Evaluation of percutaneous endoscopic lumbar discectomy in treatment of obese adolescents with lumbar disc herniation (cid:0) A retrospective study

Background: Obese patients are at risk of complications such as poor wound healing and increased infection rates after spinal surgery. Percutaneous endoscopic lumbar discectomy (PELD) has advantages over conventional open surgery in the treatment of obese adult patients with lumbar disc herniation (LDH) because it can decrease perioperative complications and enhance satisfaction degrees of patients. However, no clinical studies have evaluated the ecacy of PELD in obese adolescents with LDH. This study aimed to evaluate the ecacy of PELD in the treatment of obese ALDH. Methods: We retrospectively collected clinical data from 208 patients with single-segment ALDH who underwent PELD treatment in our hospital between January 2015 and December 2019. According to the body mass index classication standard of obesity for adolescents in our country, the patients were divided into obese and non-obese groups (control group). Based on the preoperative baseline data of the two groups, propensity score matching was performed to select patients from the two groups for the comparative study. Perioperative data included operative time, intraoperative blood loss and length of postoperative hospitalization. The visual analog scale (VAS), Oswestry disability index (ODI) and modied MacNab criteria were recorded as the main indicators of the surgical outcome, and the recurrence rate and incidence of complications were recorded as the minor indicators. Results: The obese and control groups included 45 patients each after 1:1 propensity score matching. Both groups showed improvements in VAS and ODI scores after surgery and at each follow-up time point (p < 0.05). However, there was no signicant statistical difference in VAS and ODI scores between the two groups at each follow-up time point (p >0.05). In addition, the differences in operative time, intraoperative blood loss, length of postoperative hospitalization, incidence of complications, and recurrence rate were not statistically signicant between the two groups (p >0.05). At the nal follow-up, there was no signicant difference in the excellent and good rate of MacNab classication between the two groups (p >0.05). Conclusion: PELD is a safe and effective minimally

obese adults could reduce the length of surgical incision and the occurrence of complications such as infection, thereby improving clinical e cacy and patient satisfaction. However, no studies have evaluated the short-or mid-term e cacy of PELD in obese adolescents with LDH. Therefore, we conducted a retrospective follow-up study of obese adolescents with LDH who underwent PELD at our hospital for 1 to 5 years. Our study focused on the outcomes and safety of PELD in the treatment of obese patients with ALDH.

Patients
We retrospectively analyzed patients with ALDH who underwent PELD in our hospital between January 2015 and December 2019. The diagnosis of LDH was con rmed based on medical history, physical examination, magnetic resonance imaging (MRI) of the lumbar intervertebral disc, and surgical records. The inclusion criteria included patients with the following: (1) age ≤ 21 years, (2) LDH that was clearly diagnosed by MRI in a single segmental spine, and the clinical manifestations of patients were consistent with imaging ndings, (3) no signi cant improvement of symptoms after 6 weeks of standard conservative treatment, and (4) surgery performed by a senior surgeon (with > 2 years of experience in spinal surgery). The inclusion criteria included patients with the following: (1) lumbar disc herniation in multi-segmental spine; (2) other previous lumbar spine surgeries; (3) spinal fractures, lumbar spondylolisthesis, spinal tumors, scoliosis, or some other orthopedic specialty diseases such as spinal tuberculosis or infection; or (4) severe medical diseases and mental disorders.
A total of 208 patients with ALDH met the above criteria. According to the body mass index (BMI) classi cation criteria for obese adolescents under 18 years of age in our country (Fig. 1) and the BMI classi cation criteria for obese adolescents over 18 years old (BMI ≥ 28kg/m 2 ), these patients were divided into obese and non-obese groups (control group). To ensure the balance and comparability of the clinical data between the two groups and reduce the confounding bias, propensity score matching (PSM) was performed based on the preoperative baseline data.
Propensity score matching (PSM) To reduce the impact of potential confounding factors, the patients in the obese and control groups were matched with propensity scores by using all available baseline data. PSM was established through a multivariate logistic regression model, which considered the following variables: age, sex, trauma history, herniated segment, herniated type, preoperative visual analog scale (VAS) score, and Oswestry disability index (ODI) score. Matching was performed using the nearest-neighbor matching algorithm (caliper width 0.25, standard deviation of the logit score) with a 1:1 ratio without replacement. Standardized mean differences (SMD) were compared before and after PSM to evaluate the matching of balanced potential confounders for the two study groups [9]. Covariates with a standardized difference of < 0.15, in absolute value, were considered satisfactorily balanced.
Surgical Procedure PELD After the target disc was recon rmed using C-arm X-ray uoroscopy, the puncture point was marked. The puncture needle was sent to the target position, and a guidewire was inserted. A skin incision approximately 0.8 cm in length was made around the puncture site. Then, the dilating sleeve was placed along the guidewire step-by-step. Finally, we placed the intervertebral foraminoscope after recon rming the correct position using C-arm X-ray uoroscopy. Before we formally started the core step of surgery, lumbar discography was performed to identify the brous ring crack. The diseased nucleus pulposus tissue was removed using an endoscope, and the dural sac and nerve roots were thoroughly decompressed and loosened until spontaneous pulsation was achieved. Complete hemostasis was achieved with the assistance of bipolar radiofrequency, and the brous ring crack was repaired. The endoscope was moved along the nerve root, and the operative eld was carefully examined to con rm that no residual compression was present. After endoscope removal, the incision was closed using sutures. The operation was completed without a wound drainage tube.

