Efficacy of an Herbal Cream on Patients with Hemorrhoids: a randomized triple-blind controlled trial

Abstract

Objectives

Traditional Persian medicine offers various topical treatments for hemorrhoids. This study investigated the efficacy of an herbal cream (HC) containing common myrtle, chamomile, wild leek, and Russian olive extracts.

Methods

This randomized triple-blind controlled trial compared the effects of the HC and a conventional cream (CC) on hemorrhoids. Standardization involved measuring the total phenolic content (TPC) and analyzing the HC by gas chromatography. In total, 110 outpatients with hemorrhoids were randomly assigned to intervention (n = 58) and control (n = 52) groups. The outcomes were evaluated using the visual analog scale (VAS [1-10]); anal pain, difficult defecation, and burning sensation) and four-point Likert scale anal bleeding and anal pruritus) during the intervention phase (14 days) and follow-up (28 days).

Results

The HC was safe and well tolerated, with no major adverse effects observed in toxicological or clinical evaluations. Physicochemical assays confirmed the quality parameters, and the TPC was determined. In the clinical trial involving 49 patients, both HC and CC reduced hemorrhoidal symptoms over time (p = 0.011). However, no significant treatment-group effect (p = 0.50) or treatment-by-time interaction (p = 0.77) was observed. Exploratory analyses indicated that the HC produced clinically meaningful reductions in bleeding and anal pain from baseline.

Conclusion

The HC was safe and well tolerated in patients with grade I-II hemorrhoids. Although it did not show statistical superiority over the CC, potentially due to high dropout rates and limited sample size, HC was associated with clinically meaningful improvements in anal bleeding and pain along with favorable changes in other symptoms. These findings suggest that the HC has potential as a safe herbal alternative for hemorrhoid symptom relief, without side effects and an acceptable safety profile.

INTRODUCTION

Hemorrhoids are characterized by abnormal swelling or displacement of anal cushions and represent a common anorectal condition, accounting for approximately 4 million medical visits in the United States in 2010. They are the third most frequent cause of hospital admissions related to digestive disorders [1, 2]. Treatment options range from dietary modifications to surgery, with an increasing preference for less invasive approaches [3].

Hemorrhoids are referenced in early texts such as the Old Testament and ancient Egyptian, Babylonian, and Greek texts [4]. Traditional medicines, including Traditional Persian medicine (TPM), provide valuable insights for developing new noninvasive treatments. TPM has been practiced in Iran, Afghanistan, Tajikistan, Arab countries, and Europe [5, 6]. In TPM, “Zemad” is a traditional ointment that has been used for treating hemorrhoids for centuries [7]. Notable medicinal plants in TPM for hemorrhoid treatment include Myrtus communis L. (common myrtle) [8], Matricaria chamomilla L. (chamomile) [9], Allium ampeloprasum L. (wild leek) [10], and Elaeagnus angustifolia L. (Russian olive) [11]. Their uses are documented in medieval texts, including the Canon of Medicine [12] and Kamel al-Sanāʿh al-Tebbiah [13].

These plants are recognized for their antihemorrhoidal, anti-inflammatory, analgesic, antiseptic, hemostatic, and wound healing properties. For example, Matricaria chamomilla exerts antioxidant, antibacterial, antifungal, and anti-inflammatory effects and is approved by Commission E for treating digestive spasms and for external application on mucous membranes and the skin [14]. Additionally, E. angustifolia L. is used to treat rheumatoid pain, osteoarthritis, and gastrointestinal disorders [15], and A. ampeloprasum [16] has demonstrated antihemorrhoidal effects in classical and clinical studies (Table 1) [17-22].

Table 1.

Therapeutic effects of hemorrhoid cream ingredients

References	Scientific name	Traditional name	Part used	Active component(s)	Therapeutic effects
[17-19]	Myrtus communis	Ās	Essence	Contains volatile oils (1.5%-2.0%) Major compounds: 1,8-cineole (15%-45%), α-pinene (15%-38%), flavonoids, and tannins	Analgesic, anti-inflammatory. Antihemorrhoid (relieves bleeding, persistent pain, pain during defecation, anal irritation, itching, and heaviness in patients with hemorrhoid types I and II)
[20]	Matricaria chamomilla	Bābouneh	Flower	Flavonoids, sesquiterpenes, coumarins, and polyacetylenes	Anti-inflammatory, antiseptic, antiphlogistic, neuroprotective, spasmolytic, and wound healing
[21]	Allium ampeloprasum	Korrāth	Leaf	Steroidal saponin (spirostane-type saponin, furostane-type saponins, and spirostane-type saponin), phenolics, and flavonoids	Anti-inflammatory, antiulcerogenic, antioxidant, wound healing, and antimicrobial
[22]	Elaeagnus angustifolia	Ghobayrā	Fruit	Flavonoid quercetin, quercitol isorhamnetin, isorhamnetin, 3-O-β-D-galactopyranoside, isorhamnetin-3-O-galactopyranoside, kaempferol), phenolic acids (caffeic acid, ferulic acid, p-coumaric acid, protocatechuic acid, vanillic acid, 4-hydroxybenzoic acid, 4-hydroxycinnamic acid, and benzoic acid), amino acids, steroids (β-sitosterol), carbohydrates Vitamins (C, A, E, and K)	Antinociceptive, anti-inflammatory, antimicrobial, and wound healing

This randomized, triple-blind, active-controlled trial was conducted to evaluate a potentially antihemorrhoid herbal cream (HC) and compare its efficacy with that of a conventional cream (CC). Images of plants used in the HC formulation and their active components are shown in Fig. 1.

