Comparison of Curcumin and Methylprednisolone in Preventing Epidural Fibrosis After Spinal Surgery: An Experimental Study

Aim: We aimed to compare the effectiveness of curcumin, which has been shown to inhibit cell migration and proliferation, with that of methylprednisolone, which has been shown to have a similar mechanism of action to that of curcumin and to delay tissue repair and wound healing, in the prevention of epidural brosis in an experimental laminectomy model in rats. Material and methods: The experiment was conducted with 40 male Wistar rats weighing 250-300 g which were divided into four groups. Laminectomy was performed from L1 to L4 in all groups. Curcumin and methylprednisolone were applied to the other two groups except the control and Dmso groups. The vertebral colon was excised en bloc and each section was examined for epidural brosis formation under a light microscope to evaluate histologically using the grading system. The statistical analyses indicated that the treatment groups had a signicant effect on the histological grades of epidural brosis in rats compared to the control group.


Introduction
Epidural brosis (EF) refers to the excessive formation of a scar tissue in the epidural space following lumbar laminectomy. EF is a common cause of persistent pain after spinal surgery and of failed back surgery syndrome. The EF incidence has been found to be 20-47% [1]. Various substances have been used to prevent EF; however, the use of these substances has been limited to clinical trials [2,3]. Meanwhile, it is commonly known that local administration of agents that prevent cell migration and proliferation can reduce EF formation. In this study, we aimed to compare the effectiveness of curcumin, which has been shown to inhibit cell migration and proliferation, with that of methylprednisolone, the mechanism of action of which is similar to that of curcumin, in preventing EF in experimental laminectomy rat model.

Materials And Methods
After obtaining approval from the local ethics committee for animal research, we conducted our experiment using 40 male Wistar rats weighing 250-300 g. The animals were reared in an environment with appropriate temperature and humidity levels, and they were given standard rat chow and ad libitum access to water. The rats were divided into four groups. Thirty minutes prior to the surgical procedure, prophylaxis was given via intraperitoneal administration of a single dose of cefazolin sodium (20 mg/kg) (Cefazolin, Bilim Pharmaceuticals, Istanbul, Turkey). Anesthesia was induced by intramuscular ketamine hydrochloride (25 mg/kg; Ketalar, P zer, Istanbul, Turkey) and xylazine (5 mg/kg; Rompun, Bayer, Istanbul, Turkey). The rats were xed in a prone position on an operating table. After the xation, the surgical site was brushed with povidone-iodine scrub (Medicabrush; 4% chlorhexidine soap, Medica BV, Netherlands) for 10 min and then disinfected with povidone-iodine solution (Poviod; 10% polyvinylpyrrolidone iodine complex, Saba, Turkey). The surgical site was subsequently draped with sterile sheets and then a 3-cm midline skin incision was made over the spinous processes. The paravertebral muscles were stripped through blunt and sharp microdissection. Laminectomy was performed from the L1 to the L4 vertebrae.
The ligamantum avum and the epidural fat tissue were removed, and the lumbar epidural space was exposed after hemostasis was achieved. None of the rats had any dural tears and injuries to a nerve root during the procedure. The rst group of rats received laminectomy only, and the layers were closed in the anatomical plane. The second group received 0.5 mg/kg curcumin (Sigma-Aldrich, Chemical Co., USA) diluted in DMSO solution at 1:10 ratio and applied topically on the epidural space with cotton pad for 5 min. After 5 min, the soaked cotton was removed from the surgical site, and the layers were closed in the anatomical plane. The third group received 1 mg/kg methylprednisolone applied topically on the epidural space with cotton for 5 min. After 5 min, the soaked cotton was removed from the surgical site, and the layers were closed in the anatomical plane. The fourth group received topical DMSO saline solution applied to the surgical site with cotton for 5 min. Subsequently, the soaked cotton was removed from the surgical site, and the layers were closed in the anatomical plane. The rats were maintained and followed up in a room with a temperature of 28 °C for post-anesthesia recovery. All surgical procedures were performed under an OpMi (Carl Zeiss, Germany) microscope with 16 × magni cation. None of the subjects had an infection.
After 3 weeks, the rats were sacri ced with a high dose (75-100 mg/kg) of intraperitoneal sodium thiopental (Pentothalsodium, Abbott, Italy). The vertebral column, which includes the laminectomy site, was removed cautiously.

