Tumor occupancy ratio as a predictor of neurological recovery after hemilaminectomy for intradural spinal tumors Authors: Anusorn Mungmee, Prachya Punyarat Published: 2026 ABSTRACT BackgroundHemilaminectomy is increasingly used as a less invasive alternative to conventional laminectomy for resection of intradural spinal tumors. This study aimed to evaluate whether the tumor occupancy ratio on preoperative axial magnetic resonance imaging (MRI) was associated with postoperative neurological improvement after hemilaminectomy for selected intradural spinal tumors and to describe the surgical outcomes of this approach.MethodsWe retrospectively reviewed adult patients who underwent hemilaminectomy for intradural spinal tumors between 2018 and 2024 at a single institution and had at least 6 months of follow-up. Neurological status was assessed preoperatively and at the last follow-up using the modified McCormick scale (MMS). Clinical, radiographic, and operative variables, including tumor occupancy ratio on axial MRI, were analyzed using univariate tests and multivariable logistic regression.ResultsForty-six patients were included (mean age 56.3±14.0 years; 78% female). Gross total resection was achieved in 44 of 46 patients (96%). The mean hemilaminectomy corridor ratio, measured as the hemilaminectomy width divided by the interpedicular distance, was 0.55±0.12. At the last follow-up (median 27 months), 37 patients (80.4%) showed neurological improvement. Two patients (4.3%) developed postoperative neurological deterioration. On univariate analysis, sagittal tumor extent, operative time, blood loss, and tumor occupancy ratio were associated with MMS improvement. On multivariable analysis, tumor occupancy ratio was the only variable independently associated with postoperative neurological non-improvement [odds ratio (OR) 1.31, 95% confidence interval (CI): 1.02–1.68; P=0.03].ConclusionsHemilaminectomy achieved a high rate of gross total resection with favorable neurological outcomes in selected patients with solid intradural spinal tumors. Tumor occupancy ratio was associated with postoperative neurological outcome and may be useful as a preoperative prognostic imaging marker, although further validation is required. Highlight box Key findings • Among 46 patients who underwent hemilaminectomy for selected intradural spinal tumors, 96% achieved gross total resection and 80% showed neurological improvement. • A higher preoperative tumor occupancy ratio on magnetic resonance imaging (MRI) was associated with poorer postoperative neurological recovery and was the only factor retained in the exploratory multivariable analysis. What is known and what is new? • Hemilaminectomy is a tissue-sparing alternative to conventional laminectomy that preserves the contralateral musculoligamentous complex while providing adequate exposure for resection of selected intradural spinal tumors. • This study demonstrates favorable surgical outcomes after hemilaminectomy and highlights tumor occupancy ratio as a simple preoperative MRI-based factor associated with postoperative neurological recovery. What is the implication, and what should change now? • Tumor occupancy ratio is a simple, reproducible MRI-based parameter that may help predict postoperative neurological recovery, but its clinical utility requires external validation in larger independent cohorts. Introduction Intradural spinal tumors are relatively uncommon, with an estimated incidence of approximately 0.74 per 100,000 person-years. Intradural extramedullary tumors are the most common subtype, with an incidence of approximately 0.3–0.4 per 100,000 persons (1,2). Patients commonly present with pain, myelopathy, sensory disturbance, gait impairment, or sphincter dysfunction due to progressive compression of the spinal cord or cauda equina. Because symptom progression is often insidious, many patients already have neurological deficits at the time of diagnosis. Conventional laminectomy has historically been the standard surgical approach for intradural spinal tumor resection because it provides wide posterior exposure and facilitates microsurgical tumor removal. However, this technique requires bilateral paraspinal muscle dissection and removal of posterior osteoligamentous structures, which may contribute to postoperative axial pain, muscle atrophy, and, in selected patients, delayed spinal instability or deformity. With advances in minimally invasive and tissue-sparing spinal techniques, hemilaminectomy has gained acceptance as an alternative approach for intradural spinal tumors. By preserving the contralateral musculoligamentous structures and many of the posterior bony elements, hemilaminectomy may reduce approach-related morbidity while still allowing adequate exposure for gross total resection in appropriately