Figure 1.

Images and selected constituents of plants used in the HC formulation. Matricaria chamomilla flowers (A), Myrtus communis leaves (B), Elaeagnus angustifolia fruit (C), and Allium ampeloprasum leaves (D).

MATERIALS AND METHODS

1. Drug ingredients

Medicinal plants including the dried parts of chamomile, Russian olive, and myrtle and fresh leaves of wild leek were collected from a botanical market and identified by the Department of Agriculture at the Medicinal Plants Research Center of Barij, Kashan, Iran. Their quality was evaluated using tests for foreign materials, total ash content, acid-insoluble ash content, and quantitative assays.

Wild leek juice was extracted by grinding and filtering washed leaves. Hydroalcoholic extracts of Russian olive and chamomile were prepared in a percolator using 70% ethanol, concentrated using a rotary evaporator (Heidolph, Germany), and stored under cool, dry conditions. Myrtle essential oil was hydrodistilled for 4 h in a Clevenger apparatus. For the oily phase of HC, olive oil and camel hump fat were sourced from reputable commercial stores. The physicochemical properties of the obtained extracts and essential oils were analyzed.

2. Active control cream

The control medication was a CC dispensed in tubes similar to those of the HC. Every 30 g of CC contained 5 g of lidocaine, 0.275 g of hydrocortisone acetate, 3.5 g of aluminum subacetate, and 18 g of zinc oxide. The form, color, smell, consistency, and packaging of the CC were modified to reflect those of the HC for blinding.

3. Preparation of the HC

The oil and aqueous phases of the HC were heated separately, after which the aqueous phase was gradually added to the oil phase to form an emulsion. Chamomile and Russian olive extracts, fresh wild leek juice, and myrtle essential oil were incorporated via continuous stirring until the HC was formed (batch no. 9510156).

4. Solvents and chemicals

Methanol, ethanol, Folin–Ciocalteu reagent, and sodium carbonate were purchased from Merck (Darmstadt, Germany). Gallic acid (GA) (Sigma-Aldrich, China) and 1,8-cineole (Sigma-Aldrich, USA) were also used.

5. Drug analysis

1) Total phenolic content

The total phenolic content (TPC) of HC was determined using Folin–Ciocalteu reagent and a PerkinElmer Lambda EZ-210 ultraviolet–visible spectrophotometer following the ISO 14502-1 standard. Briefly, 500 mg of the HC was mixed with 10 mL of 70% ethanol, heated at 70℃ for 15-20 min, shaken, cooled, and centrifuged. Subsequently, 0.2 mL of the supernatant was mixed with water and Folin–Ciocalteu reagent, followed by sodium carbonate. After allowing to stand for 2 h, the absorbance was measured at 760 nm. All tests were performed in triplicate. To construct a calibration curve, GA stock solution (400 µg/mL) was diluted into six concentrations (40-160 µg/mL).

2) Essential oil extraction and gas chromatography analysis

First, 15 g of the HC was hydrodistilled in a Clevenger apparatus for 4 h to extract the essential oil. The oil was dried using anhydrous sodium sulfate and stored in a dark vial at a cold temperature until analysis. The oil was analyzed using a Varian CP-3800 GC system with a flame ionization detector and a CP-WAX 52 CB fused-silica capillary column (50-m length, 0.32-mm i.d., and 0.25-µm film thickness). The oven temperature was increased from 50℃ to 200℃ at 3℃/min and held for 10 min, with nitrogen as the carrier gas at a 1 mL/min flow rate. Injector and detector temperatures were 220℃ and 230℃, respectively.

6. Study design and participants

The trial was conducted from April 2020 to June 2021, adhering to the International Council for Harmonisation-Good Clinical Practice guidelines and the Declaration of Helsinki. It was approved by the Shiraz University of Medical Sciences (SUMS) Ethics Committee (IR.SUMS.MED.REC.1398.034) and registered in the Iranian Registry of Clinical Trials on April 20, 2020 (IRCT20141011019489N3). In total, 153 outpatients with hemorrhoids were assessed at Namazi Hospital, Shiraz, Iran. Of the patients, 43 were excluded; thus, 110 were randomized into the intervention (n = 58) and control (n = 52) groups. During the 14-day intervention and 28-day follow-up, 35 intervention and 21 control patients dropped out due to COVID-19, flu-like symptoms, or unwillingness to continue in the trial. Ultimately, 20 patients in the intervention group and 29 patients in the control group completed the study. Statistical analysis was performed on the data from these 49 patients. The CONsolidated Standards of Reporting Trials (CONSORT) diagram illustrates patient allocation (Fig. 2).

Figure 2.

CONsolidated Standards of Reporting Trials (CONSORT) diagram for the process of patient allocation.

7. Sample size calculation

An estimated error of 0.05, a test power of 0.99, and a medium effect size of 0.5 were used to calculate the sample size using G*Power version 3.1. To allow for a 30% attrition rate, 110 participants were planned for enrollment.

