Search
Author
Title
Vol.
Issue
Year
1st Page

Journal of Animal Science - Animal Production

The effect of a limit-fed diet and slow-feed hay nets on morphometric measurements and postprandial metabolite and hormone patterns in adult horses1

 

This article in JAS

  1. Vol. 93 No. 8, p. 4144-4152
     
    Received: Mar 27, 2015
    Accepted: May 18, 2015
    Published: August 3, 2015


    2 Corresponding author(s): krishona@umn.edu
 View
 Download
 Share

doi:10.2527/jas.2015-9150
  1. E. C. Glunk*,
  2. M. R. Hathaway,
  3. A. M. Grev,
  4. E. D. Lamprecht,
  5. M. C. Maher§ and
  6. K. L. Martinson 2
  1. * Department of Animal and Range Science, Montana State University, Bozeman 59717
     Department of Animal Science, University of Minnesota, Saint Paul 55108
     Cargill Animal Nutrition, Elk River, MN 55330
    § College of Veterinary Medicine, University of Minnesota, Saint Paul 55108

Abstract

Modern horse management systems tend to limit a horse’s opportunity to forage, rely on meal feeding, and may contribute to the increase in equine obesity. The use of slow-feed hay nets represents an opportunity to extend foraging time while feeding a restricted diet. The objectives of this study were to determine if limit feeding combined with a slow-feed hay net would affect morphometric measurements and postprandial metabolite and hormone patterns in overweight adult horses. Eight adult Quarter horses (BW 563 kg ± 4.6 kg; BCS 7.2 ± 0.3) were used in a randomized complete block design, with 4 horses assigned to feeding hay off the stall floor (FLOOR) and 4 horses assigned to feeding from a slow-feed hay net (NET). Horses were fed in individual stalls at 1% BW each day, split evenly between 2 meals at 0700 and 1600 h. Body weight, BCS, neck and girth circumference, cresty neck score, and ultrasound measurements of average rump fat, longissimus dorsi (LD) depth, and LD thickness were taken on d 0, 14, and 28. Three 24-h blood samplings were conducted on d 0, 14, and 28 and were analyzed for glucose, insulin, cortisol, and leptin concentrations. Samplings occurred every 30 min for 3 h postfeeding, with hourly samples occurring between feedings. Horses feeding from the FLOOR took less time to consume their hay meal compared with horses feeding from the NET (P < 0.001). All horses lost weight over the 28-d period (P < 0.0001); however, no difference was observed between treatments. There was no difference in BCS, neck and girth circumference, cresty neck score, rump fat, or LD depth between days or treatments (P ≥ 0.25). There was an effect of day on LD thickness in horses feeding from the NET. Longissimus dorsi thickness was lower on d 28 compared with that on d 0 (P = 0.0257). Only time to peak insulin and peak cortisol were affected by treatment (P ≤ 0.037), with horses feeding from the NET having lower values than horses feeding from the FLOOR. Average glucose, insulin, cortisol, and leptin were affected by day (P ≤ 0.0102). Glucose and insulin values increased, whereas cortisol and leptin levels decreased throughout the 28-d study. The use of a slow-feed hay net coupled with a limit-fed diet appears to be an effective method for decreasing BW and maintaining more homeostatic levels of postprandial metabolites and hormones when feeding overweight adult horses.



INTRODUCTION

Horses are hindgut fermenters and have evolved to consume small, frequent forage meals throughout the day (Janis, 1976). However, modern management strategies including meal feeding and increased duration of stalling have led to the decreased opportunity for horses to forage (Henderson, 2007). This change in foraging behavior has led to alterations in hindgut fermentation and metabolic patterns, contributing to colic, laminitis, equine metabolic syndrome, insulin resistance, and obesity (Hudson et al., 2001; Hoffman et al., 2003; Frank et al., 2006, 2010).

Changes in postprandial metabolite and hormone concentrations highlight the impact of different feeding and management systems. Changes in blood glucose and insulin are helpful in estimating the digestibility and absorption of a meal, with elevated postfeeding values often correlated with diets that produce a more acidic pH level in the hindgut (Bailey et al., 2004; Vervuert et al., 2009). Stull and Rodiek (1988) found that different diets did not affect cortisol levels, but stress has been shown to impact cortisol levels (Fazio et al., 2008). Leptin, a hormone secreted by adipose tissue and an indicator of energy status, has been found to be positively correlated with fat mass in mammals (Considine et al., 1996; Takahashi et al., 1996; Buff et al., 2002; Kearns et al., 2006).

