The effects of feeding sodium chloride pellets on the gastric mucosa, acid-base, and mineral status in exercising horses

1 INTRODUCTION

Gastric ulcerations are common in horses. The prevalence reaches 100% in sports horses, especially in racehorses.^{1, 2} In nonracing sports horses, the prevalence of gastric ulceration ranges from 11%³ to 83.5%.⁴ The suspected risk factors for development of equine gastric ulcer syndrome (EGUS) include training, stress, use of non-steroidal anti-inflammatory drugs (NSAIDs), transportation, lack of turnout, and no contact with other horses.^{5, 6} Nutritional risk factors for EGUS, mostly equine squamous gastric disease (ESGD), include high starch intake (>2 g/kg body weight [BW]/day), low forage intake, >6 hours fasting between forage feeding, and intermittent water access.^{5, 7} Electrolyte supplementation also may increase the risk of gastric lesion formation in horses. A previous study showed that PO administration of a hypertonic solution every hour for 8 hours increased the mean number and severity of gastric ulcers in the squamous gastric region in horses.⁸

Strong clinical evidence supports an association between higher salt and salted food intake and the development of gastritis and stomach cancer in humans, as reviewed previously.⁹ In laboratory rodents, high salt intake has been shown to induce gastritis and gastric epithelial proliferation.^{10, 11}

Several studies indicate the positive effects of electrolyte supplementation on rehydration and compensation for electrolyte losses in sport horses.^12-15 However, little is known about the impact of electrolyte intake on the gastric mucosa in horses except for a single study.⁸ In that study, salt was given in the form of a concentrated paste PO hourly for 8 hours to simulate conditions during an endurance race, which might be the reason for the increased gastric mucosal lesions. Differences in application form and frequency of a palatable salt pellet given together with concentrated feed and divided into several meals throughout the day might be a safer way to compensate for sweat losses in the exercising horse.

Our aim was to investigate the effects of NaCl supplementation as a pellet on the gastric mucosa, as well as acid-base and mineral status in exercising horses. We hypothesized that PO NaCl pellets at a dosage adequate to replace the salt loss from 10 L of sweat would neither cause nor aggravate existing gastric mucosal damage and would not have negative effects on acid-base and mineral status in exercising Warmblood horses.

2 MATERIALS AND METHODS

2.4 Study protocol

The study was performed using a placebo-controlled crossover design with 2 feeding periods of 19 days and a washout period of 14 days between the feeding periods (Figure 1). Assignment to the groups (NaCl vs placebo) was done randomly before the first feeding period with half of the horses receiving the NaCl pellets and the other half receiving the placebo pellets. In the second feeding period, the groups were switched to the other treatment. Before and after each feeding period, blood was collected from the jugular vein and urine was collected. The horses were then fasted 12 hours overnight and gastroscopy was performed on the next day.

2.4.1 Treatment

The treatment period lasted 19 days. In the beginning, the horses were gradually introduced to the treatment. From day 2 (T₂) to day 6 (T₆) horses were fed half of the calculated dose (300 g NaCl pellets or 300 g of the placebo pellets) divided into 2 meals of 150 g, fed in the morning and evening, respectively. From T₇ to T₂₀, all horses were fed 600 g/day of the NaCl or placebo pellet divided into 2 meals of 300 g fed in the morning and evening, respectively. The NaCl or placebo pellets were mixed with oats and compound feed.

2.4.2 Gastroscopy

On days 0 (T₀) and 21 (T₂₁), gastroscopy was performed using a 3-meter flexible gastroscope (KARL STORZ, Tuttlingen, Germany). The horses were fasted by muzzling for 12 hours and had access to water until 2 hours before the procedure. The horses were sedated with detomidine (Cepesedan; CP Pharma Handlungsgesellschaft mbH, Burgdorf, Germany) at a dosage of 0.014-0.026 mg/kg BW. The gastric mucosa was evaluated by 2 investigators blinded to the group assignment of each horse. The regions examined included the dorsal squamous fundus, the lesser and greater curvatures of the squamous region, the lesser and greater curvatures of the glandular region, the pyloric antrum, and the pylorus. Each squamous region was scored according to the grading system recommended by the European College of Equine Internal Medicine Consensus Statement (Table 1). A modified scoring system was used for the glandular region (Table 2). In case of an initial score of 4 in 1 or both stomach regions, the horse was excluded from the study.

