Introduction
Thoracic ossification of the ligamentum flavum (TOLF) is characterized by heterotopic bone formation in the ligamentum flavum. TOLF most commonly occurs in Asian patients aged 40–70 years [1,2]. This disorder has been shown to be a major cause of thoracic spinal stenosis, frequently leading to thoracic compressive myelopathy [1–3].
Surgical treatment is indicated for symptomatic TOLF, and the most commonly used methods are posterior piecemeal and en bloc laminectomy [4–6]. Although both resection methods have been demonstrated to be safe and effective, en bloc resection is believed to result in a better surgical prognosis, particularly for multilevel TOLF; this is because of its shorter surgical time, reduced intraoperative bleeding, and reduced invasion of the spinal canal [5,6]. However, several studies have shown that en bloc resection is more likely to cause dural tear and cerebrospinal fluid leakage (CFL) than piecemeal resection [6,7]. Furthermore, a recently published study demonstrated that some patients exhibited significantly increased motor-evoked potential (MEP) amplitudes, which then gradually decreased to levels below baseline after successful OLF removal during en bloc laminectomy [6]. In contrast, these significant MEP fluctuations do not occur in piecemeal laminectomy [6]. According to previous studies [8–10], intraoperative MEP changes effectively reflect spinal cord function in real time. Thus, patients with a final reduction in MEPs usually exhibit transient or permanent motor function deterioration after en bloc laminectomy [6].
Therefore, in this study, we devised a surgical modification for laminectomy that includes both piecemeal and en bloc resection methods to reduce intraoperative spinal cord irritation. We further evaluated both the intraoperative interference and clinical prognosis of this hybrid laminectomy and compared them with those of conventional en bloc laminectomy.
Materials and Methods
The methods used in this retrospective study were conducted according to the principles of the Declaration of Helsinki, and the research protocol was approved by the Human Ethics Committee of Huashan Hospital (Fudan University, China; approval no., KY2025-068).
Study design
The preoperative demographic and perioperative medical data were obtained from all patients. The Ashworth and modified Japanese Orthopedic Association (mJOA) scores were collected from all patients with TOLF before and after surgery. Furthermore, somatosensory-evoked potentials (SEPs) and MEPs were measured using NIM-Eclipse (Medtronic; Medtronic Inc., Jacksonville, FL, USA) during surgery, and intraoperative fluctuations in SEP and MEP amplitudes were recorded after each critical manipulation.
Participants
Between January 2018 and June 2023, consecutive patients diagnosed with TOLF and treated at the Orthopedics Department of the authors’ hospital were selected as potential subjects. Informed consent was obtained from all participants.
The inclusion criteria were as follows: patients who had a confirmed diagnosis of multilevel TOLF through preoperative imaging studies and postoperative histopathology records and those with TOLF that had spinal canal-occupying ratio ≥40% in cross-sectional CT (ImageJ; National Institutes of Health, Bethesda, MD, USA) because spinal canal-occupying ratios in TOLF that exceed 40% may be associated with dural adhesion, CFL, and postoperative neurological deterioration, according to previous studies [11,12]. Furthermore, the patients included in this study underwent multilevel (≥2) hybrid laminectomy or conventional en bloc laminectomy. The exclusion criteria were as follows: patients who underwent previous spinal surgery; those with cervical or lumbar spinal disorders, spinal tumors, malformation or infections, ankylosing spondylitis, and other coexisting diseases affecting peripheral nerve function; and those who were lost to follow-up.
According to the inclusion and exclusion criteria, this retrospective study included 48 patients with multilevel TOLF who underwent postoperative multilevel laminectomy (Supplement 1). Until the end of 2020, we primarily performed en bloc laminectomy for patients with TOLF. Since 2021, hybrid laminectomy has become the main surgical method at our institution for treating multilevel TOLF. Consequently, en bloc laminectomy was performed in 28 patients with TOLF, whereas the remaining 20 patients underwent hybrid laminectomy.
