Arthrofibrosis After Adolescent Anterior Cruciate Reconstruction with Quadriceps Tendon with Bone Block Autograft

Abstract

Background:

Anterior cruciate ligament reconstruction (ACLR) utilizing quadriceps tendon autografts with a patellar bone block (QTB) has gained popularity in children and adolescents, with favorable patient-reported outcomes being reported at 2 years postoperatively. The incidence of arthrofibrosis after ACLR in pediatric patients for all graft types is between 2% and 10%. However, there is a paucity of research focused on arthrofibrosis in pediatric patients undergoing ACLR with QTB.

Purpose:

To quantify children after ACLR with QTB and compare with previously published studies evaluating an all-soft tissue quadriceps autograft.

Study Design:

Case series; Level of evidence, 4.

Methods:

This was a retrospective review of children and adolescents aged ≤17 years who underwent primary ACLR with QTB between 2019 and 2023. Demographic characteristics (age, sex, body mass index [BMI]), insurance type, and pre-, intra-, and postoperative records were reviewed. The incidence of arthrofibrosis was assessed at 3 months postoperatively and at the last follow-up. Arthrofibrosis was defined as a 20° flexion deficit and/or a 10° extension deficit at 3 months.

Results:

A total of 80 patients with a mean age of 15.9 years (range, 12.8-17 years) were included. At 3 months, arthrofibrosis occurred in 22 of 80 patients (27.5%). At the last follow-up (mean duration of 13.3 months; range, 6-47 months), 7 of 80 patients (8.8%) required procedural intervention with manipulation under anesthesia (MUA). Those with arthrofibrosis at 3 months postoperatively had significantly reduced preoperative flexion compared with those without arthrofibrosis (120.5° vs 130.7°; P = .02) and at postoperative week 6 (91.7° vs 111.9° flexion; P < .001). The presence of medial meniscal repair (MMR) was significantly higher in the arthrofibrosis group (P = .04). No significant difference in age, sex, or BMI existed (P > .05). However, patients with government insurance were disproportionately more likely to develop arthrofibrosis at 3 months, as opposed to those with private insurance (P = .02).

Conclusion:

More children and adolescents undergoing ACLR with QTB have a higher rate of arthrofibrosis at 3 months. However, return to the operating room for MUA is similar to other graft types. Reduced flexion preoperatively and at 6 weeks postoperatively, as well as concomitant MMR, were significant predictors of arthrofibrosis.

Arthrofibrosis is a known complication of anterior cruciate ligament reconstruction (ACLR) in the pediatric population; however, its definition and rates vary. Some authors advocate for a definition that relies on range of motion (ROM) deficit. In contrast, others define it as those who require further intervention to treat the arthrofibrosis, such as manipulation under anesthesia (MUA) and/or knee arthroscopy. As such, a wide range has been reported for the incidence of arthrofibrosis in pediatric patients, most commonly within a 2% to 10.1% incidence rate. ¹¹ In the literature, female sex and inability to achieve 90° of flexion at 6 weeks postoperatively have been reported as risk factors for arthrofibrosis. ¹¹ Furthermore, large graft sizes have been associated with increased risk of arthrofibrosis.^10,16

The use of quadriceps tendon autografts for ACLR has become increasingly popular, and it is shown to provide favorable outcomes for pediatric patients.^3,8,21 Recent research reports a 6.3% incidence of arthrofibrosis utilizing an all-soft-tissue quadriceps tendon autograft in a pediatric population. ¹¹ The utilization of a bone block has been an ongoing topic in ACLR, with robust efficacy demonstrated with the bone-patellar tendon-bone ligament (BTB) ACLR. In the pediatric population, decisions on whether to perform a quadriceps tendon autograft ACLR with or without (all-soft-tissue) a bone block are made in part with respect to the child’s age, with all-soft-tissue reconstructions commonly used for physeal preservation in the skeletally immature child. There has been limited research focused on arthrofibrosis risk factors after ACLR using quadriceps tendon autografts with a patellar bone block in the pediatric population. This study aimed to characterize arthrofibrosis after ACLR using quadriceps tendon autografts with a patellar bone block in children and adolescents.