8. Randomization, blinding, and allocation concealment

In this triple-blind study, gastroenterologists, patients, clinical staff, and the statistical analyst were blinded to group allocation. A balanced permuted block design with a block size of four was used for randomization, and A-B sequences were generated using R version 4.2.2. The sequences were preprinted, numbered with 4-digit codes by a third party, and inserted into medication packages for allocation concealment. After the analysis was completed, the project manager revealed the codes.

The appearance, packaging, and coding of the two creams tested were designed to be similar to minimize the risk of performance and detection bias. Throughout the trial and analysis, participants, clinicians/outcome assessors, clinic staff, and the analyst were blinded to treatment allocation.

9. Outcome measures

No single primary outcome was specified. Therefore, the following five symptom-specific outcomes were evaluated repeatedly over time: anal bleeding, anal pain, difficult defecation, pruritus ani, and burning sensation. Outcomes were assessed using a diagnostic checklist modified from Panahi et al. [23]. Anal pain, difficult defecation, and burning sensation were measured using a visual analog scale (VAS; 0-10), whereas anal bleeding and pruritus ani were measured using four-point ordinal scales.

10. Inclusion and exclusion criteria

Inclusion criteria were as follows: (i) patients aged ≥ 18 years; (ii) those with hemorrhoid grade ≤ 2; (iii) those with no history of malignancy, portal hypertension, or diabetes; and (iv) those without heart, kidney, or liver failure. Exclusion criteria included the following: (i) patients using hemorrhoid medications; (ii) those with strangulated or thrombotic hemorrhoids; (iii) those with fistula, fissure, or anal abscess; and (iv) those who were pregnant.

11. Intervention and data collection

Written informed consent was obtained from all patients, and their sociodemographic, family history, symptom duration, and clinical characteristics were recorded. An expert gastroenterologist performed rectoscopy, ruled out any concerning findings, and determined the hemorrhoid grade; eligible grade 1 or 2 cases were enrolled. The 14-day intervention included visits at baseline, 2 weeks, and 6 weeks after follow-up. Patients applied 1 g of HC or CC twice daily to the anal area after washing using a unique applicator and avoided hot baths or any systemic/local medications (e.g., phosphodiesterase inhibitors, corticosteroids, nonsteroidal anti-inflammatory drugs, analgesics, sedative-hypnotics, or antidepressants). During follow-up, the patients used conventional antihemorrhoid ointment pro re nata and recorded usage. Data collected via demographic forms and diagnostic checklists were extracted after study completion.

12. Safety monitoring

Evidence supports the topical and/or systemic safety of medicinal plants used in the HC. Myrtus communis is safe when applied topically and systemically at therapeutic doses [18]. Topically applied pure Matricaria chamomilla oil has been reported to cause no major irritation in animal and human studies and is “generally regarded as safe” by the U.S. Food and Drug Administration (FDA) [24, 25]. Topical application of E. angustifolia gel caused no side effects in a clinical trial for oral lichen planus [26].

To confirm the safety of HC, acute and subchronic toxicity tests were conducted in collaboration with Tehran University of Medical Sciences using healthy female Wistar rats (mean weight: 178 ± 23 g; approximately 7 weeks) under standard conditions. To assess acute toxicity, median lethal dose (LD50) was determined for oral exposure in six rats and topical exposure at 4 g/kg on 90% of the skin. For subchronic assessments, 10 rats per group received 1,000 mg/kg HC topically for 45 days, whereas the placebo group received the CC. Animal weight; food and water intake; behavior; hematological and biochemical parameters (e.g., glucose levels; lipid profile; liver/kidney function; and Na, K, and Ca levels); and liver, spleen, kidney, and lung histopathology were assessed at baseline, day 23, and day 45. Data were analyzed using analysis of variance (ANOVA) and Student’s t-tests, with p < 0.05 considered statistically significant.

13. Statistical analysis

Descriptive statistics were used to summarize baseline demographic and clinical characteristics and symptom scores at each visit. Continuous variables were reported as means and standard deviations, whereas categorical variables were reported as frequencies and percentages. A baseline characteristics table was prepared for the per-protocol/complete-case population included in the final analysis.

Repeated-measures ANOVA was used to evaluate changes in symptom scores over time and compare changes between treatment groups. The model included treatment group as the between-subject factor and time as the within-subject factor. Levene’s test was used to assess the homogeneity of variance. Mauchly’s test was used to assess the sphericity assumption; when it was violated, Huynh–Feldt-corrected data were used.

The primary interpretation focused on the treatment-group effect and treatment-by-time interaction. A significant time effect was interpreted as evidence of symptom change over time, not as evidence of the superiority of HC over CC. Within-group pairwise comparisons were considered exploratory and not used to support claims of comparative efficacy between treatments.

Pairwise comparisons were performed as post hoc exploratory analyses. Because multiple symptom outcomes and repeated pairwise comparisons were examined, the results were interpreted cautiously. The effect sizes were calculated to determine the magnitude of the observed effects. Analyses were performed in R version 4.2.2.

The final analysis was performed using a per-protocol/complete-case approach, including only participants who completed baseline, postintervention, and follow-up assessments. A full intention-to-treat analysis was not performed, as outcome data after dropout were unavailable for participants who discontinued follow-up. Therefore, the influence of missing outcome data and attrition bias should be considered when interpreting the findings.

All statistical tests were two-sided. p-values < 0.05 were considered statistically significant in the main analyses, whereas exploratory pairwise comparisons were interpreted in the context of multiple testing and the nonsignificant treatment-by-time interaction.