Surgical treatment
Total intravenous anesthesia was used for all patients with TOLF. Anesthesia was induced by propofol infusion combined with a single bolus of a nondepolarizing short-acting muscle relaxant [13]. Anesthesia was maintained with propofol infusion and remifentanil. Furthermore, mean arterial pressure was maintained at >75 mm Hg during surgery, and a skin temperature >34°C was ensured [13].
The surgical procedures for en bloc laminectomy were described in detail in our previous study [6], whereas those for hybrid laminectomy are presented in Fig. 1. In brief, we first performed en bloc laminectomy for patients with mild TOLF (spinal canal-occupying ratio <40%), and piecemeal laminectomy was used to treat patients with severe TOLF (spinal canal-occupying ratio ≥40%). Furthermore, simultaneous posterior segmental instrumented fusion was performed in all patients due to multilevel laminectomy. If several separated severe TOLFs were observed, we first performed en bloc laminectomy for all mild TOLFs and then piecemeal laminectomy for severe TOLFs one by one.
Variables
Intraoperative monitoring
Baseline amplitudes were obtained after exposing the surgical area, and MEPs and SEPs were recorded after each critical manipulation as a percentage of the amplitude relative to the baseline.
MEPs were elicited using a brief (50 μs), high-intensity (up to 300 mA) anodal pulse train (4–6 pulses) between two needle electrodes placed at C3 and C4 (International 10–20 system) and were recorded bilaterally from the abductor hallucis (AH) and tibialis anterior (TA) and from the abductor pollicis brevis (used as a control). Bilateral SEPs were elicited by stimulating the bilateral posterior tibial nerve using surface electrodes with a square wave (0.2 ms, 20–40 mA, 2.5 Hz). Recording was performed using needle electrodes positioned at Cz-FPz, and the SEP signal was averaged after 200 stimulations. All stimulation measurements were optimized to elicit maximal baseline amplitudes for each patient and remained unchanged during surgery.
According to our previous study [6], IOM deterioration was defined as a reduction in the MEP amplitude of ≥80% or a reduction in the SEP amplitude of ≥50% after surgery. In contrast, IOM improvement was defined as an increase in the MEP amplitude of ≥80% or an increase in the SEP amplitude of ≥50% after surgery. The IOM improvement rate in the IOM improvement group was calculated as follows: (postoperative amplitude–baseline amplitude)/baseline amplitude×100%. The other remaining patients in this study were classified into the no change group.
Clinical assessments
The operative time, intraoperative bleeding volume, and incidence of perioperative complications were assessed in all patients. Ashworth scores were recorded for all patients before surgery, 3 days after surgery, and 1 year after surgery. mJOA scores were evaluated in all patients with TOLF before and 1 year after surgery. The postoperative 1-year neurological recovery rate (NRR) was calculated as follows: (postoperative–preoperative mJOA score)/(11–preoperative mJOA score)×100%.
Statistical methods
All data were analyzed using IBM SPSS ver. 21.0 (IBM Corp., Armonk, NY, USA). The Kolmogorov-Smirnov test was used to assess data normality. The measurements between the en bloc and hybrid laminectomy groups were compared using an independent t-test or the Mann-Whitney U test, whereas measurements between preoperative and postoperative follow-up assessments were compared using the paired t-test or Wilcoxon signed-rank test. Differences in frequencies between the two laminectomy groups were evaluated using the chi-square test. One-way analysis of variance (with least-significant-difference correction) or the Kruskal-Wallis H test was used to evaluate the measurements among the IOM improvement, IOM deterioration, and no change groups. The same statistical tests were used to analyze both MEP and SEP amplitudes at baseline and after different critical manipulations in both laminectomy groups. Pearson’s correlation analysis was used to assess the correlation between IOM improvement rates and clinical assessments in the IOM improvement group. In all cases, p-values <0.05 were used to denote statistical significance.
Results
Intraoperative assessments
Table 3 presents the changes in MEP and SEP amplitudes after each critical manipulation in the en bloc and hybrid laminectomy groups.
In this study, MEP monitoring was unsuccessful in all tested lower limb muscles in five of the 48 patients at the beginning of surgery; however, these five patients were successfully monitored via SEPs (Table 3). No significant differences in mean MEP and SEP amplitudes in all tested muscles were observed among the baseline and different critical manipulations in both laminectomy groups (Fig. 2).
