1. Introduction
Chronic enteropathies (CE) represent a heterogeneous group of common diseases in dogs, typically manifesting with clinical signs such as diarrhea, vomiting, anorexia, and/or weight loss [
1]. Diet is known to play a crucial role not only in disease pathogenesis (representing a potential risk factor) but also in the therapeutic approach [
2]. Dietary strategies based on the use of “limited-ingredient” diets or novel protein sources (or protein hydrolysates) are often effective in managing the gastrointestinal symptoms [
3]. In this context, dietary medium-chain triglycerides (MCT), which are lipid molecules made up of medium-chain fatty acids (MCFAs), are proving to be interesting from a nutritional perspective. MCFAs, which represent saturated fatty acids ranging from 6 to 12 in carbon chain length and include caproic (C6:0), caprylic (C8:0), capric (C10:0), and lauric (C12:0) acids, are more soluble in water and biological liquids than long-chain fatty acids (LCFA), due to their lower molecular weight and smaller size. Consequently, MCFAs are rapidly and more efficiently absorbed through the intestinal mucosa and transported directly through the portal vein to the liver, bypassing the lymph system [
4], and therefore represent a preferential substrate for β oxidation rather than for lipid storage [
5,
6]. Because of their unique digestive and metabolic properties related to the potential role of MCTs in lipid catabolism and the reduction in body fat mass, MCTs are used to overcome metabolic and digestive diseases such as fat malabsorption and obesity [
5,
7]. In particular, MCT-enriched diets are currently considered the primary treatment strategy for people with primary intestinal lymphangiectasia, and they have been proven to reduce overall mortality [
8,
9]. In dogs, MCT supplementation has as yet been described in two cases diagnosed with protein-losing enteropathy [
10,
11]. Studies on the swine model have demonstrated the potential of these lipid molecules as feed antibiotic replacers, improving the intestinal mucosal structure and the growth performance in weaning animals [
12,
13,
14] and exhibiting interesting antibacterial effects in the pig intestinal environment. This is due to their capacity to penetrate the bacterial membrane, destroying internal structures (this effect seems to be exerted by lauric acid, in particular), with special regard to Gram-positive and potentially harmful bacteria [
13,
15].
Since the addition of non-esterified MCFA mixture in pig feed could have a negative impact on feed intake due to sensory properties (particularly over 8% of inclusion), to overcome this inconvenience, MCT-enriched ingredients have been evaluated in several studies, allowing dietary inclusion rates up to 15%, as reviewed by Zentek et al. [
16].
In this context, virgin coconut oil (VCO) has recently received great attention from the scientific community because of its relatively high content of MCT (which represents more than 50% of its total fat matter), principally formed by lauric, caprylic, and capric acids [
17,
18]. VCO is believed to offer several health benefits, including antifungal, antioxidant, antibacterial, antiviral, hepatoprotective, and anti-inflammatory effects [
19]. Moreover, beneficial effects on weight loss, glycemic index, and the immune system have been proposed for this vegetable oil [
20]. Recently, in swine species, some studies have proposed the positive influence of dietary coconut oil supplementation both on the intestinal microbiota, supported by an increase in the rectal abundance of beneficial bacteria such as
Bifidobacterium and
Lactobacillus spp., when added to feed mixture at 3 g/kg [
21], and on growth performance and immune function when it was used, at 2% of feed mixture, as an alternative to dietary antibiotics [
22]. Regarding canine species, at present, there is no strong evidence supporting the positive effects of MCFA- or MCT-enriched ingredients, such as VCO, on dog health. Nevertheless, some investigations have reported a tendency toward a slight improvement in fat digestibility in healthy dogs receiving a diet supplemented with 5.7% of MCT (dietary total lipid composition around 13%) [
23] and an increase in serum concentrations of cholesterol and fat-soluble vitamins in dogs affected by exocrine pancreatic insufficiency fed with a diet containing 35% of the total fat as MCT (dietary total lipid composition around 10–11%) [
24]. To date, the mechanism of intestinal absorption and the utility of MCT dietary supplementation have not been elucidated in canine species, while some evidence exists regarding food aversion when MCTs have been supplemented at high doses (22–25% ME) in this species [
25,
26].
