Pain and Hardware Removal After Tibial Tubercle Osteotomy: Incidence,Associated Factors,and Outcomes

Abstract

Background:

While patients may report painful or prominent hardware after tibial tubercle osteotomy (TTO), little is known about the frequency, associated factors, and outcomes after removal of symptomatic hardware.

Purpose/Hypothesis:

The purpose was to investigate the incidence of hardware removal after TTO due to pain or complications, factors associated with hardware removal, and postoperative outcomes after removal. It was hypothesized that clinical outcomes would be similar between patients who underwent TTO without hardware removal and those who underwent TTOs and subsequently hardware removal.

Study Design:

Case-control study; Level of evidence, 3.

Methods:

A retrospective analysis was performed on patients who underwent TTO at a single institution from 2000 to 2023. Age at the time of the index surgery, sex, race, body mass index, number and size of screws, tibial tubercle–trochlear groove distance, and reason for hardware removal were collected retrospectively. Knee radiographs were reviewed for measurement of soft tissue–hardware distance. Patients were contacted via email to capture final patient-reported outcome measures (PROMs). A univariate logistic regression model was used to determine factors associated with hardware removal.

Results:

A total of 152 patients representing 171 knees were included. Of the overall cohort, 38 knees (22.2%) in 32 patients underwent TTO with subsequent hardware removal. The most common reason for hardware removal was anterior knee pain (79%). Compared with patients aged 11 to 20 years, patients aged 21 to 30 years demonstrated higher odds of undergoing hardware removal (OR, 3.67; 95% CI, 1.51-9.44; P = .009). Compared with a soft tissue–hardware distance of 0 to 4.9 mm, a distance of 10.0 to 14.9 mm demonstrated lower odds of hardware removal (OR, 0.24; 95% CI, 0.07-0.84; P = .027). Visual analog scale scores (0-10 scale) improved by a mean of 3.6 points after hardware removal (P = .003). In patients undergoing hardware removal for pain, no difference in final PROMs was found compared with patients who underwent TTO without hardware removal.

Conclusion:

Hardware removal in patients undergoing TTO was mainly attributed to hardware-related pain/irritation. On average, pain scores improved after hardware removal. There was no difference in final PROMs between patients who had their hardware removed due to pain (eg, without any clinically relevant concomitant pathology) and patients who did not require hardware removal. Size and number of screws were not associated with a subsequent hardware removal procedure. The thickness of the soft tissue envelope overlying implanted hardware was inconsistently associated with lower odds of hardware removal.

Keywords

tibial tubercle osteotomy hardware removal risk factors

Malalignment of the extensor mechanism in the context of patellofemoral instability is often measured by an increased tibial tubercle–trochlear groove (TT-TG) distance and is characterized by a relatively lateral location of the tibial tubercle in relation to the trochlear groove. To correct lateralization of the tubercle, a tibial tubercle osteotomy (TTO) can be performed.⁷ The tibial tubercle fragment/shingle created by the osteotomy is often secured through the use of 2 or 3 metallic screws. This is often done in combination with a medial patellofemoral ligament reconstruction if the TT-TG distance is >20 mm. A TTO can also be performed to distalize the tubercle in the setting of patella alta or to anteriorize the tubercle to minimize contact forces across the patellofemoral joint.⁷ Thus, a TTO can be conducted to treat a broad range of patellofemoral joint disorders beyond patellofemoral instability, including correcting patella alta/baja, offloading patellar and trochlear focal chondral lesions, and addressing patellofemoral arthritis.¹⁰

While a TTO is a well-established treatment for a variety of patellofemoral conditions, 4% to 8% of cases can experience complications, such as painful hardware and delayed union that may ultimately result in hardware removal.^8,9 Historically, all-cause rates of hardware removal have been reported as high as 21% to 36%, with patient age, screw size, and prominence being associated with increased removal.^8,9,11 However, factors associated with pain after TTO have not been extensively evaluated. Some evidence has suggested that 3.5-mm screws produce less irritation of the anterior soft tissues due to a smaller and lower-profile screw head and are less likely to need removal than 4.5-mm screws.⁶

While hardware-related pain can occur after TTO, little is known about the frequency, associated factors, and outcomes after TTO hardware removal. The purpose of this study was to (1) investigate the incidence of hardware removal after TTO due to pain or complications, (2) identify factors associated with hardware removal, and (3) determine postoperative outcomes after hardware removal in comparison with patients without removal.

