Small, Short, Oblique Patellar Tunnels with Looped Graft Versus Suture Anchors for Patellar-Sided Graft Fixation During MPFL Reconstruction: A Study of Over 600 Knees

Abstract

Background:

There is a paucity of literature comparing clinical outcomes and complications between the use of 2 smaller (3.2 mm), short, oblique patellar tunnels with looped graft for patellar-sided graft fixation during medial patellofemoral ligament reconstruction (MPFLR) and the use of 2 suture anchors.

Purpose:

To compare clinical outcomes and complications in large series between the use of dual patellar suture anchors and dual, small, short, oblique bone tunnels for patellar-sided graft fixation during MPFLR.

Study Design:

Retrospective cohort study; Level of evidence, 3.

Methods:

A retrospective chart review identified all patients at a single academic institution who underwent primary MPFLR between March 2010 and December 2021, with a minimum 2-year follow-up. Operative notes, postoperative clinical follow-up notes, and radiographs were used to ascertain laterality, graft type, surgical technique, concomitant procedures, as well as the incidence of recurrent patellar instability, revision MPFLR, and patellar fracture. The follow-up was limited to a review of the postoperative clinic or emergency department visits, and patients were not contacted at the time of the study.

Results:

A total of 635 knees were included for final analysis. Patellar-sided graft fixation was achieved with small, short, oblique bone tunnels with looped grafts in 342 knees (53.9%) and suture anchors in 293 knees (46.1%). The small, oblique tunnels and suture anchor techniques both yielded a low incidence of patellar fracture, with rates of 0.3% and 0%, respectively. No significant difference in rates of patellar fracture existed between cohorts (P≥ .999). There were similar rates of recurrent patellar instability events in patients in whom patellar-sided graft fixation was performed with dual, small, short, oblique bone tunnels (n = 9; 2.6%) compared with dual suture anchors (n = 15; 5.1%; P = .103). Similarly, no significant difference was found in revision MPFLR rates between groups (dual small, short, oblique patellar tunnels: n = 5; 1.5%; dual suture anchors: n = 8; 2.7%).

Conclusion:

Two small (3.2 mm), short, oblique patellar tunnels with looped grafts and dual patellar suture anchors are similar in safety and efficacy for achieving patellar-sided graft fixation during MPFLR.

Patellar instability is a common condition that is estimated to occur in 5 to 43 cases per 10,000 persons annually.^6,12,14,15 Given its multifactorial nature, the evaluation and surgical management of patients with recurrent lateral patellar instability must be individualized to address the specific osseous and soft tissue anatomical abnormalities that contribute to its occurrence. ¹⁰ The medial patellofemoral ligament (MPFL) represents the primary soft tissue restraint to lateral patellar translation and is nearly always injured or incompetent in patients experiencing recurrent lateral patellar instability.^5,17 MPFL reconstruction (MPFLR) is the consensus surgical treatment for patients experiencing recurrent lateral patellar instability and has demonstrated excellent clinical outcomes and low rates of recurrent instability when performed in isolation or combination with other osseous procedures.^1,5,7,18,20

Despite extensive study, several controversies persist with regard to surgical techniques, graft sources, and modalities of graft fixation during MPFLR.^1,3,8-10 Several techniques exist for patellar-sided graft fixation during MPFLR—including the use of suture anchors and bone tunnels. ¹ Although no technique has emerged as being superior with regard to clinical outcomes, each modality has its respective advantages and disadvantages.^{1,4,9,10,13,17} The use of 4.5 mm, fully transverse patellar tunnels with looped grafts offers greater tendon-to-bone contact area for healing and allows for material cost savings compared with the use of suture anchors. ¹¹ It also eliminates the risk of mechanical failure due to anchor pull-out. ¹¹ Despite these advantages, some surgeons advocate for using 2 suture anchors for patellar-sided graft fixation, citing a tolerance for the use of a shorter graft and a reduced risk of articular surface violation and patellar fracture compared with 4.5 mm fully transpatellar tunnels. ¹¹ The incidence of patellar fracture with the use of 4.5 mm fully transpatellar tunnels has been found to occur between 3.3% to 3.6% of knees in 2 large case series.^16,19 A recent systematic review of MPFLR techniques demonstrated a wide range of patellar fracture rates with the use of bone sockets from 0% to 17%; however, this included various surgical techniques for bone socket formation. ²

