Tension Band Fixation for Patellar Fractures Using K-Wires Versus Cannulated Screws

Abstract

Purpose:

Tension band fixation (TBF) using screws or wires is the widely accepted standard of care for patellar fractures. This meta-analysis aims to quantitatively compare postoperative outcomes between Kirschner wire tension band fixation (KWTB) and cannulated screw tension band fixation (CSTB) to determine the optimal fixation method.

Methods:

A systematic review and meta-analysis were conducted per Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PubMed, MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials (CENTRAL) were searched for comparative studies on KWTB vs. CSTB. Primary outcomes included union rate, time to union, infection, reoperation rate and pain. Secondary outcomes included range of motion, time to weight bearing, implant complications, and functional scores.

Results:

This meta-analysis included 10 studies with 1,272 patients. KWTB was associated with a significantly longer time to union (p = 0.02), greater pain scores (p < 0.0001), and higher infection rates (p = 0.04). However, union and reoperation rates were comparable. For secondary outcomes, CSTB had significantly improved time to full weight bearing (p < 0.00001). In addition, KWTB had higher complication rates, including implant migration/loosening (p < 0.00001), implant prominence, skin irritation (p = 0.003), and implant removal (p < 0.00001). The Lysholm score (p < 0.00001) and range of motion (p = 0.03) were significantly higher in the CSTB group.

Conclusion:

While union rates are comparable, CSTB for TBF in patellar fractures is associated with improved functional outcomes and a reduced complication rate compared with KWTB.

Level of Evidence:

Level III, systematic review and meta-analysis. See Instructions for Authors for a complete description of levels of evidence.

Introduction

Patellar fractures result from direct impact or excessive quadriceps contraction with a flexed knee, often leading to extensor mechanism disruption, accounting for approximately 1% of all skeletal injuries, predominantly affecting individuals aged 20 to 50 years^1-4. While most patellar fractures heal without complications, factors such as comminution, open fractures, and smoking history are associated with poorer prognoses^5,6.

Surgical intervention is recommended for displaced fractures with an articular step-off >2 to 3 mm or a fracture gap >1 to 4 mm with extensor mechanism disruption^7,8. The primary goal of surgery is to achieve stable fixation, restore the patellofemoral articular surface, and preserve knee extensor function⁷.

Tension band wiring with Kirschner wires (KWTB) and cannulated screw tension band fixation (CSTB) are the most widely used fixation techniques¹. The tension band construct principle converts anterior tensile forces generated by the quadriceps into compression forces at the articular surface, enhancing fracture stability^1,4,5,8-10.

Although KWTB remains the mainstay of treatment, it is frequently associated with various postoperative complications contributing to high reoperation rates^1-4,7,8,10. While CSTB offers biomechanical advantages, it has been linked to implant fixation failures requiring removal^1,3,5.

Previous studies evaluating tension band constructs for patellar fractures, comparing KWTB with alternative techniques suggested that alternative fixation techniques may be as effective as KWTB in managing patellar fractures¹¹. Similarly, a recent meta-analysis comparing the use of K-wires and screws in pediatric lateral condyle fractures suggested that both techniques were effective treatment options¹². Since previous meta-analyses lacked focused analysis on CSTB, this study aims to assess and compare the safety and clinical efficacy of KWTB and CSTB to determine the optimal fixation strategy.

Methods

This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines¹³.

Search Strategy and Study Selection

Controlled trials and observational studies comparing outcomes of all patellar fracture types treated with KWTB or CSTB were included. Studies were excluded if they were not in English, lacked a direct comparison between the 2 treatment modalities, or omitted key data. In addition, reference lists of all included studies were examined to identify further relevant articles. Duplicate records were removed, and 2 independent authors screened each eligible study for final inclusion.

Two independent authors conducted a comprehensive database search of PubMed, MEDLINE, EMBASE and CENTRAL, with the final search completed on February 5, 2025. To capture unpublished or ongoing studies, searches were extended to the World Health Organization International Clinical Trials Registry and the ISRCTN Register. No language restrictions were applied. The search strategy used can be found in the Appendix (Appendix Fig. 1).