At the end of the operation, three patients exhibited a significant reduction in MEP and SEP amplitudes, whereas seven patients exhibited a reduction in MEP amplitudes without a corresponding reduction in SEP amplitudes (Table 3). Thus, 10 patients (10/48, 20.8%) were included in the IOM deterioration group. In contrast, 11 patients (11/48, 22.9%) were included in the IOM improvement group (Table 3), all of whom exhibited a marked increase in MEP amplitudes without changes in SEP amplitudes after surgery. No significant difference in the incidence of either IOM deterioration (6/28, 21.4% versus 4/20, 20.0%; p=0.904) or IOM improvement (6/28, 21.4% versus 5/20, 25.0%, p=0.772) between the en bloc and hybrid laminectomy groups (Table 3).
Furthermore, four patients who underwent en bloc laminectomy exhibited a significant increase in MEP amplitudes (≥80%) in at least one tested muscle after OLF removal; however, these increased MEP amplitudes rapidly decreased, either returning to baseline levels (2/4, 50%) or decreasing to levels below baseline (2/4, 50%) (Fig. 3). During this process, the SEP amplitudes remained relatively stable in these patients. In contrast, this condition did not occur in the hybrid laminectomy group. Equally important, in the IOM improvement group, patients who underwent hybrid laminectomy exhibited relatively greater improvement rates in MEPs in the bilateral AH (left side: 213.4%±35.9% versus 152.5%±41.0%, p=0.028; right side: 201.2%±32.0% versus 145.2%±46.3%, p=0.043) than those who underwent en bloc laminectomy (p<0.05).
Furthermore, no significant differences in the operative time, intraoperative bleeding volume, number of surgical segments, or incidence of intraoperative dural tears were observed between the two laminectomy groups (Table 1).
Postoperative assessments
Early postoperative assessment
Compared with preoperative assessments, significantly lower Ashworth scores were observed in both laminectomy groups 3 days after surgery; no significant difference in the 3-day postoperative Ashworth scores was observed between the two groups (Table 2). Furthermore, no significant difference in the incidence of postoperative epidural hematoma or transient neurological deterioration was observed between the two laminectomy groups. In contrast, a greater number of patients who underwent en bloc laminectomy had postoperative CFL than those who underwent hybrid laminectomy (Table 1).
Postoperative 1-year assessments
Compared with preoperative assessments, patients in both laminectomy groups exhibited reduced Ashworth scores and increased mJOA scores 1 year after surgery, with no significant difference in Ashworth scores, mJOA scores, or NRRs between the two laminectomy groups at the 1-year postoperative assessment (Table 2). Furthermore, no significant difference in the incidence of postoperative surgical site infection (SSI) was observed between the two laminectomy groups (Table 1).
Correlation between clinical assessment and monitoring results
Patients in the IOM improvement and no change groups exhibited significantly lower Ashworth scores at the 3-day and 1-year postoperative assessments and greater 3-day postoperative changes in the Ashworth scores than those who experienced IOM deterioration (Supplement 2). Furthermore, the 1-year postoperative change in the Ashworth scores was greater in the IOM improvement group than in the IOM deterioration group (Supplement 2). Furthermore, 1-year postoperative assessments revealed that both mJOA scores and NRR were significantly higher in the IOM improvement group than in the other two groups (Supplement 2), and the no change group had a greater NRR than the IOM deterioration group (Supplement 2). Furthermore, a clear correlation was observed between the NRR and MEP improvement rates in the bilateral TA (left TA: r=0.686, p=0.020; right TA: r=0.709, p=0.015) in the IOM improvement group (Supplement 3). Moreover, a relationship was noted between the MEP improvement rates in the right AH and the postoperative change in the Ashworth scores (postoperative 3-day change: r=0.646, p=0.032; postoperative 1-year change: r=0.695, p=0.018) in the IOM improvement group (Supplement 4).
