Digestible Energy Intake and Digestive Efficiency of Human-Managed North American River Otters (Lontra canadensis)

Diets currently provided to captive North American river otters (Lontra canadensis) are highly variable with different institutions providing various whole foods, commercial complete prepared diets, or combinations of both. This study investigated the digestible energy intake, gastrointestinal transit time, and digestive efficiency of three different diets being fed at three North Carolina institutions. Otters housed at Institution A (n = 3) were fed strictly fish. Otters housed at Institutions B (n = 3) and C (n = 2) were fed a majority fish based diet (58.5 and 74.1%, respectively), supplemented with fruits, vegetables, and supplemental protein sources as enrichment. There was an apparent trend between increased percentage of fish in the diet and faster transit time and higher digestive efficiency. As less fish was included in the diets, the GI transit time was longer (Institution A, 106 minutes; Institution B, 145 minutes; Institution C, 208 minutes). Median digestive efficiency was high for all three groups (A, 91.4%; B, 87.8%; C, 89.8%) but was higher for the institutions feeding fish. Additionally, the overall median gross energy intake for the eight animals in this study was 163.1 kcal/kgBM0.75/day (range: 92.2 to 260.7 kcal/kgBM0.75/day). While all three institutions had healthy otter populations, it appears that a higher fish diet should be further studied as the model North American river otter diet.


Introduction
e North American river otter (Lontra canadensis) is a semiaquatic opportunistic predator that feeds primarily on fish but also a wide variety of other prey items (e.g., crustaceans, amphibians, reptiles, birds, mammals, and insects) [1][2][3][4][5][6]. Once found in virtually every watershed in North America, they are being successfully reestablished in many areas after being extirpated from much of their original range [2,7,8].
e success of human-managed populations of North American river otters depends on proper nutrition [9,10]. e diets currently provided to human-managed North American river otters vary considerably. Some institutions provide a variety of whole food items (e.g., fish, shellfish, rodents, fruits, vegetables, and chicken), others feed commercial nutritionally complete pre-prepared diets, while few provide a mixture of both [11,12]. While the digestibility of different nutrients in commercial diets has been described and documented for domestic dogs and cats, comprehensive dietary evaluations for most managed zoological carnivores are limited [10,[13][14][15][16]. e objective of this study was to determine the gross energy intake, gastrointestinal transit time, and apparent digestibility of gross energy, crude protein, and crude fat of three different diets being fed to three different managed North American river otter populations.

Study Animals.
Eight adult North American river otters held in three North Carolina institutions denoted in this paper as Institution A, n � 3 (1 male; 2 females); Institution B, n � 3 (3 males); and Institution C, n � 2 (1 male; 1 female) ( Table 1), were used in this study. All of the animals appeared outwardly healthy, based on routine clinical examinations. One of the animals housed at Institution B was considered moderately overconditioned (7/9) based on the clinical body condition score (1 to 9 scale with 1 being very thin, 9 being grossly obese, and 5 being ideal), while all of the other animals in this study were considered close to an ideal body condition (5 to 6/9). All procedures in this study were reviewed and approved by the appropriate research review committees for each of the participating institutions. Except for an individually housed female at Institution A, animals were group housed at each institution. e animals exhibited at Institutions B and C were housed completely indoors with exposure to natural lighting via skylights, while the enclosure at Institution A was outdoors with indoor holding. For the duration of this study, all animals were isolated during feeding and then returned to their groups.

Animal Diets.
Otters housed at Institution A were fed a fish-only diet. Otters housed at Institutions B and C were fed different fish-based diets (Institution B, 58.5% fish; Institution C, 74.1% fish) supplemented with fruits and vegetables along with supplemental protein sources (e.g., krill, crayfish claws, shrimp, and hard boiled eggs) ( Table 2). All animals were fed twice daily as follows (wet mass): Institution A, 659.0, 751.2, and 914.2 g/animal/day; Institution B, 898.8, 1008.12, and 1087.4 g/animal/day; and Institution C, 507.4 and 575.5 g/animal/day, and provided a thiamine and vitamin E supplement (Institution A: Mazuri iamin-E Paste, Henderson, CO; Institution B: Mazuri Marine Mammal Tablet, Henderson, CO; Institution C: VPL Optima 365, Phoenix, AZ). Water was provided ad libitum.

