Abstract
Background
Microendoscopic discectomy for lumbar disc herniation has been shown to be as effective as traditional microdiscectomy or open discectomy in terms of clinical outcomes such as pain relief, and it is less invasive. Nevertheless, the reoperation rate for microendoscopic discectomy compared with microdiscectomy or open discectomy remains unclear, possibly due to difficulties in conducting follow-up of sufficient duration and in obtaining information about reoperation in other facilities.
Questions/purposes
(1) What is the rate of reoperation after microendoscopic discectomy for primary lumbar disc herniation on a large scale at a median of 4 years postoperatively? (2) Is there any difference in revision rate at a median of 4 years and within 90 days postoperatively based on surgical method?
Methods
We conducted a retrospective, comparative study of adult patients who underwent microendoscopic discectomy or microdiscectomy or open discectomy for lumbar disc herniation from April 2008 to October 2017 and who were followed until October 2020 using a commercially available administrative claims database from JMDC Inc. This claims-based database provided information on individual patients collected across multiple hospitals, which improved the accuracy of postoperative reoperation rates. We included 3961 patients who received microendoscopic discectomy or microdiscectomy or open discectomy between April 2008 and October 2017 in the JMDC claims database. After applying exclusion criteria, 50% (1968 of 3961) of patients were eligible for this study. Propensity score–weighted analyses were conducted in 646 patients in the microendoscopic discectomy group and in 1322 in the microdiscectomy or open discectomy group, with a median (IQR) of 4 years (3 to 6) of follow-up in both groups. Mean patient age was 42 ± 12 years in the microendoscopic discectomy group and 43 ± 12 years in the microdiscectomy or open discectomy group. Males accounted for 78% (505 of 646) of patients in the microendoscopic discectomy group and 79% (1050 of 1322) of patients in microdiscectomy or open discectomy group. The proportion of patients with diabetes mellitus in the microendoscopic discectomy group (10% [64 of 646]) was less than in the microdiscectomy or open discectomy group (15% [195 of 1322]). The primary outcome was Kaplan-Meier survivorship free from any type of additional lumbar spine surgery at a median of 4 years after the index surgery. The secondary outcome was survival probability using the Kaplan-Meier method with endpoints of any type of reoperation within 90 days after the index surgery. To determine which procedure had the higher revision rate, we conducted propensity score overlap weighting analysis, which controlled for potential confounding variables such as age, sex, comorbidities, and type of hospital as well as Cox proportional hazard models to estimate HRs and 95% confidence intervals (CIs).
Results
The 5-year cumulative reoperation rate was 12% (95% CI 9% to 15%) in the microendoscopic discectomy group and 7% (95% CI 6% to 9%) in the microdiscectomy or open discectomy group. After controlling for potentially confounding variables like age, sex, and diabetes mellitus, the microendoscopic discectomy group had a higher reoperation risk than the microdiscectomy or open discectomy group (weighted HR 1.57 [95% CI 1.14 to 2.16]; p = 0.004). Within 90 days of the index surgery, after controlling for potentially confounding variables like age, sex, and diabetes mellitus, we found no difference between the microendoscopic discectomy group and microdiscectomy or open discectomy group in terms of risk of reoperation (weighted HR 1.38 [95% CI 0.68 to 2.79]; p = 0.38).
Conclusion
Given the higher reoperation risk with microendoscopic discectomy compared with microdiscectomy or open discectomy at a median of 4 years of follow-up, surgeons should select microdiscectomy or open discectomy, despite the current popularity of microendoscopic discectomy. The revision risk of microendoscopic discectomy compared with microdiscectomy or open discectomy in the long term remains unclear. Future large, prospective, multicenter cohort studies with long-term follow-up are needed to confirm the association between microendoscopic discectomy and risk of reoperation.
Level of Evidence
Level Ⅲ, therapeutic study.
Introduction
Lumbar disc herniation is one of the most common indications for lumbar spine surgery. Although management initially involves nonoperative treatment for most patients, surgical treatment may be considered if pain persists [31]. The standard procedure for lumbar disc herniation is microdiscectomy or open discectomy, a surgical technique to remove the portion of intervertebral disc compressing the nerve root under direct vision or microscope magnification [24]. Microendoscopic discectomy was initially developed as an alternative discectomy method using a tubular retractor and endoscope, which minimized manipulation of the surrounding tissue, and it has recently been applied to treat various spinal diseases [21, 26, 27]. Microendoscopic discectomy has yielded clinical outcomes, such as pain relief and other patient-reported outcomes, similar to microdiscectomy or open discectomy; in addition, it offers a lower risk of postoperative systemic complications and a shorter length of stay [21, 30].
