Unilateral Biportal Endoscopic Discectomy (UBE) Versus Percutaneous Endoscopic Lumbar Discectomy (PELD) for Treating Lumbar Disc Herniation in Obese Patients: A Retrospective Study

Introduction

Lumbar disc herniation (LDH) is a common cause of chronic low back pain (LBP) and radiculopathy, significantly impairing patients’ quality of life. Obesity is a global health problem that is getting worse [1]. Over the past decade, the prevalence of obesity and high adult body mass index (BMI) in the United States has increased by 20%, posing significant public health challenges [2]. Obesity affects the musculoskeletal system by overloading the intervertebral discs, leading to abnormal inflammation and disc herniation [3]. Furthermore, being overweight increases the likelihood of experiencing LBP [4]. Conservative treatments are often less effective, and surgical intervention is more frequently required for obese individuals compared to non-obese individuals [5]. Obese patients with LDH often present unique clinical challenges, including altered anatomical landmarks, reduced intraoperative visualization, and higher risks of perioperative complications, which complicate surgical decision-making.

Currently, both minimally invasive procedures and open surgery are mainstays of clinical management for lumbar disc herniation (LDH) in overweight patients. Open surgeries such as transforaminal lumbar interbody fusion (TLIF) and posterior lumbar interbody fusion (PLIF) for LDH in obese patients often require longer incisions to better expose the anatomy, which can result in more extensive lumbar muscle injury, prolonged hospital stays, and increased complications such as infection and spinal instability [6,7]. Traditional open discectomy, while effective, carries inherent risks of tissue trauma and prolonged recovery, particularly in obese patients, and minimally invasive spinal surgeries are now more common. For obese people with LDH, percutaneous endoscopic lumbar discectomy (PELD) is a safe and effective minimally invasive method that can help improve their VAS, ODI, and SF-36 scores [8]. Unilateral biportal endoscopic discectomy (UBE), a relatively new minimally invasive procedure, utilizes an endoscope through an observation channel and introduces surgical instruments via a separate working channel. This method has been shown to significantly improve both back and leg pain in obese patients, enhance activity levels, and reduce length of hospital stays (LOS) [9]. Obesity-related factors, including subcutaneous fat thickening, narrow interlayer space, and difficult localization techniques, disproportionately affect surgical feasibility, complication rates, and postoperative recovery [10]. While both minimally invasive techniques have individually demonstrated effectiveness in obese people, their clinical efficacy in treating LDH in obese patients has not been thoroughly investigated and compared.

This retrospective study, conducted on 43 obese patients with LDH who underwent minimally invasive surgery at our hospital between January 2018 and August 2023, evaluated the clinical efficacy, perioperative outcomes, and complication profiles of UBE versus PELD. By comparing the clinical effectiveness and safety of these 2 techniques in this high-risk population, we seek to address the existing knowledge gap, thereby informing surgical strategy optimization and improving care delivery for an increasingly prevalent patient demographic.

Material and Methods

PATIENTS:

This retrospective study included 43 obese patients who had LDH treated with minimally invasive surgery at the Zhangjiagang Hospital affiliated to Soochow University from January 2018 to August 2023. Two groups were formed from the patients: the PELD group and the UBE group.

The inclusion criteria were: 1) MRI examination confirmed single-segment LDH; 2) BMI ≥30 kg/m²; 3) age >18 years old, with low back pain, leg radiation pain. The exclusion criteria were: 1) lumbar instability, lumbar tumor, lumbar deformity, or lumbar infection; 2) history of previous lumbar surgery; 3) loss to follow-up; 4) seriously ill and cannot tolerate surgery. Before the surgery, all patients underwent X-rays, CT scans, and MRI.

UBE: The patient was positioned in the prone posture after administration of general anesthesia. Fluoroscopy with a C-arm machine was used to identify and mark the surgical site, and to establish a channel for biportal endoscopic visualization (Figure 1A). Depending on the location of the disc herniation, the procedure was performed on either the left or right side of the body. A transverse incision was made, and the sleeve was gradually expanded to establish the working and viewing channels. Layer-by-layer dissection was performed to expose the articular processes, followed by opening the laminae and cutting through the ligamentum flavum to reveal the intervertebral disc, dural sac, and compressed nerve roots. After ensuring hemostasis, the adipose tissue in the spinal canal was removed, and the protruding nucleus pulposus was excised using straight forceps. Decompression of the nerve root was confirmed, with the nerve appearing relaxed and exhibiting good mobility. A negative-pressure drainage tube was inserted, and the incision was sutured.

