Fixation Strategy Does Not Affect Risk of Growth Disturbance After Surgical Treatment of Pediatric Tibial Spine Fracture

Abstract

Purpose

To characterize growth abnormalities after surgical treatment of tibial spine fractures and to investigate risk factors for these abnormalities.

Methods

A retrospective analysis of children who underwent treatment of tibial spine fractures between January 2000 and January 2019 was performed, drawing from a multicenter cohort among 10 tertiary care children’s hospitals. The entire cohort of surgically treated tibial spine fractures was analyzed for incidence and risk factors of growth disturbance. The cohort was stratified into those who were younger than the age of 13 years at the time of treatment in order to evaluate the risk of growth disturbance in those with substantial growth remaining. Patients with growth disturbance in this cohort were further analyzed based on age, sex, surgical repair technique, implant type, and preoperative radiographic measurements with χ², t-tests, and multivariate logistic regression.

Results

Nine patients of 645 (1.4%) were found to have growth disturbance, all of whom were younger than 13 years old. Patients who developed growth disturbance were younger than those without (9.7 years vs 11.9 years, P = .019.) There was no association with demographic factors, fracture characteristics, surgical technique, hardware type, or anatomic placement (i.e., transphyseal vs physeal-sparing fixation) and growth disturbance.

Conclusions

In this study, we found an overall low incidence of growth disturbance after surgical treatment of tibial spine fractures. There was no association with surgical technique and risk of growth disturbance.

Level of Evidence

Level III, retrospective comparative study.

Tibial spine fractures occur when the tibial insertion of the anterior cruciate ligament (ACL) is avulsed from the tibia.^1,2 These fractures were historically considered the pediatric equivalent of an adult ACL rupture and are thought to occur more commonly in children due to the incomplete ossification of the tibial eminence.² Fractures with substantial displacement typically are treated operatively with suture or screw fixation of the fracture, either via an open or arthroscopic approach.2, 3, 4, 5 Surgical treatment of tibial spine fractures is not without potential risk of complications, including infection, arthrofibrosis, anterior cruciate ligament injury, knee laxity, and meniscal injury.6, 7, 8, 9, 10, 11, 12

Growth disturbance after physeal injury or fracture is a well-known phenomenon.13, 14, 15, 16 Surgical violation of the physis in pediatric patients during fracture care, ACL reconstruction, and tibial spine fixation has been discouraged due to the risk of growth disturbance.17, 18, 19, 20 There is a paucity of literature regarding the occurrence and characterization of growth abnormalities as a sequela of tibial spine fracture treatment. Currently, there are only a few case reports regarding such events, which have resulted in overgrowth of the affected extremity or an angular deformity.21, 22, 23 Growth abnormalities may result in impaired long-term function and necessitate surgical correction. As such, it is important to be aware of the potential association between growth disturbance and tibial spine fracture treatment. The purposes of this study were to characterize growth abnormalities after surgical treatment of tibial spine fractures and to investigate risk factors for these abnormalities. We hypothesized that surgical treatment of tibial spine fractures would carry a small risk of growth disturbance, most likely an angular deformity in the coronal and/or sagittal planes.

Methods

Institutional review board (IRB) approval and institutional review board reliance was granted for 10 geographically diverse pediatric hospitals with The Children’s Hospital of Philadelphia as the lead site. Participating sites were The Children’s Hospital of Philadelphia, Boston Children’s Hospital, Children’s Medical Center Dallas, Johns Hopkins Children’s Center, the Hospital for Special Surgery, Hasbro Children’s Hospital in Rhode Island, Seattle Children’s Hospital, Texas Children’s Hospital, Texas Scottish Rite Hospital for Children, and Rainbow Babies & Children’s Hospital.

A retrospective review was performed using data from patients who underwent treatment of tibial spine fractures between January 2000 and January 2019. Patients older than the age of 18 years or those treated nonoperatively or with closed reductions were excluded. Data collected included demographics, injury mechanism, fracture classification (Meyers and McKeever), physical examination findings, method of treatment, surgical details, postoperative protocols, follow-up time, and presence of growth disturbance. Data from surgical treatment included surgical approach (open vs arthroscopic), fixation strategy relative to the physis (e.g., transphyseal vs physeal-sparing fixation), operative time, and fixation method (suture, screw, washer, soft-tissue anchor, Kirshner wires, or hybrid/mixed technique). Clinical follow-up data also was recorded including length of follow-up, presence of growth disturbance, mechanical axis in those with growth disturbance, treatment of growth disturbance, and leg-length discrepancy.

