Abstract
Background
The management of pediatric tibial spine fractures is evolving as the evidence base has grown. Recent evidence suggests magnetic resonance imaging (MRI) evaluation may be warranted although optimal management is unknown. The present study sought to characterize national trends in surgical management and MRI utilization of pediatric tibial spine fractures.
Methods
Patients aged 6-17 years presenting with a first diagnosis of closed tibial spine fracture were identified from the 2016-2022 PearlDiver M170 database. MRI utilization, treatment strategy, and surgical technique (arthroscopic reduction and internal fixation [ARIF], open reduction and internal fixation [ORIF], or anterior cruciate ligament reconstruction) were determined for each injury. Factors independently associated with these variables were assessed by a multivariable analysis. The year 2020 was excluded from analyses of trends due to changes to health care utilization during the COVID-19 pandemic.
Results
Overall, 3,554 pediatric tibial spine fracture patients were identified. MRI was obtained for 681 patients (19.1%), with independent predictors of MRI use being increased age (odds ratio [OR] = 1.12 per year, P < .0001), fracture displacement (OR = 2.38, P < .0001), and geographic region (Midwest relative to West, OR = 1.57, P = .0013). A total of 806 patients (22.7%) underwent surgical management, with fracture displacement (OR = 14.13, P < .0001) and preoperative MRI (OR = 7.44, P < .0001) identified as independent predictors of this treatment strategy. Of patients treated surgically, ARIF was performed for 522 patients (64.8%), with an independent predictor being preoperative MRI (OR = 3.00, P < .0001). Notably, the rate of MRI utilization increased significantly through the study period (15.7% in 2016, 23.1% in 2022, R2 = 0.795, P = .0170), while the rates of operative treatment and ARIF did not change significantly.
Conclusions
In this large database study of pediatric tibial spine fractures featuring over 3,000 patients treated over nearly a decade, we identified that MRI was obtained for 19.1% of the study population. The rate of MRI utilization increased over the course of the study period. Moreover, surgery was performed on 22.7% of the population. There were geographic variations in MRI and ARIF utilization, suggesting nonclinical factors impacting management.
Key Concepts
-
(1)
Only a minority of tibial spine fracture patients received magnetic resonance imaging, despite recognition of the frequency of associated injury, though this percentage did increase over the study period.
-
(2)
Geographic variations in imaging and surgical treatment were identified, suggesting nonclinical factors impacting management.
-
(3)
Fracture displacement was the strongest predictor of surgical treatment, which was most commonly performed arthroscopically.
Level of Evidence
III
Keywords: Tibial spine, Tibial eminence, Magnetic resonance imaging, ORIF, Arthroscopic fixation
Introduction
Tibial spine fractures are often functionally equivalent to anterior cruciate ligament (ACL) injuries in pediatric populations. These fractures are most common in children aged 7-18 years, with a mean age of 12-13 years [[1], [2], [3], [4]]. These fractures may occur from twisting-type sports injuries [3,5] or higher-energy trauma due to the lower tensile strength of the tibial spine than that of the ACL in this population [4]. While there are shifting perspectives on the treatment of tibial spine fractures, trends and drivers of these decisions are not well quantified.
Magnetic resonance imaging (MRI) is often considered in the workup of tibial spine fractures as they may occur in conjunction with other associated knee injuries [[6], [7], [8]] including meniscus injury in 21-40% of patients [[7], [8], [9], [10]] and non-ACL ligamentous injury in 32.4% of patients [8]. While many studies suggest routinely obtaining MRI as part of the workup of patients with tibial spine fractures [[10], [11], [12], [13], [14], [15]], some suggest that MRI may be unnecessary if associated injuries can be directly assessed during surgery [9,16]. Furthermore, while one multicenter study of 434 patients from 2000 to 2019 found no difference in rates of MRI utilization among children with private and public insurance [17], the impact of other patient factors on MRI utilization has not been investigated.
In patients sustaining tibial spine factures, the operative rate has been reported to be 42.1%-75% [1,18,19]. Nondisplaced and minimally displaced fractures (Meyers and McKeever types I and II) are typically managed nonsurgically [3,14,20], while displaced fractures (types III and IV) are typically managed surgically [21,22]. The goals of surgery are to restore knee stability [14,[23], [24], [25]], optimize motion [1,14,26], facilitate return to sports [27], and decrease the risk of subsequent degenerative changes. Nonoperative management may be associated with increased ACL laxity and higher risk for subsequent surgeries such as anterior cruciate ligament reconstruction (ACLR). Surgery may be associated with a higher risk of arthrofibrosis [18,19]. For those undergoing surgery for tibial spine fractures, open reduction and internal fixation (ORIF) and arthroscopic reduction and internal fixation (ARIF) are two treatment options that exist [28]. There is conflicting evidence regarding the optimal surgical technique, although ARIF is often used by surgeons with additional sports subspecialty training [[29], [30], [31], [32], [33], [34], [35], [36]].