Previous studies have suggested that between 5% and 16% of patients who undergo lumbar disc herniation surgery will have a reoperation at some point [2, 7, 9, 10, 13, 29]. Reoperation is associated with longer procedure time and length of stay, increased postoperative narcotic use, and higher risk of an additional reoperation [1, 22]. Although several studies, including randomized controlled trials, have compared the risk of reoperation of microendoscopic discectomy with that of microdiscectomy or open discectomy, results have been inconsistent [15, 24, 30], and study limitations such as insufficient length of follow-up and exclusion of reoperation in other facilities have meant that no consensus on this topic has been reached [15, 24, 30]. In contrast, only a few studies—in Korea and the United States—have reported reoperation rates after lumbar disc herniation surgery using population-based nationwide data [7, 9, 10, 29], and, to our knowledge, just one study has compared reoperation rates for different surgical techniques, including microdiscectomy or open discectomy and microendoscopic discectomy [10].
We therefore asked: (1) What is the rate of reoperation after microendoscopic discectomy for primary lumbar disc herniation on a large scale at a median of 4 years postoperatively? (2) Is there any difference in the revision rate at a median of 4 years and within 90 days postoperatively based on surgical method?
Patients and Methods
Study Design and Setting
This retrospective, comparative study used a commercially available administrative claims database from JMDC Inc [20]. The JMDC claims database has included outpatient, inpatient, and pharmacy insurance claims data for approximately 9.6 million people in Japan, mainly employed workers and their family members, since 2005. The upper age limit in the JMDC database is 74 years because the database does not cover claims associated with the health insurance system for elderly people, which is available for everyone in the Japanese healthcare system 74 years and older. The database includes patient characteristics, pharmacy data, clinical diagnoses coded using the ICD-10, medical procedures coded using Japan-specific standardized procedure codes, and specific health checkup records. Individual patients can be followed through a unique encrypted identifier, even if the patient visits or is hospitalized at multiple medical institutions. The need to obtain individual informed patient consent is waived because all patient information is anonymized; patient privacy and confidentiality are preserved because all individual identifiers were removed when the database was created. We used this claims-based database because it is known to be a major source of Japanese medical claims data and has been used in many research projects [18, 20, 25]. Further, it provides information on individual patients across multiple hospitals, which enabled us to investigate postoperative reoperation rates in patients accurately.
Participants
Patients with a diagnosis of lumbar disc herniation between 2005 and 2017 were identified. The inclusion criterion was patients who underwent microendoscopic discectomy, microdiscectomy, or open discectomy at a single level for lumbar disc herniation between April 2008 and October 2017. The date of microendoscopic discectomy or microdiscectomy or open discectomy was set as the index date. Exclusion criteria were patient age younger than 18 years at the index date, patients with a medical history of less than 2 years before and 90 days after the index date in the database, a history of lumbar surgery in the 2 years preceding the index date, simultaneous receipt of additional lumbar procedures along with microendoscopic discectomy or microdiscectomy or open discectomy, and concomitant disease (such as lumbar spinal fracture, spinal infection, or spinal tumor) (Supplementary Table 1; http://links.lww.com/CORR/A876).
Between 2005 and 2017, a total of 170,905 patients were entered into the JMDC claims database with a diagnosis of lumbar disc herniation. Of these, we considered those patients who underwent microendoscopic discectomy, microdiscectomy, or open discectomy for lumbar disc herniation between April 2008 and October 2017 as potentially eligible, making a total of 2% (3961 of 170,905) eligible on that basis. We excluded the remaining 98% (166,944 of 170,905) of patients because 96% (163,960 of 170,905) did not receive any lumbar spine surgery for lumbar disc herniation, 1% (2418 of 170,905) received other spine surgeries (such as, posterior fusion surgery) for lumbar disc herniation, and 0.3% (566 of 170,905) did not have data on procedure day in the database. A further 50% (1962 of 3961) of eligible patients were excluded because patient age was younger than 18 years at the index date, the medical history in the database was less than 2 years before the index date, a history of lumbar spine surgery before the index date, or the diagnosis included lumbar spinal fracture or spinal infection. We then set a minimum follow-up duration of 90 days because excluding patients with a follow-up duration of less than 2 years from this study could have led to selection bias. The proportion of patients with less than 2 years of follow-up was 11% (218 of 1968): 10% (67 of 646) in the microendoscopic discectomy group and 11% (151 of 1322) in the microdiscectomy or open discectomy group. As a result, a further 1% (31 of 3961: 1% [12 of 1346] in the microendoscopic discectomy group and 1% [19 of 2615] in the microdiscectomy or open discectomy group) of patients were excluded before the minimum study follow-up, leaving 50% (1968 of 3961) for analysis here.