In an effort to lessen the postprandial metabolite and hormone responses to meal feeding, researchers have attempted to increase total time of consumption of feedstuffs by decreasing intake rate (Glunk et al., 2014a) and rate of passage through the digestive tract (Glunk et al., 2013; Kutzner-Mulligan et al., 2013). The objectives of this study were to determine if a limit-fed diet combined with the use of a slow-feed hay net would affect morphometric measurements and postprandial metabolite and hormone patterns in overweight adult horses.


MATERIALS AND METHODS

Animals, Management, and Experimental Design

All experimental procedures were conducted according to those approved by the University of Minnesota Committee on Animal Use and Care. Eight client-owned adult Quarter horses (5 mares and 3 geldings) with a BW of 563 kg (SE ± 4.6 kg) and a BCS (Henneke et al., 1983) of 7.2 (SE ± 0.3) were used in a completely randomized block design for a period of 28 d. Horses had no known metabolic conditions, other than being overweight. Upon arrival, all horses were quarantined for 3 d followed by a 7-d acclimation period in individual stalls (3.0 × 3.7 m) at the University of Minnesota Large Animal Hospital. During this time, horses were fed a cool-season grass hay on the stall floor at 2.5% BW on an as-fed basis split evenly between 2 meals at 0700 and 1600 h. Stall floors were constructed of a cement base covered with rubber mats. Each stall was bedded with wood shavings, and hay was placed directly on the rubber mats in an area clear of wood shavings. After the quarantine and acclimation period, horses were blocked by BW, BCS, and gender and assigned to 1 of 2 treatments for 28 d. Treatments consisted of 4 horses consuming hay off of the stall floor (FLOOR) and the remaining 4 horses consuming hay from a slow-feed hay net (NET). The slow-feed hay net was made from webbed UV-treated Dupont fiber with 3.2-cm diamond-shaped openings (Chinch Chix LLC, North Branch, MN).

On d 0, horses were fed grass hay at 2.5% BW to obtain baseline morphometric and blood metabolite and hormone measurements. Beginning on d 1 of the data collection period, horses were fed grass hay at approximately 60% of their maintenance DE requirements (NRC, 2007), which was equivalent to 1% BW each day split evenly between 2 meals at 0700 and 1600 h. Horses had ad libitum access to water and were fed a ration balancer (ProAdvantage Grass Balancer, Progressive Nutrition, Brookville, OH) at 0.001% BW at 0700 h each day immediately before the hay meal to ensure all vitamin and mineral requirements were met for adult horses at maintenance (NRC, 2007). Horses were hand walked twice daily for 30 min immediately before receiving their meal.

Hay Analysis, Time to Consumption, and DMI

Before the start of the experiment, multiple small-square bales of the grass hay were cored using a 2 by 51 cm core sampler (Penn State Forage Sampler, University Park, PA) and were combined to determine forage nutritive value. Hay was from the same cutting and field. To confirm nutrient content, 1 representative sample of the ration balancer was collected. All units of the ration balancer were from the same manufacturing run. All samples were analyzed for nutritive content by a commercial testing laboratory (Equi-Analytical, Ithaca, NY) using the following methods. Dry matter was determined by placing samples in a 60°C forced-air oven for 24 h (method 991.01; AOAC Int., 2010). Crude protein was calculated as the percentage of N multiplied by 6.25 (method 990.03; AOAC Int., 2010). Neutral detergent fiber and ADF were nonsequentially measured using filter bag techniques (Ankom Technology, 2013a,b,c). Starch, water-soluble carbohydrates, and ethanol-soluble carbohydrates were measured using techniques described by Hall et al. (1999). Mineral concentrations were determined (Thermo Jarrell Ash IRIS Advantage HX Inductively Coupled Plasma Radial Spectrometer, Thermo Instrument Systems Inc., Waltham, MA) after microwave digestion (Microwave Accelerated Reaction System, CEM, Mathews, NC). Equine DE was calculated using an equation developed by Pagan (1998).

