Ziwu-Liuzhu-guided combined external therapy promotes wound healing in postoperative perianal fistulizing Crohn’s disease

Abstract

BACKGROUND

Perianal fistulizing Crohn’s disease (PFCD) is one of the most destructive complications of inflammatory bowel disease. Delayed postoperative wound healing, persistent inflammation, and a high risk of recurrence remain major challenges in clinical management. In recent years, external traditional Chinese medicine therapies have attracted increasing attention in postoperative recovery from perianal disorders, and time-specific intervention based on chronomedicine principles, such as the Ziwu-Liuzhu theory, may further optimize therapeutic outcomes.

AIM

To evaluate the effects of Ziwu-Liuzhu-guided combined external therapy on wound healing, disease activity, inflammation, and anxiety in postoperative PFCD patients.

METHODS

This single-center retrospective cohort study included 217 postoperative PFCD patients (2023-2025). They were divided into sitz bath-only (n = 72), sitz bath plus acupoint application (n = 69), and Ziwu-Liuzhu time-specific combined therapy (n = 76) groups. Outcomes included Crohn’s Disease Activity Index (CDAI), Perianal Disease Activity Index (PDAI), wound healing scores, exudate, pain, granulation, healing time, C-reactive protein (CRP), tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6), and Generalized Anxiety Disorder-7 scale (GAD-7). Between-group comparisons, regression, survival analysis, and propensity score matching were performed.

RESULTS

After 4 weeks, all three groups showed significant improvements in CDAI and PDAI, with the greatest reductions observed in the Ziwu-Liuzhu group. This group also exhibited significantly lower wound healing, exudate, pain, and granulation scores, along with a markedly shorter median healing time (18 days vs 24 days and 31 days). Kaplan-Meier analysis demonstrated a higher wound healing probability in the Ziwu-Liuzhu group (log-rank P < 0.001). Levels of CRP, TNF-α, and IL-6 were significantly lower in the Ziwu-Liuzhu group than in the other two groups. The reduction in GAD-7 score was also greatest in this group. Multivariable regression and propensity score matching consistently confirmed that Ziwu-Liuzhu therapy was independently associated with superior healing and overall clinical efficacy.

CONCLUSION

Ziwu-Liuzhu-guided combined external therapy promotes wound healing, reduces inflammation and anxiety, and improves disease activity in postoperative PFCD, representing a valuable adjunctive integrated management approach.

Key Words: Perianal fistulizing Crohn’s disease; Ziwu-Liuzhu; External traditional Chinese medicine therapy; Wound healing; Inflammatory response; Anxiety status

Core Tip: Perianal fistulizing Crohn’s disease (PFCD) is difficult to manage after surgery because delayed wound healing and persistent inflammation are common. In this retrospective cohort study, a Ziwu-Liuzhu-guided time-specific traditional Chinese medicine external therapy was associated with better postoperative recovery than sitz bath alone or non-time-specific combined therapy. Patients in the time-specific group showed faster wound healing, lower inflammatory marker levels, greater improvement in Crohn’s Disease Activity Index and Perianal Disease Activity Index scores, and better anxiety outcomes. These findings suggest that treatment timing may be an important adjunct in integrated postoperative PFCD management.

INTRODUCTION

Crohn’s disease (CD) is a major subtype of inflammatory bowel disease characterized by chronic relapsing progression, progressive tissue injury, and the need for long-term management, and it has become an increasing global public health burden[1]. Recent epidemiological studies have shown that the prevalence, incidence, and disability burden of inflammatory bowel disease remain high worldwide, with continued increases in regions with low and middle sociodemographic development indices[2]. Population-based data further indicate that a substantial proportion of patients already experience moderate-to-severe functional impairment and markedly reduced health-related quality of life at the time of diagnosis[3]. Among CD-related complications, perianal fistulizing CD (PFCD) is one of the most destructive phenotypes, and is closely associated with persistent pain, discharge, repeated surgery, impaired quality of life, and possible long-term malignant risk[4-6].

PFCD remains particularly challenging because it is not merely a localized fistulous lesion, but rather a complex condition involving multiple therapeutic goals, including infection control, fistula closure, remission of rectal inflammation, preservation of sphincter function, and long-term improvement in quality of life[7]. Although biologics, small-molecule therapies, and mesenchymal stem cell-based strategies have evolved substantially in recent years, the overall certainty of evidence for PFCD treatment remains limited, and high-quality studies focusing specifically on fistula healing are still insufficient[8]. Moreover, currently available PFCD evaluation systems, including clinical scores, patient-reported outcomes, and magnetic resonance imaging-based tools, still lack a unified standard with satisfactory validity, reliability, and responsiveness[9,10]. Recent consensus statements have further emphasized the marked heterogeneity of PFCD and the need for more individualized and integrated management strategies[11]. Therefore, identifying feasible adjunctive interventions that can improve the local pathological microenvironment and promote postoperative wound repair remains clinically important.

From a pathophysiological perspective, PFCD is not simply a localized infectious anal fistula, but rather the product of persistent inflammation, immune dysregulation, epithelial barrier disruption, and abnormal tissue remodeling. Increasing evidence indicates that epithelial-mesenchymal transition, intestinal barrier injury, gut microbial imbalance, and chronic inflammatory cytokine activation all contribute to disease persistence and fistula formation[12-17]. For postoperative PFCD patients, an intervention capable of simultaneously suppressing inflammation, promoting barrier restoration, and facilitating tissue repair may therefore be particularly beneficial.

PFCD also imposes a considerable psychological burden. Depression and anxiety in inflammatory bowel disease are associated with increased risks of relapse, hospitalization, surgery, and treatment escalation[18]. Even under multimodal treatment strategies, many PFCD patients continue to experience persistent symptoms or ongoing disease activity[19,20]. Thus, assessment of postoperative recovery in PFCD should not be limited to local wound-related outcomes alone, but should also incorporate disease activity, inflammatory control, and psychological status.

Chronomedicine and traditional Chinese medicine (TCM) external therapy may provide a potentially valuable complementary strategy in this setting. Circadian rhythm disruption has been linked to inflammatory bowel disease relapse and progression through effects on microbiota-host interaction, mucosal immunity, and barrier integrity[21]. Meanwhile, acupuncture, moxibustion, herbal sitz baths, and acupoint application have shown potential value in regulating inflammation, promoting tissue repair, and improving postoperative symptoms[22-24]. However, the clinical value of time-specific external therapy in postoperative PFCD recovery remains insufficiently investigated.

Accordingly, the present study compared three postoperative treatment strategies in PFCD patients: Sitz bath with compound Qinbai granules alone, sitz bath plus conventional-time acupoint application of Kuijiening plaster, and fixed-time combined external therapy based on the Ziwu-Liuzhu theory. By evaluating wound healing, disease activity, inflammatory markers, psychological status, and overall clinical efficacy, this study aimed to assess the clinical value of time-specific TCM external therapy in postoperative PFCD management and to provide evidence supporting optimization of integrated postoperative care.

MATERIALS AND METHODS

Study design

This was a single-center retrospective observational cohort study based on real-world clinical data and conducted in The Second Affiliated Hospital of Hunan University of Chinese Medicine. The study population consisted of patients with PFCD who underwent surgery and completed perioperative management at the study center. By systematically reviewing historical clinical records, this study evaluated the associations between different postoperative TCM external treatment strategies and clinical recovery outcomes, with particular emphasis on wound healing-related outcomes, while also assessing inflammatory response, disease activity, psychological status, and overall clinical efficacy.