Patients in the IOM improvement group and/or no change groups had fewer surgical segments, shorter operative time, less intraoperative bleeding, and lower incidences of intraoperative dural tear, postoperative CFL, and transient neurological deterioration than those in the IOM deterioration group (Supplement 5).
Discussion
In this study, four patients with TOLF who underwent conventional en bloc laminectomy exhibited significant abnormal fluctuations in MEPs, with no concurrent changes in SEPs after OLF removal. According to previous studies [14,15], both MEPs and SEPs are mediated by different spinal pathways, and MEPs may be more sensitive than SEPs in detecting changes in spinal cord function related to intraoperative manipulations [8–10,13–15]. Therefore, the initial increase in MEP amplitudes after OLF removal in these patients reflects reduced temporal dispersion and improved conduction in the motor pathways due to the recovery of spinal cord blood supply, as supported by previous studies [6,16,17]. The subsequent reduction in the MEP amplitudes may be attributed to ischemia–reperfusion injury following rapid decompression. Increasing evidence indicates that rapid blood supply restoration can lead to a significant burst of oxygen-derived free radicals within the first few minutes of reperfusion, which are toxic to nerve tissues and subsequently decrease MEP amplitudes [18,19]. Importantly, as shown in Fig. 2, relatively reduced mean MEP amplitudes after surgery were observed in all tested muscles in the en bloc laminectomy group compared with those observed after OLF removal (though the difference did not reach statistical significance). This result indicates that delayed reduction in MEP amplitudes occurs not only in patients with significant recovery of MEPs after OLF removal but also in those undergoing en bloc laminectomy.
However, the delayed reduction in MEP amplitudes after OLF removal was not observed in the hybrid laminectomy group. One possible reason for this finding could be the gradual release of spinal cord compression, as supported by a previous study involving piecemeal laminectomy for TOLF. Furthermore, recently published studies have shown that if surgical manipulations are temporarily halted after MEP and/or SEP deterioration, these potentials often recover spontaneously [9,10], suggesting spinal cord adaptability. Thus, gradually restoring blood supply to allow the spinal cord to adapt may help prevent SCI caused by ischemia–reperfusion injury. Furthermore, this possibility provides an explanation for the difference in the final improvement rates of MEPs between patients who underwent en bloc laminectomy and those who underwent hybrid laminectomy in the IOM improvement group. These findings collectively suggest that hybrid laminectomy causes less intraoperative interference with spinal cord function.
Although no aggravated neural dysfunction or serious complications occurred in either laminectomy group at the last follow-up assessment, a relatively greater number of patients who underwent en bloc laminectomy had postoperative CFL than those who underwent hybrid laminectomy. Previous studies have reported that the incidence of CFL ranged from 11% to 62% in patients with TOLF [20,21], and this complication was closely associated with both postoperative SSI and neurological deterioration [20,21]. Importantly, the occupying ratio of OLF has been identified as a significant risk factor for CFL occurrence [21,22], suggesting that dural tears and subsequent CFLs are more likely to occur in areas with the most severe compression in patients with TOLF, where the dura often adheres to the OLF. Therefore, the dura attached to the OLF is frequently torn during en bloc laminectomy because of the limited field of view. In contrast, during hybrid laminectomy, we performed en bloc resection of the OLF only in areas with mild compression and used piecemeal resection to relieve segments with the most severe OLF compression. This approach may reduce the risk of dural tears and subsequent CFLs. Another possibility that could cause the different incidences of CFL between the two laminectomy groups is that the range of dural tears may be relatively smaller in the hybrid laminectomy group than in the en bloc laminectomy group due to the different resection methods used for severe OLF. Consequently, more patients who undergo en bloc laminectomy may experience inadequate repair.