Total time to consumption of the hay meal was measured by trained personnel on d 14 and 28 using a stopwatch. The stopwatch was started when the horses began eating their hay meal and was stopped when the horses consumed their entire hay meal (Glunk et al., 2014a). Dry matter intake rate (DMIR) was determined by dividing the total amount of hay consumed (kg) by the total time to consumption (h).

Morphometric and Ultrasound Measurements

On d 0, 14, and 28, horse BCS (Henneke et al., 1983), morphometric measurements, and ultrasound measurements of rump fat, longissimus dorsi (LD) muscle depth, and LD thickness were taken. Body weight was measured using a calibrated livestock scale (Fairbanks Scales, Kansas City, MO). Morphometric measurements were taken by trained personnel and included neck circumference located halfway between the poll and withers (Carter et al., 2009), girth circumference at the base of the mane hairs (Martinson et al., 2014), and a cresty neck score on a scale of 0 to 5 (Carter et al., 2009). To measure subcutaneous fat and muscle thickness, LD depth and LD thickness, respectively, were measured (D’Angelis et al., 2007). The LD muscle depth and thickness were taken 5 cm lateral from the spinous processes between the 12th and 13th ribs (D’Angelis et al., 2007). Using ultrasound, rump fat was measured 5 cm lateral from the midline on both the left and right sides of the rump and averaged. On the basis of rump fat, the following equation was used to determine extractable fat (Westervelt et al., 1976):where Y = percent extractable fat and X = cm of average rump fat.

Blood Sampling and Laboratory Analysis

Jugular catheters were placed approximately 2 h before blood sampling on d 0 (baseline), 14, and 28. Blood samples were drawn 1 h before feeding (baseline), immediately after the morning meal (0700 h) was fed, and every 30 min for the next 3 h. Sampling continued hourly until the evening feeding at 1600 h. The procedure was then repeated after the evening meal and stopped at 0600 h the following day (n = 30). Approximately 10 mL of blood were drawn from each horse at every sampling using a 20-G syringe. Catheter lines were flushed with 10 mL of saline, followed by 10 mL of heparinized saline. Blood samples were then aliquotted into sterile Vacutainer tubes containing heparin (BD Diagnostics, Franklin Lakes, NJ) and placed on ice for transport to the laboratory. Upon arrival in the laboratory, samples were immediately centrifuged at 4°C for 15 min at 2,000 × g, and the plasma was collected and frozen immediately at −20°C until the laboratory analysis could be completed.

Glucose concentrations were determined using a spectrophotometric glucose assay (Coat-a-Count, Diagnostic Products, Los Angeles, CA) in duplicate. The interassay and intra-assay coefficients of variation were 5% and 3%, respectively. Serum insulin concentrations were determined using a radioimmunoassay (Coat-a-Count PITKIN-9, Diagnostic Products, Los Angeles, CA) that had been validated for equine plasma (Reimers et al., 1982). All samples were run in duplicate. The interassay and intra-assay coefficients of variation were 1% and 11%, respectively. Plasma cortisol concentrations were determined using radioimmunoassay (Coat-a-Count PITKCO-10, Diagnostic Products) in duplicate. The interassay and intra-assay coefficients of variation were 1% and 9%, respectively. Plasma leptin was measured using a multispecies radioimmunoassay (Millipore Research, Billerica, MA) previously validated for equine plasma (McManus and Fitzgerald, 2000; Cebulj-Kadunc and Cestnic, 2005). The assay used I25I-labeled human leptin and multispecies leptin antiserum. The interassay and intra-assay coefficients of variation were 0.7% and 6%, respectively. All tubes were counted with a Packard Cobra II Gamma Counter (Packard Biosciences, Boston, MA).

Statistical Analysis

Morphometric measurements, BCS, and ultrasound measurements were analyzed using PROC MIXED of SAS (version 9.3; SAS Inst., Cary, NC). The model included day, treatment, gender, and the day × treatment interaction. Area under the curve (AUC) for glucose, insulin, and cortisol were analyzed using the trapezoidal method. Area under the curve was calculated for a period of 8 h after each feeding, the approximate amount of time required for horses to return to baseline (Glade et al., 1984).

Total time to consumption and average, time to peak, and peak values for insulin, glucose, cortisol, and leptin were analyzed using PROC MIXED of SAS. The model included day, treatment, and day × treatment. To confirm effects of treatments and feeding time (morning vs. afternoon), results from d 14 and 28 were analyzed. These data were analyzed using PROC MIXED of SAS. The model included day, feeding, and day × feeding.