14. Ethical considerations

The SUMS Ethics Committee approved this study. All participants were fully informed of the terms and conditions of the investigation and were asked to complete and sign an informed written consent form. Participants incurred no costs for their visits or medication. Data records, patient histories, and patient identities were kept confidential throughout the study.

RESULTS

1. Physicochemical data

Physicochemical properties and quantitative assays of raw materials are presented in Table 2, and the results of the analysis of the extracts and essential oil of myrtle are presented in Table 3.

Table 2.

Physicochemical properties of the raw materials

Plant name	Herbarium code	Pharmacopoeia standards
		Foreign material			Total ash (%w/w)			Acid-insoluble ash (%w/w)			Assay		Reference
		Standard	Result		Standard	Result		Standard	Result		Standard	Result
Allium ampeloprasum	249-1	≤ 2%	0.7%		-	-		-	-		-	-	In house
Elaeagnus angustifolia	248-1	< 2%a	1.5%		< 4%b	3.2%		-	-		-	-	aISIRI 6257 bGhandiasl & Nourani, 2019
Matricaria recutita	210-1	≤ 2%	0.5%		≤ 13%	8%		-	0.5%		Apigenin-7-glucoside: ≥ 0.3% Volatile oil: ≥ 0.4%	0.5% 0.6%	USP 44, HPLC
Myrtus communis	142-1	≤ 2%	1.5%		≤ 5%	8%		≤ 3%	1%		Volatile oil: ≥ 1.5%	1.5%	Iranian Herbal Pharmacopoeia

Table 3.

Results of the analysis of the extracts and essential oil of myrtle

Plant name	Tests
	Appearance			Odor			Clarity			Density (g/mL)			Dry residue (%w/w)			Assay		Reference method
	Standard	Result		Standard	Result		Standard	Result		Standard	Result		Standard	Result		Standard	Result
Allium ampeloprasum	Dark green liquid	Dark green liquid		Characteristic to Allium	Characteristic to Allium		Turbid	Turbid		1.00-1.05	1.01		≥ 3	3.7		Total phenols as gallic acid (mg/mL): > 0.90	1.18	USP 841, method 1;EP 8.0, 2.8.16 and 2.2.25
Elaeagnus angustifolia	Brown viscous liquid	Brown viscous liquid		Special	Special		Turbid	Turbid		-	-		≥ 75	80		Total phenols as gallic acid (mg/mL): ≥ 0.70	0.81	EP 8.0, 2.8.16 and 2.2.25
Matricaria recutita	Dark Brown liquid	Dark Brown liquid		Special	Special		Turbid	Turbid		1.150-1.170	1.164		≥ 40	46.5		Apigenin (%w/w): ≥ 0.17	0.19	USP 841, method 1;EP 8.0, 2.8.16 and 2.2.29
	Appearance			Odor			Refractive index			Density (g/mL)			Optical rotation			Assay
	Standard	Result		Standard	Result		Standard	Result		Standard	Result		Standard	Result		Standard	Result
Myrtus communis	Yellow to greenish yellow liquid	Yellow liquid		Characteristic to myrtle	Characteristic to myrtle		1.464-1.468	1.4641		0.879-0.904	0.879		–3 to +26	+15.8		1,8-cineole (%w/w): ≥ 15	24.3	USP 841, method 1; EP 8.0, 2.2.6, 2.2.7 and 2.2.28 (GC)

2. Total phenolic content

TPC was quantified using a standard curve generated with GA solutions measured at 760 nm using the following equation:

Abs = 0.0043C + 0.0371 (r² = 0.997),

where Abs represents absorbance, and C is the GA concentration (μg/mL) in the sample solution. The results were expressed as milligrams of TPC as GA equivalents per gram of cream. The absorbance spectrum shows the TPC (measured in GA) in creams derived from three plants E. angustifolia, Matricaria chamomilla, and A. ampeloprasum (Fig. 3).

Figure 3.

Absorption spectrum of herbal cream (HC) in the reaction with Folin–Ciocalteu reagent measured at 760 nm to determine the total phenolic content.

The essential oil in HC is derived from Myrtus communis leaves, which contain 0.1%-0.5% volatile oils according to PDR, and one primary component is 1,8-cineole. Fig. 4 shows the comparison of the standard 1,8-cineole chromatogram with that of the essential oil of the HC, confirming the presence of 1, 8-cineole.

Figure 4.

Gas chromatography of 1,8-cineole in a standard (A) and essential oil extracted from herbal cream (B).

3. Safety monitoring

Toxicological assessments of rats showed no significant changes in weight, food and water intake, grooming, social behavior, or hematologic and biochemical factors with HC exposure. At 23 and 45 days, minor, reversible histological changes were observed in the liver, spleen, kidneys, and lungs. The changes included hematopoietic/inflammatory foci, a slight increase in red blood cell count, lymphocyte/eosinophil infiltration, and mild edema, all of which were considered nontoxic (Fig. 5). Overall, the HC showed no major side effects relative to the CC.

Figure 5.

Histopathological changes in rat livers, spleens, kidneys, and lungs following local administration of the herbal cream for 45 days.

4. Participants

In this study, 29 patients (including 21 females and 8 males) participated in the CC and 20 patients (including 17 females and 3 males) in the HC group. The mean ages of the participants in the CC and HC groups were 45.79 ± 14.25 and 48.50 ± 13.26 years, respectively. Table 4 summarizes the baseline characteristics of participants included in the per-protocol/complete-case analysis, summarizing the analyzed population. The data do not confirm that the balance achieved by randomization was preserved after attrition. p-values are descriptive and cannot be used for formal inference regarding randomization.

