2.1 Literature search methodology

A comprehensive search of PubMed, Embase, and Cochrane Library databases was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines in August 2024. The following search strategy was utilized: “('patellar tendon repair' OR 'patella tendon repair').”

Studies were included if they evaluated males and females of any age group who underwent acute, primary patellar tendon repair (as defined by surgical repair without graft augmentation within 6 weeks from injury, including for mid-substance tears and avulsion injuries), were prospective or retrospective studies, reported functional outcomes such as ROM on knee flexion and extension, or complications in patients undergoing patellar tendon repair, and were published after 2004. Translational studies or cadaveric studies, revision PT repair patients, PT repairs after 6 weeks, studies that did not report complications, functional outcomes, or had study designs that were systematic reviews, case-series, narrative reviews, conference abstracts, technical notes, letters to editors, or meta-analyses were excluded. Two authors (initials blinded for review) independently screened titles, abstracts, and full article texts using the online software program Covidence (Veritas Health Innovation Ltd; Melbourne, Australia). Any disagreements were resolved with discussion leading to consensus between the two screening authors and a third senior author (initials blinded for peer review).

2.2 Data extraction and quality assessment

Data items extracted from each study included demographic information such as gender, age, and body mass index (BMI) distribution of included patients, mean follow-up time, functional outcomes such as mean post-operative flexion and extension, complications, and failure or re-operation rates. This study ultimately only included non-comparative studies, thus a non-randomized retrospective study quality assessment was performed with the Methodological Index for Non-Randomized Studies (MINORS) criteria.²⁷

2.3 Statistical analysis

Pooling of data was avoided due to high risk of bias inherent with retrospective studies and heterogeneity among included studies. A qualitative data comparison was conducted. For studies only reporting ranges, the standard deviation was approximated as the range divided by four. Eligible studies were entered into Open Meta Analyst (Brown University; Providence, RI) to create single leg forest plots illustrating mean post-operative extension and flexion for each study. These forest plots served as visual aids to summarize these means.

3.1 Search results and study quality/risk of bias

A total of 1737 studies were identified in the initial search, 103 of which were duplicates and were subsequently excluded. The remaining 1634 studies underwent a title and abstract screening, 1616 were found to be irrelevant to the study aims and therefore excluded. The remaining 18 studies were assessed for eligibility with full-text review. After excluding 9 studies for having an incorrect intervention, study design, or patient population, 9 studies were ultimately included for data extraction (Fig. 1). Six studies were excluded for having an incorrect intervention, with one study including patients with patellar fractures, one study comparing acute versus chronic PT ruptures, and four studies utilizing autograft or allograft augmentation. One study was excluded for having incorrect study design as it was a novel technique and case report. One study was excluded for having an incorrect patient population, evaluating only professional athletes. Each of the studies included in this systematic review evaluated outcomes and complications following PT repair.

Close

Fig. 1 PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) study selection flow diagram.

Export to PPT

Table 1 summarizes study quality based on Methodological Index for Non-Randomized Studies (MINORS) criteria for non-randomized studies.²⁷ The ideal MINORS score for non-comparative studies is 16, with scores ≤8 being the accepted cut-off for poor study quality. Each of the included studies had a score ≥9, indicating sufficiently low risk of bias.²⁷ Of note, none of the included studies had an unbiased assessment of the primary endpoint or prospective calculation of study size. Although each included study had low risk of bias, pooling was avoided due to low levels of evidence and high heterogeneity across included studies.

