To avoid interference from colonic microflora and to allow repeated sampling of digesta over time, ileo-rectal anastomoses (IRA), have in the past been used to collect pre-caecal digesta from the terminal ileum. Numerous IRA configurations have been tried since they were first proposed in the early 1980s (Fuller and Livingstone 1982; Laplace et al. 1985; Picard et al. 1984). Digestibility estimates obtained using these procedures vary depending on the configuration used (Darcy-Vrillon and Laplace 1990). In all of the ileo-rectal configurations the normal functions associated with the large intestine are interrupted, and will therefore affect the absorption of water and minerals (Köhler et al. 1991). Since the first use by Fuller and Livingstone (1982) little research has been conducted to determine the effects of IRA on intestinal morphology. Salgado et al. (2002) found that anastomosed pigs had higher spleen and small intestine weights and lower large intestine weights than intact pigs. They also discovered that IRA influenced intestinal villus and crypt architecture although it had no significant effect on the activities of intestinal enzymes. Marinho et al. (2007) also found that IRA modified the intestinal villus and crypt architecture in the duodenum and ileum.

Although there are advantages of IRA there are also many disadvantages. The method has been deemed unethical by some researchers and is not permitted in some jurisdictions. Despite this the method still finds application (Branner et al. 2004; Bohmer et al. 2005; Fontes et al. 2007; Hennig et al. 2006, 2008; Hackl et al. 2010).

Classically, methods to obtain ileal digesta samples have involved the use of cannulae, where the lumen of the distal ileum has been exteriorised.

Cannulation techniques can be broadly divided into two categories, those where the experimental animal is fitted with a single cannula and those fitted with two, the so called re-entrant variants. Compared with single cannulation techniques, re-entrant cannulation has been thought to give more reliable nutrient digestibility values since this technique allows the collection of total ileal digesta samples within the experimental time period and this obviates the need for an indigestible marker (Yin and McCracken 1996). However, both approaches have been criticised on the grounds that such procedures have a modifying effect on the processes of digestion and absorption (Sauer and Ozimek 1986) and that differences in the determined digestibility may occur depending on the technique used (Darcy-Vrillon and Laplace 1990). Double re-entrant cannulae are rarely used today because of the complex and expensive surgical procedures.

Studies comparing the different techniques have been undertaken by a number of researchers who have all reported some disparity in the calculated digestibility coefficients (Donkoh et al. 1994b; Köhler et al. 1990; Yin and McCracken 1996; Yin et al. 2000a, b), although some of these differences may be attributable to other secondary factors including the composition of the diet, the amount of dietary fibre or the nature of the indigestible marker (Laplace et al. 1994; Sauer and Ozimek 1986; Yin et al. 2000a, b). Further discussion on the merits of these techniques is therefore required.

First proposed by Livingstone et al. (1977a) this technique utilised a “T”-shaped cannula made of annealed Pyrex glass that had an internal diameter of just 16–20 mm (Fig. 17.1). The cannula was secured from becoming internalised by one or more Perspex washers and a self-locking nylon strap. The shank of the cannula was threaded to take an impact-resistant plastic cap. The cannula was inserted 15 cm proximal to the ileo-caecal valve. Livingstone reported that the cannulae remained intact and functional when fitted to pigs, housed in smooth-walled pens, growing to a live weight of 175 kg. Trauma from the surgery of fitting the cannulae was reported to be minimal and the intestinal activity, appetite, and growth rate of the animals was unaffected by the cannulation. They concluded that pigs fitted with such a cannula were physiologically similar to intact pigs (Livingstone et al. 1977a, b). As with all single cannula techniques, not all of the digesta flows through the cannula and the incorporation of an indigestible dietary marker to determine the total flow of digesta is essential. It is assumed that the composition of the digesta collected via the T-cannula is representative of the total quantity of digesta flowing through the distal part of the ileum, and that no fractionation of its contents, particularly between the solid and liquid phases occurs. However, this has been questioned by a number of researchers (Fuller et al. 1994; Sauer et al. 1989; Yin and McCracken 1996). Differences in the recovery of indigestible markers have indicated that phase separation may indeed occur and that this may be one reason why a greater variability in results was observed by Köhler et al. (1990) when they compared a simple T-cannula with a post-valve T-caecum cannula and a re-entrant cannula. They suggested that a reduction in the homogeneity of the samples recovered through the T-cannula may be attributable to changes in pressure at the base of the open cannula, forcing a separation of coarse and fine insoluble particles (Schröder 1988), and this might explain differences in the digestibility coefficients observed in diets rich in fibre. Although no incidence of the cannula blocking was reported by Livingstone et al. (1977a, b) this has been noted by Potkins et al. (1991), who had to change their experimental design owing to the frequency of blockages at the site of the cannula and the consequential inability to collect digesta when their experimental animals were fed a diet rich in bran fibre. Fernández et al. (2001) also reported problems with the simple T-cannulation technique, finding unreasonably large deviations in results when studying fibre rich materials. However, and despite its few disadvantages the use of simple T cannulae, for the determination of nutrient digestibilities, is one of the most commonly used cannulation techniques (Goerke et al. 2012). Today, flexible plastic and rubber T cannulae have replaced the early glass models.

