Introduction

Lumbar disc herniation (LDH) is the leading cause of sciatica. The prevalence of symptomatic LDH has been reported to be 1–3%1. While the majority of patients can be relieved by conservative treatment, approximately 20% of patients require surgical intervention when conservative treatment fails or neurological symptoms worsen2,3. Lumbar discectomy is the most common spinal surgical procedure for the treatment of LDH. Compared with interbody fusion techniques, lumbar discectomy offers a variety of advantages, including simpler surgical procedure, shorter operation time, quicker postoperative recovery, lower surgery cost, and preservation of mobility of the operated segments4. Percutaneous endoscopic lumbar discectomy (PELD) is a novel surgical technique that can be performed with endoscopic assistance and is increasingly used in patients with LDH due to its less invasive nature, faster recovery and less blood loss5.

Many clinical studies related to PELD have been reported, including clinical outcomes, radiographic changes, incidence and risk factors of recurrent disc herniation and residual symptoms6. However, the studies focused on the subjective patient satisfaction is relatively few compared with which focused on the objective clinical outcomes. Additionally, some previous studies have examined risk factors for patient dissatisfaction after microdiscectomy and intervertebral fusion techniques such as old age, smoking, prolonged duration of clinical symptoms, obesity, preoperative leg numbness, et al., but few studies have focused on the prognostic factor for patient satisfaction after PELD7,9.

Since patient satisfaction is an important indicator of the success and outcome of lumbar spine surgery and is associated with improved quality of life10,11, to the best of our knowledge, few studies have focused on patient satisfaction and its prognostic factor for PELD, this study aims to demonstrate the patient satisfaction index (PSI) followed by PELD and to explore the prognostic factors that associated with PSI. This will help identify patients who are most likely to improve postoperatively and help doctors develop surgical strategies for PELD for LDH.

Methods

Study design

This is a retrospective study of patients who underwent PELD at a single medical center. This study was carried out in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board at our institution (2023KYLL055, 960th Hospital of PLA, Jinan, China).

Patient selection

The analysis included consecutive surgical patients undergoing PELD between November 2015 and October 2020. The inclusion criteria were: (1) patients diagnosed with LDH below L3/L4; (2) sciatica with or without low back pain and failure to conservative treatments for over 6 weeks; and (3) treated with PELD. The exclusion criteria included: (1) presence of 3 or more levels of disc herniation; (2) lumbar deformity or severe canal stenosis; (3) cauda equina syndrome; (4) patients with tumors, trauma, history of spinal surgery and (5) patients with missing data or follow-up < 24 months. All the enrolled patients were operated on by the same senior surgeon at our medical center. Informed consent was obtained from all subjects and their legal guardians.

Data collection

Baseline characteristics, clinical symptom features, neurological physical examinations, and radiographic manifestation were collected from the enrolled patients. Baseline characteristics included gender, age, height, weight, body mass index (BMI), cigarette smoking, alcohol consumption, and history of diabetes. In terms of clinical manifestation, the main clinical complaints were categorized into three types, including low back pain, radiculopathy, and low back pain combined with radiculopathy. The symptom duration was categorized into an acute subgroup (less than 3 months), a subacute subgroup (greater than 3 months and less than 1 year) and a chronic subgroup (greater than 1 year). Physical examinations included Lasegue sign, lower limb hypesthesia, and tendon flex. Preoperative radiological images included standard anterior-posterior and lateral fluoroscopy, CT, and MRI of the lumbar spine. Modic sign, annulus fibrous calcification, and lesion site in the sagittal line and horizontal line were documented.

Clinical outcomes were assessed preoperatively and at final follow-up by Visual Analog Scales (VAS) and the Oswestry Disability Index (ODI). Patient satisfaction was evaluated using the PSI proposed by the North American Spine Society (NASS), with a score of 1 (“the treatment met my expectations”) or 2 (“I did not improve as much as I had hoped, but I would undergo the same treatment for the same outcome”) defined as satisfactory and a score of 3 (“I did not improve as much as I had hoped, and I would not undergo the same treatment for the same outcome”) or 4 (“I am the same or worse than before treatment”) were defined as dissatisfactory12. Residual symptoms are defined as low back pain (LBP) or leg pain and/or numbness at the minimum of two years of follow-up. An LBP-VAS ≥ 2 was considered as having postoperative residual LBP13. Residual leg numbness symptoms are defined as a JOA score of 1 according to the Japanese Orthopedic Association (JOA) scoring system14. Patients were interviewed by telephone during the 2–3 years postoperative period.