Study data were obtained from the hospital’s integrated information platform, including the electronic medical record system, nursing record system, surgical anesthesia system, laboratory information system, and outpatient and inpatient follow-up databases. All study variables were derived from routine clinical records and were extracted, integrated, cross-checked, and logically validated through a unified data management platform to ensure data completeness, accuracy, and traceability.

To ensure case continuity and stability of the study findings, the case inclusion period was defined from January 1, 2023 to November 30, 2025. All participants were consecutive PFCD patients who were admitted and surgically treated during this period. February 28, 2026 was used as the unified data cutoff date for follow-up completion, endpoint verification, and database locking. Within this time window, potential cases were retrieved through the hospital information system and systematically reviewed according to a predefined screening process. After removal of duplicate records, confirmation of diagnostic validity, assessment of the completeness of key clinical variables, and verification of follow-up information, a total of 217 PFCD patients were ultimately included in the analytic cohort.

Because this was a retrospective observational study, no treatment allocation was performed by the investigators. Instead, exposure was defined according to the postoperative external treatment strategy that patients had actually received during routine clinical practice. Treatment implementation and adherence were assessed retrospectively on the basis of cross-verified nursing execution records, treatment sheets, medication orders, and follow-up documentation. Only patients with mutually consistent records regarding treatment type, timing, and frequency were included in the final analysis, in order to improve the reliability of exposure ascertainment. Based on historical nursing records, treatment sheets, and related medical orders, the included cases were classified into three exposure groups: The conventional sitz bath group (72 cases), the sitz bath plus acupoint application group (69 cases), and the Ziwu-Liuzhu time-specific external treatment group (76 cases). These groups reflected the natural distribution of treatment decisions in real-world clinical practice rather than investigator-assigned interventions.

Inclusion and exclusion criteria

Inclusion criteria: (1) They fulfilled the diagnostic criteria for PFCD according to the “expert consensus on the diagnosis and treatment of Crohn’s disease-associated anal fistula”, and the diagnosis was jointly confirmed by two anorectal specialists with associate chief physician rank or above based on clinical manifestations, imaging findings, and relevant laboratory examinations; (2) They were aged between 18 years and 65 years; (3) They underwent fistula-related surgery at the study center, including but not limited to fistulotomy, seton drainage, or fistula debridement, and complete operative records were available; (4) They received standardized postoperative TCM external treatment during recovery, including sitz bath with compound Qinbai granules, acupoint application of Kuijiening plaster, or time-specific external therapy based on the Ziwu-Liuzhu theory; (5) The duration of TCM external intervention was at least 2 weeks, which was used as the minimum requirement for exposure classification; (6) For the primary outcome analysis, only patients who completed 4 weeks of postoperative follow-up and had evaluable week-4 outcome data were included; (7) Complete clinical data were available, including baseline demographic information, operative records, nursing records, and primary outcome-related data; and (8) They completed 4 weeks of postoperative follow-up to allow uniform assessment of the primary study outcomes at week 4.

Exclusion criteria: (1) Severe systemic diseases that could substantially affect wound healing or inflammatory responses, including severe hepatic insufficiency (Child-Pugh class C), end-stage renal disease, active malignancy, or severe hematologic disorders; (2) Pregnancy or lactation; (3) Severe psychiatric disorders or cognitive impairment, resulting in inability to cooperate with follow-up evaluation or inability to obtain reliable clinical data; (4) Definite allergy to the external herbal preparations or related ingredients involved in the study; (5) Receipt of other special therapies during the postoperative observation period that might significantly affect wound healing or inflammatory status, such as stem cell therapy, intensified biologic treatment, or other experimental regimens; (6) Repeat perianal surgery during postoperative follow-up; (7) Missing key clinical variables, including primary outcome indicators, treatment exposure information, or follow-up data, making valid statistical analysis impossible; and (8) Follow-up duration shorter than 4 weeks or loss to follow-up during the observation period.

Study procedures

Principles for treatment exposure determination: This study was retrospective and observational, and the investigators did not assign any treatment. All cases were grouped according to the postoperative TCM external nursing regimen actually received during routine clinical practice. To ensure the accuracy and traceability of exposure determination, perioperative clinical records were extracted from the hospital’s integrated information system, and electronic medical orders, nursing execution records, TCM nursing documents, treatment sheets, prescription records, and available follow-up records were cross-verified. In this retrospective study, treatment adherence was not measured prospectively; therefore, adherence was operationally determined according to whether the scheduled sitz bath sessions and plaster applications were documented as completed in routine nursing execution records, treatment sheets, and follow-up records. For hospitalized patients, adherence was mainly identified from nursing execution documentation; for patients managed after discharge, adherence was additionally verified on the basis of outpatient follow-up notes and patient-reported treatment records documented in the charts. A case was assigned to the corresponding exposure category only when records from different sources were mutually consistent and clearly documented the principal TCM external intervention received during postoperative recovery. Cases with incomplete adherence documentation, major discrepancies between prescribed and recorded treatment frequency, or inability to verify implementation status were not used for definitive exposure classification.

To reduce information bias in treatment exposure classification, a uniform procedure was used. First, nursing records within the first 4 postoperative weeks were systematically reviewed to determine whether TCM external therapy had been administered and in what form. In this process, a treatment duration of at least 2 weeks was used to define stable exposure, whereas week-4 data were used for outcome evaluation to ensure comparability across groups. Second, different treatment models were classified according to intervention type, timing, and duration. If a patient received more than one treatment modality during the observation period, the modality that accounted for the greatest proportion of treatment records and had the longest duration was defined as the principal treatment exposure. Cases with substantial discrepancies across records, inability to confirm the main treatment modality, or insufficient documentation of treatment implementation were excluded from the final analysis to ensure consistency of grouping and reliability of the findings.

Definition of conventional sitz bath exposure: Conventional sitz bath exposure was defined as postoperative herbal sitz bath treatment with compound Qinbai granules without concomitant acupoint application or other study-related TCM external interventions. Compound Qinbai granules were an in-house preparation produced by the hospital pharmacy and mainly consisted of Scutellaria baicalensis, Phellodendron chinense, Rehmannia glutinosa, the tail of Angelica sinensis, persica seed, corydalis, Areca catechu, Saposhnikovia divaricata, Gentiana macrophylla, Alisma orientale, and Rheum palmatum. The formula is traditionally used to clear heat, dry dampness, activate blood circulation, resolve stasis, regulate qi, and relieve pain, and it has been widely used in postoperative nursing care for anorectal inflammatory disorders to promote local inflammation resolution and wound recovery.

According to the hospital’s routine nursing pathway, 12 g of compound Qinbai granules were dissolved in approximately 1000 mL of warm water, with the temperature maintained at about 40 °C to ensure local tolerance and facilitate release of active ingredients. Patients received local fumigation and washing in the sitz-bath position for approximately 10-15 minutes each time, twice daily. This intervention was continued for 4 weeks during postoperative recovery. Cases that received only this sitz bath regimen throughout the observation period, without acupoint application or other TCM external interventions, were classified as conventional sitz bath exposure. In the retrospective exposure assessment, sitz bath adherence was judged according to documented completion frequency in nursing execution records during hospitalization and corresponding follow-up documentation after discharge.

Definition of sitz bath plus acupoint application exposure: Sitz bath plus acupoint application exposure was defined as postoperative care consisting of sitz bath treatment with compound Qinbai granules combined with acupoint application of Kuijiening plaster, but without a fixed Ziwu-Liuzhu time-specific administration pattern. Kuijiening plaster was a commonly used external TCM preparation at the study center, and its main ingredients included processed aconite, Asarum, clove, white mustard seed, corydalis, red peony root, and ginger juice. The formula is traditionally considered to warm yang, dispel cold, activate the channels and collaterals, and promote tissue repair, and it has been commonly used in postoperative anorectal care to improve local circulation and facilitate wound healing.