Table 4.

Baseline characteristics of participants in the per-protocol/complete-case analysis

Variable	Conventional cream (n = 29)	Herbal cream (n = 20)	Descriptive p-value
Age, years, mean ± SD	45.79 ± 14.25	48.50 ± 13.26	0.500
BMI, kg/m2, mean ± SD	25.32 ± 3.70	25.38 ± 2.50	0.949
Male, n (%)	8 (27.6%)	3 (15.0%)	0.488
Female, n (%)	21 (72.4%)	17 (85.0%)	0.488
Baseline bleeding score, mean ± SD	0.69 ± 0.97	1.05 ± 0.83	0.169
Baseline anal pain score, mean ± SD	4.48 ± 3.71	4.40 ± 3.41	0.936
Baseline difficult defecation score, mean ± SD	3.93 ± 3.68	3.90 ± 3.96	0.978
Baseline pruritus ani score, mean ± SD	1.31 ± 2.12	2.00 ± 1.45	0.183
Baseline burning sensation score, mean ± SD	1.66 ± 2.14	1.80 ± 1.44	0.778

Continuous variables are presented as means ± standard deviations (SDs); categorical variables are shown as counts and percentages. p-values are descriptive.

5. Outcomes

Table 5 summarizes the descriptive statistics (mean and standard deviation) for the examined signs and symptoms in each group at three measurement points: study onset, after 2 weeks, and after 6 weeks.

Table 5.

Means (M) and standard deviations (SDs) of manifestations at each visit and all visits for treatment groups and all patients

	Manifestations	First visit			Second visit			Follow-up			All visits combined
		M	SD		M	SD		M	SD		M	SD
Conventional cream	Bleeding	0.69	0.97		0.55	0.21		0.55	0.58		0.60	0.34
	Anal pain	4.48	1.71		3.72	1.81		3.38	1.84		3.86	1.77
	Difficult defecation	3.93	1.68		2.69	1.69		3.03	1.56		3.22	1.64
	Pruritus ani	1.31	0.88		0.69	1.17		0.72	0.36		0.91	0.55
	Burning sensation	1.66	1.14		1.52	0.86		1.03	0.48		1.40	0.55
Herbal cream	Bleeding	1.05	0.83		0.50	0.44		0.50	0.33		0.68	0.44
	Anal pain	4.40	1.41		2.80	1.22		2.80	1.32		3.33	1.35
	Difficult defecation	3.90	2.96		2.80	1.61		2.65	1.83		3.12	1.78
	Pruritus ani	2.00	1.45		1.30	0.78		1.60	0.70		1.63	0.65
	Burning sensation	1.80	1.44		1.55	0.50		1.15	0.27		1.50	0.41
All patients	Bleeding	0.84	0.92		0.53	0.21		0.53	0.33		0.63	0.44
	Anal pain	4.45	3.55		3.35	1.57		3.14	1.61		3.65	1.60
	Difficult defecation	3.92	1.76		2.73	1.62		2.88	1.64		3.18	1.68
	Pruritus ani	1.59	0.89		0.94	0.46		1.08	0.44		1.20	0.63
	Burning sensation	1.71	0.87		1.53	0.91		1.08	0.77		1.44	0.86

6. Clinical effectiveness

The effectiveness of the creams was evaluated using a repeated-measures multivariate design. The assumptions for ANOVA were assessed. Levene’s test did not show significant heterogeneity of variance across treatment groups. Mauchly’s test was used to assess the sphericity assumption, and Huynh–Feldt corrections were applied where appropriate.

The primary multivariate repeated-measures analysis showed no significant treatment-group effect (F[5,43] = 0.89, p = 0.50, effect size = 0.09). Additionally, the treatment-by-time interaction was nonsignificant (F[10,38] = 0.64, p = 0.77, effect size = 0.14). This indicates that symptom changes over time did not differ significantly between the HC and CC groups. A significant time effect was observed (F[10,38] = 2.78, p = 0.011, effect size = 0.42), indicating that symptom scores changed over time in the analyzed participants. However, a significant time effect alone indicates overall symptom change during the study period and does not demonstrate the superiority of HC over CC (Table 6).

Table 6.

Multivariate repeated-measures analysis of symptom outcomes

Effect source	Value	F	Hypothesis df	Error df	p-value	Effect size
Treatment	0.91	0.89	5	43	0.50	0.09
Time	0.58	2.78	10	38	0.011	0.42
Treatment × time	0.86	0.64	10	38	0.77	0.14

1) Univariate repeated-measures analysis

Univariate repeated-measures analyses showed no significant between-group effect for any individual symptom (bleeding: F = 0.13, p = 0.72; anal pain: F = 0.30, p = 0.59; difficult defecation: F = 0.01, p = 0.92; pruritus ani: F = 3.45, p = 0.07; burning sensation: F = 0.04, p = 0.85).

