Introduction
Acute postoperative pain is a critical issue in endoscopic spine surgery (ESS). A national survey found that approximately 80% of patients experienced acute pain after surgery, and of those, 86% had moderate, severe, or extreme pain [1]. Effective pain control is essential for facilitating early mobilization, reducing perioperative morbidity, shortening hospital stays, and preventing the transition to chronic postsurgical pain syndromes [2]. Sometimes, early severe pain may indicate complications such as a hematoma or infection. For patients without such complications, less pain is generally associated with a more favorable surgical outcome. Inadequately controlled postoperative pain is associated with an increased risk of long-term opioid dependence, functional disability, and delayed recovery time [3].
ESS, a minimally invasive technique with limited soft tissue dissection, has emerged as an alternative that may reduce postoperative pain and related complications [4]. It potentially diminishes wound-related complications, including wound dehiscence, decreases the incidence of surgical site infections, and exhibits a safety profile comparable to that of microscopic spine surgery [5,6]. Decreased surgical time, blood loss, and faster recovery due to narrow incision and decreased collateral tissue damage made it possible [7]. However, even with a minimally invasive approach, it is not possible to prevent all postoperative pain. Persistent radicular pain is common after lumbar surgery and remains a clinically important concern for recovery [8].
Traditional pain management approaches have relied on systemic opioids, which have clear limitations, including severe side effects like nausea, vomiting, respiratory depression, constipation, and sedation, and have contributed to the current opioid crisis [7]. Against this backdrop, multimodal analgesia (MMA) has emerged as the current standard of care. The core principle of MMA is to simultaneously target multiple pain pathways by combining non-opioid analgesics like non-steroidal anti-inflammatory drugs and acetaminophen, adjuvants such as gabapentin or ketamine, and regional anesthetic techniques. This synergistic approach aims to maximize analgesic efficacy while minimizing opioid consumption and its associated side effects [9].
Local drug delivery at the surgical site has emerged as a key, evolving component of MMA. The rationale for this approach lies in achieving high local drug concentrations at the source of pain generation while minimizing systemic exposure and side effects [10]. There are studies indicating that the local co-administration of bupivacaine and methylprednisolone during lumbar surgery can reduce postoperative analgesic consumption. Such local administration may help alleviate patients’ symptoms following surgery [11]. Although certain studies have demonstrated the efficacy of local infiltration anesthesia, its effects are often short-lived, and prior studies have reported conflicting outcomes. Furthermore, procedures such as epidural catheters come with a risk of complications, including systemic toxicity and issues related to the catheter, such as infection and dislodgement [12,13].
Acellular dermal matrix (ADM), originally applied in other surgical fields as an adhesion barrier, is a decellularized extracellular dermal matrix derived from human, bovine or porcine dermis. With its porous structure, it has been widely used in soft tissue reconstruction, including breast reconstruction, abdominal wall, hernia, and dural repair [14,15]. It consists primarily of collagen and elastin, forming a matrix that supports fibroblast proliferation, neovascularization, and host cells repopulation, while promoting structural stability, modulating inflammation, and facilitating tissue integration [14,15]. These properties underlie ADM’s capacity to function as an adhesion barrier, both by physically separating tissues and by modulating the inflammatory and fibrotic responses that drive adhesion formation.
Clinical experience in spine and head and neck surgery, including large dural repair series [16,17] and randomized controlled trials in thyroidectomy patients [18,19]. has indicated that ADM may help reduce postoperative adhesions. In this context, postoperative epidural fibrosis—representing dense scar adhesions around the dura—remains a major concern in spine surgery. These adhesions contribute to persistent pain, increase the risk of failed back surgery syndrome, and complicate revision procedures by raising the likelihood of dural tears [20]. Based on these considerations, we hypothesized that placement of an ADM patch soaked in a combination of ropivacaine and fentanyl over the exposed dura during ESS (lumbar-predominant)—leveraging its dual role as an adhesion barrier and a local drug delivery matrix—would be associated with reduced early postoperative pain and a decreased need for rescue analgesics.
Material and Method
Study design
This was a single-institution retrospective chart review conducted at Daejeon Woori Hospital, between September 5, 2024, and January 31, 2025. This design was chosen to retrospectively assess postoperative pain outcomes in patients who had received an analgesic-soaked ADM patch as part of surgical care.
This retrospective chart review analyzed de-identified data from routine care. In accordance with institutional policy, formal institutional review board review and individual informed consent were not required.
