Introduction

Most burst fractures of the spine are associated with retropulsion of bony fragments into the spinal canal. The clinical significance of bony encroachment of the canal with reference to neurological outcome is not clear.

Consequently, the role of surgery in the treatment of these fractures is controversial. Some surgeons advocate surgical treatment with a view of restoring the spinal canal and stabilizing the spine,1,2 while others recommend non-operative treatment.3,4,5 Some reports suggest that although surgical removal of bony fragments may restore the spinal canal, it does not improve the chance of neurological recovery.6,7,8

This study was undertaken to determine whether neurological damage and subsequent neurological recovery following burst fractures of the thoracolumbar spine depend on the extent of canal compromise.

Methods

Two hundred and forty-eight cases of spinal injuries at the thoracolumbar and lumbar levels were treated at this centre between January 1996 and January 1999. Of these, 61 cases had burst fractures. Patients with multiple level fractures, pathological fractures and those treated surgically were excluded from the study. The remaining 45 patients formed the basis of this study.

There were thirty-four patients with fractures in the thoracolumbar region and eleven patients with fractures in the lumbar region. The male to female ratio was 10 : 1. The mean age at the time of injury was 35.6 years (range 17–60 years).

The most common mechanism of injury was falling from a height (96%) and the most common vertebra involved was T12 vertebra. Fourteen patients presented within 6 h following injury; 26 patients were admitted within 24 h whereas the rest were seen between 24–40 h after the injury (mean 11.6 h).

Neurological evaluation

After admission neurological examination was carried out every 2 h for 24 h in all patients. The neurological deficit noted after the return of the bulbocavernous reflex (end of the spinal shock) was considered as the initial neurological status. The neurological status was classified according to American Spinal Injury Association's modified Frankel's grading of traumatic paraplegia.9

Radiological evaluation

Antero-posterior and lateral roentgenograms and computerized tomographic scans were performed in all patients.

The least mid sagittal diameter of the spinal canal at the level of injury was measured. The normal mid sagittal diameter of the spinal canal was estimated by calculating the average of the corresponding measurements at the adjacent uninjured level above and below the injury. The percentage of the spinal canal compromise at presentation (a) was calculated using the formula of Hashimoto et al2 as shown below:

where a=percentage of canal compromise,

x=mid -sagittal diameter of spinal canal at the level of injury

y=average mid- sagittal diameter of the spinal canal.

Treatment

All 45 patients were treated non-operatively in the form of postural reduction, immobilization in polyethylene moulded body jacket and recumbency for 3 months. Associated injuries and complications were treated accordingly.

Follow-up

The neurological status was reassessed at regular intervals of 3 months. At the end of the first 3 months, the patients were made to sit and by the end of six months following injury they were mobilized. The final neurological evaluation was performed at the time of repeat CT scan.

CT scans were repeated in all cases after a minimum of 1 year follow up. The maximum follow up period was 2 years. The percentage of spinal canal narrowing at follow-up (b) was calculated with the above formula. The percentage of remodeling that had occurred was calculated using the formula:

where a=the spinal canal compromise at the time of injury

b=the spinal canal compromise at follow up

Analysis of variance was used to compute statistical differences at the 0.05 level of significance.

Results

Neurological status

The neurological status of the patients at admission and at the follow up is depicted in the Table 1. Twenty-two patients (58%) had incomplete neurological deficit, 11 (24%) had complete neurological deficit and the remaining patients had no neurological deficit. Among patients with neurological deficit, in 11 patients no appreciable improvement in neurological status was noted. Among the patients who did improve neurologically, the median improvement was one grade in the Frankel's scale.

Table 1 Neurological status at admission and follow up
Full size table

Neurological impairment and canal compromise

Table 2 shows the association between the neurological deficit and the extent of canal compromise at the time of admission. The mean spinal canal compromise in patients with neurologic deficit was 46.2% while in patients with no neurological deficit it was 36.3%. There was no statistically significant difference between the severity of neurological deficit and the extent of canal compromise.

