Adverse reactions to food (food sensitivities) include

Transcription

1 J Vet Intern Med 200;5:7 Food Sensitivity in Cts with Chronic Idiopthic Gstrointestinl Problems W. Grnt Guilford, Boyd R. Jones, Peter J. Mrkwell, Donld G. Arthur, Mrk G. Collett, nd John G. Hrte The objectives of this study were to investigte the prevlence of food sensitivity in cts with chronic idiopthic gstrointestinl problems, to identify the food ingredients responsible, nd to chrcterize the clinicl fetures. Seventy cts tht presented for chronic gstrointestinl signs underwent dignostic investigtion. Fifty-five cts hd idiopthic problems nd were entered into the study. Dignosis of food sensitivity ws mde by dietry elimintion-chllenge studies by using commercil selected-protein diets s the elimintion diet. Sixteen (29%) of the 55 cts with chronic idiopthic gstrointestinl problems were dignosed s food sensitive. The clinicl signs of nother cts (20%) resolved on the elimintion diet but did not recur fter chllenge with their previous diet. The foods or food ingredients responsible for the clinicl signs were dietry stples. Fifty percent of ffected cts were sensitive to more thn food ingredient. The clinicl feture most suggestive of food sensitivity ws concurrent occurrence of gstrointestinl nd dermtologicl signs. Weight loss occurred in of the ffected cts, nd lrge-bowel dirrhe ws more common thn smll-bowel dirrhe. Assy of serum ntigen-specific immunoglobulin E (IgE) hd limited vlue s screening test, nd gstroscopic food sensitivity testing ws not helpful. In conclusion, dverse rections to dietry stples were common in this popultion of cts, nd they responded well to selected-protein diets. Dignosis requires dietry elimintion-chllenge trils nd cnnot be mde on the bsis of clinicl signs, routine clinicopthologicl dt, serum ntigen-specific IgE ssy, gstroscopic food sensitivity testing, or gstrointestinl biopsy. Key words: Dignosis; Dirrhe; Food llergy; Vomiting. Adverse rections to food (food sensitivities) include those medited by the immune system (food llergies) nd those without n immunologicl bsis (food intolernces). Clinicl signs ttributed to food sensitivity by veterinrins usully re dermtologicl or gstrointestinl. The prevlence of chronic dermtologicl bnormlities resulting from food sensitivity in cts hs been estimted to be 5.8% in university prctice. 2 Food sensitivity is thought to be the second most common cuse of llergic dermtitis in cts nd is considered responsible for up to % of cts with miliry dermtitis. In contrst, the frequency with which chronic gstrointestinl complints in cts re cused by food sensitivity is unknown. 4 A number of observtions suggest tht the prevlence of food sensitivity in cts with gstrointestinl problems my be higher thn in cts with skin problems. For exmple, evidence from other species suggests tht not only cn food sensitivity produce gstrointestinl problems, but conversely tht gstrointestinl diseses cn led to food sensitivity by compromising orl tolernce. 5 Furthermore, food intolernces from such disorders s brush border enzyme bnormlities nd ingested toxins would more commonly ffect the bowel nd result in gstrointestinl clinicl signs rther thn dermtologicl signs. From the Institute of Veterinry, Animl, nd Biomedicl Sciences, Mssey University, Plmerston North, New Zelnd (Guilford, Jones, Collett); Wlthm Centre for Pet Nutrition, Wlthm-on-the-Wolds, UK (Mrkwell, Hrte); nd the Ministry of Agriculture nd Forestry, Qulity Mngement, P.O. Box 24, Lincoln, New Zelnd (Arthur). Presented in prt t the 4th Americn College of Veterinry Internl Medicine Forum, 996. Reprint requests: Professor W. Grnt Guilford, BVSc, PhD, Dipl ACVIM, Institute of Veterinry, Animl, nd Biomedicl Sciences, Mssey University, Privte Bg -222, Plmerston North, New Zelnd; e-mil: W.G.Guilford@mssey.c.nz. Submitted December 2, 999; Revised My 7, 2000; Accepted August 2, Copyright 200 by the Americn College of Veterinry Internl Medicine /0/50-000/$.00/0 It is uncertin to which food ingredients cts with gstrointestinl problems re