TABLE 1. European College of Equine Internal Medicine grading system for equine squamous gastric disease (ESGD).⁵

Grade	Characteristics
0	Epithelium intact and no appearance of hyperkeratosis
1	Mucosa intact, but areas of hyperkeratosis
2	Small, single, or multifocal lesions
3	Large single or extensive superficial lesions
4	Extensive lesions with areas of apparent deep ulceration

TABLE 2. Modified grading system for equine glandular gastric disease (EGGD).¹⁷

Grade	Characteristics
0	Epithelium intact and no appearance of hyperemia (reddening) or fibrinosupperative areas
1	Intact flat mucosa, but with small single or multifocal areas of reddening
2	Raised mucosa with large single or multifocal areas of reddening or fibrinosupperative areas, no signs of bleeding
3	Raised mucosa with hemorrhagic and fibrinosupperative areas
4	Ridged or depressed mucosa with severe signs of bleeding or with large and distinct fibrinosupperative areas

2.4.3 Water intake

All horses had access to a water bucket with a scale. The bucket was cleaned and filled 4 times per day. Water intake was measured from T₁ to T₄ and from T₁₈ to T₂₀ by reading the scale before refilling the bucket. The filled water buckets were placed in the boxes 1 day before starting the measurements to acclimate the horses. The minimum and maximum ambient temperatures were recorded on each day of water intake measurement. The horses had access to automatic water dispensers between intake measurement days.

2.4.4 Blood and urine sampling

On T₋₁ and T₂₀, blood was taken from the jugular vein via an 18 G needle (SARSTEDT; Nümbrecht, Germany) and collected in serum analysis tubes (Monovetten; SARSTEDT; Nümbrecht, Germany). Syringes containing electrolyte-compensated dried heparin (Pico; RADIOMETER; Krefeld, Germany) were used for the blood gas analyses.

Blood used in the serum analyses was centrifuged (3000 rpm, 10 min) after 30 minutes clotting and the sera were stored at −20°C until analysis. Urine was taken on T₋₁ and T₂₀ by collecting midstream voided urine in a container. Before each urine sample collection, the horses were confined without water on a paddock for 2 hours and urine was collected by spontaneous urination immediately after horses returned to the boxes. The urine samples were stored at −20°C until analysis. Urine could not be collected from all horses. Consequently, less data for urine were available than for blood and gastric variables (Table 6).

3 RESULTS

All included animals were clinically healthy based on results of physical examinations conducted at the start and end of the study. Body weights did not change significantly during the study (NaCl: T₋₁:547 ± 29 kg; T₂₀: 559 ± 31 kg; placebo: T₋₁:549 ± 31 kg; T₂₀: 559 ± 32 kg). None of the horses refused the pellets.

4 DISCUSSION

Our study was designed to examine the effects of a NaCl supplementation on the gastric mucosa in horses, dosed to replace the loss of NaCl in 10 L sweat. For this purpose, 15 3-year-old Warmblood horses received a pelleted supplement along with their compounded feed daily for 19 days. Before the NaCl treatment, the prevalence of gastric mucosal lesions was 83.33%. The NaCl treatment did not have any clinically relevant adverse effects on the gastric mucosa.

The NaCl pellets generally were well accepted and none were left over. Pelleted NaCl feeding is a practical form of electrolyte supplementation and a superior alternative to table salt and salt blocks, because voluntary intake of the latter is usually low.¹⁸

Interestingly, the prevalence of gastric mucosal lesions with a total overall score ≥1 was already 83.33% in both squamous and glandular regions at the start of the study, but with relatively low scores (Table 3). The median lesion scores were 2 for ESGD and 1 for EGGD, respectively. The high total prevalence observed accorded with results reported in earlier studies. In 1 study, 83.5% of nonracing Danish pleasure horses with unequal workloads (pasture only, light or hard work) had EGUS scores ≥1.⁴ Nevertheless, studies on the prevalence of EGUS in nonracing horses have had equivocal results. One study reported a prevalence of nonglandular gastric lesions (ESGD) of only 11% in riding horses with different riding disciplines.³ Horses of different breeds and in the age range of 2 to 23 years were included in that study and the prevalence of nonglandular gastric lesions was higher in horses aged 2 to 6 years (21.4%) compared to the overall prevalence. In our study, all horses were only 3 years old, which might have led to relatively higher prevalences. However, 2-year-old Thoroughbred racehorses in training had the largest increase in mean maximum lesion scores compared with those aged 3 years and older.¹ In a necropsy study on 3715 horses of various breeds aged >1 year, the number of gastric lesions decreased with age especially in Standardbred horses.¹⁹

