance, indicating that these differences may not correspond to clinical differences. As for the comparison of postoperative outcomes between the subgroups with type B TD in the PAM + MPFLR and MPFLR groups, no significant differences were found in VAS, KOOS, Lysholm score, and Tegner activity score (P > 0.05; Table 2). Preoperatively, the MPFLR group had significantly more pain than the PAM + MPFLR group in the VAS - U score and pain of KOOS (P < 0.001, Table 2). It may suggest baseline imbalance. Since the PAM + MPFLR group showed more severe multiple dysplasia than the MPFL group, the PAM + MPFLR group may have a more incompetent and lax MPFL. So, in the episode of patellar dislocation or subluxation, the MPFLR group with a relatively tight ligament may lead to a more serious acute MPFL tear and associated pain. Post - operatively, the pain scores showed no significant difference between the two groups (P > 0.05, Table 2).

Open trochleoplasty remains a technically demanding procedure for many surgeons due to its complexity and the potential for various complications. Survey data from Germany and Australia[8,9] indicated that surgeons often refrained from performing trochleoplasty, even in cases of severe TD, despite acknowledging its importance in managing patellar instability. A 15-year follow-up study on trochleoplasty reported a total failure rate of 20% and postoperative stiffness in 23% of cases[11]. A systematic review evaluating the three most commonly used trochleoplasty techniques demonstrated an overall complication rate of 40%, including increased pain, reduced range of motion, and osteoarthritis, although the rate of patellar redislocation was only 2%[13]. Additionally, a systematic review and meta-analysis of 1000 trochleoplasty procedures[5] reported postoperative patellofemoral osteoarthritis in 27% of knees, stiffness in 7%, and subsequent surgical intervention in 17%. Over the past decades, trochleoplasty techniques have continued to evolve to reduce complication rates, a goal that also underlies the development of the arthroscopic approach.

Blønd and Barfod[14] was the first surgeon to report an arthroscopic trochleoplasty procedure. The PAM trochleoplasty technique[15] was inspired by Blønd and Barfod’s procedure[14] and the Bereiter-Gautier subchondral sulcus-deepening technique. PAM trochleoplasty is a minimally invasive approach in which the osteotomy is limited to the supratrochlear spur and the proximal aspect of the trochlear bump. This targeting may influence the impingement between the postoperatively reduced patella and the trochlea. Prolonged impingement can contribute to the development of patellofemoral osteoarthritis. As a minimally invasive procedure, the PAM technique emphasizes maximal preservation of the original condylar ridge height, which is critical for maintaining patellar stability (a2 in Figure 2A and b2 in Figure 2B). After preserving a 2-mm subchondral bone layer and rendering the osteochondral flap sufficiently malleable, the flap was firmly pressed onto the underlying bone using a suture bridge fixation technique to promote rapid bone healing (a3 in Figure 2A and b3 in Figure 2B). The PAM technique increases the cost of 3 anchors, and the suture bridge technique does not require a long learning curve for a knee arthroscopic surgeon. Compared with open trochleoplasty, the PAM technique shows better generalizability because it does not require more expense and long - term learning. In this study, no cases of knee arthrofibrosis were observed among 30 patients who underwent PAM trochleoplasty, resulting in a stiffness rate of 0. In contrast, stiffness remains one of the most common complications of open trochleoplasty during short- to mid-term follow-up[11,13].

Compared to open trochleoplasty, MPFLR is associated with relatively low complication rates, and complications such as knee arthrofibrosis and increased patellofemoral pain are rare. Most surgeons consider MPFLR a safe procedure with favorable outcomes and low complication rates[8,9,21,22]. In this current study, preoperatively, compared with MPFLR group, the PAM + MPFLR group had multiple dysplasia concomitantly, including more severe TD (P < 0.001, Table 1) and extensor malalignment (abnormal and larger TT-TG distance, P = 0.002, Table 3), and other potential dysplasia probably, because of more severe condition for an integrative case. Nevertheless, the PAM + MPFLR group did not show higher complication rates than the MPFLR-only group (P > 0.05, Table 4). The same result was observed in the subgroup analysis of patients with type B TD (P > 0.05, Table 4). The residual instability rate was 7.1% in the PAM + MPFLR type B TD subgroup, whereas it was 40% in the MPFLR subgroup. Even with such a large difference in percentages, however, no statistically significant difference was found (P = 0.075, Table 4). The P values just higher than the threshold (0.05) of statistical significance may imply meaningful clinical differences. Since the total sample was only 24 cases, a type II error in statistical analysis may lead to a non-significant difference. In the future, a bigger sample may address this underpowered issue and eliminate the type II error.

