Evaluation of a Chinese herbal supplement on equine squamous gastric disease and gastric fluid pH in mares

Abstract Background Wei Le San (WLS) is a Chinese herbal formula comprised of 9 herbs selected for their putative anti‐inflammatory effects. Objectives To evaluate the effects of WLS administration in horses with nonglandular gastric ulcers. Animals Ten mixed breed mares (aged 7‐21 years, 401‐567 kg body weight). Methods Experimental design was a blinded, prospective, 2‐period crossover study. All horses received a placebo (25 mL dextrose‐based syrup; n = 10) and the treatment (WLS, 5 g in 25 mL dextrose‐based syrup; n = 10), administered twice daily. Horses underwent a 1‐week, alternating feed‐deprivation period to induce or worsen existing ulcers; treatment began on day 7. Gastroscopic examination was performed on d0, d6, and d35, with gastric fluid pH obtained on d6 and d35. Gastric ulcer scores assigned by 3 masked observers were averaged for each examination. Results Ulcer number scores for horses treated with WLS (median = 0; range, 0‐4) was not different from the untreated controls (median = 0.5; range, 0‐4; P = .81) by the end of the treatment period. Ulcer severity score for treated horses (median = 0; range, 0‐1) was also unchanged compared to the control group (median = 0.5; range, 0‐1; P = .85). Gastric pH was not altered by either treatment, with a median of 2.1 (range, 1.9‐4.1) for the horses treated with WLS and 2.8 (range, 1.6‐7.2) in the untreated controls (P = .46). Conclusions and Clinical Importance The experimental model used to induce gastric ulceration was unable to discern a difference between the herbal supplement and the placebo in normal horses.

Treatment of ESGD centers on medications that reduce gastric acid secretion. These include proton pump inhibitors that inhibit the H + K + ATPase pump in the secretory membrane of the gastric parietal cell, or H 2 -receptor antagonists that block the parietal cell receptor. 4 The most commonly used treatment protocol for ESGD is to administer medications to maintain gastric pH above 4, and omeprazole is the current treatment of choice. 4 Medical management of gastric ulceration in horses involves daily administration of these drugs for up to 28 days or more, which is a financial burden for owners of horses. 14,15 Recently, a Chinese herbal formulation, Wei Le San (WLS), has been recommended for the treatment of mild to moderate gastric ulceration in horses. [16][17][18] This proprietary herbal formula is derived from a combination of 2 traditional herbal formulas: Xiao Yao San and Er Chen Tang.
It contains 9 herbs selected for their putative individual abilities, in addition to their proposed synergistic effects on the gastrointestinal tract. A commercially manufactured preparation is available but has limited clinical evidence to support its use as a supplement in horses with gastric ulceration. 16 The purpose of this study was to determine the efficacy of WLS for equine nonglandular gastric ulcers induced by an intermittent feeding regime. We hypothesized that the administration of WLS would significantly decrease the nonglandular gastric ulceration scores compared to the control treatment, without an alteration in the gastric fluid pH.

| MATERIALS AND METHODS
All procedures performed on the horses in this study were evaluated and approved for ethical use of animals by the University of Wisconsin Institutional Animal Care and Use Committee (Protocol #V005953), and the protocol followed the NIH Guide for Care and Use of Laboratory Animals.
A total of 10 healthy, adult mares (7-21 years of age; 400.9-567.2 kg body weight) from the resident herd at the University of Wisconsin Charmany teaching facility were obtained as a convenience sample. A minimum sample size of 10 was determined using ordinal logistic regression for the differences in ulcer number scores, using an alpha of 0.05, power of 0.8, and a 2-tail test. 19 Mares were not screened based on endoscopic ulcer scores before the study. Breeds of horse included 3 American Quarter Horses, 3 Thoroughbreds, and 1 each of a warm blood, Paint, Tennessee walking horse, and Paso Fino. At the time of enrollment in the study, a physical examination was performed to exclude those with clinical disease. All horses were vaccinated for Eastern Equine Encephalitis, Western Equine Encephalitis, West Nile Virus Encephalitis, Tetanus, and Influenza, and an anthelmintic (ivermectin, 0.2 mg/kg, PO) was provided twice yearly as part of the routine health management program.

