Modified Hu-Lian-Bi-Guan decoction for wound healing in complex anal fistula: A randomized controlled trial

Abstract

BACKGROUND

Complex anal fistulas (AFs) frequently require extensive surgical excision, leading to large, slow-healing wounds that significantly affect the quality of life. Therefore, the optimization of the postoperative wound microenvironment to accelerate healing is an urgent clinical priority.

AIM

To evaluate the clinical efficacy and mechanism of modified Hu-Lian-Bi-Guan decoction (mHLBGD) in promoting postoperative wound healing for complex AF.

METHODS

In this randomized controlled trial, 56 adults with complex AF were randomly assigned (1:1) to receive standard surgical care (control group, n = 28) or standard care with oral mHLBGD (treatment group, n = 28) for 14 days. Clinical outcomes included wound healing rate, Visual Analog Scale, wound exudate, wound edema, and Wexner incontinence scores, and complete healing time. Mechanistically, paired tissue samples were analyzed to assess the expression of inflammatory cytokines, TGF-β/Smad pathway components, and epithelial-mesenchymal transition (EMT) markers.

RESULTS

Baseline characteristics were comparable between the two groups. The mHLBGD group demonstrated a significantly shorter median time to complete wound healing than the control group (32 days vs 50 days; P < 0.001), and achieved higher wound healing rates at all postoperative follow-ups. Postoperative pain, wound exudate, and edema scores were significantly lower in the mHLBGD group (P < 0.001) with no adverse effects on continence. Per histological analysis, mHLBGD treatment significantly diminished inflammatory infiltration, edema, and hemorrhage while enhancing granulation tissue maturation and promoting reparative epithelial proliferation. At the molecular level, mHLBGD treatment significantly downregulated the expression of IL-1β, IL-6, TGF-β, Smad2, N-cadherin, and vimentin, and upregulated E-cadherin by day 14 compared to controls.

CONCLUSION

Oral mHLBGD effectively accelerates postoperative wound healing in complex AF by reducing local inflammation and TGF-β/Smad signaling pathway regulation to inhibit pathological EMT, suggesting its potential for improving recovery post-surgery.

Key Words: Modified Hu-Lian-Bi-Guan decoction; Complex anal fistula; TGF-β/Smad signaling pathway; Epithelial-mesenchymal transition; Inflammatory factors

Core Tip: This randomized controlled trial evaluates the efficacy of modified Hu-Lian-Bi-Guan decoction (mHLBGD) in enhancing postoperative wound healing for complex anal fistula (AF). The study demonstrates that mHLBGD significantly accelerates healing time and reduces postoperative pain compared to standard care. Mechanistically, the research highlights that mHLBGD promotes tissue repair by modulating inflammatory cytokines, regulating the TGF-β/Smad signaling pathway, and influencing epithelial-mesenchymal transition. These findings provide compelling clinical and molecular evidence supporting mHLBGD as an effective adjunctive therapy for complex AF recovery.

INTRODUCTION

An anal fistula (AF) is a widespread chronic inflammatory disease of the perianal area, pathologically defined as a tract linking the anal canals to the perianal skin. Epidemiological data show a remarkable sex difference, revealing an incidence of 12.3 per 100000 among men compared with 5.6/100000 among women, with a peak age of onset of approximately 40 years[1]. Its pathogenesis is primarily attributed to the cryptoglandular hypothesis, indicating that the infection originates from the anal glands in the intersphincteric space[2]. When complex AF occurs, surgical intervention frequently results in large complex wounds, and healing is a tedious and challenging process. Therefore, refining postoperative wound healing is the key to clinical practice. It directly influences critical outcomes, including the preservation of sphincter function, risk of infection recurrence, and long-term prognosis.