Patient population
All adult patients who underwent ESS at a single institute were retrospectively screened for eligibility. Patients were included if they had undergone primary single- or multi-level spinal (predominantly lumbar) decompression (such as laminectomy or discectomy) with or without spinal fusion. Patients were excluded if they had surgery for infection or tumor, underwent revision surgeries, had incomplete medical records for postoperative pain scores or analgesic consumption, or had documented allergies to ADM, ropivacaine, or fentanyl.
Patients were divided into 2 cohorts based on the intraoperative intervention received: the ADM group, who received the analgesic-soaked ADM patch, and the control group, who received standard surgical closure without the patch.
Intervention: analgesic-soaked ADM patch preparation and application
NOVOGRID (CG Bio Co., Ltd.) was used as an ADM patch source. Analgesic solution was mixed as a sterile solution was prepared by mixing ropivacaine 0.2% (10 mL, 20mg) and fentanyl 50 µg (1 mL). The rationale for this combination was to provide multimodal local analgesia, with ropivacaine providing prolonged peripheral sensory blockade at the surgical site, while fentanyl, absorbed locally, may also exert a potent analgesic effect on the neural elements in direct proximity to the dura. Ropivacaine is an amide-type local anesthetic that provides predominant sensory blockade while preserving motor function. It is known for its excellent safety profile due to its lower cardiotoxicity [21]. In addition, ropivacaine offers long-lasting analgesia for up to 3 days, making it particularly effective for controlling acute postoperative pain. Furthermore, fentanyl acts by blocking pain at the level of the central nervous system [22]. Therefore, the combination of ropivacaine and fentanyl was expected to produce a synergistic effect, resulting in superior pain control after surgery.
Immediately prior to application, a sheet of ADM, sized to cover the exposed dura at the laminectomy site (about 1.5 cm×1 cm, Fig. 1), was soaked in the prepared analgesic solution for a minute to ensure complete saturation of entire analgesic dose. After neural decompression was complete and before fascial closure, the analgesic-soaked ADM patch was carefully placed over the exposed dura (Fig. 2), ensuring full coverage of the at-risk neural elements.
Standard perioperative pain management
All patients in both cohorts received a standardized perioperative management protocol, including spinal anesthesia and postoperative patient-controlled analgesia (PCA). Two PCA regimens were used during the study: combination A (fentanyl 300 µg, nefopam hydrochloride 10 mg, tramadol 200 mg, and ramosetron hydrochloride 0.6 mg) and combination B (ketorolac tromethamine 120 mg, tramadol 200 mg, and ramosetron hydrochloride 0.6 mg). From September 5 to December 18, 2024, both groups received combination A. On December 19, 2024, an institutional protocol update switched the control group to combination B, while the ADM group continued combination A. Because this change applied only to controls and introduced potential calendar-time confounding, we performed post hoc sensitivity analyses restricted to patients managed with a uniform PCA regimen (combination A).
PCA was patient-activated with a 15-minute lockout. When pain was not controlled with PCA alone, additional analgesics (including ketorolac, tramadol, or diclofenac β-dimethylaminoethanol) were administered on request.
Data collection and outcome measures
Data were retrospectively collected from the electronic medical record system. Baseline demographic data (such as age and sex), as well as perioperative clinical variables including the number of operated spinal levels, the specific surgical procedure performed, and use of a surgical drain, were reviewed.
All procedures were performed using the biportal endoscopic technique. Surgical cases were classified into the following categories: endoscopic lumbar microdiscectomy, which involved partial laminectomy and discectomy; unilateral bilateral foraminotomy (UBF), performed a laminectomy and a sublaminar contralateral approach with partial laminectomy, flavectomy, and foraminotomy; partial hemilaminectomy, consisting of unilateral partial laminectomy and flavectomy; and the paraspinal (para) approach, based on the Wiltse technique, to perform foraminotomy of the exiting root and, when indicated, discectomy according to the patient’s pathology. Two-root decompression (TRM3) was defined as an extension of the UBF procedure, in which both the traversing root and the superior exiting root were decompressed. Posterior cervical discectomy was conducted with partial laminectomy, flavectomy, and discectomy. In certain cases, combined procedures were performed (e.g., OLM-TRM3, OLM-UBF). Cases in which the surgical classification was ambiguous or procedural applicability was limited were categorized as “not applicable.”
The following clinical parameters were evaluated. Postoperative pain intensity was assessed using the 11-point numeric rating scale (NRS), where 0 indicated no pain and 10 represented the worst imaginable pain. NRS scores were separately recorded for radicular pain (radiculopathy) and surgical site (incisional) pain at 6, 24, 48, and 72 hours following surgery. The administration of additional (non-PCA) rescue analgesics during the first 72 postoperative hours were reviewed to determine whether patients required supplemental pain control. Cumulative PCA usage was also documented for postoperative days 1, 2, and 3.