Statistical analysis

Statistical analyses were performed using SPSS for Windows (SPSS 17; SPSS Inc.). Shapiro–Wilk’s W test was used for all continuous variables. Statistical comparisons between groups of continuous variables were made using the independent samples t-test. The chi-square test or Fisher’s exact test was used for binary variables. Wilcoxon signed-rank test was used for ranked variables. Multivariate logistic regression was used to assess the relationship between the data and the NASS index and to determine the effect of covariates. P < 0.05 was considered statistically significant.

Results

A total of 337 patients with PELD attending our department were included in the current study. Out of these 337 patients, 195 were female and 142 were male. The mean age was 46.1 ± 14.3 years old, ranging from 14 to 83 years. After a mean follow-up of 28.7 ± 3.6 months, 310 patients were satisfied (NASS score 1:268, 2:42) and 27 patients were dissatisfied (NASS score 3:13, 4:14), giving a satisfaction rate of 91.99%.

The mean age of the satisfaction group was 45.55 ± 14.33 years, which was lower than that of the dissatisfaction group (52.30 ± 12.37) years (P = 0.018). However, this difference did not exist in the subgroup (NASS score 1, 2, 3, 4). Other demographics (including gender, BMI, smoking status, alcohol drinking, and diabetes) were similar between the four groups. Clinical symptom profiles (including main clinical complaints, and duration of disease) were similar among the four groups. The rate of positive Lasegue sign and preoperative VAS was higher in the satisfaction group (P = 0.010), whereas other neurological physical examinations, including lower limb hypesthesia, Saddle area hypesthesia, muscle strength, tendon flex were not significantly different among the four groups. More contiguous double-level disc herniation (P = 0.003) and left paracentral disc herniation (P = 0.020) were detected in the dissatisfaction group, whereas no significant differences were found in other radiological data. The contiguous double-level (L3/4 and L4/5, L4/5 and L5/S1) accounts for 52.8% in dissatisfaction group and only 7.7% in satisfaction group. (Table 1). Using univariate ordinal logistic regression and considering the NASS scores 1 through 4 individually to analyze the preoperative data, the results showed that the odds of having a positive Lasegue sign, an increase in preoperative VAS, and choosing the next NASS score were 0.51 (95% confidence interval [CI] 0.28–0.94, P = 0.031) and 0.71(95% confidence interval [CI] 0.56–0.89, P = 0.004).

Table 1 Baseline characteristics and preoperative data of enrolled patients.
Full size table

The dissatisfaction group had a higher incidence of recurrent lumbar disc herniation, postoperative VAS, and incidence of residual back and leg numbness (Table 2).

Table 2 Comparison of clinical outcomes.
Full size table

Using univariate ordinal logistic regression and considering the NASS scores 1 through 4 individually to analyze the preoperative and postoperative data, the odds ratios for preoperative VAS, postoperative VAS, residual leg numbness, residual back pain was 0.62(95% confidence interval [CI] 0.43–0.89, P = 0.009), 5.37(95% confidence interval [CI] 3.44–8.38, P < 0.001), 4.89(95% confidence interval [CI] 1.82–13.23, P = 0.002), 3.21(95% confidence interval [CI] 1.33–7.75, P = 0.009) (Table 3).

Table 3 Odds of increased patient satisfaction index for all preoperative and postoperative data.
Full size table

Discussion

Patient-reported outcome measures include the VAS, ODI, SF-12, NASS-defined PSI. The VAS was used to assess back pain and leg pain; the ODI and SF-12 were used to determine functional outcomes and quality of life, and PSI was used to assess the patients` satisfaction. Improvements in ODI correlate with increases in the Patient Satisfaction Index, and for every 10-point improvement in the ODI, the odds of an improvement in patient satisfaction almost double15. Although many studies have reported ODI after PELD, few reports have focused on PSI after PELD16,17. The current study shows that patients’ PSI and its prognostic factors improve after at least 2 years of follow-up after PELD.

91.99% of patients in our study expressed satisfaction, which is similar to the previous PELD study18 and higher than the open lumbar microdiscectomy studies8,9. Although open lumbar microdiscectomy is considered the gold standard procedure for the treatment of symptomatic lumbar disc diseases19,20, its disadvantages include muscle damage, removal of the ligamentum flavum, and nerve retraction, which can lead to instability and scarring of the epidural space, resulting in residual back pain and/or leg pain in 10% or more of patients21. The rate of patient dissatisfaction with open lumbar microdiscectomy has been reported to be 30–40%7,9,22.