The healing of cutaneous wounds involves a delicate balance of biological processes, among which epithelial-mesenchymal transition (EMT) plays a dual role. Under physiological conditions, moderate and reversible EMT is essential for re-epithelialization and wound closure[3]. Nonetheless, this process is often dysregulated in chronic non-healing wounds, such as those in complex AF. The TGF-β/Smad signaling pathway is the canonical cascade driving EMT[4]. Chronic local inflammation results in persistent overexpression of TGF-β, which subsequently activates the Smad2/3 signaling pathway. This cascade prompts epithelial cells to lose their apical-basal polarity and intercellular junctions, as evidenced by the downregulation of epithelial markers such as E-cadherin (E-cad) and the simultaneous upregulation of mesenchymal markers like vimentin (VIM) and N-cadherin (N-cad)[5-8]. Ultimately, this sustained EMT overactivation compromises the integrity of the basement membrane and facilitates cellular transition toward a fibroblast-like phenotype. This transition results in excessive extracellular matrix deposition and fibrosis rather than the regeneration of functional tissue. Therefore, targeting the TGF-β/Smad pathway to rebalance EMT is a promising therapeutic strategy to improve the healing of complex AF.

Traditional Chinese medicine (TCM) is extensively employed as a crucial adjunctive therapy alongside conventional surgery for the postoperative management of AF. TCM employs various modalities, such as herbal fumigation, sitz baths, topical applications, and oral administration, to deliver comprehensive therapeutic effects[9]. These effects include heat and dampness elimination, blood circulation enhancement to resolve stasis, and facilitation of debridement to promote tissue regeneration[10]. Hu-Lian-Bi-Guan decoction (HLBGD) is a classical formula composed of six herbs: Rhizoma Picrorhizae (Hu Huang Lian), Radix Ampelopsis (Bai Lian), Concha Haliotidis (Sheng Shi Jue Ming), Flos Sophorae (Huai Hua), Radix Angelicae Sinensis (Dang Gui), Raw Radix Astragali (Sheng Huang Qi), and Spina Gleditsiae (Zao Jiao Ci). The prescription has multifaceted therapeutic effects, including clearing heat, eliminating dampness, stimulating blood circulation, tonifying qi and blood, and astringing wounds for tissue recovery. Building on this foundation, we developed a modified HLBGD (mHLBGD) by adding Spina Gleditsiae (Zao Jiao Ci). This modification was performed to enhance the pharmacological effect of “unblocking collaterals” and ensure the delivery of more targeted therapeutic action to the local wound site. The aim of the study was to systematically evaluate the effects of an optimized formulation (mHLBGD) on postoperative wound healing and elucidate the underlying biological mechanisms.

MATERIALS AND METHODS

Study design

Patients were grouped stochastically in a 1:1 ratio using a random number table. The allocation sequence was concealed using the “envelope method”. Patients were either assigned to the treatment group (n = 28) or the control group (n = 28).

Patient recruitment and selection criteria

Based on preliminary confirmation of the effectiveness of HLBGD in complex AF, this randomized controlled trial was approved by the Medical Ethics Committee of the China-Japan Friendship Hospital (Approval No. 2023-KY-363). All patients provided written informed consent before enrollment. The patients who participated in the study met the inclusion and exclusion criteria. The inclusion criteria are as follows: (1) Age 18-70 years old; (2) Meeting the classification criteria for complex AF according to the Clinical Guidelines for the Diagnosis and Treatment of Anal Fistula[11]; (3) Meeting the diagnostic criteria for the TCM syndrome of downward flow of damp heat in AF per the Guidelines for the Diagnosis and Treatment of Common Diseases in Anorectal Medicine of TCM[12]; and (4) Fistulectomy was the surgical method used. The exclusion criteria are as follows: (1) The presence of immunological disorders, metabolic diseases, malignant tumors, or coagulation dysfunction; (2) Concomitant anorectal diseases known to affect wound healing; (3) Poor compliance; (4) Concomitant perianal skin diseases; (5) Presence of psychiatric disorders; (6) History of perianal surgery; (7) Pregnancy or lactation; (8) Presence of sexually transmitted or other infectious diseases; and (9) Morphological or structural abnormalities in the anus.

Treatment method

All patients underwent surgical treatment for AF performed by the same surgical team under total intravenous anesthesia. During surgery, fistulotomy was performed on low-level tracts and seton placement on high-level tracts. Rubber band drains were used to formulate counter-drainage between the main tract and secondary openings, which guarantees credible, sustained drainage. In the control group, all patients underwent postoperative routine antibiotic therapy. Dressing changes were also performed daily. The wound care protocol consisted of irrigating the wound with a chlorhexidine acetate solution, disinfecting the surrounding area with povidone-iodine, and applying Vaseline gauze strips locally. Subsequently, the site was covered with a sterile gauze and immobilized using surgical tape. In the treatment group, apart from the standard care provided for the controls, the patients received oral mHLBGD. The herbal formulation was administered orally twice daily. All decoctions were centrally fabricated and allocated to participants following the standardized protocols of the Department of Pharmacy at the China-Japan Friendship Hospital. The intervention period was 2 weeks for both groups.