The entire cohort of patients was analyzed for risk factors for growth disturbance. Growth disturbance was defined by either a leg-length discrepancy or an angular deformity (as documented in the medical record) that was asymmetric to the contralateral side. Anatomic and mechanical angles were measured and compared with the uninjured leg. Those with growth disturbance were analyzed for risk factors and compared with those with no documented postoperative growth disturbance. Given that the risk of growth disturbance is greater in younger patients with more potential growth remaining, a subanalysis of patients younger than the age of 13 years was performed.

Statistical Analysis

Descriptive statistics were performed to examine demographic characteristics of all patients in the cohort. Growth disturbance rates were calculated for the entire patient cohort, as well as the cohort of patients who were younger than the age of 13 years. Potential risk factors were assessed using either χ² analysis or Fisher exact testing for categorical data or a 2-sample independent t-test for continuous data, as appropriate. For the subgroup of patients younger than 13 years old, risk factors also were assessed using either χ² analysis or Fisher exact testing for categorical data or a 2-sample independent t-test for continuous data. A binary multivariate logistic regression model was also used to determine whether any risk factors were associated with growth disturbance. Statistical significance for hypothesis testing was set to an alpha level of 0.05 (SPSS Statistics, version 25.0; IBM Corp., Armonk, NY).

Results

Entire Cohort

A total of 661 patients were identified. Of the 645 patients who underwent operative treatment of tibial spine fractures, 425 were male (65.9%). The mean age was 11.9 ± 2.89 years (range, 3.5- 18 years). The average follow-up time was 10.8 months (interquartile range 10.3 months). In the entire cohort, there were 9 patients who experienced growth disturbance, a 1.4% incidence rate. Eight of the 9 patients developed overgrowth in the operative leg with an average of 1.075 cm (standard deviation [SD] 0.50 cm), whereas 1 patient developed a valgus angular growth disturbance/deformity. Patients who experience a growth disturbance were statistically significantly younger at the time of their tibial spine fracture compared with those with no growth disturbance (9.7 years old [SD 1.9 years] vs 11.9 years old [SD 2.8 years], respectively, P = .019).

Younger Than 13 Years Old Cohort

In total, 404 patients younger than the age of 13 years were analyzed (Table 1). The mean age in this cohort was 10.1 years (SD 1.99 years). There were 221 male patients (54.7%). The average follow-up time was 10.8 months (interquartile range 11.2 months). In this group, there were 9 patients with growth disturbance, an incidence of 2.2%. There was no statistically significant relationship based on multivariate logistic regression between the occurrence of growth disturbance and demographic factors or preoperative injury characteristics, including sex (P = .822), age (P = .843), Meyers and McKeever fracture type (P = .085), injury mechanism (P = .176), surgical time (P = .157), surgical approach (open vs arthroscopic vs closed reduction) (P = .999), implants used (P = .179), the use of resorbable or non-resorbable implants – screw (P = .998), suture (P = .073), or whether hardware was placed epiphyseal or transphyseal (P = .867) (Table 2).

Table 1.

Patient Demographics for Patients Younger Than the Age of 13 Years Who Underwent Surgical Treatment for Tibial Spine Fractures

	Mean or No.	% or SD
Age, y	10.1	1.9
Sex, n (%)
Male	221	54.7%
Female	183	45.3%
Fracture type (Myers and McKeever), n = 335)
Type I	24	7.2%
Type II	145	43.3%
Type III	142	42.4%
Type IV	24	7.2%

NOTE. n value denotes the number of patients available for analysis.

SD, standard deviation.

Table 2.

Surgical Characteristics for Tibial Spine Fracture Fixation in Those With and Without the Occurrence of Growth Disturbance (Younger Than 13 Years Cohort)^∗

	Growth Disturbance	No Growth Disturbance	P Value
Surgical approach (n = 324)
ARIF	8	279
ORIF	1	36
Surgical time (n = 196)
Average time, min (SD)	99.00 (42.0)	124.16 (48.3)	P = .157
Implant
Suture
Resorbable	5	89
Nonresorbable	2	65
Both	0	7
			P = .653
Screw
Resorbable	1	17
Nonresorbable	0	82
Both	0	1
			P = .097
Hardware placement
Transphyseal	3	60
Epiphyseal	4	110
			P = .900
Implants used
Suture	5	147
Screw	2	84
Screw + washer	0	11
Anchor	1	36
Suture + screw	1	17
Anchor + screw/washer	0	11
Suture + screw/washer	0	1
Pins	0	4
			P = .991

NOTE. P values based on binary logistic regression (<13 years old cohort).