Descriptive Data
A total of 1968 patients (646 in the microendoscopic discectomy group and 1322 in the microdiscectomy or open discectomy group) were included in the analysis (Fig. 1). Median (IQR) follow-up duration was 4 years (3 to 6) in the entire group: 4 years (3 to 6) in the microendoscopic discectomy group and 4 years (3 to 6) in the microdiscectomy or open discectomy group. Compared with the microdiscectomy or open discectomy group, patients in the microendoscopic discectomy group were younger, had a lower BMI, and a lower proportion of patients with diabetes mellitus. After propensity score overlap weighting adjustment, both groups were balanced in all baseline characteristics (Table 1). The covariate balance for all variables was achieved after propensity score overlap weighting (Supplementary Fig. 1; http://links.lww.com/CORR/A877).
Fig. 1.
This figure is a flow chart of the selection of study participants; LDH = lumbar disc herniation; MED = microendoscopic discectomy; MD = microdiscectomy; OD = open discectomy.
Table 1.
Baseline characteristics before and after propensity score weighting
| Characteristic | Before weighting | After weightinga | ||||
| MED (n = 646) | MD/OD (n = 1322) | Standardized mean differenceb | MED (n = 431) | MD/OD (n = 431) | Standardized mean differenceb | |
| Men | 78 (505) | 79 (1050) | 0.03 | 79 (339) | 79 (339) | < 0.01 |
| Age in years | 42 ± 12 | 43 ± 12 | 0.09 | 43 ± 12 | 43 ± 12 | 0.03 |
| Age category in years | 0.10 | 0.05 | ||||
| ≤ 29 | 15 (95) | 14 (184) | 14 (61) | 15 (65) | ||
| 30-39 | 28 (178) | 25 (336) | 27 (117) | 27 (114) | ||
| 40-49 | 30 (194) | 28 (375) | 30 (130) | 29 (123) | ||
| 50-59 | 21 (133) | 23 (310) | 21 (91) | 22 (94) | ||
| 60-75 | 7 (46) | 9 (118) | 8 (33) | 8 (35) | ||
| BMI in kg/m2 | 24 ± 3 | 24 ± 4 | 0.11 | 24 ± 3 | 24 ± 4 | 0.08 |
| Missing | 44 (282) | 38 (497) | 43 (186) | 39 (166) | ||
| Diabetes mellitus | 10 (64) | 15 (195) | 0.15 | 11 (48) | 11 (48) | < 0.01 |
| Osteoporosis | 3 (17) | 2 (26) | 0.04 | 2 (10) | 2 (10) | < 0.01 |
| Charlson comorbidity index | 0.08 | 0.06 | ||||
| 0 | 77 (499) | 77 (1015) | 77 (332) | 78 (336) | ||
| 1 | 16 (106) | 15 (198) | 16 (71) | 15 (63) | ||
| ≥ 2 | 6 (41) | 8 (110) | 6 (28) | 8 (33) | ||
| Teaching hospital | 5 (35) | 4 (54) | 0.06 | 5 (21) | 5 (21) | < 0.01 |
Data presented as % (n) or mean ± SD.
Frequency numbers were rounded to integers based on weight. Weighted to adjust for age, gender, diabetes mellitus, osteoporosis, Charlson comorbidity index, and hospital type.
Standardized mean difference measured the normalized difference between the means of each covariate between the two groups; MED = microendoscopic discectomy; MD = microdiscectomy; OD = open discectomy.
Exposure
The main exposure variable was microendoscopic discectomy versus microdiscectomy or open discectomy, whichever occurred earlier. If some patients underwent both microendoscopic discectomy and microdiscectomy or open discectomy during follow-up, we regarded the earlier surgery as the primary surgery and the later one as a reoperation.
Outcome Measures
The primary outcome was the incidence of any type of second lumbar spine surgery before the end of the follow-up period. This was necessary because details on surgical level were not identifiable in the claims data [9, 10] (Supplementary Table 2; http://links.lww.com/CORR/A878). The secondary outcome was the incidence of any lumbar surgery within 90 days after the index date because, in general, the reason for reoperation before versus after 90 days may differ [10]. Reoperation before 90 days may be related to recurrent disc herniation, insufficient decompression, or wrong level of operation, whereas reoperation after 90 days may be caused by recurrent disc herniation, instability, and exacerbation of degenerative disease. Furthermore, we performed an additional analysis that used debridement for postoperative complications, including surgical site infection and postoperative epidural hematoma, as an extra secondary outcome in addition to reoperation within 90 days (Supplementary Table 3; http://links.lww.com/CORR/A879).
Other Variables
The following data were extracted from the database: age, gender, BMI, diabetes mellitus, osteoporosis, Charlson comorbidity index, and hospital type. The Charlson comorbidity index was calculated by algorithms developed by Quan et al. [23] and categorized as 0, 1, and ≥ 2. We included only comorbidities diagnosed within 1 year before the index date. BMI measured as part of a health checkup conducted within 1 year before the index date was used. When multiple BMI data obtained at different timepoints were available, the data obtained closest to the index date were adopted.