All models included day as a repeated effect. Data were checked for normalcy using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Results were considered significant at P ≤ 0.05.


RESULTS AND DISCUSSION

Forage Nutritive Value, Time to Consumption, and DMIR

Nutritive values for the cool-season grass hay and ration balancer are listed in Table 1. When compared to a national hay nutritive value database (Common Feed Profiles, 2011), the hay was within or near normal ranges for all nutrients tested for cool-season grass hay.


View Full Table | Close Full ViewTable 1.

Nutritive value of cool-season grass hay and ration balancer fed to overweight adult horses

 
Nutrient1 Hay Ration balancer2
DM, % 92 89
CP, % DM 12 39
ADF, % DM 38 9
NDF, % DM 63 18
Starch, % DM 1 5
WSC, % DM 11 10
ESC, % DM 6 8
Ca, % DM 0.47 2.10
P, % DM 0.31 1.09
Equine DE, Mcal/kg 2.0 1.5
1WSC, water-soluble carbohydrates; ESC, ethanol-soluble carbohydrates.
2ProAdvantage Grass Balancer, Progressive Nutrition, Brookville, OH.

The horses fed from the FLOOR took less time to consume their hay meal than horses fed from the NET (P < 0.001). The mean total times to consumption for horses fed from the FLOOR and NET were 120 min (SE ± 13 min) and 193 min (SE ± 13 min), respectively. No differences were observed between morning and evening feedings (P = 0.5895). Dry matter intake rate was affected by treatment (P = 0.0012). Mean DMIR was 1.42 and 0.87 kg/h (SE ± 0.1 kg/h) for horses fed from the FLOOR and NET, respectively.

The differences in DMIR between horses fed from the FLOOR and NET were similar to results found by Glunk et al. (2014a). They determined that DMIR from horses on a non-limit-fed diet fed from the stall floor was 1.49 kg/h, compared with the rate from the same slow-feed hay net, which resulted in 0.88 kg/h. The DMIR remained similar, even though horses in the previous study were fed hay at 2.0% BW and not a limit-fed diet. The current study confirms the effectiveness of slow-feed hay nets at extending total time to consumption and slowing DMIR. These results are useful when managing stalled horses and when feeding a limit-fed diet. Increasing the time horses spend foraging each day promotes gut health and hindgut fermentation (Siciliano and Schmitt, 2012) and has been shown to reduce stereotypical behaviors (McBride and Hemmings, 2009) and the incidence of colic (Hudson et al., 2001). Future research should explore the use of slow-feed hay nets in combination with multiple hay meals (vs. only 2 hay meals) in horses on a limit-fed diet and resulting impacts on DMIR, horse behavior, and health parameters.

Previous research has shown that horses can acclimate and adapt to changes in length of feeding time. Longland et al. (2011) observed that ponies were able to increase their daily DMIR over a period of 6 wk, whereas Glunk et al. (2013) determined that horses were able to increase their DMIR as the grazing period was reduced from 24 to 3 h. In the current study, there was an effect of day. Horses took less time to consume their hay meal on d 28 than on d 14 (P = 0.047). These results confirm horses can acclimate to different feeding and management strategies over time.

BCS, Morphometric, and Ultrasound Measurements

All horses lost BW between d 0 and 28 (P < 0.001); however, BW loss was not affected by treatment (P = 0.326; Table 2). On average, horses fed on the FLOOR lost 32 kg (±7.4 kg), whereas horses fed on the NET lost 40 kg (±7.4 kg). Body weight on d 14 was similar to BW on d 0 and 28 for both treatments (P = 0.691). There was no difference in horses’ BCS, girth circumference, neck circumference, cresty neck score, average rump fat, extractable fat, or LD depth between d 0 and 28 (P ≥ 0.42) or between treatments (P ≥ 0.13). There was an effect of day on LD thickness for horses feeding from the NET. Longissimus dorsi thickness on d 28 was less than LD thickness on d 0 (P = 0.0257). No difference in LD thickness was observed from horses fed from the FLOOR between d 0 and 28. The absence of differences in LD depth from d 0 to 28 and the limited differences in morphometric measurements suggest that BW loss was likely due to a combination of factors. It has been well established that a sustained, moderate-energy deficient diet results in the generalized loss of both fat and lean body mass (Pasiakos et al., 2015; Rotella and Dicembrini, 2015), which, when coupled with reduced gut fill and loss of visceral organ weight associated with a limit-fed diet (Connysson et al., 2010; Webb and Weaver 1979), could explain the BW changes observed in this study. Future research should investigate components contributing to BW loss when horses are exposed to a limit-fed diet.


