Generation Change of Practice in Spinal Surgery: Can Endoscopic Spine Surgery Expand its Indications to Fill in the Role of Conventional Open Spine Surgery in Most of Degenerative Spinal Diseases and Disc Herniations: A Study of 616 Spinal Cases 3 Years
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.299145
Source of Support: None, Conflict of Interest: None
Keywords: Cervical spine, decompression, degenerative spinal disease, endoscopic spine surgery, lumbar spine, open spine surgery, spinal fusion, thoracic spine
Degenerative spinal conditions and disc herniations are major causes of back pain, leg pain, and disability. Minimally invasive spine surgery had benefits of less hospital stay, blood loss, infection rate, and perioperative pain with comparable long-term results compared to open spinal surgery. There is an evolution of endoscopic spine surgery from initial percutaneous positioned based needle nucleotomy to transforaminal endoscopic lumbar discectomy. Transforaminal endoscopic approaches involve working around Kambin's triangle which is bound by caudal margin as vertebral body, height as traversing nerve root, and hypotenuse as exiting nerve root in the neural foramen. This working space allows docking of working channel and scope within the disc which provides stability for discectomy, and subsequently the technique evolved to docking outside the disc which does not violate the integrity of the disc but required more skilled control of the scope as there is less stability of the working channel. Interlaminar approaches were developed to address the limitations of transforaminal approaches especially in L5/S1 region which is limited by iliac crest obstruction in transforaminal approaches as interlaminar approaches enter through the ligamentum flavum in midline, they also have the advantage of addressing more medial located disc and other pathologies. Earlier phase of interlaminar endoscopic approaches was focused on discectomy with subsequent expansion to include posterior decompression procedures in lumbar spine. As one gets more confident with interlaminar approaches, decompression procedure extending to thoracolumbar spine and cervical spine.,, The recent development of interest in lumbar endoscopic spinal fusion had allowed further expansion in the indications of degenerative spinal conditions in the practice of endoscopy in spine surgery., However, there is a paucity of literature documenting the effect on the expansion of endoscopic spinal surgery in surgical practice and its overall effect on patients' outcome. Our objective is to evaluate the effect of expansion of endoscopic practice in degenerative spinal conditions and disc herniations on patients' outcome as we shifted from first-generation endoscopic discectomy practice to second-generation decompression practice and finally third generation of endoscopic spinal fusion practice and each generation change encompasses expansion of a new technique together with previous generation surgeries.
Between 2016 and 2019, a total of 616 patients underwent operations for degenerative spinal conditions and disc herniations by senior author HSK's team. Informed consent was obtained from each patient prior to surgery. Three generations of cases reflected authors' evolving experience in spinal endoscopic practice with initial endoscopic discectomy practice progressing to decompression and finally involving fusion. All patients had significant disabling clinical symptoms and had failed a minimum 6 weeks of conservative treatments prior to surgical treatment.
134 patients from the first generation (Generation 1) of spinal procedures done in January to June 2016 consisted of all open spinal procedures, except surgeries involving only endoscopic procedures, for most, if not all, lumbar discectomy procedures [Figure 1] and [Figure 2]. Generation 1 focused on endoscopic discectomy procedures.
254 patients from January to June 2018 consisted of the second generation (Generation 2 of spinal procedures comprising open microscopic spinal fusion and endoscopic decompression and discectomy procedures for most, if not all, cervical thoracic and lumbar regions) [Figure 3] and [Figure 4]. Generation 2 had endoscopic decompression procedures in addition to endoscopic discectomy.
228 patients from January to June 2019 consisted of the third generation (Generation 3) of spinal procedures which involved endoscopic procedures in most of cervical, thoracic, and lumbar decompression, discectomy, and endoscopic spinal fusion procedures except a small percentage of open procedures [Figure 5]. Generation 3 had endoscopic fusion procedures in addition to endoscopic discectomy and decompression.
Over the three generations of procedures, we included adult patients more than 18 years of age, presenting with a diagnosis of grade two and below spondylolisthesis, spinal stenosis, prolapsed intervertebral disc, degenerative disc disease, and spinal instability for lumbar region. For endoscopy cases involving cervical and thoracic regions, we treated patients who presented with radiculopathy and/or myelopathy with radiological evidence of degenerative spinal stenosis or disc herniations. We excluded patients with tumor, trauma, deformity, and infection. Data were collected and anonymized by HSK and analysis was performed by study collaborator, PHW who had no access to patients' personal information.