Postoperative treatment
In addition to the temporary use of antibiotics during surgery, the patients did not routinely use antibiotics after the operation. The patients were regularly administered symptomatic treatments for nutritional nerves, dehydration, and analgesia. Patients wearing waist circumference were allowed to leave the sickbed 6 h postoperatively. Patients were instructed to rest in bed for 1 month after surgery and to wear waist circumference during daily activities, and to avoid standing for long periods of time or performing physical labor within 3 months of surgery.
Patients were followed up in the outpatient clinic at 3 months, 6 months, and 1 year after surgery. Afterwards, it was recommended that patients be followed at least once per year. The re-examination items mainly included computed tomography or MRI of the lumbar spine. Clinical Evaluation and Follow-Up Demographic and perioperative clinical data were obtained by consulting surgical records and electronic medical records. In addition, we collected data on disease recurrence and complications through telephone and outpatient follow-up. The demographic data included sex, age, BMI, duration of symptoms, and follow-up time. Perioperative data included operative time, intraoperative blood loss and length of postoperative hospitalization. The major indicators for evaluating the e cacy of PELD included the VAS score [10] for lower back and leg pain, the ODI score [11], and modi ed MacNab criteria [12] obtained at the last follow-up timepoint. The minor indicators included the complication and recurrence rates obtained from the medical records or the last follow-up timepoint. The VAS score was collected before surgery, the rst day after surgery, discharge). The judgment of recurrence was mainly based on the lumbar MRI diagnosis of the patient after the onset of symptoms or the second surgical treatment.

Statistical Analysis
All data analyses were performed using SPSS version 26.0 (IBM, Armonk, New York, USA). Categorical data are expressed as the number of cases or percentages, and the differences between the two groups were analyzed using the χ 2 test or Fisher's exact probability method. Normally distributed measurement data are expressed as the mean ± standard deviation, and Student's t-test was used to compare differences between the two groups. One-way analyses of variance were used for the comparison of VAS or ODI scores within the same group at different times, and the least signi cant difference method was used for pairwise comparison.
Moreover, measurement data that were not normally distributed were expressed by the median and interquartile range, and analyzed using the Mann-Whitney U test to observe difference between the two groups. Differences were considered to be statistically signi cant at p < 0.05.