A retrospective cohort study by Deasey et al ¹ demonstrated the use of smaller (3.2 mm), short, oblique patellar tunnels with looped grafts to result in a significantly decreased rate of recurrent patellar instability compared with the use of 2 suture anchors for patellar-sided graft fixation. In addition, the authors did not observe the same risk of patellar fracture previously observed with the use of larger (4.5 mm) fully transpatellar tunnels. To date, this study represents the only comparison of using 2 small (3.2 mm), short, oblique patellar tunnels to other modalities of patellar-sided graft fixation. Therefore, the primary objective of the present study was to compare clinical outcomes and complications in the largest series to date between the use of dual patellar suture anchors and dual, small (3.2 mm), short, oblique bone tunnels for patellar-sided graft fixation during MPFLR. We hypothesized that there would be no significant difference in rates of recurrent patellar instability, revision MPFLR surgery, or patellar fracture between groups.

Methods

Study Population and Design

Institutional review board approval was acquired to perform this retrospective cohort study of consecutive patients undergoing primary MPFLR at a single academic institution between March 2010 and December 2021. Patients were identified retrospectively by querying the Current Procedural Terminology (CPT) code for MPFLR (27427, 27424, 27422, and 27420) and subsequently reviewing the surgical indications and procedures performed for all potentially eligible patients. Outcomes were monitored via chart review. The mean follow-up was 41.3 months. Inclusion in this retrospective cohort study was limited to primary MPFLR and a minimum 2-year follow-up. Patients undergoing revision MPFLR or those who underwent concomitant femoral osteotomies or trochleoplasty procedures were excluded.

Indications and Technique for MPFLR

Patients were indicated for MPFLR in the setting of recurrent patellar dislocations that failed a trial of conservative management consisting of activity modification, physical therapy, and bracing, or a first-time patellar instability event resulting in an osteochondral lesion that was determined would benefit from surgery. All surgeries were performed by 1 of 6 fellowship-trained, high-volume orthopaedic surgeons on an outpatient basis (D.R.D.). If indicated by the surgeon, concomitant tibial tubercle osteotomy (TTO) with or without distalization was performed. This decision considered such criteria as tibial tubercle-trochlear groove distance (typically >20 mm), patellar tendon-lateral trochlear ridge distance (typically >9 mm), the presence of patella alta, or a combination of these factors. The need for lateral Z-lengthening was made intraoperatively if the lateral tissue about the knee was found to be tight after the MPFLR. Graft type—whether ipsilateral gracilis semitendinosus autograft or allograft—was similarly utilized based on surgeon preference.

Patellar Tunnel Technique

A full description of the surgical technique for MPFLR utilizing 2 small (3.2 mm), short, oblique patellar tunnels is described in detail by Deasey et al. ¹ In short, for patellar tunnels and looped grafts, the tunnels were created 1 cm apart in the proximal 40% of the patella using a 3.2 mm drill starting at the medial border of the patella at the osteochondral junction and directed anteriorly, emerging approximately one-third of the diameter across the patella while maintaining a bony bridge. A beath pin is then utilized to shuttle the hamstring tendon graft in a looped configuration through the tunnels. The graft is passed between layers 2 and 3 of the medial knee in anticipation of subsequent fixation to the femur at Schöttle point (Figure 1).

Figure 1.

Posteroanterior (left) and lateral (right) postoperative radiographs of the left knee MPFLR utilizing patellar tunnels with concomitant TTO. MPFLR, medial patellofemoral ligament reconstruction; TTO, tibial tubercle osteotomy.

Suture Anchor Technique

Two suture anchors (Gryphon Suture Anchor; Johnson & Johnson) are placed along the medial patellar border at the upper one-third to one-half of the patella with approximately 1 cm of space between which is prepared using a burr to encourage soft tissue to bone healing by creating a shallow trough. The hamstring tendon graft is then secured in each anchor. This technique is provided in further description by Deasey et al. ¹

After patellar-sided graft fixation, the graft ends are shuttled between layers 2 and 3 of the medial knee. Schottle point is then identified with a beath pin on a perfect lateral radiograph. After pin placement and confirmation of graft isometry, the beath pin is over-reamed with a 7 mm reamer. The graft ends are then pulled into the blind-ended tunnel and an interference screw is placed for femoral-sided graft fixation with approximately ∼0.5 pounds (2 N) of force being held through the graft with the knee at approximately 40 degrees of flexion, and while the patella is ensured to be centered within the trochlea (Figure 2).