Methods

Upon institutional review board (IRB) approval (Mayo IRB No. 15-000601), a retrospective chart review was conducted on patients at a single institution. Inclusion criteria consisted of all patients regardless of age who underwent a TTO with or without hardware removal at a single institution from the years 2000 to 2023. Indications for surgery included either patellofemoral instability, patellofemoral pain syndrome, or patellofemoral chondral damage. It was not our institutional standard for all patients to undergo hardware removal after TTO. Hardware removal status was determined by examining the patient’s full medical record from index surgery to latest orthopaedic note. They were then categorized into either TTO with hardware removal or TTO without hardware removal. Exclusion criteria consisted of patients who underwent conversion to subsequent total knee arthroplasty of any type, which resulted in the exclusion of 42 patients. All relevant information was obtainable through patient medical records through their last known orthopaedic follow-up. Patient baseline characteristics of age at the time of the index surgery, sex, race, body mass index (BMI), smoking status, number of screws, size of screws, and TT-TG distance were collected retrospectively through the institutional electronic medical record. An institution-wide standardized rehabilitation protocol was utilized for patients who underwent TTO. Time from index surgery to hardware removal and postoperative complications were identified. The incidence and reason for hardware removal was determined utilizing electronic medical records and subsequently confirmed via a follow-up survey for those who responded.

Plain radiographs of the knee in the lateral view were reviewed to confirm the number of screws and measure the soft tissue–hardware distance (Figure 1). The contrast of the plain radiograph was adjusted accordingly to ensure accurate capture of the soft tissue envelope. Secondarily, review of knee radiographs allowed for validation of symptomatic painful hardware as the primary reason for hardware removal rather than loose screws, screw pullout, infection, or fracture. Measurements were taken of the shortest distance from the most superficial aspect of the screw to the skin. In patients with >1 screw for fixation of their TTO, the soft tissue distances were averaged, and statistical analysis was conducted based on the subsequent calculation.

Figure 1.

Plain radiograph in the lateral view of the right knee in flexion showing measurement of the soft tissue–hardware distance.

Interobserver reliability was conducted with 20 random knee radiographs between 2 authors (X.P., F.M.), one of whom is a medical student and the second a fellowship-trained surgeon. Intrarater reliability was conducted between the index measurement and the second measurement (8 months later) of the 20 random knee radiographs. Reliability was calculated based on the correlation coefficients between the 2 measurements.

Surgical Rehabilitation

All patients followed standard institutional protocol after TTO, which was divided into 4 phases. Phase 1 was between 0 and 6 weeks, and restrictions included touch weightbearing with the knee locked in extension while bearing weight. Range of motion was restricted to 90°. Phase 2 was between 6 and 12 weeks, and restrictions were weightbearing as tolerated with a gradual range of motion progression. Phase 3 was between 12 and 20 weeks, and restrictions included slow progression to impact and plyometric activities with special attention to avoiding excessive loading of the anterior knee. Finally, phase 4 was ≥20 weeks with no restrictions and clearance for sports when fit.

