Table 2.
General characteristics of the studies included in the systematic review and meta-analysis
| Reference | Study design | Details of participants | Tumor type | LP type and level | Follow-up (mo) | Main results | Complications (events per sample) | Newcastle-Ottawa score |
|---|---|---|---|---|---|---|---|---|
| Hao et al. [14] (2021) | Retrospective cohort study | Mean age: 45 yr; LP (n=34), LE (n=60); M (n=58), F (n=36) | Ependymoma; low-grade astrocytoma; malignant astrocytoma; cavernoma; schwannoma; angiolipoma; meningioma; neurofibroma; epidermoid cyst; enterogenous cysts; epidural simple cyst | LP with screws and connectors; cervical, thoracic, lumbar | 12 | No significant difference between the LE and LP in terms of operative time, infection, CSF leak, and length of hospitalization (p>0.05). The incidence of postoperative spinal deformity was 15.0% in the LE and 11.7% in the LP (p>0.05). LP vs. LE was associated with a similar risk of progressive deformity. For the cervical subgroup, there is significant difference (p<0.05) in postoperative spinal deformity. | LP 4/34; LE 7/60 | ***/**/*** |
| Tatter et al. [15] (2020) | Retrospective cohort study | Mean age: 52 yr; LP (n=34), LE (n=50); M (n=37), F (n=75) | Ependymoma; hemangioblastoma; meningioma; neurofibroma; schwannoma; other; intradural chordoma; hemangiopericytoma; histology inconclusive; intramedullary lipoma; neurilemmoma; neurothekeoma; dermoid | LP with miniplates; cervical | 47±36 | Two patients (2.4%) required delayed posterior fixation. The mean radiological kyphotic increase after surgery was 3.0°, which was significantly associated with LE of C2 and C3. Of these, C3 LE demonstrated independent risk association. | Not divided into groups: 1/84 | ***/**/*** |
| Kumar et al. [16] (2020) | Retrospective case series | Mean age: 35.28 yr; LP (n=14); M (n=5), F (n=9) | Schwannoma; meningioma; ependymoma; lymphoma; neurofibroma; teratoma | Flipped reposition LP technique; thoracic, lumbar | 17 (6–26) | Two patients were assessed with ASIA grade A neurology, one patient improved to grade B, whereas the other did not improve. Two patients improved from grade B to grade D, and seven patients with grades C and D improved to grade E. Fusion at the osteotomy site was seen in 92.85% (13 out of 14 cases) on one side within 6 months postoperation. Fusion was seen in all the cases within 1-year postoperation. | LP 2/14 | NA |
| Cofano et al. [17] (2020) | Retrospective cohort study | Mean age: 58 yr; LP (n=23), LE (n=226); M (n=88), F (n=161) | Schwannoma; meningioma; ependymoma; paraganglioma; neurofibroma; hemangiotelioma; hemangiopericitoma; dermoid cyst; epidermoid cyst | Laminotomy with ultrasonic aspirator; LP with titanium screws and plates; cervical, thoracic, lumbar | 48.3 | Functional outcomes were not associated with the approach. No associations were seen between surgical approach and the incidence of incidental durotomy or the degree of resection. Minimally invasive approaches (monolateral LE) resulted to be equally effective for tumor resection compared to bilateral LE or LP and no major complications were observed (e.g., revisions for dural leakage). | Not divided into groups: 19/249 | ***/**/*** |
| Song et al. [18] (2019) | Retrospective cohort study | Mean age: LP (53.3 yr), LE (49.4 yr); LP (n=27), LE (n=32); M (n=27), F (n=32) | LP group: Meningioma; Neurilemmoma; Neurofibroma; Lipoma; Ependymoma LE group: Neurilemmoma; Neurofibroma; Ependymoma; Meningioma; Lipoma; Neuroepithelial cyst |