Nutritional Analysis.
Gastrointestinal transit time and total fecal collection digestibility assessments were performed over a 3-day period. Uniquely colored markers (McCormick & Company, Inc., Sparks Glencoe, MD) were added to each diet to identify individual otter scat in the group-housing situations. e rate of food transit was measured on Day 1 of the diet trial. e time of ingestion of the marked diet was recorded, and the animal was observed every 2½ minutes until the appearance of the marked feces.
Gastrointestinal transit time was calculated as the interval between the first ingestion of the diet and the appearance of marked feces.
Prior to each feeding on Days 1 and 2, the wet weight of each diet component was recorded. All of the food left uneaten (ort) was immediately collected and weighed. Total feces were collected throughout the day and the following mornings and weighed. Four separate samples of each fish species fed, each of the supplemental enrichment portions of the diets, and feces from individual animals were individually homogenized in a kitchen blender (Oster, Inc., Neosho, MO), and the homogenates were placed in an ultralow freezer ( ermo Fisher Scientific, Inc., Waltham, MA) set at − 70°C and stored up to 4 months until analyzed.
To determine percent dry matter, aliquots of frozen samples were thawed and then dried under forced air at 60°C for 72 hrs. To determine adequate drying, repetitive weights were acquired after each drying, and when weight change between weighings was less than 0.1 g, a sample was considered dry matter. Percent dry matter was calculated by dividing dry weight by wet weight multiplied by 100 (AOAC method 950.46). Dried samples were ground using a handheld coffee grinder (Krups, Inc., Millville, NJ), and gross energy was determined for each fish species fed, the supplemental diet enrichment component, and each fecal sample in duplicate using an IKA C5000 bomb calorimeter (IKA ® Works, Inc., Wilmington, NC). Duplicate subsamples were averaged to yield a single data point. Variation in duplicate subsamples was less than 4.0% for all diet items and feces samples analyzed. Fish, supplemental diet enrichment, and feces were analyzed for crude fat by Soxhlet ether extraction (AOAC method 2003.05) and crude protein using a Leco nitrogen/protein determinator (Leco FP-528, LECO Corporation, St. Joseph, MI; AOAC method 992.15) in duplicate by a commercial laboratory service (Zooquarius, Inc., Ithaca, NY). All nutritional compositions were calculated and presented on a dry matter basis (DMB). Percent apparent nutrient digestibility was then calculated using the following formula [17]:

Results
e nutrient composition of the commercially available fish used in the diets varied (Table 3). Gross energy (DMB) ranged from 5136 to 6955 cal/g across the species of fish sampled. Whole catfish was the fattiest, with a crude fat of 56.0%, and had the lowest protein with a crude protein of 39.2%. Tilapia and salmon fillets used in some diets had the lowest crude fat (9.9% and 9.3%, respectively) and the highest crude protein (89.0% and 88.6%, respectively) of all fish species tested.
Gross energy for each of the actual overall diets fed to the animals was within 10% of each other (range: 5272 to 5622 cal/g; Table 4). e crude protein of overall diets varied more widely (range: 53.7 to 73.6%; Table 3) and was nearly 35% higher for the diet composed of entirely fish compared to the two diets composed of fish supplemented with fruits and vegetables. e fat composition of the Institution C diet (29.9%) was similar to the crude fat of the Institution B diet (23.4%), but the diet for Institution A (entirely fish) contained just over half the amount of crude fat (15.7%) of diets fed at Institutions B and C (fish supplemented with fruits and vegetables). e median gross energy intake for all eight animals was 163.1 kcal/kgBM 0.75 /day (range: 92.2 to 260.7 kcal/ kgBM 0.75 /day). Median gross energy was the highest for otters held at Institution B (228.5 kcal/kgBM 0.75 /day), followed by Institution A (154.5 kcal/kgBM 0.75 /day) and Institution C (144.2 kcal/kgBM 0.75 /day; Table 5). e gastrointestinal transit time for animals fed only fish tended to be faster than that for animals fed fish along with vegetables and fruits (Institution A, 106 minutes; Institution B, 145 minutes; Institution C, 208 minutes; Table 6).
Apparent digestibility of gross energy, crude protein, and crude fat differed only slightly between the three institutions (ranges: gross energy, 87.8 to 91.4%; crude protein, 87.8 to 92.1%; crude fat, 90.4 to 96.5%; Table 6). Apparent digestibility for gross energy, crude protein, and crude fat was highest for those animals fed only fish diet (Institution A).