ARIF, arthroscopic reduction internal fixation; ORIF, open reduction internal fixation; SD, standard deviation.

^∗Patients with incomplete data were not included in analyses. n values correlate to the number of patients included in analysis for each variable.

Case Analysis for Those With Growth Disturbance

The 9 patients who developed growth disturbance were further inspected for specific characteristics (Table 3). The mean follow-up time for these 9 patients was 44.8 months (median 32.6 months). The mean age of the growth disturbance group was 9.64 years (range, 8.1-12.8). Among growth disturbance patients, 2 were female and 7 were male.

Table 3.

Descriptive Factors Associated With Patients Experiencing Growth Disturbance After Tibial Spine Fracture Treatment

	Patient 1	Patient 2	Patient 3	Patient 4	Patient 5	Patient 6	Patient 7	Patient 8	Patient 9
Age, y	12.8	11.1	10.3	6.7	10.8	7.4	9.6	10	8.1
Sex	Male	Male	Female	Male	Male	Male	Male	Female	Male
Follow-up time, mo	17.3	50.3	81.9	115.5	18.0	57.4	9.5	20.7	32.6
Surgical treatment	ARIF	ORIF	ARIF	ARIF	ARIF	ARIF	ARIF	ARIF	ARIF
Fixation: screw or suture (type)	Suture	Screw and Suture	Screw	Screw	Suture	Suture	Suture	Suture	Suture
Placement of screw or suture	E	N/A∗	E	E	T	T	T	T	E
Type of suture	R	NR	None	None	R	R	R	NR	R
Type of screw	None	R	NR	NR	None	None	None	None	None
LLD, cm	+0.9	+1.3	+1.5	+2	+0.5	+1	+0.8	Valgus angular deformity	+0.6
Treatment to address growth arrest?	None	Shoe lift	1/4-in lift	3/8-in lift	None	None	None	Proximal medial tibial hemiepiphysiodesis	None

ARIF, arthroscopic reduction internal fixation; E, epiphyseal; LLD, leg-length discrepancy; NR, nonresorbable suture or screw, ORIF, open reduction internal fixation; R, resorbable screw/suture, T, transphyseal.

^∗N/A correlates with data not available.

Among patients with growth disturbance, 8 were treated with arthroscopic reduction internal fixation and 1 with open reduction internal fixation as a means of surgical repair. Seven patients had sutures placed for fixation, with 5 of these patients having resorbable sutures and 2 having nonresorbable. Two patients had nonresorbable metal screws placed during fixation, and one had a resorbable screw placed. One patient received both sutures (nonresorbable) and screws (resorbable). One patient had a soft-tissue anchor placed. In terms of fixation location, 4 patients had the fixation placed transphyseally, and 4 patients had the fixation placed all-epiphyseal, 1 was unknown/not recorded. Eight of 9 patients with growth disturbance had a difference in leg length, and the other patient had a valgus coronal angular deformity. The average difference between legs was 1.075 cm (SD 0.50 cm), where the operative leg developed an overgrowth compared to the contralateral leg. There were no recurvatum/procurvatum deformities seen in the series. Limb-length discrepancy was treated as follows: 4 patients received no treatment, and 3 patients were given shoe lifts in the contralateral leg. One patient with an angular deformity was treated with implant-mediated guided growth.

Discussion

This study found growth disturbance following operatively treated tibial spine fractures in 9 patients in a large multicenter cohort, yielding an incidence of 1.4% in the entire cohort and 2.2% incidence in patients younger than the age of 13 years. Eight of the 9 patients experienced a leg-length discrepancy that consisted of overgrowth of the injured, operative leg. The other patient developed a valgus deformity. Patients who were younger at the time of their tibial spine fracture were more likely to experience growth disturbance. However, in the cohort younger than the age of 13, there was no statistically significant association with age at the time of surgery and risk of growth disturbance. In this younger cohort, there was no association with demographic factors, injury factors, fracture characteristics or any surgical technique and the development of growth disturbance.