selected cases. The primary aim of this study was to assess tumor occupancy ratio as a preoperative imaging factor associated with postoperative neurological improvement after hemilaminectomy for intradural spinal tumors. The secondary aim was to report the clinical outcomes, extent of resection, and perioperative morbidity in this selected cohort. We present this article in accordance with the STROBE reporting checklist (available at https://jss.amegroups.com/article/view/10.21037/jss-2026-0128/rc). Methods After approval by the Human Research Ethics Committee of Thammasat University (Medicine) (No. MTU-EC-SU-0-278/67, date March 4, 2025), we retrospectively reviewed consecutive adult patients who underwent surgery for intradural spinal tumors between January 2018 and December 2024 at a single tertiary referral center. Inclusion criteria were age 18 years or older, histologically confirmed intradural spinal tumor, tumor resection performed using a hemilaminectomy approach, and a minimum follow-up of 6 months. Exclusion criteria were intradural cystic lesions, treatment with laminectomy, and incomplete clinical or radiographic records (Figure 1). Flowchart of patient selection for hemilaminectomy. Because of the retrospective nature of the study and the use of de-identified data, written informed consent for study participation was waived. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Clinical assessment Baseline data included age, sex, body mass index (BMI), diabetes mellitus (DM), duration of neurological symptoms, and neurological status. Neurological function was graded using the modified McCormick scale (MMS) (3). For analysis, postoperative neurological improvement was defined as improvement by at least one MMS grade between the preoperative assessment and the last follow-up. Because patients with preoperative MMS grade I could not improve by at least one grade, improvement in these patients was defined as documented resolution of mild preoperative symptoms, including dysesthesia, hyperreflexia, or gait-related symptoms, by the last follow-up evaluation. Radiographic assessment Preoperative magnetic resonance imaging (MRI) was reviewed for tumor location and size. Sagittal tumor location was categorized as cervical, thoracic, or lumbar. Craniocaudal extent was recorded as the number of vertebral levels involved. On axial MRI, tumor position relative to the spinal cord was classified as anterior, lateral, or posterior. Tumors primarily located between the 10 and 2 o’clock positions were classified as anterior; those between 2 and 4 o’clock or 8 and 10 o’clock were classified as lateral; and those between 4 and 8 o’clock were classified as posterior (4). The tumor occupancy ratio was calculated on axial MRI as previously described by Ahn et al. (5), using the following: tumor occupancy ratio (%) = [(a + b)/(A + B)] × 100, where a is the transverse diameter of the tumor, b is the longitudinal diameter of the tumor, A is the transverse diameter of the intradural space, and B is the longitudinal diameter of the intradural space (Figure 2). Tumor occupancy ratio (%) was calculated as [(a + b)/(A + B)] × 100, where a and b represent the transverse and longitudinal tumor diameters, respectively, and A and B represent the corresponding diameters of the intradural space. The hemilaminectomy approach was selected based on preoperative MRI findings, including tumor location, craniocaudal extent, axial position, and the anticipated surgical corridor. This approach was generally chosen for tumors considered accessible through a unilateral posterior or posterolateral corridor without excessive spinal cord manipulation. Conventional laminectomy was preferred when wider bilateral exposure was considered necessary, particularly for extensive multilevel tumors, large intramedullary lesions, unclear surgical planes, or cases with preoperative instability or deformity. Surgical technique All procedures were performed under general anesthesia with the patient in the prone position. A midline skin incision was made over the involved segment. Unilateral subperiosteal paraspinal muscle dissection was performed on the side selected to approach the tumor while minimizing spinal cord manipulation. A hemilaminectomy was performed using a high-speed drill to obtain an adequate operative corridor. For tumors located anteriorly or laterally, the bony exposure could be extended laterally while preserving at least two-thirds of the facet joint to reduce the risk of postoperative instability. After dural opening and arachnoid dissection, cerebrospinal fluid was released at the rostral and caudal aspects of the lesion to facilitate spinal cord relaxation. Under microscopic visualization, tumor resection was performed by identifying and developing the dissection plane between