Peng-Yuan Chang

Endoscopic transnasal odontoidectomy without resection of nasal turbinates: clinical outcomes of 13 patients

Object The goal of the study was to report a series of consecutive patients who underwent endoscopic transnasal odontoidectomy (ETO) without resection of nasal turbinates. The techniques for this minimally invasive approach are described in detail. Methods The authors conducted a retrospective review of consecutive patients who underwent ETO for basilar invagination. All the patients had myelopathy caused by compression at the cervicomedullary junction, which required surgical decompression. Preoperative and postoperative data, including those from radiographic and clinical evaluations, were compared. Morbidity and mortality rates for the procedure are also reported in detail. Results Thirteen patients (6 men and 7 women) with a mean age of 52.7 years (range 24-72 years) were enrolled. The basilar invagination etiologies were rheumatoid arthritis (n = 5), trauma (n = 4), os odontoideum (n = 2), ankylosing spondylitis (n = 1), and postinfectious deformity (n = 1). The average follow-up duration was 51.2 months (range 0.3-105 months). One patient died 10 days after the operation as a result of meningitis caused by CSF leakage. Among the other 12 patients, the average postoperative Nurick grade (3.2) was significantly improved over that before the operation (4.1, p = 0.004). The mean (± SD) duration of postoperative intubation was 1.5 ± 2.1 days, and there was no need for perioperative tracheostomy or nasogastric tube feeding. There also was no postoperative velopharyngeal insufficiency. There were 6 (46%) intraoperative and 2 (15%) postoperative CSF leaks in the 13 patients in this series. Conclusions ETO is a viable and effective option for decompression at the ventral cervicomedullary junction. This approach is minimally invasive and causes little velopharyngeal insufficiency. The pitfall of this approach is the difficulty in repairing dural defects and subsequent CSF leakage.

Dynamic stabilization for L4-5 spondylolisthesis: comparison with minimally invasive transforaminal lumbar interbody fusion with more than 2 years of follow-up.

OBJECTIVE In the past decade, dynamic stabilization has been an emerging option of surgical treatment for lumbar spondylosis. However, the application of this dynamic construct for mild spondylolisthesis and its clinical outcomes remain uncertain. This study aimed to compare the outcomes of Dynesys dynamic stabilization (DDS) with minimally invasive transforaminal lumbar interbody fusion (MI-TLIF) for the management of single-level spondylolisthesis at L4-5. METHODS This study retrospectively reviewed 91 consecutive patients with Meyerding Grade I spondylolisthesis at L4-5 who were managed with surgery. Patients were divided into 2 groups: DDS and MI-TLIF. The DDS group was composed of patients who underwent standard laminectomy and the DDS system. The MI-TLIF group was composed of patients who underwent MI-TLIF. Clinical outcomes were evaluated by visual analog scale for back and leg pain, Oswestry Disability Index, and Japanese Orthopaedic Association scores at each time point of evaluation. Evaluations included radiographs and CT scans for every patient for 2 years after surgery. RESULTS A total of 86 patients with L4-5 spondylolisthesis completed the follow-up of more than 2 years and were included in the analysis (follow-up rate of 94.5%). There were 64 patients in the DDS group and 22 patients in the MI-TLIF group, and the overall mean follow-up was 32.7 months. Between the 2 groups, there were no differences in demographic data (e.g., age, sex, and body mass index) or preoperative clinical evaluations (e.g., visual analog scale back and leg pain, Oswestry Disability Index, and Japanese Orthopaedic Association scores). The mean estimated blood loss of the MI-TLIF group was lower, whereas the operation time was longer compared with the DDS group (both p < 0.001). For both groups, clinical outcomes were significantly improved at 6, 12, 18, and 24 months after surgery compared with preoperative clinical status. Moreover, there were no differences between the 2 groups in clinical outcomes at each evaluation time point. Radiological evaluations were also similar and the complication rates were equally low in both groups. CONCLUSIONS At 32.7 months postoperation, the clinical and radiological outcomes of DDS were similar to those of MI-TLIF for Grade I degenerative spondylolisthesis at L4-5. DDS might be an alternative to standard arthrodesis in mild lumbar spondylolisthesis. However, unlike fusion, dynamic implants have issues of wearing and loosening in the long term. Thus, the comparable results between the 2 groups in this study require longer follow-up to corroborate.