Figure 2.

Posteroanterior (left) and lateral (right) postoperative radiographs of the right knee MPFLR utilizing suture anchors with concomitant TTO. MPFLR, medial patellofemoral ligament reconstruction; TTO, tibial tubercle osteotomy.

Evaluation Methods

A retrospective chart review was performed to ascertain demographic information, operative techniques, and relevant clinical outcomes for each identified patient. Demographic characteristics collected from chart review included patient age and sex. Operative notes were utilized to ascertain laterality, graft type, the technique utilized for patellar-sided graft fixation, and any concomitant procedures performed. The decision to perform revision MPFLR procedures was made on an individualized basis, with shared decision-making based on the clinical judgment of the attending surgeon and the respective patient's preferences.

Postoperative Data Collection

Postoperative follow-up was achieved via chart review of postoperative clinic or emergency department visits. For each patient included in the study, all notes available through the electronic medical record (EMR) system were reviewed for patient-reported episodes of recurrent patellar instability. This was done through a combination of manual and semi-automated data extraction via the EMR system, which utilizes key terms and CPT codes. Clinic notes and radiologic imaging were reviewed for evidence of patellar fracture and revision MPFLR. Patients were not contacted at the time of the study.

Statistical Analysis

Statistical analysis was performed with SPSS Statistics Version 24 (IBM) and R Version 3.6.0 (R Core Team). Given the use of categorical variables, chi-square or Fischer exact tests were utilized to examine associations between variables. In addition, because of the relative infrequency of the outcomes of interest, a Firth logistics regression was also performed to further validate the statistical analyses. Subgroup analysis was performed for patients who underwent isolated MPFLR without concomitant procedures to reduce confounding variables allowing for more direct comparison of the 2 differing techniques. Odds ratios (ORs) with student t tests, along with firth logistics regression, were performed on this subgroup of isolated MPFLR patients. Patients with hybrid gracilis/semitendinosus grafts (n = 3), hybrid autografts/allografts (n = 1), and in whom an unknown graft type (n = 1) was used were excluded from the analysis. P < .05 was required to reject the null hypothesis and to be considered statistically significant.

Results

A total of 635 knees in 560 patients were identified for inclusion in this study (98.76%). Overall demographic data for included patients is reported in Table 1. A total of 342 (53.9%) knees underwent patellar-sided graft fixation with small (3.2 mm), short, oblique bone tunnels, and 293 (46.1%) knees received suture anchors for patellar-sided graft fixation. The mean duration of follow-up was 41.3 months. Also, 337 of 635 (53.1%) knees underwent isolated MPFLR, whereas a tibial tubercle osteotomy was performed in 298 of 635 (46.9%) knees (anteromedializing TTO: n = 273 [43%]; anteromedializing and distalizing TTO: n = 87 [13.7%]) (Table 1). An outline of graft type utilized in the overall study cohort, and within each study group, is provided in Table 1.

Table 1.

An Overview of the Overall Demographic Information for Included Patients Undergoing Primary MPFLR ^a

Characteristics	Tunnel Cohort (n = 342; 53.9%)	Suture Cohort (n = 293; 46.1%)	P
Age, y	23.53 ± 9.51	22.57 ± 9.02	.1979
Sex, female	217 (63.3)	197 (67)	.3182
Laterality, left	186 (54.2)	152 (51.7)	.5842
Mean follow-up, mo	43.9	39.5	.4756
Concomitant tibial tubercle osteotomies	210 (61.2)	88 (39.9)	.0001
Anteromedializing	186 (54.2)	87 (29.6)
Distalizing	64 (19)	23 (7.9)
Concomitant lateral Z-lengthening	168 (49.1)	43 (14.7)	.0001
Graft type			.0001
Gracilis	320 (94.2)	122 (44.2)
Semitendinosus	22 (5.8)	171 (55.8)
Graft source			.0001
Autograft	322 (93.6)	130 (41.8)
Allograft	20 (6.4)	163 (58.2)

^aData are presented as mean ± SD or n (%). An outline of the incidence of the respective modality of patellar-sided graft fixation is provided, as well as the graft types utilized and rates of performance of concomitant anteromedializing and distalizing TTO. MPFLR, medial patellofemoral ligament reconstruction; TTO, tibial tubercle osteotomy.