Discussion
e crude nutrient compositions for the commercially available fish species analyzed in this study were in the same range as previously published values [18,19]. While whole fish is a good source of most nutrients, the nutrient parameters vary a great deal among fish species, and within fish species based on diet, season, and lifecycle/physiological stage of the fish [18][19][20]. e values for the nutrient parameters presented here provide only a snap shot of the varying nutrient composition of fish provided to these animals over time. Basic diet analysis such as gross energy,   Veterinary Medicine International crude fat, crude protein, vitamins, and minerals should be performed with every lot of fish being used in the diet preparation for otters. is ensures that the animals are being provided a balanced diet, as seasonal changes in these nutrient parameters can lead to a large inconsistency in the intake gross energy, crude fat, and crude protein with similar quantities of fish being consumed [19]. e nutritional needs of the North American river otters are often extrapolated from the requirements of domestic cats and diets that have been successfully used to maintain captive otters [12,14]. e gross energy found in the three primarily fish-based diets of this study was similar to those reported for other primarily teleost fish diets fed to harp seals (Phoca groenlandica) (range: 4680 to 6690 cal/g), Hawaiian monk seal (Monachus schauinslandi) (range: 4760 to 5740 cal/g), and Steller sea lion (Eumetopias jubatus) (6001 cal/g) as well as several commercially prepared diets (polar bear, exotic feline, and domestic cat) (range: 4720 to 6690 cal/g) fed North American river otters [11,[21][22][23]. e high gross energy needs suggested for North American river otters and provided in the diets of this study address a concern for a high energy demand for this species. is concern for a high energy demand is based on the high metabolic rate for this species, which is approximated at 20 to 48% greater than that expected for most other mammals [24,25].
Protein, carbohydrate, and fat all provide energy in a diet, with fat providing energy on average twice that of protein and carbohydrates [12]. e fish-only diet provided to the otters at Institution A contained the highest concentration of protein of the three diets analyzed, but the crude protein concentration of all three diets was well above the minimum recommendations reported in the North American River Otter Husbandry Manual (24 to 32.5% crude protein) and similar to that of several diets being fed to both captive North American river otters (33.3 to 54.1% crude protein) and Asian small-clawed otters (42.9 to 68.1% crude protein) [11,12,26]. Crude fat concentration in the three diets analyzed for this study was consistent with the current minimum recommendation for the North American river otters (15 to 30% crude fat) and that of numerous diets being fed to captive otter species (11.3 to 37.9% crude fat) [11,12,26]. Crude fat concentration varies more than 50% in diets being fed to otter species with no reported clinical signs of steatorrhea, pancreatitis, or significant pathological changes in the cardiovascular system suggesting that otters have the capacity to tolerate and utilize variable levels of dietary fats [11,12,26]. e median gross energy intake was numerically higher for the otters housed at Institution B when evaluated against the animals at either Institution A or C. is was expected because the otters at Institution B were fed a greater amount of diet by wet weight when compared to the animals in the other institutions. e median gross energy intake established from the otters in this study was consistent with reports on gross energy intake for both North American otters (177.0 kcal/kgBM 0.75 /day) and Eurasian river otters (Lutra lutra) (171.9 and 225.5 kcal/kgBM 0.75 /day) and fishers (Martes pennanti) (131.4 to 188.8 kcal/kgBM 0.75 /day) fed several different diets comprising various whole food items and slightly higher than those reported for other comparable sized terrestrial mammals (domestic cat (Felis silvestris catus), 42.0-134.7 kcal/kgBM 0.75 /day; bobcat (Lynx rufus), 46.3 kcal/kgBM 0.75 /day; African wildcats (Felis silvestris lybica), 91.5 to 120.5 kcal/kgBM 0.75 /day) [27][28][29][30][31][32][33][34][35]. e gastrointestinal transit time of the North American river otter