Indicator monitoring

Baseline characteristics: The patients’ baseline characteristics included age, sex, total wound length (cm), and baseline Wexner incontinence score (WIS).

Efficacy outcomes: The outcome measures were as follows: (1) For wound-healing rate, in which the extent of wound contraction and tissue recovery was evaluated, the total wound length was determined by measuring the full extent of each fistulous tract using a sterile cotton swab and summing the tract lengths. The rate was calculated using postoperative day 1 as the baseline: (Baseline length - length at follow-up)/baseline length × 100%. Assessments were performed on postoperative days 14, 28, and 42; (2) To determine postoperative pain and Wound Exudate Scores, a Visual Analog Scale (VAS; 0-10) was used to assess postoperative pain. The wound exudate was quantified based on the number of gauze layers permeated by drainage 12 hours after dressing changes. Both measures were assessed on postoperative days 1, 7, 14, and 28; (3) The Wound Edema Score reflected the severity of postoperative inflammatory swelling at the wound margin and was graded from 0-3 (0: No edema; 1: Mild edema including ≤ 1/3 of the wound edge; 2: Moderate edema including > 1/3 to ≤ 2/3; 3: Severe edema including > 2/3 of the wound edge). Edema was evaluated on postoperative days 1, 7, and 14; (4) The WIS was used to assess anorectal continence function, which comprises five items with a total score between 0 and 20. Elevated scores indicate worse continence. Measurements were obtained at baseline and on postoperative days 14, 28, and 42; and (5) Time for complete wound healing was used to determine full epithelial coverage with a stable scar and the absence of local discomfort. Participants were followed until complete healing or up to 60 days postoperatively; wounds that did not heal by day 60 were recorded as 60 days. This outcome was monitored continuously from surgery until completion of healing or day 60, whichever occurred first.

Experimental methods

Clinical sample collection: Preoperative tissue specimens were obtained intraoperatively from the fistulous internal opening at the dentate line, ensuring the inclusion of the fistula wall and trimming the superficial squamous epithelium. The specimens were collected from the same anatomical site and labeled on postoperative day 14. They comprised freshly formed granulation tissue, and the anal canal mucosa was carefully removed. Subsequently, sterile saline was used to wash the specimens. Specifically, a small portion was immobilized in 4% paraformaldehyde for histological analysis, and the remaining tissue was snap-frozen under liquid nitrogen, incubated for half a minute in RNase-free conditions at -80 °C to extract RNAs and proteins. Each patient donated corresponding preoperative and postoperative tissue specimens.

Histopathological analysis: Formalin-fixed tissues were dehydrated, cleared, and embedded in paraffin to obtain blocks. The cells were sliced and stained. Hematoxylin-eosin (HE) staining was performed, and histopathological changes were examined using light microscopy. This evaluation highlighted inflammatory cell infiltration, granulation tissue progression, and interstitial fibrosis. Digital micrographs were acquired and archived using dedicated image analysis software.

Quantitative real-time PCR: Total RNA was isolated from frozen tissue specimens using the TRIzol method, and complementary DNA was synthesized following the manufacturer’s specifications in the reverse transcription kit. Quantitative real-time PCR was performed to assess mRNA expression levels of pro-inflammatory cytokines (IL-1β and IL-6), key components of the TGF-β/Smad signaling pathway (TGF-β and Smad2), and EMT-related markers, such as E-cad, N-cad, and VIM.

Western blot: Tissue samples were digested on ice in radioimmunoprecipitation assay buffer containing protease and phosphatase suppressors. Following centrifugation, the supernatant was collected, and protein concentrations were measured using the bicinchoninic acid method and normalized across specimens before denaturation by boiling. Equal proteins (20-30 μg per lane) were isolated through sodium dodecyl sulfate-polyacrylamide gel electrophoresis and delivered to polyvinylidene difluoride membranes under wet-transfer conditions at 4 °C. Membranes were sealed with 5% nonfat milk for 2 hours and incubated all night at 4 °C using primary antibodies against IL-1β, IL-6, TGF-β, Smad2, E-cad, N-cad, and VIM (1:1000). After rinsing with TBST, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000) for 1 hour at room temperature. The protein bands were visualized using improved chemiluminescence and imaged using a chemiluminescent detection system. Densitometric analysis was performed using ImageJ software (version 2.2.0, National Institute of Health, Bethesda, MD, United States).