Developed by van Leeuwen et al. (1988) in the late 1980s at the Institute for Animal Nutrition and Physiology in the Netherlands, the Post-valve T-caecum cannula (PVTC) is made of medical grade silicon tubing, with an internal diameter of 25 mm and is secured by an exterior ring and self-locking nylon straps (Fig. 17.2). The technique involves the removal of part of the caecum and installation of the cannula opposite the ileo-caecal valve. One hour before the collection of digesta starts, the cannula is opened to allow the ileo-caecal valve to protrude into the lumen of the cannula. Although this ensures that more of the digesta flowing from the ileum is collected, this technique cannot guarantee a complete quantitative collection of ileal digesta and so the inclusion of an indigestible dietary marker is advised (van Leeuwen et al. 1991). Having studied more than 100 animals fitted with PVTC cannulae, with live weights ranging from 8 to 100 kg, van Leeuwen’s group found that post-operative recovery was swift and that the animals displayed no signs of discomfort or a reduction in appetite, even over long periods of time. Following the partial caecectomy and insertion of the cannula into what remains of the caecum, the physiology of both the ileum and the colon appear to be normal. As the gut is not transected, there appears to be no interference with the myoelectric innervation of the intestines that may reduce intestinal motility, points that have been corroborated by a number of other researchers (Köhler et al. 1990, 1992a, 1992b; Yin and McCracken 1996; Yin et al. 2000b). Although few digestibility studies have compared PVTC cannulae with other methods of collection, those that have concluded that results, using the PVTC technique, are comparable with those obtained using simple T-cannula (Köhler et al. 1990; Yin and McCracken 1996; Yin et al. 2000b).

In one long-term study by Kohler et al. (1992a, b) using 23 pigs (9 with IRAs, 7 with PVTCs and 6 intact pigs), no differences were found in growth performance, body nitrogen retention, blood variables and mineral balances, between intact animals and those fitted with PVTCs.

Greater precision can be achieved using the PVTC method for most diets and this is thought to be attributable to greater consistency in the recovery of the indigestible marker, specifically titanium dioxide. However, with high fibre diets some variability in the amount of digesta recovered has been recorded, which confirms the necessity of using an indigestible marker (Köhler et al. 1990, 1991; van Leeuwen et al. 1991; Yin et al. 2000a) Although the PVTC cannulation technique has been found to have a minor effect on total tract nutrient digestibility (Lindberg 1997), it is at present considered to give the most satisfactory results for ileal digestibility determination (Hodgkinson and Moughan 2000a; Presto et al. 2011).

Mroz et al. (1994, 1996) developed a variant of the PVTC technique which they termed the steered ileo-caecal valve (SICV), cannulation (Fig. 17.3). This technique permits the collection of digesta, with both low and high fibre diets, and maintains the normal physiological function of the gut as it does not require the removal or transection of any part of the GIT. Although this technique was thought to allow the quantitative collection of digesta, the recovery of a chromium marker from the faeces of cannulated pigs was approximately 20 % less than in intact pigs. Therefore under the conditions of their study, the use of an inert marker was deemed advisable for determination of the total tract digestibility and ileal digestibility of nutrients. Changes proposed by Radcliffe et al. (2005) resulted in a decrease in post-surgical complications, notably fewer and less severe adhesions, together with less leakage of digesta around the cannulae. The surgery for this technique is more complex than the PVTC method, but it appears to be less prone to blockages.

In comparison to the simple T cannula, Zhang et al. (2004) concluded that ileal digesta samples collected using the SICV-cannula were more homogenous.

Although re-entrant fistulae have been commonly used to study the digestive process in ruminants (Phillipson 1952), Cunningham et al. (1962, 1963) used a similar re-entrant cannula to compare the digestibility and rate of passage of a variety of purified and natural feed materials in the gut of the pig (Fig. 17.4). Blockages would often occur in the cannulae, particularly, if coarsely ground feeds were administered, but the cannulae were well tolerated by the pigs. Kowalik et al. (2004) studying the digestion of starch and crude fibre in segments of the digestive tract of sheep reported no problems with blockages.

Both Cunningham et al. (1963) and later Cho et al. (1971) found that the flow of digesta through the externally joined ileo-ileal re-entrant cannulae was not uniform. Easter et al. (1972) and Easter and Tanksley (1973) believed that the surgically disrupted small intestine was not capable of forcing digesta through the externalised cannulae and the ileo-caecal valve. As the ileo-caecal valve was not essential to the digestion process (Davenport 1966), changing the configuration of the re-entrant cannula from ileo-ileal to ileo-caecal offered a solution to this resistance (Fig. 17.5). Although the use of single caecal cannulae in the digestion studies of pigs had been demonstrated by earlier researchers (Henderickx et al. 1964; Redman et al. 1964), this was the first time that a double, ileo-caecal re-entrant had been used. Although not widely used, ileo-caecal re-entrant cannulae are still employed particularly in studies involving dietary fibre (Bartelt et al. 1999; Rapp et al. 2001).

First described by Darcy et al. (1980), Ileo-colic post-valve fistulation (ICPV), which they considered to be a reference method (Fig. 17.6), is surgically difficult and too time consuming for routine ileal digestibility measurement. Early suggestions that this method could be used in diets rich in fibre (Sauer and Ozimek 1986) were recanted by Fuller et al. (1994) who reported evidence to the contrary for diets rich in barley.

There is little agreement between researchers as to which digesta collection technique is the best, although lately the PVTC method has been popular. Variations in determined ileal amino acid digestibility values have been noted due to differences in:

Indeed Fuller et al. (1994), state that each method has its strengths and limitations such that no single method is suited for all purposes, a point echoed more recently by Knudsen et al. (2006) when reviewing in vivo methods of studying the digestion of starch in pigs and poultry. Such variation is also compounded by the differing performance of indigestible markers necessary to quantify the total flow of digesta (Yin et al. 2000b). A comparative summary of the advantages and disadvantages of the various collection techniques is given in Tables 17.2 and 17.3.