Statistical analysis
All statistical analyses were performed using R ver. 4.4.2 (https://www.r-project.org/). A P-value of less than 0.05 was considered statistically significant. Categorical variables (e.g., gender, rescue analgesic use) were compared using the chi-square test or Fisher exact test. Continuous variables were assessed for normality. Normally distributed data were compared using an independent t-test, while non-normally distributed data (e.g., NRS scores, age) were compared using the Wilcoxon rank-sum test. Post hoc subgroup analyses were conducted to account for differences in surgical invasiveness (decompression-only cases), PCA regimen, and their combination.
Results
Baseline demographic and perioperative characteristics
Table 1 summarizes the demographic and perioperative characteristics of the study cohorts. A fundamental requirement of a cohort study is to demonstrate baseline similarity between groups, and this table presents all major demographic and surgical variables collected. The data show that most variables were indeed similar (P>0.05), strengthening the validity of the comparison.
A total of 426 patients met the inclusion criteria and were assigned to either the ADM group (n=216) or the control group (n=210). The 2 groups were similar in baseline demographic characteristics, including age (ADM: 63.51 vs. control: 63.40 years, P=0.65) and gender distribution (P=0.86). There were no significant differences between the groups in terms of multi-level surgery, surgical region, number of surgical portals, surgical side, or drain placement (all P>0.05). Most procedures were performed at the lumbar level in both groups (>96%). Overall, the 2 cohorts were well matched, supporting the validity of further analyses.
Clinical outcomes in the overall population
Clinical outcomes showed distinct differences between the 2 groups. Postoperative radicular pain was significantly lower in the ADM group at 6 hours (ADM: 0.97 vs. control: 1.48, P<0.01), 24 hours (ADM: 1.49 vs. control: 1.86, P=0.02), 48 hours (ADM: 1.61 vs. control: 1.91, P=0.03), and 72 hours (ADM: 1.67 vs. control: 2.19, P<0.01). Surgical site pain was also significantly lower in the ADM group at the 48-hour time point (ADM: 2.89 vs. control: 3.08, P=0.0144), while no significant differences were observed at 6, 24, or 72 hours.
All patients received spinal anesthesia and PCA as part of a standardized perioperative management protocol. Although the PCA regimen used in the control group was changed mid-study due to inadequate pain control, there were no statistically significant differences in cumulative PCA consumption between the groups. Since PCA was activated only when pain became intolerable, the lack of difference in PCA usage suggests similar baseline access to and responsiveness to PCA between the groups. Despite this comparable baseline, a clear difference emerged in the need for additional analgesia. The proportion of patients requiring additional rescue analgesics within 72 hours postoperatively was significantly lower in the ADM group (13.0%) compared to the control group (27.1%, P<0.01). Notably, there was no significant difference between the groups in the distribution of analgesic regimens administered (P=0.87), suggesting that the reduced use of rescue analgesics in the ADM group was not due to differences in baseline analgesic protocols but rather reflects more effective pain control.
These findings are summarized in Table 2 and demonstrate a consistent pattern of improved pain control in the ADM group throughout the early postoperative period.
Subgroup analyses to control for confounding variables
To address potential bias from the inclusion of all cases regardless of whether fusion was performed, as well as from the change in PCA regimen that occurred during the data collection period in the control group, a series of post hoc subgroup analyses were performed. These analyses aimed to minimize confounding and clarify the independent association between ADM use and postoperative pain outcomes.
Decompression-only subgroup
This analysis included 195 patients in the control group and 216 in the ADM group who underwent decompression without fusion, decreasing bias in surgical invasiveness. By limiting the cohort to decompression-only cases (without fusion), a more homogenous population was achieved, and the potential variability in surgical invasiveness between groups was minimized.
Even within the decompression-only subgroup, the ADM group continued to experience significantly less postoperative pain, suggesting that the observed analgesic benefit reflects an independent pain control effect of the ADM patch. Radicular pain was significantly lower in the ADM group at all time points—6 hours (ADM: 0.97 vs. control: 1.49, P<0.01), 24 hours (ADM: 1.49 vs. control: 1.87, P=0.01), 48 hours (ADM: 1.61 vs. control: 1.93, P=0.02), and 72 hours (ADM: 1.67 vs. control: 2.18, P<0.01). Surgical site pain was also lower in the ADM group at 24 hours (ADM: 2.89 vs. control: 3.09, P=0.02), with no significant differences at other time points. As in the overall cohort, cumulative PCA consumption showed no significant difference between groups (all P>0.05), while the need for rescue analgesics remained significantly lower in the ADM group (ADM: 13.0% vs. control: 25.1%, P<0.01). These findings are summarized in Table 3.