According to previous studies [4,5], piecemeal laminectomy likely increases the risk of iatrogenic SCI due to spinal cord irritation from a Kerrison rongeur entering the spinal canal to remove the lamina around the OLF. However, similar incidences of MEP/SEP deterioration between the hybrid and en bloc laminectomy groups challenge this view. The similar mean amplitudes of both MEPs and SEPs during different critical manipulations in hybrid laminectomy further suggest that the Kerrison rongeur entering the spinal canal did not significantly interfere with the spinal cord in patients undergoing hybrid laminectomy. Pre-decompression of mild OLF may be one possible reason. The relatively increased mean MEP amplitudes after en bloc resection of mild OLF in hybrid laminectomy indicate that initial en bloc resection of mild OLF compression restores partial blood supply to the spinal cord and creates a buffer space, potentially enhancing the spinal cord’s ability to tolerate further damage from subsequent decompression (Fig. 2). Therefore, hybrid laminectomy may not only avoid the drawbacks of en bloc resection but also improve the safety of piecemeal resection. Importantly, compared with conventional en bloc laminectomy, staged decompression in hybrid laminectomy did not significantly increase surgical time or trauma.
The findings of this study should be interpreted with caution. Although this study suggested that hybrid laminectomy combines the advantages of both en bloc and piecemeal laminectomy, the mJOA and Ashworth scores at the last follow-up assessment were similar between the hybrid and en bloc laminectomy groups. This finding may be because adequate decompression of the spinal cord in both laminectomy groups may promote plastic processes within the spinal cord below the injury site and “extraspinal” complementary mechanisms [23–25], leading to relatively good clinical outcomes. Although en bloc laminectomy may not immediately recover impulse transmission within the spinal cord as hybrid laminectomy can, clinical assessment may be suboptimal for detecting subtle responses to treatment. Although functional improvement is crucial for patients with TOLF in clinical practice, even mild differences in MEP responses (e.g., the difference in final improvement rates of MEPs between the two laminectomy groups in the IOM improvement group) may provide additional unique insights for refining surgical techniques. A potential correlation between MEP improvement and postoperative functional recovery has been reported in both the present study and previous studies [15,16].
When reviewing the results of this study, another clinical limitation identified was that the warning criteria of MEP amplitudes used to detect MEP abnormalities varied across different studies [8–10,13–17]. Unlike deformity correction, an excessive number of false-positive changes can easily lead to insufficient decompression during laminectomy, which may result in poor prognosis for patients with TOLF. Therefore, instead of the more common rate of 50%–75% used in spinal deformity correction [26,27], this study employed a cutoff of 80% as the warning criterion, and similar criteria were also used in spinal surgeries [6,13]. Furthermore, there is a lack of identified criteria for the spinal canal-occupying ratio in TOLF used to determine the type of laminectomy required. However, previous studies have demonstrated that when the spinal canal-occupying ratio of TOLF exceeds 40% or 50%, surgeries are prone to cause various complications, including CFL and neurological deterioration [12,13,20,28], and the sample size in our institution limits the selection of larger spinal canal-occupying rates of TOLF to explore the feasibility of hybrid laminectomy. Therefore, this study employed a cutoff of 40% for the use of different laminectomy approaches. Furthermore, this study was a retrospective data review from a single center, which may have caused selection bias in patient enrollment, and more significant results may be achieved in a future prospective (preferably blinded) randomized controlled trial with a larger number of cases.
Conclusions
The current study showed that hybrid laminectomy is a safe and effective surgical technique for treating patients with multilevel TOLF. This approach can reduce the intraoperative delayed reduction in MEP amplitudes (spinal cord irritation) and incidence of postoperative CFL compared with conventional en bloc laminectomy. Therefore, hybrid laminectomy can be an alternative surgical method for treating multilevel TOLF.
Key Points
Hybrid laminectomy may result in relatively better recovery of spinal pulse transmission in patients with thoracic ossification of the ligamentum flavum (TOLF) and reduce the incidence of cerebrospinal fluid leakage without increasing surgical trauma than conventional en bloc laminectomy.
En bloc laminectomy may result in significant fluctuations in motor-evoked potential (MEP) amplitudes associated with ischemia–reperfusion injury of the spinal cord after decompression, and hybrid laminectomy may effectively prevent intraoperative spinal cord irritation.
Intraoperative changes in MEPs and somatosensory-evoked potentials potentially provide a valid method for quantitatively evaluating the safety of different intraoperative manipulations and their prognostic impacts on the spinal cord.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement1