View Full Table | Close Full ViewTable 2.

Morphometric measurements and ultrasound values (±SE) of horses fed a limit-fed diet from either the stall floor (FLOOR) or a slow-feed hay net (NET) on d 0 and 28

 
Day BW, kg BCS1 Neck circ.,2 cm Girth circ.,2 cm Cresty neck score3 Rump fat, cm Extractable fat,4 % LD thickness,5 cm LD depth,5 cm
Floor 0 569a 7 39 76 1.5 1.6 13 0.61a 4.0
28 537b 7 40 76 1.5 1.8 14 0.51a 3.8
Net 0 561a 7 41 76 2.3 2.3 16 1.44a 4.0
28 521b 7 41 75 2.0 1.7 13 0.44b 4.0
SEM 18 0.4 0.8 0.5 0.4 0.3 0.05 0.2
a,bMeans without a common superscript within a column differ (P ≤ 0.05).
1Body condition score was assessed on a scale of 1 to 9 (Henneke et al., 1983).
2Circ.: Circumference.
3Cresty neck score was assessed on a scale of 0 to 5 (Carter et al., 2009).
4Extractabel fat was estimated on the basis of an equation by Westervelt et al. (1976).
5LD: longissimus dorsi.

Caloric intakes were designed to be at approximately 60% of DE for adult horses at maintenance to achieve a reduction in 1 BCS unit in 1 mo (NRC, 2007). Recently, Martinson et al. (2014) found that the differences between each unit of BCS averaged 15, 10, and 17 kg for Arabians, ponies, and stock horses, respectively. In the current study, horses lost 32 to 40 kg of BW in 28 d; however, BCS did not change. Dugdale et al. (2010) also observed BW losses in ponies without observing a change in BCS. McGowan et al. (2013) did observe a decrease in BCS when horses were fed a limit-fed diet for 6 wk. However, mean BCS was reduced by only 0.4 (7.6 to 7.2). One possible reason for the inability to detect a change in BCS is the system itself. The BCS systems evaluates adipose tissue in 6 areas, including the ribs, behind the shoulder, along the neck and withers, in the crease of the back, and tailhead (Henneke et al., 1983). Most of the horses in the current study had noticeable visual losses of adipose tissue in their lower abdominal area, a region not considered when assessing horse BCS. This further supports the concept that BW loss observed in this study likely included muscle mass loss, reduced gut fill, and loss of visceral organ weight.

Other measurements of adipose tissue include girth and neck circumference, cresty neck score, rump fat, and LD depth. None of these measurements changed over the 28-d period. This is similar to the results of Dugdale et al. (2010), who found that neither rump fat depth nor neck circumference decreased with decreasing BW when horses were fed hay at 1% BW. However, Dugdale et al. (2010) did observe a decrease in girth circumference and LD depth with decreasing BW, which was not observed in the current study. It is likely that horses in the current study lost adipose tissue in other regions or that the time frame was not long enough to see significant changes in morphometric measurements. Gordon et al. (2009) found that a significant amount of rump fat was not lost until approximately week 6 of a calorically restricted diet and forced exercise program.

Postprandial Metabolite and Hormone Patterns

Results from d 14 and 28 were similar (P ≥ 0.05) among the postprandial metabolites and hormones measured; therefore, only results from d 0 and 28 will be presented and discussed. Average, AUC, peak, and time to peak (TTP) for glucose insulin and cortisol are shown in Fig. 1 and Table 3.

Figure 1.
Figure 1.

Changes in glucose, insulin, and cortisol over a 24-h sampling period when overweight adult horses were fed a limit-fed diet at 0700 (hour 1) and 1600 h (hour 12) from either the stall floor (FLOOR) or a slow-feed hay net (NET) on d 0 and 28.

 

View Full Table | Close Full ViewTable 3.