Statistical analysis

All statistical analyses were performed using R software (v 4.4.1, R Foundation for Statistical Computing, Vienna, Austria). Categorical variables are summarized as frequencies and percentages, and between-group comparisons were performed using the χ² test; Fisher’s exact test was used when any expected cell count was < 5. Continuous variables are expressed as mean ± SD, and compared using an independent-samples t-test after verifying the assumptions of normality and homogeneity of variance. Survival results were compared using the Kaplan-Meier method, and differences between groups were evaluated using the log-rank test. Repeated-measures data were analyzed using repeated-measures analysis of variance (ANOVA) and Greenhouse-Geisser correction when the sphericity assumption was violated. All the tests were two-sided. Statistical significance was set at P < 0.05.

RESULTS

Baseline characteristics comparison

No remarkable differences were observed in the baseline features between the groups, including age, sex, WIS, and total wound length (P > 0.05; Table 1).

Table 1 Demographic and clinical information of each group, n (%).

Items	Total patients, n = 56	Control group, n =28	mHLBGD group, n =28	P value
Age (year)	37.32 ± 11.67	36.29 ± 11.78	38.36 ± 11.68	0.5
Gender				0.9
Male	49 (88)	25 (89)	24 (86)
Female	7 (13)	3 (11)	4 (14)
Wexner score				0.4
0	54 (96)	27 (96)	27 (96)
2	1 (1.8)	1 (3.6)	0 (0)
3	1 (1.8)	0 (0)	1 (3.6)
Total wound length (cm)	12.17 ± 5.69	11.59 ± 4.82	12.68 ± 6.42	0.5

mHLBGD: Modified Hu-Lian-Bi-Guan decoction.

Comparison of wound healing rate

The wound-healing rate was consistently higher in the treatment group than in the control group on postoperative days 14, 28, and 42. Repeated-measures ANOVA revealed significant between-group differences, within-group changes over time, and group-by-time interaction (all P < 0.001), as shown in Table 2.

Table 2 Comparison of wound healing rate in two groups (%).

Group	Postoperative days 14	Postoperative days 28	Postoperative days 42
Control (n = 28)	23 ± 6.3	35 ± 13.48	46 ± 15.74
mHLBGD (n = 28)	32 ± 5.73c	40.8 ± 11.82c	53.53 ± 17.92a,b,c
F	Fgroup = 27.950, Ftime = 28.258, Finteraction effect = 27.755
P value	Pgroup < 0.001, Ptime < 0.001, Pinteraction effect < 0.001

^aP < 0.05 vs postoperative day 14.

^bP < 0.05 vs postoperative day 28.

^cP < 0.05 vs the control group.

mHLBGD: Modified Hu-Lian-Bi-Guan decoction.

Comparison of postoperative pain score and Wound Exudate Score

VAS or Wound Exudate Scores did not significantly differ between groups (P > 0.05) on postoperative day 1. Meanwhile, on postoperative days 7, 14, and 28, the VAS and exudate scores were consistently lower in the treatment group than in the control group. Repeated-measures ANOVA revealed significant between-group differences and within-group changes over time (all P < 0.05) as well as a significant group-by-time interaction (P < 0.001), as shown in Table 3.

Table 3 Comparison of postoperative pain score, Wound Exudate Score and Wound Edema Score in two groups.

