Study design, setting, protocol approval and consent: A simple randomized parallel open-labelled comparative study was conducted at Govt Nizamia Tibbi College, Telangana India from November 2005 to May 2006. The protocol was approved (Reg No.10/250/03 DRNTRUHS dt: 26/12/2006) by Dr NTR University of Health Sciences and all the patients gave written consent before initiation of the study.

Participants: The participants were recruited based on the signs and symptoms of abnormal uterine bleeding and endometrial thickness in pelvic ultrasonography and endometrial biopsy reports.

Inclusion and exclusion criteria: Female married patients of premenopausal age with changes in the menstrual pattern, and abnormal uterine bleeding with thickened endometrium>8mm in transabdominal or transvaginal pelvic ultrasonography were included in the study.¹⁰ Participants who underwent diagnostic dilation and curettage (DD&C) for diagnosis of the type of endometrial hyperplasia were included. The exclusion criteria were patients who showed cytological atypia on DD&C, blood dyscrasias, and other medical disorders.

Procedure: All the participants underwent assessment including history, physical examination, and blood investigations such as haemogram, random blood sugar, HIV, HbsAg, VDRL, Serum T3, T4, TSH, bleeding time, clotting time and platelet count before treatment. Transabdominal or transvaginal pelvic ultrasonography was carried out for endometrial thickness before treatment. Before treatment, all participants underwent diagnostic dilation and curettage (DD&C) for diagnosis of the type of endometrial hyperplasia. Post-treatment transabdominal or transvaginal pelvic ultrasonography was repeated in all participants for endometrial thickness. DD&C was carried out in participants who were willing for the procedure or who had an endometrial thickness of more than 11 mm after treatment. Follow-up visits were scheduled for 20 days for 3 consecutive cycles commencing from the 3rd day of the menstrual cycle.

Data collection tool: For data collection endometrial thickness was measured by transabdominal or transvaginal pelvic ultrasonography before and after treatment. The cut-off value for endometrial thickness (ET) was 8-10 mm. Previous studies showed that the cut-off ET value was 8 mm with sensitivity and specificity of 83.9%, and 58.8%, respectively, and 90.4% negative predictive value for abnormal endometrium. 10 Haemoglobin percentage was measured by Sahli’s method before and after treatment. The duration of menstrual flow was observed in the days before and after treatment. The normal cut-off for the duration of menstrual flow was taken as 6 days.

Group A: Medicine included in group A were Itrifal Aftimoon, Marham Dakhilyun and Roghan Gul (Table 1).

Preparation and dosage: Itrifal Aftimoon was prepared by the Institute pharmacy. All the ingredients were dried and powdered in the grinder. Then the powder was sieved. Itrifal was prepared as mentioned in the traditional Pharmacopoeia. Marham Dakhilyun was directly purchased from the local market by the Hamdard company. Group A (n=20) received Itrifal Aftimoon, 5g orally twice daily after meals from day 3 of menses for 20 days plus suprapubic liniment application of Marham Dakhilyun and per vaginally Ḥamūl of Marham Dakhilyun (5g) and Roghan Gul (10ml) advised for 10 days from day 5 of menses. Table 1 Summarizes the ingredients of Itrifal Aftimoon and Marham Dakhilyun.

Group B: Medicine included in group B were powder of Gulnar Farsi, Phitakri Biryan, Dammul Aqwain, Geru, Bargh Sambhalu, Safuf Mazu, Kalijiri and Roghan Gul.

Preparation and dosage: Group B (n=20) received 2.5g powder of Gulnar Farsi (2g), Phitakri Biryan (0.25g), Dammul Aqwain (0.25g), and Geru (2g), orally twice daily after meals from day 3 of menses for 20 days plus Douche of Bargh Sambhalu followed by Ḥamūl of Safuf Mazu (2g), Kalijiri (2g) and Roghan Gul (10ml) for 10 days from day 5 of menses for 3 consecutive cycles.

Assessment of efficacy: The primary outcome included a change in pelvic ultrasound findings of endometrial thickness. The secondary outcome included improvement in haemoglobin and decrease in the duration of menstrual flow and a change in menstrual pattern.

Randomization and allocation: A total of 40 patients with AUB were recruited at random and assigned to either Group A or Group B in a 1:1 ratio using a lottery strategy. We used an open list of random numbers.