List of all surgeries
First generation open procedures
Anterior cervical discectomy and fusion, anterior cervical disc replacement, posterior cervical laminoplasty, thoracic and lumbar laminectomy with or without discectomy, anterior and lateral approach thoracolumbar interbody fusion, and microscopic transforaminal lumbar interbody fusion.
First Generation: Endoscopic Procedures—Endoscopic Discectomy Only [Figure 1] and [Figure 2].
Transforaminal endoscopic lumbar discectomy and interlaminar endoscopic lumbar discectomy.,,
Second Generation Endoscopic Procedures: Inclusion of Endoscopic Decompression in addition toFirst Generation of Endoscopic Procedures [Figure 3] and [Figure 4].
Posterior endoscopic cervical foraminotomy and discectomy and thoracic/lumbar endoscopic unilateral laminotomy for bilateral decompression which was known as percutaneous endoscopic stenotic thoracic or lumbar decompression (T/L E-ULBD also known as PESTD/PESLD), interlaminar contralateral endoscopic lumbar foraminotomy.,,,
Third Generation of Endoscopy: Inclusion of Endoscopic Lumbar Interbody Fusion in addition to Second Generation of Endoscopic Procedures [Figure 5].
Posterolateral endoscopic transforaminal lumbar interbody fusion.,
Collection of data
Overall, 616 patients underwent operation over the 3 separate intervals of 6 months, of which 26 are thoracic patients, 51 are cervical patients, and 539 are lumbar patients. We analyzed the baseline parameters of age, sex, diagnosis, and number of lumbar segments of surgery performed per patient, and follow-up period is recorded. The operative details involving the nature of operations, perioperative complications, recurrence, and results of operations were evaluated. Clinical outcomes were measured with visual analog scale; Oswestry disability index was evaluated at time of preoperative, postoperative 1 week, 6 months, and final follow-up. MacNab's criteria for pain relief were recorded at final follow-up.
Clinical data was analyzed with SPSS version 18 statistical analysis software (IBM Corporation, New York). The continuous variables were expressed as mean and standard deviation (SD). The pairedt-test is used for comparison clinical parameters of visual analog scale and Oswestry disability index within each group, and analysis of variance (ANOVA) test was used for comparison of statistical difference between each generation of endoscopic practice. Bonferroni correction was used for multiple comparisons. A P value of <0.05 considered significant.
Baseline demographics and nature of operations
We had 134 cases in Generation 1, 254 cases in Generation 2, and 228 cases in Generation 3 group. The distribution of cases is shown in [Figure 6] and [Figure 7]. There is no statistical difference between basic demographics and levels of operated on in each group. In the Generation 1 group, 24.62% of the operations are done using open technique and 75.38% endoscopic procedures. There are no cervical and thoracic endoscopic surgeries in Generation 1 group. Among the endoscopic discectomy, there is a small number of 14.2% of percutaneous endoscopic stenotic lumbar decompression of ipsilateral side only with discectomy. The open surgery mean operation time per level was 83.3 min (range: 40–120 min), while the endoscopic surgery mean operation time per level was 36 min (range: 25–50 min).
In the Generation 2 group, a further increase in the proportion of endoscopic surgery gives a proportion of 8.65% open surgeries. There is a concordant increase in endoscopic procedures involving decompression, which were percutaneous endoscopic posterior cervical discectomy, percutaneous endoscopic thoracic decompression, percutaneous endoscopic stenotic lumbar decompression with corresponding decrease in number of percutaneous endoscopic transforaminal lumbar discectomy and percutaneous endoscopic interlaminar lumbar discectomy procedures. The open surgery mean operation time per level was 101 min (range: 60–120 min), while the endoscopic surgery mean operation time per level was 58 min (range: 35–100) min.
In Generation 3 group, the inclusion of endoscopic interbody fusion in the practice had further decreased the percentage of open spinal surgeries to a total of 2.64% of the surgeries. 97.4% of degenerative cases were treated by endoscopic spinal surgery and the newly introduced endoscopic spinal fusion procedure: Endoscopic transforaminal lumbar interbody fusion (ETLIF) took up 21.5% of the endoscopic practice. The open surgery mean operation time per level was 135 min (range: 120–150 min), while the endoscopic surgery mean operation time per level open was 62 min (range: 35–140).