PELD: After administering local anesthesia, the patient was positioned laterally with the affected side elevated. C-arm fluoroscopy was used to accurately position and mark the puncture site. A 20-G puncture needle was then inserted horizontally into the target intervertebral foramen, followed by insertion of a guide wire, placement of a sleeve, and gradual expansion to establish the surgical channel. Next, the endoscope was inserted and connected to the spinal endoscopic system. Guided by the endoscope, a bipolar radiofrequency knife was used to remove soft tissue, perform foraminal enlargement and decompression, and place the working cannula (Figure 1B). The protruding nucleus pulposus was excised. The nerve root was assessed to ensure adequate decompression, the vertebral space was irrigated, and hemostasis was confirmed. Finally, a negative-pressure drainage tube was inserted, and the incision was sutured.

CLINICAL OUTCOMES:

Baseline demographic data were collected for patients considered for inclusion. BMI was computed as (kg/m²)=weight (kg)/(height)² (m²). To evaluate surgery-related outcomes for PELD and UBE, we recorded the duration of surgery, LOS, Hb loss, and postoperative complications. VAS was used to assess the degree of pain in the lower back and lower limbs. ODI scores were collected to determine the impact on daily living activities. Clinical assessments were conducted at admission, 1 month, 3 months, and the final follow-up after surgery.

STATISTICAL ANALYSIS:

Data analysis was carried out using SPSS version 29.0 (SPSS, Inc., Chicago, IL, USA). Statistical significance was determined using the independent-samples t test and the chi-square test on for count data, with a significance level of P<0.05.

Results

PATIENT DEMOGRAPHICS:

Among the 43 patients eligible for inclusion, 29 were assigned to the PELD group and 14 to the UBE group. At admission, the 2 groups were similar with respect to age, sex, BMI, VAS, ODI scores, and relevant components of LDH (Table 1).

CLINICAL OUTCOMES:

Table 2 shows the perioperative results. The UBE group had longer surgery duration (142.1±28.9 min vs 110.3±23.9 min, P<0.001), more Hb loss (11.2±4.2 g/L vs 7.0±4.3 g/L, P=0.005), and more postoperative drainage (57.5±24.4 ml vs 27.1±14.0 ml, P<0.001). No statistically significant difference in LOS was observed between the 2 groups.

FUNCTION OUTCOMES:

Table 3 presents the functional results. The last follow-up time of UBE group was 9.9±2.4 months, and 10.7±3.4 months for the PELD group. Both groups showed considerable improvement in postoperative back and leg pain as measured by VAS and ODI scores. When comparing the 2 groups at the 1-month, 3-month, and the last follow-up points after surgery, no statistically significant changes were seen in VAS score or ODI. Both groups’ VAS and ODI scores improved considerably from admission to the last follow-up, as shown in Figure 2, but this improvement was not statistically significant (VAS back: 4.6±1.3 vs 3.8±1.6, P=0.12; VAS leg: 5.2±1.7 vs 5.4±1.8, P=0.78; ODI: 34.1±6.3 vs 38.2±8.8, P=0.13).

COMPLICATIONS:

In the UBE group, 2 patients had 1 small intraoperative dural tear each. Consequently, there was no need for a repair, postoperative drainage was normal, and problems like cerebrospinal fluid leakage did not arise. One patient in the UBE group had an asymptomatic hematoma, while no such cases were observed in the PELD group (P=0.71). However, none of the patients required surgery to remove the hematoma. Infection was more common in PELD group (2 cases) but the difference did not reach statistical significance (P=0.82). Additionally, 1 patient in the UBE group and 4 patients in the PELD group were treated nonoperatively (with drugs or lumbar injections) for lumbar disc herniation recurrence. However, 2 patients in the PELD group eventually underwent reoperation. The 2 groups had similar rates of postoperative complications and recurrence (Table 4).