Methods

Data collection and manuscript creation were conducted using the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. ¹⁸ This study is an institutional review board-approved retrospective review of prospectively collected data from a single urban tertiary institution’s database (Research Electronic Data Capture). The inclusion criteria were patients aged <18 years who underwent a primary ACLR with quadriceps tendon autografts with a bone block between 2019 and 2023 and had a minimum follow-up of 6 months. All surgeries were conducted by 1 fellowship-trained orthopaedic surgeon (J.S.) specializing in pediatrics and sports medicine. The exclusion criteria were previous ACLR, multiligamentous injury, concomitant fracture sustained at the time of injury, and/or patients aged ≥18 years at the time of surgery.

Data Collection

Medical records of pediatric patients who met the inclusion criteria were reviewed. Descriptive, clinical, and operative data were obtained. Based on patient residence, the Area Deprivation Index (ADI) ¹⁷ and the Childhood Opportunity Index (COI) ⁹ were determined and used as a proxy of socioeconomic status. The graft size was measured using the size of the bone block and soft tissue. The graft diameter was recorded as the diameter of the bone block, as the bone block and soft tissue component of the graft were uniform in all cases. Postoperative records were reviewed for any complications—including arthrofibrosis and graft failures. All patients were placed in a ROM brace locked in extension until quadriceps control returned. Patients were allowed to remove the brace immediately to begin home ROM exercises and during physical therapy.

Definition of Arthrofibrosis

For this paper, we defined arthrofibrosis in 2 discrete ways as reported previously in the literature. The first definition used is based on the ROM deficits reported by Ouweleen et al. ¹¹ This definition of arthrofibrosis includes any patient who has a flexion deficit of 20° and/or an extension deficit of 10° compared with the contralateral leg. The second definition used, determined as “clinically significant arthrofibrosis,” is defined as any patient requiring additional interventions postoperatively to manage any ROM deficits effectively. ROM was identified via the senior author’s assessment (J.S.) on physical examination with the use of a handheld goniometer with a 6-inch arm length with the patient in a supine position. Hyperextension was performed passively with the extremity held in maximal hyperextension and measured. The assessment for arthrofibrosis was made starting at the 6-week follow-up time point postoperatively.

Socioeconomic Analysis

The ADI is a socioeconomic measure that utilizes census data of block groups as a representation of neighborhood geography.^6,17 The ADI ranks neighborhoods from 1 to 10 by examining such factors as income, education, employment, and housing quality. Neighborhoods with an ADI of 1 represent the least disadvantaged, while those with an ADI of 10 represent the most disadvantaged.

The COI is another socioeconomic measure that examines 29 indicators relating to education, health, and environment, and social and economic domains at the census tract level to assess opportunity for child development in a particular zip code. ⁹ Each census tract is ranked into 5 categories outlining the relative opportunities for the average child: (1) Very High; (2) High; (3) Moderate; (4) Low; and (5) Very Low. For example, a child with a COI of “Very High” has the highest amount of opportunity for child development and is thereby the least disadvantaged.

Statistical Analysis

Frequencies or percentages were reported for categorical variables. The mean and 95% CIs were reported for interval data. Categorical data were compared utilizing the chi-square test or the Fisher exact test. Interval data were compared between the groups using analysis of variance. If parametric assumptions were violated, the nonparametric Mann-Whitney U test was utilized. Additional univariable analyses were conducted on the various factors in relation to the outcome of arthrofibrosis. Univariate binary logistic regression was performed to obtain the odds ratios (ORs). The alpha was set at P < .05 to declare significance, and analyses were performed with SPSS Version 28 (IBM Corp).