Patient-Reported Outcomes

After determining which patients had undergone hardware removal and which had not, patient medical records were queried for baseline preoperative visual analog scale (VAS) scores (0-10 scale) before their index surgeries. Follow-up clinic notes were reviewed for postoperative VAS scores within 1 year of index surgery in patients who did not undergo hardware removal. In patients who underwent hardware removal, VAS scores within 1 year of their hardware removal surgery were recorded. VAS scores from patient charts did not discriminate between rest and use. However, VAS scores from Research Electronic Data Capture (REDCap) (Version 13.1.37; Vanderbilt University) surveys did discriminate between pain at rest and with use. Patients were then subsequently contacted via email through our institutional REDCap to capture patient-reported outcome measures. These consisted of a subjective 5-point scale of perceived knee function (a higher number denotes higher perceived knee function), VAS pain score with use, VAS pain score at rest, and a 5-point Likert scale determining patient satisfaction with their procedure(s) and willingness to undergo the same series of procedure(s) given their current state. A higher number on the Likert scale represents higher patient satisfaction and willingness to undergo the same series of procedures. Follow-up time was determined as the time from the index surgery (in the retained hardware group) or from hardware removal surgery (in the hardware removal group) to survey completion date.

Statistical Analysis

Descriptive statistics are presented as mean ± standard deviation, percentages, and frequency distribution. Pearson chi-square or Fisher exact tests were used for categorical data. Wilcoxon rank-sum tests were conducted for age, BMI, TT-TG distance, VAS scores, and responses on the Likert scale. The Wilcoxon rank-sum test was chosen as it was determined through the Shapiro-Wilk test that all continuous variables in this study were not normally distributed. Variables on the Likert scale were considered ordinal. A univariate logistic regression model was used to determine factors associated with hardware removal. These data are reported using odds ratios. Continuous variables were stratified to create categorical variables, as it was uncertain whether the effect of age, BMI, and soft tissue distances on hardware removal was a linear relationship. Analyzing these as categorical variables allowed for improved interpretation of continuous variables at opposite ends of the spectrum. Missing data and outliers were omitted in the regression model to ensure output fidelity. A priori power analysis with respect to a 2-tailed Wilcoxon rank-sum test was conducted with an alpha of .05 and power of 0.8, and an effect size of 0.79 for the VAS score based on previously published values for patellofemoral pain syndrome.³ A sample size calculation demonstrated that 54 patients would be sufficient to detect a difference in VAS scores for our population. The reliability and reproducibility of the soft tissue–hardware measurements were excellent. Intrarater reliability in a subset of 20 random patients demonstrated a mean correlation coefficient of 0.943, and interrater reliability in a separate subset of 20 random patients demonstrated a coefficient of 0.86. P values <.05 were considered statistically significant. All statistical analysis was performed in the R statistical environment and G*Power (Version 3.1.9.6).⁴

Results

A total of 152 patients, representing 171 knees undergoing TTO, at our institution were identified. Of the overall cohort, 120 patients representing 133 knees underwent TTO without hardware removal, with a mean age of 24.4 ± 9.4 years at the index procedure (Table 1). In total, 93 patients endorsed no tobacco use ever, 11 were former smokers, 10 were current smokers, and 6 had an unknown smoking history. A total of 82 patients had available preoperative TT-TG distance measurements, and the mean preoperative TT-TG distance was 19.1 ± 4.5 mm. After their TTO procedure, the patient VAS pain score decreased, on average, 3.9 points (P < .001).

Table 1

Procedure Characteristics^a

Characteristic	TTO Without Hardware Removal (n = 133)	Hardware Removal (n = 38)	P Value
Sex			.241
Male	51 (38.3)	10 (26.3)
Female	82 (61.7)	28 (73.7)
Race
White	123 (92.5)	34 (89.5)	.516
Black/African American	4 (3.0)	4 (10.5)	.0741
Asian	5 (3.8)	0 (0)	.589
Other	1 (0.8)	0 (0)	>.999
Age at index surgery, y	24.4 ± 9.4	27.8 ± 10.2	.070
Body mass index, kg/m²	27.42 ± 7.3	29.3 ± 6.2	.242
Current smoker	10 (7.5)	1 (2.6)	.479
TT-TG distance, mm	19.1 ± 4.53	20.0 ± 3.0	.410

Data are presented as n (%) or mean ± SD. TT-TG, tibial tubercle–trochlear groove; TTO, tibial tubercle osteotomy.