In routine nursing practice at the study center, Kuijiening plaster was usually applied to acupoints including Shangjuxu, Tianshu, Zusanli, Mingmen, and Guanyuan. The plaster was evenly placed at the center of the patch material, applied over the corresponding acupoints, and fixed in position. Each application lasted approximately 6 hours and was administered once daily for 4 consecutive weeks. Patients who received this combined treatment during postoperative recovery, but whose nursing records did not indicate a fixed time-specific pattern or showed considerable fluctuation in daily application timing, were classified as receiving conventional combined external treatment. Adherence to plaster application was retrospectively evaluated according to the recorded execution of patch application, removal time, and daily treatment frequency in nursing and follow-up records.

Definition of Ziwu-Liuzhu time-specific external treatment exposure: Ziwu-Liuzhu time-specific external treatment exposure was defined as treatment in which patients received sitz bath with compound Qinbai granules combined with acupoint application of Kuijiening plaster, with the acupoint application being carried out according to a relatively fixed schedule corresponding to the meridian qi circulation timing described in the Ziwu-Liuzhu theory. Based on previous nursing pathways and treatment execution records, this study defined cases in which plaster application was stably administered at approximately 18:00 and remained relatively consistent during the treatment course as belonging to the Ziwu-Liuzhu time-specific external treatment group. The choice of approximately 18:00 was based on the Ziwu-Liuzhu meridian-flow theory in TCM, according to which the period from 17:00 to 19:00 corresponds to the Foot Shaoyin Kidney Meridian[25]. The use of Ziwu-Liuzhu theory in time-based intervention has also been described in recent clinical and theoretical literature[26]. According to the theoretical framework of TCM chronomedicine and the Ziwu-Liuzhu theory, the circulation of qi and blood follows specific circadian characteristics, and different meridians are considered to exhibit relatively predominant activity at different times of day. In the conventional meridian clock system, the interval from 17:00 to 19:00 is assigned to the Foot Shaoyin Kidney Meridian. The time period corresponding to Kidney Meridian predominance is generally believed in TCM theory to be closely related to the storage of essence, regulation of water metabolism, and facilitation of recovery processes. Therefore, plaster application at approximately 18:00 was selected as the operational time point for the Ziwu-Liuzhu intervention in this study. Acupoint application during this period may therefore facilitate postoperative wound recovery by promoting meridian qi-blood circulation, enhancing transdermal absorption of active compounds, and improving the local immune microenvironment. In this study, the continuity and stability of plaster application timing were retrospectively reviewed in nursing execution records, and cases with clear fixed-time characteristics were assigned to the Ziwu-Liuzhu time-specific external treatment group.

Control of exposure consistency: To ensure accuracy of study grouping and consistency of exposure definition, uniform adjudication rules were established during case screening. When a patient received multiple treatment modalities during the observation period and the treatment records showed clear changes over time, the principal exposure was determined according to treatment duration and implementation frequency. Cases receiving sitz bath plus acupoint application but with incomplete timing records, or cases in which stable time-specific features could not be confirmed, were uniformly assigned to the sitz bath plus acupoint application group rather than the Ziwu-Liuzhu time-specific external treatment group, in order to improve the specificity of the time-specific exposure definition. Cases in which the principal treatment exposure could not be clearly determined, in which obvious discrepancies existed across records, or in which treatment adherence could not be adequately verified were excluded from the final analysis.

Through the above exposure definition and quality control procedures, the study established a treatment exposure classification system with clear operational definitions and reproducible adjudication criteria, while preserving real-world variation in clinical practice as much as possible and providing a reliable methodological basis for subsequent analyses.

Outcome measures

To systematically evaluate the associations between different postoperative TCM external treatment strategies and recovery in PFCD patients, this study assessed outcomes across multiple dimensions, including disease activity, wound healing, inflammatory response, psychological status, and overall clinical efficacy. All indicators were extracted from historical clinical records and assessed according to unified criteria to ensure comparability across cases.

Clinical symptom assessment: The Crohn’s Disease Activity Index (CDAI) was used to assess overall disease activity in CD and is one of the most widely used activity indices internationally. In this study, disease activity was evaluated according to the standard CDAI calculation method. The index was derived from weighted scoring of clinical symptoms and signs recorded over 7 consecutive days. In this retrospective study, the 7-day stool diary data were obtained from patient-completed symptom records documented in the medical charts or follow-up records; when such diaries were unavailable as separate forms, the corresponding symptom information was extracted from routine clinical records by the investigators according to predefined rules.

During scoring, the daily number of liquid stools was recorded and averaged over 7 days on the basis of the documented 7-day symptom diary or equivalent information available in the medical records. Abdominal pain severity was evaluated on a four-grade scale, where 0 indicated no abdominal pain, 1 mild pain without affecting daily activities, 2 moderate pains with some effect on daily activities, and 3 marked pain requiring analgesic measures or clearly impairing activity. Patients’ subjective general well-being was recorded on a five-grade scale, where 0 indicated generally well, 1 mild discomfort, 2 moderate discomfort, 3 obvious discomfort, and 4 severe discomforts. Extraintestinal manifestations, including arthralgia, skin lesions, and ocular inflammation, were also recorded and scored according to predefined criteria. Hematocrit, abdominal mass, and body weight change were additionally incorporated into the scoring system. All variables were weighted and summed to yield the total CDAI score. According to standard criteria, a CDAI score < 150 indicates remission, 150-450 indicates active disease, and > 450 indicates severe disease activity. Higher scores indicate greater disease activity.

Wound healing indicators: To comprehensively evaluate postoperative wound recovery, this study systematically assessed wound healing status, exudate, pain severity, granulation tissue growth, and wound healing time. All scores were determined according to unified criteria by two medical staff members experienced in anorectal nursing based on the nursing records. Inter-rater agreement for the wound healing score, wound exudate score, and granulation tissue growth score was evaluated using Cohen’s kappa statistic based on the independent ratings of the two assessors before final adjudication. (1) Wound healing score: The wound healing score was used to comprehensively assess local wound recovery. Scoring was based mainly on the degree of redness and swelling, tissue repair, signs of infection, and exudate status. The score ranged from 0 to 4. A score of 0 indicated good wound recovery, with a dry wound surface, no obvious redness, swelling, or signs of infection, and uniform bright-red granulation tissue. A score of 1 indicated mild redness or edema without obvious infection, slight wound moisture, and basically normal granulation growth. A score of 2 indicated moderate redness and swelling with a small amount of exudate and relatively slow granulation growth. A score of 3 indicated obvious redness and swelling with more exudate and uneven granulation growth. A score of 4 indicated obvious infection, marked local redness and swelling, purulent discharge, and impaired granulation growth. Higher scores indicated poorer wound healing; (2) Wound exudate score: The wound exudate score was used to assess wound seepage. The score was determined according to the extent to which wound exudate permeated the dressing. A score of 0 indicated extremely little exudate, with only slight dressing moisture and no obvious penetration. A score of 2 indicated that wound exudate could infiltrate approximately four layers of gauze. A score of 4 indicated infiltration of approximately eight layers of gauze. A score of 6 indicated a large amount of exudate sufficient to soak through more than sixteen layers of gauze. Higher scores indicated greater wound exudation; (3) Wound pain score: The Visual Analogue Scale was used to assess postoperative wound pain on a 0-10 scale. A score of 0 indicated complete absence of pain. Scores of 1-3 indicated mild pain with little effect on daily activities. Scores of 4-6 indicated moderate pain with some interference with activities. Scores of 7-10 indicated severe pain that clearly interfered with daily life or required analgesics. Higher scores indicated more severe pain; (4) Granulation tissue growth score: The granulation tissue growth score was used to evaluate wound tissue repair and ranged from 0 to 3. A score of 0 indicated good granulation growth with uniform coverage, a bright-red wound surface, and no obvious necrotic tissue. A score of 1 indicated basically normal granulation growth but with a small amount of necrotic tissue remaining. A score of 2 indicated relatively slow and uneven granulation growth. A score of 3 indicated markedly restricted granulation growth with a relatively large amount of necrotic tissue still present. Higher scores indicated poorer tissue repair; and (5) Wound healing time: Wound healing time was defined as the interval from surgery completion to complete epithelialization of the wound without exudative discharge, recorded in days. Shorter wound healing time indicated faster recovery.

