A lesion of the lumbar posterior vertebral end plate, termed an apophyseal ring fracture, in children and adolescents causes symptoms similar to those of a herniated lumbar intervertebral disc. The end-plate lesion has been described as an apophysis, an ossified vertebral rim, an avulsion of the rim, or a fracture of the vertebral body1-6. In one study, it was reported that adult patients with disc herniation had an associated posterior end-plate lesion5. However, the pathology of end-plate lesions and the pathology of herniated discs differ7. Apophyseal ring fractures likely result from mechanical stress applied to the apophyseal ring, which can lead to this fracture of the vertebral growth plate8,9. This fracture is more common in children and adolescents who are involved in sports.
The purpose of this study was to clarify the long-term clinical and radiographic outcomes in children and adolescents who had been treated either surgically or conservatively for a lumbar posterior apophyseal end-plate lesion.
From our medical records, we identified and contacted twenty-seven consecutive patients who had been treated in the sports clinic of our hospital from April 1989 to April 1995. Letters were sent to these twenty-seven patients, but only twenty-four responded.
The mean age at the first medical examination was 14.5 years (range, nine to eighteen years). The mean follow-up time was 13.8 years (range, six to nineteen years). The mean age at the final follow-up evaluation was 28.4 years (range, twenty-two to thirty-two years). All twenty-four patients had symptomatic low back pain with sciatica at the initial examination. All but two participated in sports (Table I). Sixteen patients (twelve male and four female) were treated conservatively, and eight patients (six male and two female) were treated surgically. Indications for surgical treatment were (1) intolerable pain or a progressive neurological deficit after conservative treatment for six months, (2) severe neurological motor weakness in the lower extremities, and/or (3) urinary symptoms.
Radiographic Assessment
Lesions of the posterior end plate were diagnosed from lateral radiographs. Skeletal maturity was evaluated on the basis of the appearance of the secondary ossification center of the L3 vertebra. The maturity stage of the vertebral body was classified as (1) the cartilaginous (“C”) stage, indicating that the secondary ossification center of the vertebral body was not visible on the radiograph; (2) the apophyseal (“A”) stage, indicating that the secondary ossification center of the vertebral body was visible on the radiograph and an ossified fragment was confirmed at the corner of the vertebral body; or (3) the epiphyseal (“E”) stage, meaning that the apophyseal ring was fused to the vertebral body, suggesting that the vertebra had reached maturation9-11. The staging was performed by two orthopaedic surgeons (K.H. and K.S.) using radiographs. If the two orthopaedic surgeons differed in their assessment, a discussion ensued, and a final score was decided. Representative radiographs and computed tomography (CT) scans of the posterior end-plate lesion in different stages are shown in Figure 1.
Oblique radiographs in neutral position and CT scans were obtained to evaluate for spondylolysis and spondylolisthesis. Instability of each segment was evaluated on lateral flexion and extension radiographs. Lumbar spine instability was defined according to two criteria: (1) translational vertebral motion of >5 mm and (2) a cutoff value of −5° for the end-plate angle (the angle between one line drawn from the inferior margin of the superior vertebral body and another line drawn from the superior margin of the inferior vertebral body)12-14.
The posterior end-plate lesion (ossified fragment) was evaluated on CT. The lesions were classified into two types: large fragment and small fragment. Eighteen patients had a large-fragment type, with a large osseous fragment positioned centrally within the spinal canal, and the remaining six patients had a small-fragment type, with small osseous fragments positioned laterally within the spinal canal.
Magnetic Resonance Imaging (MRI) Equipment and Protocol
MRI was used to identify disc degeneration according to the criteria described by Schneiderman et al.15 and to assess degenerative changes at the adjacent end plate according to the criteria described by Modic et al.16.
For the Schneiderman criteria15, the midline sagittal section on the T2-weighted image was used. Sagittal intensity was assigned one of four grades: grade 1 indicated a normal disc height and signal intensity; grade 2 (intermediate), a speckled pattern or heterogeneously decreased signal intensity; grade 3, a marked diffuse loss of signal; and grade 4, signal void.
For the Modic criteria, signal intensity changes in the vertebral body bone marrow adjacent to the end plates were identified. Type 1 showed decreased signal intensity on T1-weighted images and increased signal intensity on T2-weighted images. Type 1 showed disruption and fissuring of the end plate with regions of degeneration and regeneration and vascular granulation tissues. Type 2 showed increased signal intensity on T1 and T2-weighted images. Type 2 showed chronic repetitive trauma. No signal intensity changes and only small signal changes were classified as normal.