Outcome	Group	Postoperative days 1	Postoperative days 7	Postoperative days 14	Postoperative days 28	F	P
VAS	Control, (n = 28)	6.71 ± 0.46	5.89 ± 0.88a	2.96 ± 0.69a,b	1.61 ± 0.74a,b,c	Fgroup = 78.890, Ftime = 108.494, Finteraction effect = 20.671	Pgroup < 0.001, Ptime < 0.001, Pinteraction effect < 0.001
	mHLBGD, (n = 28)	7.39 ± 0.69	3.61 ± 0.88a,d	0.96 ± 0.92a,b,d	0.32 ± 0.48a,b,c,d
Wound Exudate Score	Control, (n = 28)	4.89 ± 0.57	5.50 ± 0.69a	3.68 ± 0.77a,b	2.86 ± 0.65a,b,c	Fgroup = 104.532, Ftime = 113.432, Finteraction effect = 114.272	Pgroup < 0.001, Ptime < 0.001, Pinteraction effect < 0.001
	mHLBGD, (n = 28)	5.36 ± 0.68	4.43 ± 0.63a,d	2.89 ± 0.69a,b,d	1.04 ± 0.39a,b,c,d
Wound Edema Score	Control, (n = 28)	5.43 ± 1.07	5.25 ± 0.80a	3.86 ± 0.36a,b	/	Fgroup = 25.963, Ftime = 45.106, Finteraction effect = 142.316	Fgroup < 0.001, Ftime < 0.001, Finteraction effect < 0.001
	mHLBGD, (n = 28)	5.81 ± 0.69	3.36 ± 0.49a,d	1.82 ± 0.39a,b,d	/

^aP < 0.05 vs postoperative day 1.

^bP < 0.05 vs postoperative day 7.

^cP < 0.05 vs postoperative day 14.

^dP < 0.05 vs the control group.

VAS: Visual Analog Scale; mHLBGD: Modified Hu-Lian-Bi-Guan decoction.

Comparison of Wound Edema Score

Wound Edema Scores did not differ significantly between the groups (P > 0.05) on postoperative day 1. On postoperative days 7 and 14, the treatment group consistently showed lower edema scores than the control group did. Repeated-measures ANOVA revealed significant between-group differences, within-group changes over time, and group-by-time interaction (all P < 0.001), as shown in Table 3.

Comparison of WIS

At enrollment and on postoperative days 28 and 42, no significant differences in WIS were observed between the two groups (P > 0.05). The repeated-measures ANOVA showed no significant between-group effects, time effects, or group-by-time interactions (all P > 0.05), as shown in Table 4.

Table 4 Comparison of Wexner incontinence score in two groups.

Group	Enrollment day	Postoperative days 28	Postoperative days 42
Control (n = 28)	0.07 ± 0.38	0.11 ± 0.31	0.04 ± 0.20
mHLBGD (n = 28)	0.11 ± 0.57	0.22 ± 0.70	0.04 ± 0.20
F	Fgroup = 0.597, Ftime = 1.056, Finteraction effect = 0.229
P value	Pgroup = 0.443, Ptime = 0.352, Pinteraction effect = 0.796

mHLBGD: Modified Hu-Lian-Bi-Guan decoction.

Comparison of time to complete wound healing

Follow-up analysis showed that the Kaplan-Meier-estimated median times to wound healing were 32 and 50 days in the treatment and control groups, respectively. As illustrated by the Kaplan-Meier healing curve, the treatment group achieved significantly faster wound healing than the control group did (P < 0.001; Figure 1).

Figure 1 Kaplan-Meier analysis of time to wound healing in the modified Hu-Lian-Bi-Guan decoction group compared with the control group. mHLBGD: Modified Hu-Lian-Bi-Guan decoction.

Histopathological results

On postoperative day 14, HE staining revealed granulation tissue formation in both groups; however, marked discrepancies were observed in the degree of inflammation and tissue maturation. The control group exhibited dense neutrophilic infiltration, prominent stromal edema, red blood cell extravasation, and irregularly formed neovessels with loosely organized granulation tissue. Conversely, the treatment group showed substantially fewer inflammatory cells, reduced edema, and more uniform mature granulation tissue. Fibroblasts and newly formed capillaries were arranged in a more orderly pattern, and areas of reparative epithelial hyperplasia were evident, collectively reflecting a more advanced stage of wound healing relative to the controls, as illustrated in Figure 2.

Figure 2 Histopathological results of postoperative wound tissue in two groups (hematoxylin-eosin staining). Low-power (left, scale bar = 20 μm) and high-power (right, scale bar = 10 μm). mHLBGD: Modified Hu-Lian-Bi-Guan decoction.