LP with with a reconstruction plate (Fuller Inc., Beijing, China); thoracic, lumbar | 12 | Patients in the LP group had lower blood loss (LP: 281.5±130.2 mL; LE: 450.0±224.3 mL; p=0.001), shorter surgical time (LP: 141.7±26.2 min, LE: 175.3±50.4 min; p=0.003), lower volume of drainage (LP: 1,578.9±821.7 mL; LE: 2,621.2±1,351.0 mL; p=0.001), shorter drainage time (LP: 6.6±2.5 days; LE: 9.7±1.8 days; p=0.000), and a shorter hospital stay (LP: 16.9±4.9 days; LE: 21.0±4.4 days; p=0.002) compared with patients in the LE. There were significant differences of ODI between the two groups. The incidence of secondary spinal stenosis in the LP was significantly reduced (p=0.029). | LP 3/27; LE 7/32 | ***/**/*** |
| Onyia et al. [19] (2018) | Retrospective cohort study | Mean age: 40.6 yr; LP (n=24), LE (n=36); M (n=37), F (n=23) | Arachnoid cyst; hamartoma; hemangioblastoma; meningioma; neurofibroma; schwannoma; astrocytoma; ependymoma I; ependymoma II; ependymoma III; ganglion-cell tumor/paraganglioma; glomerular basal membrane; tuberculosis granuloma; plasma cell granuloma; other inflammatory lesions (nonspecific); lipoma; drop metastasis | Not reported; cervical, thoracic, lumbar | Mean LP (21), LE (13.1) | The incidence of neurologic function remaining unchanged at the end of follow-up was similar between LP and LE (12.5% vs. 11.1%). LP not more or less likely to have any better functional outcome or need for revision compared to LE in the resection of long‐segment intradural lesions. | None | ***/**/*** |
| Miyakoshi et al. [20] (2018) | Retrospective case series | Mean age: 76 yr, 60 yr; LP (n=2); M (n=1), F (n=1) | Schwannoma | Open-door LP with hydroxyapatite spacers; cervical | 72 | Both patients showed a combination of spondylotic changes in the cervical spine and stenosis at the level of the tumor. Both tumors were successfully resected through open-door LP with hydroxyapatite spacers, with the tumor located on the side of the LP. Cervical spine alignment was maintained at the final follow-up of 6 years in both cases. | LP 0/2 | NA |
| Kobayashi et al. [21] (2018) | Retrospective cohort study | Mean age: conical LP (45.3 yr), T-saw LP (48.6 yr), LE (53.8 yr); conical LP (n=21), T-saw LP (n=17), LE (n=16); M (n=30), F (n=24) | Ependymoma; schwannoma; hemangioma; meningioma; hemangioblastoma; astrocytoma; neurofibroma; others | Conical LP (the bone graft fixing by sutures); T-saw LP (posterior arches are reconstructed with sutures); thoracic | 41 | Association recovery rates did not differ significantly among the three groups. Sagittal kyphosis significantly worsened at final follow-up in LE group, but did not change significantly in LP groups. | Not reported | ***/**/*** |
| Shirosaka et al. [22] (2018) | Retrospective case series | Mean age: 41.6 yr; n=14; M (n=8), F (n=6) | Anaplastic ependymoma; hemangioblastoma; germ cell tumor; ependymoma; cavernoma; schwannoma; meningioma; subependymoma | Lift-up LP with titanium basket plates; cervical | 17.2 | In the 14 patients, a total of 32 cervical laminae were reconstructed after resection of intradural tumors. Mean operation time including resection of the tumor was 473 min (range, 250–637 min). Mean estimated blood loss including the resection of the tumor was 261 mL (range, 40–700 mL). Imaging analysis demonstrated no significant change in C2–C7 angle cervical range of motion between before and after surgery, suggesting sequential spinal stability at the base of the laminae. | None | NA |
| Emel et al. [23] (2017) | Retrospective cohort study | Mean age: 45.8 yr; LP (n=19), LE (n=23), HLE (n=7); M (n=21), F (n=26) | Schwannoma | LP with bilaterally sutured laminae using a strong non-absorbable suture; cervical, thoracic, lumbar, sacral | 61.4 | LP was better than hemi-LE and LE approaches in the length of stay with averages of 2.9±1.05 days, versus 5±2.16 days, and 9.5±7.17 days, respectively. The mean intraoperative blood loss was less in LP with an average of 395 mL and in hemi-LE with average of 435 mL versus an average of 1,130 mL in cases who underwent LE. | LP 1/19; LE 2/23; HLE 1/7 | **/*/*** |
