Among spinal tumors, lumbar dumbbell-shaped tumors are relatively rare, and the preferred treatment is surgical resection of the tumor. In the treatment of lumbar dumbbell-shaped tumors, total resection of the tumor and preservation of spinal stability are the main concerns of the surgeon. Surgeons choose more traditional surgeries and rarely choose minimally invasive surgery, in order to balance total resection of the tumor and preservation of spinal stability. Traditional surgery often requires resection of the facet joint to adequately expose and completely remove the tumor, which results in the need to apply instrumentation to preserve spinal stability. Intraoperative instrumentation leads to additional tissue damage, prolonging the patient’s recovery time after surgery and potentially adding new symptoms. In the treatment of lumbar dumbbell-shaped tumors, we used the paravertebral approach and micro-tubular tumorectomy (PAMT) for one-stage tumor resection. The Eden classification and tumor pathological type of lumbar dumbbell-shaped tumors are decisive factors in determining surgical options. The Eden classification method is currently one of the most common methods for the clinical classification of spinal dumbbell-shaped tumors. Based on the combination of MRI and CT findings, the EDEN classification is more accurate in identifying the type of thoracolumbar dumbbell tumors and helps surgeons to select a better surgical approach (1, 2). Most patients with lumbar dumbbell-shaped tumors had a history of low back pain or leg pain, and improvement of symptoms after surgery is one of the main objectives of patients and surgeons. the VAS score and the JOA score were often used as indicators to assess the degree of pain symptoms and recovery in spine patients, so we chose these two scoring systems to evaluate and analyze the improvement of patients’ symptoms before and after surgery. We collected clinical data on 46 patients with lumbar dumbbell-shaped tumors over the last 8 years. In this group of cases, slightly different tubular spinal approaches were selected by applying Eden classification to achieve total resection of the tumor while preserving spinal stability. The feasibility and efficacy of the procedure were evaluated by analyzing the perioperative data, postoperative neurological function, and complications. On the other hand, the limitations of the minimally invasive tubular spinal approach were discussed to provide a reference for surgeons in selecting treatment options.

Traditional open surgery is the main choice for lumbar dumbbell tumors, where total tumor resection and instrument fixation is accomplished by excising the facet joints and opening the intervertebral foramen (5). For lumbar dumbbell-shaped tumors, most surgeons currently believe that minimally invasive spinal surgery (MISS) can achieve the same therapeutic effect as open surgery (6–8). Among them, the paravertebral approach and micro-tubular technique is promising for the treatment of lumbar dumbbell tumors.

Tubular retractors have different diameters sizes (14-28 mm) and are also divided into non-expandable and expandable types. The choice of tubule during surgery is based on the size, location, and pathological type of the lesion. Tubular spine surgery is performed using a tubular retractor to reach the lesion and provide sufficient space for manipulation to remove the lesion while reducing normal tissue damage without increasing the risk of nerve injury.

Zairi et al. applied a non-expandable tubule and an expandable tubule (24 mm in diameter) respectively to complete 8 cases of lumbar dumbbell-shaped tumor resection with an average maximum tumor diameter of 3-4 cm (9). Lee et al. completed GTR of subdural extramedullary schwannomas using a tubular retractor (6 cases) or unilateral hemilaminectomy (18 cases) (10). We have previously performed GTR of 56 subdural extramedullary schwannomas under the tubule (11) and also performed 31 resections of thoracic dumbbell-shaped tumors (Eden II-IV) under the micro-tubule (12). Thavara et al. reported 13 subdural tumor tubule (diameter: 22 mm-30 mm) surgeries, of which 8 (67%) completed GTR and 4 cases (33%) completed NTR, mainly because a small section of tumor tissue adhered tightly to the spinal cord and nerve roots (13). In Eden II group, the intradural nerve root of three cases was invaded by the tumor and could not be dissected; an intraoperative electrophysiological monitoring stimulator was applied to confirm that invaded nerve root by the tumor was a sensory nerve root. Then, the sensory nerve root was resected along with tumors, and there was no new symptom after surgery.