most frequently sensitive. Limited informtion from studies of cts nd more compelling evidence from other species suggest tht the most prevlent food llergens re proteins commonly included in the diet.,5 In contrst, the food ingredients most likely to be responsible for food intolernces re more vried nd do not require prior exposure. These include discchrides (such s lctose), food dditives (such s coloring gents), phrmcologiclly ctive products (such s histmine), nd food toxins.,5,6 The gstrointestinl signs of food sensitivity in cts re poorly described. Vomiting nd dirrhe usully re reported,,5,7,8 but the specific chrcteristics of the pttern of vomiting nd type of dirrhe re rrely mentioned. Furthermore, the presence or bsence of prticulr clinicl fetures suggestive of gstrointestinl food sensitivity hve not been methodiclly investigted. Adverse rections to food usully re suspected when n ssocition is mde between the ingestion of certin food nd the ppernce of clinicl sign. The dignosis is confirmed by dietry elimintion-chllenge studies. Alterntive methods of dignosis hve been proposed in other species, including ssy of serum ntigen-specific immunoglobulin E (IgE) nd gstroscopic food sensitivity testing (GFST). 5,9 A commercil ssy of ct ntigen-specific IgE in serum now is vilble. A technique to perform GFST recently hs been developed for use in dogs nd wits ppliction in cts. The primry dvntge of serum ntigen-specific IgE ssy over elimintion-chllenge trils is convenience. The principl dvntge of GFST testing is tht the response of the gstrointestinl mucos to severl food llergens cn be directly nd simultneously observed. An importnt disdvntge of serum ntigen-specific IgE ssy (nd most likely GFST) is tht these tests focus on the dignosis of just type of food sensitivity immedite (type ) hypersensitivity. Therefore, the effectiveness of these tests s screening tests for the entire rnge of dverse rections to foods requires investigtion.

2 8 Guilford et l The principl objective of this study ws to investigte the prevlence of dverse rections to foods in cts with chronic idiopthic gstrointestinl problems nd to identify the foods or food ingredients to which the cts were sensitive. In ddition, we imed to chrcterize the clinicl fetures of gstrointestinl food sensitivity in cts nd to evlute the concordnce of serum ntigen-specific IgE ssy nd GFST with the elimintion-chllenge tests by which the dignosis ws mde. Mterils nd Methods Cts Seventy cts tht were presented to the Mssey University Veterinry Teching Hospitl (MUVTH) with chronic gstrointestinl signs underwent thorough dignostic investigtion (see below) to determine their suitbility for entry into the study. Thirty-seven cts were femle, nd cts were mle. All of the cts hd dirrhe or vomiting for more thn 2 weeks (6 cts with vomiting, 22 cts with dirrhe, nd 2 cts with both vomiting nd dirrhe). The cts showed vriety of other clinicl signs, including weight loss (40 cts), norexi (22 cts), fltulence (8 cts), nd bdominl bloting (8 cts). Of the 70 cts, 5 were dignosed with renl disese, liver disese, hyperthyroidism, prsitism, infection with feline leukemi virus (FeLV) or feline immunodeficiency virus (FIV), obstruction of the gstrointestinl trct, neoplsi, or infectious gstroenteritis nd were eliminted from the study t vrious stges of the dignostic workup. Fifty-five cts completed workups without identifiction of specific dignosis to explin their clinicl signs. These cts were dignosed s hving chronic idiopthic gstrointestinl problems nd were entered into the study. The gstrointestinl biopsy specimens of the mjority of these 55 cts were subjectively ssessed to hve mild to moderte increses in the number of lymphocytes, plsm cells, or eosinophils in the lmin propri (ie, inflmmtory bowel disese), but other cts hd no evidence of incresed numbers of mucosl inflmmtory cells. Gstrointestinl Workup Tble. GFST. Solution No Food dditive nd histmine solutions used in Additive Sodium nitrite Red oxide Titnium oxide Gur gum Petset b Liquid crmel Histmine Percentge Mixture 