| Siller et al. [24] (2016) | Retrospective case series | Mean age: 36.8 yr; LP (n=6), LE (n=15), HLE (n=5); M (n=12), F (n=12) | Hemangioblastoma | Not reported; cervical, thoracic, lumbar | 94.8 | No surgical approach-related outcomes reported. | Not divided into groups: 5/24 | NA |
| Montano et al. [25] (2014) | Retrospective cohort study | Mean age: 51.07 yr; LP (n=24), LE (n=19); M (n=18), F (n=25) | Schwannoma; cavernoma; ependymoma; meningioma; hemangioblastoma; low-grade astrocytoma; metastasis; fibrous tumor | LP with titanium microplates; cervical, thoracic, lumbar | 24.41 | Nine patients developed deformity or experienced a worsening of preoperative deformity at latest follow-up. Among the considered potential prognostic factors, LE (p=0.03) and evidence of pre-operative spinal deformity (p=0.009) were significantly associated with new-onset or worsening of spinal deformity. With logistic regression analysis, only the performed surgical procedure emerged as an independent prognostic factor (p=0.044). Mean hospital stay was 7.73±2.53 days in the LE and 6.62±1.83 days in the LP. Mean operation length was 171.57±41.66 min in the LE and 175.41±36.47 min in the LP. | LP 0/24; LE 2/19 | ****/**/*** |
| Parker et al. [26] (2013) | Retrospective case series | Mean age: 38.0 yr; LP (n=40); M (n=20), F (n=20) | Ganglioglioma; dermoid; ependymoma; arachnoid cyst; germ-cell tumor; lipoma; cavernous malformation; pilocytic astrocytoma; gliosis; exophytic solitary fibrous tumor; paraganglioma; schwannoma; meningioma; arteriovenous; malformation; neuropathic pain syndrome; dermal sinus tract resection and cord; untethering; myxopapillary; ependymoma; neuroenteric cyst; hemangioblastoma; intradural lipoma and tethered cord release; low-grade glioma; neurofibroma | LP with Leibinger titanium fixation plates; cervical, thoracic, lumbar | 6.4 | Successful LP was carried out in all 40 patients. Median (interquartile range) intraoperative blood loss in this series was 150 mL (75–300 mL). | LP 2/40 | NA |
| Zhou et al. [27] (2013) | Retrospective case series | Mean age: 40.7 yr; LP (n=13); M (n=5), F (n=8) | Neurilemmoma; hemangioma; meningioma; ependymoma; astrocytoma | LP with arch plate fixation; thoracic, lumbar | 9–22 | The VAS and ODI results 2 weeks following surgery and at the final follow-up examination demonstrated a significant improvement compared with the corresponding preoperative results. Radiographs demonstrated satisfactory positioning of instrumentation without any evidence of fracture, lumbar scoliosis, kyphosis, or instability. | LP 1/13 | NA |
| Menku et al. [28] (2010) | Prospective case series | Mean age: 42.3 yr; LP (n=45); M (n=25), F (n=20) | Astrocytoma; ependymoma; dermoid; lipoma; neurinoma; meningioma; cavernous angioma; syringomelia | LP with laminar flap and fixation with mini-plates; thoracic, lumbar | 12 | No patients had kyphosis and/or instability on static or dynamic plain X-ray films done at 3, 6, and 12 months. There were no cases of dural, nerve root, or spinal cord injuries attributable to laminotomy or laminar flap replacement. Bony healing was confirmed radiologically in 91% of the laminae. There was no scar tissue invasion in the spinal canal based on the MRI findings. | None | NA |