Inflammatory markers: (1) C-reactive protein (CRP): CRP is an important marker of acute inflammatory response. Fasting venous blood samples were collected before and after treatment for CRP measurement. Elevated CRP generally indicates increased inflammatory burden; (2) Tumor necrosis factor-α (TNF-α): TNF-α is an important proinflammatory cytokine that plays a key role in the immune-inflammatory response of inflammatory bowel disease. Serum TNF-α levels were measured to assess the degree of inflammatory activation, with higher levels indicating stronger inflammatory activity; and (3) Interleukin-6 (IL-6): IL-6 is one of the key cytokines involved in inflammatory responses. Serum IL-6 levels were measured to evaluate changes in inflammatory status, with higher levels indicating more active inflammation. All inflammatory markers were measured using enzyme-linked immunosorbent assay.

Psychological status assessment: The Generalized Anxiety Disorder-7 (GAD-7) scale was used to assess the severity of anxiety symptoms. The scale includes 7 items, each scored according to symptom frequency over the previous 2 weeks: 0 for not at all, 1 for several days, 2 for more than half the days, and 3 for nearly every day. Total scores range from 0 to 21. According to standard cutoffs, scores of 0-4 indicate no obvious anxiety symptoms, 5-9 mild anxiety, 10-14 moderate anxiety, and 15-21 severe anxiety. Higher scores indicate greater severity of anxiety.

Overall clinical efficacy evaluation: Overall clinical efficacy was determined on the basis of wound healing status and the Perianal Disease Activity Index (PDAI). Efficacy was classified into three categories: Clinical cure, effective, and ineffective. Clinical cure was defined as complete wound healing without exudative discharge and a PDAI score of 0. Effective was defined as obvious wound contraction, marked reduction in exudation, and a PDAI score ≤ 4. Ineffective was defined as no obvious improvement in wound recovery or worsening symptoms together with a PDAI score > 4. The overall response rate was calculated as the number of clinically cured plus effective cases divided by the total number of cases, multiplied by 100%.

Statistical analysis

All statistical analyses were performed using SPSS 26.0 (IBM Corp., Armonk, NY, United States) and R software (version 4.3.0). All statistical tests were two-sided, and a P value < 0.05 was considered statistically significant. Continuous variables were first assessed for normality using the Shapiro-Wilk test. Data conforming to a normal distribution were expressed as mean ± SD, whereas non-normally distributed variables were presented as median with interquartile range. Categorical variables were described as frequencies and percentages. For between-group comparisons of continuous variables, one-way analysis of variance was used when data met assumptions of normality and homogeneity of variance, followed by Bonferroni correction for post hoc pairwise comparisons. When these assumptions were not met, the Kruskal-Wallis rank-sum test was applied. For categorical variables, the Pearson χ² test was used, and Fisher’s exact test was employed when the expected frequency in some cells was < 5. Inter-rater agreement for wound healing-related ordinal scores, including wound healing score, wound exudate score, and granulation tissue growth score, was assessed using Cohen’s kappa statistic. Kappa values were interpreted according to conventional criteria, with higher values indicating better agreement.

To evaluate the associations between different postoperative TCM external treatment strategies and clinical outcomes while controlling for potential confounding factors, multivariable regression models were constructed. For continuous outcomes, such as changes in inflammatory marker levels or score changes, multiple linear regression models were used. For binary outcomes, such as overall clinical efficacy, multivariable logistic regression models were applied. Covariates included age, sex, disease duration, preoperative disease activity scores, and other major clinical characteristics that could influence postoperative recovery. Adjusted effect estimates and corresponding 95% confidence intervals (CIs) were calculated.

Because wound healing time was a time-to-event variable, survival analysis was also performed. Kaplan-Meier methods were used to generate wound healing curves for different treatment groups, and the log-rank test was used to compare wound healing times across groups. Cox proportional hazards regression models were then constructed to estimate the associations between treatment strategy and wound healing hazard, with hazard ratios and 95%CIs reported. Potential confounders were included in the Cox models to obtain adjusted estimates.

Given the retrospective observational nature of the study and the possibility of baseline imbalance among treatment groups, propensity score matching was conducted as a supplementary analysis. Propensity scores were estimated using logistic regression, with covariates including age, sex, disease duration, preoperative disease activity scores, and other major baseline characteristics. After calculation of propensity scores, nearest-neighbor matching was performed using a caliper width of 0.2 standard deviations. The matched cohort was then reanalyzed using the same statistical framework to test the robustness of the primary findings.

Sensitivity analyses were additionally performed by comparing the results from the original cohort and the propensity score-matched cohort, in order to evaluate the influence of potential confounding on the study conclusions. Effect estimates and 95%CIs were reported for the primary analyses to improve interpretability and clinical relevance.

Ethics statement

This study was approved by the ethics committee of The Second Affiliated Hospital of Hunan University of Chinese Medicine (No. 2025-KY-060-01). The study was conducted in strict accordance with the ethical principles of the Declaration of Helsinki and relevant international ethical standards. As this was a retrospective observational study based on historical clinical records, all study data were derived from previously documented medical information. During data extraction and analysis, all patient information was anonymized and de-identified to protect privacy and ensure information security. For cases analyzed solely on the basis of historical medical records, informed consent could be waived in accordance with institutional ethics regulations, provided that patient privacy was not compromised and clinical care was not affected.

RESULTS

Case screening and baseline characteristics

Among 312 potentially eligible postoperative PFCD patients identified from the hospital databases, 29 were excluded because the diagnosis could not be confirmed, 24 because no study-related postoperative TCM external intervention had been used, 18 because of missing key variables, 13 because follow-up was shorter than 4 weeks, and 11 because treatment exposure could not be clearly determined. A total of 217 patients were finally included in the analytic cohort, including 72 in the sitz bath-only group, 69 in the sitz bath plus acupoint application group, and 76 in the Ziwu-Liuzhu time-specific external treatment group. The detailed screening process is shown in Figure 1.

Figure 1 Flowchart of patient screening, eligibility assessment, exclusion, and final group allocation. PFCD: Perianal fistulizing Crohn’s disease; TCM: Traditional Chinese medicine.

Baseline demographic characteristics, disease-related features, inflammatory markers, and psychological status were generally comparable across the three groups (Table 1). There were no statistically significant differences in age, sex, disease duration, preoperative CDAI, preoperative PDAI, CRP, TNF-α, IL-6, GAD-7, body mass index, smoking history, alcohol use, or surgical procedure distribution (all P > 0.05), indicating acceptable baseline comparability.

Table 1 Baseline characteristics of perianal fistulizing Crohn’s disease patients in the three treatment groups, mean ± SD/n (%).