Sagittal MRIs were used to evaluate for spinal stenosis due to the lesion at the end-plate level. All of the MRIs were analyzed independently by one orthopaedic surgeon (K.H.) and one radiologist (S.T.). MRIs were obtained with use of a 1.5-T instrument (Signa, GE Healthcare, USA) with a surface coil. T1, T2, and STIR-weighted sagittal, coronal, and axial images were used for the spinal studies. The repetition and echo times for the sagittal, coronal, and axial images were TR450/TE8 ms (T1-weighted), TR3500/TE102 ms (T2-weighted), and TR3000/TE21.5/TI105 (STIR). The slice thickness and interslice gap were 4 mm and 1 mm for sagittal and axial images. Other MRIs were acquired with use of a 1.0-T instrument (MRT-100A; Toshiba, Japan) with a surface coil. Repetition and echo times for sagittal and axial images were TR500/TE30 ms (T1-weighted) and TR1800/TE120 ms (T2-weighted). The slice thickness and interslice gap were 10 mm and 1 mm for both the sagittal and the axial images.
Clinical Assessment at Time of Final Follow-up
To assess low back pain at the time of final follow-up, we used the Roland-Morris Disability Questionnaire (RDQ)17,18. The RDQ is a validated self-administered back-specific questionnaire. It contains twenty-four yes/no questions describing difficulties in accomplishing activities because of low back or leg pain on the day of questionnaire completion. The total score ranges from 0 (no disability) to 24 (severe disability). Each patient filled out a RDQ at the time of latest follow-up. The neurological outcome was assessed with a straight-leg-raising test as well as sensory and motor evaluation of the lower extremities at the time of latest follow-up.
Histological Examination
Histological specimens from six patients who received surgical treatment were evaluated.
Statistical Analysis
The RDQ scores were compared between the patients who had received surgical treatment and those who had received conservative treatment with use of the Mann-Whitney U test. The analyses were performed with use of StatView, version 5.0 (Abacus Concepts, Berkeley, California), with p < 0.05 considered significant.
Source of Funding
We did not receive any outside funding or grants in support of our research for this study.
At the time of final follow-up, the average RDQ for the sixteen patients treated conservatively was 1.3 (range, 0 to 8), and the average RDQ for the eight patients treated surgically was 1.8 (range, 0 to 7). These two averages did not differ significantly (Table I).
Twelve patients (six in the surgical group and six in the conservative group) had a positive straight-leg-raising test at <60° at the initial examination. The average angle on the affected side was 33.8° (range, 20° to 70°) for the patients treated conservatively and 64.0° (20° to 90°) for the patients treated surgically; this was a significant difference (p = 0.02). In all patients, the initial neurological findings had resolved and lower-extremity sensory and motor findings were normal at the final follow-up evaluation. Of the twenty-two patients initially active in sports, eighteen returned to their active sports participation.
A lesion of the inferior apophyseal rim was more prevalent than a lesion of the superior rim. A lesion at the inferior rim of L4 was seen in nine patients (two treated surgically and seven treated conservatively), a lesion of the inferior rim of L5 was seen in nine patients (five treated surgically and four treated conservatively), a lesion of the superior rim of L5 was seen in four patients (one treated surgically and three treated conservatively), and a lesion of the superior rim of S1 was seen in two patients (both treated conservatively) (Table I). Skeletal maturity at the initial presentation was “C” stage in three patients (mean age, 11.0 years), “A” stage in fourteen (mean age, 14.3 years), and “E” stage in seven (mean age, 17.0 years). The apophyseal stage (Fig. 2) was seen most frequently (in fourteen patients).
According to the CT classification, eighteen patients had a large fragment and seven of these eighteen patients underwent surgical intervention. Six patients had small fragments, and one of these patients underwent surgical intervention. The other patients were treated conservatively. The bone fragments did not naturally resorb if the patients were treated conservatively.
Disc degeneration was assessed with the MRI grading system. Both the surgically treated group and the conservatively treated group showed progressive disc degeneration. In the surgically treated group, the mean difference in MRI scores (the score at the latest examination minus the score at the first examination) was an increase of 1.6 points. In the conservatively treated group, the mean difference in MRI scores was an increase of 1.1 points. According to the Modic classification, signal abnormality was observed at the first examination in five patients treated conservatively. Both the surgically treated group and the conservatively treated group showed progressive bone marrow changes except for one patient who had been treated surgically. One patient, who had been treated surgically, had increased lumbar spinal stenosis (Table II). One patient developed spinal stenosis after twelve years of conservative treatment (see Case 2 below). There were no significant associations between the MRI findings and RDQ scores.