the tumor and neural structures (Figure 3). Internal decompression followed by piecemeal removal was used for larger lesions. For selected ventral tumors, sectioning of the dentate ligament was performed to improve exposure and allow tumor mobilization without direct spinal cord retraction. The dura was closed primarily with 7-0 Prolene, and fibrin sealant was generally applied to reinforce the closure. Intraoperative neurophysiological monitoring was used selectively, particularly in intramedullary lesions and selected schwannomas in the cervical and lumbar regions. Intraoperative views during hemilaminectomy. (A) Development of the dissection plane at the tumor-spinal cord interface. (B) Re-expansion of the spinal cord after gross-total tumor resection. Perioperative and follow-up evaluation Perioperative variables included operative time, estimated blood loss, length of hospital stay, and perioperative complications. Patients were routinely evaluated at 4 weeks postoperatively for the early follow-up and at 6 months and longer intervals thereafter for late follow-up. Postoperative MRI was routinely obtained at approximately 3 months to evaluate residual tumor. To quantify the surgical exposure, the hemilaminectomy corridor ratio was calculated on postoperative axial MRI as the ratio of hemilaminectomy width (B) to interpedicular distance (A), where B was measured from the lateral edge of the bony resection to the base of the spinous process (Figure 4). Postoperative spinal radiographs were obtained during follow-up when patients had symptoms suggestive of possible spinal instability, such as new axial back pain, progressive deformity, or neurological deterioration. Postoperative axial MRI illustrating measurement of the hemilaminectomy corridor ratio (B/A). Hemilaminectomy width (B) is measured from the lateral edge of the bony resection to the base of the spinous process. Interpedicular distance (A) is measured between the medial cortical margins of the pedicles at the same vertebral level. MRI, magnetic resonance imaging. Statistical analysis All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (version 4.5.1; R Foundation for Statistical Computing, Vienna, Austria). The Shapiro-Wilk test was used to assess normality of continuous variables. Normally distributed data are presented as mean ± standard deviation (SD), whereas non-normally distributed variables are presented as median with range or interquartile range (IQR), as appropriate. Continuous variables were compared using the independent-samples t-test or Mann-Whitney U test. Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Variables with P<0.05 on univariate analysis were entered into a multivariable binary logistic regression model to identify independent predictors of postoperative neurological outcome. Results are reported as odds ratios (ORs) with 95% confidence intervals (CIs). A P value <0.05 was considered statistically significant. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the discriminative ability of the tumor occupancy ratio for postoperative neurological non-improvement. Discrimination was quantified using the area under the ROC curve (AUC) with its 95% CI. The optimal cutoff value was determined using the maximum Youden index, calculated as sensitivity + specificity − 1. Sensitivity and specificity at the selected cutoff value were reported. The reliability of tumor occupancy ratio measurements was assessed using the intraclass correlation coefficient (ICC). Intraobserver reliability was evaluated by comparing repeated measurements obtained by the same observer at two separate time points. Interobserver reliability was evaluated by comparing measurements obtained independently by two observers. ICC values and their corresponding 95% CIs were calculated using a two-way random-effects model with absolute agreement {ICC [2,1]}. ICC values were interpreted as follows: <0.50, poor reliability; 0.50–0.74, moderate reliability; 0.75–0.89, good reliability; and ≥0.90, excellent reliability. Results A total of 46 patients met the inclusion criteria. The mean age was 56.3±14.0 years, and 36 patients (78%) were female. The median duration of neurological symptoms prior to surgery was 3 months (range, 1 day to 120 months). The median preoperative MMS was grade II. The thoracic spine was the most common tumor location (24/46, 52.2%), followed by the cervical spine (16/46, 34.8%) and lumbar spine (6/46, 13.0%). The most common pathological diagnoses were meningioma in 22 patients (47.8%) and schwannoma in 21 patients (45.7%). One patient each had ependymoma, myxopapillary ependymoma, and hemangioblastoma. Baseline demographic, radiographic, and operative variables are summarized in Table 1. Table 1Patient characteristicsVariableValue (n=46)Age (years)56.3±14.0Female36 (78.0)BMI (kg/m2)24.2±4.2DM5 (10.1)Duration of neurological deficit (months)3 [0–120]Preoperative MMSII [I–V]Tumor location Cervical16 (34.8) Thoracic24 (52.2) Lumbar6 (13.0)Sagittal tumor extent 1 level22 (47.8) 2 levels22 (47.8) >2 levels2 (4.4)Axial tumor location Anterior19 (41.3) Lateral14 (30.4) Posterior13 (28.3)Tumor occupancy ratio (%)84.5 [42.9–98.4]Hemilaminectomy corridor ratio0.55±0.12Gross total resection44 (95.7)Operative time (min)240±112Blood loss (mL)161.8±95.5Pathology Meningioma22 (47.8) Schwannoma21 (45.7) Ependymoma1 (2.2) Myxopapillary ependymoma1 (2.2) Hemangioblastoma1 (2.2)Length of hospital stay (days)7.8±2.9Follow-up (months)27 [6–98]Postoperative neurological deterioration2 (4.3)Values are presented as number (%), median [range] or mean ± standard deviation. BMI, body mass index; DM, diabetes mellitus; MMS, modified McCormick scale. Patient characteristics Values are presented as number (%), median [range] or mean ± standard deviation. BMI, body mass index; DM, diabetes mellitus; MMS, modified McCormick scale. Gross total resection was achieved in 44 of 46 patients (96%). The mean hemilaminectomy corridor ratio was 0.55±0.12. Mean operative time was 240±112 minutes, mean estimated blood loss was 161.8±95.5 mL, and mean length of hospital stay was 7.8±2.9 days. Two patients underwent subtotal resection. In one patient with an upper thoracic intramedullary ependymoma, no safe cleavage plane was identifiable between the tumor and the spinal cord. In the other patient, who had an L3 schwannoma, the tumor could not be safely separated from an intact nerve root. Immediate postoperative complications were limited. Two patients (4.3%) experienced postoperative neurological deterioration. One had a laterally located cervical schwannoma and the other had a posteriorly located upper thoracic meningioma. Both patients gradually improved during follow-up; however, neither returned to their preoperative neurological baseline. In addition, no patient developed postoperative cerebrospinal fluid leakage, wound infection, or epidural hematoma. One patient developed recurrent meningioma requiring reoperation 5 years after the index surgery. No patient required delayed spinal instrumentation or reoperation for clinically suspected postoperative instability during follow-up. At the last follow-up (median 27 months; range, 6–98 months), 37 of 46 patients (80.4%) showed neurological improvement compared with their preoperative status. The distribution of MMS grades improved substantially from preoperative assessment to the last follow-up (Table 2). Table 2Distribution of modified McCormick scale grades before surgery and at last follow-upTime pointModified McCormick scaleIIIIIIIVVPreoperative neurological status327592Last follow-up neurological status343531Values are presented as number. Distribution of modified McCormick scale grades before surgery and at last follow-up Values are presented as number. On univariate analysis, sagittal tumor extent, tumor occupancy ratio, operative time, and blood loss were significantly associated with postoperative neurological improvement (Table 3). Patients without neurological improvement had a greater median tumor occupancy ratio than those with improvement (90.8% vs. 82.0%). Table 3Factors associated with postoperative neurological improvementVariableNeurological improvementNo neurological improvementP valueNumber37 (80.4)9 (19.6)Age (years)57.8±14.149.9±12.40.12Female29 (78.4)7 (77.8)>0.99DM5 (13.5)0 (0)0.56BMI (kg/m2)24.2±4.023.5±4.80.65Duration of neurological deficit (months)3 [2–4]4 [3–7]0.11Preoperative MMS0.10 I3 (8.1)0 (0.0) II23 (62.2)4 (44.4) III2 (5.4)3 (33.3) IV8 (21.6)1 (11.1) V1 (2.7)1 (11.1)Tumor location0.46 Cervical13 (35.1)3 (33.3) Thoracic18 (48.6)6 (66.7) Lumbar6 (16.2)0 (0)Sagittal tumor extent0.04* 1 level21 (56.8)1 (11.1) 2 levels14 (37.8)8 (88.9) >2 levels2 (5.4)0 (0)Axial tumor location0.65 Anterior16 (43.0)3 (33.3) Lateral10 (27.0)4 (44.4) Posterior11 (29.7)2 (22.2)Tumor occupancy ratio (%)82.0 [76.4–89.1]90.8 [82.9–97.6]0.02*Hemilaminectomy corridor ratio0.55±0.120.55±0.150.96Operative time (min)227 [114–275]310 [262–348]0.01*Blood loss (mL)105 [100–200]200 [150–287]0.04*Length of hospital stay (days)7 [6–9]10 [7–11]0.07Pathology0.40 Meningioma17 (45.9)5 (55.6) Schwannoma18 (48.6)3 (33.3) Ependymoma0 (0)1 (11.1) Myxopapillary ependymoma1 (2.7)0 (0) Hemangioblastoma1 (2.7)0 (0)Follow-up (months)27 [16–40]18 [6–37]0.56Values are presented as number (%), median [IQR] or mean ± standard deviation. *, P<0.05. BMI, body mass index; DM, diabetes mellitus; IQR, interquartile range; MMS, modified McCormick scale. Factors associated with postoperative neurological improvement Values are presented as number (%), median [IQR] or mean ± standard deviation. *, P<0.05. BMI, body mass index; DM, diabetes