Patellar Fracture

One patellar fracture was recorded, occurring 16 weeks postoperatively. The patient had undergone primary MPFLR using 2 small (3.2 mm), short, oblique bone tunnels with looped semitendinosus autografts for patellar-sided graft fixation with a concomitant anteromedializing tibial tubercle osteotomy (Figure 3).

Figure 3.

Posteroanterior (left), lateral (middle), and sunrise view (right) radiographs depicting the left knee with postoperative patellar fracture.

The rate of patellar fracture in the cohort of patients in which 2 small (3.2 mm), short, oblique bone tunnels with looped grafts were utilized for patellar-sided graft fixation was 1 of 342 (0.3%) knees (Table 2). There were no occurrences of patellar fracture in the cohort in which 2 suture anchors were utilized (Table 2). No statistically significant difference was found in rates of patellar fracture between groups (P≥ .999) (Table 2).

Table 2.

Comparison of MPFLR Outcomes Between the Use of Small (3.2 mm), Short, Oblique Tunnels and Suture Anchors for Patellar Fixation ^a

	Oblique Patellar Tunnels	Suture Anchors	Odds Ratio	P
Patellar fracture	1 (0.3 )	0	NA	NA
Revision MPFLR	5 (1.5 )	8 (2.7 )	1.892	.261
Recurrent instability	9 (2.6 )	15 (5.1 )	1.989	.103

^aData are presented as n (%). A comparison of the primary study outcomes of interest between the primary study cohorts of patients undergoing patellar-sided graft fixation during MPFLR with the use of 2 small (3.2 mm), oblique patellar tunnels or suture anchors. MPFLR, medial patellofemoral ligament reconstruction; NA, not applicable.

Recurrent Instability

Subsequent recurrent patellar instability events were identified in 24 of 635 (3.8%) knees in our overall study population. Also, 9 of 342 (2.6%) knees in the small (3.2 mm), short, oblique patellar tunnel cohort experienced recurrent instability events compared with 15 of 293 (5.1%) knees in the suture anchor cohort. The use of suture anchors was not associated with a statistically significant increased risk of symptomatic instability or dislocation compared with small, oblique tunnels (OR, 1.989; CI, 95%; P = .103) (Table 3). These results were validated by a Firth logistic regression (P = .109) (Table 3).

Table 3.

Firth Logistics Regression Analysis Evaluating the Incidence of Subsequent Patellar Instability and Revision MPFLR ^a

	Recurrent Instability	Revision Surgery
	P	P
Patellar tunnels vs suture anchors	.109	.35
Gracilis vs semi-T graft	.376	.25
Autograft vs allograft	.746	.88
TTO vs isolated MPFLR	.014 b	.034 b
Concomitant lateral Z-lengthening vs no lateral Z-lengthening	.389	.187

^aFirth logistics regression analysis was performed relative to the modality of patellar-sided graft fixation, the performance of concomitant osteotomies, and the graft type utilized. MPFLR, medial patellofemoral ligament reconstruction; Semi-T, semitendinosus; TTO, tibial tubercle osteotomy.

^bIndicates P < .05.

Thirteen knees in our overall study population underwent revision MPFLR. Also, 5 of 342 (1.5%) knees in the small (3.2 mm), short, oblique patellar tunnel cohort underwent revision MPFLR versus 8 of 293 (2.7%) knees in the suture anchor cohort. No statistically significant difference was found in the need for revision MPFLR between the 2 methods for patellar fixation (OR, 1.89; P = .261) (Table 2).

Subgroup Analysis

Subanalyses were performed to examine confounding variables with regard to the impact of patellar-sided graft fixation on the primary outcomes of interest. The impact of graft type (autograft vs allograft; gracilis vs semitendinosus) and the performance of a concomitant tibial tubercle osteotomy on the incidence of patellar fracture, recurrent instability events, and the need for revision MPFLR were investigated.

No statistically significant correlations were found with the primary outcomes concerning graft type or source (Tables 3, 4, and 5). The performance of a concomitant tibial tubercle osteotomy correlated with a statistically significant decrease in rates of recurrent instability events (P = .009) and revision MPFLR (P = .021) (Table 6). These results were validated by a Firth logistic regression (Table 3). No significant differences were observed between the small (3.2 mm), short, oblique patellar tunnel and suture anchor cohort in subgroup analysis undergoing isolated MPFLR without concomitant procedures (Tables 7 and 8).