Ethical Approval
We obtained ethical review board approval for this study.
Statistical Analysis
Kaplan-Meier estimates of survivorship free from any type of subsequent lumbar spine surgery were used to report the cumulative incidence with 95% confidence intervals (CIs).
To determine which procedure had the higher revision rate, we used a propensity score overlap weighting analysis to adjust for measured confounders such as age, sex, comorbidities, and hospital type between the microendoscopic discectomy and microdiscectomy or open discectomy groups [14]. After propensity score overlap weighting, we used Cox proportional hazards models to estimate HRs for outcomes. The 95% CIs in weighted analyses were calculated using robust variance estimators. Because propensity score overlap weighting accounts for confounding, we performed no further adjustment in the Cox models. Interactions between exposure variables and time were verified by the proportionality assumption. We did not perform a competing-risks analysis because death occurred in 0.2% of patients.
We created a propensity score for receiving microendoscopic discectomy using a logistic regression model. Based on our clinical experience, we included the following covariates: demographics (age and gender), preexisting comorbidities (diabetes mellitus, osteoporosis, and Charlson comorbidity index), and hospital type [10, 16, 17, 19] (Supplementary Table 4; http://links.lww.com/CORR/A880). We did not include BMI because these data were missing for a large proportion of patients. Although we did not adjust for BMI, we show these values in both groups to illustrate the degree of balance in this value before and after adjustment. Overlap weighting is a technique that assigns less weight to those with an outlier propensity score, and those with a propensity score close to 0.5 make the largest contribution to the effect estimate [14]. Specifically, treated patients are weighted by 1 – propensity score and untreated patients are weighted by propensity score. The target of inference is the average treatment effect in the overlap population. Compared with propensity score matching, overlap weighting retains all patients and can therefore improve precision [11]. Compared with the inverse probability of treatment weighting, overlap weighting improves balance and precision by reducing the influence of extreme values in the tails of the propensity score distribution [6]. Standardized differences were used to evaluate the balance of covariates between the microendoscopic discectomy and microdiscectomy or open discectomy groups in the weighted cohorts. A standardized difference of less than 0.10 indicated good balance [4].
To test the robustness of the results, we conducted a sensitivity analysis in which the study population was restricted to those with a look-back period of at least 3 years to ensure these patients did not have a history of lumbar spine surgery before the index surgery was performed.
All statistical analyses were conducted using SAS version 9.4 (SAS Institute) and R 4.1.1 (R Foundation).
Results
What Is the Reoperation Rate After Microendoscopic Discectomy for Primary Lumbar Disc Herniation?
The propensity score–weighted Kaplan-Meier curve showed that 1-year, 3-year, 5-year, and 7-year cumulative reoperation rates were 4% (95% CI 2% to 5%), 8% (95% CI 6% to 10%), 12% (95% CI 9% to 15%), and 15% (95% CI 11% to 19%), respectively, in the microendoscopic discectomy group, and 3% (95% CI 2% to 4%), 6% (95% CI 4% to 7%), 7% (95% CI 6% to 9%), and 9% (95% CI 7% to 10%), respectively, in the microdiscectomy or open discectomy group (Fig. 2).
Fig. 2.
This propensity score–adjusted Kaplan-Meier plot shows the cumulative incidence of reoperation during the follow-up period; MED = microendoscopic discectomy; MD = microdiscectomy; OD = open discectomy. A color image accompanies the online version of this article.
Is There Any Difference in Revision Rate Based on Surgical Method?
The reoperation incidence in the microendoscopic discectomy group was higher than that in the microdiscectomy or open discectomy group at a median of 4 years potsoperatively (weighted HR 1.57 [95% CI 1.14 to 2.16]; p = 0.004) (Table 2). Sixty-four percent (98 of 153) of patients underwent discectomy as the revision surgery, and approximately half of the patients in both groups received the same surgery as their primary surgery. Fifteen percent (10 of 65) of patients in the microendoscopic discectomy group and 28% (25 of 88) of patients in the microdiscectomy or open discectomy group received fusion surgeries as reoperation (Table 3). Similarly, after controlling for potentially confounding variables like age and diabetes mellitus, we found no difference between the microendoscopic discectomy group and the microdiscectomy or open discectomy groups in terms of the reoperation risk within 90 days of the index procedure (weighted HR 1.38 [95% CI 0.68 to 2.79]; p = 0.38) (Table 4). Additionally, when debridement was added to reoperations within 90 days of the index procedure, no difference was found between the two groups (weighted HR 1.05 [95% CI 0.69 to 1.60]; p = 0.81) (Supplementary Table 5; http://links.lww.com/CORR/A881). Generally, consistent results were obtained in the sensitivity analysis, in which study participants were limited to those with records extending more than 3 years before the index surgery. Both groups were balanced in all baseline characteristics in the sensitivity analysis (Supplementary Table 6; http://links.lww.com/CORR/A882). There was no difference in reoperation rate between the two groups at a median of 4 years postoperatively (Supplementary Table 7; http://links.lww.com/CORR/A883) and within 90 days of the index procedure (Supplementary Table 8; http://links.lww.com/CORR/A884). Furthermore, when debridement was added to reoperations within 90 days of the index procedure, no difference was found between the two groups (Supplementary Table 9; http://links.lww.com/CORR/A885).