3.2 Study characteristics and demographics

Nine total studies were included in this systematic review, including three retrospective cohort studies,^17,28,30 three retrospective case series,^18,29,33 one retrospective case control study,⁶ and two prospective case series studies.^31,32 A total of 970 patients met inclusion criteria and were included in this systematic review. Aside from Bushnell et al., all studies included patient sex, reporting 897 (93.8 %) male patients and 59 (6.2 %) female patient.^6,17,28–33 Of note, Fredericks et al. only included patients who were active duty military member.⁶ The mean age among all included patients ranged from 32.1 years to 44.6 years of age.^{6,17,18,28–33} Four included studies reported BMI with mean values ranging from 25.7 kg/m² to 29.1 kg/m².^6,28,30,33 O'Dowd et al. separately reported mean BMI for patients who received transosseous tunnels or suture anchors as 31.2 kg/m² and 30.9 kg/mm², respectively.¹⁷ Five included studies reported mean follow-up times.^{6,17,18,32,33} Roudet et al. assessed outcomes in two phases: early complications and results at mean 7.1 months, and long-term functional outcomes and satisfaction at mean 111.5 months.³³ The other four studies had mean follow-up times ranging from 29.0 months to 51.2 months.^6,17,18,32 Beranger et al., Hinz et al., and Massoud et al. reported median follow-up times of 47.7, 17.0, and 45.0 months respectively.^28,30,31 Core et al. included patients that followed up for a minimum of 12.0 months.²⁹

Seven studies included patients who underwent PT repair with transosseous sutures.^{6,17,28,30–33} Hinz et al. and Otsubo et al. specified the utilization of high-tensile strength suture or suture tapes in their transosseous approach in each of their patients.^30,32 Three studies evaluated patients who underwent PT repair with suture anchors.^6,17,18 Three studies evaluated patients who underwent PT repair with end-to-end tendon repair with non-absorbable sutures.^28,29,33 Of note, Core et al. used end-to-end tendon repair technique augmented with synthetic ligaments.²⁹ Roudet et al. also included patients who underwent PT fixation by staples.³³Table 2 summarizes demographic factors and operative techniques used for all patients included in this systematic review.

3.3 Post-operative range of motion

Four studies reported on mean knee extension at follow-up with values ranging from 0° to 6.4°.^18,28,30,33 Of note, Fredericks et al. reported on 5 included patients with a mean extensor lag of 16° (range 5°–30°) but did not report mean post-operative extension across all included patients in their study.⁶ Five studies reported degrees of flexion achieved at follow-up with values ranging from 128.5° to 143.3°.^{18,28,30,32,33}Table 3 summarizes range of motion at follow-up after patellar tendon repair for patients included in this systematic review. Fig. 2 is a visual aid that summarizes post-operative extension across patients included in this systematic review. Fig. 3 is a visual aid that summarizes post-operative flexion across included patients.

Close

Fig. 2 Forest plot illustrating post-operative extension across included studies.

Export to PPT

Close

Fig. 3 Forest plot illustrating post-operative flexion across included studies.

Export to PPT

3.4 Complications

Complications reported by studies included in this systematic review consisted of re-rupture or failure, re-operation, and VTE.^{6,17,18,28,29,31–33} Five of the nine included studies reported no cases of patellar re-rupture or surgical failure.^28–32 There were 3 (21.4 %) cases of surgical failure in the Bushnell et al. study, all attributed to patient non-adherence with prescribed bracing, weight bearing and activity recommendations.¹⁸ There were 10 (2.1 %) cases of re-rupture in the Fredericks et al. study.⁶ There were 24 (6.6 %) cases of re-rupture in the O'Dowd et al. study, in which all patients received transosseous tunnels.¹⁷ There was one (2.7 %) case of re-rupture following a fall in the Roudet et al. study.³³ Ranges of re-rupture or surgical failure rates across studies were 0–21.4 %. The overall re-rupture rate was 3.9 %.