Average (Avg) values, area under the curve (AUC), peak, and time to peak (TTP) in minutes for blood glucose, insulin, and cortisol and average leptin values for overweight adult horses fed a limit-fed diet at 0700 h (am) and 1600 h (pm) from either the stall floor (FLOOR) or a slow-feed hay net (NET) on d 0 and 28

 
FLOOR
NET
d 0 d 28 d 0 d 28 P-value
Item am feeding pm feeding am feeding pm feeding am feeding pm feeding am feeding pm feeding Treatment Day
Glucose
    Avg, mg/dL 106y 108 y 120x 113 x 105 y 100 y 115 x 116 x 0.4695 <0.0001
    AUC 4506 6644 5064 7246 4397 5948 4857 6939 0.5552 0.2430
    Peak, mg/dL 120 136 143 124 117 111 128 129 0.16 0.28
    TTP 112 97 187 135 172 112 217 82 0.67 0.08
Insulin
    Avg, IU 12.1y 13.0y 20.6x 10.0x 12.3y 12.7y 18.1x 9.2x 0.8229 0.0218
    AUC 514 785 911 579 500 709 733 506 0.6548 0.3832
    Peak, IU 15y 16y 44x 16x 18y 18y 34x 15x 0.82 0.0277
    TTP 120a 90a 187a 120a 112b 97b 106b 67b 0.037 0.330
Cortisol
    Avg, μg/dL 6.1x 6.7x 5.1y 3.7,y 3.0x 2.6x 2.9y 2.5y 0.0642 0.0002
    AUC 260x 387,x 198,y 217,y 123x 156x 115y 140,y 0.0684 0.0102
    Peak, μg/dL 5.8a 3.3a 8.2a 5.0a 4.3b 3.1b 4.4b 3.9b 0.0207 0.0679
    TTP 82 97 67 142 120 97 112 67 0.9568 0.9389
Leptin
    Avg, ng/mL 2.52x 2.23x 0.85y 1.56y 4.15x 3.82x 1.79y 2.26y 0.4545 <0.0001
a,bWithin a row, means without a common superscript differ by treatment (P ≥ 0.05).
x,yWithin a row, means without a common superscript differ by period (P ≥ 0.05).

Average glucose was affected by day (P < 0.0001). Average glucose values increased from d 0 to 28. A possible explanation for the increased glucose level is an increase in cytokines over time. When overweight horses are placed on a limit-fed diet, FFA are mobilized from lipid stores to meet caloric requirements, which can increase cytokines (Gordon et al., 2009; Samuel et al., 2010; Van de Woestijne et al., 2011) and stimulate gluconeogenesis through increased circulating levels of amino acids resulting from muscle catabolism (Pasini et al., 2003). There was no effect of treatment or day on AUC, peak, or TTP for glucose (P ≥ 0.08). Although Pagan et al. (1999) found an increase over time for TTP glucose when horses were fed a limit-fed diet, the current results agree with previous researchers who found no effect on TTP glucose when horses lost BW (Stull and Rodiek, 1988) or were fed a limit-fed diet (Glade et al., 1984).

Average and peak insulin values were affected by day (P ≤ 0.0218) and feeding (P < 0.001) with values being greater on d 28 than on d 0 during the morning feeding. These results disagree with those of Buff et al. (2002) and Frank et al. (2006), who observed decreased insulin values in ponies after BW loss. Time to peak insulin was affected by treatment (P = 0.037), with horses fed from the FLOOR having a longer TTP compared with horses fed from the NET. This indicates that the increased time to consumption from horses fed from the NET likely caused the differences observed in TTP insulin. Area under the curve and TTP insulin were not affected by day (P ≥ 0.330), and average, AUC, and peak insulin values were not affected by treatment (P ≥ 0.6548).