Histopathologic Examination
The vertebral colon was excised en bloc, xed in 10 mM phosphate buffer, and decalci ed in 10% formic acid for 10 days. Subsequently, samples were obtained from the laminectomy site, washed for 6 h under running water, and then subjected to routine para n tissue procedure. Para n blocks were dissected into 5-µm-thick sections and then stained with hematoxylin and eosin. Each section was examined for EF formation under a light microscope, and the EF scar tissues were graded histologically using the grading system proposed by He  ). The frequency distribution of brosis scores were compared using the Chisquare test (χ 2 ), and the likelihood ratio Chi-square test was performed given that values in one or more cells was smaller than 5. A p value of < 0.05 was considered signi cant. Given that the proportion of cells with an expected value of < 5 is more than 20%, the Fisher Exact test results were considered.

Results
In the histological examination, dense epidural scar tissue formation and hemorrhage were detected in the laminectomy site in the control group. The scar tissue adhered to the dura mater, and the epidural space was covered by a brous tissue (Fig. 1a). A signi cant difference in the distribution of EF grades was observed in the different groups (p < 0.05). In the control, methylprednisolone, and DMSO groups, 100%, 70.0%, and 80.0%, respectively, of the EF cases are grade 3; in the curcumin group, 50.0% of the EF cases are grade 2 ( Table 2). The control and curcumin groups signi cantly differed in terms of grade 2 EF (p < 0.05), wherein the incidence rate was higher in the curcumin group (50.0%) ( Table 3) No signi cant difference was observed between the control and methylprednisolone groups (p > 0.05) ( Table 4) as well as between the control and DMSO groups (p > 0.05) ( Table 5) in terms of the distribution of EF grades. By contrast, a signi cant difference was observed between the curcumin and methylprednisolone groups in terms of the distribution of EF grades (p < 0.05) The EF grade with a high incidence rate in the curcumin group was the grade 2 EF (50.0%) whereas that in the methylprednisolone group was grade 3 EF (70.0%) ( Table 6). Similarly, a signi cant difference was observed between the curcumin and DMSO groups in terms of the distribution of EF grades (p < 0.05). The EF grade that had a high incidence rate in the curcumin group was grade 2 EF (50.0%) whereas that in the DMSO group was grade 3 EF (80.0%) ( Table 7). No signi cant difference was found between the methylprednisolone and DMSO groups in terms of the distribution of EF grades (p > 0.05) ( Table 8; Fig. 2).          Discussion EF is de ned as the excessive formation of a scar tissue in the epidural space following a lumbar laminectomy; it is a major cause of persistent pain after spinal surgery, and it often causes hospital readmissions. Due to brosis, the adhesions in the epidural region prevent the movement of a nerve in a nerve root canal [5]. Following reoperations due to epidural brosis, adhezions may develop again. New and varied treatment options that prevent the development of epidural brosis have become necessary as the current treatments are not su ciently effective with prolonged medical treatments and re-surgical interventions.
The formation of an epidural scar tissue is an indication of wound healing. EF results from the migration of broblasts induced by the chemotactic agents released as a result of the degradation of erythrocytes and thrombocytes after hemorrhage. Fibrosis formation begins in the early stages of in ammation and consists of four phases, namely, broblast migration and proliferation, extracellular matrix deposition, angiogenesis, and nally reorganization of the scar tissue [6]. The resulting brotic tissue causes distension or compression of a root nerve, ultimately leading to a radicular pain after a spinal surgery [3].
Fibroblast and vascular endothelial cells begin to proliferate within 24 h after tissue damage and also after a surgery, leading to granulation tissue formation, which is an indication of wound healing. Given that the volume of uid and protein leakage from new vessels induced by angiogenesis is relatively high, the granulation tissue formed in this stage displays an edematous appearance [7]. Fibroblast migration is induced by growth factors, including platelet-derived growth factor, epidermal growth factor, broblast growth factor, and transforming growth factor, as well as by macrophages and lymphocyte-derived cytokines. Some cytokines, such as interleukin 1 (IL-1) and tumor necrosis factor (TNF), stimulate collagen synthesis. In advanced stages of brosis, the cellular components of a granulation tissue diminish, leading to the formation of a rigid tissue that is denser than collagen and elastic bers and that involves a small number of vessels [7]. In our study, intense brosis was observed in the entire control laminectomy group, and the brosis caused adhesions to the duramater (Fig. 1a).
Although medical agents such as analgesics, myorelaxants, steroids, and/or local anesthetic injections are commonly used in the treatment of EF to eliminate the symptoms associated with brosis, a standard treatment for the underlying cause of EF remains unavailable. The abovementioned medical therapies are supported by symptomatic treatment methods, such as physical therapy, bed rest, and exercise [5]. However, given that these treatments do not eliminate a scar tissue, patients' complaints are only reduced and not completely gone.