mellitus; IQR, interquartile range; MMS, modified McCormick scale. In the exploratory multivariable logistic regression model, tumor occupancy ratio was associated with postoperative neurological outcome. The OR for postoperative neurological non-improvement was 1.31 for each 1% increase in the tumor occupancy ratio (95% CI, 1.02–1.68; P=0.03) (Table 4). Given the limited number of patients without neurological improvement, this result should be interpreted cautiously. Table 4Multivariable logistic regression analysis of predictors of postoperative neurological non-improvementVariableOdds ratio (95% CI)P valueSagittal tumor extent2.29 (0.39–13.3)0.36Tumor occupancy ratio (%)1.31 (1.02–1.68)0.03*Operative time1.02 (0.99–1.05)0.06Blood loss1.00 (0.99–1.02)0.49For tumor occupancy ratio, the odds ratio is reported per 1% increase. *, P<0.05. CI, confidence interval. Multivariable logistic regression analysis of predictors of postoperative neurological non-improvement For tumor occupancy ratio, the odds ratio is reported per 1% increase. *, P<0.05. CI, confidence interval. ROC analysis demonstrated an AUC of 0.757 (95% CI, 0.569–0.918) for postoperative neurological non-improvement. The exploratory optimal cutoff value for tumor occupancy ratio was 81.15%, with a sensitivity of 100.0% and a specificity of 48.6% (Figure 5). ROC curve evaluating preoperative tumor occupancy ratio for postoperative neurological non-improvement. The area under the curve was 0.757 (95% CI, 0.569–0.918). The exploratory optimal cutoff value for tumor occupancy ratio was 81.15%, with a sensitivity of 100.0% and a specificity of 48.6%. AUC, area under the curve; CI, confidence interval; ROC, receiver operating characteristic. Measurement reliability of the tumor occupancy ratio was evaluated using ICCs. Intraobserver reliability was excellent for both observers, with ICCs of 0.910 (95% CI, 0.773–0.962) and 0.961 (95% CI, 0.925–0.976), respectively. Interobserver reliability was moderate for single measurements but improved to good reliability when averaged measurements were used, with an average-measure ICC of 0.827 (95% CI, 0.590–0.919). Discussion Surgical resection remains the cornerstone of treatment for symptomatic intradural spinal tumors, with the primary goals of achieving maximal safe resection, preserving neurological function, and maintaining spinal stability. Although conventional laminectomy provides broad posterior exposure and facilitates microsurgical tumor removal, it requires more extensive paraspinal muscle dissection and greater removal of the posterior stabilizing structures. As a result, this approach may increase the risk of postoperative axial pain, paraspinal muscle injury, and delayed spinal deformity or instability, particularly in younger patients and in the cervical spine (6,7). To minimize these approach-related morbidities, hemilaminectomy was developed as a less invasive alternative that preserves the contralateral musculoligamentous complex and a greater proportion of the posterior osseous tension band. Early microsurgical series by Eggert et al. and subsequent reports of unilateral posterior approaches established the feasibility of hemilaminectomy for spinal intradural extramedullary tumors (8,9). Since then, this technique has become increasingly accepted as an effective surgical option for selected intradural lesions. In appropriately selected cases, hemilaminectomy can provide sufficient exposure for tumor removal while potentially reducing tissue trauma and preserving postoperative spinal stability (10). One of the major concerns regarding hemilaminectomy is whether the limited operative corridor may compromise the extent of resection. Chen et al. demonstrated that satisfactory exposure can still be achieved through a unilateral approach, reporting approximate exposure widths of 1.5–2.5 cm in the cervical spine, 1.0–1.5 cm in the thoracic spine, and 1.5–2.0 cm in the lumbar spine (11). Additional exposure may be obtained, when necessary, by undercutting the base of the spinous process or partially resecting the medial facet while preserving spinal stability (12). These technical considerations highlight that hemilaminectomy is not merely a narrow exposure technique, but rather a tailored approach that can be safely expanded when required. In the present study, we quantified the operative exposure using the hemilaminectomy corridor ratio and found that a mean ratio of 0.55 was sufficient to achieve gross total resection in 96% of cases. This result is consistent with previously published hemilaminectomy series reporting gross total resection rates ranging from 90% to 100% (12-15). Among intradural extramedullary tumors, gross total resection was achieved in 43 of 44 cases (97%). The single subtotal resection in this subgroup occurred