The largest study of pediatric tibial spine fractures to date analyzed 876 patients from 2016 to 2018, assessing a limited number of epidemiologic variables and management/outcomes variables with the private insurance Truven database [2]. This database was limited by an inability to compare different surgical methods. Noting the limitations of prior studies assessing the workup and management of patients with pediatric tibial spine fractures, the current study aimed to leverage the scale of a large, national, mixed-insurance, administrative database over the extended time period of 2016-2022 to examine the MRI utilization and treatment (nonoperative vs operative and ORIF vs ARIF) trends in the management of pediatric tibial spine fractures.
Methods
Study population
Data were retrieved from the PearlDiver M170 database, a national, multi-insurance administrative claims database comprised of 170 million covered patient lives from 2010 to April 2023. To our knowledge, this is the only pediatric tibial spine fracture manuscript utilizing this database. As all data accessed in PearlDiver are aggregated and deidentified, this study was exempt from review by our institutional review board.
Patients with tibial spine fracture were identified with International Classification of Diseases 10 (ICD-10) codes, which stratify for displaced and nondisplaced fractures. This provided additional granularity that was a noted limitation of the previously mentioned Truven database study [2]. As the ICD-9 coding system used prior to 2015 lacks the granularity to separate tibial spine fractures from other proximal tibial fractures (eg, tibial plateau or tuberosity fractures), the study was restricted to patients with an ICD-10 code for tibial spine fracture from the years 2016 to 2022. The ICD-10 codes for displaced or nondisplaced tibial spine fractures came into effect on October 1, 2015, and as such, for temporal simplicity and accuracy, the study period began in 2016.
To focus on a pediatric population with similar injury patterns, and in line with the literature demonstrating these injuries being most common in patients aged 7-18 years [[1], [2], [3], [4]], patients were excluded if they were younger than 6 years or older than 17 years. Additional exclusion criteria were open fractures, prior tibial spine fractures or tibial spine fracture treatment, less than 90 days of follow-up after diagnosis, and having a fracture outside of the knee (defined as distal to the tibial tuberosity). Fractures distal to the tibial tuberosity were excluded to reduce potential confounding of MRI utilization results for some nonspecific lower-extremity codes of MRI described later. Patients with a Current Procedural Technology (CPT) code for ACLR before or concurrent with a CPT code for ARIF or ORIF were also excluded due to ambiguity of this treatment.
Patient and injury characteristics
Patient characteristics were abstracted from the database. These included age, sex Elixhauser Comorbidity Index (ECI; an estimate of patient comorbidity burden based on ICD-10 diagnosis codes) [37], insurance (commercial or Medicaid), and geographic region (West, Midwest, Northeast, or South). Fracture displacement was determined by the ICD-10 code and categorized as nondisplaced or displaced.
MRI utilization was assessed based on CPT codes 73721, 73722, and 73723. This was tracked in an interval limited to 90 days after diagnosis and the 90 days prior to surgery in patients treated operatively.
Surgical treatment was defined as ORIF (CPT-27540), ARIF (CPT-29850, CPT-29851), or ACLR (CPT-29888) based on the respective CPT codes present within 90 days of fracture diagnosis. Patients with both CPT codes 27540 and 29850 were assumed to be surgeries that were converted from arthroscopic to open and were only included in the ORIF population. Patients with an initial code for ARIF or ORIF and a later code for ACLR were considered to have an adverse outcome of their initial surgery necessitating ACLR and were thus considered in their respective ARIF/ORIF groups. Of note, patients with isolated ACLR without ORIF or ARIF of tibial spine fractures were not included in analyses comparing open vs arthroscopic fixation for tibial spine fractures as they were not deemed to be the primary treatment recipients for this.
Data analysis
Statistical analysis was performed using PearlDiver Bellwether software (PearlDiver Technologies, Colorado Springs, CO, USA) and Prism 9 (GraphPad Software, Boston, MA, USA). Statistical significance was defined as a P value <.05 for all analyses, with Bonferroni corrections applied for each separate analysis to account for multiple comparisons.
A univariable analysis was conducted to compare characteristics of patients who received nonoperative, ARIF, or ORIF treatment and who did or did not receive MRI. Differences in categorical variables were compared using chi-square tests. Student’s t-tests were used to compare continuous variables. Multivariable logistic regression was used to determine factors independently associated with MRI utilization, surgical treatment, and ARIF utilization. Linear regression was used to evaluate trends in MRI utilization, surgical treatment, and ARIF utilization over the course of the study period. The year 2020 was excluded from analyses of trends due to changes to health care utilization during the COVID-19 pandemic.
Results
Patient cohort/MRI utilization
Overall, 3,554 patients aged 6-17 years with a diagnosis of tibial spine fracture between 2016 and 2022 were identified as meeting study criteria (Fig. 1).The three most common ages for tibial spine fractures in our study population were 13, 14, and 15 years. MRI was obtained in 681 of the 3,554 (19.2%) patients in this cohort, and the percent of patients receiving MRI significantly increased from 2016 to 2022 (P = .0170, R2 = 0.795, Fig. 2).
Figure 1.
Agedistribution ofpediatrictibialspinefracture. Histogram of patient age at diagnosis of tibial spine fracture in 6- to 17-year-old patients.
Figure 2.