Sample size estimation: The sample size was calculated using sample size calculator software and was based on an earlier study's proportion value of cure rate of 20% and 39%.¹¹ The study would require a total sample size of 44 (n1=22 and n2=22), 5% significance, and 80% power. As a result, in the current study, a sample size of 40 patients was chosen, with a 20% dropout rate allowed.

Data analysis: The data was analyzed utilizing the statistical software Graph Pad Instat version 3.00 for Windows (Graph Pad Software, San Diego, Calif, USA). P0.05 was deemed significant for all statistical tests. All deviations from the baseline were compared between groups.

Participants flow: During the study period, a total of 73 patients were screened for abnormal uterine bleeding. Thirty-three patients were omitted from the trial for various reasons. Then, 40 patients were assigned to groups A and B at random (Figure 1).

Socio-economic, gynaecological and obstetrics parameters at baseline in groups A and B: The variables were comparable between groups (age, socio- economic status, religion, occupation, contraceptive history, per vaginal examination, parity, and last childbirth) at baseline. The mean age in group A was 40±5.1 and B was 41±6 years. Maximum participants were between 36-45 years of age [group A: n=10/20 (50%) and group B: n=7/20 (35%)]. Maximum participants were from middle socio-economic status [group A: n=12/20 (60%) and group A: n=10/20 (50%)]. There was no statistical difference in mean baseline measurements between the groups (Table 2).

Duration of illness, menstrual bleeding pattern and associated symptoms in groups A and B at baseline: Maximum participants had a duration of illness between one to three months in both groups (Group A: n=8/20, 40%; Group B: n=8/20, 40%) at baseline. Maximum participants had heavy menstrual bleeding with prolonged duration of menstrual flow in both groups (Group A: n=12/20, 60%; Group B: n=11/20, 55%) at baseline. Other details are summarized in Table 3.

Investigations in both groups at baseline: The haematological, biochemical, histopathological and pelvic Ultrasonography variables were comparable between groups before treatment (Hb%, T3, T4, and TSH). HIV, HBsAg, and VDRL were normal in all patients. Maximum participants had simple endometrial hyperplasia in both groups (Group A: n=14/20, 70%; Group B: n=13/20, 65%) and thickened endometrium in pelvic ultrasonography (Group A: n=13/20, 65%; Group B: n=15/20, 75%) (Table 4).

Primary and secondary outcomes in groups A and B at baseline and after treatment

Primary outcome: The primary outcome was a change in endometrial thickness in pelvic ultrasonography

Endometrial thickness in pelvic Ultrasonography: The mean endometrial thickness at baseline and after treatment in group A was 14.95 (3.00) and 7.75(3.12) mm respectively with a significant difference in P value <0.001. The mean endometrial thickness at baseline and after treatment in group B was 13.8(1.79) and 6.85 (2.58) mm respectively with a significant difference in P value <0.001. Group A and B comparisons at baseline (P=0.14) and after treatment (P=0.15) showed no statistical difference (see Table 5).

Secondary outcome: Secondary outcomes included improvement in haemoglobin and decrease in the duration of menstrual flow and a change in menstrual pattern.

Haemoglobin percentage: The mean haemoglobin per cent at baseline and after treatment in group A was 10.75 (1.29) and 11.62(0.99) per cent respectively with a significant difference in P value <0.0001. The mean haemoglobin per cent at baseline and after treatment in group B was 10.08(1.19) and 11.25(1.25) per cent respectively with significant differences in P value <0.0001. At baseline, between the group comparisons A and B showed not quite a statistical difference (P=0.09). After treatment, group A and B comparisons showed no statistical difference (P=0.68) (see Table 5).

Duration of menstrual flow and menstrual cycle in groups A and B at baseline and after treatment: Maximum participants had a duration of menstrual flow between 9 to 12 days in both groups (Group A: n=7/20, 35%; Group B: n=9/20, 45%) at baseline. After treatment duration of menstrual flow was less than 6 days in 50% (n=10) and 60% (n=12) participants respectively showing 50% and 60% of participants had normal duration and menstrual blood loss after normal menstrual bleeding was seen in 85% (n=17) and 90% (n=18) participants respectively showing 35% and 30% change after treatment from baseline in groups A and B respectively (see Table 6). treatment from baseline in groups A and B respectively. Maximum participants had a duration of the cycle between 25 to 35 days in both groups (Group A: n=9/20, 45%; Group B: n=12/20, 60%) at baseline. After treatment duration of the cycle between 25 to 35 days normal menstrual bleeding was seen in 85% (n=17) and 90% (n=18) participants respectively showing 35% and 30% change after treatment from baseline in groups A and B respectively (see Table 6).