Dynesys dynamic stabilization–related facet arthrodesis

OBJECTIVE Dynamic stabilization devices are designed to stabilize the spine while preserving some motion. However, there have been reports demonstrating limited motion at the instrumented level of the lumbar spine after Dynesys dynamic stabilization (DDS). The causes of this limited motion and its actual effects on outcomes after DDS remain elusive. In this study, the authors investigate the incidence of unintended facet arthrodesis after DDS and clinical outcomes. METHODS This retrospective study included 80 consecutive patients with 1- or 2-level lumbar spinal stenosis who underwent laminectomy and DDS. All medical records, radiological data, and clinical evaluations were analyzed. Imaging studies included pre- and postoperative radiographs, MR images, and CT scans. Clinical outcomes were measured by a visual analog scale (VAS) for back and leg pain, the Oswestry Disability Index (ODI), and Japanese Orthopaedic Association (JOA) scores. Furthermore, all patients had undergone postoperative CT for the detection of unintended arthrodesis of the facets at the indexed level, and range of motion was measured on standing dynamic radiographs. RESULTS A total of 70 patients (87.5%) with a mean age of 64.0 years completed the minimum 24-month postoperative follow-up (mean duration 29.9 months). Unintended facet arthrodesis at the DDS instrumented level was demonstrated by CT in 38 (54.3%) of the 70 patients. The mean age of patients who had facet arthrodesis was 9.8 years greater than that of the patients who did not (68.3 vs 58.5 years, p = 0.009). There were no significant differences in clinical outcomes, including VAS back and leg pain, ODI, and JOA scores between patients with and without the unintended facet arthrodesis. Furthermore, those patients older than 60 years were more likely to have unintended facet arthrodesis (OR 12.42) and immobile spinal segments (OR 2.96) after DDS. Regardless of whether unintended facet arthrodesis was present or not, clinical evaluations demonstrated improvement in all patients (all p < 0.05). CONCLUSIONS During the follow-up of more than 2 years, unintended facet arthrodesis was demonstrated in 54.3% of the patients who underwent 1- or 2-level DDS. Older patients (age > 60 years) were more likely to have unintended facet arthrodesis and subsequent immobile spinal segments. However, unintended facet arthrodesis did not affect the clinical outcomes during the study period. Further evaluations are needed to clarify the actual significance of this phenomenon.

The Effect of Lumbar Lordosis on Screw Loosening in Dynesys Dynamic Stabilization: Four-Year Follow-Up with Computed Tomography

Introduction. This study aimed to evaluate the effects of Dynesys dynamic stabilization (DDS) on clinical and radiographic outcomes, including spinal pelvic alignment. Method. Consecutive patients who underwent 1- or 2-level DDS for lumbar spondylosis, mild degenerative spondylolisthesis, or degenerative disc disease were included. Clinical outcomes were evaluated by Visual Analogue Scale for back and leg pain, Oswestry Disability Index, and the Japanese Orthopedic Association scores. Radiographic outcomes were assessed by radiographs and computed tomography. Pelvic incidence and lumbar lordosis (LL) were also compared. Results. In 206 patients with an average follow-up of 51.1 ± 20.8 months, there were 87 screws (8.2%) in 42 patients (20.4%) that were loose. All clinical outcomes improved at each time point after operation. Patients with loosened screws were 45 years older. Furthermore, there was a higher risk of screw loosening in DDS involving S1, and these patients were more likely to have loosened screws if the LL failed to increase after the operation. Conclusions. The DDS screw loosening rate was overall 8.2% per screw and 20.4% per patient at more than 4 years of follow-up. Older patients, S1 involvement, and those patients who failed to gain LL postoperatively were at higher risk of screw loosening.

Risk of spinal cord injury in patients with cervical spondylotic myelopathy and ossification of posterior longitudinal ligament: a national cohort study