Of the overall cohort, 32 patients (21.1%) representing 38 knees (22.2%) subsequently underwent hardware removal, with a mean age of 27.8 ± 10.2 years (Table 1). In total, 24 patients endorsed no tobacco use ever, 6 were former smokers, 1 was a current smoker, and the smoking history was unknown for 1. Fourteen of the 38 patients requiring hardware removal had available preoperative TT-TG distance measurements before their index surgery, and the mean preoperative TT-TG distance from before their index surgery was 20.0 ± 3.0 mm. The most common indication for hardware removal was anterior knee pain (79%) (Table 2). Removal of hardware occurred at a mean of 18.7 months (570 days) after TTO. After hardware removal, general VAS pain scores were found to decrease by 3.6 points on average (P = .003). One patient experienced deep vein thrombosis after hardware removal.

Table 2

Reasons for Hardware Removal

Reason for Removal	Hardware Removal, %
Anterior knee pain	79
Clinical nonunion	8
Infection	6
Sustained tibial fracture	3
Screw pullout	3

Patient-Reported Outcome Measures

Preoperative VAS pain scores, postoperative VAS pain scores, and Likert scale satisfaction were not significantly different between the 2 groups at the final follow-up (P > .078). The time from surgery to final follow-up between the TTO without hardware removal group and the TTO with hardware removal group was not statistically significant (1977 days vs 1708 days; P = .552). At the most recent follow-up, VAS pain score with use was greater in the hardware removal group than in the nonremoval group (4.0 vs 2.6; P = .035). A subgroup analysis was conducted in which patients who received isolated hardware removal for pain were solely included. This subgroup consisted only of patients with hardware removal without concomitant complications of infection, fracture, screw pullout, or nonunion and without concomitant revision procedures. This analysis was conducted for improved interpretation of elective hardware removal. No measures were significant in this subgroup analysis in comparison with patients without hardware removal (Table 3).

Table 3

Patient-Reported Outcome Measures^a

Characteristic	TTO Without Hardware Removal	Hardware Removal	P Value^b	Hardware Removal Subgroup^c	P Value^b
VAS pain score
Preoperative	5.3 ± 2.8 (93)	5.3 ± 2.3 (22)	.976	5.1 ± 2.8 (11)	.830
Postoperative	1.4 ± 2.2 (93)	1.7 ± 2.4 (17)	.559	0.3 ± 0.9 (9)	.179
At rest at most recent follow-up	1.0 ± 1.76 (61)	1.9 ± 2.4 (17)	.078	2.0 ± 2.6 (11)	.117
With use at most recent follow-up	2.6 ± 2.3 (61)	4.0 ± 2.2 (17)	.035	3.7 ± 2.5 (11)	.161
5-point Likert scale score
Knee function compared with before surgery	4.4 ± 1.1 (61)	4.2 ± 1.2 (17)	.500	4.4 ± 1.2 (11)	.957
Surgery satisfaction	4.1 ± 11.1 (61)	3.9 ± 1.2 (17)	.537	4.2 ± 1.2 (11)	.798
“Would you redo the same series of procedures?”	4.1 ± 1.3 (61)	3.6 ± 1.1 (17)	.130	3.5 ± 1.2 (11)	.178

Data are presented as mean ± SD (n). Bold P value indicates statistical significance. TTO, tibial tubercle osteotomy; VAS, visual analog scale.

Comparison conducted with respect to TTO without hardware removal group.

Hardware removal subgroup consisted only of patients with hardware removal without concomitant complications of infection, fracture, and nonunion and without concomitant revision procedures.