Table 2.
Reoperation event counts and HR for MED and MD/OD
| Surgery type | Patients, n | Person-years | Reoperation, n | Weighteda HR (95% CI) | p value |
| MED | 646 | 2681 | 65 | 1.57 (1.14-2.16) | 0.004 |
| MD/OD | 1322 | 5754 | 89 | Reference |
Weighted to adjust for age, gender, diabetes mellitus, osteoporosis, Charlson comorbidity index, and hospital type; MED = microendoscopic discectomy; MD = microdiscectomy; OD = open discectomy.
Table 3.
Types of revision surgery
| Surgery type | MED (n = 65) | MD/OD (n = 88) |
| MD/OD | 26 (17) | 48 (42) |
| MED | 48 (31) | 8 (7) |
| Percutaneous discectomy | 0 | 1 (1) |
| Anterior lumbar interbody fusion | 2 (1) | 0 |
| Posterior or posterolateral lumbar fusion | 0 | 1 (1) |
| Posterior lumbar interbody fusion | 14 (9) | 26 (23) |
| Anterior and posterior interbody fusion | 0 | 1 (1) |
| Laminectomy | 3 (2) | 6 (5) |
| Laminoplasty | 2 (1) | 9 (8) |
| Microendoscopic laminectomy | 3 (2) | 0 |
| Microendoscopic laminoplasty | 3 (2) | 0 |
Data presented as % (n); MED = microendoscopic discectomy; MD = microdiscectomy; OD = open discectomy.
Table 4.
Reoperation event counts and HR within 90 days postoperatively for MED and MD/OD
| Surgery type | Patients, n | Person-years | Reoperation, n | Weighteda HR (95% CI) | p value |
| MED | 646 | 157 | 13 | 1.38 (0.68 to 2.79) | 0.38 |
| MD/OD | 1322 | 322 | 19 | Reference |
Weighted to adjust for age, gender, diabetes mellitus, osteoporosis, Charlson comorbidity index, and type of hospital; MED= microendoscopic discectomy; MD = microdiscectomy; OD = open discectomy.
Discussion
Microendoscopic discectomy has been introduced for the surgical management of symptomatic lumbar disc herniation to reduce muscle and soft tissue damage. A few studies have demonstrated that microendoscopic discectomy was associated with comparable pain relief and lower risk of systemic complications and shorter hospital stay than microdiscectomy or open discectomy [21, 30]. However, previous studies examining the risk of reoperation after microendoscopic discectomy compared with microdiscectomy or open discectomy have yielded inconsistent findings [24, 30]. In this large, claims-based database study comparing reoperation rates for patients receiving microendoscopic discectomy or microdiscectomy or open discectomy, we found that microendoscopic discectomy was associated with a higher reoperation risk than microdiscectomy or open discectomy at a median of 4 years postoperatively. Within 90 days after the index surgery, in contrast, we saw no difference in the incidence of reoperation between the two surgical methods.
Limitations
Our study has several limitations. First, our results might have been affected by unmeasured confounders such as patient occupation, surgeon experience, or herniation type; radiologic information, such as the degree of degenerative change or presence of Modic change (pathological change in the end plate of the neighboring disc); and disc height. However, using propensity score overlap weighting, we could effectively balance other measured confounders. Additionally, although a previous study showed that high BMI was associated with an increased risk of recurrent lumbar disc herniation [16], we could not adjust for this because of a lack of BMI data in approximately 40% of participants. However, we believe this did not influence our results because among patients for whom data were available, the mean BMI after adjustment for other confounders was almost fully balanced between the two groups. Although this does not guarantee that other unmeasured confounders such as patient occupation, surgeon experience, and herniation type were balanced after weighting, we did not consider that these unmeasured confounders eliminated the association between microendoscopic discectomy and reoperation risk.