However, based on the evidence collected in pigs, MCTs may represent interesting energy sources and intestinal microbial modulators in dogs affected by CE, potentially those presenting with fat malabsorption and/or microbial dysbiosis [
2].
We, therefore, hypothesized that VCO, as a source of MCT, may have a beneficial influence on the clinical scores, intestinal microbiomes, and metabolomes of CE dogs receiving a homemade limited-ingredient diet.
4. Discussion
The objective of the present study was to investigate the effects of a home-cooked diet supplemented with VCO as a source of MCT on clinical scores and fecal microbiota and metabolome of CE dogs. The inclusion level of VCO provided in this trial was relatively low (10% of ME) to ensure a good acceptance of the diet by dogs. Indeed, the owners did not report food refusal problems. In this regard, a worsening of diet palatability has been previously described with a higher concentration of MCT supplemented in dogs (22–25% of ME) [
25,
26]. In accordance with our evidence, Beyen et al. [
23] reported a total diet consumption in three dogs receiving a diet containing MCT at 11% of ME. In a recent study, Berk et al. [
36] reported no differences in food intake in 19 healthy dogs when MCT oil was fed at 10% of ME in a short-term (5 day) palatability test.
All dogs responded positively to the home-cooked diet and were classified as food-responsive (FRE). Accordingly, in the literature, it is reported that FRE represents a very common condition, as it comprises approximately two-thirds of CE dogs [
27,
37]. In FRE dogs, an improvement in clinical signs is typically expected within a few days and usually no additional treatments are necessary [
2,
3]. In this study, clinical parameters such as CCECAI and FS showed a significant improvement throughout the study, and most of the dogs, by the end of the study, reached scores typically associated with healthy animals. However, at the end of the present study, higher CCECAI scores (5 or 6 were recorded, compared to baseline, in two out of the eighteen dogs, expressing, therefore, a moderate/severe disease still persistent after 37 days of dietary treatment. Moreover, these dogs were clearly dysbiotic (DI > 6) with a very low level of
C. hiranonis (<2 log DNA) at baseline, and one showed suboptimal cobalamin concentration (the value was in the lower part of the reference range, <300 ng/L). Serum cobalamin concentrations of <350–400 ng/L have been associated with increased serum methylmalonic acid (MMA) concentrations [
38]. MMA is known to accumulate as a result of the decreased intracellular availability of cobalamin and can impair the urea cycle [
39]. However, MMA is currently not routinely performed in companion animals; therefore, cyanocobalamin supplementation is recommended when serum levels are <400 ng/L to avoid intracellular cobalamin deficiency in CE dogs, [
39,
40].
Concerning clinical scores, a beneficial effect attributable to VCO supplementation can be supposed only regarding FS, since this parameter was assessed before VCO supplementation (HCD) and at the end of the study (HCD + VCO). Conversely, CCECAI was assessed only at the beginning (baseline) and at the end of the study; consequently, it was not possible, regarding CCECAI, to differentiate the effect of HCD from that of VCO.