Univariate Logistic Regression

In comparison with those aged 11 to 20 years, univariate logistic regression analysis demonstrated significance for an increased odds of hardware removal in patients aged 21 to 30 years (OR, 3.67; 95% CI, 1.51-9.44; P = .009) (Table 4). Age comparisons were not statistically significant for all other age groups (P > .072). In comparison with a BMI of 18.5 to 24.9 kg/m², no significant odds for hardware removal were demonstrated (P > .382). With regard to race, in comparison with White patients, significantly increased odds of hardware removal were demonstrated in Black/African American patients (OR, 6.26; 95% CI, 1.46-31.82; P = .015). The odds were unable to be reported with respect to Asian race as no patients in this category underwent hardware removal. No significant odds were demonstrated with respect to smoking status, number of screws, and screw head size. In comparison with soft tissue–hardware distance of 0 to 4.9 mm, distances of 20 to 24.9 mm (OR, 0.06; 95% CI, 0.00-0.73; P = .043) and 10 to 14.9 mm (OR, 0.12; 95% CI, 0.01-0.85; P = .035) demonstrated decreased odds of hardware removal.

Table 4

Univariate Logistic Regression Results for Odds of Hardware Removal^a

Characteristic	OR (95% CI)	P Value
Age, y
11-20	Reference	Reference
21-30	3.67 (1.51-9.44)	.009
31-40	2.67 (0.90-7.83)	.072
41-50	0.67 (0.03-4.10)	.713
BMI, kg/m²
18.5-24.9	Reference	Reference
25.0-29.9	1.29 (0.42-3.91)	.654
30.0-39.9	1.62 (0.55-4.88)	.382
≥40.0	0.90 (0.04-6.69)	.928
Race
White	Reference	Reference
Black/African American	6.26 (1.46-31.82)	.015
Smoking status
Never	Reference	Reference
Former	1.42 (0.43-4.11)	.537
Active	1.23 (0.17-5.67)	.806
No. of screws
1	Reference	Reference
2	1.77 (0.29-34.01)	.602
No. of screws and sizes
One 4.5-mm screw	Reference	Reference
Two 3.5-mm screws	6.00 (0.60-143.41)	.165
One 3.5-mm screw and one 4.5-mm screw	6.67 (0.90-139.07)	.106
Two 4.5-mm screws	1.21 (0.19-23.47)	.862
Soft tissue–hardware distance, mm
0-4.9	Reference	Reference
5.0-9.9	0.37 (0.12-1.12)	.073
10.0-14.9	0.24 (0.07-0.84)	.027
15.0-19.9	0.49 (0.12-1.89)	.308
≥20.0	0.42 (0.09-1.73)	.243

Bold P values indicate statistical significance. BMI, body mass index.

Discussion

The current study set out to investigate the incidence of hardware removal after TTO. Secondarily, the study attempted to identify factors associated with hardware removal, while also demonstrating postoperative outcomes after hardware removal. Notably, 21.1% of patients who underwent TTO had undergone subsequent hardware removal. These patients cited anterior knee pain (79%) as the most common reason influencing the decision to seek hardware removal. Patients aged 21 to 30 years and Black/African American patients had higher odds of undergoing hardware removal. VAS scores improved by a mean of 3.6 points after hardware removal. At the final follow-up, the hardware removal group reported VAS pain scores at rest that were similar to those of the retained hardware group; however, VAS pain scores with use were significantly higher in the hardware removal group compared with the retained hardware group. A subgroup analysis of the hardware removal group without concomitant infection, fracture, hardware failure, or nonunion demonstrated that this difference in pain with use was not significant compared with patients with retained hardware. Interestingly, the number of screws and screw head size were not significant factors for undergoing hardware removal. These findings shed light on the incidence and factors contributing to hardware removal after TTO in patients experiencing pain or complications. These results also describe associated outcomes, something that is not well described in the current literature.