is much faster than that reported for many other terrestrial carnivores [36]. is would presumably allow less time for complete hydrolysis and absorption of nutrients. e otters fed a strictly fish diet in this study had a notably faster transit time than the otters fed fish supplemented with both fruits and vegetables. e median transit time for the otters at Institution A was also faster than previous data from North American otters fed a diet of various whole food items (202 minutes) and those fed several different commercially available diets (167 to 188 minutes) [11,30]. Diets with a   [37,38]. is trend was also observed in North American river otters fed several different commercially available diets with varying degrees of crude fat [11]. e addition of both insoluble and soluble fibers from the supplementation of both fruits and vegetables may also have affected the transit time of the animals in this study though regrettably analysis of the total dietary fiber was not performed. e median apparent digestibility of gross energy for all three groups in this study was consistent with reports measuring apparent digestibility of gross energy in North American river otters fed Nebraska Feline Diet (90.2%) (North Platte, NE), Mazuri Polar Bear Diet (83.1%) (Henderson, CO), and Hill's Science Diet cat food (86.6%) (Topeka, KS), as well as fishers fed several different diets comprising various whole food items (80.9 to 92.6%) and both walrus (Odobenus rosmarus) (92.7%) and Steller sea lions fed a diet of herring (95.5%) [11,23,27,39]. e animals in this study fed strictly fish appeared to be slightly more efficient at the digestion of gross energy, crude fat, and crude protein than those animals supplemented with both fruits and vegetables. e decrease in apparent digestibility of nutrients (gross energy, crude fat, and crude protein) in the animals at both Institutions B and C could be related in part to the supplementation of fruits and vegetables and therefore an increase in fibers or the less digestible fat in these food items. Our results were consistent with findings that an increase in fibers in the diets of mink (Mustela vison), blue fox (Alopex lagopus), and both the domestic cat and dog decreased the apparent digestibility of crude protein, which corresponded with the increase in concentration of plant protein, which is less likely digestible than the animal proteins it replaced [40][41][42]. It was anticipated that the higher concentration of crude fat in the diets at both Institutions B and C would have increased the apparent digestibility for the nutrients analyzed (gross energy, crude fat, and crude protein) for those diets similar to what has been reported for dogs and blue foxes with the addition of fat to their diets [43,44]. at was not the case in our study. While not measured in our study, the addition of fibers to the diets through the supplementation of fruits and vegetables may explain the increase in the excretion of fecal fat, thereby decreasing the apparent digestibility of crude fat [45].
Digestibility trials can be difficult to perform on animals housed in more natural habitats in zoological collections due to the difficulty of conducting total feces collection. is difficulty can be overcome by comparing the changes in concentrations of an inert dietary marker [17]. e animals in this study used the same latrine sites which allowed us to collect a complete fecal sample making dietary markers unnecessary to determine apparent digestibility.
is study indicated an apparent trend between an increase in the percentage of fish offered in the diet and a faster transit time and higher digestive efficiency. Knowledge on energy content, gastrointestinal transit time, and apparent digestibility of nutrients associated with a more fish-based diet is important if zoological institutions are to provide a nutrient dense diet sufficient for maintenance, growth, and reproduction in this highly active species. is study did not examine the digestibility of any of the diets in regard to vitamins and minerals, and further studies should be performed to evaluate these nutrient parameters and to ensure that these animals are being provided a completely balanced diet.
Data Availability e full data set which was used for this research is available from the corresponding author on reasonable request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.