Decompression-only subgroup with consistent PCA regimen
To further minimize potential confounding, a final subgroup analysis was performed including only patients who underwent decompression only (without fusion) and received the consistent combination A PCA regimen. This resulted in 166 patients in the control group and 216 in the ADM group.
In this subgroup, the ADM group continued to show significantly improved pain outcomes. As shown in Table 4, radicular pain was significantly lower in the ADM group at 6, 24, 48, and 72 hours (P=<0.01, P=0.01, P=0.02, and P<0.01, respectively). Lower surgical site pain was also observed in the ADM group at 24 hours and was statistically significant (P=0.02). Importantly, although cumulative PCA use was comparable between groups in this subgroup analysis, the proportion of patients requiring rescue analgesics in the ADM group (13.0%) was significantly less than half of that in the control group (27.7%, P<0.01).
This consistently observed and statistically significant reduction in radicular pain across all subgroup analyses suggests a localized and sustained analgesic effect of the ADM patch. In contrast, the effect on surgical site pain was less consistent—statistically significant at 48 hours in the overall cohort, and at 24 hours in both subgroup analyses. Given that radicular pain originates from irritation of neural elements within the spinal canal and the ADM patch is applied directly over the dura near the nerve roots, these findings raise the possibility of a targeted analgesic effect on the neural elements (reducing radiculitis). The underlying mechanism will be further discussed in the following section.
Discussion
This retrospective chart review found that intraoperative placement of an ADM patch soaked in ropivacaine and fentanyl during ESS was associated with significantly lower early radicular pain scores and a substantially reduced need for rescue analgesics. These associations persisted in post hoc analyses that accounted for surgical invasiveness and PCA regimen, despite comparable cumulative PCA use, supporting a clinically meaningful benefit beyond standard PCA-based care. The reduction in rescue analgesic use was approximately half in the decompression-only subset with a uniform PCA regimen.
As mentioned above, the more consistent reduction in radicular pain compared with surgical site pain suggests a key functional mechanism of the analgesic-soaked ADM. Radicular pain arises from irritation and inflammation of the nerve roots within the spinal canal, whereas incisional pain originates from skin and muscle incision. Because the ADM patch is placed directly over the dura in close proximity to the nerve roots, local drug release is likely to act more strongly on neural elements than on skin and muscle. Therefore, these findings support the interpretation that the primary effect of the ADM patch is a targeted, local anti-inflammatory and analgesic action on the neural elements, thereby reducing radiculitis.
The use of ADM in spine surgery represents an example of interdisciplinary technology transfer, where a biomaterial proven to improve outcomes in soft tissue reconstruction is repurposed to address a neuro-analgesic problem. The rationale is straightforward: first, ADM is widely used in plastic surgery to provide a scaffold, modulate inflammation, and promote healing [23]. Second, the post-laminectomy space is particularly vulnerable to inflammation, scarring (epidural fibrosis), and persistent pain. Third, these 2 clinical contexts share a common pathophysiological mechanism—uncontrolled local inflammation and tissue response to injury. Applying a material known to modulate this response in one field to another is thus a logical and innovative step. In this sense, ADM should not be regarded as a passive “sponge” for drugs but rather as an active biological interface. Synthesizing these considerations—the challenge of postoperative pain, the need for opioid-sparing MMA, and the unique biological properties of ADM—the local application of an analgesic-soaked ADM patch emerges as a theoretically compelling strategy. The choice of ropivacaine and fentanyl reflects this logic, combining peripheral nerve blockade from a long-acting local anesthetic with central modulation of pain from a potent opioid. This dual pharmacological targeting, integrated into the ADM scaffold, supports this rationale that the intraoperative placement of an analgesic-soaked ADM patch would reduce early postoperative pain and the need for rescue analgesics in ESS.