Table 4.

Comparison of MPFLR Outcomes Relative to Graft Source ^a

	Autograft	Allograft	Odds Ratio	P
Patellar fracture	0	1 (0.5)	NA	NA
Revision MPFLR	10 (2.2)	3 (1.6)	1.357	.766
Subsequent instability	19 (4.2)	5 (2.7)	1.684	.301

^aData are presented as n (%). A comparison of the primary study outcomes of interest between patients receiving an autograft versus an allograft for MPFLR. MPFLR, medial patellofemoral ligament reconstruction; NA, not applicable.

Table 5.

Comparison of MPFLR Outcomes Relative to Graft Source ^a

	Gracilis	Semitendinosus	Odds Ratio	P
Patellar fracture	1 (0.2)	0	NA	NA
Revision MPFLR	8 (1.8)	5 (2.6)	1.443	.548
Subsequent instability	16 (3.6)	8 (4.1)	1.239	.628

^aData are presented as n (%). A comparison of the primary study outcomes of interest between patients receiving gracilis versus semitendinosus grafts for MPFLR. MPFLR, medial patellofemoral ligament reconstruction; NA, not applicable.

Table 6.

Comparison of Surgical Outcomes Between the Performance of a Concomitant Tibial Tubercle Osteotomy Versus Isolated MPFLR ^a

	No Concomitant Osteotomy	Concomitant Osteotomy	Odds Ratio	P
Patellar fracture	0	1 (0.3)	NA	NA
Revision MPFLR	11 (3.3)	2 (0.7)	5.000	.021
Recurrent instability	19 (5.6)	5 (1.7)	3.484	.009

^aData are presented as n (%). A comparison of the primary study outcomes of interest between patients undergoing a concomitant tibial tubercle osteotomy versus those who underwent an isolated MPFLR, medial patellofemoral ligament reconstruction; NA, not applicable.

Table 7.

Subgroup Analysis of Isolated MPFLR Without Concomitant Procedures ^a

	Oblique Patellar Tunnels, n = 132	Suture Anchors, n = 205	Odds Ratio	P
Patellar fracture	0	0	NA	NA
Revision MPFLR	4	7	0.884	.846
Recurrent instability	5	14	0.537	.244

^aA subgroup analysis of patients undergoing isolated MPFLR without any concomitant procedures. Comparison of potential complications after patellar-sided graft fixation during MPFLR with the use of 2 small (3.2 mm), oblique patellar tunnels or suture anchors. MPFLR, medial patellofemoral ligament reconstruction; NA, not applicable.

Table 8.

Subgroup Analysis of Patients Undergoing Isolated MPFLR Without Concomitant Procedures ^a

	Recurrent Instability, P	Revision Surgery, P
Patellar tunnels vs suture anchors	.244	.846
Gracilis vs semi-T graft	.416	.249
Autograft vs allograft	.650	.709

^aFirth logistics regression analysis evaluating the incidence of subsequent patellar instability and revision MPFLR relative to the modality of patellar-sided graft fixation, the performance of concomitant osteotomies, and the graft type utilized. MPFLR, medial patellofemoral ligament reconstruction; semi-T, semitendinosus.

Discussion

The most important finding from this large, retrospective cohort study was that the use of 2 small (3.2 mm), short, oblique patellar tunnels with looped grafts is not associated with an increased risk of patellar fracture compared with the use of 2 suture anchors for patellar-sided graft fixation during MPFLR. The use of small (3.2 mm), short, oblique patellar tunnels had comparable overall complications and risk of recurrent lateral patellar instability compared with the use of suture anchors for patellar-sided graft fixation. Collectively, these findings suggest that the use of 2 small (3.2 mm), short, oblique patellar tunnels with looped grafts is a safe and efficacious means of achieving patellar-sided graft fixation during MPFLR while conferring material cost savings compared with the use of 2 suture anchors.