| McGirt et al. [29] (2010) | Retrospective cohort study | Mean age: 46 yr; LP (n=58), LE (n=180); M (n=108), F (130) | Ependymoma; low-grade astrocytoma; hemangioblastoma; ganglioglioma; malignant astrocytoma; metastasis; cavernoma; medulloblastoma; schwannoma; meningioma; lipoma; neurofibroma; dermoid | LP with titanium microplates; cervical, thoracic, lumbar | 24 | All baseline clinical, radiographic, and operative variables were similar between the LP and LE cohorts. LP was associated with a decreased mean length of hospitalization (5 days vs. 7 days; p=0.002) and trend of decreased incisional CSF leak (3% vs. 9%; p=0.14). Following LP vs. LE, 5 (9%) vs. 21 (12%) patients developed progressive deformity (p=0.728) a mean of 14 months after surgery. The incidence of progressive deformity was also similar between LP vs. LE in pediatric patients <18 years of age (43% vs. 36%), with preoperative scoliosis or loss of cervical/lumbar lordosis (28% vs. 22%), or with intramedullary tumors (11% vs. 11%). | LP 9/58; LE 25/180 | ***/**/*** |
| Iplikcioglu et al. [30] (2010) | Retrospective case series | Mean age: 44.3 yr; LP (n=17); M (n=8), F (n=9) | Astrocytoma; meningiomas; neurinomas; ependymomas; Ewing’s sarcoma, metastasis; abscess; hemangioblastoma; arachnoid cyst; lipoma | Open-door LP with silk sutures or titanium miniplates with screws; cervical, thoracic, lumbar | 30 | All lesions were exposed using the open-door LP technique and were successfully removed for intraspinal mass lesions. An average of 3.7 level LP was performed. Neither spinal malalignment on the coronal plane nor displacement of bone flap (LP flap) were observed on postoperative CT and MR examinations. | LP 1/17 | NA |
| Liu et al. [31] (2009) | Retrospective case series | Mean age: 36.2 yr; LP (n=24); M (n=15), F (n=9) | Neurofibroma; ependymoma; lipoma; teratoma | LP with silk or nylon sutures; thoracic, lumbar | 34.2 | Sixty-six laminae were re-implanted in 24 patients. The average operative time was 137 min (range 90–205 min). Mean blood loss was 320 mL (260–600 mL). According to postoperative X-rays, MRI, and/or CT, fusion of laminae was achieved in 3 to 6 months (mean, 4.5 months). The preoperative and postoperative Cobb angles were 16.5°±5.0° and 18.4°±11.4°, respectively, and there was no significant difference (p>0.05). Unilateral healing of the incision lines was observed in seven cases, including one with two-level laminotomy, three with three-level laminotomy, and three with four-level laminotomy. The other 17 cases underwent bilateral fusion. | LP 2/24 | NA |
| Sciubba et al. [32] (2008) | Retrospective case series | Mean age: 41 yr; LP (n=6), LE (n=26); M (n=16), F (n=16) | Astrocytoma; ependymoma; hemangioblastoma; ganglioglioma; schwannoma; subependymoma; chordoma; meningioma; neurofibroma; intradural metastatic lesions; leiomyoma; cyst | Not reported; cervical | 25.2 | Each additional level of LE performed was associated with a 3.1-fold increase in the likelihood of subsequent vertebral instability (odds ratio, 3.114; 95% confidence interval, 1.207–8.034; p=0.02). 33% (4 of 12) of the patients who had undergone a ≥3-level LE required subsequent fusion compared with 5% (1 of 20) who had undergone a ≤2-level LE (p=0.03). Four (36%) of 11 patients initially presenting with myelopathic motor disturbance required subsequent fusion compared with 1 (5%) of 21 presenting initially with myelopathic sensory or radicular symptoms (p=0.02). Age, the presence of a syrinx, intramedullary tumor, C-2 LE, C-7 LE, and LP were not associated with subsequent symptomatic instability requiring fusion. | Not divided into groups: 5/32 | NA |
| Hida et al. [33] (2006) | Retrospective case series | Mean age: 60.1 yr; LP (n=8); M (n=5), F (n=3) | Neurilemoma; hemangioma; ossification of yellow ligament; spontaneous spinal cord herniation | Transverse placement LP with titanium miniplates; thoracic, lumbar | 24 | Mean blood loss was 219 g (range, 45–600 g), and the mean duration of surgery was 3 hr 54 min (range, 2 hr 50 min–6 hr 20 min). Bony fusion of reconstructed laminae was noted at a mean of 3.2 months. No cases of spinal deformity, epidural hematoma, scar tissue in the spinal canal, or back pain were noted. | Not reported | NA |