Variable	Sitz bath group (n = 72)	Sitz bath plus acupoint application group (n = 69)	Ziwu-Liuzhu time-specific external treatment group (n = 76)	Statistic	P value
Age (years)	34.8 ± 9.1	35.6 ± 8.7	33.9 ± 8.4	F = 0.71	0.493
Male	42 (58.3)	39 (56.5)	44 (57.9)	χ2 = 0.05	0.973
BMI (kg/m2)	21.9 ± 2.6	22.3 ± 2.8	22.1 ± 2.5	F = 0.42	0.66
Smoking history	15 (20.8)	13 (18.8)	15 (19.7)	χ2 = 0.10	0.95
Alcohol consumption	12 (16.7)	11 (15.9)	13 (17.1)	χ2 = 0.05	0.973
Disease duration (months), median IQR	19.0 (12.0-28.0)	18.0 (11.0-27.0)	20.0 (13.0-29.0)	H = 0.86	0.651
Age at diagnosis (years)	31.2 ± 8.6	32.0 ± 8.3	30.7 ± 8.1	F = 0.52	0.596
Previous fistula surgery	21 (29.2)	20 (29.0)	22 (28.9)	χ2 = 0.00	0.999
Number of fistula tracts	1.6 ± 0.7	1.5 ± 0.6	1.6 ± 0.7	F = 0.48	0.62
Complex fistula	31 (43.1)	29 (42.0)	33 (43.4)	χ2 = 0.04	0.981
Preoperative CDAI score	256.4 ± 41.2	252.7 ± 39.6	254.1 ± 40.5	F = 0.16	0.853
Preoperative PDAI score	6.4 ± 1.2	6.3 ± 1.1	6.2 ± 1.2	F = 0.53	0.592
Preoperative VAS score	6.1 ± 1.3	5.9 ± 1.2	6.0 ± 1.3	F = 0.34	0.714
Preoperative GAD-7 score	9.2 ± 2.6	9.0 ± 2.4	9.1 ± 2.5	F = 0.13	0.878
CRP (mg/L)	31.8 ± 8.6	30.6 ± 8.1	31.1 ± 8.4	F = 0.34	0.713
TNF-α (pg/mL)	41.7 ± 9.5	40.9 ± 9.2	41.2 ± 9.4	F = 0.13	0.879
IL-6 (pg/mL)	33.6 ± 8.3	32.8 ± 7.9	33.1 ± 8.0	F = 0.16	0.85
White blood cell count (× 109/L)	8.9 ± 2.1	8.7 ± 2.0	8.8 ± 2.2	F = 0.14	0.867
Hemoglobin (g/L)	118.6 ± 15.2	120.1 ± 14.7	119.5 ± 15.0	F = 0.22	0.804
Albumin (g/L)	37.9 ± 4.5	38.4 ± 4.3	38.1 ± 4.6	F = 0.28	0.758

BMI: Body mass index; IQR: Interquartile range; CDAI: Crohn’s Disease Activity Index; PDAI: Perianal Disease Activity Index; VAS: Visual Analogue Scale; GAD-7: Generalized Anxiety Disorder-7 scale; CRP: C-reactive protein; TNF: Tumor necrosis factor; IL: Interleukin.

Clinical symptom improvement

Before treatment, CDAI and PDAI scores were similar among the three groups, suggesting comparable baseline disease severity. All post-treatment comparisons reported in this study were based on the unified week-4 assessment time point (Table 2). After 4 weeks of treatment, both CDAI and PDAI decreased significantly in all groups, but the degree of improvement differed markedly.

Table 2 Changes in Crohn’s Disease Activity Index and Perianal Disease Activity Index scores from baseline to week 4 among the three treatment groups, mean ± SD.

Variable	Sitz bath group (n = 72)	Sitz bath plus acupoint application group (n = 69)	Ziwu-Liuzhu time-specific external treatment group (n = 76)	Statistic	P value	Effect size (η2)
CDAI score
Baseline	256.4 ± 41.2	252.7 ± 39.6	254.1 ± 40.5	F = 0.16	0.853	0.001
Week 4	183.7 ± 34.9	151.8 ± 30.6	121.6 ± 27.4	F = 57.32	< 0.001	0.347
Δ CDAI	72.7 ± 28.6	100.9 ± 30.2	132.5 ± 35.7	F = 61.08	< 0.001	0.362
95%CI for Δ CDAI	66.1-79.3	93.8-108.0	124.5-140.5
Within-group comparison (paired t)	t = 21.56	t = 27.75	t = 32.34		< 0.001 for all
PDAI score
Baseline	6.4 ± 1.2	6.3 ± 1.1	6.2 ± 1.2	F = 0.53	0.592	0.005
Week 4	4.3 ± 1.0	3.1 ± 0.9	2.0 ± 0.8	F = 98.46	< 0.001	0.478
Δ PDAI	2.1 ± 0.9	3.2 ± 1.0	4.2 ± 1.1	F = 74.92	< 0.001	0.411
95%CI for Δ PDAI	1.89-2.31	2.96-3.44	3.95-4.45
Within-group comparison (paired t)	t = 19.80	t = 26.59	t = 33.28		< 0.001 for all

Within-group comparisons were performed using paired t tests. Between-group comparisons were performed using one-way analysis of variance. Bonferroni correction was applied for post hoc pairwise comparisons. Effect size is expressed as eta-squared (η²). Δ indicates the change from baseline to week 4. CDAI: Crohn’s Disease Activity Index; PDAI: Perianal Disease Activity Index; CI: Confidence interval.

Post-treatment CDAI scores were 183.7 ± 34.9 in the sitz bath-only group, 151.8 ± 30.6 in the sitz bath plus acupoint application group, and 121.6 ± 27.4 in the Ziwu-Liuzhu time-specific external treatment group, with a significant between-group difference (F = 57.32, P < 0.001) (Table 2). The corresponding mean reductions in CDAI were 72.7 ± 28.6, 100.9 ± 30.2, and 132.5 ± 35.7, respectively (F = 61.08, P < 0.001) (Table 2).

Similarly, post-treatment PDAI scores were 4.3 ± 1.0, 3.1 ± 0.9, and 2.0 ± 0.8, respectively (F = 98.46, P < 0.001), while the mean reductions were 2.1 ± 0.9, 3.2 ± 1.0, and 4.2 ± 1.1, respectively (F = 74.92, P < 0.001) (Table 2). Taken together, all three treatment strategies improved overall disease activity and perianal symptom burden, but the Ziwu-Liuzhu time-specific combined therapy produced the most pronounced improvement.

Wound healing-related indicators

Clear between-group differences were observed in wound healing-related outcomes. The inter-rater agreement between the two assessors was good for wound healing score (κ = 0.78), wound exudate score (κ = 0.74), and granulation tissue growth score (κ = 0.72), supporting the reliability of these retrospectively evaluated wound indicators. At week 4, wound healing score, exudate score, pain score, and granulation tissue growth score were all significantly better in the Ziwu-Liuzhu time-specific external treatment group than in the other two groups, while the sitz bath plus acupoint application group also generally performed better than the sitz bath-only group (Figure 2A-D).