Patellar fractures are a rare but significant complication that could occur after MPFLR. While patellar fractures have been described to occur with the use of 2 suture anchors for graft fixation, it is much more commonly observed in the literature when two 4.5-mm fully transpatellar tunnels with looped grafts are utilized.^2,3,16,19 In a large retrospective series evaluating complications after MPFLR, Schiphouwer et al ¹⁹ reported that 7 of 179 (3.6%) patients sustained a patellar fracture when two 4.5-mm fully transpatellar tunnels were utilized. ¹⁹ Parikh et al ¹⁹ similarly found a 3.35% incidence of patellar fracture in 6 of 179 patients when this technique was utilized. ¹⁶ This increased risk of fracture is believed to be due to a stress riser effect of the larger bone tunnels, particularly when the integrity of the anterior cortex is disrupted.^3,16,19 Because of this risk of fracture, some surgeons favor the use of 2 suture anchors for patellar-sided graft fixation. The findings of the present study are therefore significant, as they demonstrate in the largest clinical series to date that the risk of patellar fracture with the use of smaller (3.2 mm), short, oblique patellar tunnels does not appear to be increased compared with suture anchors, nor does it bear the similar risk as with the historical use of 4.5 mm, fully transpatellar tunnels.

The use of 2 small (3.2 mm), short, oblique patellar tunnels also demonstrated no significant difference from patellar suture anchors when comparing the risk of recurrent lateral patellar instability after surgery. In this study, the use of 2 small (3.2 mm), short, oblique patellar tunnels led to lower rates of recurrent patellar instability and revision MPFLR; however, this difference did not reach statistical significance. Previous literature has likewise failed to demonstrate a clear benefit in the risk of recurrent patellar instability between methods of patellar-sided graft fixation. In 2 systematic reviews comparing the use of patellar tunnels to suture anchors for patellar-sided graft fixation, Migliorini et al ¹⁰ and Kang et al ⁹ found no differences in rates of recurrent patellar instability. While these reviews are limited in heterogeneity in study population and technique, both studies concluded a lack of significant difference in outcomes between techniques.

Of the 24 knees that had recurrent patellar instability after MPFLR, 10 of the knees had instability after a traumatic injury—including ground-level fall (n = 6) and blunt trauma (n = 4). Nine knees had an episode of instability after a noncontact sports or dynamic activity—including a noncontact twisting mechanism (n = 6), running downstairs (n = 1), hyperextension of the knee from housework (n = 1), and running during sports (n = 1). One knee was determined to have continued maltracking due to chronic quadricep laxity and weakness. Four knees had recurrent instability due to uncorrected trochlear dysplasia, malalignment, patellar alta, or a combination of these mechanical factors.

While the present study did not demonstrate a significant difference in rates of recurrent patellar instability between groups, there are theoretical advantages and disadvantages conferred with each technique. For example, proponents of utilizing bone tunnels have theorized that anchor pull-out could lead to more incidences of mechanical failure in the suture anchor cohort. ¹¹ The increased tendon-to-bone contact area could also improve the healing rates and integrity of the reconstruction. ¹¹ Importantly when considering rates of recurrent instability in the current series, another potential contributing factor is that concomitant tibial tubercle osteotomies were performed more frequently in the small (3.2 mm), short, oblique patellar tunnel cohort in this study than in the suture anchor cohort. Although controversial, some studies have previously suggested isolated MPFLR as an effective treatment modality, even in the presence of other anatomical risk factors—including tibial tubercle-to-trochlear groove distance, Caton-Deschamps index, and trochlear dysplasia. ⁵ However, the subgroup analysis of isolated MPFLR between groups showed no difference in the rates of revision MPFLR surgery between groups.

Limitations

This study is not without limitations. First, confounding variables, specifically the inclusion of a concomitant tibial tubercle osteotomy, could influence the study findings because of the retrospective and nonrandomized nature of the study. However, the effects of the variables of interest in isolation were investigated and validated using a Firth logistic regression analysis. Given the nature of the retrospective chart review, it also remains possible that patients would have been evaluated at a different facility for 1 of the particular complications or outcomes of interest and this would not have been identified. However, we believe it is unlikely that confounding variables or isolated lack of capture of outcomes would preferentially influence either study cohort to a differing degree. Last, approximately 50% of the patients in this cohort were included in the study of Deasey et al ¹ but at shorter time points and with less comprehensive analysis.

Conclusion

The use of 2 small (3.2 mm), short, oblique patellar tunnels with looped grafts and dual patellar suture anchors have similar safety and efficacy for achieving patellar-sided graft fixation during MPFLR.