Second, our median of 4 years of follow-up might not be long enough to assess the risk of revision surgery accurately [9]. We set a minimum follow-up duration of 90 days and included patients with less than 2 years of follow-up duration to avoid selection bias caused by excluding these patients. However, there was no difference in follow-up duration (median 4 years in the microendoscopic discectomy group and 4 years in the microdiscectomy or open discectomy group) and the rate of the patients with less than 2-year follow-up duration (10% in the microendoscopic discectomy group and 11% in the microdiscectomy or open discectomy group) between the two groups. Third, the reoperation rate included any type of lumbar surgery, and thus may have differed from that at the index level. In practical terms, however, this information is likely to be as meaningful to patients as the recurrence rate at the same level. Furthermore, it appears almost impossible to deny the possibility that primary lumbar disc herniation surgery may have caused degeneration at the adjacent level. Additionally, as this study is based on administrative claims of ICD-10 codes, standard disease codes, and procedure codes to assess the reoperation rate, there are weaknesses associated with the accuracy and completeness of administrative claims coding. Fourth, although we used a 2-year look-back period, which was the period before the index procedure, it is not possible to ensure that the index surgery for lumbar disc herniation was the primary procedure. Although our sensitivity analysis restricting patients to those with a 3-year look-back period showed no difference in reoperation rate between microendoscopic discectomy and microdiscectomy or open discectomy, this did not overturn our main findings. Thus, we do not think this substantially affected the results. Fifth, our results may have limited generalizability regarding age because our database is based on health insurance claims for employed workers and their family members, and the mean age of patients with lumbar disc herniation in our study population was approximately 43 years. Nevertheless, the fact that most lumbar disc herniation surgeries were performed on a working-age population might render our study informative for spinal surgeons [12]. Lastly, we could not assess patient-reported outcomes such as the Oswestry disability index because this type of outcomes information was not available in the database. Furtheremore, we could not investigate length of stay, readmission, and medication cost because the database does not allow determination of what types of diseases cause patients to receive medications and/or visit the emergency department or obtain accurate data about length of stay.
What Is the Reoperation Rate After Microendoscopic Discectomy for Primary Lumbar Disc Herniation?
Studies have reported the occurrence of reoperation after primary lumbar discectomy using nationwide data. A study using claims data from Korea reported a 5-year reoperation risk of 13% [10], and in two studies from the United States, one found a 4-year reoperation risk of 12% [7] and the other, a 7-year reoperation incidence of 6% [29]. Our results—12% in the microendoscopic discectomy group and 7% in the microdiscectomy or open discectomy group at 5 years after the index surgery—are consistent with the results of these previous studies. These findings may be helpful for establishing accurate expectations for patients scheduled for lumbar disc herniation surgery.
Is There Any Difference in Revision Rate Based on Surgical Method?
Although several studies have compared microendoscopic discectomy with microdiscectomy or open discectomy with regard to clinical outcomes and perioperative complications [10, 24, 30], very few studies have compared long-term reoperation rates between microendoscopic discectomy and microdiscectomy or open discectomy. Unlike one of these previous studies [10], our data showed that microendoscopic discectomy had a higher reoperation risk at a median of 4 years of follow-up than microdiscectomy or open discectomy. We speculate that this difference is because of our definition of reoperation. Unlike the previous study [10], our definition did not include debridement for perioperative complications. Microendoscopic discectomy was associated with a lower risk of surgical site infection than microdiscectomy or open discectomy [24]. Thus, adding debridement to the definition of reoperations could make the relative reoperation risk of microendoscopic discectomy lower than that for reoperations without debridement. Most reoperations after primary surgery for lumbar disc herniation were attributable to recurrent or new herniations, especially after microendoscopic discectomy [2, 5]. According to previous randomized controlled trials, microendoscopic discectomy was associated with a higher risk of recurrent herniation [3, 24, 28]. One advantage of microdiscectomy or open discectomy is that it enables surgeons to maintain three-dimensional vision, in contrast to microendoscopic discectomy, which provides two-dimensional vision. The poor depth perception during microendoscopic discectomy may increase the incidence of dural tears and root injuries and decrease the chance of finding and removing disc fragments [8, 24]. Furthermore, the restricted field of work induced by a tubular retractor may be related to residual fragments [28]. Lastly, microendoscopic discectomy was associated with less decompression than microdiscectomy or open discectomy, which could make recurrent herniation more symptomatic and result in reoperations. These drawbacks of microendoscopic discectomy may account for the higher reoperation rates we observed with this procedure than with microdiscectomy or open discectomy. Based on our findings, surgeons should reconsider which method to choose in spite of the recent popularity of microendoscopic discectomy. Future studies with long-term follow-up might be warranted to substantiate these findings.
Conclusion
Our comparative study using a large insurance claims–based database demonstrated that the microendoscopic discectomy group had increased reoperation rates over a median 4-year duration than the microdiscectomy or open discectomy group. Although microendoscopic discectomy has some advantages—for instance, it is less invasive to surrounding soft tissues—our findings should make surgeons reconsider the indications for this procedure by weighing its higher reoperation risk. However, given our insufficient follow-up duration, these findings need to be confirmed by a large, multicenter, prospective cohort study with at least a 10-year follow-up period.
Acknowledgment
We thank Guy Harris DO, of DMC Corp (www.dmed.co.jp), for his support with the writing of the manuscript.