Titles and abstracts of all studies were independently screened by 2 authors. Full-text articles of relevant studies were obtained and evaluated against eligibility criteria. Discrepancies in study selection were resolved by consensus among authors.

Outcomes

Primary outcomes included reoperation rate, union rate, infection rate, need for implant removal, time to union (weeks), and postoperative pain score (measured on a 10-point scale).

Secondary outcomes included range of motion (ROM), implant migration, implant-related complications, time to full weight bearing, and functional scores.

Data Extraction and Analysis

Data extraction followed the Cochrane data collection form for intervention reviews. A structured spreadsheet was developed, pilot-tested, and refined using randomly selected articles.

Data were analyzed using Review Manager (RevMan) version 5.4. Two independent authors entered data into the software. A random-effects model was applied to calculate 95% confidence intervals (CIs). Odds ratios (ORs) were computed for dichotomous outcomes, and mean differences (MDs) for continuous outcomes.

Heterogeneity was evaluated using the Cochran Q test (χ²) and I² statistics, interpreted as follows:

• 0%–25%: Low heterogeneity

• 25%–75%: Moderate heterogeneity

• 75%–100%: High heterogeneity

Risk of bias in randomized studies was assessed using the Cochrane risk-of-bias (ROB2) tool¹⁴. For nonrandomized studies, the risk of bias in nonrandomized studies of interventions (ROBINS-I) tool was used¹⁵. Two independent authors conducted the quality assessments, resolving disagreements by consensus.

Each study was sequentially excluded to assess its influence on result significance. This analysis confirmed that no single study or any study with a high risk of bias significantly altered the overall outcomes, further validated by funnel plot analysis.

Availability of Data and Materials

The data generated and analyzed in this study can be obtained from the corresponding author upon reasonable request.

Results

Literature Search Results

The search identified 10 studies which met the inclusion criteria and were included in the final analysis (Fig. 1). These studies collectively evaluated 1,272 patients (786 in KWTB, 504 in CSTB) (Appendix Table 1). A full summary of patient outcomes can be found in Appendix Table 2.

Fig. 1.

PRISMA Flow Diagram. PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Baseline Characteristics

A total of 1,272 patients were included in this systematic review. The mean age of participants ranged from 35.96 ± 10.75 to 63.65 ± 12.91 years. BMI values ranged from 22.07 ± 1.49 to 26.17 ± 4.92 kg/m². Gender distribution was provided in most studies, with varying male-to-female ratios.

Primary Outcomes

Infection rates were reported in 10 studies, including 1,272 patients (Appendix Fig. 2). The incidence of infection was significantly higher in patients treated with KWTB compared with CSTB (OR = 2.31; 95% CI = 1.06-5.05; p = 0.04). Low heterogeneity was observed (I² = 0%, p = 0.87).

Reoperation rates were reported in 9 studies, including 1,212 patients (Appendix Fig. 2). No significant difference was observed between the groups (OR = 0.98; 95% CI = 0.29-3.78; p = 0.95). Moderate heterogeneity was detected (I² = 73%, p = 0.01).

Need for implant removal was reported in 10 studies with a total of 1,272 patients (Appendix Fig. 2). The KWTB group demonstrated a significantly higher rate of implant removal (OR = 2.28; 95% CI = 1.64-3.35; p < 0.00001). Low heterogeneity was observed (I² = 0%, p = 0.15).

Knee pain was reported in 10 studies, totaling 1,272 patients (Appendix Fig. 3). A significantly higher incidence of knee pain was observed in the KWTB group compared with the CSTB group (OR = 6.67; 95% CI = 2.48-17.92; p = 0.0002). Low heterogeneity was detected (I² = 0%, p = 0.25).