Trend inpercent ofpediatrictibialspinefracturepatientsundergoing MRI. Trend in utilization of MRI from 2016 to 2022. R2 = 0.795, P = .0170. R2 and P values were calculated by linear regression analyses. MRI, magnetic resonance imaging.
Predictors of patients having had an MRI for the workup of these injuries are delineated in Table 1. Significant predictors of MRI utilization by the univariable analysis included older age (P < .0001), male sex (P = .0034), geographic region (P = .0005), and fracture displacement (P < .0001). Significant factors by the multivariable analysis included older age (odds ratio [OR] = 1.12, P < .0001), fracture displacement (OR = 2.38, P < .0001), and Midwest geographic regions (relative to West, OR = 1.57, P = .0013). There were no significant differences by the multivariable analysis with regards to sex, ECI, or insurance type.
Table 1.
Univariable and multivariable analyses of MRI utilization relative to no utilization in pediatric patients with tibial spine fractures.
| Univariable | Multivariable | ||||
|---|---|---|---|---|---|
| No MRI | MRI | P value | Odds ratio (95% CI) | P value | |
| Total | 2,873 (80.8%) | 681 (19.2%) | |||
| Age (mean ± SD) | 12.41 ± 2.91 | 13.28 ± 2.54 | P < .0001 | 1.12 (1.08,1.15) | P < .0001 |
| Sex | P = .0034 | ||||
| Female | 937 (32.6%) | 182 (26.7%) | Reference | ||
| Male | 1936 (67.4%) | 499 (73.3%) | 1.15 (0.95,1.40) | P = .1582 | |
| ECI (mean ± SD) | 0.90 ± 1.11 | 0.98 ± 1.14 | P = .1245 | 1.03 (0.95,1.11) | P = .4686 |
| Insurance | P = .3721 | ||||
| Commercial | 2,232 (77.7%) | 525 (77.1%) | Reference | – | |
| Medicaid | 570 (19.8%) | 137 (20.1%) | 1.01 (0.81,1.25) | P = .9635 | |
| Region | P = .0005 | ||||
| West | 501 (17.4%) | 107 (15.7%) | Reference | – | |
| Midwest | 601 (20.9%) | 187 (27.5%) | 1.57 (1.20,2.07) | P = .0013 | |
| Northeast | 515 (17.9%) | 138 (20.3%) | 1.34 (1.01,1.79) | P = .0484 | |
| South | 1,223 (42.6%) | 244 (35.8%) | 1.02 (0.79,1.33) | P = .8608 | |
| Fracture type | P < .0001 | ||||
| Nondisplaced | 695 (24.2%) | 81 (11.9%) | Reference | – | |
| Displaced | 2,178 (75.8%) | 600 (88.1%) | 2.38 (1.86,3.07) | P < .0001 | |
CI, confidence interval; SD, standard deviation; ECI, Elixhauser Comorbidity Index; MRI, magnetic resonance imaging.
Bonferroni correction applied, P < .0083 considered significant and represented in bold.
Surgical treatment of pediatric tibial spine fractures
Among the 3,554 patients with tibial spine fractures, 806 (22.7%) underwent operative treatment, while 2,748 (76.9%) received nonoperative treatment. Surgery was performed in the first week following injury for 51.5% and within 4 weeks following injury for 91.6% (Fig. 3). Of the patients undergoing the initial surgery 29-90 days after fracture diagnosis, 50 of 68 (73.5%) had ACLRs. There was no significant change in the rate of operative treatment between 2016 and 2022 (R2 = 0.4354, P = .1539).
Figure 3.
Time fromdiagnosis tosurgery inpediatrictibialspinefracture. Elapsed time from diagnosis to surgery in pediatric patients with operatively treated tibial spine fractures. Total number of patients per interval <11 not shown to protect patient privacy. ARIF, arthroscopic reduction and internal fixation; ORIF, open reduction and internal fixation; ACLR, anterior cruciate ligament reconstruction.
Factors significantly associated with surgical management by univariable analyses were age (P = .0045), geographic region (P = .0027), fracture displacement (P < .0001), and MRI utilization (P < .0001). Independent predictors of surgery by multivariable regression analysis were fracture displacement and MRI use (OR = 14.13 and OR = 7.44, respectively, P < .0001 for both) (Fig. 4). There were no significant differences in rates of surgical management by the multivariable analysis with regards to sex, ECI, or insurance type. Predictors of surgery are shown in Table 2.
Figure 4.
MRIutilization inpediatrictibialspinefracturemanagement. MRI utilization in operatively and nonoperatively treated pediatric tibial spine fracture from 2016 to 2022. MRI, magnetic resonance imaging.
Table 2.