Footnotes
One of the authors (KK) certifies receipt of personal payments or benefits, during the study period, in an amount of USD 100,001 to USD 1,000,000 from Eisai Co, Ltd, Kyowa Kirin Co Ltd, Sumitomo Dainippon Pharma Co Ltd, Pfizer Inc, Stella Pharma Corporation, CMIC Co Ltd, and Suntory Beverage & Food Ltd; in an amount of USD 10,000 to USD 100,000 from Mitsubishi Corporation, Real World Data Co Ltd, LEBER Inc, JMDC Inc, and Shin Nippon Biomedical Laboratories Ltd; and in an amount of less than USD 10,000 from Kaken Pharmaceutical Co Ltd and Advanced Medical Care Inc; executive compensation from Cancer Intelligence Care Systems Inc; and honoraria from Mitsubishi Chemical Holdings Corporation, Mitsubishi Corporation, Pharma Business Academy, and Toppan Inc.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Ethical approval for this study was obtained from Kyoto University Graduate School and Faculty of Medicine (number R3017).
This work was performed at Kyoto University, Kyoto, Japan.
Contributor Information
Soichiro Masuda, Email: masuda.soichiro.77a@st.kyoto-u.ac.jp.
Toshiki Fukasawa, Email: fukasawa.toshiki.4a@kyoto-u.ac.jp.
Masato Takeuchi, Email: takeuchi.masato.3c@kyoto-u.ac.jp.
Shunsuke Fujibayashi, Email: shfuji@kuhp.kyoto-u.ac.jp.
Bungo Otsuki, Email: bungo@kuhp.kyoto-u.ac.jp.
Koichi Murata, Email: kchm@kuhp.kyoto-u.ac.jp.
Takayoshi Shimizu, Email: takayosh@kuhp.kyoto-u.ac.jp.
Shuichi Matsuda, Email: smat522@kuhp.kyoto-u.ac.jp.
References
- 1.Ahn J, Tabaraee E, Bohl DD, Aboushaala K, Singh K. Primary versus revision single-level minimally invasive lumbar discectomy: analysis of clinical outcomes and narcotic utilization. Spine (Phila Pa 1976) . 2015;40:E1025-30. [DOI] [PubMed] [Google Scholar]
- 2.Aihara T, Endo K, Sawaji Y, et al. Five-year reoperation rates and causes for reoperations following lumbar microendoscopic discectomy and decompression. Spine (Phila Pa 1976). 2020;45:71-77. [DOI] [PubMed] [Google Scholar]
- 3.Arts MP, Brand R, van den Akker ME, et al. Tubular diskectomy vs conventional microdiskectomy for sciatica: a randomized controlled trial. JAMA . 2009;302:149-158. [DOI] [PubMed] [Google Scholar]
- 4.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28:3083-3107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cheng J, Wang H, Zheng W, et al. Reoperation after lumbar disc surgery in two hundred and seven patients. Int Orthop . 2013;37:1511-1517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Desai RJ, Franklin JM. Alternative approaches for confounding adjustment in observational studies using weighting based on the propensity score: a primer for practitioners. BMJ . 2019;367:l5657. [DOI] [PubMed] [Google Scholar]
- 7.Heindel P, Tuchman A, Hsieh PC, et al. Reoperation rates after single-level lumbar discectomy. Spine (Phila Pa 1976). 2017;42:E496-E501. [DOI] [PubMed] [Google Scholar]
- 8.Kang TW, Park SY, Oh H, Lee SH, Park JH, Suh SW. Risk of reoperation and infection after percutaneous endoscopic lumbar discectomy and open lumbar discectomy: a nationwide population-based study. Bone Joint J . 2021;103:1392-1399. [DOI] [PubMed] [Google Scholar]
- 9.Kim CH, Chung CK, Choi Y, et al. The long-term reoperation rate following surgery for lumbar herniated intervertebral disc disease: a nationwide sample cohort study with a 10-year follow-up. Spine (Phila Pa 1976) . 2019;44:1382-1389. [DOI] [PubMed] [Google Scholar]
- 10.Kim CH, Chung CK, Park CS, Choi B, Kim MJ, Park BJ. Reoperation rate after surgery for lumbar herniated intervertebral disc disease: nationwide cohort study. Spine (Phila Pa 1976). 2013;38:581-590. [DOI] [PubMed] [Google Scholar]
- 11.King G, Nielsen R. Why propensity scores should not be used for matching. Polit Anal . 2019;27:435-454. [Google Scholar]
- 12.Koenig L, Dall TM, Gu Q, Saavoss J, Schafer MF. How does accounting for worker productivity affect the measured cost-effectiveness of lumbar discectomy? Clin Orthop Relat Res . 2014;472:1069-1079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Leven D, Passias PG, Errico TJ, et al. Risk factors for reoperation in patients treated surgically for intervertebral disc herniation: a subanalysis of eight-year SPORT data. J Bone Joint Surg Am . 2015;97:1316-1325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Li F, Thomas LE, Li F. Addressing extreme propensity scores via the overlap weights. Am J Epidemiol . 2019;188:250-257. [DOI] [PubMed] [Google Scholar]