Discussion

The rising epidemic of obesity has received significant attention due to its associated health risks. Obese individuals have a higher prevalence of LDH compared to normal-weight people, often necessitating surgical intervention [5]. Traditional open surgery remains a common treatment for LDH, effectively alleviating mechanical compression within the spinal canal [11]. However, previous studies have shown that obese patients face challenges, including prolonged hospital stays, increased risk of bleeding, and a higher susceptibility to surgical site infections following open surgery [12]. In contrast, minimally invasive surgery provides precise decompression through small incisions while minimizing compromise of the integrity of the back of the spine and mitigating potential complications [13]. Consequently, this retrospective study evaluated whether clinical outcomes differed between obese patients undergoing PELD versus UBE surgery.

We found no difference in LOS between the UBE group and the PELD group. However, postoperative drainage volume, Hb loss, and operation time in the UBE group were greater than in the PELD group. A previous study found the average operation time was significantly longer in the UBE group (96.15±16.97 min) than in the PELD group (85.52±17.79 min) [14]. However, in our study, they were 142.1±28.9 min and 110.3±23.9 min, respectively (P<0.05). One potential factor to consider is that obese patients have a larger skin-to-surgical-field distance, making it more difficult to define the surgical field. Additionally, the use of long, precise spinal instruments in a confined space increases the surgical time compared to conventional LDH procedures [15]. Hao et al revealed that the shorter operation time in the PELD group compared to UBE may be due to simpler surgical steps and greater visual range [16]. UBE requires the sequential creation of 2 different channels – an observation portal and a working portal – unlike PELD, which uses a single-channel approach. Due to the need for precise positioning, continuous expansion, and continuous adjustments to maintain optimal visualization and instrument triangulation, this 2-channel setup inherently prolongs surgical time [17]. In obese patients, thickened subcutaneous fat layers obscure anatomical markers, further complicating the placement and dilation of channels. Fluoroscopic guidance, while essential, requires repeated adjustments and prolonged surgical preparation due to reduced imaging clarity due to changes in adipose tissue. In addition, UBE often needs to remove partial laminae and ligamentum flavum to enter the spinal canal, a step that PELD does not have [18]. These procedures require meticulous attention to anatomy to avoid nerve damage, especially in obese patients with distorted anatomy, thereby prolonging surgical time. Hb loss is often overlooked in minimally invasive spinal surgery. Ao et al found that there was greater Hb loss during percutaneous endoscopic surgery [19]. Our study found that Hb loss in the UBE group was considerably higher than in the PELD group. Firstly, PELD directly accesses the herniated disc via natural pathways without the need for lamina removal, whereas UBE necessitates partial lamina resection, potentially causing cancellous bone bleeding [20]. Secondly, UBE surgery is more extensive, and while beneficial for decompression, it also increases the risk of unintentional soft tissue and epidural venous plexus injury, especially in patients with epidural fat thickening. Thirdly, compared with the single puncture site of PELD, the 2-channel technique expands the surgical wound area, resulting in greater capillary seepage in the subcutaneous and muscular layers. Wu et al reported that UBE surgery involved significantly longer incisions than PELD, thereby increasing the likelihood of bleeding from the surgical wound [21]. Lastly, the creation of 2 channels in UBE surgery leads to greater injury to the paravertebral muscles, especially the erector spinae muscles, which may be exacerbated in obese patients. The UBE group also had higher postoperative drainage volumes compared to the PELD group. Due to the characteristics of the endoscopic device, saline used during the procedure may flow into the drainage tube postoperatively, consistent with findings by Choi et al [22].

In our study, we found that postoperative pain and daily function improved in both groups, and both groups showed no statistically significant difference. Regardless of whether LDH was treated with open surgery or minimally invasive surgery, obese patients appeared to show improvements in pain and daily functioning independent of the type of surgery. Willems et al reported that patients showed clinical improvement within the first year following lumbar discectomy, regardless of preoperative BMI [23]. Additionally, a similar proportion of obese patients who received open TLIF and those who received MIS-TLIF achieved clinically meaningful improvements at 2 years [6]. Furthermore, ODI and VAS scores in obese patients treated with UBE and tubular microscopic discectomy showed no statistically significant differences at 3, 6, and 12 months postoperatively [22].