Similarly, no significant treatment-by-time interaction was observed for any symptom: bleeding: F = 1.95, p = 0.15; anal pain: F = 0.73, p = 0.47; difficult defecation: F = 0.29, p = 0.74; pruritus ani: F = 0.21, p = 0.75; burning sensation: F = 0.05, p = 0.94. However, the time effect was significant for bleeding (F = 5.44, p = 0.006, effect size = 0.10), anal pain (F = 8.83, p < 0.001, effect size = 0.16), difficult defecation (F = 7.55, p = 0.001, effect size = 0.14), and burning sensation (F = 6.25, p = 0.004, effect size = 0.12). For pruritus ani, the time effect was p = 0.014, which is below the conventional 0.05 threshold but does not meet the Bonferroni-adjusted threshold of 0.01 for five symptom outcomes; thus, this finding should be interpreted cautiously (Table 7).

Table 7.

Univariate repeated-measures results for symptom outcomes

Outcome	Effect	F	p-value	Effect size	Interpretation
Bleeding	Between group	0.13	0.72	0.003	Not significant
Bleeding	Time	5.44	0.006	0.10	Significant time effect
Bleeding	Treatment × time	1.95	0.15	0.04	Not significant
Anal pain	Between group	0.30	0.59	0.006	Not significant
Anal pain	Time	8.83	< 0.001	0.16	Significant time effect
Anal pain	Treatment × time	0.73	0.47	0.015	Not significant
Difficult defecation	Between group	0.01	0.92	0.000	Not significant
Difficult defecation	Time	7.55	0.001	0.14	Significant time effect
Difficult defecation	Treatment × time	0.29	0.74	0.006	Not significant
Pruritus ani	Between group	3.45	0.07	0.07	Not significant
Pruritus ani	Time	5.21	0.014	0.10	Cautious interpretation after multiple testing
Pruritus ani	Treatment × time	0.21	0.75	0.004	Not significant
Burning sensation	Between group	0.04	0.85	0.001	Not significant
Burning sensation	Time	6.25	0.004	0.12	Significant time effect
Burning sensation	Treatment × time	0.05	0.94	0.001	Not significant

2) Exploratory within-group comparisons

In the HC group, bleeding decreased from baseline to the postintervention visit (mean difference, 0.55; 95% CI: 0.05-1.05, p = 0.027). Bleeding also decreased from baseline to follow-up (mean difference, 0.55, 95% CI: 0.06-1.04, p = 0.022). No significant change was observed between the postintervention visit and follow-up, suggesting that the mean bleeding score remained stable after the intervention period. In the CC group, changes in bleeding were not significant. Fig. 6 shows the mean bleeding scores over time.

Figure 6.

Mean bleeding scores in treatment groups from the first visit to follow-up.

In the HC group, anal pain decreased from baseline to the postintervention visit (mean difference, 1.60; 95% CI: 0.28-2.92, p = 0.013). Anal pain also decreased from baseline to follow-up (mean difference, 1.60; 95% CI: 0.05-3.15, p = 0.041). No significant change was observed between the postintervention visit and follow-up. In the CC group, changes in anal pain were not significant in the pairwise comparisons. Fig. 7 shows the mean anal pain scores over time.

Figure 7.

Mean anal pain scores in treatment groups from the first visit to follow-up.

Difficult defecation in the CC group decreased from baseline to the postintervention visit (mean difference, 1.24; 95% CI: 0.04-2.44; p = 0.041). Other pairwise changes in the CC group were not significant. In the HC group, the reduction from baseline to follow-up showed a marginal trend (mean difference, 1.25; 95% CI approximately crossing zero, p = 0.06). Fig. 8 shows the mean difficult defecation scores over time.

Figure 8.

Mean difficult defecation scores in treatment groups from the first visit to follow-up.

Burning sensation exhibited no significant pairwise change in the HC group. In the CC group, the reduction from baseline to follow-up was marginal (mean difference, 0.62; 95% CI: –0.05 to 1.29; p = 0.078). Fig. 9 shows the burning sensation scores over time.

Figure 9.

Mean burning sensation scores in treatment groups from the first visit to follow-up.

No significant pairwise differences in pruritus ani were observed between the two groups. Fig. 10 shows the pruritus ani scores over time.

Figure 10.

Mean pruritus ani scores in treatment groups from the first visit to follow-up.

3) Exploratory comparison of symptom change scores

To clarify the magnitude and direction of symptom changes, exploratory change-score comparisons from baseline to follow-up were examined descriptively. Mean changes in the HC group versus the CC group were as follows: bleeding, –0.55 vs. –0.14; anal pain, –1.60 vs. –1.10; difficult defecation, –1.25 vs. –0.90; pruritus ani, –0.40 vs. –0.59; burning sensation, –0.65 vs. –0.62. No significant differences were observed between the groups. The estimated between-group differences in change from baseline to follow-up were as follows: bleeding, –0.41 (95% CI: –0.92 to 0.09), p = 0.108; anal pain, –0.50 (95% CI: –2.19 to 1.20), p = 0.557; difficult defecation, –0.35 (95% CI: –1.73 to 1.02), p = 0.607; pruritus ani, 0.19 (95% CI: –0.72 to 1.09), p = 0.680; and burning sensation, –0.03 (95% CI: –0.90 to 0.84), p = 0.946. These results are consistent with the repeated-measures analysis and do not indicate a significant superiority of HC over CC (Table 8).

Table 8.