| Experimental design
The experimental protocol was designed as a 2-period, crossover design. Each horse completed the 5-week protocol twice, once with herbal treatment and once with the placebo to serve as their own controls. The horses were ranked and paired by ulcer number and severity combined scores (Table 1), 14,20,21 after the first week of feed deprivation in period 1, and the pairs then randomly allocated by coin flip to 1 of 2 groups for the remainder of the study. During the study period, the horses were housed in a dry lot and fed a free choice, native grass hay ration. The treatment period was 5 weeks (35 days) in duration, where the first week consisted of a modified intermittent feed-deprivation protocol designed to induce nonglandular gastric ulceration or worsen existing ulcers. 22 During the first week of each 5-week period, the horses were deprived of feed for 24 hours, then fed their normal grass hay ration free choice for 24 hours, until a total of 96 hours of cumulative feed deprivation were achieved. Throughout the protocol, the horses had free access to water, except for 4 hours immediately before gastroscopy.
The treatment (placebo or herbal supplement) was administered after this feed deprivation period, starting on day 7, for a total of  (Table 1). Scores were assigned by 3 evaluators, 2 during the endoscopy and 1 from recorded images, who were blinded to the treatment group.

| Nonglandular gastric ulcer scores
The resulting interrater reliability was excellent for both ulcer number and P = .85, respectively) was not different between horses in the treatment and control groups of the study at the end of the treatment periods, but both scores increased significantly at day 6 after 1 week of feed deprivation for both groups (P < .001). (Figures 1 and 2) However, after the 4 weeks of washout between the 2 study periods, gastric ulceration resolved in all horses. Scores at the start of the second period were not different than the rankings on day 0 of the first study period (ulcer number scores: P = .95; ulcer severity scores: P = .43).  A limitation of this study was that although the intermittent feed deprivation model was successful in inducing nonglandular gastric ulcers, not all horses responded to the protocol. This was not unexpected, based on a previous work that used a similar model. 22 One difference within our study was that the horses were allowed to remain on free choice hay between periods of intermittent fasting, which followed the original protocol that was designed to produce consistent ulceration of the squamous mucosa. 27 It is likely that if we had used a pelleted feed or reduced the hay intake to <2% of body weight, the squamous ulcers after feed deprivation could have been more severe.

| Gastric fluid pH
Gastric fluid pH was variable and was not significantly different between WLS-treated horses and the untreated controls. Although a single gastric fluid sample is not as likely to note changes in pH that occur over a 24-hour period, continuous monitoring might have been able to better identify change in gastric fluid pH. 28 sampling protocol was uniform with regards to fluid depth location and obtained in fasted horses where the pH should have been at its nadir due to the absence of buffering feed and lower levels of saliva production. 30 An additional limitation of the pH measurements in this study was the use of alpha-2 agonists and butorphanol for the gastroscopy procedure, which could have resulted in reduced gastrointestinal motility. 31 Although these medications are known to reset the migrating motility complex, the effects of butorphanol are believed to be minimal, whereas those of detomidine can markedly reduce duodenal motility. 32,33 The time frame from sedation to sample collection was minimized (<10 minutes), in an effort to reduce the influence of duodenal stasis on gastric contents.
The results of this study indicate that the Chinese herbal WLS was unsuccessful in reducing the severity of experimentally induced, nonglandular gastric ulcers in horses exposed to intermittent feeding compared to a placebo control. It is possible that the model used to produce gastric ulceration in these horses was not severe enough to produce a significant difference between the groups on a free choice hay diet by the end of the study period, leading to a type II error. It is also likely that the diet could have contributed to the improvement in gastric ulcer scores in both groups, as a hay diet is known to increase saliva production that can buffer stomach contents and reduce acid exposure. 34 Further investigations could involve a feed deprivation challenge at the end of the administration period, to determine the herbal supplement's effect in the face of a feed deprivation challenge.
Additional evaluation is needed in a larger number of horses, and in clinical cases, where feeding protocols and husbandry practices are not as tightly controlled.