First, primary thoracic spinal stenosis is rare . The formation of T-OLF is strongly considered due to degenerativewear and tear, as well as external triggers such as biomechanical stress due to posttraumatic ossification [16, 17].There are reports showing that OLF manifestationsare related to DM, heavymanual labour in males, highBMI values in females, increased bone density,and DISH , and these findings are in accordance with the demographic results in our study. Thus, T-OLF coincides with spinal osteoarthritis and spondylosis deformans, leading to a high incidence of tandem T-OLF and other stenosis lesions in the cervical or lumbar area, sometimes with the ossification of ligaments. Cases with the ossification of ligaments have a tendency to develop into tandem spinal stenosis (TSS) ; therefore, T-OLF seems to be no exception, as T-OLF is frequently found in C-OPLL patients . Although spondylotic process is mainly observed in the lower lumbar spine of T-OLF patients, the incidence of lumbar ossifiedlesions (L-OLF or L-OPLL) is relatively low (5 cases, 8.2%).In the present study, up to 75.4% of thoracic myelopathic patients who underwent T-OLF surgeries had LCS, which was higher than that reported in a study in cervical myelopathic patients (57.9%)  but consistent with that reported in a previous study about thoracic myelopathy .
Patients with compressive myelopathy due to especially lower thoracic lesions often present with pain, numbness, or motor disturbance in the lower limbs and neurologic claudication, which is consistent with the symptoms of lumbar radiculopathy or cauda equine lesions. Actually, with or without LCS, most of the T-OLF patients in this study suffered from leg pain, numbness, or dysfunction. Spasticity of the lower limbs with hyperreflexia is often masked by cauda equine or lumbar nerve root disorders [6, 20]. In patients with complicated characteristics, it is difficult to diagnose thoracic pathologies. Despite the rate of combined spinal stenosis being high, there have been few reports describing the clinical course of tandem thoracic and lumbar stenosis [5, 6, 20]. The present study includes a reasonable number of tandem T-OLF and LCS subjects and discusses several aspects of these peculiar cases.
Some reports have mentioned that a short preoperative symptom duration, a single-level lesion, and the unilateral type were favourable prognostic factors, while the bridged (fused) type, the beak type, intramedullary HIA on T2-weighted images, concomitant intervertebral degeneration (Modic change), and the presence of a ventral compressive lesion were poor prognostic factors for T-OLF surgery [11, 12, 21]. Contrary to our expectations, we revealed that these factors were not statistically associated, but that age, a short stature, and coexistent LCS were related to a poor prognosis. Considering that the number of patients with cervical or upper thoracic lesions was smaller in the poor outcome group than in the good outcome group (Figure3), the dynamic biomechanical factor based on their predilection in the lower thoracic spine could be attributed to progressive thoracic myelopathy through compression of the spinal cord.
Second, compared with the non-LCS patients, the patients with tandem T-OLF and LCS in this studyexhibited an older age, a lower preoperative MMT grade, a greater SVA, and degenerative Modic changes in thoracic lesions, resulting in lower postoperative T-JOA scores.Interestingly, the younger non-LCS group exhibited a higher coexistence rate of C-OPLL.The early formation of tandem ossifications, early detection, and younger age at the time ofT-OLF surgery could partially explain the good T-JOA scores found in the non-LCS group.An older age might be not only a determining factor of tandem T-OLF/LCS occurrence (Figure 2)  but also one of the most influential factors of neurological improvement in patients withT-OLF. Eskander et al. reported that age increases the risk of major and minor complications, regardless of the surgical algorithm used to manage tandem stenosis lesions . Other reports have concluded that the coexistence of LCS in T-OLF patients has adverse effects on their surgical outcomes, and vice versa [5, 6]. These findings seem to support our conclusion that the prognosis for T-OLF paralleled that of isolated lumbar stenosis, rather than the preoperative symptom duration, the surgical procedures, the local T-OLF size, morphology, the number of affected OLF lesions, or presence of radiographic cervical stenosis. Although the causal relationship between compressive lesions at levels below the operative site and the appearance of symptoms could not be confirmed in the present study, the presence of LCS itself might partially augment the risk of thoracic cord compression, with this lumbar stenosis lesion serving as a fulcrum according to the lever principle; in addition, asymptomatic LCS might naturally progress, leading to the development of lumbar-related symptoms within a few years after thoracic surgery.