Overall, complications in this study were infrequent, although some studies suggest that obese patients tend to experience a higher incidence of complications. Dural tear is the most common complication associated with UBE [24]. Two cases of dural tears occurred in our UBE group, but repair was not performed due to the small size of the tears. Research indicates that overweight individuals are at an increased risk of dural tears compared to those with normal weight [25]. The risk of dural tear is heightened in UBE procedures due to the longer surgical incision and broader exposed area, which result from variations in surgical techniques and tools [26]. Infection is a significant concern for orthopedic surgeons. Previous studies have shown that overweight individuals are more likely to have an infection after spine surgery [27]. Infection may occur due to thicker subcutaneous fat layers, reduced blood circulation, and fat tissue’s promotion of inflammatory cytokine production, leading to delayed healing [28]. At the same time, the likelihood of dead cavities forming in larger incisions during open surgery is higher, leading to an increased risk of infection and inadequate vascularization in adipose tissue [29]. Two patients in the PELD group developed postoperative infection, with an incidence of 7.0%, slightly higher than the 2.7% reported by Xiao et al, and they found obesity was an independent risk factor for infection after PELD [30]. However, surprisingly, we found no discernible variation in wound infection rates among obese patients who had UBE versus PELD. Nevertheless, infection remains an important issue for surgeons to address. Relapses occurred in 4 patients treated with PELD and 1 patient treated with UBE in our study. According to the literature, the postoperative recurrence rate for PELD ranges from 7.0% to 17.6%, and obesity increases the probability of postoperative recurrence of PELD [31,32]. There are no data available on the recurrence rate of UBE. Consequently, drawing comparisons between the findings of our research and other studies is challenging. In our study, the recurrence rate was 7.1% in the UBE group and 13.8% in the PELD group, which is consistent with previous studies, but the results may be influenced by the limited sample size. PELD relies on a single working channel, which restricts the endoscopic field of view and instrument maneuverability, particularly in obese patients with thickened subcutaneous fat and narrowed intervertebral foramina. This constrained visualization may impede complete removal of herniated disc material, leaving residual fragments that predispose to recurrence. In contrast, UBE utilizes dual channels, enabling a wider surgical field and enhanced triangulation of instruments. This technical advantage likely facilitates more thorough decompression and excision of herniated nucleus pulposus, reducing residual fragments – a critical factor in recurrence prevention [33]. Additionally, the anatomical distortion caused by obesity, such as obscured landmarks and compressed spinal spaces, may disproportionately affect PELD’s efficacy, as precise targeting of the herniation becomes more challenging without the broader exposure afforded by UBE. Obesity exacerbates technical difficulties due to increased epidural fat, thickened soft tissues, and reduced working space. These factors may limit PELD’s ability to achieve adequate foraminal enlargement or address central or migrated herniations, which require extensive decompression. Notably, incomplete resection of disc material in such scenarios is a well-documented risk factor for recurrence. In contrast, UBE’s ability to perform partial laminectomy and ligamentum flavum resection allows direct access to the spinal canal, enabling comprehensive decompression even in anatomically challenging cases. Furthermore, variations in herniation location (eg, central vs paracentral) may influence outcomes. For instance, central herniations in obese patients may be less accessible via PELD’s transforaminal approach, whereas UBE’s posterior interlaminar route provides more direct visualization and instrument control. The greater mechanical load in obese patients increases intradiscal pressure, potentially accelerating re-herniation at inadequately decompressed levels. PELD’s reliance on indirect decompression via foraminal expansion may be insufficient to mitigate this risk compared to UBE’s direct neural decompression and broader annular defect management. Additionally, the inflammatory milieu in obese patients, characterized by adipokine dysregulation, may impair tissue healing and exacerbate residual disc degeneration, further predisposing to recurrence [34].