Unexpectedly, while FS displayed an improvement during the study, fecal water content slightly increased when dogs started to be fed with HCD. Water content is the main, but not the only, factor which influences fecal consistency; the fraction of fecal insoluble solids, its water holding capacity (WHC) properties, and the ratio of fecal water to fecal solids, are supposed to be other determinants of fecal firmness [
41,
42]. In this regard, the WHC of fecal insoluble solids has shown high variability in people with different dietary habits, showing that the ratio of fecal water to fecal solids depends mainly on diet composition [
41]. For instance, potato fiber, the main dietary fiber source in HCD in this study, is known to possess a WHC capacity similar to psyllium (around 22 g water/g fiber source) [
41,
43,
44]. Psyllium has been proven to improve fecal consistency in human diarrhea models, even though, at the same time, its ingestion was associated with an increase in fecal water content, probably due to its water-binding property [
41,
42]. The improvement of the FS and the higher fecal water content observed in this study might thus be explained by an increase in fecal viscosity depending on potato fiber intake, given the similar WHC capacity of psyllium and potato fiber. However, neither fecal viscosity nor the amount of fecal insoluble solid fraction was measured in this study. Consistent with our results, a linear decrease in fecal dry matter with no effect on FS was reported following an increase in potato fiber intake in healthy dogs [
45]. Moreover, the high water content of HCD might have influenced fecal water excretion, as the increase in fecal moisture was more remarkable in dogs previously fed low-water foods, such as extruded dry kibbles. Factors other than diet, such as dog size and time of feces collection during the day, may cause variation in fecal consistency through the effect on transit time [
46,
47,
48,
49]. In this study, small- and large-sized dogs were included, and no impact of dog size on fecal water content was found. Unfortunately, the time of feces collection was not recorded during the day; therefore, we cannot exclude that this may have influenced fecal water content. To the best of the authors’ knowledge, to date, no studies have assessed differences in fecal characteristics in dogs fed home-cooked or commercial diets, but no differences in fecal water content in healthy dogs fed commercial wet or dry diets have been observed [
50,
51]. This latter finding is consistent with data showing that fluids are largely reabsorbed through the colonic mucosa in healthy dogs [
52], but a reduction in intestinal absorptive capacity might be present in dogs with chronic gastrointestinal disorders, due to a more rapid colonic transit time [
50,
53].
The chemical analysis highlighted some interesting modulations on fecal fatty acid profile during the study, presumably induced by dietary treatments. In particular, an increase in the fecal proportion of MCFAs (C10:0 and C12:0) and C14:0, together with a reduction in C18:1c9 (MUFA), was observed as a consequence of VCO supplementation. The VCO is a lipidic source containing more than 92% SFAs (MCFAs, in particular), as previously reported [
18]. On the other hand, the increase in the fecal proportions of C16:0, C18:3n3, and C20:4−n6 following HCD consumption can be associated with the fatty acid profile of horse meat [
54,
55]. In humans, SFAs are generally considered to be hypercholesterolemic, whereas MUFAs and PUFAs are considered hypocholesterolemic [
18]. In this study, VCO consumption increased total fecal SFAs without affecting serum cholesterol levels in CE dogs.
A digestibility assessment was outside of the scope of this study, but MCFAs are considered to be almost completely absorbed by the intestinal mucosa as free fatty acids, without requiring the action of bile salts and pancreatic lipase [
56]. However, this absorptive mechanism is still controversial in dogs, since MCFAs have been found to be absorbed in canine lymph [
57]. Among MCFAs, C8:0 and C10:0 have short half-lives and rapidly enter hepatic mitochondria; therefore, they are presumably not incorporated into chylomicrons [
58]. Partially in agreement with this evidence, in the present study, an increase in the fecal proportion of C10:0, but not of C8:0, was exerted by VCO supplementation. Therefore, along with our results, an impaired intestinal passive absorption of MCFAs with carbon chains longer than eight carbons might be suspected in CE dogs. In this regard, because VCO contains prevalently C12:0 (and C14:0), purified MCT oils (high in C8:0 and C10:0), rather than VCO, could represent a preferable dietary MCFA source to be used in CE dogs [
59]. Moreover, investigations are needed to understand if MCFA absorption takes place in dogs via the portal venous system rather than via the lymphatic system, as has been observed in other animal species.