0 mg in 00 ml bse 00 mg in 00 ml bse 200 mg in 00 ml bse 00 mg in 00 ml bse 600 mg in 00 ml bse 200 mg in 00 ml bse 27.5 mg histmine phosphte in 00 ml bse GFST, gstroscopic food sensitivity testing. Bse sterile sline with 0.2% phenol. b Petset crgeenn : locus bengum : potssium chloride 40 : 40 : 20 (lso clled crob gum). Stndrdized ptient history forms were completed to determine the clinicl signs, the stple diet, the temporl reltionship of the clinicl signs to food intke (if ny), nd the time between bouts (if the clinicl signs were episodic). A complete physicl exmintion ws performed nd the results recorded on stndrdized physicl exmintion forms to ensure consistency in exmintion. A dtbse ws collected consisting of complete blood count, serum chemistry profile, FeLV nd FIV test, urinlysis, nd t lest 2 zinc nd sugr fecl flottions for Girdi nd nemtode prsites, respectively. A rdiopque mrker b study ws performed on ll vomiting cts to rule out prtil obstructions of the bowel nd to evlute gstrointestinl motility. Totl serum thyroxine concentrtion ws mesured in ll cts older thn 6 yers. Cts with dirrhe dditionlly underwent rectl scrping for cytology nd cid-fst stin of the feces to detect Cryptosporidium spp. All cts underwent gstroduodenoscopy, during which biopsy smples were obtined from the stomch nd duodenum for histopthology nd duodenl fluid ws spirted for quntittive erobic nd nerobic culture. In ddition, incisionl biopsy ws obtined from the rectl mucos of ll cts, nd colonoscopy nd biopsy were performed on those cts with history of blood or mucus in their feces. All endoscopic biopsy smples were pinch biopsy specimens, nd 8 to 2 biopsy smples were collected from ech region of the gstrointestinl trct exmined. The biopsy specimens were viewed by the pthologist on duty t the time of the dignostic workup. All biopsy smples subsequently were pooled nd reviewed by pthologist (D.G.A.) with subjective grding system bsed on the number nd type of inflmmtory cells in the lmin propri. The degree of mucosl inflmmtion ws clssified s bsent, mild, moderte, or severe. The pthologist ws unwre of whether the cts hd been dignosed s food sensitive. Serum IgE Assy Serum smples were tken t dmission from ll cts nd stored t 20 C for mximum of 6 months before nlysis by ntigen-specific enzyme-linked immunodsorbent ssy by commercil lbortory. Cts with food-specific IgE concentrtions greter thn 200% of control ser were considered positive. Gstroscopic Food Sensitivity Tests All cts underwent GFST ccording to previously described technique. The stple diets of the cts were withheld for t lest dys before the GFST in the hope of incresing the sensitivity of the test. The procedure ws performed fter induction of nesthesi with tiletmine-zolzepm c nd mintennce with hlothne. All cts were tested with 0.5-mL volumes of 5 diluted food protein extrcts, d dilute histmine solution, nd 4 solutions of food dditives (Tble ). Four of the food proteins (milk, beef, lmb, nd chicken) were tested in ll cts. Milk, beef, nd lmb proteins re dietry stples for most New Zelnd cts. Chicken ws used s the negtive control protein becuse it is infrequently fed s dietry stple in New Zelnd. The 5th food protein ws vried ccording to the dietry history of the individul ct under study. Most commonly, this protein ws whet, corn, or mixed-fish extrct. d The food extrcts were pplied t finl concentrtion of 5,000 protein nitrogen units (PNUs) per milliliter with the exception of whet nd corn extrcts, which were pplied t concentrtions of 5,000 PNU/mL. The dditives tested were those included frequently in commercil ct foods in New Zelnd. The dditives were donted by locl pet food mnufcturer nd were diluted in sline to obtin the concentrtions used by this mnufcturer in their products. All cts were tested with the sme dditives. Some of the dditives were tested in combintion to minimize the nesthesi time required to test the dditives. Dietry Elimintion-Chllenge Tril to Dignose Food Sensitivity After the dignostic procedures were completed, the cts were fed sufficient food to stisfy their ppetites with commercilly vilble chicken- or venison-bsed selected-protein diet e,f for minimum of 4 weeks. The choice between these diets ws bsed on the dietry history of the individul ct. Most cts were plced on the chicken-bsed diet, but ny cts tht hd eten chicken within the previous 6 months were fed the venison-bsed diet. The owners were given the option of