The choice between UBE and PELD in obese patients with LDH should be guided by a nuanced evaluation of surgeon expertise, herniation characteristics, and anatomical constraints. First, surgeon experience and preference play a pivotal role. UBE, with its dual-channel approach, demands advanced endoscopic skills for channel triangulation and instrument manipulation, particularly in obese patients with distorted anatomy. Surgeons less familiar with UBE may prefer PELD, which utilizes a single-channel system and may align more closely with traditional endoscopic techniques. Conversely, surgeons proficient in UBE may favor its broader field of view for complex cases. Second, the location and severity of herniation significantly influence technique selection. Central or migrated herniations, which require direct spinal canal access and extensive decompression, may be better addressed via UBE. The posterior interlaminar approach of UBE allows direct visualization of the dural sac and nerve roots, facilitating complete removal of migrated fragments. In contrast, PELD’s transforaminal approach is highly effective for paracentral or foraminal herniations but may struggle with central or sequestered fragments due to limited visualization and instrument maneuverability. For smaller, localized herniations, PELD’s minimally invasive nature and shorter operative time may suffice. Third, anatomical challenges specific to obesity must be considered. Thickened subcutaneous fat, narrowed intervertebral foramina, and obscured bony landmarks in obese patients can hinder PELD’s transforaminal access. UBE’s dual-channel system mitigates these issues by enabling sequential dilation and improved instrument triangulation, even in patients with excessive epidural fat. Additionally, UBE’s ability to perform partial laminectomy and ligamentum flavum resection provides a critical advantage in cases of severe spinal stenosis or foraminal narrowing. Patient-specific factors also inform decision-making. For high-risk patients with comorbidities predisposing them to bleeding or infection, PELD’s shorter operative time and reduced soft-tissue trauma may lower perioperative risks. However, in patients with a history of recurrent herniation or biomechanical instability due to obesity, UBE’s thorough decompression and lower observed recurrence rates (7.1% vs 13.8% in our study) may justify its longer operative duration. Prospective studies stratifying patients by herniation type, surgeon experience, and BMI subcategories could refine selection criteria. Additionally, cost-effectiveness analyses comparing long-term outcomes (eg, reoperation rates, chronic pain) may further guide clinical decisions.

Although this study provides preliminary evidence for minimally invasive surgical options for LDH in obese patients, there are several limitations that need to be considered. First, small sample sizes (14 in the UBE group and 29 in the PELD group) may reduce statistical power, especially in the analysis of secondary outcomes such as complications and recurrence rates. Small sample sizes may not adequately capture subtle differences between the 2 groups, such as 7.1% in the UBE group and 13.8% in the PELD group in terms of recurrence rates. Although the trend was consistent with expectations, it was not statistically significant due to the sample size. In addition, the small sample size may limit the external validity of the results, and future multi-center, large-sample studies are needed to verify the universality of the conclusions. Second, retrospective study design can introduce selection bias. The grouping of patients was based on historical data and was not randomized, which could have led to potential imbalances in baseline characteristics (eg, type of disc herniation), and although there was no significant difference in baseline measures between the 2 groups in this study, there was a risk that unmeasured confounders (eg, patient activity level) could have affected the results. In addition, reliance on data from medical records can lead to information bias, such as inadequate completeness of postoperative care details or complication records. Third, the shorter follow-up time may not be sufficient to assess long-term efficacy. Obese patients are at higher risk of long-term postoperative complications (such as adjacent segment degeneration and chronic pain) due to metabolic abnormalities and increased mechanical load, and short-term follow-up may not detect such problems. Future studies need to extend the follow-up period to fully assess the durability and safety of the procedure. In addition, the study did not control for the potential impact of surgeon experience and postoperative care. Both UBE and PELD are technically sensitive operations, and the proficiency level of different surgeons may significantly affect the operative time, blood loss, and complication rate. For example, inexperienced surgeons dealing with the complex anatomy of obese patients may prolong UBE surgery time or increase the risk of dural tears, and differences in postoperative rehabilitation protocols (such as activity restriction, physical therapy) can also affect functional recovery and recurrence rates. In the future, it is necessary to reduce the interference of such confounding factors by standardizing surgical procedures and postoperative management. Despite the above limitations, this study still provides an important reference for the selection of minimally invasive surgery methods in obese patients. UBE achieves sufficient decompression of nerves under a wider surgical field and may reduce the risk of recurrence. PELD, on the other hand, is more minimally invasive and reduces intraoperative trauma. Prospective multi-center studies, combined with long-term follow-up and hierarchical analysis of surgeon experience, are needed to further identify the best-suited patient population and indications of the 2 surgical methods.

Conclusions

For obese patients requiring surgical intervention, UBE and PELD demonstrated similar clinical improvements. While UBE was linked to greater Hb loss, longer surgery duration, and increased postoperative drainage, it tended to result in fewer complications in the short term and a lower recurrence rate.

Data Availability

Data are available on request.