OBJECTIVE This study aimed to estimate the risk of spinal cord injury (SCI) in patients with cervical spondylotic myelopathy (CSM) with and without ossification of posterior longitudinal ligament (OPLL). Also, the study compared the incidence rates of SCI in patients who were managed surgically and conservatively. METHODS This retrospective cohort study covering 15 years analyzed the incidence of SCI in patients with CSM. All patients, identified from the National Health Insurance Research Database, were hospitalized with the diagnosis of CSM and followed up during the study period. These patients with CSM were categorized into 4 groups according to whether they had OPLL or not and whether they received surgery or not: 1) surgically managed CSM without OPLL; 2) conservatively managed CSM without OPLL; 3) surgically managed CSM with OPLL; and 4) conservatively managed CSM with OPLL. The incidence rates of subsequent SCI in each group during follow-up were then compared. Kaplan-Meier and Cox regression analyses were performed to compare the risk of SCI between the groups. RESULTS Between January 1, 1999, and December 31, 2013, there were 17,258 patients with CSM who were followed up for 89,003.78 person-years. The overall incidence of SCI in these patients with CSM was 2.022 per 1000 person-years. Patients who had CSM with OPLL and were conservatively managed had the highest incidence of SCI, at 4.11 per 1000 person-years. Patients who had CSM with OPLL and were surgically managed had a lower incidence of SCI, at 3.69 per 1000 person-years. Patients who had CSM without OPLL and were conservatively managed had an even lower incidence of SCI, at 2.41 per 1000 person-years. Patients who had CSM without OPLL and were surgically managed had the lowest incidence of SCI, at 1.31 per 1000 person-years. The Cox regression model demonstrated that SCIs are significantly more likely to happen in male patients and in those with OPLL (HR 2.00 and 2.24, p < 0.001 and p = 0.007, respectively). Surgery could significantly lower the risk for approximately 50% of patients (HR 0.52, p < 0.001). CONCLUSIONS Patients with CSM had an overall incidence rate of SCI at approximately 0.2% per year. Male sex, the coexistence of OPLL, and conservative management are twice as likely to be associated with subsequent SCI. Surgery is therefore suggested for male patients with CSM who also have OPLL.

Differences between C3-4 and other subaxial levels of cervical disc arthroplasty: more heterotopic ossification at the 5-year follow-up.

OBJECTIVE Several large-scale clinical trials demonstrate the efficacy of 1- and 2-level cervical disc arthroplasty (CDA) for degenerative disc disease (DDD) in the subaxial cervical spine, while other studies reveal that during physiological neck flexion, the C4-5 and C5-6 discs account for more motion than the C3-4 level, causing more DDD. This study aimed to compare the results of CDA at different levels. METHODS After a review of the medical records, 94 consecutive patients who underwent single-level CDA were divided into the C3-4 and non-C3-4 CDA groups (i.e., those including C4-5, C5-6, and C6-7). Clinical outcomes were measured using the visual analog scale for neck and arm pain and by the Japanese Orthopaedic Association scores. Postoperative range of motion (ROM) and heterotopic ossification (HO) were determined by radiography and CT, respectively. RESULTS Eighty-eight patients (93.6%; mean age 45.62 ± 10.91 years), including 41 (46.6%) female patients, underwent a mean follow-up of 4.90 ± 1.13 years. There were 11 patients in the C3-4 CDA group and 77 in the non-C3-4 CDA group. Both groups had significantly improved clinical outcomes at each time point after the surgery. The mean preoperative (7.75° vs 7.03°; p = 0.58) and postoperative (8.18° vs 8.45°; p = 0.59) ROMs were similar in both groups. The C3-4 CDA group had significantly greater prevalence (90.9% vs 58.44%; p = 0.02) and higher severity grades (2.27 ± 0.3 vs 0.97 ± 0.99; p = 0.0001) of HO. CONCLUSIONS Although CDA at C3-4 was infrequent, the improved clinical outcomes of CDA were similar at C3-4 to that in the other subaxial levels of the cervical spine at the approximately 5-year follow-ups. In this Asian population, who had a propensity to have ossification of the posterior longitudinal ligament, there was more HO formation in patients who received CDA at the C3-4 level than in other subaxial levels of the cervical spine. While the type of artificial discs could have confounded the issue, future studies with more patients are required to corroborate the phenomenon.