The incidence for hardware removal after TTO has been documented in previous investigations. Johnson et al⁶ described a cohort undergoing TTO in which 32 of 153 patients (21%) cited painful hardware. Furthermore, 38 patients in their cohort required reoperation; 32 of these reoperations were hardware removal. In total, 21% of their TTO cohort underwent hardware removal. This is comparable to the findings of the current study at a 22.2% incidence of hardware removal. Meanwhile, recent systematic reviews of the established literature have documented hardware removal incidence after TTO to range anywhere from 21% to 36%.^8,9 A closer look at the studies included in these systematic reviews demonstrates significant heterogeneity in indications, surgical technique, rehabilitation, and follow-up.

In direct comparison, patients undergoing hardware removal demonstrated a statistical trend toward older age (at the time of the index TTO procedure) compared with patients who did not have hardware removed, although this was not statistically significant. To further investigate this, the youngest patient group (11-20 years of age) was designated as the reference group for defining age as a factor leading to hardware removal. In this analysis, 21- to 30-year-old patients had 3.67 times higher odds of undergoing hardware removal. While this was statistically significant, the 31- to 40-year age group only trended toward significance, with 2.67 times higher odds of hardware removal. A similar age range (21-30 years) appears to be associated with hardware removal across other studies within the published literature.⁸ Further investigation is required to elucidate whether activity level, work status, perception of pain, or other factors contribute to the variation of likelihood of hardware removal within different age or other demographic groups.

Osteotomy fragment fixation after TTO can be highly variable. Biomechanical studies have demonstrated equal fixation strength using two 4.5-mm screws compared with three 3.5-mm screws.¹² In addition to screw size, screw trajectory, and configuration also play a role in fixation strength.¹ As a result, choosing the optimal hardware size and orientation can be up to surgeon discretion but might have implications on symptoms due to the hardware construct. The current study highlights that most patients undergoing hardware removal have a high rate of anterior knee pain (79%). Reducing pain related to hardware might be an area of opportunity to reduce the reoperation rate and has been an area of ongoing research interest. A previous investigation by Johnson and colleagues⁶ demonstrated that smaller screws are 10 times less likely to be symptomatic (eg, three 3.5 mm–diameter screws vs two 4.5 mm–diameter screws). While this might be intuitive, the current study did not find screw size or number of screws to be a statistically significant factor influencing the odds of hardware removal. This finding might indicate that hardware size and configuration may be less important among the several factors contributing to a decision to undergo hardware removal, such as age, race, and distance from hardware to skin.

The current study also found that hardware prominence (as measured by soft tissue thickness superficial to the screws) was inconsistently a factor that led to an effect in the odds of hardware removal. In general, there was a trend for lower odds of undergoing hardware removal with a thicker soft tissue envelope overlying hardware. Yet only the group with a soft tissue–hardware distance ranging from 10 to 14.9 mm displayed a statistically significantly lower odds of hardware removal compared with the comparison group (0- to 4.9-mm soft tissue–hardware distance). The lack of consistency in the presented data might indicate less clinical importance on the prominence of hardware as it may not always correlate with symptoms. Greenberg and colleagues investigated hardware prominence as it relates to patellar fracture fixation constructs and found no correlation between prominence measures and the improvement of knee function or pain after surgery.⁵ Perhaps hardware prominence plays a larger role in areas of the body with thinner dermal/subcutaneous layers and more sensitive cutaneous innervation (eg, hand, foot, clavicle, etc).

Patients in this investigation reported higher pain scores with use of their knee after hardware removal than patients who had asymptomatic/retained hardware. This difference was not noted with pain scores at rest. Because of the nature of data collection, the patient activity level after surgery was not recorded and thus might play a confounding role in higher pain scores with use in the hardware removal group if there was an underlying difference in activity. Similarly, anterior knee pain after TTO could be related to the initial underlying knee pathology (eg, concomitant presence of chondromalacia) or complications during postoperative recovery rather than painful hardware. This was demonstrated in our subgroup analysis of hardware removal in patients without concomitant infection, fracture, hardware failure, or nonunion, in which no discernable difference was detected in any patient-reported outcomes compared with patients who did not require hardware removal. Additionally, persistent postoperative pain with use despite hardware removal may indicate there are also dynamic factors at play (eg, overconstraint of the knee, muscle imbalance, etc), which patients should be counseled about before hardware removal.