| Casha et al. [34] (2004) | Retrospective case series | Mean age: 57 yr; LP: tumor case (n=9), non-tumor case (n=19); M (n=18), F (n=10) | Ependymoma; astrocytoma; sarcoidosis; metastatic carcinoma; schwannoma | En bloc LP with titanium miniplates; cervical | 15 | The mean angular extension–flexion displacement measured between C1 and C7 was unchanged postoperatively, with preserved mobility across LP-treated segments in all patients. The anteroposterior diameter of the spinal canal increased 3.6 mm (27.2%) postoperatively (p=0.004). In one patient an asymptomatic postoperative kyphosis developed. | 5/28 | NA |
| Asazuma et al. [35] (2004) | Retrospective cohort study | Mean age: 49.0 yr; LP (n=22), LE (n=14), HLE (n=15); M (n=25), F (n=26) | Neurinoma; meningioma; ependymoma; neurofibroma | Expansive open-door LP; cervical | 52 | Worsening of cervical curvature type was statistically less frequent with hemi-LE than LP or LE (p<0.05), although no significant difference was evident overall between pre- and postoperative cervical curvature indices or ROM in any group. Cases of C2 surgery had more frequent curvature worsening (80%) in the LP group than in other groups (p<0.05). Postoperative changes of the cervical curvature in patients with cervical spinal cord tumors were greatly influenced by the surgical method; the cervical curvature was particularly unfavorable when patients had C2 decompression. | Not reported | */*/** |
| Asazuma et al. [36] (2003) | Retrospective case series | Mean age: 55.7 yr; LP (n=7); M (n=1), F (n=6) | Neurinoma; meningioma; neurofibroma | 90° rotational LP; thoracic, lumbar | 29 | The tumor was totally removed in all cases. The JOA score improved from 62.5% to 100% (average: 81.9%). The change in the postoperative intervertebral range of motion was a 5.0° decrease to a 5.0° increase (average: 0.4° decrease). The change in lordosis varied from a 9° decrease to a 11° increase (average: 1.3° increase). Bony union was confirmed and the rotated laminae remained in situ in all seven cases. | Not reported | NA |
| Kawahara et al. [37] (1999) | Prospective case series | Mean age: 46.1 yr; LP (n=24); M (n=14), F (n=10) | Neurinoma; ependymoma; hemangioma; epidermoid cyst; meningioma | T-saw LP; thoracic, lumbar | 47 | Primary bony union in 23 patients by 6 months after surgery, and in one patient by 12 months after surgery. No complications such as postoperative spinal canal stenosis, facet arthrosis, or kyphosis were observed. | None | NA |
| Inoue et al. [38] (1996) | Retrospective cohort study | Mean age: 42.5 yr; LP (n=10), LE (n=18), HLE (n=7); M (n=22), F (n=14) | Astrocytoma; neurinoma; ganglio-neurinoma; neurofibroma; ependymoma; meningioma; lipoma; syringomyelia; cavitation of spinal cord; enterogenous cyst | Not reported; cervical | 48 | In the LE group, kyphosis of the upper cervical spine and compensatory increased lordosis of the lower cervical spine were observed in the C2 LE patients. Localized kyphosis of the spine at the cervicothoracic junction and compensatory increased lordosis of the upper cervical vertebrae were noted in the C7 LE patients. In the LP group, spinal deformities were less frequently observed, and when present, the deformity was limited to a slight increase of lordosis, even in patients who had the facetectomy. | Not reported | **/**/*** |
LP, laminoplasty; LE, laminectomy; HLE, hemilaminectomy; M, male; F, female; CSF, cerebrospinal fluid; ASIA, American Spinal Cord Injury Association; NA, not applicable; ODI, Oswestry Disability Index; VAS, Visual analog Scale; MRI, magnetic resonance imaging; CT, computed tomography; MR, magnetic resonance; ROM, range of motion; JOA, Japanese Orthopaedic Association.