In Generation 1, 2, and 3, mean age was 53.6 (34–82), 59.5 (27–86), and 61.5 (28–87) years, mean follow-up was 50.5 (48–53) months, 14.6 (12–17) months, 9 (6–11) months, and a number of segments were 1.16 (1–3), 1.35 (1–4), and 1.33 (1–4), respectively [Table 1].
The percentage of surgeries performed is shown in [Figure 6] and [Figure 7].
Details of the three generations of procedures are shown in [Table 1].
Clinical outcome analysis
In terms of clinical outcomes, compared to preoperative scores, the 1 week, 6 months, and final follow-up VAS and ODI scores all statistically significantly improved in each generation using paired t-test [Table 2]. ANOVA statistical analysis with Bonferroni correction of the 3 generations of endoscopy was performed for clinical outcomes of VAS and ODI [Table 3]. There was a statistically significant improvement in terms of both VAS and ODI for Generation 2 as compared to Generation 1 and in terms of only VAS for Generation 3 as compared to 1. There was no statistically significant difference in clinical outcomes between Generations 2 and 3.
In terms of MacNab's Criteria of pain improvement clinical outcome, in Generation 1 group, there were 39 cases of excellent outcomes, 78 cases of good outcomes, 16 with fair outcomes, and 1 with poor outcomes, making a good-to-excellent rate of outcome as 87.3%.
In Generation 2 group, there were 83 cases of excellent outcomes, 161 cases of good outcomes, 9 with fair outcomes, and 1 with poor outcomes, making a good-to-excellent rate of outcome as 96.0%. In Generation 3 group, there were 76 cases of excellent outcomes, 149 cases of good outcomes, 3 with fair outcomes, and 0 poor outcomes, making a good-to-excellent rate of outcome as 98.7%. Comparing across the 3 generations, clinical outcomes improved in each generation for MacNab's Criteria but it did not reach statistical significance.
[Figure 8] shows a summary of clinical outcomes over the three generations of endoscopic spinal practice.
Overall complications rate was 8.2%, 9%, and 4% in Generation 1, 2, and 3, respectively. [Table 4] shows the detail breakdown and sequelae of complications of three generations of endoscopic spinal practice.
In an aging population, there is an increasing need for minimally invasive surgical options to reduce duration and intensity of perioperative pain, inpatient hospital stay, and blood loss and infection rate., Endoscopic spine surgery techniques, instruments, optics, and scientific understanding had improved in a rapid phase in the past decade in line with the goal of achieving surgical targets of decompression and fusion., There is a steep learning curve in endoscopic spine surgery, as it is a subspecialty in spine which is not commonly covered in residency training of both neurosurgical and orthopedic curricula.,,, Kim et al. described that the evolution of endoscopic spine surgery gives rise to 4 generations of endoscopic spine procedure with the first generation focused mainly on transforaminal approaches, second generation introduced interlaminar approaches, third generation had included stenosis decompression, and cervical and thoracic procedures and contralateral approaches,,,, while fourth generation is the current generation which is evolving and involves spinal fusion procedures. We started our endoscopic practice being trained in both transforaminal and interlaminal discectomy. Hence, unlike their description of 4 generations, our practice pattern had evolved into three generations of practice in the treatment of degenerative spinal conditions and disc herniations which were, namely, endoscopic discectomy, endoscopic decompression, and endoscopic fusion. In treatment of disc herniations, many open microscopic discectomies were replaced by endoscopic spinal discectomy in our first generation. Open anterior cervical discectomy and fusion/artificial disc replacement were replaced by percutaneous endoscopic posterior cervical decompression, and open laminectomy and decompression in thoracolumbar spine procedures were replaced by endoscopic thoracic and lumbar decompression in second generation. Majority of open lumbar spinal fusion procedures were replaced by endoscopic spinal fusion in third generation. As spinal stenosis and instability had a higher incidence than prolapsed disc, the endoscopic decompression and fusion procedure percentages also increased over the generations., We performed increasing percentage of endoscopic procedures as compared to open surgery with 75.4%, 91.4%, and 97.4% of the cases in Generation 1, 2, and 3, respectively.