The overall fat content measured in fecal samples collected after the beginning of HCD treatment displayed a reduction in fat excretion compared to the baseline. The availability of data on fecal fat content is limited in dogs [
60,
61,
62], and a threshold value for excessive fecal fat excretion, identifiable as steatorrhea, has been not established so far. In human beings, an excretion of 4–7 g/fecal lipid/day (or fat <5% fecal wet weight) is considered normal, while a fecal fat content >9.5% is considered risky for steatorrhea [
63,
64]. In a study with healthy dogs fed a dry extruded diet, fecal fat loss depended on dog size, ranging from 1.2 to 14 g/fecal lipid/day for small to large size dogs, respectively [
61]. In another study, fecal fat excretion did not change in response to a progressive increase in dietary fat content (from 15 to 50 g/lipids/day) in healthy dogs, while a linear increase in fecal fat loss was observed when those dogs were deprived of the gallbladder [
60]. More recently, fecal fat loss was evaluated in healthy Beagle dogs in relation to different dietary treatments that consisted of a basal diet with increasing levels of poultry fat [
62]. In that study, when dogs were fed a diet containing 7.8% fat on a DM basis, a value close to that of the HCD diet used in the present study (7.9% fat on a DM basis), the fat content of feces was twofold higher than in our study (5% vs. 2.6% on a DM basis). This discrepancy could be explained by the composition of the different diets; in fact, differences in fat and fiber sources and levels of inclusion can result in considerable variation in fecal fat losses in dogs and pigs [
65,
66]. Interestingly, in the cited study with Beagle dogs [
62], fecal fat content was significantly higher when dogs were fed diets with fat content equal to or higher than 13% on a DM basis, compared to a 7.8% fat diet. The fat content of the diets that were fed to dogs before the HCD diet was not investigated in the present study, but a fat content higher than that of HCD could be supposed for dog adult maintenance diets; therefore, our results might mirror the study by Marx et al. [
62]. Other authors [
67] observed no changes in the fecal fat content of dogs as dietary fat levels increased, but a lower fecal output was registered in that case, and this might have accounted for the lack of difference in the fecal fat content. The total fecal output was not weighed in the present study; thus, speculation about the reduction in indigestible fecal matter content secondary to HCD, as well as an improvement of fat digestibility in CE dogs fed HCD, cannot be confirmed, even though the latter would support the clinical improvement also highlighted by CCECAI. However, in addition to unabsorbed dietary fat, fecal fat also comprised endogenous fat loss, which are molecules originating from bile salts, cholesterol, and the structural lipids of mucosal cells that pass through the intestinal tract to be finally excreted with feces [
68].
In this study, fecal sterol composition was affected by dietary treatments. In fact, the introduction of HCD, a diet rich in proteins from an animal source, horse meat, lead to a reduction in phytosterols and the phytosterol-to-zoosterol ratio in dog feces. On the other hand, a reduction in fecal campesterol concentration was observed secondary to VCO supplementation; campesterol is commonly found in the highest amounts in vegetable oils, with those extracted from seeds having around five-fold higher amounts of campesterol than VCO [
69].
Results from recent studies in dogs suggest a decreased concentration of β-sitosterol and sitostanol in CE dogs compared to healthy subjects, with no difference regarding zoosterols [
33,
70]. Compared to cholesterol, phytosterols are bioactive compounds poorly absorbed by the intestine (2–3% vs. 30–60% absorption rate for phytosterols and cholesterol, respectively [
71]); therefore, fecal phytosterol concentration better reflects dietary composition than zoosterols. In this study, total phyto- and zoosterol fecal concentrations were similar to values observed in CE Yorkshire Terrier dogs [
33].