Univariable and multivariable analyses of surgical treatment relative to nonoperative treatment in pediatric tibial spine fracture patients.
| Univariable | Multivariable | ||||
|---|---|---|---|---|---|
| Nonoperative | Operative | P value | Odds ratio (95% CI) | P value | |
| Total | 2,748 (77.3%) | 806 (22.7%) | |||
| Age (mean ± SD) | 12.57 ± 2.93 | 12.81 ± 2.64 | P = .0045 | 1.00 (0.97,1.04) | P = .8471 |
| Sex | P = .1150 | ||||
| Female | 884 (33.2%) | 235 (29.2%) | 0.99 (0.81,1.21) | P = .9437 | |
| Male | 1,864 (67.8%) | 571 (70.8%) | Reference | – | |
| ECI (mean ± SD) | 0.92 ± 1.12 | 0.90 ± 1.08 | P = .5851 | 0.97 (0.89,1.05) | P = .4454 |
| Insurance | P = .5181 | ||||
| Commercial | 2,134 (77.7%) | 626 (77.7%) | Reference | – | |
| Medicaid | 545 (19.8%) | 159 (19.7%) | 1.02 (0.81,1.29) | P = .8350 | |
| Region | P = .0027 | ||||
| South | 1,164 (42.4%) | 303 (37.6%) | Reference | – | |
| Midwest | 571 (20.8%) | 217 (26.9%) | 1.29 (1.03,1.63) | P = .0286 | |
| Northeast | 505 (18.4%) | 148 (18.4%) | 1.04 (0.81,1.35) | P = .7372 | |
| West | 475 (17.2%) | 133 (16.5%) | 1.02 (0.78,1.31) | P = .9037 | |
| Fracture type | P < .0001 | ||||
| Nondisplaced | 756 (27.5%) | 20 (2.5%) | Reference | – | |
| Displaced | 1,992 (72.5%) | 786 (97.5%) | 14.13 (9.14,23.15) | P < .0001 | |
| MRI utilization | P < .0001 | ||||
| No MRI | 2,455 (89.3%) | 418 (51.9%) | Reference | ||
| MRI | 293 (10.7%) | 388 (48.1%) | 7.44 (6.11,9.08) | P < .0001 | |
CI, confidence interval; SD, standard deviation; ECI, Elixhauser Comorbidity Index; MRI, magnetic resonance imaging.
Bonferroni correction applied, P < .0071 considered significant and represented in bold.
MRI utilization rates among those managed nonoperatively and those treated surgically are additionally displayed in Table 2. Of nonoperatively managed patients, 2,455 patients (89.3%) did not receive MRI, while 388 operatively managed patients (48.1%) received MRI. Among 388 operatively managed patients who received MRI, surgery was performed within 7 days after MRI for 186 patients (47.9%). Ninety-nine patients received surgery 7-14 days after MRI (25.5%). Hence, the majority of patients (73.4%) managed operatively and who received MRI underwent surgery within 2 weeks after imaging.
Open vs arthroscopic surgery
Of 806 patients treated surgically, ARIF was pursued for 522 patients (64.8%), while an open procedure was done for 158 patients (19.6%). The proportion of operative patients treated with ARIF compared to ORIF did not increase or change over the study period from 2016 to 2022 (R2 = 0.3237 P = .2387).
Among patients treated with ARIF or ORIF, predictors of arthroscopic approach are shown in Table 3. The remaining 126 surgical patients were treated with isolated ACLR and were not included in this analysis. The only factor significantly associated with arthroscopic approach by the univariable analysis was MRI utilization (P < .0001). The acquisition of a preoperative MRI was also a significant predictor of ARIF relative to ORIF by the multivariable analysis (OR = 3.00, P < .0001). There were no significant differences in ARIF utilization with regards to sex, ECI, or insurance type.
Table 3.
Univariable and multivariable analyses of surgical technique (ARIF vs ORIF) in surgically treated patients.
| Univariable | Multivariable | ||||
|---|---|---|---|---|---|
| ORIF | ARIF | P value | Odds ratio (95% CI) | P value | |
| Total | 158 (23.2%) | 522 (64.8%) | |||
| Age (mean ± SD) | 12.75 ± 2.56 | 12.51 ± 2.62 | P = .3228 | 0.985 (0.88,1.03) | P = .2197 |
| Sex | P = .5655 | ||||
| Female | 44 (27.8%) | 160 (30.7%) | 1.17 (0.77,1.81) | P = .4650 | |
| Male | 114 (72.2%) | 362 (69.3%) | Reference | – | |
| ECI (mean ± SD) | 0.92 ± 1.18 | 0.90 ± 1.06 | P = .8541 | 0.95 (0.80,1.12) | P = .5197 |
| Insurance | P = .6023 | ||||
| Commercial | 128 (81.0%) | 406 (77.8%) | Reference | – | |
| Medicaid | 27 (17.1%) | 103 (19.7%) | 1.26 (0.78,2.09) | P = .3484 | |
| Region | P = .0972 | ||||
| Midwest | 54 (34.2%) | 126 (24.1%) | Reference | – | |
| Northeast | 21 (13.3%) | 103 (19.7%) | 1.74 (0.97,3.20 | P = .0674 | |
| South | 55 (34.8%) | 201 (38.5%) | 1.45 (0.92,2.29) | P = .1128 | |
| West | 27 (17.1%) | 88 (16.9%) | 1.41 (0.81,2.48) | P = .2322 | |
| MRI | P < .0001 | ||||
| No MRI | 117 (74.1%) | 263 (50.4%) | Reference | – | |
| MRI | 41 (25.9%) | 259 (49.6%) | 3.00 (2.01, 4.56) | P < .0001 | |
CI, confidence interval; SD, standard deviation; ECI, Elixhauser Comorbidity Index; MRI, magnetic resonance imaging; ARIF, arthroscopic reduction and internal fixation; ORIF, open reduction and internal fixation.