Exploratory change-score comparison from baseline to follow-up

Outcome	Mean change in CC	Mean change in HC	Between-group difference in change	95% confidence interval	p-value
Bleeding	–0.14	–0.55	–0.41	–0.92 to 0.09	0.108
Anal pain	–1.10	–1.60	–0.50	–2.19 to 1.20	0.557
Difficult defecation	–0.90	–1.25	–0.35	–1.73 to 1.02	0.607
Pruritus ani	–0.59	–0.40	0.19	–0.72 to 1.09	0.680
Burning sensation	–0.62	–0.65	–0.03	–0.90 to 0.84	0.946

7. Side effects

Based on the Common Terminology Criteria for Adverse Events version 5.0, no adverse effects were observed during the treatment or follow-up periods in either treatment group.

DISCUSSION

This study evaluated the standard local treatment for hemorrhoids with a multicomponent HC formulation developed based on traditional texts and previous experimental and clinical evidence. Specifically, this randomized, triple-blind, active-controlled trial revealed that hemorrhoid-related symptom scores changed over time among participants in the per-protocol/complete-case analysis. However, the main multivariate repeated-measures analysis revealed no significant treatment-group effect or treatment-by-time interaction. Therefore, the findings do not demonstrate a statistically significant superiority of the HC over the CC.

This distinction is crucial for interpreting the trial results. A statistically significant improvement within a single treatment group without a corresponding between-group difference or treatment-by-time interaction cannot be used as evidence that one intervention is superior to the other. Accordingly, the observed within-group reductions in symptoms should be considered exploratory and hypothesis-generating rather than confirmatory evidence of comparative efficacy.

In the analyzed population, both HC and CC were associated with improvements in various hemorrhoid-related symptoms over time. Exploratory pairwise comparisons suggested that participants receiving HC experienced reductions in bleeding and anal pain, whereas those receiving CC experienced improvements in difficult defecation during the intervention period. However, these findings should be interpreted as descriptive symptom changes within the analyzed participants rather than conclusive evidence that HC is more effective than CC.

The beneficial effects observed in the HC group over time could be attributed to the anti-inflammatory, analgesic, antioxidant, constipation-relieving, and antimicrobial properties of the creams. Previous studies have confirmed the healing effects of chamomile on wounds [27, 28]. Flavonoids, such as quercetin and apigenin, are known to inhibit matrix-metalloproteinase-1 (MMP-1), downregulating its expression by suppressing activator protein-1 activation. This suppression can promote increased collagen production, aiding in wound healing [29]. Although these mechanisms provide a plausible pharmacological rationale for the effects of HC, they should not be interpreted as direct evidence of clinical superiority without significant comparative effects.

Elaeagnus angustifolia exerts antioxidant, anti-inflammatory, analgesic, antimicrobial, antitumor, and antimutagenic effects [30, 31]. It lowers levels of inflammatory cytokines such as TNF-α and MMP-1 while increasing the levels of anti-inflammatory cytokines such as interleukin-10. Furthermore, reducing MMP-1 expression inhibits collagen degradation, potentially accelerating wound healing [32]. These effects may help explain the symptom improvements observed in participants receiving the HC over time. Nevertheless, because this study evaluated the HC as a multicomponent product, the independent contributions of each plant ingredient could not be isolated.

The therapeutic effects of A. ampeloprasum have been assessed in previous clinical studies [16, 17], in which leek was applied topically as a cream. The results showed that local leek treatment was effective for treating the primary symptoms of hemorrhoids, likely by inhibiting phagocytosis through reduced nitro-oxidative stress [33]. Our findings are consistent with those of previous reports suggesting that topical herbal preparations can improve hemorrhoid-related symptoms. Nevertheless, this does not confirm that the combined HC formulation was superior to topical CC therapy in the present trial.

Myrtus communis has been traditionally used in Iranian medicine for hemorrhoid management, and its efficacy in modern formulations has been supported by clinical trials. Clinical studies have indicated that M. communis essential oil, whether in ointment or lotion form and standardized for 1,8-cineole, is as effective as conventional antihemorrhoid treatments in relieving key symptoms, particularly bleeding and pain, even in patients refractory to standard therapy. These benefits are attributed to its anti-inflammatory, analgesic, anesthetic, and vasoactive properties, making Myrtus communis a safe and effective botanical option for hemorrhoid therapy [18]. The observed reductions in bleeding and anal pain in the HC group are consistent with this biological and clinical rationale. However, because the treatment-by-time interaction was nonsignificant, these reductions should be interpreted cautiously and not as definitive evidence that HC is more effective than CC in reducing bleeding or pain.

Clinical evidence suggests that organic chamomile flower extract (Matricaria chamomilla), which is rich in flavonoids, can effectively alleviate symptoms in patients with acute hemorrhoidal bleeding. In a study of 24 patients, a daily intake of 200 µL/kg of body weight for 2 months markedly reduced the frequency and severity of hemorrhoidal symptoms in 21 patients, with symptoms almost fully resolved after 1 month. Bleeding, pain, and itching were markedly reduced. Compounds in the extracts of chamomile flowers potentially inhibit inflammation and improve microcirculation in the bowel, thereby ameliorating hemorrhoids and bleeding. Overall, clinical data suggest that organic chamomile extract effectively alleviates acute hemorrhoidal symptoms, including bleeding, pain, and itching, likely through flavonoid-mediated anti-inflammatory effects and enhanced local microcirculation [34].