3 Food Sensitivity in Cts 9 tking the ct home or leving it in the hospitl for this phse of the study. The mjority of owners (90%) took their cts home. Owners were wrned of the importnce of feeding the elimintion diet exclusively. If the owner hd more thn ct, the owner ws instructed to feed ll of the cts the elimintion diet. If the ct hd ccess to the outdoors, the owners were sked to visit their neighbors to sk them not to feed the ct. During the elimintion tril, the owner ws sked to record on dily bsis the ct s food intke nd clinicl signs on stndrdized logs. The clinicl signs recorded by the owners included the number of times the ct vomited nd the grde of the feces s compred with photogrphic fecl grding chrt. The 5-point chrt grded the fecl consistency from firm stools (grde 5) to liquid dirrhe (grde ). Those cts with clinicl signs tht did not become reduced in frequency or severity fter 4 weeks were considered not to be suffering from food sensitivity. Cts with clinicl signs tht completely resolved on the elimintion diet were returned to their stple diet for 4 to 7 dys. If more thn food hd formed prt of ct s stple diet, the ct ws chllenged with ech of these foods for 4-dy period followed by -dy wshout period, during which the ct ws fed the elimintion diet before the next diet chllenge. The 4-dy chllenge period ws considered the minimum desirble but ws chosen to llow new food to be tested on weekly bsis. The wshout period minimized the chnce of misdignosis cused by delyed food sensitivity rection ppering during the subsequent week of testing. These chllenge tests were performed either t home or in the hospitl, depending on the owner s preference. A ct ws considered to hve recrudescence of its clinicl signs if fecl consistency deteriorted to the grde ssigned before the elimintion tril or if the vomiting incresed to frequency similr to tht recorded before the elimintion tril. Cts were gin fed the elimintion diet for 2 to 4 weeks until signs resolved. Cts were dignosed s food-sensitive if their clinicl signs resolved when they were fed the elimintion diet, recrudesced when they were chllenged with their stple diet, nd then resolved for 2nd time when they were returned to the elimintion diet. Dietry Elimintion-Chllenge Trils to Identify the Responsible Food Ingredient The owners of 2 of the cts dignosed s food-sensitive greed to dmit their cts to the MUVTH for more detiled dietry chllenge studies, with the im of identifying the specific food ingredient to which the ct ws sensitive. Dietry chllenge studies were undertken by dding smll quntity of food protein (minimum 0.6 grms of protein per kilogrm body weight) or food dditives to the elimintion diet on weekly bsis. The food proteins chosen were those tht formed the mjority of the ct s diet in the previous months. Most cts were tested with lmb nd beef (50 g diced fresh met per dy), whet (25 g cooked cerel per dy), cnned viscer (50 g of sheep nd beef lung, testicles, kidney, nd liver minced on n equl weight bsis), nd milk (5 g of whole milk powder in slurry). Up to 2 other protein sources were tested, depending on the owner s willingness to ccept prolongtion of the hospitliztion period. The other protein sources (dependent on diet history) included tun (50 g), whitefish (50 g), pork (50 g), corn gluten (5 g), ots or brley (25 g), or egg ( whole egg cooked nd diced into diet). The food dditives chosen were the sme s those used in the GFST. The dditives were mixed in combintion into the elimintion diet t concentrtions commonly found in New Zelnd commercil pet foods (Tble 2). If n dverse rection to the combintion of dditives ws observed, cts then underwent series of