Results

A total of 80 patients (53 girls and 27 boys) who underwent ACLR with QTB met the inclusion criteria. A total of 22 children (27.5%) were diagnosed with arthrofibrosis at 3 months postoperatively, given a ROM deficit, while the other 58 children had normal flexion or extension. Seven patients (8.8%) did not improve with nonoperative management and required MUA. At the time of index surgery, the mean age was 15.9 years (range, 12.8-17 years). No significant difference in age, sex, or body mass index existed (P > .05). However, patients with government insurance were more likely to develop arthrofibrosis at 3 months, compared with those with private insurance (P = .02). No differences existed between these groups regarding socioeconomic indexes, ADI and COI, (P > .05). Descriptive data and socioeconomic indexes are outlined in Table 1.

Table 1.

Descriptive Data and Socioeconomic Status of the 2 Groups ^a

	Arthrofibrosis Group (n = 22)	No Arthrofibrosis Group (n = 58)	P
Age, years	16.2 ± 1.1	15.8 ± 1.3	.18
Sex, female/male	18/4	35/23	.07
BMI, kg/m2	23.7 ± 4	24 ± 4.8	.81
Insurance type, government/private	10/12	11/47	.02
ADI, 1-10	4.4 ± 2.2	4 ± 2.2	.44
COI, VL/L/M/H/VH	0/3/4/4/11	1/8/8/14/27	.99

^aQuantitative data are presented as mean ± SD values, and qualitative data are presented with integer values. The bold P value indicates statistical significance. Group analysis was conducted for the 2 groups. The definition used for arthrofibrosis was the ROM definition, including any patient with a flexion deficit of 20° and/or an extension deficit of 10° compared with the contralateral leg. The ADI is measured on a decile scale from 1 to 10, with 10 indicating the most disparity. The COI is categorized into pentiles, VL, L, M, H, VH, with VH indicating the most opportunity or the least disparate. ADI, Area Deprivation Index; BMI, body mass index; COI, Childhood Opportunity Index; H, high; L, low; M, moderate; ROM, range of motion; VH, very high; VL, very low.

Pre- and Intraoperative Characteristics

During the preoperative physical examination, the mean flexion was 127.9° (range, 65°-140°) and the mean extension was 0° (range, –20° to 45°). Only preoperative flexion demonstrated an association with arthrofibrosis (P = .02). In contrast, preoperative extension demonstrated no such relationship (P = .70). No difference was found in time to surgery or graft size between study cohorts (P > .05). Separate analysis investigating concomitant meniscal work found that undergoing additional medial meniscal repair (MMR) was associated with a significantly higher incidence of arthrofibrosis (P = .04), while patients with lateral meniscal repair, and/or partial lateral meniscectomy were not. No patients received a partial medial meniscectomy—additional data are represented in Table 2.

Table 2.

Pre- and Intraoperative Characteristics ^a

	Arthrofibrosis Group (n = 22)	No Arthrofibrosis Group (n = 58)	P
Preop flexion, deg	120.5 ± 21.1 (65140)	130.7 ± 15 (85140)	.02
Preop extension, deg	−0.68 ± 5.4 (–20 to 10)	0.21 ± 9.4 (–20 to 45)	.70
Time to initial surgery, weeks	16.5 ± 23.8	14.3 ± 16.7	.64
Graft size, mm	9 ± 0.3	9.1 ± 0.3	.41
Concomitant MMR, yes/no	16/6	27/31	.04
Concomitant LMR, yes/no	11/11	23/35	.40
Concomitant PLM, yes/no	1/21	8/50	.43

^aQuantitative data are presented as mean ± SD (range) values. Qualitative data are presented using integer values. The bold P values indicate statistical significance. Between-group analysis was performed on the 2 groups. The definition used for arthrofibrosis was the ROM definition, including any patient with a flexion deficit of 20° and/or an extension deficit of 10° compared with the contralateral leg. LMR, lateral meniscal repair; MMR, medial meniscal repair; PLM, partial lateral meniscectomy; Preop, preoperative; ROM, range of motion.

Postoperative Characteristics

Patients who had arthrofibrosis at 3 months postoperatively were more likely to have a diminished flexion angle at 6 weeks (P < .001). However, no difference in early postoperative knee extension was noted between groups. The mean final follow-up was 13.3 months (range, 6-47 months), with no difference in follow-up duration between the 2 groups (P = .96). On average, children who experienced arthrofibrosis at 3 months had slightly lower mean flexion at the final follow-up (136°) than those who did not (139°) (P = .007). No difference was noted in knee extension at the final follow-up (P = .08). Additional data are represented in Table 3.