Bonferroni correction applied, P < .0083 considered significant and represented in bold.
Discussion
In this study, we utilized a large data set of pediatric tibial spine fractures and highlight recent (within the past decade) nationwide trends in MRI utilization as well as treatment strategies for this patient population. MRI was utilized in the workup of these injuries 19.1% of the time, a rate that was found to increase over the course of the study period. While this change is in line with recent literature proposing an MRI-based classification system [13] and expert consensus recommending advanced imaging for pediatric tibial spine fractures so as not to miss associated injuries [11], it was still the minority of patient who had such imaging. Though some have argued MRI is unnecessary when other structures of the knee will inevitably be assessed during surgical treatment [9,16], the majority of patients not receiving MRI also did not have surgery. Low rates of imaging may then be driven to a greater extent by outdated perceptions of tibial spine fractures as isolated injuries, hesitancy to pursue advanced imaging in young children, or concern that all relevant injuries are not well visualized on MRI [6,15].
Furthermore, MRI is utilized both to evaluate for associated injuries and also to better assess fracture morphology to make treatment decisions. While computed tomography may characterize the fracture, it does not provide the granular detail of chondral and meniscal pathology to the same extent as MRI [[6], [7], [8], [9]]. The greatest independent predictor of MRI utilization in our population was fracture displacement. Additional factors independently associated with MRI utilization were older age and geographic region. Older age being an independent predictor could potentially be explained by older patients being more likely to sustain sports-related injuries that may also involve ligament, meniscal, or cartilage damage [12]. As such, providers would be inclined to obtain MRI to guide treatment in these athletes. Additionally, physicians may be less willing to obtain MRI in younger children who could require sedation. Geographic variation suggests nonclinical factors weigh in on such decision-making and that there is lack of consensus on approaches. These differences may further reflect disparities in access to imaging resources, regional referral patterns, or institutional protocols, all contributing to regional-level differences in care.
Surgical intervention was performed in 22.7% of the identified injuries, with most of these surgeries being done early in the course of management. Approximately 92% of operatively treated patients underwent surgery within 28 days of diagnosis. Moreover, 73.4% of operatively managed patients who received an MRI underwent surgery within 2 weeks after imaging. There was no change in the rate of surgical management over the years of the study by linear regression, despite studies in this period suggesting that surgery may be warranted in patients with minimally displaced fractures [28].
Fracture displacement was the greatest predictor of surgical intervention for these pediatric tibial spine fractures. Current guidelines recommend surgery for fractures with greater than minimal displacement [20,22] as displacement is associated with nonunion in fractures managed nonoperatively [35]. Additionally, nonoperatively managed tibial spine fractures with displacement have poor ACL tension and are associated with higher risk of ACL tear or injury [28]. Notably, age was not independently associated with surgical management, despite survey data of pediatric orthopaedists reporting hesitancy to operate on minimally displaced fractures in younger patients [11]. Beyond displacement, the only independent predictor of surgery was MRI utilization. This is also consistent with expectations and may represent a selection bias for patients with severe fractures on plain radiographs or high clinical suspicion for associated injuries, given the utility of MRI for surgical planning and identifying other injuries that may need to be addressed in the operating room. Additionally, obtaining advanced imaging adds opportunity to diagnose potentially missed ligamentous injuries that could encourage surgical treatment, irrespective of the tibial spine fracture.
Of the 806 patients undergoing surgery, 522 had ARIF, 158 had ORIF, and 126 had ACLR. The prevalence of ACLR suggests that associated intrasubstance tears may be detected [6], leading to the decision to pursue ACLR, or that this population may represent patients receiving surgery for persistent ACL insufficiency rather than initial fixation [38]. ARIF was performed for the majority of cases, but open surgery was done for 23.2%, a rate that did not significantly change over the course of the study. This was despite recent expert consensus recommending ARIF as the standard surgical approach to pediatric tibial spine fractures [11]. The high proportion of arthroscopic procedures suggests that many of these surgeries may be performed by surgeons with additional sports medicine training, which was also proposed by Shimberg et al. [34]. However, arthroscopy is not universally used and regional differences exist within the USA. Rates of ARIF were significantly different by geographic region by the univariable analysis, though not by the multivariable analysis, suggesting that there are different rates of surgeon comfort, preference, or training but that this may be better explained by variables other than region.
Notably, none of the decision points assessed (MRI utilization, treatment strategy, or surgical approach) were independently associated with patient insurance. Consistent with prior literature, we found no significant differences in MRI utilization rates or surgical technique on the basis of insurance status [17]. Additionally, there were no differences on the basis of sex by the multivariable analysis. However, geography did impact the likelihood of MRI utilization, suggesting nonclinical factors affect treatment.