By adjusting the position and angle of the paravertebral intramuscular pathway to achieve resection of intraspinal, foraminal and paravertebral tumor tissue, microtubular surgery was performed in 46 (100%) cases of total resection of lumbar dumbbell-shaped tumors (Eden I-IV types) in one stage in Eden type II and III tumors, the establishment of 2 paravertebral muscle tubular paths (dual-tubule path) was the key to complete lumbar intraspinal and paravertebral tumor resection, and it can avoid damage to the joints facet and isthmus, avoid excessive stretch and separation of muscles, reduce muscle damage, and reduce the risk factors for spinal instability (14, 15). However, we still recommend combined surgery or mini-open surgery for tumors that may invade the spinal cord, have large tumor size, or have abundant blood supply (5, 16).

Among the lumbar spine dumbbell-shaped tumors of different Eden classifications, PAMT surgery was slightly different in terms of operative time, EBL and PHS. We found that the operative time, EBL, and PHS of Eden II were higher than those of Eden III or Eden IV. The main reasons were more time required for resecting the intradural tumor, suturing of the dura, more bleeding from the epidural plexus, and longer bed rest after surgery, respectively, in Eden II. Since the tumor in Eden IV was located in the paravertebral region, the operation did not need to deal with intraspinal lesions, the intravertebral venous plexus was not affected, and operative time and EBL were the lowest, and postoperative recovery was also the fastest. When comparing the improvement in VAS scores 12 months after PAMT, types I and II showed more improvement than types III and IV, respectively, probably because the preoperative tumor portion of the tumor located in the spinal canal compressed and irritated the nerve roots more severely in types I and II, leading to a higher chance of back or leg pain symptoms, with greater severity. For types III and IV, some patients did not have back and leg pain before surgery, so the VAS score improved less after surgery.

During the surgery on 2 schwannomas (Eden IV), we exposed the outer border of the tumor capsule and then separated the tumor along with the tumor capsule and resected the tumor in pieces, and found more bleeding from the tumor capsule and tumor tissues compared with previous surgeries. Subsequently, in three other schwannomas (Eden IV), we intentionally separated and resected the tumor along with the medial border (near the spinal canal) of tumor capsule and found that intraoperative bleeding could be significantly reduced. Therefore, we performed the intracapsular resection sequence for paravertebral tumors by first separating and resecting the tumor from the medial border (near the spinal canal) of tumor, separating towards the cranial and caudal ends, respectively, and finally reaching the lateral border of the capsule. This sequence of resection operations reduced intraoperative bleeding during the subsequent resection of paravertebral tumor tissue for lumbar dumbbell-shaped tumors (Eden II-IV). We considered that it may be that the vascular network formed by the paravertebral arteries and veins was compressed and pushed out of place by the tumor, and that removal of the tumor starting from the lateral border of the tumor reduced the compression of the vascular network, making it easier to bleed distally and that the tumor blood supplying vessels and draining veins were mainly located on the paravertebral side and more often anastomose and communicating with the normal vascular network on the paravertebral side. Therefore, intracapsular resection of the tumor from the paravertebral side may reduce intraoperative bleeding by blocking the tumor blood supplying vessels and draining veins earlier, and reducing bleeding in the normal paravertebral vascular network by compression, but there was no evidence to confirm this conclusion.