Results
A total of 208 subjects were included in the study, including 57 patients in the obese group and 151 patients in the control group, and 174 patients (83.7%) were followed up. There were 9 patients (15.8%) and 25 patients The basic preoperative characteristics of each group before and after PSM are shown in Table 1. There were signi cant differences in preoperative VAS scores between the two groups before PSM (p < 0.05). However, the balance of each variable was signi cantly improved after PSM (|SMD|<0.15, Table 1), and the baseline data of the two groups were consistent (p > 0.05, Table 1). Table 1 Comparison of preoperative basic characteristics between the two groups before and after 1:1 PSM A comparison of preoperative data between the two groups is shown in Table 2. There was no signi cant difference between the two groups in terms of operative time, intraoperative blood loss, and length of postoperative hospitalization (p > 0.05).  In terms of the VAS score changes during the follow-up period (Fig. 2, Table 3), the VAS scores on the rst postoperative day, 3 months postoperatively, and at the nal follow-up timepoint were signi cantly lower than the preoperative VAS scores in both the obese and control groups (p < 0.05). Regarding the ODI score changes during the follow-up period (Fig. 3, Table 3), the ODI scores at 3 months postoperatively and at the nal follow-up timepoint were signi cantly lower than the preoperative VAS scores in both groups (p < 0.05). There was no signi cant difference in the postoperative VAS score and ODI score between the two groups ( Fig. 2 and Fig. 3).
Based on the minor indicators of clinical outcomes of the two groups, there was no signi cant difference in the overall incidence rate of complications between the two groups (20% vs. 15.6%, p = 0.581). There were no intraoperative complications in either group. However, one patient had immediate postoperative complications. The patient felt that the symptoms of preoperative lower back pain were basically relieved after returning to the ward, but the pain of the right lower limb was not relieved completely, with the symptom of hyperesthesia on the lateral side of right lower leg but no neurological symptoms or cauda equina syndrome.

