Lumbar spinal stenosis (LSS) is a common spinal degenerative disease in middle-aged and older adults. Open laminectomy for spinal stenosis is a safe and cost-effective procedure, and its outcomes are superior to those of nonsurgical management [1,2]. More recently, uniportal and biportal endoscopic decompressive approaches provide safe and outstanding clinical outcomes. One of the primary benefits of endoscopic methods is a further reduction in the disruption of the surrounding soft tissue and direct visualization of the pathologic process. The most superior procedure among minimally invasive spine surgeries remains unknown; however, in several meta-analyses, biportal endoscopic procedures have shown benefits [3–6].

Several sequences for endoscopic decompression have been described. The most commonly used technique is the “N” technique, which follows craniocaudal decompression. Some authors have recently suggested that the “Z” or “side-to-side” technique may yield better outcomes [7,8]. Regardless of the decompression sequence used, the primary goal is to remove the hypertrophic part of the ligamentum flavum. Over the years, various methods have been developed to remove the hypertrophic ligamentum flavum. The “en bloc” or “butterfly-shaped” removal technique is a novel approach that is recommended for piecemeal resection [9,10]. Some authors reported several advantages of this “one-piece” technique: the ligament acts as a protective barrier for the dura during drilling, the incidence of epidural bleeding is lower, which improves visibility, and any inadvertent durotomy typically occurs later in the procedure [9]. However, one-piece resection of the ligament can be challenging in patients with severe LSS. In accordance with the “en bloc” principle, the resection of the ligamentum flavum in two pieces, or “two-wings,” by dissection from the midline toward the lateral recess was deemed to be a more straightforward procedure. Unfortunately, no comparative clinical results regarding these approaches have been published. Therefore, this study aimed to evaluate and compare the clinical results of three surgical techniques for the treatment of lumbar stenosis: the conventional piecemeal flavectomy, “en bloc” one-piece resection with a butterfly technique, and the novel “two-wings” method in performing biportal unilateral en bloc flavectomy and laminectomy (BUTTERFLY) in patients with LSS.

This study was approved by the ethics committee of a Hospital Universitario Virgen del Rocío and was conducted in accordance with the Declaration of Helsinki (SICEIA-2024-003561). The retrospective study included prospectively collected data from patients who underwent endoscopic biportal surgery for LSS between May 2022 and May 2023. Written informed consent was obtained from all patients.

Patients diagnosed with lumbar degenerative diseases that met the surgical indications for LSS, specifically types C or D according to the classification by Schizas et al. [11], were included. Conversely, patients with spinal deformities, such as scoliosis, lumbar disc herniation without accompanying bony spinal canal stenosis, spinal tumors, history of lumbar surgery, spondylolisthesis (Meyer grade ≥II), and any other neurological lesions or conditions that could disrupt accurate clinical assessments preoperatively and postoperatively were excluded.

The ligamentum flavum, a bifid ligament within the posterior ligamentous complex of the spine, is frequently encountered during lumbar surgery and serves as a crucial anatomical landmark [12]. The concept of “en bloc” resection of the ligamentum flavum, that is, surgically searching peripherally for the insertions of the ligament, preserving most of the surgery and removing it at the end of the procedure in one-piece like a butterfly, was previously described [10]. In this study, we propose a novel surgical variant involving the ligament’s detachment at upper and lower insertions with the same “en bloc” principle. Thereafter, a midline dissection, analogous to an “open-book” maneuver, is performed to remove the ligament in two large fragments, termed “two-wings.” This approach allows meticulous dissection in both lateral recesses, detaching the ligament from the most medial part of the superior articular process on both sides.

In patients with LSS, the objective of the surgical procedure is successful decompression through complete resection of the ligamentum flavum [13–15]. In this study, all groups demonstrated clinical improvement in most clinical variables; however, no substantial differences in the extent of improvement were identified. Therefore, this study suggests that the technical approach to its removal does not affect the ultimate clinical outcome (VAS, ODI, and MacNab). Although this conclusion appears reasonable, it has not been previously assessed.

Significant differences were not identified among the three groups with regard to perioperative variables, although a significant reduction in surgical time was observed. The resection of the ligamentum flavum using the piecemeal technique was nearly twice as long as the “two-wings” resection. A previous study already indicated that the “en bloc” resection of the ligamentum flavum could reduce surgical time in open surgery [9].

Surgical time is becoming a more important variable due to not only the increased efficiency of shorter surgeries but also the fatigue experienced by the surgeon and surgical team [5,6]. Multilevel lumbar canal stenosis is very common in older populations, and the shortened surgical time may allow for the decompression of more levels within the same period. The results of this study show that the “two-wings” technique shortens the surgical time by >45% compared with the piecemeal technique per level and by 30% compared with the “en bloc” technique.

In this study, no significant differences in postoperative complications were found. However, the piecemeal group exhibited higher complication rates. The ligamentum flavum acts as a natural barrier, protecting the dura and nerve root from direct injury, and can be strategically used during surgery. Injuries to the nerve root or dura typically occur during spinal canal decompression. The “en bloc” or “two-wings” removal of the ligamentum flavum may provide better protection for the nerve root and dura. In this study, all incidental durotomies occurred while manipulating a Kerrison rongeur to release the ligamentum flavum. The use of the “en bloc” or “two-wings” techniques allows for the release of the ligamentum flavum in larger sections, reducing the number of Kerrison rongeur maneuvers needed and maintaining protective coverage over the dura throughout most of the procedure.

Compared with piecemeal resection, “en bloc” resection of the ligamentum flavum helps prevent the inevitable expansion of the dural sac that occurs during decompression. This expansion increases the risk of dural tears, which can be reduced by maintaining compression on the dural sac until all insertion points are released and the ligamentum flavum is completely removed [9,10].

This retrospective study has several key limitations. First, the study was conducted at a single center and involved a relatively small cohort, and the surgical techniques were not randomly assigned to each patient. To reduce the risk of selection bias, only patients with severe or critical central canal stenosis were included. However, a randomized, multicenter, controlled clinical comparison study with a larger sample size and an extended follow-up period is required. In addition, new surgeons must be aware of the learning curve associated with removing the ligamentum flavum “en bloc.”

The results revealed that removing the ligament flavum “en bloc” (either in two pieces or one) may reduce the surgical time and incidence of dural tears. However, randomized and prospective studies are warranted to establish definitive conclusions.