- 15.Martin BI, Mirza SK, Comstock BA, Gray DT, Kreuter W, Deyo RA. Reoperation rates following lumbar spine surgery and the influence of spinal fusion procedures. Spine (Phila Pa 1976). 2007;32:382-387. [DOI] [PubMed] [Google Scholar]
- 16.Meredith DS, Huang RC, Nguyen J, Lyman S. Obesity increases the risk of recurrent herniated nucleus pulposus after lumbar microdiscectomy. Spine J . 2010;10:575-580. [DOI] [PubMed] [Google Scholar]
- 17.Miwa S, Yokogawa A, Kobayashi T, et al. Risk factors of recurrent lumbar disk herniation: a single center study and review of the literature. J Spinal Disord Tech. 2015;28:E265-E269. [DOI] [PubMed] [Google Scholar]
- 18.Mizuno K, Takeuchi M, Kishimoto Y, Kawakami K, Omori K. Indications and outcomes of paediatric tracheotomy: a descriptive study using a Japanese claims database. BMJ Open . 2019;9:e031816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Mobbs RJ, Newcombe RL, Chandran KN. Lumbar discectomy and the diabetic patient: incidence and outcome. J Clin Neurosci. 2001;8:10-13. [DOI] [PubMed] [Google Scholar]
- 20.Nagai K, Tanaka T, Kodaira N, Kimura S, Takahashi Y, Nakayama T. Data resource profile: JMDC claims database sourced from health insurance societies. J Gen Fam Med . 2021;22:118-127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ohya J, Oshima Y, Chikuda H, et al. Does the microendoscopic technique reduce mortality and major complications in patients undergoing lumbar discectomy? A propensity score-matched analysis using a nationwide administrative database. Neurosurg Focus . 2016;40:E5. [DOI] [PubMed] [Google Scholar]
- 22.Osterman H, Sund R, Seitsalo S, Keskimäki I. Risk of multiple reoperations after lumbar discectomy: a population-based study. Spine (Phila Pa 1976). 2003;28:621-627. [DOI] [PubMed] [Google Scholar]
- 23.Quan H, Li B, Couris CM, Fushimi K, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173:676‐682. [DOI] [PubMed] [Google Scholar]
- 24.Rasouli MR, Rahimi-Movaghar V, Shokraneh F, Moradi-Lakeh M, Chou R. Minimally invasive discectomy versus microdiscectomy/open discectomy for symptomatic lumbar disc herniation. Cochrane Database Syst Rev. 2014;(9):CD010328. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Seki T, Takeuchi M, Kawakami K. Eating and drinking habits and its association with obesity in Japanese healthy adults: retrospective longitudinal big data analysis using a health check-up database. Br J Nutr . 2021;126:1585-1591. [DOI] [PubMed] [Google Scholar]
- 26.Soma K, Kato S, Oka H, et al. Influence of incidental dural tears and their primary microendoscopic repairs on surgical outcomes in patients undergoing microendoscopic lumbar surgery. Spine J . 2019;19:1559-1565. [DOI] [PubMed] [Google Scholar]
- 27.Tamai K, Kaneda K, Iwamae M, et al. The short-term outcomes of minimally invasive decompression surgery in patients with lumbar ossification or calcification of the ligamentum flavum. J Neurosurg Spine. Published online November 6, 2020. DOI: 10.3171/2020.6.SPINE20946. [DOI] [PubMed] [Google Scholar]
- 28.Teli M, Lovi A, Brayda-Bruno M, et al. Higher risk of dural tears and recurrent herniation with lumbar micro-endoscopic discectomy. Eur Spine J . 2010;19:443-450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Virk SS, Diwan A, Phillips FM, Sandhu H, Khan SN. What is the rate of revision discectomies after primary discectomy on a national scale? Clin Orthop Relat Res . 2017;475:2752-2762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wang Y, Liang Z, Wu J, Tu S, Chen C. Comparative clinical effectiveness of tubular microdiscectomy and conventional microdiscectomy for lumbar disc herniation: a systematic review and network meta-analysis. Spine (Phila Pa 1976) . 2019;44:1025-1033. [DOI] [PubMed] [Google Scholar]
- 31.Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical vs nonoperative treatment for lumbar disk herniation: the Spine Patient Outcomes Research Trial (SPORT): a randomized trial. JAMA . 2006;296:2441-2450. [DOI] [PMC free article] [PubMed] [Google Scholar]