Figure 2 Comparison of postoperative wound healing-related outcomes at week 4 among perianal fistulizing Crohn’s disease patients in the three treatment groups. A: Distribution of wound healing scores, the Ziwu-Liuzhu time-specific external treatment group had the lowest scores, indicating the most favorable wound recovery; B: Distribution of wound exudate scores; The Ziwu-Liuzhu time-specific external treatment group showed markedly less wound exudation than the other two groups; C: Comparison of pain intensity assessed by the Visual Analogue Scale (VAS); The Ziwu-Liuzhu time-specific external treatment group had significantly lower VAS scores than both the sitz bath plus acupoint application group and the sitz bath-only group; D: Distribution of granulation tissue growth scores; The Ziwu-Liuzhu time-specific external treatment group had substantially lower granulation tissue growth scores than the other two groups; E: Comparison of wound healing time; The Ziwu-Liuzhu time-specific external treatment group had the shortest wound healing time; F: Kaplan-Meier analysis of wound healing time across the three treatment groups. The log-rank test showed a statistically significant difference among the wound healing curves, with the Ziwu-Liuzhu time-specific external treatment group demonstrating the highest probability of wound healing. ANOVA: Analysis of variance; VAS: Visual Analogue Scale.

Median wound healing times were 31 days in the sitz bath-only group, 24 days in the sitz bath plus acupoint application group, and 18 days in the Ziwu-Liuzhu time-specific external treatment group, with a highly significant difference among groups (H = 84.63, P < 0.001) (Figure 2E).

Kaplan-Meier analysis showed significant differences in wound healing probability over time among the three groups (log-rank χ² = 38.72, P < 0.001) (Figure 2F). Using the sitz bath-only group as the reference, the hazard ratio for wound healing was 1.69 (95%CI: 1.19-2.40, P = 0.004) in the sitz bath plus acupoint application group and 2.58 (95%CI: 1.83-3.64, P < 0.001) in the Ziwu-Liuzhu time-specific external treatment group. After adjustment for age, sex, body mass index, disease duration, and baseline CDAI, the Ziwu-Liuzhu time-specific external treatment group remained significantly associated with faster wound healing (adjusted hazard ratio = 2.31, 95%CI: 1.62-3.29, P < 0.001). These findings indicate that time-specific combined external therapy may confer a meaningful advantage in postoperative wound recovery.

Changes in inflammatory markers

All three treatment strategies reduced postoperative inflammatory marker levels, but the greatest reductions were consistently observed in the Ziwu-Liuzhu time-specific external treatment group. After 4 weeks, CRP levels were 18.7 ± 6.3 mg/L in the sitz bath-only group, 12.9 ± 5.4 mg/L in the sitz bath plus acupoint application group, and 8.4 ± 4.1 mg/L in the Ziwu-Liuzhu group (F = 62.74, P < 0.001) (Figure 3A). TNF-α levels were 28.6 ± 7.8 pg/mL, 22.1 ± 6.7 pg/mL, and 16.3 ± 5.5 pg/mL, respectively (F = 68.93, P < 0.001) (Figure 3B). IL-6 levels were 21.4 ± 6.9 pg/mL, 15.8 ± 5.9 pg/mL, and 10.7 ± 4.8 pg/mL, respectively (F = 74.26, P < 0.001) (Figure 3C).

Figure 3 Changes in inflammatory markers and multivariable regression results among the three treatment groups. A: Trends in C-reactive protein (CRP) levels before and after treatment. CRP levels decreased significantly after treatment in all three groups, with the greatest reduction observed in the Ziwu-Liuzhu group; B: Trends in tumor necrosis factor-α (TNF-α) levels. The Ziwu-Liuzhu time-specific external treatment group had the lowest post-treatment TNF-α level; C: Trends in interleukin-6 (IL-6) levels. The Ziwu-Liuzhu time-specific external treatment group showed the most pronounced decline in IL-6; D: Forest plot of multivariable linear regression analysis showing the associations of treatment strategy and related variables with changes in inflammatory markers. After adjustment for potential confounders, including age, sex, body mass index, and disease severity, the Ziwu-Liuzhu time-specific external treatment group strategy remained significantly associated with lower inflammatory levels. ANOVA: Analysis of variance; CRP: C-reactive protein; TNF: Tumor necrosis factor; IL: Interleukin; CI: Confidence interval; BMI: Body mass index.

Multivariable linear regression further showed that, compared with sitz bath alone, both combined treatment and Ziwu-Liuzhu treatment were independently associated with lower CRP, TNF-α, and IL-6 levels, with the largest effect sizes consistently observed in the Ziwu-Liuzhu time-specific external treatment group (Figure 3D). Overall, these results suggest that the time-specific combined regimen may be associated with a stronger anti-inflammatory effect during postoperative recovery.

Improvement in psychological status

Baseline GAD-7 scores were similar among the three groups (Table 3). After 4 weeks, GAD-7 scores decreased significantly in all groups, but the magnitude of improvement was greatest in the Ziwu-Liuzhu group.

Table 3 Changes in Generalized Anxiety Disorder-7 scale scores and anxiety severity at baseline and week 4 among the three treatment groups, mean ± SD/n (%).

Group	Sitz bath (n = 72)	Sitz bath plus acupoint application group (n = 69)	Ziwu-Liuzhu time-specific external treatment (n = 76)
Baseline GAD-7	9.2 ± 2.6	9.0 ± 2.4	9.1 ± 2.5
Week 4 GAD-7	7.9 ± 2.4	5.8 ± 2.0	4.1 ± 1.7
Δ GAD-7	1.3 ± 1.6	3.2 ± 1.7	5.0 ± 1.9
Δ GAD-7 95%CI	0.9-1.7	2.8-3.6	4.6-5.4
Effect size (Cohen d)	0.5	1.36	2.22
Within-group t value	6.18	15.37	21.48
Within-group P value	< 0.001	< 0.001	< 0.001
No anxiety 0-4	13 (18.1)	22 (31.9)	40 (52.6)
Mild 5-9	38 (52.8)	34 (49.3)	29 (38.2)
Moderate 10-14	17 (23.6)	11 (15.9)	6 (7.9)
Severe ≥ 15	4 (5.5)	2 (2.9)	1 (1.3)

GAD-7: Generalized Anxiety Disorder-7 scale; CI: Confidence interval.

Post-treatment GAD-7 scores were 7.9 ± 2.4 in the sitz bath-only group, 5.8 ± 2.0 in the sitz bath plus acupoint application group, and 4.1 ± 1.7 in the Ziwu-Liuzhu time-specific external treatment group (F = 48.63, P < 0.001) (Table 3). The mean reductions were 1.3 ± 1.6, 3.2 ± 1.7, and 5.0 ± 1.9, respectively (F = 66.92, P < 0.001) (Table 3).

In addition, the Ziwu-Liuzhu time-specific external treatment group had the highest proportion of patients with no obvious anxiety after treatment, whereas the proportion of moderate-to-severe anxiety was lowest in this group (χ² = 24.57, P < 0.001) (Table 3). These findings suggest that better postoperative recovery under the time-specific regimen may also have translated into greater psychological benefit.

Overall clinical efficacy

At week 4, the distribution of clinical cure, effective response, and ineffective response differed significantly among the three groups (Figure 4A). The overall response rates were 61.1% in the sitz bath-only group, 78.3% in the sitz bath plus acupoint application group, and 92.1% in the Ziwu-Liuzhu time-specific external treatment group (χ² = 18.74, P < 0.001) (Figure 4B).