chllenges with the individul dditives dded to the elimintion diet. For ll chllenge trils with food ingredients, 4-dy chllenge period nd -dy wshout period were used. Both the food proteins nd the food dditives were tested in rndom order. Clinicl signs evluted during in-hospitl chllenge trils were the sme s those evluted by the owners (vomiting frequency nd Tble 2. Food dditive nd histmine doses used in dietry chllenge studies. Additive Sodium nitrite Red oxide Titnium oxide Gur gum Petset Liquid crmel Histmine Dose 2 mg 00 mg 600 mg 00 mg,800 mg 600 mg 20 mg Petset Crgeenn : locus bengum : potssium chloride 40 : 40 : 20 (lso clled crob gum). fecl grde) but with the ddition of defection frequency (which could not be ssessed relibly in nonhospitlized indoor-outdoor cts). Results Sixteen (29%) of the 55 cts with chronic idiopthic gstrointestinl problems were dignosed s food sensitive. In ddition to the 6 food-sensitive cts, the clinicl signs of nother cts (20%) entered into the study resolved on the selected-protein diet but did not recrudesce on chllenge with their old diet. The clinicl signs of the food-sensitive cts resolved quickly on the elimintion diet. Vomiting stopped immeditely in lmost ll ffected cts, nd dirrhe resolved in most ffected cts within 2 or dys. Recrudescence of clinicl signs in the food-sensitive cts ws rpid lso, occurring within to 4 dys of chllenge with their previous diet. Commercil cnned food comprised some or ll of the diet of 9% of the food-sensitive cts nd 94% of the non food-sensitive cts. Commercil dry diets were fed to 60% of the food-sensitive cts nd 54% of the non food-sensitive cts. Six percent of the food-sensitive cts nd none of the non food-sensitive cts were fed exclusively dry diets. Tble foods occsionlly were fed to 7% of food-sensitive cts nd 67% of non food-sensitive cts. The foods or food ingredients to which the ffected cts were sensitive re listed in Tble. The most common llergens were beef, whet, nd corn gluten. One of the cts with whet sensitivity ws determined to hve trnsient sensitivity to whet gluten, which resolved fter pproximtely 6 weeks of dietry mngement with selectedprotein diet. One ct vomited immeditely fter ingestion of the mixture of food dditives but demonstrted no dverse rections to ny of the food dditives when they were fed seprtely. The ct ws rechllenged with the mixture of dditives nd immeditely vomited 2nd time. Of the 8 food-sensitive cts tht underwent elimintion-chllenge testing with multiple food ingredients, 4 (50%) were found to be sensitive to more thn food ingredient. Nine of the food-sensitive cts were femle, nd 7 were mle. Their ges rnged from 6 months to 4 yers (medin, 5 yers), with only cts being less thn yer. The mjority of cts were domestic short hirs. Their gender, ge, nd breed distributions were similr to the non foodsensitive cts in the study. Nine of the food-sensitive cts hd history of vomiting (56%), 4 hd history of dirrhe (25%), nd hd

4 0 Guilford et l Tble. Foods or food ingredients responsible for food sensitivity. Food or Ingredient Commercil dry diet Commercil cnned diet Beef Corn gluten Whet Whet gluten Brley Chicken Lmb Srdines Viscer Lctose Food dditives No. Cts 5 b b Fifty percent of the cts were multiply llergic. b The llergen in the dry or cnned foods fed to these cts ws not identified either becuse the owners refused further testing or becuse further elimintion-chllenge testing filed to identify the llergen. history of both vomiting nd dirrhe (9%). The vomiting ws reltively infrequent, occurring less thn once per dy in most cts nd up to mximum of times per dy in the most severely ffected ct. The nture of the vomitus nd its timing fter eting ws vrible. Some cts vomited food within minutes of eting, wheres others vomited food 2 hours to more thn 2 hours fter eting. In some cts, the timing of the vomiting ws seemingly unrelted to eting, nd the vomitus consisted primrily of bile. Four of the 7 food-sensitive cts (57%) with dirrhe hd evidence of lrge-bowel dysfunction (mucus nd fresh blood in the