Both groups were equally effective in reducing endometrial thickness regularizing menstruation and decreasing heavy menstrual bleeding. Unani scholars said that the initial stages of Waram-i-Sulb sometimes are the melancholic type and are the result of chronic Balghamī Waram. This type of swelling may progress into carcinoma. Unani medicine that is helpful in Amrade Bārid, Waram-i-Sawdāwī such as Itrifal Aftimoon, Marham Dakhilyun Ointment, ⁸ Phitakri Biryan, Dammul Aqwain, Geru, Bargh Sambhalu, Mazu, Kalijiri, and Roghan Gul ⁹ are useful to treat endometrial hyperplasia and abnormal uterine bleeding that possess Muhallil Waram, Habis, Qabiz, Mundij Sawdā’, etc properties. ⁹ Most of the aforementioned medicines have Bārid wa Yābis temperament including Geru, Mazu, and Phitakri Biryan ¹² which helps in haemostasis. Moreover, Kathrat-i-Hayd is triggered by Ḍu‘f Quwwat Māsika (weak retentive power) and Qāwi Quwwat Dāfi‘a (strong expulsive power) and it is supposed that Bārid wa Yābis drugs tone up the Quwwat Ghādhiya of Rahim (nutritive power of uterus) and ultimately rectify the abnormality of Quwwat Māsika and Quwwat Dāfi‘a. ^{12, 13} Furthermore, these medicines are pharmacologically proven for anti-inflammatory, anti-estrogenic, anti-proliferative, styptic as they possess tannins, flavonoids isoflavonoids, saponins, alkaloids, and other bioactive components (see table 7). ^{14, 15, 16} The response of the trial drugs in both groups was due to the Qabiz (astringent) property which helps to control excessive bleeding and these herbs with astringent activity also produce a protective coating on the tissue surface.¹⁷

Although the particular mechanism of action of these herbs is unknown, it has been hypothesized that these plant components and minerals are useful because they have been demonstrated for astringent, anti-inflammatory, blood purifier, antioncogenic, anti-proliferative and hemostatic properties attributed to bioactive phytoconstituents. Oestrogen is the main reason for the increase in the thickness of the endometrium leading to endometrial hyperplasia. Geru contains calcium that helps to maintain the hemostatic mechanism.¹² Gulnar (Punica granatum) possess strong anti-oestrogenic, anti-inflammatory and antioncogenic activities.¹⁸According to Kim et al., polyphenols from aqueous pericarp extract can suppress the activity of 17--hydroxysteroid dehydrogenase. Polyphenols from the pericarp of pomegranate juice reduced the growth of ER+ MCF-7 and ER- MB-MDA-231 breast cancer cell lines in terms of anti-estrogenic actions. ¹⁹ According to new research, ellagic acid may have both estrogenic and anti-estrogenic effects depending on the oestrogen receptor to which it binds. ²⁰Sambhalu (Vitex negundo Linn) possess anti-inflammatory and anti-oestrogenic properties.21Mazu (Quercus infectoria) possess a high concentration of tannins (50-70%) and is used for the treatment of menorrhagia.²² Murdarsang (Litharge) possesses astringent and anti-inflammatory properties.²³