VAS pain scores were reported in 3 studies, including 412 patients (Appendix Fig. 3). Pain scores at 3 and 6 months were significantly higher in the KWTB group (MD = 1.16; 95% CI = 0.59-1.72; p < 0.0001). High heterogeneity was observed (I² = 96%, p < 0.00001)

Union rates were reported in all studies, encompassing 1,272 patients (Fig. 2). A comparison between KWTB and CSTB demonstrated a higher likelihood of union with screw fixation (OR = 0.66; 95% CI = 0.20-2.11; p = 0.48); however, this difference was not statistically significant. Moderate heterogeneity was observed (I² = 71%, p = 0.48).

Fig. 2.

Comparison of Union Rates Between KWTB and CSTB. CSTB = cannulated screw tension band fixation, and KWTB = Kirschner wire tension band fixation.

Time to union was reported in 7 studies, including 696 patients (Appendix Fig. 4). Patients treated with KWTB required a significantly longer time to achieve union compared with those treated with CSTB (MD = 0.89; 95% CI = 0.12-1.65; p = 0.02). High heterogeneity was observed (I² = 84%, p < 0.00001).

Secondary Outcomes

ROM was reported in 3 studies with 236 patients in total (Appendix Fig. 3). KWTB was associated with a significant reduction in knee flexion CSTB (MD = −11.22; 95% CI = −21.11 to −1.32; p = 0.03). High heterogeneity was present (I² = 96%, p < 0.00001).

The Lysholm score was reported in 2 studies with a total of 206 patients (Appendix Fig. 3). At the 6-month follow-up, the score was significantly lower in patients treated with KWTB (MD = −10.35; 95% CI = −11.75 to −8.94; p < 0.00001). Low heterogeneity was observed (I² = 0%, p = 0.46).

The IOWA knee score was reported in 2 studies with a total of 161 patients (Appendix Fig. 3). Patients in the KWTB group were significantly less likely to achieve a good or excellent IOWA Knee Score at final follow-up compared with those treated with CSTB (OR = 0.06; 95% CI = 0.01-0.44; p = 0.006). Low heterogeneity was observed (I² = 0%, p = 0.93).

Appendix Fig. 2 summarizes implant migration rates across 10 studies including 1,272 patients. A significantly higher occurrence of implant migration and/or loosening was observed in the KWTB group (OR = 10.32; 95% CI = 4.25-25.08; p < 0.00001). Low heterogeneity was observed (I² = 0%, p = 0.49).

Appendix Fig. 2 demonstrates implant prominence and skin irritation in 10 studies with a total of 1,272 patients. Patients treated with KWTB experienced significantly higher rates of implant prominence and skin irritation (OR = 3.88; 95% CI = 1.58-9.55; p = 0.003). Minimal heterogeneity was observed (I² = 0%, p = 0.46).

Operation duration was reported in 4 studies with a total of 362 patients (Appendix Fig. 2). No significant difference in operative duration was observed between both groups (MD = −2.83; 95% CI = −6.43-0.77; p = 0.12). Substantial heterogeneity was observed (I² = 81%, p = 0.0009).

Both Poh et al. and Liu et al. reported longer hospital stays in the screw fixation group, suggesting a potential advantage of KWTB in reducing hospitalization duration.

There was no clear pattern regarding blood loss. Liu et al. reported greater blood loss in the CSTB group, whereas Liu et al. observed higher blood loss in the KWTB group.

Time to full weight bearing was reported in 2 studies totalling 219 patients (Appendix Fig. 4). Patients in the KWTB group required significantly longer times to reach full weight bearing compared with those in the CSTB group (MD = 14.10; 95% CI = 11.68-16.52; p < 0.00001). Low heterogeneity was observed (I² = 0%, p = 0.63).

Quality Assessment Results

The risk of bias in randomized controlled trials (RCTs) was assessed using the ROB 2 tool, while the ROBINS-I tool was used to evaluate nonrandomized studies (Appendix Fig. 5). In ROB 2, both randomized studies demonstrated a low risk of bias, with only minor concerns in certain domains.