In a triple-blind randomized controlled trial, Myrtus communis ointment significantly reduced hemorrhoidal symptoms in females with grade I-II disease, with efficacy comparable to standard antihemorrhoid ointment [35]. Symptom severity and quality of life improvements were similar between groups; however, anal itching was better alleviated with Myrtus communis ointment. Patient satisfaction was significantly higher in the Myrtus communis group, supporting its effectiveness and acceptability as an herbal alternative for hemorrhoid management.

This evidence suggests that herbal topical therapies may have clinical value. The present study provides comparative data on a multicomponent formulation, although it shows symptom improvement over time rather than statistical superiority of the HC over conventional treatment.

Among the studies conducted in the colorectal region, experimental and clinical evidence suggest that the local effects of Matricaria chamomilla, E. angustifolia, A. ampeloprasum, and Myrtus communis on hemorrhoid symptoms can be comparable to those of conventional topical medications [36]. The findings of the present study should be positioned within this cautious interpretation: HC may be a useful topical herbal formulation for symptom relief; however, based on the current data, its comparative advantage over CC remains uncertain.

Overall, experimental and clinical evidence on the four aforementioned ingredients confirms the hypothesis tested in this trial. These findings suggest that the local effects of chamomile, Russian olive, wild leek, and myrtle on hemorrhoid manifestations are comparable with those of conventional topical medications.

Recent studies have shown that Unani compounds improve hemorrhoids. For example, Habb-e-Bawaseer reduced grade 1-2 hemorrhoid symptoms such as bleeding, pain, itching, and constipation and improved proctoscopic findings and hemoglobin levels [37, 38].

Habb-e-Bawaseer Khooni has been reported to be effective for grades 1-3 internal hemorrhoids compared with Euphorbia prostrata and contains bearberry, neem, chinaberry, pomegranate, and seeds of leek and Centratherum anthelminticum [39].

Sixty-day use of Habb-e-Rasaut, containing bearberry, and Habb-e-Muqil, containing guggul, has also been reported to have therapeutic effects [40]. However, these studies have methodological and ethical limitations [41].

Therefore, although traditional and complementary medicine systems provide useful therapeutic leads, robust randomized trials with transparent reporting, adequate sample size, prospectively defined outcomes, and appropriate handling of missing data are necessary before making firm comparative claims.

The lack of significant between-group differences in the present trial may be attributed to several factors. CC was an active comparator containing lidocaine, hydrocortisone acetate, aluminum subacetate, and zinc oxide; therefore, it was not an inert control. Demonstrating superiority over an active conventional therapy would require sufficient statistical power and a well-defined comparative endpoint. Additionally, only 49 of the 110 randomized participants completed all assessments, resulting in a small final sample. This reduced the precision of effect estimates and limited the ability to detect small or moderate between-group differences.

The safety results were promising but should be interpreted with caution. No adverse events were observed during treatment or follow-up among participants included in the per-protocol/complete-case analysis, and no significant toxic effects were observed in the preclinical safety assessment. However, because of the limited final sample size, larger studies are necessary to obtain more reliable safety data, particularly regarding repeated or prolonged use of the herbal formulation.

Overall, this study suggests that both HC and CC were associated with improvements in hemorrhoid-related symptoms over time among participants who completed the trial. The herbal formulation is supported by traditional use and pharmacological rationale, and the observed within-group reductions in bleeding and anal pain are clinically relevant indications for further research. Nevertheless, the current evidence is preliminary. The trial does not establish that HC is superior to CC, and future studies should determine whether HC provides comparable or superior outcomes under more rigorous conditions, such as larger sample sizes, enhanced retention, prospectively defined primary outcomes, and intention-to-treat analysis.

LIMITATIONS

This study has several limitations. The final analysis was per-protocol/complete-case because only 49 of 110 randomized participants completed all assessments, which was largely attributable to the disruptions during the COVID-19 pandemic. This high attrition rate may have introduced attrition bias and reduced statistical power. Furthermore, an intention-to-treat analysis was not feasible because outcome data after dropout were unavailable. Moreover, no single primary outcome was predefined; multiple symptom outcomes were evaluated, most of which were patient-reported. Consequently, these results should be interpreted as exploratory and require confirmation through larger randomized trials with prospectively specified primary outcomes, improved retention, and intention-to-treat analysis.

CONCLUSION

The herbal formulation evaluated in this study, containing Myrtus communis, Matricaria chamomilla, A. ampeloprasum, and E. angustifolia, was safe and well tolerated in patients with grade I-II hemorrhoids, with no treatment-related adverse effects observed. Although it did not demonstrate statistical superiority over CC, the HC was associated with clinically meaningful improvements in anal bleeding and pain along with favorable changes in several other hemorrhoidal symptoms. These findings support the therapeutic potential of this herbal formulation as a safe, natural approach to symptom management in patients with hemorrhoids. Larger, adequately powered, randomized clinical trials are needed to further assess its efficacy and define its role in clinical practice.

Overall, the HC containing Myrtus communis, Matricaria chamomilla, A. ampeloprasum, and E. angustifolia was safe and well tolerated in patients with grade I-II hemorrhoids, with no treatment-related adverse effects observed during the study period. Although it did not show significant superiority over the CC, patients receiving the HC experienced clinically meaningful improvements in anal bleeding and pain as well as benefits in several other symptom outcomes. Collectively, these findings suggest that this herbal formulation may be a safe and promising natural option for managing hemorrhoidal symptoms. Further adequately powered randomized clinical trials are necessary to clarify its therapeutic efficacy and clinical utility.