Table 3.

Postoperative Characteristics ^a

	Arthrofibrosis Group (n = 22) b	No Arthrofibrosis Group (n = 58) c	P
6-week postop flexion	91.7 ± 18.3	111.9 ± 19.4	<.001
6-week postop extension	3.5 ± 7.2	1.2 ± 4.1	.08
3-month postop flexion	109.3 ± 16.3	136.3 ± 5.6	<.001
3-month postop extension	3.5 ± 4.7	−0.8 ± 4.4	<.001
Follow-up duration, years	13.3 ± 7.9	13.2 ± 8.5	.96
Knee flexion at last follow-up	136 ± 5.4	139 ± 3.9	.007
Knee extension at last follow-up	0.91 ± 2	−1.57 ± 6.3	.08

^aQuantitative data are presented as mean ± SD. Bold P values indicate statistical significance. Between-group analysis was performed on the 2 groups. The definition used for arthrofibrosis was the ROM definition, including any patient with a flexion deficit of 20° and/or an extension deficit of 10° compared with the contralateral leg. Postop, postoperative; ROM, range of motion.

^bOne patient lacked 6-week follow-up data.

^c4 patients lacked 6-week follow-up data.

Postoperatively, all patients were placed in a knee brace and attended physical therapy. Patients with arthrofibrosis at 3 months spent a mean of 5.3 weeks in a postoperative brace (range, 4-8 weeks). This was significantly longer than those without arthrofibrosis, as they spent a mean of 4.6 weeks in a postoperative brace (range, 1-8 weeks) (P = .03). The first postoperative physical therapy appointment took place within 13.2 days, and patients received 18.02 visits within the first 3-month postoperative period. No difference existed with regard to time to physical therapy after surgery (P = .71) or amount of physical therapy within the first 3 months postoperatively (P = .51).

Odds Ratios

Binary logistic regression analysis of significant differences between arthrofibrosis incidences demonstrated multiple significant ORs. Qualitatively, the OR for the presence of MMR association with arthrofibrosis was 3.1 (1.05-8.93) (P = .04). Government insurance has an increased risk of arthrofibrosis with an OR of 3.6 (1.2-10.3) (P = .02). Quantitatively, for every 1-degree increase in flexion preoperatively, we expect to see a 3% increase in the odds of not having arthrofibrosis (P = .03). For every 1-degree rise in flexion at 6 weeks, we expect to see a 5.5% increase in the odds of not having arthrofibrosis (P < .001). Finally, for every 1-week increase in time in the knee ROM brace, there is a 45% increase in odds of having arthrofibrosis (P = .04).

Discussion

Our findings of 27.5% arthrofibrosis at 3 months are significantly higher than those of reported outcomes after ACLR, especially when compared with findings of Ouweleen et al, ¹¹ which had a 6.3% arthrofibrosis in ACLR with quadriceps tendon soft tissue autografts. However, most patients (15/22 [68.2%]) who had reduced ROM at 3 months would eventually regain full motion without additional operative intervention. Furthermore, based on a mean follow-up of 13.3 months, all patients regained functional ROM, acknowledging that this varies between approximately 120° to 140° depending on the patient, with only 7 patients (8.8%) of the entire study requiring MUA. This demonstrates that while ACLR with QTB may be at greater risk of stiffness earlier in the postoperative period, by 1 year, most patients will regain full mobility of their knee, similar to other reported grafts. ¹⁰

The use of quadriceps tendon autografts with or without a patellar bone block is a viable and safe option for ACLR. ⁸ Proponents of QTB claim that it has similar properties to the BTB with its bone-to-bone healing, without the increased risk of anterior knee pain. ⁷ Setliff et al ¹⁴ reported no difference in clinical outcomes between quadriceps tendon autograft ACLR performed with or without a bone block. As such, multiple studies support strong patient-reported outcomes with low graft failure.^3,8 Complications such as arthrofibrosis after ACLR are rare, but they can carry a significant effect. However, the number of patients experiencing arthrofibrosis has been steadily increasing as more youth are getting involved in sports and subsequently requiring ACLR.^15,19