As with all studies utilizing insurance databases, this work is limited by the accuracy and specificity of provider coding in the PearlDiver database. Due to lack of granularity in the coding, we are unable to separate fractures by Meyers and McKeever classification (ie, degree of fracture displacement). Our temporal trends were also based on 6 years of data, which limits the power to detect longer-term temporal trends. Additionally, the CPT codes used for MRI were not entirely specific to the knee, but by excluding patients with distal fractures and investigating MRI utilization within 90 days of fracture diagnosis, we aimed to reduce capture of other lower extremity MRIs. Future studies may consider determining MRI utilization rates for concomitant injuries with pediatric tibial spine fractures and how this may affect operative management, as well as stratifying MRI utilization rates within each region by institution type.
Conclusion
In this large national database study, we found MRI utilization rates to be lower than expected given recent literature promoting its utility in pediatric tibial spine fractures. Meanwhile, surgery was performed for 22.7% of the population, with fracture displacement being a strong predictor (OR = 14.13). Finally, most fractures treated surgically were performed arthroscopically (76.8%). Geographic variations in MRI suggest that nonclinical factors impact management, warranting further investigation.
Additional links
Funding
Katie Zehner received funding for this project from the Arons-Millard Student Research Fund and James G. Hirsch Endowed Medical Student Research Fund at the Yale School of Medicine. Meera Dhodapkar received funding from the Richard K. Gershon Student Research Fund at Yale School of Medicine.
Author contributions
Katie M. Zehner: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Conceptualization. Meera M. Dhodapkar: Writing – review & editing, Methodology, Investigation, Formal analysis. Ashraf N. Nawari: Writing – review & editing, Data curation. Estevao D. Santos: Writing – review & editing, Validation, Conceptualization. Elizabeth C. Gardner: Writing – review & editing, Supervision, Project administration, Conceptualization. Jonathan N. Grauer: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.
Ethics approval and consent
The authors declare that no patient consent was necessary as no images or identifying information are included in the article.
Declaration of competing interests
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: KZ reports financial support was provided by James G Hirsch Endowed Medical Student Research Fund at Yale School of Medicine and by Arons-Millard Student Research Fund. M.D. reports financial support was provided by Richard K Gershon Student Research Fund at Yale School of Medicine. E.S. reports a relationship with Gotham Surgical. K.Z. is the Associate Editor of Visual Abstracts, North American Spine Society Journal. M.D. is a former Associate Editor of Visual Abstracts, North American Spine Society Journal. E.G. is a member of Team Physician Committee of the American Orthopaedic Society for Sports Medicine and a member of National Collegiate Athletic Association Committee on Competitive Safeguards and Medical Aspects of Sports. J.G. is the Editor-in-Chief of North American Spine Society Journal and Deputy Editor of Journal of the American Academy of Orthopaedic Surgery. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.DeFrancesco C.J., Tananbaum A., LeBrun D.G., Fabricant P.D. The incidence of pediatric tibial spine fractures is greater and peaks later in male patients. Arthrosc Sports Med Rehabil. 2022;4:e639–e643. doi: 10.1016/j.asmr.2021.12.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.DeFrancesco C.J., Wilson L., Lebrun D.G., Memtsoudis S.G., Fabricant P.D. Pediatric tibial spine fractures: exploring case burden by age and sex. Orthop J Sports Med. 2021;9 doi: 10.1177/23259671211027237. 23259671211027237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Meyers M.H., McKeever F.M. Fracture of the intercondylar eminence of the tibia. JBJS. 1959;41:209. [PubMed] [Google Scholar]
- 4.Wiley J.J., Baxter A.P. Tibial spine fractures in children. Clin Orthop Relat Res. 1990;255:54. [PubMed] [Google Scholar]
- 5.Kocher M.S., Micheli L.J., Gerbino P., Hresko M.T. Tibial eminence fractures in children: prevalence of meniscal entrapment. Am J Sports Med. 2003;31:404–407. doi: 10.1177/03635465030310031301. [DOI] [PubMed] [Google Scholar]