in a large lumbar schwannoma and was not attributable to insufficient exposure, but rather to dense adhesion of the tumor to a functioning motor nerve root identified intraoperatively with triggered electromyography. This finding is clinically important because it suggests that incomplete resection, when it occurs, may be more closely related to the need for neural preservation than to limitations of the unilateral surgical corridor. The applicability of hemilaminectomy to intramedullary lesions appears to be more selective (16,17). In our series, one small subpial hemangioblastoma was completely removed through a hemilaminectomy approach using a direct subpial and dorsal root entry zone-based technique. In contrast, a large ependymoma could not be safely removed through the limited corridor because no clear dissection plane could be established between the tumor and the spinal cord. In that case, only partial resection was performed, followed by reoperation through a wider laminectomy approach. These findings suggest that while hemilaminectomy may be suitable for carefully selected intramedullary lesions, particularly small or subpial tumors, it should not be considered universally applicable to all intramedullary tumors. Postoperative neurological deterioration remains one of the principal concerns in surgery for intradural spinal tumors. In the present study, postoperative neurological worsening occurred in only two patients (4.3%), and both improved during follow-up. No patient developed cerebrospinal fluid leakage, wound infection, or epidural hematoma, and no patient required further instrumentation for delayed instability. These findings compare favorably with previously published hemilaminectomy series. In a study of 901 patients treated with hemilaminectomy for spinal tumors, the reported incidences of cerebrospinal fluid leakage, wound infection, and long-term neurological deterioration were 1.6%, 1.3%, and 1.1%, respectively (12). Other contemporary series have similarly reported postoperative neurological worsening in approximately 1–4% (10,15). Furthermore, comparative evidence also supports the safety of hemilaminectomy. In a recent meta-analysis of 12 studies including 1,047 patients, hemilaminectomy was associated with significantly lower neurological deterioration than laminectomy [risk ratio (RR) 0.53, 95% CI: 0.35–0.81] (18). Taken together, these findings suggest that hemilaminectomy is a safe and potentially preferable approach for selected intradural spinal tumors. The principal finding of this study was that the tumor occupancy ratio was associated with postoperative neurological recovery in patients undergoing hemilaminectomy for intradural spinal tumors. Patients without neurological improvement had a higher median tumor occupancy ratio than those with improvement, suggesting that greater occupancy of the intradural space is associated with a lower likelihood of postoperative neurological recovery. This association is clinically reasonable. A higher tumor occupancy ratio may reflect more severe compression of the spinal cord or cauda equina, which can result in irreversible neural injury before surgical decompression. In addition, marked canal occupancy may reduce the effective working space available for microsurgical dissection, even when adequate bony exposure is achieved. Therefore, tumor occupancy ratio may represent not only the severity of preoperative neural compression but also the technical complexity of tumor resection. In the present study, ROC analysis showed acceptable discriminatory ability of tumor occupancy ratio for postoperative neurological non-improvement, with an exploratory cutoff value of 81.15%. This finding suggests that patients with a tumor occupancy ratio above this threshold may be at increased risk of limited neurological recovery after surgery. The reliability analysis also supported the reproducibility of tumor occupancy ratio measurement. Intraobserver reliability was excellent for both observers, while interobserver reliability was moderate for single measurements but was good when averaged measurements were used. These findings suggest that tumor occupancy ratio is a reasonably reproducible MRI-based parameter, particularly when repeated or averaged measurements are used. Our findings are consistent with those of Singh et al., who reported that patients with a tumor occupancy ratio below 80% had greater improvement in gait disability and better functional outcome at 1 year than those with a higher degree of canal occupancy (4). In our cohort, a similar pattern was observed, with patients in the improvement group showing a lower median tumor occupancy ratio than those without improvement. However, because the number of patients without neurological improvement was small, the proposed cutoff