Scoliosis may increase the risk of recurrence of lumbar disc herniation after microdiscectomy

OBJECT The aim of this paper was to investigate the risk of recurrence of lumbar disc herniation (LDH) in patients with scoliosis who underwent microdiscectomy. METHODS A series of consecutive patients who underwent microdiscectomy for LDH was retrospectively reviewed. The inclusion criteria were young adults younger than 40 years who received microdiscectomy for symptomatic 1-level LDH. An exclusion criterion was any previous spinal surgery, including fusion or correction of scoliosis. The patients were divided into 2 groups: those with scoliosis and those without scoliosis. The demographic data in the 2 groups were similar. All medical records and clinical and radiological evaluations were reviewed. RESULTS A total of 58 patients who underwent 1-level microdiscectomy for LDH were analyzed. During the mean follow-up of 24.6 months, 6 patients (10.3%) experienced a recurrence of LDH with variable symptoms. The recurrence rate was significantly higher among the scoliosis group than the nonscoliosis group (33.3% vs. 2.3%, p = 0.001). Furthermore, the recurrence-free interval in the scoliosis group was short. CONCLUSIONS Young adults (< 40 years) with uncorrected scoliosis are at higher risk of recurrent LDH after microdiscectomy.

Can segmental mobility be increased by cervical arthroplasty

OBJECTIVE Many reports have successfully demonstrated that cervical disc arthroplasty (CDA) can preserve range of motion after 1- or 2-level discectomy. However, few studies have addressed the extent of changes in segmental mobility after CDA or their clinical correlations. METHODS Data from consecutive patients who underwent 1-level CDA were retrospectively reviewed. Indications for surgery were medically intractable degenerative disc disease and spondylosis. Clinical outcomes, including visual analog scale (VAS)-measured neck and arm pain, Neck Disability Index (NDI), and Japanese Orthopaedic Association (JOA) scores, were analyzed. Radiographic outcomes, including C2-7 Cobb angle, the difference between pre- and postoperative C2-7 Cobb angle (ΔC2-7 Cobb angle), sagittal vertical axis (SVA), the difference between pre- and postoperative SVA (ΔSVA), segmental range of motion (ROM), and the difference between pre- and postoperative ROM (ΔROM), were assessed for their association with clinical outcomes. All patients underwent CT scanning, by which the presence and severity of heterotopic ossification (HO) were determined during the follow-up. RESULTS A total of 50 patients (mean age 45.6 ± 9.33 years) underwent a 1-level CDA (Prestige LP disc) and were followed up for a mean duration of 27.7 ± 8.76 months. All clinical outcomes, including VAS, NDI, and JOA scores, improved significantly after surgery. Preoperative and postoperative ROM values were similar (mean 9.5° vs 9.0°, p > 0.05) at each indexed level. The mean changes in segmental mobility (ΔROM) were -0.5° ± 6.13°. Patients with increased segmental mobility after surgery (ΔROM > 0°) had a lower incidence of HO and HO that was less severe (p = 0.048) than those whose ΔROM was < 0°. Segmental mobility (ROM) was significantly lower in patients with higher HO grade (p = 0.012), but it did not affect the clinical outcomes. The preoperative and postoperative C2-7 Cobb angles and SVA remained similar. The postoperative C2-7 Cobb angles, SVA, ΔC2-7 Cobb angles, and ΔSVA were not correlated to clinical outcomes after CDA. CONCLUSIONS Segmental mobility (as reflected by the mean ROM) and overall cervical alignment (i.e., mean SVA and C2-7 Cobb angle) had no significant impact on clinical outcomes after 1-level CDA. Patients with increased segmental mobility (ΔROM > 0°) had significantly less HO and similarly improved clinical outcomes than those with decreased segmental mobility (ΔROM < 0°).