Table 2 The relationship between the neurological impairment at the time of admission and the extent of canal compromise
Full size table

Canal compromise and neurological recovery

The mean percentage of spinal canal compromise in patients with neurological recovery was 46.1±20.9 and 48.4±21.1 in those with no recovery. This difference was not statistically significant.

Neurological recovery and canal remodeling

The mean percentage of the degree of canal remodeling in patients who recovered was 51.7±13.7 and 46.1±11.0 in patients who did not recover. This difference again was not statistically significant.

Discussion

Neurologic injury has been reported to occur in 30%–60% of the patients with thoracolumbar burst fractures.10 Various authors have reported a relationship between the degree of canal compromise and the extent of neurologic deficit.1,2,6,10,11,12 Others have noted that there is no correlation between the initial neurological impairment and the degree of spinal canal narrowing.8,13,14,15

In the present study, there was no statistically significant correlation between the degree of canal encroachment and initial neurologic deficit. A patient with 75% canal compromise at T12 level was in Frankel D group whereas a patient with mere 14% canal compromise at T12 level had complete neurologic deficit (Figure 1).

Figure 1
figure 1

The relationship between neurological deficit and initial canal compromise

Full size image

There was also no correlation between the neurological recovery and initial canal compromise. This is similar to the views of El Masry et al6,13 who are of the opinion that there is no correlation between the degree of canal encroachment, initial degree of neurological impairment or degree of neurological recovery. However Li-Yang Dai16 found the recovery rate to be significantly related to the first examination stenotic ratio.

It is generally accepted that patients with incomplete injury to the cord or the cauda equina initially have better chances of neurological improvement than patients who initially have complete cord injury.2,12,17 Our study also demonstrated this phenomenon. Of 11 patients with complete neurological deficit, only four recovered, while out of 22 patients with incomplete deficit, 18 improved upon their pretreatment Frankel grade. These findings support the view of Rosenberg et al18 that the initial impact to the spinal cord determines the future neurological outcome regardless of the spinal segment injured.

Limb et al19 suggest that the static image of the canal obtained by the computerized tomography scans hours or days after the injury does not necessarily reflect the displacement at the time of injury, which is what determines the initial neurological insult. The degree of spinal canal narrowing reflects the final resting position of the vertebral body fragments after the trauma. This would explain why the present study failed to show any association between the extent of canal compromise and the extent of neurological deficit (Figure 1).

Several studies have shown that the spinal canal remodels as the time progresses.16,20,21,22,23,24 de Klerk et al24 observed that the process of remodeling usually takes place during the first year after injury; after this period, there is little further remodeling. It is for this reason that we analyzed the follow-up canal compromise at 1 year. The percentage of remodeling noted in this study ranged from 28% to 100%. Canal remodeling was seen in all patients irrespective of the neurological involvement. However the extent of canal remodeling was not related to the degree of neurological recovery (Figures 2a,b and 3a,b).

Figure 2
figure 2

(A) CT scan of a 52 year old man with a burst fracture at L1 showing canal compromise of 69%. He had neurological deficit of Frankel grade ‘D’ (B) CT scan at 1 year follow-up showing 29% canal compromise (remodeling of 58%). He recovered to Frankel grade ‘E’

Full size image
Figure 3
figure 3

(A) CT scan of a 47 year old male with a burst fracture at L1 showing canal compromise of 32%. He had Frankel grade ‘A’ paraplegia. (B) CT scan at 1 year follow-up showing 20% canal compromise (remodeling of 37%). He did not show any recovery

Full size image

In the light of our results the rationale of decompression of the spinal canal to remove any retropulsed bony fragments with a hope of improving the neurological status advocated by some surgeons25,26 seems questionable.

Conclusion

We conclude that in burst fractures of the thoracolumbar and lumbar spine there is no correlation between the neurologic deficit or the recovery pattern with the extent of canal compromise. Like other bones, fractured vertebrae also undergo substantial remodeling so that the size and shape of the spinal canal improves with time. However this remodeling does not have a bearing on neurological recovery.