Several other factors may have had an impact on the prevalences found in our study. The daily concentrate intake was >4 kg and oats comprised the main feed (2850 g/day). Different studies suggested that the amount of grain fed is an important contributing factor in ESGD development.^{7, 20, 21} Furthermore, the horses in our study only had paddock access 2 days per week and for 2 hours each time. This condition also might have influenced the observed prevalence of ulcerations in both squamous and glandular gastric regions. However, other studies reported equivocal effects of paddock turnout on ulcer development.^{22, 23} One study suggested that time periods above 6 hours between forage feedings could increase lesion scores in the squamous region of the stomach to a greater extent than intervals <6 hours between forage feedings.⁷ In the former case, there is less buffering effect of salivary bicarbonate on stomach acid. The horses in our study usually were subjected to intervals of >10 hours between hay feedings. This management strategy might partially account for the relatively high prevalence of lesions in the squamous regions of the stomachs of the horses. In addition, the study population consisted of stallions, which are predisposed to the development of gastric ulcerations as reported previously.¹⁹

Electrolyte supplementation provides beneficial effects in maintaining physiological acid-base and water balance in exercising horses. It increased water consumption and decreased net electrolyte loss in exercising horses and in a furosemide-induced dehydration model in horses.^12-15 It also promoted muscle glycogen resynthesis in horses after exercise.²⁴ Most feed rations do not meet the Na and Cl requirements of sweating horses.¹⁶ Hence, safe electrolyte supplementation strategies must be determined because voluntary NaCl intake is quite variable in horses.¹⁸

Consistent with our hypothesis, no significant changes were observed in the gastric mucosa regions of the horses fed NaCl pellets for 19 days consecutively. In contrast, a previous study reported a significant increase in the median lesion scores for the gastric squamous regions in horses given 56.7 g hypertonic electrolyte solution (11 056 mg Na; 23 772 mg Cl; 7314 mg K; 1508 mg Ca; and 306 mg Mg) PO q1h for 8 hours consecutively.⁸ The hypertonic solution administration method and quantity used in that study were designed to simulate conditions in endurance racing. Hourly electrolyte paste supplementation is commonly performed during competitions to compensate for electrolyte loss and manage rehydration. However, this high frequency of electrolyte administration over a short time period might have exacerbated the effects of NaCl on the gastric mucosa. In our study, however, the horses received NaCl pellets (30 g Na plus 39 g Cl/horse/day) only twice daily.

There is strong clinical evidence for an association between higher salt and salted food intake and the development of gastritis and stomach cancer in humans, as previously reviewed.⁹ In laboratory rodents, high salt intake has been shown to induce gastritis and gastric epithelial proliferation.^{10, 11} In comparison, studies in species other than horses used an estimate of 7.5% to 12.5% NaCl in the diet, either as a single application or as total daily intake. Our study used a pellet 11.5% NaCl, which, together with the amount in the basal diet, translated to approximately 1.2% NaCl in the total diet. This amount is substantially less than in studies of other species.

Our findings suggest that NaCl supplementation in the form of pellets had no negative effect on the gastric mucosa when fed for 19 consecutive days. With respect to the gastric mucosa, it is a safe, convenient, and effective way to replace electrolyte losses in horses. We speculate that even a longer supplementation period of NaCl pellets would not harm the gastric mucosa. The additional NaCl supplementation necessitates a free-choice supply of water to ensure voluntary water intake by horses.

The positive effects of NaCl supplementation on rehydration and compensation for electrolyte loss have been demonstrated repeatedly.^{12-14, 24} In contrast, another study reported mild metabolic acidosis in horses receiving NaCl supplementation.²⁵ The authors supplemented moderately exercising horses with 100 g NaCl daily for 21 days. After 15 days, however, blood pH significantly decreased from 7.414 to 7.406. As a rule, however, a blood pH of 7.406 is not considered acidotic (normal pH range for horses: 7.38-7.44).²⁶ In our study, no significant changes in blood pH were observed after 19 days of NaCl treatment at 30 g Na/day plus 39 g Cl/day.

Another difference between our study and the previous study²⁵ was the amount of forage fed. In the previous study, a low-forage diet (1 kg/100 kg BW/day) was fed with the expectation that large quantities of hay could compensate for the impact of acidifying agents on blood pH.²⁷ In our study, the hay intake rate was 1.5 kg/100 kg BW/day. Nonetheless, it is generally recommended to provide forage at ≥1.5 kg/100 kg BW/day based on dry matter intake.²⁸ Low-forage diets should be avoided because they may cause behavioral, gastrointestinal, and muscular disorders in horses.^{5, 28}

In our study, serum Na⁺ concentration significantly decreased after both placebo and NaCl pellets. However, no significant difference was found between treatments in terms of the relative impact on serum Na⁺ concentration. In contrast, earlier studies showed that different levels of Na intake did not alter serum Na⁺ concentration in horses.^{27, 29} In our study, the cause of the decreases in serum Na⁺ concentrations in both treatment groups is not fully understood, but the moderate decrease in serum Na⁺ concentration did not appear to have any clinically relevant impact.