Patients did not demonstrate a difference in satisfaction with hardware removal compared with hardware retention in the current study. Previous literature regarding hardware removal in other regions of the body has shown comparable results. Wyles and colleagues¹³ demonstrated a trend toward inferior patient-reported outcomes in patients who underwent hardware removal after periacetabular osteotomy. Therefore, hardware removal is not always a solution for pain at the hardware site. In the instance of TTOs, perhaps other factors are at play, such as radiographically undetected incomplete union or extensor mechanism tendinopathy due to an altered vector of force after realignment. This is merely theoretical, however, as this study was not designed to identify this.

This study is not without limitations. Although a power analysis demonstrated an adequate sample size with respect to the VAS pain score, it may have been underpowered to detect differences between the 2 groups with respect to screw size, number of screws, and other secondary aims. Given that multiple surgeons contributed to this operative cohort over a 23-year time period, there exists the potential for varying surgical techniques, and hence varying outcomes after both TTO and hardware removal surgery. Additionally, soft tissue–hardware distance was collected by only 1 unblinded reviewer (X.P.), which introduces inherent bias. However, these measurements were not exceedingly complex, and measurements from 1 reviewer allowed for consistency in measurement technique with excellent intrarater reliability. The radiographs examined for these measurements were from clinical visits rather than for research purposes. Thus, there may be some slight deviations in the positioning of the lower extremity that affect image quality and cause deviations from a true lateral projection. Similarly, TT-TG distance can be measured by computed tomography or magnetic resonance imaging, and both modalities were documented by surgeons in the electronic medical record for clinical decision-making. Some literature has demonstrated a difference in the obtained values between the 2 imaging modalities, which might affect their interchangeability.² Additionally, time from dislocation to surgery was not evaluated because it was not consistently documented in the clinical notes (eg, exact date), which is another limitation attributed to the retrospective nature of the study. Likewise, a detailed social history of each patient was unavailable, and variables such as patient activity level and occupation were unable to be evaluated. These factors may play a significant role in contributing to pain resulting in hardware removal.

Conclusion

Hardware removal in patients undergoing TTO was mainly attributed to hardware-related pain/irritation. On average, pain scores improved after hardware removal. There was no difference in final patient-reported outcome measures between patients who had their hardware removed due to pain (eg, without any clinically relevant concomitant pathology) and patients who did not require hardware removal. Size and number of screws were not associated with a subsequent hardware removal procedure. The thickness of the soft tissue envelope overlying implanted hardware was inconsistently associated with lower odds of hardware removal.

Footnotes

Final revision submitted September 29, 2024; accepted November 12, 2024.