Figure 4 Effects of different treatment strategies on overall postoperative efficacy in perianal fistulizing Crohn’s disease patients and evaluation of the efficacy prediction model. A: Composition of clinical efficacy outcomes across the three groups, including clinical cure, effective response, and ineffective response. The Ziwu-Liuzhu time-specific external treatment group had the highest proportion of clinical cure; B: Comparison of overall response rates among the three groups; The Ziwu-Liuzhu time-specific external treatment group showed the highest overall response rate; C: Forest plot of multivariable logistic regression analysis showing that the Ziwu-Liuzhu treatment strategy was significantly associated with treatment response and served as an independent predictor of favorable efficacy; D: Receiver operating characteristic curve of the efficacy prediction model constructed from the multivariable analysis. The curve was clearly above the diagonal reference line, indicating good discriminative performance of the model. OR: Odds ratio; CI: Confidence interval; AUC: Area under the curve; CRP: C-reactive protein; CDAI: Crohn’s Disease Activity Index; PDAI: Perianal Disease Activity Index; BMI: Body mass index.

Multivariable logistic regression showed that, compared with sitz bath alone, sitz bath plus acupoint application was significantly associated with treatment response [odds ratio (OR) = 2.31, 95%CI: 1.12-4.77, P = 0.024], while the Ziwu-Liuzhu regimen showed an even stronger association (OR = 4.86, 95%CI: 1.96-6.04, P = 0.001). Baseline CDAI was also an independent predictor of treatment response (OR = 0.94, 95%CI: 0.90-0.98, P = 0.009) (Figure 4C). Receiver operating characteristic analysis indicated good discriminatory performance of the predictive model (Figure 4D).

Propensity score matching and sensitivity analysis

To reduce possible selection bias, propensity score matching was performed. After matching, baseline balance was acceptable, with standardized mean differences below 0.10 (Table 4).

Table 4 Propensity score matching and sensitivity analysis results, mean ± SD/median interquartile range.

Comparison	Indicator	Ziwu-Liuzhu time-specific external treatment group	Comparator group	Statistic	P value
Ziwu-Liuzhu time-specific external treatment vs sitz bath (PSM, n = 58 pairs)	Healing time (day)	19 (16-23)	29 (25-34)	Z = -5.21	< 0.001
	CRP after treatment (mg/L)	9.1 ± 4.3	17.8 ± 6.1	t = -8.64	< 0.001
	TNF-α after treatment (pg/mL)	17.1 ± 5.4	27.4 ± 7.6	t = -7.93	< 0.001
	IL-6 after treatment (pg/mL)	11.4 ± 4.7	20.6 ± 6.8	t = -8.11	< 0.001
	Total effective rate (%)	53/58 (91.4)	37/58 (63.8)	χ2 = 12.37	< 0.001
	Logistic regression OR (95%CI)	4.12 (1.89-8.97)	Reference		< 0.001
Ziwu-Liuzhu time-specific external treatment vs sitz bath plus acupoint application group (PSM, n = 61 pairs)	Healing time (day)	18 (15-22)	24 (20-28)	Z = -4.28	< 0.001
	VAS pain score	2.2 ± 0.9	3.3 ± 1.1	t = -5.43	< 0.001
	CRP after treatment (mg/L)	8.7 ± 4.0	13.4 ± 5.2	t = -5.58	< 0.001
	TNF-α after treatment (pg/mL)	16.5 ± 5.3	22.4 ± 6.4	t = -5.01	< 0.001
	IL-6 after treatment (pg/mL)	10.9 ± 4.5	16.2 ± 5.7	t = -5.32	< 0.001
	Total effective rate (%)	57/61 (93.4)	49/61 (80.3)	χ2 = 5.29	0.021
	Logistic regression OR (95%CI)	3.26 (1.28-8.29)	Reference		0.013

PSM: Propensity score matching; CRP: C-reactive protein; TNF: Tumor necrosis factor; IL: Interleukin; OR: Odds ratio; CI: Confidence interval; VAS: Visual Analogue Scale.

In the matched comparison between the Ziwu-Liuzhu time-specific external treatment group and the sitz bath-only group, the Ziwu-Liuzhu time-specific external treatment group continued to show shorter wound healing time, lower CRP, TNF-α, and IL-6 levels, and a higher overall response rate. Logistic regression in the matched cohort confirmed that Ziwu-Liuzhu treatment remained significantly associated with treatment response (OR = 4.12, 95%CI: 1.89-8.97, P < 0.001) (Table 4).

In the matched comparison between the Ziwu-Liuzhu time-specific external treatment group and the sitz bath plus acupoint application group, the Ziwu-Liuzhu time-specific external treatment group likewise showed shorter wound healing time, lower pain scores, lower inflammatory marker levels, and a higher response rate. Ziwu-Liuzhu treatment remained an independent predictor of treatment response (OR = 3.26, 95%CI: 1.28-8.29, P = 0.013) (Table 4). These sensitivity analyses indicate that the main findings were robust and not merely driven by observable baseline imbalances.

DISCUSSION

The present study compared three postoperative management strategies for PFCD and found that Ziwu-Liuzhu time-specific combined external therapy was associated with greater improvements in disease activity, wound healing, inflammatory markers, psychological status, and overall clinical efficacy than sitz bath alone or non-time-specific combined therapy. These findings suggest that integrating chronomedicine concepts with TCM external therapy may provide a useful adjunctive approach in postoperative PFCD management; however, given the absence of a placebo or sham-controlled comparator, the magnitude of the specific treatment effect should be interpreted with caution.

One important finding of this study was the more pronounced reduction in both CDAI and PDAI in the Ziwu-Liuzhu time-specific external treatment group. This suggests that the benefit of time-specific combined external therapy may not be confined to the local wound itself, but may also extend to broader disease activity control. In PFCD, CDAI reflects the overall systemic disease burden, whereas PDAI more directly captures local perianal inflammatory severity, tenderness, discharge, and the impact of disease on daily activities. Therefore, the simultaneous improvement in both indices supports the possibility that this intervention may influence both local lesion status and the broader inflammatory milieu[27].

This interpretation is biologically plausible. Increasing evidence indicates that inflammatory bowel disease activity is shaped not only by cytokine dysregulation but also by circadian rhythm-related immune modulation. Circadian disruption may alter the recruitment and activation of innate and adaptive immune cells and thereby amplify intestinal inflammation[28]. Experimental studies have also shown that core clock molecules such as BMAL1 and CLOCK are involved in maintaining intestinal epithelial homeostasis and barrier function, and that disruption of this regulatory network may aggravate inflammatory injury and impair mucosal repair[29]. In clinical settings, disturbances in sleep rhythm and neuroendocrine fluctuations are also common in patients with inflammatory bowel disease and have been linked to increased disease activity[30]. Against this background, the better improvement in CDAI and PDAI observed in the Ziwu-Liuzhu time-specific external treatment group may reflect a more favorable alignment between treatment delivery and biological rhythms governing inflammation and tissue repair. Although this study cannot directly establish a mechanistic pathway, the clinical pattern observed here is consistent with the rationale of chronomedicine.

Another major finding was the superiority of the Ziwu-Liuzhu regimen across multiple wound-related indicators, including wound healing score, exudate score, pain score, granulation tissue growth, and wound healing time. This pattern suggests that the intervention may act on the wound repair process in a multidimensional manner rather than merely alleviating a single symptom. Wound healing is a dynamic biological process involving coordinated control of inflammation, angiogenesis, fibroblast activation, epithelial regeneration, and extracellular matrix remodeling[31]. In PFCD, postoperative wounds are particularly difficult to heal because they exist in a complex microenvironment characterized by persistent inflammation, high exudation, local bacterial burden, and impaired tissue regenerative capacity[32].