Prior research on water consumption by horses supplemented with salt showed equivocal results. A linear correlation between NaCl and water intake has been reported.²⁹ A 2-fold increase in Na intake (from 50 mg/kg BW to 100 mg/kg BW) in horses increased water intake by 53% and urinary output by 47%.³⁰

On the other hand, 2 other studies reported no correlation between NaCl intake and water consumption.^{18, 25} In our study, NaCl pellets likewise did not increase water intake in the horses. However, water intake measurements and outcome were limited because of the switch between automatic water dispensers and water buckets during the feeding periods. It has been shown that horses prefer water from buckets rather than from automated dispensers.^31-33 Here, water provision from a bucket during the feeding periods likely increased water intake despite salt intake. Also, the baseline daily water intake (T₀) was a single time point, which may not have been an accurate measure, compared with the post-treatment measures with an average of 3 days (T_1-3 and T_18-20). Nevertheless, NaCl treatment did not have a significant effect on water intake. We speculate that the effect of NaCl supplementation on water intake was weaker than the effect of water provision by buckets.

The decrease in USG also may reflect changes in water intake associated with water provision by buckets. Hence, the latter might have masked the effects of NaCl treatment. In our study, USG decreased over time in both the NaCl and placebo treated horses but was not influenced by treatment. In contrast, USG significantly decreased from 1.034 to 1.027 after 15 days of feeding 100 g NaCl.²⁵ Another study reported a significant decrease in urinary osmolality and significant increase in urine production after PO administration of 0.5 g NaCl/kg BW after dehydration induced by furosemide administration and withholding water for 12 hours.¹⁵ Additionally, water intake significantly increased during the rehydration phase of horses receiving NaCl compared with those recieving a placebo.¹⁵

Another limitation of our study was that the ambient temperature differed by approximately 10°C between trials and might have masked the effects of NaCl treatment on water intake. To be able to detect an effect of NaCl on water intake in future research, differences in ambient temperature and changes between types of water provision must be avoided. Air-conditioned stabling together with automatic water dispensers with a water meter would be ideal to prevent such limitations.

Urine could not be spontaneously collected from all horses in the study. Nevertheless, urinary Na significantly increased with Na intake in response to the NaCl treatment. Similar findings were reported previously.^{18, 25, 27} Urinary Cl^- excretion also was significantly higher after NaCl supplementation in the previous studies.^{18, 25, 27} In our study, although urinary Na⁺ concentrations changed in response to the treatments, urinary Cl⁻ concentrations did not. The reason for this discrepancy is unknown despite the fact that the kidney is the principal route of both Na⁺ and Cl⁻ excretion. Because of the response of the kidney to a higher Na intake, the Na⁺ concentration in urine might be a good tool to assess the adequacy of the dietary intake of Na in performance horses. In our study, 31 g of Na was supplemented to compensate 10 L of sweat. The replacement of Na required for 10 L sweat losses was estimated for practical reasons to investigate the effect of NaCl supplementation by a pellet on the gastric mucosa. Therefore, the high amount of Na⁺ excretion after feeding the NaCl pellet could be a result of oversupply because horses in our study may have had lower sweat losses than estimated.

Serum and urinary Ca²⁺ concentrations decreased significantly in both treatment groups over time. However, treatment did not have a significant effect on Ca²⁺ concentrations. We speculate that the increases in water intake in both feeding groups lowered serum, blood and urinary Ca²⁺ concentrations. Similar results in Ca²⁺ concentrations after NaCl supplementation have been reported previously.¹⁸ On the other hand, increases in urinary Ca²⁺ concentration after NaCl feeding also have been reported.²⁵ The authors associated the observed increase in urinary Ca²⁺ excretion with changes in bone remodeling in response to metabolic acidification induced by NaCl intake. In our study, blood pH remained constant. Therefore, changes in bone metabolism by NaCl feeding remained possible but seemed unlikely.

5 CONCLUSIONS

We demonstrated that palatable NaCl pellets, fed together with compound feed and divided into 2 equal meals, effectively and safely compensated for electrolyte losses in sweating horses. In agreement with our hypothesis, NaCl supplementation to compensate for 10 L sweat losses (50 g Na/kg pellet and 65 g Cl/kg pellet) did not negatively affect the gastric mucosa, acid-base balance, and mineral status of exercising horses. In summary, we recommend supplementation using a palatable NaCl pellet in horses to compensate for electrolyte losses in sweat.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION

Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Approved by the Ethics Committee for Animal Rights Protection of the LAVES (No. TVV 33.19-42 502-04-20/3572) in accordance with German legislation for animal rights and welfare.

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.