Only peak cortisol was affected by treatment (P = 0.0207), with horses fed from the FLOOR having greater peaks than horses fed from the NET. Horses fed from the FLOOR took less time to consume their hay meal compared with horses fed from the NET. When finished with the hay meal, horses fed from the FLOOR were still able to see horses feeding from the NET. This likely resulted in elevated stress, which may have led to the increased peak cortisol levels observed in horses fed from the FLOOR. The greater peaks of cortisol observed in horses fed from the FLOOR could also help explain the longer TTP of insulin observed in the same group. It is possible the longer TTP of insulin was a result of cortisol regulating glucose uptake (Hoffis et al., 1970; Stull and Rodiek, 1988; Mårin et al., 1992; Frank et al., 2010). Average and AUC cortisol were affected by day (P ≤ 0.0102). Average and AUC cortisol were greater on d 0 than on d 28. Other researchers have observed a decrease in cortisol over time when horses were subject to a limit-fed diet (Glade et al., 1984; Sticker et al., 1995; Gordon et al., 2009). Sticker et al. (1995) determined that cortisol levels decreased by d 9 when horses were fed a limit-fed diet. The reduction in cortisol between d 0 and 28 confirms that horses can acclimate to different feeding scenarios over time (Longland et al., 2011; Glunk et al., 2013). Although not observed in this study, others have observed a decrease over time in average cortisol contributes to a decrease in average glucose. Cortisol has been identified as an important glucose regulator (Hoffis et al., 1970; Stull and Rodiek, 1988; Mårin et al., 1992; Frank et al., 2010). However, it is likely that many other factors contributed to the elevated glucose levels observed in the current study, including meal feeding and the limit-fed diet.

Some researchers have reported a daily circadian rhythm in cortisol with peaks observed in the morning and valleys observed in the evening (Larsson et al., 1979; Stull and Rodiek, 1988; Irvine and Alexander, 1994). This pattern was not observed in the current study and may be due to meal feeding, the limit-fed diet, or housing the animal indoors with minimal access to natural light. Altered light perception has been found to impact cortisol circadian rhythms in humans and rats (Orth and Island, 1969; Irvine and Alexander, 1994; Scheer and Buijs, 1999).

Average leptin values were affected by day (P < 0.0001) but were not affected by treatment (P = 0.4545). Similar to previous research (Buff et al., 2002; Kearns et al., 2006; Gordon et al., 2009), average leptin values decreased with decreasing BW (Table 3). Since leptin is a hormone secreted by adipose tissue (Considine et al., 1996; Takahashi et al., 1996; Buff et al., 2002; Kearns et al., 2006), a decrease in leptin helps to confirm a loss in adipose tissue during the 28-d time period.

A visual difference in metabolic response between morning and evening feedings was observed (Fig. 1). There was an effect of AUC glucose and insulin when horses were fed in the morning vs. evening from both the NET and FLOOR (P ≤ 0.0056). The effect of feeding time could be attributed to a few factors, including the addition of the ration balancer to the morning feeding and the amount of time between meals. On average, horses were fed 0.56 kg of the ration balancer each morning. In the current study, the feeding effect of the ration balancer appears to cause an increase in blood glucose and insulin similar to what is observed after a concentrate meal is fed (Stull and Rodiek, 1988). However, it is possible the increase in blood glucose and insulin from the addition of the ration balancer during the morning feeding was exacerbated by the limit-fed diet. In the current study, horses were fed at 0700 and 1600 h, resulting in 9 and 15 h, respectively, between feedings. It is possible the extended time between the evening and morning feedings could have contributed to the greater AUC glucose and insulin observed in the morning. Although a difference was observed, insulin and glucose values after the morning feeding remained within normal ranges of 90 to 160 mg/dL for glucose and 5 to 46 mIU/L insulin (Williams et al., 2001; Eiler et al., 2005).

Conclusion

All horses lost BW when subjected to the restricted diet; however, no differences were observed between horses feeding from the FLOOR and NET. Horses feeding from the FLOOR took less time to consume their hay meal than horses feeding from the NET. There were no differences in BCS, morphometric measurements, or ultrasound measurements between the treatments throughout the study period, except LD thickness in horses feeding from the NET. Only TTP insulin and peak cortisol levels were affected by treatment, with horses feeding from the NET having lower values than horses feeding from the FLOOR. Average insulin, glucose, cortisol, and leptin; AUC cortisol; and peak insulin were affected by day. Glucose and insulin values increased, whereas cortisol and leptin levels decreased during the study. An effect of morning vs. evening feeding was observed for insulin and glucose and was attributed mostly to the addition of a ration balancer to the morning feed ration. However, insulin and glucose levels remained within normal ranges for adult horses. The use of a slow-feed hay net coupled with a limit-fed diet appears to be an effective method for decreasing BW and maintaining more homeostatic levels of postprandial metabolites and hormones when feeding overweight adult horses.

 

References

Footnotes


Comments
Be the first to comment.



Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.