Discussion
According to the characteristics of the spinal growth and development process, the vertebral epiphyseal cartilage is normally completely fused with the vertebral body around 21 years of age [14]; hence, 21 years of age was de ned as the upper age limit for patients with ALDH in this study. However, the lower age limit has not been clearly de ned. Raghu et al. [15] reported that LDH patients under 12 years of age were very rare, and the youngest patient in this study was a 10-year-old girl. BMI is one of the commonly used standards for measuring human body fat. A BMI ≥ 28kg/cm 2 is used as the diagnostic criterion for obese patients over 18 years of age in our country. However, since adolescents under 18 years of age are still growing and developing, the above criteria for obesity are not suitable for the adolescent population. Therefore, we classi ed obese adolescent patients according to the obesity classi cation criteria for adolescents and children under 18 years of age in this study.
Over the past 30 years, obesity has become a global epidemic that threatens public health. The prevalence of suggested that obesity is associated with the incidence of various diseases such as cardiovascular disease, cancer, and bone and joint diseases [17][18][19], and there is evidence that obesity is also an important risk factors for the onset of ALDH [4,20]. Obesity applies an excessive load to the intervertebral disc, and it leads to an abnormal in ammatory response and endocrine regulation in the human body, which eventually resulted in accelerated degeneration or damage to intervertebral disc. Obese adolescents accounted for 27.4% of patients with ALDH patients in our study, so we should closely observe obese people with such diseases whether they are adults or adolescents.
spinal surgery in obesity is a challenging endeavor for many reasons, including anesthesia, intravenous access, positioning, and wound exposure. Most spine surgeons would agree that surgical intervention is di cult in this population. Before the emergence of minimally invasive spinal surgery, this group of patients usually required longer surgical incisions than the general population to fully expose the surgical area during lumbar disc herniation, but many complications may occur, including wound infection or poor healing, which affected the surgical satisfaction of these individuals and surgical e cacy [21,22]. With the development of minimally invasive techniques and the requirement of medical apparatus, PELD has emerged and is widely used in adult and adolescent spinal surgery. Numerous clinical studies have suggested that PELD not only has comparable e cacy compared with conventional spinal surgery, but also has the advantages of reduced blood loss, reduced tissue destruction, and faster postoperative recovery [23][24][25]. Moreover, several studies [26,27] have suggested that the application of PELD in obese adults with LDH could reduce the incidence of complications such as wound infection or poor healing. We assume that the decreased incidence of complication in obese patients with PELD can be attributed to technological breakthroughs in the deep surgical eld; operations require less time and thus, decrease the chances of contamination and paraspinal muscles trauma. In addition, PELD is usually performed under local anesthesia; therefore, general anesthesiarelated adverse events can be avoided effectively. However, no studies have evaluated the e cacy of PELD in obese patients with ALDH.
Compared with adult patients with LDH, patients with ALDH also have the following characteristics. First, the surgical methods is cautiously selected. Regardless of the surgical method used, we need to focus on minimizing the surgical impact on spinal growth and development and the possibility of secondary adjacent segment degeneration or recurrent disc herniation after surgery [28]. Compared with open surgery, PELD minimizes structural damage in the normal spine, such as muscles and facet joints, and reduces the recurrence rate of postoperative iatrogenic instability. Second, the scope of surgical resection is limited.
Whether it is PELD or open surgery, the scope of discectomy in adolescents should be controlled to create conditions for the regeneration of intervertebral discs [29] and maximize the retention of the remaining disc function. Finally, it is necessary to consider whether the growth and development of the adolescent spine affects the e cacy of surgery. Gulati et al. [30] believed that the growth of adolescent spine may affect the e cacy of surgery. However, our study showed that in obese or non-obese patients, the excellent and good rate of surgery was more than 93%. Given that PELD is more complicated than traditional open spinal surgery and there may be complications such as residual nucleus pulposus, intraspinal hematoma, infection, and more [31], it often requires a longer learning curve to maximize its performance [32].
Based on the basic characteristics of the two groups before surgery, we found that there were no signi cant differences in age, sex, history of trauma, segment and type of herniation, severity of preoperative lower back and leg pain, and ODI scores after PSM, so the comparison between the two groups was more reliable. In terms of perioperative data such as operative time, intraoperative blood loss and length of postoperative hospitalization, there was no statistical difference between the obese and control groups ( Table 3, P > 0.05), indicating that compared with non-obese patients, obese patients with ALDH may accept PELD without signi cant di culties. We believe that because PELD caused less surgical trauma and required no wound drainage tubes, it can effectively shorten the postoperative bedridden recovery time, reduce the occurrence of bed-related complications and hospitalization costs, improve postoperative quality of life, and help patients return to their normal life or work faster, which is in line with the current concept of enhanced recovery after surgery in the eld of spinal surgery [33]. However, this was contrary to the results reported in some studies that microdiscectomy or open surgery increased the amount of intraoperative blood loss and length of hospitalization in obese adults [7,34], indicating that PELD was advantageous in the treatment of obese patients with LDH from the other side.
Based on the main indicators of clinical outcome, the postoperative VAS and ODI scores of both groups were signi cantly lower than the preoperative scores (Fig. 2, Fig. 3, p < 0.05), and there was no difference in VAS and ODI scores between the two groups at each follow-up time (  (Table 3, p > 0.05). However, Meredith et al. [35] believed that obesity was a strong and independent predictor of recurrence in patients who underwent lumbar discectomy, which was also inconsistent with our results. We believe that the following two aspects may explain this difference. First, the follow-up time in our study was not su ciently long, and there was a certain proportion of patients who were lost to follow-up. Second, most adolescents were highly compliant and may adopt more strict postoperative rehabilitation plans under the supervision of their parents. Moreover, the postoperative complications of PELD in our study were mainly manifested as recurring lower back pain or insigni cant relief of postoperative pain symptoms, which existed in both the obesity and control groups. However, pain in most patients had little effect on their daily lives or required analgesic intervention. We believe that this nding may be related to abnormal expression of pro-in ammatory cytokines in the blood of patients after surgery or adjacent joint diseases [36,37].
However, here are some limitations to our study. First, a retrospective single-center study design was used. The number of patients included and the follow-up time were limited, and there were some patients that we could not obtain follow-up data on, which lessened the accuracy of the present cohort. Therefore, a prospective randomized controlled trial with a larger sample size is needed to con rm our results.

Conclusion
In summary, this retrospective study suggested that PELD is a safe and effective minimally invasive technique for the treatment of obese patients with ALDH. The improvement of pain and disability in obese patients with ALDH were comparable with that in non-obese patients with ALDH, and there was no difference in intraoperative blood loss, operative time, and length of postoperative hospitalization between these groups.
Obese patients with ALDH who undergo PELD do not face a greater risk of complications and recurrence, indicating that PELD has good short-and mid-term effects. However, a prospective randomized controlled trial with a larger sample size is still needed to further con rm our results.

Declarations
Competing interests The authors declare they have no competing or con ict of interest.
Ethics approval and consent to participate The study was conducted in accordance with the Declaration of Helsinki and was approved by the Peking University Third Hospital Medical Science Research Ethics Committee. Because of the retrospective nature of the study, patient consent for inclusion was waived.

Consent for publication
Not applicable.

Funding
The study was not funded