One or more of the authors has declared the following potential conflict of interest or source of funding: M.E.D. has received hospitality payments from Stryker. F.J. has received hospitality payments from Exactech and Smith & Nephew, and education payments from Pinnacle. A.J.T. has received hospitality payments from Stryker, Zimmer Biomet Holdings, Arthrex, Medical Device Business Services, DePuy Synthes Sales, and Wright Medical Technology; nonconsulting fees from Smith & Nephew, Synthes GmbH, and Arthrex; consulting fees from Smith & Nephew and Endo Pharmaceuticals; education payments from Foundation Medical, Gemini Medical, Arthrex, Pinnacle, Medwest Associates, and Smith & Nephew; and a grant from Arthrex. A.J.K. has received research support from Aesculap/B.Braun, Arthritis Foundation, Ceterix, and Histogenics; consulting fees, IP royalties, and research support from Arthrex; consulting fees from JRF Ortho, Vericel, and Responsive Arthroscopy; royalties from Responsive Arthroscopy; a grant from DJO; honoraria from Joint Restoration Foundation and Musculoskeletal Transplant; and hospitality payments from Gemini Mountain Medical LLC and Smith & Nephew. M.H. has received support for education from Arthrex, Foundation Medical, Medwest Associates, and Smith & Nephew; honoraria from Encore Medical; hospitality payments from Orthalign, Stryker, and Medical Device Business Services; consulting fees from DJO-Enovis, Moximed, and Vericel; education payments from Arthrex, Foundation Medical, Medwest Associates, and Smith & Nephew; and a grant from Medical Device Business Services. K.R.O. has received consulting fees from Arthrex, Smith & Nephew, and Endo Pharmaceuticals; support for education from Gemini Medical, Smith & Nephew, Arthrex, Medwest Associates, and Pinnnacle; hospitality payments from Stryker, Wright Medical Technology, Medical Device Business Services, and Zimmer Biomet; and a grant from Arthrex. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

Ethical approval for this study was obtained from Mayo Clinic (PR15-000601-10).

ORCID iDs

Adam J. Tagliero

Aaron J. Krych

Mario Hevesi

Kelechi R. Okoroha

References

Aykanat

Kose

Guneri

, et al. Comparison of four different screw configurations for the fixation of Fulkerson osteotomy: a finite element analysis. J Orthop Traumatol. 2023;24(1):30.

Camp

Stuart

Krych

, et al. CT and MRI measurements of tibial tubercle-trochlear groove distances are not equivalent in patients with patellar instability. Am J Sports Med. 2013;41(8):1835-1840.

Crossley

Bennell

Cowan

Green

. Analysis of outcome measures for persons with patellofemoral pain: which are reliable and valid? Arch Phys Med Rehabil. 2004;85(5):815-822.

Faul

Erdfelder

Buchner

Lang

. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods. 2009;41(4):1149-1160.

Greenberg

Kadar

Drexler

, et al. Functional outcomes after removal of hardware in patellar fracture: are we helping our patients? Arch Orthop Trauma Surg. 2018;138(3):325-330.

Johnson

Wolfe

Mintz

, et al. Complications after tibial tuberosity osteotomy: association with screw size and concomitant distalization. Orthop J Sports Med. 2018;6(10):2325967118803614.

Koh

Stewart

. Patellar instability. Orthop Clin North Am. 2015;46(1):147-157.

Payne

Rimmke

Schmitt

Flanigan

Magnussen

. The incidence of complications of tibial tubercle osteotomy: a systematic review. Arthroscopy. 2015;31(9):1819-1825.

Saltzman

Rao

Erickson

Cvetanovich

Levy

Cole

. A systematic review of 21 tibial tubercle osteotomy studies and more than 1000 knees: indications, clinical outcomes, complications, and reoperations. Am J Orthop (Belle Mead NJ). 2017;46(6):E396-E407.

10.

Sherman

Erickson

Cvetanovich

, et al. Tibial tuberosity osteotomy: indications, techniques, and outcomes. Am J Sports Med. 2013;42(8):2006-2017.

11.

Unal

Demirayak

Ertan

Kilicaslan

Kose

. Bioabsorbable magnesium screw fixation for tibial tubercle osteotomy; a preliminary study. Acta Biomed. 2022;92(6):e2021263.

12.

Warner

Kamath

Spang

Weinhold

Creighton

. Comparison of fixation methods after anteromedialization osteotomy of the tibial tubercle for patellar instability. Arthroscopy. 2013;29(10):1628-1634.

13.

Wyles

Statz

Hevesi

Chalmers

Sierra

Trousdale

. Incidence and risk factors for hardware removal following periacetabular osteotomy and its association with clinical outcomes. Hip Int. 2021;31(3):410-416.