The question of how to treat these elderly patients with tandem combined thoracic and lumbar stenosis who have neurological deficits in the lower limbs remains. As a rule, in the operative sequence for tandem T-OLF and LCS, the most clinically symptomatic area should be decompressed first. Radiological LCS itself does not always affect the clinical symptoms. If neurological symptoms originate equally from both the thoracic and lumbar spine and there are severe radiological findings, both lesions require surgical decompression. We generally addressed the thoracic spine first. Thoracicdecompression improve lumbar symptoms because the lumbar neural fibres may also be under compression by the processes T-OLF formation. Among these cases, in cases without severe systemic problems, simultaneous thoracic and lumbar decompression is sometimes performed after an adequate and careful explanationof its invasiveness to the patient.
We further investigated the clinical outcomes of lumbar decompression in OLF patients to clarify the impact of concurrent LCS. A total of 21.3% of the T-OLF patients required additional lumbar surgery. Although the recovery rate of the T-JOA score did not reach significance with or without lumbar decompression for LCS, we determined the effectiveness of additional lumbar decompression in cases of symptomatic tandem T-OLF and LCS after thoracic treatment. It remains unknown which procedure, two-stage or single-stage surgery, is more effective for achieving postoperative neurologic improvements in T-OLF patients with symptomatic LCS. The degree of neurological recovery after simultaneous single-stage surgery in tandem T-OLF and LCS cases was also equal to that after staged surgery. This result suggested that lumbar operations can be staged if the single-stage surgery is considered to be too invasive for the patient. Careful observation is needed after the first thoracic surgery for these patients to undergo an additional lumbar surgery at the appropriate time.
As one of the limitations, this analysis was based on a simple comparison of the widely accepted T-JOA score in evaluating thoracic surgical outcomes, and we did not comparethe T-JOA score to other tests. As shown in Table 2 and Table 3, the improvement in the VAS score did not always reflect that in the T-JOA score. These results indicated that good improvement in the T-JOA score could reflect improvement in the neurological functional status, whereas subjective symptoms were expressed in the VAS score.This study has several other limitations, including the following: 1) the patient baseline was not controlled, and the number of subgroup analysis was limited; 2) this was a retrospective investigation; and 3) the assessment of the correlation between the dynamic factor in OLF levels and clinical outcomes was inadequate. As mentioned above, these limitations could be partially ascribed to the challenges of assessing and treating the neurological symptoms related to OLF and those of performing biomechanical evaluations in the thoracic spinal system, which has a limited range of segmental motion. Other sophisticated evaluation methods and a large numberof prospective data need to be collected in the future to study the clinical outcomes of OLF cases in more detail.
Given these results, T-OLF may be related to an individual’s age and lifestyle, and is likely accompanied by other spinal stenosis lesions. The identification of the predominant lesion and the interpretation of clinical outcomes seems to be controversial and difficult in T-OLF cases.We focused on tandem T-OLF and LCS, which are highly prevalent. The prognosis for patients with T-OLF and concurrent LCS was worse than that for patients without LCS.The present study results concluded that the severity of lumbar lesions could have great influence on neurological improvement after thoracic surgery for OLF, and an additional surgery for another lumbar lesion could significantly improve the neurological outcomes. Therefore, surgical management in these tandem stenosis should be tailored to the patient’s age and general condition, whether the procedures are performed simultaneously or in stages, as noted previously [10, 11]..