Some studies demonstrated that the combination of arthroscopic deepening trochleoplasty and MPFLR significantly improved knee function in 37 cases with a follow-up period of 1-5 years[35], and in 16 cases with an average follow-up of 5 years[19]. In Blønd and Barfod’s 5-year follow-up study[19], LPT and LTI were significantly improved. In 2024, Bin Zainuddin et al[17] reported similar findings in 13 knees at a 1.5-year follow-up. However, the only study comparing arthroscopic trochleoplasty with open trochleoplasty was conducted by Riedl et al[20] in 2024. In that study, 20 patients who underwent arthroscopic trochleoplasty demonstrated outcomes comparable to those of 15 patients treated with open trochleoplasty at 1-year follow-up, and no significant MRI parameters difference was found between the two procedures. The MPFLR is the most common used and well accepted concomitant procedure with either open or arthroscopic trochleoplasty[36-39]. To our knowledge, no study has compared combined arthroscopic trochleoplasty and MPFLR with MPFLR alone. Therefore, it is still unknown whether combining arthroscopic trochleoplasty would increase complication rates compared with MPFLR alone. In this study, the PAM + MPFLR group had more severe dysplasia and more complex surgical procedures than the MPFLR-only group (P < 0.001, Table 1, P = 0.002, Table 3). Theoretically, more severe multiple dysplasia and more complex surgical procedures are associated with worse outcomes and higher complication rates; however, this study found comparable complication rates between the two groups, and PAM trochleoplasty did not increase the complication rate. The PAM + MPFLR procedure achieves the purpose of good outcomes and low complications for the surgical treatment of RPD with severe TD. PAM trochleoplasty offers a choice for surgeons who hesitate to decide on open trochleoplasty.

Regarding radiological evaluation, a study on open trochleoplasty[40] reported significant reductions in both TT-TG distance and sulcus angle following trochleoplasty combined with MPFLR. The present arthroscopic study similarly demonstrated that the PAM + MPFLR procedure significantly reduced the TT-TG distance (P < 0.001, Table 3). In addition, other risk factors, including LPT and LTI, were also improved by the PAM + MPFLR procedure. In contrast, isolated MPFLR did not result in significant changes in TT-TG distance or LTI (P > 0.05, Table 3). These findings differ from those reported by Blønd and Barfod[19], in which the TT-TG distance was not reduced following arthroscopic trochleoplasty.

Limitations

This study has several limitations. First, it involved a relatively small sample size and a follow-up period of only two years. The postoperative rate of patellar redislocation, subluxation, or positive apprehension sign (Table 4) was 10% in the PAM + MPFLR group and 25% in the MPFLR group, with corresponding rates of 7.1% in the PAM + MPFLR subgroup with type B TD (n = 14) and 40% in the MPFLR subgroup with type B TD (n = 10). Although the likelihood of preventing postoperative patellar instability appeared higher in the PAM + MPFLR group and subgroup than in the isolated MPFLR, statistical analysis revealed no significant differences between these groups (P = 0.075 in the subgroups, Table 4). This may be due to the limited sample size, which might not have been sufficient to detect a statistically significant difference. Additionally, our 24-month follow-up is insufficient to determine whether trochleoplasty accelerates or protects against osteoarthritis. There is still a need for at least a 5-year radiological and clinical review, ideally with sequential MRI to detect late cartilage deterioration.

A 15-year follow-up study of open trochleoplasty reported that osteoarthritis greater than Iwano grade 2 developed in 65% of cases[11]. In contrast, the two-year follow-up period in the present study is insufficient to adequately assess the progression of osteoarthritis. A longer follow-up and a larger sample size are needed to provide more definitive and reliable results. These are currently being pursued and will be reported in the future. Second, the study employed a non-randomized design, which may introduce biases and limit the ability to control for confounding variables. Third, due to the coronavirus disease 2019 pandemic during the study period, some cases and follow-up evaluations were lost. Fourth, because of limited access to MRI scanners and long patient wait times for MRI examinations at our institution, CT imaging was used for radiological evaluation. If both CT and MRI had been assessed, the conclusions drawn from this study would likely have been more robust.

CONCLUSION

In this level III, non-randomized study with a relatively short 2-year follow-up, even in PAM trochleoplasty cases with more severe multiple dysplasia, both the combined PAM trochleoplasty and MPFLR procedure and MPFLR alone yielded favorable clinical outcomes with similarly low complication rates at the 2-year follow-up. However, PAM trochleoplasty resulted in significantly greater improvements in radiological parameters, including TT-TG distance, LTI, and LPT, whereas MPFLR alone improved only LPT. These findings suggest PAM + MPFLR may be considered as a viable surgical option for treating recurrent patellar instability in patients with severe trochlear dysplasia.