Advantages of PELD over open microdiscectomy include preservation of normal para-spinal muscles and minimization of the risk of postoperative epidural scar formation and instability23,24,25. Patient satisfaction with PELD has been reported to be 74–91%18. Intraoperative pain and discomfort during the foraminotomy portion of the procedure are some of the reasons for patient dissatisfaction, as epidural anesthesia or general anesthesia reduces the pain and thus improves satisfaction with PELD26. Although local anesthesia (a mixture of ropivacaine (0.15%) and lidocaine (0.67%), 20-30 ml) was used for all the patients enrolled in our study, they expressed a relatively high satisfaction rate of 91.99%. In our group, preemptive analgesia with Ketoprofen tromethamine or parecoxib sodium, and gradient local anesthesia were used to reduce intraoperative pain. Epidural injection and nerve root block are sometimes required for pain relief prior to nerve root stretching and disc treatment27, but may cause transient limb weakness and dysesthesia that gradually resolve within 24 h28. Overuse of the bipolar radiofrequency scalpel and stimulation of the rotating working channel may cause unbearable nerve stimulation, postoperative nerve root edema, and neuritis, resulting in a lower PSI, which should be addressed during the surgery.

Patients are advised to get out of bed 6 h after surgery, and early activities reduce the risk of nerve root adhesion and deep venous thrombosis. In the absence of complications, patients can be discharged from the hospital 1 day after surgery, and compared with open discectomy, PELD has a shorter hospital stay and return-to-work time, which shortens the duration of disability and thus improves patient outcome and productivity25.

In addition, timely pain feedback during local anesthesia surgery can protect the integrity of the neural structure, prevent serious neurological damage, and lower the incidence of leg numbness. Previous studies have reported that preoperative leg resting numbness7, BMI, symptom duration8, Modic type 1 change, preoperative VAS for LBP, and female sex9 are prognostic factors that reduce PSI. The relationship between patients’ satisfaction and the preoperative factor, and the impact on postoperative outcome is complex and results are conflicting. We did not find any previous factors affecting PSI, and we found that a positive Lasegue sign and a higher preoperative VAS was the prognostic factor of satisfaction, and a positive Lasegue sign was the only predictor of satisfaction among all physical examinations. Lim also found that patients with more pre-operative leg pain were more likely to be satisfied at the 2-year follow-up29, which is consistent with our findings. Yamashita found that the only independent predictor of postoperative satisfaction was the degree of subjective walking difficulty30. We found a higher prevalence of disc herniation recurrence, postoperative VAS, residual leg numbness, and residual back pain in the dissatisfaction group, and all of these factors were associated with increasing PSI except for recurrence.

Yan found a higher incidence of central disc herniation in the dissatisfaction group31. However, we found that the dissatisfaction rate was higher in left paracentral disc herniation than in central disc herniation and foraminal disc herniation, although this lesion location did not correlate with PSI. The reason for this confusing result might be the operator is dextromanual, the left lumbar disc might be ignored because of visual field and habits. We also found a significantly higher incidence of double-level disc herniation in the dissatisfaction group. PELD may not be a good option for contiguous double-level disc herniation due to a long operation time, increased fluoroscopies, and impaired patient tolerance. We advocate selective nerve root blocks to determine the symptomatic level, and then performed PELD at both levels in 13 patients with L4/5 and L5/S1 and in 9 patients with L3/4 and L4/5, with 31.8% (7/22) reporting dissatisfaction at the last follow-up. Injuries to the ligamentum flavum, the posterior longitudinal ligament, and the posterior lumbar disc structure at contiguous double-level may increase the risk of instability, resulting in an increased likelihood of recurrence or back pain.

This study has some limitations. First, this was a retrospective cohort study, and prospective studies with more patients are needed in the future. Second, the follow-up period in this study was medium duration. However, a minimum follow-up of 2 years would reflect the situation of patients` physical function and PSI. Third, although the mental health status affects PSI and depressed patients have a worse postoperative outcome, including worse physical function, pain, and disability32, we did not record patients’ mental health status because they may deny depression and provide inaccurate data due to traditional Chinese culture. Additionally, the data was obtained from a single surgeon’s registry, and the generalizability and external validity of our findings may be weakened.

Conclusion

Overall patient satisfaction with PELD was 91.99%. Positive Lasegue sign and high preoperative VAS are prognostic factors for patient satisfaction. Postoperative VAS, residual leg numbness, and back pain are risk factors for dissatisfaction. PELD may not be the first choice for contiguous double-level disc herniation.