Case Reports
Case 1. A thirteen-year-old boy who played table tennis underwent removal of the apophyseal end-plate lesion at L4. Postoperative lateral radiographs obtained in the neutral position six months and thirteen years after surgery showed progressive instability. MRIs (T2-weighted) made thirteen years after surgery showed Grade-3 disc degeneration, Type-1 degeneration of bone marrow, and severe spinal canal stenosis with dural sac compression at the inferior L4 level. The RDQ score was 7 at the latest examination (Fig. 3).
Case. 2 A fourteen-year-old boy who played baseball was treated conservatively. The radiographs at initial examination showed an end-plate lesion at the inferior rim of L4. He was able to return to playing baseball after conservative treatment. He needed surgical intervention when he was twenty-six years old (after twelve years of conservative treatment) because of severe pain and a neurological deficit as determined by progressive muscle weakness and hypesthesia. The RDQ score was 8. At the time of follow-up, when he was twenty-six years old, CT showed osseous fragments at the inferior and superior L4 rims. MRIs showed severe degeneration of the inferior and superior rims of the end plate of L4 (Fig. 4). He underwent decompression and removal of the osseous fragments at the L4 rims. The RDQ score was 0 after surgery. Histological examination of the surgical specimen showed an abnormal end plate with degeneration and some chondrocytes without a nucleus. The adjacent osseous fragment of the cartilage at the posterior corner showed irregular woven bone without lamellar structure and loose fibrous tissue (Fig. 5). The end-plate lesion seemed not able to remodel and become resorbed.
The vertebral apophyseal ring is a unique structure seen only in the pediatric immature spine (the “A” stage). It has been reported that stresses during lumbar motion concentrate in the apophyseal ring19. Our results suggest that when a patient is in the “A” stage, there is more risk of an apophyseal ring fracture.
The present study shows that the long-term outcome is good after either conservative or surgical treatment of an apophyseal ring fracture. The majority of patients had no permanent neurological deficit and returned to sports activity and ordinary activities of daily living. However, degenerative changes in the disc tissue and the bone marrow adjacent to the cartilaginous end plate were observed on MRIs in twelve cases at the time of final follow-up. It seems that when the end plate is injured, it can cause disc degeneration and an entire end-plate lesion (Modic change). Histological examination of specimens from one patient who needed surgery after twelve years of conservative treatment indicated that the initial severe degeneration of the end-plate lesion was not restored to normal enchondral ossification. Several in vivo studies have suggested that most of the blood supply and nutrient supply to the apophysis come from the vertebral body20-23. The posterior end-plate lesion is not absorbed24, whereas herniated disc material in the lumbar spine usually is absorbed25-27. The apophyseal lesion is often detected in association with herniation of disc tissue15,16. In our previous study, we described the end-plate lesion as one involving the cartilage or osseous tissue of the growth plate and the tissue of the anulus fibrosus and nucleus pulposus7. Disc herniation in adults involves the nucleus pulposus and/or the fibers of the anulus fibrosus, whereas an end-plate lesion involves the growth plate tissue and disc tissue. Histological examination of the material removed at surgery showed irregular alignment of the anulus, plus degenerative matrix and chondrocytes without a nucleus. These histological findings of the end-plate lesion showed that an avulsion segment of the end-plate lesion was not able to be remodeled. Once the end plate has fractured, the lesion in these cartilaginous growth sites causes abnormal chondrocyte proliferation and maturation. We suggest that failure of the posterior end plate may produce abnormal mechanical stress and alter the local biology and that, over time, these could produce degenerative changes in the end plate and disc28,29.
During lumbar posterior surgery, the posterior elements and the lumbar muscles can be injured. It is important to avoid surgery-related back muscle injury, as well as interspinous ligament disruption, and to preserve the facet joints. One patient who received conventional open (rather than less invasive) surgery showed severe lumbar instability at the time of final follow-up.
We concluded that bone fragments do not naturally resorb if the patient is treated conservatively and that bone fragments can cause spinal canal stenosis, either early or late. However, most of our patients were asymptomatic for long periods of time. The surgical outcomes were also acceptable, although late lumbar spine instability and severe disc degeneration were observed in two patients.
In summary, this study shows that the long-term outcome for patients with a posterior end-plate lesion is favorable.
Note: The authors thank Dr. William C. Hutton (Emory University School of Medicine, Atlanta, Georgia) for help in preparing the manuscript.
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.