Comparison of mRNA expression of inflammatory cytokines, TGF-β/Smad signaling molecules, and EMT markers

As shown in Figure 3, baseline mRNA expression levels of IL-1β and IL-6, TGF-β and Smad2, and E-cad, N-cad, and VIM were not obviously different between the two groups (all P > 0.05). By postoperative day 14, IL-1β and IL-6 expression levels were substantially higher in controls than in the treatment group (t = 3.112, P = 0.008; t = 2.545, P = 0.023, respectively). Similarly, mRNA levels of TGF-β and Smad2 were markedly elevated in controls relative to the treatment group (t = 3.166, P = 0.007; t = 2.654, P = 0.019, respectively). Regarding EMT-related genes, E-cad expression was significantly higher in the treatment group (U = 1, P < 0.001), whereas N-cad and VIM levels were significantly higher in the control group (t = 2.503, P = 0.025; t = 3.886, P = 0.002, respectively).

Figure 3 Comparison of mRNA expression of inflammatory cytokines, TGF-β/Smad signaling molecules, and epithelial-mesenchymal transition markers in two groups. ^aP < 0.05. Pre: Pre-treatment; Post: Post-treatment; mHLBGD: Modified Hu-Lian-Bi-Guan decoction; N-cad: N-cadherin; VIM: Vimentin; E-cad: E-cadherin.

Comparison of protein expression levels of inflammatory cytokines, TGF-β/Smad signaling molecules, and EMT markers

Protein expression levels of IL-1β, IL-6, TGF-β, Smad2, E-cad, N-cad, and VIM were not apparently different between the two groups at baseline (all P > 0.05). By postoperative day 14, IL-1β and IL-6 protein levels were markedly lower in the treatment group than in controls (both P < 0.01), and expression of TGF-β and Smad2 was likewise significantly reduced in the treatment group (both P < 0.001). Among the EMT-related proteins, N-cad and VIM were downregulated in the treatment group (P < 0.05, P < 0.001, respectively), whereas E-cad expression was upregulated (P < 0.05), as shown in Figure 4.

Figure 4 Comparison of protein expression levels of inflammatory cytokines, TGF-β/Smad signaling molecules, and epithelial-mesenchymal transition markers in two groups. A: Relative protein expression levels of each group; B: Western blot analysis of wound tissue protein expression. ^aP < 0.05. Pre: Pre-treatment; Post: Post-treatment; Con: Control; mHLBGD: Modified Hu-Lian-Bi-Guan decoction; N-cad: N-cadherin; VIM: Vimentin; E-cad: E-cadherin.

DISCUSSION

In complex AF, postoperative wound healing still presents a large clinical challenge, such as low healing proportions, delayed healing time, escalated incidence of complications, and a substantial risk of relapse. In some patients, the duration of wound healing may exceed 6-12 weeks. Postponed healing lengthens overall recovery and increases the risk of relapse, while undermining quality of life and increasing financial burdens[13]. Postoperative wounds are often large and broadly exposed, requiring a longer duration for granulation tissue ingrowth and reepithelialization. Concurrently, prolonged exposure of the wound to fecal contamination and inflammatory exudates creates a highly complex biological microenvironment for healing. Refining the postoperative wound microenvironment for healing is an urgent clinical priority in colorectal surgery.

This study revealed that mHLBGD can remarkably boost postoperative wound healing in patients with complex AF. The potential mechanisms may be twofold: (1) mHLBGD downregulates pro-inflammatory cytokine expression such as IL-1β and IL-6, thereby attenuating excessive inflammatory responses; and (2) It modulates the TGF-β/Smad signaling pathway to inhibit excessive activation of EMT, ultimately ameliorating wound repair. Normal wound healing is an orchestrated cascade comprising four overlapping stages: Hemostasis, inflammatory reactions, proliferation, and remodeling[14]. In the inflammatory stage, clearing pathogens and necrotic tissue and secreting diverse cytokines facilitates angiogenesis and cell transfer. This stage lays the foundation for future proliferative and remodeling stages[15]. Nevertheless, delayed or excessive inflammation impairs the migration and efficacy of reparative cells, such as fibroblasts and keratinocytes. This disruption blocks key transitions from the inflammatory stage to the proliferative phase, ultimately prolonging healing[16,17].