should be interpreted as hypothesis-generating rather than definitive, and external validation in larger independent cohorts is required. Tumor pathology should also be considered when interpreting these findings. Although this study included different histological entities, most tumors were meningiomas and schwannomas. These intradural extramedullary tumors usually compress the spinal cord or cauda equina from outside the neural parenchyma, whereas intramedullary tumors directly involve the spinal cord and may have different surgical planes, operative risks, and recovery potential. Therefore, the association between tumor occupancy ratio and neurological recovery in this study is likely applicable to solid intradural extramedullary tumors, particularly meningiomas and schwannomas. This pathological heterogeneity may have introduced selection bias and confounding because tumor biology, surgical planes, and baseline recovery potential differ between intradural extramedullary and intramedullary tumors. Several previous studies have identified other factors associated with postoperative neurological outcome, including thoracic location, ventral tumor position, duration of symptoms, and preoperative neurological status (19-21). Prognostic tools such as the spinal meningioma prognostic evaluation scale (SPES) have also been proposed to predict postoperative recovery (22). In the present study, sagittal tumor extent, operative time, and blood loss were associated with outcome on univariate analysis, but these factors lost significance after multivariable adjustment. This suggests that longer operative time and greater blood loss likely reflect technically more demanding cases rather than directly determining postoperative neurological recovery. In contrast, tumor occupancy ratio remained significant after adjustment, underscoring its potential value as a direct imaging marker of disease severity and postoperative prognosis. Intraoperative neurophysiological monitoring was used selectively in this series. In our practice, monitoring was routinely employed for intramedullary tumors and for selected schwannomas in the cervical and lumbar regions. Triggered electromyography was particularly valuable in assessing the functional significance of involved nerve roots during schwannoma resection. Although the impact of routine neurophysiological monitoring on outcomes in all intradural extramedullary tumors remains uncertain in the existing literature, it may provide additional value in cases where preservation of functional neural structures is critical (23). Overall, the present study suggests that tumor occupancy ratio may be a useful preoperative MRI-based marker of postoperative neurological recovery in patients treated with hemilaminectomy. In addition, the present study supports hemilaminectomy as an effective and safe approach for selected intradural spinal tumors, particularly solid intradural extramedullary lesions. The high rate of gross total resection, low complication rate, and favorable neurological recovery observed in this cohort are in line with the growing body of literature supporting tissue-sparing posterior approaches. Limitations This study has several limitations. First, its retrospective, single-center design introduces the possibility of selection bias and limits generalizability. Second, the sample size was relatively small, with few patients experiencing no neurological improvement, which reduced statistical power and increased the risk of overfitting; therefore, the regression findings should be considered exploratory. Third, the cohort was histologically heterogeneous, although most tumors were meningiomas or schwannomas, and differences between intradural extramedullary and intramedullary tumors may affect surgical planes and recovery. Pathology-specific subgroup analysis was not feasible; therefore, the findings are most applicable to solid intradural extramedullary tumors. Fourth, the absence of a contemporaneous laminectomy group precludes conclusions regarding the superiority of hemilaminectomy over laminectomy and limits generalizability to patients requiring wider bilateral exposure. Finally, spinal alignment and stability were not systematically assessed using serial or dynamic radiographs; therefore, subclinical postoperative deformity or instability could not be excluded. Conclusions Hemilaminectomy provided favorable surgical and neurological outcomes in selected patients with intradural spinal tumors. The tumor occupancy ratio on preoperative axial MRI was associated with postoperative neurological outcome and may serve as a simple prognostic imaging marker in patients undergoing hemilaminectomy. Supplementary The article’s supplementary files as Available at https://jss.amegroups.com/article/view/10.21037/jss-2026-0128/prf