Patient and disease factors played a significant role in endoscopic spine surgery practice. The incidence of comorbidities, risk of general anesthesia, complications secondary to immobility after painful procedure, and blood loss and infection increased as patients are aging. All these factors made minimally invasive surgery a favored option., We can explore more expansion of indications in area of sinuvertebral and basivertebral neuropathy, tumor debulking, and treatment of spinal infection. However, one needs to be aware of the limits of endoscopic practice, when patients present with congenital deformity of spine and degenerative deformity with stiff spine; these pathological features can lead to disorientation of the anatomical structures which can lead to higher risk of complications if one is not experienced in dealing with such deformity. Minimally invasive decompression such as endoscopy can be clinically effective in deformity and low-grade spondylolisthesis patients but patients with significant scoliosis and lateral listhesis have a higher revision rate. In such situations, open spinal surgery and perhaps deformity correction may be better options. Anatomical variations such as furcal nerve, low lying nerve root in foramen, and conjoined nerve roots are possible though uncommon situations, which can be encountered during transforaminal and fusion procedures. Careful preoperative evaluation can lower the risk of complications and perhaps relative contraindications to endoscopic surgery. Fortunately, majority of the degenerative spinal conditions and disc herniations are not associated with anatomical variants or deformity, which are ideal indications for endoscopic practice.
Surgical and technical factors varied due to the experience of the surgeon in dealing with endoscopic spine surgery. Wang et al. showed that there was significant learning curve of endoscopic spine surgery to perform transforaminal lumbar discectomy for disk herniation with varied outcomes based on experience. However, Xu et al. demonstrated, the rate of attaining interlaminar proficiency was reasonably fast without significant outcome differences though the experienced surgeon tended to preserve bone and soft tissue more. Our authors found in our series that there was significant clinical outcomes improvement as we moved from Generation 1 to 2 and 3. But there were no significant differences in clinical outcomes in Generation 2 and 3 even though we had broadened our indications for endoscopic spine surgery to involve more complex cases that required fusion. This might be confounded by the fact that there were no significant differences in fusion and decompression procedures in low-grade spondylolisthesis and instability., Understanding the anatomical features of the landmarks is important in orientation during endoscopic surgery. Step by step guidance by experts and structured teaching might be helpful at the beginning of practice. There are techniques to resolve complications in endoscopic spine surgery such as incidental durotomy. However, it is advised that the operating endoscopic surgeon be familiar with open procedures in case of complications occurred during spinal endoscopic surgery that required open surgical conversion.
Despite the steep learning curve, there was a significant and consistently less operation time for endoscopic spine surgery as compared to the open surgery across all 3 generations of endoscopic practice. In our opinion, that is due to less time consumed in soft tissue and bone dissections and the additional time required for closure of these tissues. As spinal endoscope has small diameter typically ranging from 7-13mm outer diameter with its surgical vision at the distal tip of the endoscope directly over the tissue and working area, less dissection than open surgery where tissues are retracted to allow visualization of the surgical field, allowing the ease of closure of wound to decrease the time for surgery and more importantly achieving its surgical goals with less damage to tissues.
Limitations of the study
The authors' institution focusses on spine surgeries, especially endoscopic spine surgery; hence, there is a selection bias in both the authors' as well as the patients' choice of the type of surgery. The first generation of endoscopic surgery was done with more basic endoscopic instruments as compared to the second and third generation as industries developed better endoscopic instruments over time confounding the results of the surgeries performed. The follow-up period varied between each generation as the authors continued to follow up without discharges for all the post-surgical patients. Final follow-up data were collected at the latest clinical visit which the authors felt it reflected the most updated status of the patient. Preoperative comorbidities were present in some of the patients; however, analysis of the comorbidities data was not performed which might be a confounder in this study.
Generation change of increasing percentage of endoscopic surgeries and expansion of endoscopic spinal indications over open surgeries in degenerative spinal conditions and disc herniations are possible as a surgeon gets more experience with endoscopic spine surgery producing a good clinical outcome.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed. The IRB number is NR-IRB 2020-012.
Hyeun Sung Kim and Pang Hung Wu contributed equally to the article as first co-authors.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7], [Figure 8]
[Table 1], [Table 2], [Table 3], [Table 4]