- 6.Mayo M.H., Mitchell J.J., Axibal D.P., Chahla J., Palmer C., Vidal A.F., et al. Anterior cruciate ligament injury at the time of anterior tibial spine fracture in young patients: an observational cohort study. J Pediatr Orthop. 2019;39:e668–e673. doi: 10.1097/BPO.0000000000001011. [DOI] [PubMed] [Google Scholar]
- 7.Mitchell J.J., Sjostrom R., Mansour A.A., Irion B., Hotchkiss M., Terhune E.B., et al. Incidence of meniscal injury and chondral pathology in anterior tibial spine fractures of children. J Pediatr Orthop. 2015;35:130. doi: 10.1097/BPO.0000000000000249. [DOI] [PubMed] [Google Scholar]
- 8.Rhodes J.T., Cannamela P.C., Cruz A.I., Mayo M., Styhl A.C., Richmond C.G., et al. Incidence of meniscal entrapment and associated knee injuries in tibial spine avulsions. J Pediatr Orthop. 2018;38:e38–e42. doi: 10.1097/BPO.0000000000001110. [DOI] [PubMed] [Google Scholar]
- 9.Johnson A.C., Wyatt J.D., Treme G., Veitch A.J. Incidence of associated knee injury in pediatric tibial eminence fractures. J Knee Surg. 2014;27:215–219. doi: 10.1055/s-0033-1360656. [DOI] [PubMed] [Google Scholar]
- 10.Shea K.G., Grimm N.L., Laor T., Wall E. Bone bruises and meniscal tears on MRI in skeletally immature children with tibial eminence fractures. J Pediatr Orthop. 2011;31:150–152. doi: 10.1097/BPO.0b013e3182093df1. [DOI] [PubMed] [Google Scholar]
- 11.Brenner M., Johnson B., Shea K., Ganley T., Wilson P.L., Ellis H.B., et al. Perioperative management of tibial spine fractures in pediatric patients: a Delphi study to identify principles of treatment. J Pediatr Orthop. 2025;45:57. doi: 10.1097/BPO.0000000000002844. [DOI] [PubMed] [Google Scholar]
- 12.Feucht M.J., Brucker P.U., Camathias C., Frosch K.-H., Hirschmann M.T., Lorenz S., et al. Meniscal injuries in children and adolescents undergoing surgical treatment for tibial eminence fractures. Knee Surg Sports Traumatol Arthrosc Off J ESSKA. 2017;25:445–453. doi: 10.1007/s00167-016-4184-0. [DOI] [PubMed] [Google Scholar]
- 13.Green D., Tuca M., Luderowski E., Gausden E., Goodbody C., Konin G. A new, MRI-based classification system for tibial spine fractures changes clinical treatment recommendations when compared to Myers and Mckeever. Knee Surg Sports Traumatol Arthrosc. 2019;27:86–92. doi: 10.1007/s00167-018-5039-7. [DOI] [PubMed] [Google Scholar]
- 14.LaFrance R.M., Giordano B., Goldblatt J., Voloshin I., Maloney M. Pediatric tibial eminence fractures: evaluation and management. J Am Acad Orthop Surg. 2010;18:395. doi: 10.5435/00124635-201007000-00002. [DOI] [PubMed] [Google Scholar]
- 15.Shimberg J.L., Aoyama J.T., Leska T.M., Ganley T.J., Fabricant P.D., Patel N.M., et al. Tibial spine fractures: how much are we missing without pretreatment advanced imaging? A multicenter study. Am J Sports Med. 2020;48:3208–3213. doi: 10.1177/0363546520957666. [DOI] [PubMed] [Google Scholar]
- 16.Severyns M., Odri G.A., Vendeuvre T., Marchand J.-B., Germaneau A., Dramé M. Meniscal injuries in skeletally immature children with tibial eminence fractures. Systematic review of literature. Int Orthop. 2023;47:2439–2448. doi: 10.1007/s00264-023-05787-w. [DOI] [PubMed] [Google Scholar]
- 17.Smith H.E., Mistovich R.J., Cruz A.I., Leska T.M., Ganley T.J., Aoyama J.T., et al. Does insurance status affect treatment of children with tibial spine fractures? Am J Sports Med. 2021;49:3842–3849. doi: 10.1177/03635465211046928. [DOI] [PubMed] [Google Scholar]
- 18.Edmonds E.W., Fornari E.D., Dashe J., Roocroft J.H., King M.M., Pennock A.T. Results of displaced pediatric tibial spine fractures: a comparison between open, arthroscopic, and closed management. J Pediatr Orthop. 2015;35:651–656. doi: 10.1097/BPO.0000000000000356. [DOI] [PubMed] [Google Scholar]
- 19.Prasad N., Aoyama J.T., Ganley T.J., Ellis H.B., Mistovich R.J., Yen Y.-M., et al. A comparison of nonoperative and operative treatment of type 2 tibial spine fractures. Orthop J Sports Med. 2021;9 doi: 10.1177/2325967120975410. 2325967120975410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Adams A.J., Talathi N.S., Gandhi J.S., Patel N.M., Ganley T.J. Tibial spine fractures in children: evaluation, management, and future directions. J Knee Surg. 2018;31:374–381. doi: 10.1055/s-0038-1636544. [DOI] [PubMed] [Google Scholar]
- 21.Leie M., Heath E., Shumborski S., Salmon L., Roe J., Pinczewski L. Midterm outcomes of arthroscopic reduction and internal fixation of anterior cruciate ligament tibial eminence avulsion fractures with K-Wire fixation. Arthrosc J Arthrosc Relat Surg Off Publ Arthrosc Assoc N Am Int Arthrosc Assoc. 2019;35:1533–1544. doi: 10.1016/j.arthro.2018.11.066. [DOI] [PubMed] [Google Scholar]
- 22.Tuca M., Bernal N., Luderowski E., Green D.W. Tibial spine avulsion fractures: treatment update. Curr Opin Pediatr. 2019;31:103. doi: 10.1097/MOP.0000000000000719. [DOI] [PubMed] [Google Scholar]