Six of the nine included studies reported zero re-operations among their patients.^28–33 Bushnell et al. reported 3 (21.4 %) cases of revision repair using transosseous suture tunnels for the three patients with surgical failure.¹⁸ Fredericks et al. reported 15 (3.1 %) revision surgeries for the ten patients who experienced re-rupture and for five additional patients with chronic weakness and extensor lag post-operatively.⁶ O'Dowd et al. reported 37 (10.2 %) cases of re-operation, in which 35 of the 37 patients had undergone repair with transosseous tunnels; based off logistic regression, the authors found patients who were treated with the transosseous tunnel approach were 3.24 times more likely to require re-operation compared their counterparts who received suture anchor fixation.¹⁷ For transosseous tunnel patients, re-operation was performed for indications of patellar tendon re-rupture, lysis of adhesions, removal of uncomfortable sutures, and wound debridement.¹⁷ For suture anchor patients, re-operations included one late wound infection and one knee arthroscopy.¹⁷ Ranges of re-operation rate across studies were 0–21.4 %. The overall re-operation rate was 5.7 %.

VTE was not observed in six of the nine included studies.^6,18,29–32 One (5.0 %) patient in the Beranger et al. study suffered a deep vein thrombosis (DVT).²⁸ Nine (2.5 %) patients in the O'Dowd et al. study suffered DVTs, of whom two patients died due to pulmonary embolism (PE).¹⁷ One (2.7 %) patient with no known risk factors in the Roudet et al. study died from a massive PE.³³ Overall, the range of VTE was 0–5.0 % across included studies.

Less common complications observed in patients included across this systematic review included post-operative stiffness, hardware discomfort, and infection. There were zero reported cases of post-operative stiffness in seven of the nine studies included in this systematic review.^{6,17,18,30–33} Beranger et al. reported seven (35.0 %) cases of post-operative knee stiffness, with a minimum flexion of 110°.²⁸ Core et al. reported one (3.3 %) case of post-operative knee stiffness, which they defined as knee flexion less than 90°.²⁹ Overall, the range of stiffness was 0–35 % across included studies. In addition, two studies reported on prominent hardware in patients. There were zero reported cases of hardware discomfort in seven of the nine studies included in this systematic review.^6,18,29–33 Beranger et al. reported one (5.0 %) case of hardware discomfort leading to hardware removal.²⁸ O'Dowd et al. reported nine (2.5 %) cases of hardware discomfort, each of which was treated with re-operation for removal.¹⁷ Hardware discomfort rates ranged from 0 to 5% across included studies. There were zero reported cases of infection in seven of the nine included studies.^6,18,29–33 One (5.0 %) patient in the Beranger et al. study suffered a major deep infection that was treated with muscle flap and skin graft.²⁸ Nine (2.5 %) patients in the O'Dowd et al. study experienced superficial surgical site infections, of whom four required operative debridements while five patients were managed with antibiotics.¹⁷ Overall, infection rates ranged from 0 to 5.0 % across all included studies. Table 4 summarizes the complications from patellar tendon repairs included in this systematic review.

4.1 Limitations

This study is not without limitations. First, the study is limited by the relatively small sample sizes of the included studies, the retrospective nature of each of the included studies, as well as heterogeneity with regards to the surgical technique (transosseous versus suture anchor versus end-to-end tendon repairs). Studies evaluating end-to-end repair were ultimately included in this systematic review because their exclusion would have resulted in a paucity of studies for an effective systematic review. Furthermore, we were unable to perform a subgroup analysis based on the specific type of acute patellar tendon injury (i.e. midsubstance tear, avulsion injury, etc.). Additionally, there was a paucity of patient reported outcome data across all included studies, which limited our ability to make meaningful conclusions regarding patient baseline and subsequent improvement post-operatively, especially from a subjective patient perspective. Finally, we were unable to further stratify our data based on demographic factors such as patient BMI and athletic status, both of which are important surgical considerations and factors that affect post-operative outcomes and potential for patellar tendon re-tear.

4.2 Conclusions

Patellar tendon repair with both transosseous and suture anchor technique restores range of motion in patients post-operatively with a low complication rate. Decisions regarding operative technique should be tailored to individual patient characteristics and surgeon preferences.

Each of the other authors declare that they have no competing interests.