The main methods of neurophysiological monitoring in spinal surgery are EMG, SSEP, and MEP. Intraoperative evoked potential monitoring can provide important warning signals early, but each electrophysiological monitoring technology has its own limitations, such as the high false-negative rate of SSEP, especially in patients with preoperative nerve damage, the reliability of SSEP is significantly reduced (17, 18). Our combined intraoperative continuous monitoring of EMG, SSEP, and TcMEP can determine whether the sensory and motor conduction pathways are damaged, so as to adjust or suspend the surgical operation and take protective measures to prevent postoperative neurological dysfunction. We performed INMO on 317 intravertebral lesions and found that intraoperative monitoring improved in 22 cervical patients and 5 thoracic patients, and only one case of lumbar giant neurogenic tumor had improved intraoperative monitoring (19). In 46 lumbar dumbbell-shaped tumor surgeries, there was no significant difference in the postoperative JOA scores and VAS scores between the 20 patients with combined intraoperative neurophysiological monitoring and those without intraoperative electrophysiological monitoring, probably because the tumors were small in size, and it did not invade the conus medullaris, most of the tumors originated from sensory roots, and the tumor-bearing nerve functions overlapped and crossed with the functions of the nerve roots of the adjacent segments.

As with other spinal surgical methods, microtubular spine surgery carried the risk of perioperative complications, including tumor residual or recurrence, hemorrhage, neurological damage, cerebrospinal fluid leakage, thrombosis, infection, and instability. Among these complications, cerebrospinal fluid leakage was a more serious surgical complication that may cause mental disorders, headaches, dizziness, and central system infection in patients (20–22).

Thavara et al. reported 1 case (7.7%, 13 cases) of cerebrospinal fluid leakage and pseudomeningocele during tubular surgery, which resolved after 10 days of conservative treatment and 1 case of wound infection (7.7%, 13 cases) that healed after re-suturing (13). Zairi et al. reported 1 case (12.5%,8 cases) of intraoperative cerebrospinal fluid leakage that resolved after fat tissue packing and glue reinforcement (9). Xu et al. had 2 cases (11.8%, 17 cases) of cerebrospinal fluid leakage during expandable tubular surgery for thoracic meningiomas, which were cured after conservative treatment including drainage (23). Of the 46 cases in this group, 2 cases (4.3%) had postoperative cerebrospinal fluid leakage, mainly due to inadequate suture of dural incision, diabetes, and muscle atrophy and weakness, and the patients were cured after 7-9 days of conservative treatment.

Our experience in reducing cerebrospinal fluid leakage after surgery was as follows (1): 6-0 nylon thread was used to suture the dura intermittently in the dural incision, and the incision was reinforced with epidural bioprotein glue and a meningeal patch (2); using the muscle splitting technique to separate the paravertebral muscles to reduce muscle damage, which helps to reset the muscles after exiting the tubule, and to close the surgical space and surgical path (3); The muscle fascia and skin are tightly sutured (4); The drainage tube does not go through the tubular path, and it was removed as early as possible within 2 days. 5 cases (10.9%) in this group were found to have cavity effusion in the operation area by MRI re-examination, considering the large tumor volume and slow muscle retraction after resection, the effusion caused by the residual cavity disappeared after 6 months of follow-up. In addition, in tubular spine surgery the significantly reduced tissue damage and the layer-by-layer suturing of muscle, subcutaneous, and skin could reduce the cavity in the surgical area, which was helpful for healing of postoperative cerebrospinal fluid leakage and effusion in the surgical area (24).

This study is the first to comprehensively validate PAMT for different types of small lumbar dumbbell-shaped tumors, including Eden types I-VI. and PAMT can achieve complete resection of the tumor in one stage and a good prognosis. In addition, PAMT does not require additional internal fixation, and no lumbar instability was found during follow-up. The promotion of this technology is expected to reduce the financial burden on patients.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Approval for this study was granted by the Ethics committee of Fujian Medical University Union Hospital, Fuzhou, China (2020YF023-01). The patients/participants provided their written informed consent to participate in this study.

CC and RW developed the concept of the manuscript. ZL and RW conducted the analysis of the data, then RW, ZL, XX, YC composed and edited the manuscript text, figures, and tables. All authors contributed to the article and approved the submitted version.

This work was funded by Technology and Innovation Foundation of Fujian, China (grant no 2021Y0021 to RW).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer KL declared a shared parent affiliation with the authors to the handling editor at the time of review.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.