Various studies have steadily confirmed the importance of oestrogens in regulating endometrial cell proliferation, angiogenesis and inflammation.²⁴ The endometrium includes a balanced cytokine system with various linkages during the proliferative and secretory stages of the menstrual cycle. Although inflammation is the most frequent feature in most hyperplasia situations, some research has concentrated on the involvement of various pro- and anti-inflammatory cytokines in EH development. EH was related to “reduced production of tumour necrosis factor-α (TNF-α), proliferating cell nuclear antigen, and epithelial growth factor mRNA and enhanced production of Fas mRNA”. Also, TNF- was also shown to be expressed in normal endometrium as well as simple and complicated hyperplasia, while it was downregulated in atypical hyperplasia and endometrial ca. The transcription factor nuclear factor-κB was also found in proliferative endometrium and EH. ¹ This shows that anti-inflammatory herbs may have the potential to treat EH and thereby regularize menstruation. Table 7 summarizes that most of the ingredients of both groups have anti-inflammatory properties. Soy isoflavonoids are well-known inhibitors of protein-tyrosine kinases and topoisomerase-II. ²⁵ Isoflavonoid can be found in genistein. Through cytokine and ER-mediated mechanisms, genistein inhibits the internal cytokines IL-1 α and TNF- α.²⁶ Likewise herbal medicine that contains isoflavonoids are beneficial in suppressing inflammation. Tannins in Mazo possess anti-inflammatory potential, which is positively related to their antioxidant activities. Tannins in experimental studies modulate the inflammatory cytokine release and inhibit the production of nitric oxide (NO) and prostaglandins. Besides Mazo also possesses anti-proliferative effects in vitro conditions. ²²

Numerous authors have documented the link between inflammation and oxidative stress. Evidence indicates that oxidative stress plays a pathogenic role in chronic inflammatory diseases. ²⁷ Antioxidants have anti-inflammatory actions that limit nociceptor activity and reduce the production and/or release of prostaglandins that act as inflammatory pain mediators. By blocking the NF-kB pathway, a substance can exhibit both antioxidant and anti-inflammatory characteristics. ²⁸ Plants metabolites such as tannin have anti-inflammatory, haemostatic analgesic, and effects. ²⁹ Flavonoids have anti-inflammatory, antioxidant, and analgesic effects ²⁹. Flavonoids can scavenge lipid peroxyl radicals, superoxide anion radicals and hydroxyl radicals, and play a key role in preventing illnesses caused by oxidative damage to membranes, proteins and DNA. Saponins and alkaloids have anti-inflammatory properties. Anti-inflammatory activity aids from anti-oxidant characteristics.³⁰ Polyphenols also have numerous biological activities. Before cell viability is seriously affected, phenolic compounds and flavonoids can interact with ROS/RNS and thus terminate the chain reaction.

Currently, research has established that alum has antihemorrhagic, anti-inflammatory, and antimicrobial properties. Alum is documented to inhibit inflammation via several mechanisms, and its effects include immune cell function inhibition (reduction in lymphocyte infiltration, decrease in dilation, blood vessel congestion and inhibition of goblet cell proliferation). ³¹ Hence, this study validates that both group Unani regimens were beneficial in abnormal uterine bleeding due to endometrial hyperplasia.

The research Unani medicines reduced excessive menstrual bleeding, reduced inflammation and reduced the thickness of hyperplastic endometrium. As a result of the aforementioned features, both groups were equally effective.

Strengths of the study: This is the first kind of study that evaluated the effectiveness of Holistic Unani therapy in abnormal uterine bleeding due to endometrial hyperplasia. Besides, it was a randomized, parallel comparative study and no adverse effects were reported.

Limitations and recommendations: The limitations include no follow-up assessment after treatment and no assessment of progression to uterine cancer. A double-blind study could not be carried out due to a lack of facilities, equipment, resources, and staff. To evaluate the efficacy of trial medications on endometrial hyperplasia, additional phase II and III studies, double-blind, placebo/standard controlled with longer treatment duration and follow-ups are needed. The purpose of this study was to validate the efficacy and safety of the Unani formulations in abnormal uterine bleeding. The authors also suggest conducting standardization and quantitative analysis of Unani formulations as well as stability evaluation of the finished product. Furthermore, they recommend checking the presence of active constituents in the bloodstream to assess their absorption and safety. Therefore, comprehensive pharmacokinetics and pharmacodynamics studies are recommended. Furthermore, research is required to determine the precise mechanism of action of these Unani compositions and qualitative analysis of the formulations.

This study reveals that Group A and B were equally beneficial in the treatment of abnormal uterine bleeding due to endometrial hyperplasia as research medicines regularized abnormal uterine bleeding and normalized endometrial thickness by their anti-inflammatory, anti-proliferative and astringent properties. Furthermore, experiments comparing the efficacy of both groups with conventional control are recommended.