Previous studies have shown that, in complex perianal wounds, bacterial colonization, sustained inflammatory stimulation, and altered interstitial pressure may jointly inhibit angiogenesis and fibroblast function, thereby delaying granulation maturation and wound contraction[33]. From this perspective, the shorter healing time and better granulation status observed in the Ziwu-Liuzhu time-specific external treatment group may reflect not only improved local anti-inflammatory control, but also a more favorable microenvironment for tissue repair. Related studies have suggested that some topical TCM formulations may upregulate repair-related mediators such as vascular endothelial growth factor and transforming growth factor-β and promote granulation tissue formation, thereby facilitating angiogenesis and tissue remodeling[34]. Clinical evidence has also shown that herbal sitz bath therapy can reduce postoperative edema, pain, and exudation in anorectal wounds, creating more favorable conditions for transition from the inflammatory phase to the proliferative phase of healing[35]. Taken together, these data support the view that the observed wound-healing advantage of Ziwu-Liuzhu time-specific therapy is biologically credible and clinically meaningful.

The combined use of compound Qinbai granules and Kuijiening plaster may constitute an important therapeutic basis for the observed effect. Compound Qinbai granules mainly emphasize heat-clearing, dampness-drying, detoxifying, and swelling-reducing actions. The core herbal components, particularly Scutellaria baicalensis and Phellodendron chinense, contain flavonoid compounds with recognized anti-inflammatory and antimicrobial potential, and these constituents have been shown to inhibit inflammatory signaling pathways such as nuclear factor kappa-B (NF-κB)[36]. Baicalin, one of the representative active compounds, has been reported to reduce the release of proinflammatory mediators such as TNF-α and IL-6 and to regulate macrophage polarization, thereby helping shift tissues from a persistent inflammatory state toward a reparative state[37].

By contrast, Kuijiening plaster is more oriented toward warming the channels, promoting circulation, and facilitating tissue repair. Its components, including aconite, clove, and white mustard seed, are traditionally used to improve local blood flow and relieve stagnation, and their modern pharmacological relevance may lie in promoting local perfusion and microcirculatory reconstruction[38]. Since inadequate oxygen and nutrient supply caused by persistent hypoperfusion is a recognized contributor to delayed wound healing, improving local hemodynamics is particularly relevant to granulation growth, epithelial regeneration, and wound contraction[39]. In addition, some volatile and lipophilic constituents of topical preparations may penetrate through the skin and mucosal barrier, thereby enhancing local anti-inflammatory and analgesic effects[40]. Therefore, the combination of compound Qinbai granules and Kuijiening plaster may act through complementary mechanisms one primarily targeting inflammation control and the other favoring circulation and repair which may help explain why combined therapy, especially when delivered in a time-specific manner, showed superior outcomes.

The findings regarding inflammatory markers further support this interpretation. In the present study, CRP, TNF-α, and IL-6 all declined in the three groups, but the reduction was greatest in the Ziwu-Liuzhu time-specific external treatment group. Because CD is driven by persistent immune activation and reciprocal amplification among inflammatory cytokines, particularly through networks such as NF-κB and Janus kinase-signal transducer and activator of transcription, a greater decline in these markers may indicate attenuation of both local inflammatory stimulation and systemic inflammatory burden[41]. This is clinically relevant because postoperative PFCD wounds are not isolated local lesions; rather, they remain embedded within the broader inflammatory context of CD.

Several previous studies support this line of reasoning. Certain plant-derived active compounds have been shown to reduce inflammatory burden in inflammatory bowel disease by modulating inflammatory signaling, oxidative stress, and immune cell function[42]. Baicalin has also been reported to inhibit the IL-23/T helper 17 axis and reduce inflammatory cell infiltration, suggesting a possible regulatory effect on persistent mucosal inflammation[43]. In addition, experimental studies have found that some TCM compound formulations may ameliorate intestinal inflammation by reshaping gut microbiota composition, restoring epithelial barrier function, and improving mucosal immune homeostasis[44]. Moreover, increasing evidence indicates that persistent local lesion activity and systemic inflammatory activation are bidirectionally linked in inflammatory bowel disease, so that better local wound control may itself contribute to a reduction in systemic inflammatory markers[45]. Therefore, the greater decline in CRP, TNF-α, and IL-6 observed in the Ziwu-Liuzhu time-specific external treatment group may reflect a broader anti-inflammatory benefit rather than a purely local symptomatic effect.

Another clinically meaningful finding was the greater improvement in GAD-7 scores in the Ziwu-Liuzhu time-specific external treatment group. Anxiety and depression in inflammatory bowel disease are not merely secondary emotional reactions, but are closely intertwined with disease activity, symptom burden, and deterioration in quality of life[46]. For PFCD patients in particular, persistent pain, exudation, odor, repeated dressing changes, and the chronic risk of recurrence may substantially worsen psychological distress and reinforce a sense of disease uncontrollability[47]. Against this background, the more pronounced reduction in GAD-7 scores observed in the Ziwu-Liuzhu time-specific external treatment group may reasonably be interpreted as the downstream result of better local wound recovery, less pain, lower inflammatory burden, and a smoother recovery experience.

This interpretation is supported by previous evidence suggesting that TCM external therapies in chronic disease settings may indirectly improve anxiety and stress through modulation of inflammatory status, neuroendocrine responses, and somatic symptoms[48]. In the context of the present study, time-specific combined external therapy may also have provided a more structured and predictable treatment rhythm, which could enhance patients’ confidence in treatment and strengthen their perception of recovery control. Thus, the psychological benefit observed here should not be viewed as an isolated outcome, but rather as part of the broader integrated clinical benefit associated with improved postoperative recovery.

The present study also has important clinical implications. First, it suggests that postoperative PFCD management may benefit not only from the choice of external therapeutic modality but also from optimization of treatment timing. Second, it supports the value of evaluating PFCD recovery through a multidimensional framework that includes wound repair, inflammatory burden, disease activity, and psychological status, rather than relying solely on local anatomical outcomes. Third, the findings provide preliminary clinical evidence that chronomedicine-based TCM external therapy may serve as a practical adjunct in real-world postoperative care pathways.

Several limitations should nevertheless be acknowledged. First, this was a single-center retrospective observational study, and residual confounding cannot be fully excluded despite multivariable adjustment and propensity score matching. In addition, no placebo or sham external treatment was available in this real-world retrospective study. Because the study compared routinely used postoperative management strategies rather than investigator-assigned interventions, the observed benefits should be interpreted as comparative associations between treatment models rather than definitive placebo-controlled efficacy estimates. Second, neither patients nor outcome assessors were blinded because treatment exposure had already occurred in routine clinical practice and outcome data were retrospectively extracted from historical records; therefore, performance bias and assessment bias cannot be ruled out. Third, the follow-up period was relatively short, precluding evaluation of long-term recurrence, sustained remission, or reoperation risk. Fourth, although the clinical findings are compatible with mechanisms involving circadian immune regulation, local microcirculatory improvement, and inflammatory suppression, these mechanisms were not directly tested in this study. In addition, although wound-related scores were assessed according to unified criteria by two trained staff members, formal inter-rater reliability statistics were not available for all historical records, which may have introduced measurement variability. Future multicenter prospective randomized studies are therefore needed to validate the efficacy of Ziwu-Liuzhu time-specific external therapy and to clarify its biological basis through immunological, microecological, and tissue repair-related investigations.

CONCLUSION

In conclusion, compared with sitz bath alone or conventional combined external therapy, Ziwu-Liuzhu time-specific combined external treatment was associated with faster wound healing, better improvement in disease activity, lower inflammatory marker levels, reduced anxiety symptoms, and a higher overall response rate in postoperative PFCD patients. These findings suggest that time-specific TCM external therapy may serve as a valuable adjunctive strategy in the integrated postoperative management of PFCD. Further prospective multicenter studies with long-term follow up and mechanistic investigations are warranted to confirm its efficacy and clarify its biological basis.