The importance of atlantoaxial fixation after odontoidectomy

OBJECT Although anterior odontoidectomy has been widely accepted as a procedure for decompression of the craniovertebral junction (CVJ), postoperative biomechanical instability has not been well addressed. There is a paucity of data on the necessity for and choice of fixation. METHODS The authors conducted a retrospective review of consecutively treated patients with basilar invagination who underwent anterior odontoidectomy and various types of posterior fixation. Posterior fixation included 1 of 3 kinds of constructs: occipitocervical (OC) fusion with atlantoaxial (AA) fixation, OC fusion without AA fixation, or AA-only (without OC) fixation. On the basis of the use or nonuse of AA fixation, these patients were assigned to either the AA group, in which the posterior fixation surgery involved both the atlas and axis simultaneously, regardless of whether the patient underwent OC fusion, or the non-AA group, in which the OC fusion construct spared the atlas, axis, or both. Clinical outcomes and neurological function were compared. Radiological results at each time point (i.e., before and after odontoidectomy and after fixation) were assessed by calculating the triangular area causing ventral indentation of the brainstem in the CVJ. RESULTS Data obtained in 14 consecutively treated patients with basilar invagination were analyzed in this series; the mean follow-up time was 5.75 years. The mean age was 53.58 years; there were 7 males and 7 females. The AA and non-AA groups consisted of 7 patients each. The demographic data of both groups were similar. Overall, there was significant improvement in neurological function after the operation (p = 0.03), and there were no differences in the postoperative Nurick grades between the 2 groups (p = 1.00). According to radiological measurements, significant decompression of the ventral brainstem was achieved stepwise in both groups by anterior odontoidectomy and posterior fixation; the mean ventral triangular area improved from 3.00 ± 0.86 cm2 to 2.08 ± 0.51 cm2 to 1.68 ± 0.59 cm2 (before and after odontoidectomy and after fixation, respectively; p < 0.05). The decompression gained by odontoidectomy (i.e., reduction of the ventral triangular area) was similar in the AA and non-AA groups (0.66 ± 0.42 cm2 vs 1.17 ± 1.42 cm2, respectively; p = 0.38). However, the decompression achieved by posterior fixation was significantly greater in the AA group than in the non-AA group (0.64 ± 0.39 cm2 vs 0.17 ± 0.16 cm2, respectively; p = 0.01). CONCLUSIONS Anterior odontoidectomy alone provides significant decompression at the CVJ. Adjuvant posterior fixation further enhances the extent of decompression after the odontoidectomy. Moreover, posterior fixation that involves AA fixation yields significantly more decompression of the ventral brainstem than OC fusion that spares AA fixation.

Letters to the editor: Hypermobility accelerates adjacent-segment disease after ACDF?

To The ediTor: We read with great interest the article by McDonald et al.5 (McDonald CP, Chang V, McDonald M, et al: Three-dimensional motion analysis of the cervical spine for comparison of anterior cervical decompression and fusion versus artificial disc replacement in 17 patients. Clinical article. J Neurosurg Spine 20:245– 255, March 2014). The authors compared data of 7 patients who underwent cervical arthroplasty with those of 10 patients who underwent anterior cervical discectomy and fusion (ACDF) and concluded that after surgery there was increased motion at nonoperative segments in ACDFtreated patients. Although the result appears to corroborate most surgeons’ inference, there are several caveats. First, these two groups of patients are not similar. Their cervical spines had variable degeneration. For example, the arthroplasty-treated patient whose images are shown in Fig. 3 had significant spondylosis and anterior osteophytes in both the cranial and caudal adjacent levels. On the other hand, the ACDF-treated patient whose images are shown in Fig. 4 had little spondylotic changes throughout the entire cervical spine. Several studies have demonstrated the differences of cervical arthroplasty among patients with and without spondylosis.11 In nonrandomized studies, mildly different indications for arthroplasty could cause variable outcomes.9,10 The “less motion” detected in every other level of the cervical discs among the arthroplasty group compared to the ACDF group of this study, as demonstrated in Figs. 5–10, can be attributed to the spondylosis per se, rather than the effect of the artificial disc. Second, the postoperative motion of adjacent segments after anterior cervical discectomy should be compared to the preoperative condition in the same patient. Comparing two patients who have undergone different procedures inadvertently allows for confounding results. Ideally, the mean difference between preand postoperative segmental motion of each patient in the ACDF group could be compared to that of the arthroplasty group. The authors are commended for shedding light on the change in physiological motion after cervical discectomy. The true incidence and cause of adjacent-segment disease after cervical discectomy is still uncertain.12 Furthermore, whether artificial disc replacement can overcome adjacent-segment disease remains controversial, despite many randomized control trials having been published.1–4,6–8 Motion analysis of the cervical spine in healthy, ACDFtreated, and arthroplasty-treated individuals might provide insight into future spine care. Peng-Yuan Chang, M.d.1,2 Yu-Shu Yen, M.d.1,2 Jau-Ching Wu, M.d., Ph.d.1,2 Wen-Cheng huang, M.d., Ph.d.1,2 TSung-hSi Tu, M.d.1,2 Li-Yu FaY, M.d.1,2 henriCh Cheng, M.d., Ph.d.1–3 1Neurological Institute Taipei Veterans General Hospital Taipei, Taiwan 2School of Medicine National Yang-Ming University Taipei, Taiwan 3Institute of Pharmacology National Yang-Ming University Taipei, Taiwan