- 23.Aderinto J., Walmsley P., Keating J.F. Fractures of the tibial spine: epidemiology and outcome. Knee. 2008;15:164–167. doi: 10.1016/j.knee.2008.01.006. [DOI] [PubMed] [Google Scholar]
- 24.Horibe S., Shi K., Mitsuoka T., Hamada M., Matsumoto N., Toritsuka Y. Nonunited avulsion fractures of the intercondylar eminence of the tibia. Arthrosc J Arthrosc Relat Surg. 2000;16:757–762. doi: 10.1053/jars.2000.4823. [DOI] [PubMed] [Google Scholar]
- 25.Vargas B., Lutz N., Dutoit M., Zambelli P.Y. Nonunion after fracture of the anterior tibial spine: case report and review of the literature. J Pediatr Orthop B. 2009;18:90. doi: 10.1097/BPB.0b013e328329895b. [DOI] [PubMed] [Google Scholar]
- 26.Panni A., Milano G., Tartarone M., Fabbriciani C. Arthroscopic treatment of malunited and nonunited avulsion fractures of the anterior tibial spine. Arthrosc J Arthrosc Relat Surg. 1998;14:233–240. doi: 10.1016/s0749-8063(98)70137-9. [DOI] [PubMed] [Google Scholar]
- 27.Quinlan N.J., Hobson T.E., Mortensen A.J., Tomasevich K.M., Adeyemi T., Maak T.G., et al. Tibial spine repair in the pediatric population: outcomes and subsequent injury rates. Arthrosc Sports Med Rehabil. 2021;3:e1011–e1023. doi: 10.1016/j.asmr.2021.03.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Shimberg J.L., Leska T.M., Cruz A.I., Ellis H.B., Patel N.M., Yen Y.-M., et al. Is nonoperative treatment appropriate for all patients with type 1 tibial spine fractures? A multicenter study of the Tibial Spine Research Interest Group. Orthop J Sports Med. 2022;10 doi: 10.1177/23259671221099572. 23259671221099572. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Jääskelä M., Turati M., Lempainen L., Bremond N., Courvoisier A., Henri A., et al. Long-term outcomes of tibial spine avulsion fractures after open reduction with osteosuturing versus arthroscopic screw fixation: a multicenter comparative study. Orthop J Sports Med. 2023;11 doi: 10.1177/23259671231176991. 23259671231176991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Limone B., Zambianchi F., Cacciola G., Seracchioli S., Catani F., Tarallo L. Management and outcomes of tibial eminence fractures in the pediatric population: a systematic review. Child Basel Switz. 2023;10:1379. doi: 10.3390/children10081379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Pailhé R., Bauer T., Flecher X., Bonnevialle N., Roussignol X., Saragaglia D., et al. SOFCOT: better functional outcomes for ORIF in tibial eminence fracture treatment: a national comparative multicentric study of ORIF vs ARIF. Knee Surg Sports Traumatol Arthrosc Off J ESSKA. 2020;28:653–657. doi: 10.1007/s00167-019-05785-9. [DOI] [PubMed] [Google Scholar]
- 32.Watts C.D., Larson A.N., Milbrandt T.A. Open versus arthroscopic reduction for tibial eminence fracture fixation in children. J Pediatr Orthop. 2016;36:437–439. doi: 10.1097/BPO.0000000000000476. [DOI] [PubMed] [Google Scholar]
- 33.Chandanani M., Jaibaji R., Jaibaji M., Volpin A. Tibial spine avulsion fractures in paediatric patients: a systematic review and meta-analysis of surgical management. Child Basel Switz. 2024;11:345. doi: 10.3390/children11030345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Shimberg J.L., Leska T.M., Cruz A.I.J., Patel N.M., Ellis H.B.J., Ganley T.J., et al. A multicenter comparison of open versus arthroscopic fixation for pediatric tibial spine fractures. J Pediatr Orthop. 2022;42:195. doi: 10.1097/BPO.0000000000002049. [DOI] [PubMed] [Google Scholar]
- 35.Tibial Spine Research Interest Group, Orellana K.J., Houlihan N.V., Carter M.V., Baghdadi S., Baldwin K., et al. Tibial spine fractures in the child and adolescent athlete: a systematic review and meta-analysis. Am J Sports Med. 2024;52:1357–1366. doi: 10.1177/03635465231175674. [DOI] [PubMed] [Google Scholar]
- 36.Gans I., Baldwin K.D., Ganley T.J. Treatment and management outcomes of tibial eminence fractures in pediatric patients: a systematic review. Am J Sports Med. 2014;42:1743–1750. doi: 10.1177/0363546513508538. [DOI] [PubMed] [Google Scholar]
- 37.Menendez M.E., Neuhaus V., van Dijk C.N., Ring D. The elixhauser comorbidity method outperforms the Charlson index in predicting inpatient death after orthopaedic surgery. Clin Orthop. 2014;472:2878–2886. doi: 10.1007/s11999-014-3686-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Mitchell J.J., Mayo M.H., Axibal D.P., Kasch A.R., Fader R.R., Chadayammuri V., et al. Delayed anterior cruciate ligament reconstruction in young patients with previous anterior tibial spine fractures. Am J Sports Med. 2016;44:2047–2056. doi: 10.1177/0363546516644597. [DOI] [PubMed] [Google Scholar]