The tibial spine is the attachment site of the ACL and the bony prominence is located in the epiphysis of the proximal tibia. Fracture of the tibial spine itself does not usually infer injury to the physis, unlike tibial tubercle or Salter–Harris type proximal tibia fractures.24, 25, 26, 27 However, the surgical treatment of tibial spine fractures can potentially cause growth issues if any hardware or instrumentation violates the physis. Physeal-sparing surgery has been a point of emphasis in pediatric ACL reconstruction for this reason.^17,19,28 More recently, there has also been interest in physeal-sparing tibial spine fixation with both screw and suture fixation in an emphasis to prevent any potential injury to the growth plate.^20,29

There is a scarcity of literature about growth disturbance following tibial spine injury and/or surgery and overall, it appears to be a relatively rare complication. Ahn and Yoo²¹ describe a case series of 14 patients who underwent arthroscopic suture fixation of tibial spine fracture with long term follow-up (3-6 years). In this series, 2 patients developed leg-length discrepancy with overgrowth of 1 cm on the operative extremity. Similarly, the case of a 12-year-old female patient who underwent suture fixation of a tibial spine fracture, complicated by postoperative arthrofibrosis requiring manipulation under anesthesia, who developed a 1.75-cm leg-length discrepancy more than 2.5 years after surgery has been published.²² The phenomenon of overgrowth seen in these previous case reports is similar to the growth disturbances we observed in this study. Overgrowth of an injured extremity may be observed in other pediatric injuries and is a well-known phenomenon in femoral shaft fractures.^16,30 However, it is much less commonly investigated in tibia fractures, especially tibial spine fractures. The mechanism of overgrowth of the injured leg is unclear, both in the current study as well as in previous studies. Previous literature suggests it may be a combination of chondrocyte proliferation, increased blood flow, or increased expression of antiapoptosis genes in osteoblasts.^22,30, 31, 32 In femoral shaft fractures, younger age at the time of fracture had an influence on the amount of overgrowth seen. This may parallel the current study, as growth disturbance was more likely to occur if the patient was younger at the time of injury.³⁰ Interestingly, in the current study in the entire cohort, there was a correlation between age at the time of surgery and occurrence of growth disturbance but not in the cohort younger than the age of 13 years. Due to the scarcity of literature on this complication, risk of age and growth disturbance has not been discussed in cohort studies to this point. In the pediatric ACL literature, a systematic review by the PLUTO (Pediatric ACL: Understanding Treatment Options) group found growth disturbances from patients ages 7-16 years with no clear delineation of risk based on the age.³³

One of the 9 patients with growth abnormalities in our study had an angular deformity that ultimately required proximal medial tibial hemiepiphysiodesis. Typically, in tibial spine fracture fixation, the major concern is a sagittal plane deformity due the anteriorly based tunnel for suture repair of the tibial spine, similar to what can be seen with an ACL reconstruction in pediatric patients.^17,19 However, none of the patients in our cohort developed a sagittal plane deformity. In the series by Ahn and Yoo,²¹ one of the patients with a leg-length discrepancy also developed a genu recurvatum deformity.²¹ Mylle et al.²³ also presented a case report of an 11-year-old girl who developed a 25° recurvatum deformity after transepiphyseal screw fixation of a tibial spine fracture. There is one case report that described a coronal plane deformity after tibial spine fixation, similar to the patient in our series. Fabricant et al.³⁴ report a case of a 13-year-old male patient who developed a valgus deformity 19 months after tibial spine fixation with a transphyseal screw. The patient later underwent a guided growth surgery, similar to the patient in our cohort

Limitations

There are several limitations in this study. Given the retrospective nature of the study, there is a potential for inherent bias. This study also encompasses many centers and many different surgeons. The decision-making for surgical approach, implant choice, and postoperative care is surgeon and facility dependent, and there was no standardization for treatment in this retrospective cohort. The follow-up data for all patients is also not uniform. This is apparent when comparing the patients with growth abnormalities, where the follow-up average is 44.8 months and the follow-up of the entire cohort is 10.8 months. In order to clinically appreciate leg-length discrepancies or angular deformities, it may take years of follow-up data. Since the average follow-up time in our cohort was less than 1 year, it is possible that we may be underestimating the true risk of growth disturbance. In our study, the average leg-length discrepancy was 1.075cm. Discrepancies less than 2 cm may be well tolerated without any intervention, or with conservative management with a shoe lift.³⁵ It is unclear what the clinical impact may be on patients who develop postoperative limb overgrowth. Another limitation may be that long leg films to investigate growth disturbance or leg-length discrepancy were only completed on patients who had a suspected clinical asymmetry and were not routinely performed, potentially underestimating the true number of patients with growth disturbance. It also should be noted that 24 patients with Myers and McKeever type I fracture underwent surgical intervention, whereas type I fractures typically are managed nonoperatively. All decisions to pursue operative treatment were surgeon-dependent. Given the retrospective nature of the study, it is unclear the exact indication in each case.

Conclusions

In this study, we found an overall low incidence of growth disturbance after surgical treatment of tibial spine fractures. There was no association with surgical technique and risk of growth disturbance.