Except for Zhu et al., all nonrandomized studies exhibited a low to moderate risk of bias.

Discussion

The most significant overall findings of our meta-analysis suggested that postoperative VAS pain scores and rate of knee pain were significantly higher in the KWTB group. Time to union and time to full weight bearing were also found to be significantly higher in the KWTB group.

Knee range of motion and functional outcomes were significantly better in the CSTB group, although union rates were comparable between the 2 groups.

Our findings are consistent with recent literature evaluating screw-based fixation methods for patellar fractures. Tijare et al. reported reduced hardware-related complications and comparable union rates when cannulated screws were used, reinforcing the clinical relevance of screws as a viable alternative to tension band wiring alone¹⁶. Liu et al. demonstrated that CSTB fixation resulted in faster union times, lower pain scores, and quicker weight-bearing ability². These advantages may be attributed to cannulated screws providing primary compression forces. The threaded end of cannulated screws enhances fracture site compression, promoting faster healing^10,17,18. CSTB enhances compression by applying pressure to tighten the tension band wire rather than exerting force directly on the patella. This mechanism generates a more effective force at the fracture site, ensuring stable fixation of the bone fragments¹⁹. Both fixation techniques yielded similar union rates, reinforcing their overall effectiveness.

Across studies, KWTB fixation consistently demonstrated higher complication rates than CSTB fixation including infection, implant migration/loosening, implant prominence/skin irritation, and implant removal. K-wires extend beyond the fracture site, often causing skin irritation due to contact with surrounding soft tissues. By contrast, screw fixation allows size customization, reducing implant prominence¹⁸. The reduced hardware irritation at the anterior patella and decreased need for secondary implant removal may be a contributing factor to the increased ROM reported in CSTB patients. CSTB provide stable fixation with less soft-tissue irritation, which may facilitate earlier active mobilization protocols and minimize flexion discomfort during rehabilitation. Unlike smooth-surfaced Kirschner wires, cannulated screws provide superior compression forces, reducing postoperative complications^4,19. CSTB fixation may also be more cost-effective due to fewer complications, leading to shorter recovery times and lower healthcare costs²⁰.

In terms of functional outcomes, KWTB fixation resulted in lower Lysholm scores and fewer patients achieving good or excellent Iowa Knee Scores. These findings align with the better recovery rates and lower complication rates observed in the CSTB group. Such findings may reflect enhanced quadriceps function and earlier restoration of knee mechanics, which are critical for everyday tasks such as stair climbing and returning to occupational or athletic activity.

This review was not without its challenges. Selection bias may have influenced outcomes, as cannulated screws are preferred in less comminuted fractures, whereas K-wire constructs are commonly used in complex fractures. Fracture morphology was not consistently stratified across studies, preventing subgroup analyses based on fracture complexity. Measurement of time to union also lacks precision in patellar fractures, as radiographic healing is difficult to standardize. Furthermore, time to weight bearing and initiation of knee motion were largely driven by surgeon preference and fixation confidence rather than uniform protocols, introducing variability in rehabilitation timelines. Finally, quality-of-life and patient-reported outcomes remain underreported in the current literature, with only isolated studies incorporating validated measures of physical and mental function. This represents a meaningful gap, as pain perception, activity level, and patient satisfaction may differ between fixation constructs and are not fully captured by traditional functional scores. These factors contribute to the observed heterogeneity and may limit definitive conclusions; however, the overall direction of effect remained consistent across analyses. Future research should focus on implementing uniform postoperative outcome measures, follow-up timelines, and knee scoring systems to establish evidence-based clinical guidelines.

Conclusion

CSTB fixation is associated with better functional outcomes, faster recovery, and lower complication rates compared with KWTB fixation. Large multicenter trials would aid in the development of standardized clinical protocols to optimize treatment strategies.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Appendix

Supporting material provided by the authors is posted with the online version of this article as a data supplement at jbjs.org (http://links.lww.com/JBJSOA/B137). This content was not copyedited or verified by JBJS.