Previous studies have demonstrated alterations in intestinal BA metabolism in CE dogs [
72,
73]. In the present study, concentrations of fecal BA, cholesterol, and its intermediates were not affected by HCD and VCO. However, a numerical decrease in the fecal excretion of primary bile acids (mainly due to a decrease in cholic acid) occurred when HCD was fed to dogs, but the high variability of fecal BA concentrations among CE dogs might have weakened the statistical power. Differences in the fecal BA concentrations of healthy and CE Yorkshire Terrier dogs have recently been described by Galler et al. [
33]. Moreover, Pezzali et al. [
74] hypothesized differences in BA concentration among different dog breeds. Studies on alterations of the intestinal BA metabolism have been suggested to be a promising target for fecal analysis because of their important role in host metabolism and their relationship with gut bacteria [
75]. In particular, the intestinal microbiota affects bile acid metabolism through deconjugation and dihydroxylation of primary BA to form secondary BA [
73]. In dogs,
C. hiranonis represents the main bacterial converter of primary to secondary BA [
76]. Since the abundance of this bacterial species is often depleted in CE dogs or, in general, in dogs presenting with intestinal dysbiosis, in these situations an increase in primary BA (potentially the cause of secretory BA diarrhea) and a decrease in secondary BA has been often observed [
75]. In this study, fecal concentrations of
C. hiranonis were not significantly affected by HCD or VCO supplementation, and this result is in accordance with the lack of changes in BA metabolism. In fact, the results from the present study confirmed the inability to convert primary to secondary BA in dogs with low
C. hiranonis abundance.
The CE dogs involved in this study were not all affected by gut dysbiosis since the median DI was lower than 0, which is considered normal [
34]. Moreover, DI was not significantly affected by dietary treatments, and all dogs classified as clearly dysbiotic had low
C. hiranonis abundance at each sampling time. Contrary to the clinical score (CCECAI), which includes symptoms such as vomiting and diarrhea that might improve rapidly in response to dietary treatment, the DI rather reflects the changes on a mucosal level which might take longer to recover.
The abundance of total bacteria, as well as
Fusobacterium spp., increased as a result of feeding with the HCD. Fusobacterium is a co-dominant phylum of fecal microbiota in healthy dogs, and its abundance is associated with intestinal health [
77], with a subset of CE dogs showing a lower abundance of
Fusobacterium spp. compared to healthy dogs [
78]. Some bacterial species belonging to Fusobacteria are known to produce butyrate from dietary protein [
77], and its growth in dogs is stimulated by high-protein diets, such as raw meat-based diets [
79,
80]. Among short-chain fatty acids (SCFAs), butyrate is the most important energy source for colonocytes and has shown local and systemic anti-inflammatory effects, as well as increased expression of immunosuppressive cytokines in human ulcerative colitis [
81,
82]. Even though the fecal abundance of
Fusobacterium spp. in CE dogs in this study was within the reference range previously attributed to healthy dogs [
77], the HCD significantly increased the fecal concentration of this bacterial genus only a few days after starting the dietary treatment. Conversely, the fecal depletion of
Fusobacterium spp. caused by severe dysbiosis or antibiotics treatments would probably take longer to restore, as shown in a previous study where Fusobacteria abundance remained depleted in healthy dogs treated with metronidazole, even 4 weeks after metronidazole discontinuation [
83]. In this study, we did not evaluate the effects exerted by HCD or VCO on fecal SCFA and other postbiotics, even though future nutritional interventions involving substances released by or produced through the metabolic activity of intestinal bacteria may significantly benefit the host in disease conditions linked to the microbiome, such as chronic intestinal inflammation [
84].
The study has some limitations that should be considered. Firstly, the dogs used in this study were client-owned and were fed different habitual diets that could have resulted in a potential confounding factor in the assessment of the effects exerted by the HCD treatment on the fecal microbiota, sterol, and fatty acid profile of CE dogs. In addition, HCD was fed to CE dogs for only 7 days before adding VCO; therefore, it cannot be differentiated if the improvement in FS observed at the end of the study was associated with VCO supplementation or a delayed effect associated with HCD. Finally, we cannot exclude that an improvement in parameters other than clinical signs, such as the DI, would have been seen if the dietary treatment was tested long-term.