IL-1β and IL-6 are critical proinflammatory cytokines during the inflammatory stage of wound healing. Elevated IL-1β not only induces its own synthesis and activates other cytokines, such as TNF-α, but also inhibits M2 macrophage polarization and granulation tissue formation, thereby trapping the wound in a state of chronic inflammation[18-20]. This process prolongs healing or leads to refractory chronic wounds occurring. In our study, mHLBGD treatment significantly reduced IL-1β and IL-6 expression levels in wound tissues. This finding suggests that mHLBGD may reduce the inflammatory stage and promote early entry into the proliferative restoration stage. Contemporary pharmacological evidence indicates that extracts of Picrorhiza scrophulariiflora enhance IL-10 production and reduce IL-1β, IL-6, and TNF-α levels[21]. Additionally, these extracts inhibited inflammatory signaling pathways, including those involving NF-κB, iNOS, and COX-2. Consequently, these compounds exhibit significant immunomodulatory and anti-inflammatory properties. Notably, oxidative stress and tissue damage are mitigated. These findings provided a pharmacological basis for the observed outcomes.

EMT plays a dual role in wound healing. During physiological repair, keratinocytes undergo a reversible “partial EMT”, acquiring migratory potential while retaining essential epithelial traits to facilitate rapid re-epithelialization[18,19]. Conversely, “pathological EMT”, sustained by chronic inflammation or persistent TGF-β signaling, impairs wound healing. This aberrant overactivation confers apoptotic resistance to cells and drives excessive collagen and matrix deposition, culminating in fibrosis and scarring[3,20,22]. Our study shows that mHLBGD effectively reverses this pathological trajectory. Molecular analysis (PCR/Western blot) revealed restoration of E-cad expression and significant downregulation of N-cad and VIM expression in the treatment group. Collectively, these findings suggest that mHLBGD promotes high-quality wound repair by curbing excessive EMT and preserving epithelial phenotype stability.

Crucially, our study shows that the TGF-β/Smad signaling pathway is the mechanistic nexus bridging inflammation and aberrant EMT. We observed a marked downregulation of TGF-β and Smad2 expression in the mHLBGD-treated group. It is well-established that inflammatory cytokines, such as IL-1β and IL-6, can induce TGF-β secretion and receptor upregulation. This process results in a positive feedback loop that sustains Smad2/3 phosphorylation and drives pathological EMT[18,19,23,24]. By simultaneously suppressing pro-inflammatory cytokines and inhibiting TGF-β/Smad signaling, mHLBGD likely disrupts this “inflammation-TGF-β/Smad-EMT” axis. This dual action prevents signal amplification responsible for pathological fibrosis, thereby accelerating wound closure and attenuating fibrotic progression.

This study has some limitations. First, the sample sizes were comparatively small, limited by factors such as the research duration and patient adherence. In addition, serial tissue biopsies in the internal opening of a clinic pose remarkable challenges. Unlike prior animal studies, we were confined to constructing tissues at a single key time point (postoperative day 14). Furthermore, we observed significant changes in the expression of EMT-related markers and in the TGF-β/Smad signaling pathway in clinical tissue samples; these findings reflect a correlation rather than a confirmed causal relationship. Therefore, the specific regulatory role of mHLBGD in this pathway warrants further investigation. Additionally, the follow-up period was limited to 60 days, which may be insufficient to detect delayed healing in a small subset of patients or evaluate long-term recurrence rates. Therefore, future research with extended follow-up periods is necessary to substantiate the long-term efficacy and recurrence-prevention potential of mHLBGD. Moreover, follow-up surveys should focus on addressing these problems using multicenter, large-scale clinical tests. Animal studies are required to clarify the molecular mechanisms underlying mHLBGD. Therefore, a more robust basis is warranted for clinical practice.

CONCLUSION

In summary, mHLBGD showed robust clinical efficacy in patients undergoing surgery for complex AF, mitigating pain, edema, and wound exudation, and improving wound healing synergistically. Its potential mechanisms may include inhibiting excessive inflammatory reactions, as shown by downregulation of IL-1β and IL-6. Significantly, the regulation of the TGF-β/Smad signaling pathway suppressed aberrant EMT activation, enhancing the quality of tissue recovery. Our research provides valuable insights into the biological mechanisms by which mHLBGD facilitates postoperative wound healing in complex AF.

ACKNOWLEDGEMENTS

We would like to thank the clinical team for their dedicated support in recruiting patients and collecting biological samples.