Comparison of surgical invasiveness, hidden blood loss, and clinical outcome between unilateral biportal endoscopic and minimally invasive transforaminal lumbar interbody fusion for lumbar degenerative disease: a retrospective cohort study

In this study, the VAS pain and ODI scores improved significantly in both groups at each time point compared with preoperatively. The clinical results at three months and one year were similar between groups, without significant differences. It appears that both surgical techniques are beneficial for LDD patients, with similar medium and short-term clinical results. However, the low back pain VAS scores at three days and one month were better in BE-LIF than in MIS-LIF patients. This suggests that BE-LIF better reduces lower back pain in the early postoperative period. CRP indicates the level of inflammation and tissue trauma [20]. In the absence of infection, the peak postoperative CRP value is assumed to reflect the extent of tissue damage [21, 22]. CRP levels tend to peak on POD three and decrease rapidly to baseline between PODs ten and 14 [22, 23]. In our study, CRP peaks were similar in the two groups. CRP values were generally lower in BE-LIF than in MIS-TLIF patients and were significantly reduced on POD three. CK is a quantitative indicator of muscle damage, and there is a relationship between CK levels and the pressure exerted on paraspinal muscles by retraction [24, 25]. In a prospective cohort study, Arts et al. [26] observed a dose-response relationship between CK and the degree of surgical invasiveness. In our study, the peak postoperative CK level was measured on POD one in both groups, consistent with previous study results [27]. The CK value was generally lower in the BE-LIF group than in the MIS-TLIF group, and this was more obvious at day one. These results suggest that BE-LIF causes less trauma and muscle damage than MIS-TLIF, which may also explain the different lower back pain scores in the early postoperative period (P < 0.001). In addition, Heemskerk et al. [28] have reported that the effects of pressure and duration of the tubular retractor and retraction of the paraspinal muscles during MIS-LIF will lead to muscle atrophy and denervation, thus increasing the possibility of early postoperative pain. However, the initial BE-LIF technique working area is established within the trigone of the multifidus muscle, the potential gap between the multifidus and spinous process of the posterior lamina [29]. Therefore, BE-LIF preserves the spinal structural integrity as much as possible and reduces intraoperative trauma and bleeding; these are other explanations for reduced early postoperative lower back pain in the BE-LIF group. Hospitalisation was also shorter in the BE-LIF group than in the MIS-TLIF group, implying that the BE-LIF group had better early clinical outcomes.

The concept of HBL was first proposed by Sehat et al. in 2000 [30] and mainly includes tissue extravasation, residual drain blood, and BL caused by haemolysis, which is often ignored by surgeons. Sehat et al. [30, 31] found that after total hip arthroplasty, HBL reached 49% of true TBL. In a prospective analysis of 114 patients, Smorgick et al. [32]reported substantial HBL during posterior spinal fusion surgery. Many studies have shown that ignoring HBL may not only result in postoperative anaemia not matching perioperative blood loss, but also in medical complications including delayed wound healing, infection, and prolonged hospitalisation [10, 17]. Therefore, clarifying HBL allows for a more accurate assessment of TBL. In previous studies [33, 34], HBL ranged from 194.4 to 782.4 ml during MIS-TLIF, consistent with our results. In our study, BE-LIF patients had significantly lower TBL, intraoperative and postoperative blood loss, and HBL than MIS-TLIF patients (Table 3). One possible explanation is that BE-LIF causes less muscle damage than MIS-TLIF and may reduce postoperative haemolysis. Additionally, maintaining adequate saline pressure perfusion and water patency is essential for clear vision under BE-LIF endoscopy. We usually place the saline perfusion solution approximately 70–90 cm above the surgical field surface and maintain pressure at 30–50 mmHg, which is higher than bone and small blood vessel bleeding pressures (approximately 15–25 mmHg). It simultaneously maintains clear vision and reduces bleeding.

Some studies [2, 35] have reported that continuous saline perfusion flushing of the fusion bed within the intervertebral space during BE-LIF surgery, which leads to decreased blood supply and osteogenic factors, may lead to decreased fusion rates. In our study, however, both groups achieved good fusion rates and did not differ, similar to previous studies [5, 8, 36]. BE-LIF can identify and completely remove cartilage endplates under direct endoscopic vision; this provides a good environment for intervertebral fusion [4, 37]. However, such delicate endoscopic manipulation not only requires surgical experience but may also be time-consuming; this might explain why the BE-LIF group had longer surgical times. In addition, a meta-analysis [7] reported that the learning curve is also a factor, especially in the curve’s early stages, because decompression takes longer. Kim et al. [38] found that with increasing case numbers, the BE-LIF operation time will gradually decrease; proficiency is reached at around 34 cases. Although the operation time was longer in the BE-LIF than in the MIS-TLIF group, the incidence of perioperative complications did not differ. No severe complications requiring revision occurred in either group. According to the modified MacNab criteria, the excellent and good rates were 97.3% in BE-LIF and 95.4% in MIS-TLIF at one-year follow-up. This indicates that both techniques are safe and effective.

We believe that the BE-LIF technique offers several advantages. First, it has independent endoscopic and instrument channels. Therefore, the endoscopic lens has a wide range of movement and flexibility, and identification of intraspinal structures is clearer and more convenient. Second, the space between the muscle and lamina is used to establish the initial working channel, and there is no prolonged paravertebral muscle traction, causing less lower back muscle trauma. Continuous irrigation with water pressure during the operation also helps to reduce bleeding and to maintain a clear view with endoscopy, reducing the infection risk. However, hospitalisation costs were significantly greater in the BE-LIF group (66813.68 ± 3734.79 RMB) than in the MIS-TLIF group (58968.95 ± 4757.98 RMB; P < 0.001). Higher prices and an inadequate Chinese social-medical insurance system related to BE-LIF surgery have somewhat limited its development.

Some limitations of our study should be mentioned. First, this is a single-centre, retrospective cohort study, and inherent selection bias is inevitable. We did our best to balance the two groups at the beginning of the study, and they did not differ based on basic demographics. Second, we calculated true TBL using HCT at three days postoperatively (HCT_post). If the patient has not yet hemodynamically stabilized, fluid shifts may be incomplete, resulting in low values for calculated TBL and HBL. This factor did not affect the results of our two groups, because the HCT_post values of both groups were collected on the same day after surgery. Third, the case number was small and follow-up time short; thus, a prospective randomized controlled trial with longer follow-up and larger sample size is still needed to further evaluate this technique.