For the prevention of EF progression, the most commonly used and recommended technique that causes less tissue damage is microsurgery [8]. Given that microsurgery leads to less dead space in the paraspinal region and epidural area, it is likely to result in less scar tissue. Moreover, given that the blood elements remaining in the epidural area may aggravate a scar tissue, a rapid hemostasis is needed. Another technique used to prevent EF is the preservation of the ligamentum avum along with the epidural adipose tissue, although this technique demonstrates limited effectiveness [9].
Literature reviews have shown that although numerous clinical and experimental studies on EF prevention have been conducted, no de nitive outcomes have been obtained. Additionally, some experimental studies have used various synthetic and organic substances in combination with drugs to prevent scar tissue formation and adhesion in laminectomy defects [3,4,10,11] Methylprednisolone is a type of synthetic glucocorticoid (steroid) drug. Steroids are known to exhibit antiin ammatory, anti-allergic, and immunosuppressive activity, and they also affect the hematopoietic system. Steroids prevent the migration of cells to areas of in ammation, and they inhibit the synthesis of acute-phase reactants as well as the synthesis and release of cytokines [12]. Moreover, they inhibit further recruitment of neutrophils and monocytes/macrophages in an in ammation site and reduces the number, the effect, and the proliferation of broblasts in a connective tissue.
A previous study reporting on the effect of long-acting steroids on EF prevention has revealed that the intraoperative irrigation of long-acting dexamethasone in the laminectomy site signi cantly reduced hospitalization and narcotics usage as a result of the inhibition of anti-in ammatory effects and pain mediators [13]. Consistent with these ndings, our results showed that compared with the control and DMSO treatments, the administration of epidural methylprednisolone prevented EF formation. Also, the incidence rates of brosis grades were signi cantly low under this treatment.
Curcumin is the main bioactive component derived from the rhizome (Rhizoma Curcumae) of turmeric (Curcuma longa), which belongs to the ginger family Zingiberaceae. Curcumin is used in wound healing and in the treatment of arthritis owing to its antibacterial and anti-in ammatory properties; it is also used in the treatment of cancer patients, particularly in preventing the development of lymph node metastasis and the progression of tumors [7,14]. Additionally, curcumin has been shown to inhibit hydrogen peroxide-induced damage in human keratinocytes and broblasts [14,15]. This effect indicates that curcumin exhibits antioxidant activity in wound healing. Studies have evaluated the effect of curcumin on wound healing in rats, and they found that the administration of curcumin enhanced wound healing, improved cell proliferation, and demonstrated an e cient free radical scavenging activity compared with the control and collagen treatments [16,17]. Furthermore, curcumin signi cantly suppresses TNF-αinduced neuroin ammation and the expression levels of IL-6, PGE2, and COX-2. Our results showed that the rate of grade 2 EF was higher in the curcumin group than in the control group. Conversely, the rate of grade 3 EF was higher in the control group than in the curcumin group. This nding may have been caused by the various effects of curcumin. Moreover, curcumin has been found to decrease posttraumatic in ammation owing to its anti-in ammatory properties in experimental spinal cord trauma studies [17,18].
In a study, curcumin was considered to provide therapeutic bene ts by suppressing lung broblasts through the inhibition of the protein kinase C epsilon, leading to the absence of apoptosis in normal broblasts [19]. The abovementioned studies strengthen the theory that broblasts may be inhibited by curcumin and thus may play a role in preventing EF. Our study also showed that curcumin exerted a signi cant effect in the prevention of EF formation by inhibiting further recruitment of broblasts to the laminectomy site compared with the control and methylprednisolone treatments (Fig. 1b).

Limitations
Our study showed that various degrees of brosis occurred in the laminectomy area. Moreover, our results showed that curcumin signi cantly reduced brosis, but further studies are needed to identify which antiin ammatory mechanisms play a role in this phenomenon. In addition, visualizing the adhesions in the epidural area through radiological examinations will render future investigations sounder.

Conclusion
EF is a major spinal surgery complication that causes pain. Numerous studies have reported on the surgical techniques and treatment methods used to prevent the onset of this complication. However, most of these studies have not provided de nitive or clinically positive outcomes. The present study indicated that the administration of curcumin remarkably reduced the incidence of EF after laminectomy, suggesting that curcumin is an agent that can effectively reduce EF formation owing to its potent antiin ammatory and anti-broblastic activity as demonstrated in the experiments. Declarations Ethics approval and consent to participate: Institutional review board approval was granted for the study.

Abbreviations
Consent for publication: The authors give the permission the publisher consent for publication.
Availability of data and materials: The datasets analysed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no con ict of interest Funding: The authors received no speci c funding for this work.