In addition to this targeted neural effect, the overall clinical benefits are likely reinforced by a broader dual-action mechanism inherent to the ADM patch. The first mechanism is mechanical protection and anti-fibrotic action: the ADM functions as a physical barrier, isolating the thecal sac and nerve roots from postoperative hematoma and inflammatory exudate. This is consistent with preclinical studies showing that ADM can reduce epidural fibrosis and adhesion formation, a major cause of failed back surgery syndrome [20]. Clinically, ADM has been reported to reduce postoperative adhesions across several surgical fields. In large intracranial dural repair series using acellular human dermis (AlloDerm; LifeCell Corp.) [16], grafts were stably incorporated without adhesion or scar formation, and similar results were observed in posterior fossa decompression cases [17]. Randomized controlled trials in thyroidectomy patients further confirmed that ADM significantly reduced postoperative adhesions and improved functional outcomes [18,19]. Supportive evidence has also been reported with porcine ADM (Strattice; Acelity), successfully applied in skull base and sacral dural defect repairs without adhesion-related complications [24]. Taken together, these findings suggest that ADM may exert adhesion-preventing effects in spinal surgery as well, potentially mitigating postoperative epidural fibrosis—a major contributor to failed back surgery syndrome [20]. The second mechanism is sustained local drug delivery: the porous collagen scaffold of ADM (Fig. 3) can act as a drug reservoir, enabling slow and sustained release of ropivacaine and fentanyl directly at the site of neural irritation. This targeted delivery achieves high local concentrations of analgesic and anti-inflammatory agents while minimizing systemic exposure, likely attenuating local nociceptor sensitization and neuroinflammation more effectively than systemic administration [10]. This mechanism aligns with the consistent effect observed on radicular pain, which is closely related to nerve root inflammation.
Placed in the context of modern pain management, this approach represents an alternative form of local drug delivery for spine surgery. Wound infiltration with local anesthetics provides only short-term relief due to rapid systemic absorption, while epidural catheters can prolong analgesia but carry risks of infection, migration, and nerve root irritation [25]. The ADM patch combines targeted local delivery with sustained release, without requiring an indwelling catheter, functioning as a “biodegradable, single-shot system.” As the collagen matrix degrades in vivo, it provides potent, non-opioid baseline analgesia at the surgical site. By reducing reliance on systemic opioids, this method may help mitigate opioid-related side effects and facilitate faster mobilization and recovery [9,26].
Study limitations
The primary limitation is that this work was a retrospective chart review of medical records rather than a prospectively designed clinical trial, which precludes definitive causal inference and introduces the potential for selection bias and unmeasured confounding variables. Although we performed robust post hoc subgroup analyses to control for major confounders, residual bias may remain. The decision to apply the ADM patch was not randomized but based on the surgeon’s judgment, which could have introduced systematic differences in unmeasured baseline characteristics between groups. In addition, the postoperative PCA regimen was changed mid-study for the control group due to inadequate pain relief. Although we conducted post hoc subgroup analyses restricted to patients managed with a uniform PCA regimen, this protocol change introduces the possibility of calendar-time confounding that cannot be fully excluded.
Furthermore, outcome assessment was limited to variables consistently documented in routine records, namely pain scores (NRS) and rescue analgesic use. Other potentially relevant endpoints such as length of stay, functional recovery, or quality-of-life measures could not be evaluated. Nevertheless, the large sample size and systematic data capture across more than 400 patients provide robustness to the primary findings.
The study's single-institution nature limits the generalizability of the findings to other institutions with different techniques and patient populations. Moreover, this study focused on early postoperative pain (up to 72 hours). Long-term outcomes, such as the incidence of chronic postsurgical pain, functional recovery, or rates of epidural fibrosis, were not assessed.
Because this study was conducted exclusively on endoscopic spine surgeries, specifically biportal procedures, further validation is required to determine the effectiveness of this approach in other surgical techniques, including uniportal endoscopy. In addition, reoperations were not included in the analysis of this study; therefore, further investigation is warranted to assess whether the ADM, if not fully degenerated postoperatively, or even after degeneration, might cause a mass effect or other related complications.
Although no ADM-related complications were observed in this study, potential concerns such as delayed resorption of the ADM, granuloma formation, or mass effect on nerve root should be carefully monitored in future studies. Further randomized, multicenter studies with long-term follow-up are needed to validate the generalizability and sustained efficacy of this approach.
Conclusion
The intraoperative application of an ADM patch soaked in a ropivacaine and fentanyl may represent a novel and safe adjunct for early pain control following ESS. In this retrospective review, the technique was associated with consistent reductions in radicular pain and a lower need for rescue analgesics, suggesting a meaningful advance in multimodal, opioid-sparing pain management strategies. While no ADM-related complications were observed, further prospective multicenter trials are needed to confirm safety, validate efficacy, and explore long-term outcomes before integration into routine clinical practice.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