In many cases, arthrofibrosis can be effectively prevented and treated nonoperatively with measures such as intensive physical therapy regimens and dynamic splinting. ¹² If nonoperative management fails, MUA with lysis of adhesions may be required to restore a patient’s ROM; however, their subjective level of functionality may remain slightly diminished. ²⁰ Therefore, it is important to analyze potential risk factors for arthrofibrosis to prevent unnecessary surgery as well as improve long-term outcomes after ACLR.

Another risk factor for arthrofibrosis discussed in the literature is concomitant meniscal injury.^2,5 These data also demonstrate that children and adolescents who underwent concomitant MMR in addition to ACLR were found to be at a statistically higher risk for arthrofibrosis (P = .04). Haley et al ⁴ recorded similar findings, with meniscal repair proving to be a risk factor for arthrofibrosis in their cohort, which may be due to discomfort at higher degrees of knee flexion in patients who underwent meniscal repair.

In this study, children who had government-funded insurance were more likely to develop arthrofibrosis (P = .02). This may be due to difficulty obtaining required authorizations from insurance or difficulty scheduling an appointment postoperatively at a facility that accepts the patient’s insurance. This coincides with findings from Patel et al ¹³ who demonstrate that having government-funded insurance results in a higher risk for postoperative complications.

In addition, the insurance type may affect the quality and quantity of physical therapy sessions. In this study, there was no association in time to physical therapy or amount of physical therapy visits in the first 3 months (P = .71 and P = .51, respectively). However, the overall expertise, quality of care, and level of direct attention by the physical therapist may be indirectly skewed depending on the practice’s insurance policy.

In this study, there was no difference in socioeconomic status between those who developed arthrofibrosis and those who did not, according to 2 metrics—ADI and COI. However, the methods of this study were not equipped to thoroughly analyze this effect specifically. Additional studies using large patient sizes are needed, as there is a paucity of literature investigating the impact of common health care disparities of ACLR arthrofibrosis, especially in the pediatric population.

Interestingly, in this study, some factors that have previously been reported to be risk factors for the development of arthrofibrosis were found to have no significant relationships. Female sex has been a predictor of arthrofibrosis in the literature, with its significance noted with ACLR with all-soft-tissue quadriceps tendon autografts. ⁴ Our data did not support any association with patient sex, as the data approached but did not reach statistical significance (P = .07). Increased graft size has been another common predictor for arthrofibrosis.^1,4,16 However, in this study, no such association existed (P = .41), likely because of the surgeon’s preference in graft harvesting and standardization.

Limitations

This study had several limitations: it was retrospective; a single surgeon performed the ROM measurements; it had a relatively small sample size considering the amount of ACLR performed annually; and the follow-up period was only 13.3 months. The lack of a standardized arthrofibrosis definition was a significant limitation. In addition, the lack of contralateral knee ROM might allow for unnecessary error, as the assumed ROM of 0° to 140° might not be representative of each patient’s baseline ROM.

Because of the study window, an interesting point from this study is the potential effect of the coronavirus disease 2019 (COVID-19) pandemic on the development of arthrofibrosis. The COVID-19 pandemic may have affected many of our patients' rehabilitation, whether due to direct sickness or the effect on the business itself due to contact precautions. While, to our knowledge, none of the patients had any recorded COVID-19 + diagnoses, our data were limited because of the family's willingness to offer that information and the subsequent effect on the patient's life.

Conclusion

This study reports that more children and adolescents undergoing ACLR with QTB have a higher rate of arthrofibrosis at 3 months. However, return to the operating room for MUA is similar to other graft types. Reduced flexion preoperatively, at 6 weeks postoperatively, and MMR were significant predictors of arthrofibrosis.