feces, excessive strining to defecte, or both). The frequency of vomiting (6%), dirrhe (4%), nd concurrent vomiting nd dirrhe (%) in the non foodsensitive cts ws similr to tht of the food-sensitive cts. Fifty-six percent of non food-sensitive cts with dirrhe hd evidence of lrge-bowel dysfunction. Weight loss nd fltulence were observed in (69%) nd 6 (8%) of the food-sensitive cts nd 5% nd 2%, respectively, of the non food-sensitive cts. The weight loss ws clssified s moderte to mrked in 50% of the food-sensitive cts. Appetite ws vribly reported s decresed, unchnged, or incresed in food-sensitive cts. Some of the food-sensitive cts with weight loss hd moderte or severe histologic bnormlities (20%) of the smll intestinl mucos, but most hd mild histologic bnormlities (40%) or no histologic bnormlities (40%). The demenor of 6 (8%) of the food-sensitive cts ws reported s irritble. Another 4 cts (25%) were reported s lethrgic. Collectively, therefore, ltered demenor ws reported in 0 (6%) food-sensitive cts. Altered demenor ws noted in 45% of the non food-sensitive cts. Concurrent dermtologicl disese ws observed in 4 cts (25%) with gstrointestinl problems cused by food sensitivity. These cts suffered from miliry dermtitis, pruritus, nd lopeci. The dermtologicl signs were cused by food sensitivity in of these 4 cts. In contrst, concurrent dermtologicl signs were reported in 5% of the non food-sensitive cts. Of totl of 0 cts presented with combintion of dermtologicl nd gstrointestinl disese, 4 of these cts (40%) were dignosed s food sensitive. The durtion of clinicl signs in the food-sensitive cts rnged from to 70 months (medin, 7.5 months). The clinicl signs were episodic in 4 (25%) of the cts, with n intervl free of clinicl signs for up to months. A similr percentge of non food-sensitive cts hd episodic clinicl signs, but in these cts longer intervls free of clinicl signs were reported (s long s yer). No consistent bnormlities were detected in the CBC nd serum chemistry profiles of the food-sensitive cts. However, 29% of the food-sensitive cts hd eosinophili. The eosinophil counts rnged from 0.07 to per liter (medin, per liter). A smller percentge (0%) of non food-sensitive cts hd eosinophili. In these cts, the eosinophil counts rnged from 0. to per liter (medin, per liter). Fourteen percent of the food-sensitive cts hd lymphopeni. The lymphocyte counts rnged from.6 to per liter (medin, per liter). Twenty-four percent of the non foodsensitive cts hd lymphopeni. In these cts, the lymphocyte counts rnged from 0.57 to per liter (medin, per liter). Serum ntigen-specific IgE test results were vilble for 2 of the food-sensitive cts. Of these 2 cts, 7 cts (58%) hd one or more positive test results. Two cts with dverse rections to corn gluten nd ct with n dverse rection to dry food contining corn hd positive serum tests for corn-specific IgE. Two food-sensitive cts (7%) hd positive serum IgE tests to ntigens to which they showed no dverse rections. Three food-sensitive cts (25%) hd positive serum IgE tests to ntigens to which they were not tested but to which they were unlikely to be sensitive becuse the ntigens were novel proteins to which these cts were unlikely to hve been exposed. The remining 5 foodsensitive cts hd no positive serum ntigen-specific IgE tests. Serum ntigen-specific IgE tests were vilble for 24 of the non food-sensitive cts. Of these 24 non food-sensitive cts, 6 (25%) hd or more positive tests. Eighty-three percent of these positive results were to soy ( cts), rice (4 cts), nd corn ( cts) or some combintion of these foods. One ct hd positive test to penut nd nother to potto. The lbels of the stple diets of the non food-sensitive cts with positive serum ntigen-specific IgE tests did not list rice, soy, penut, or potto. However, trnsient exposure to these food ntigens could not be discounted becuse ll of these cts hd very vried diets, including tble foods nd cnned nd dry foods purchsed from supermrkets nd veterinry prctices. Gstroscopic food sensitivity testing ws not helpful. Only one suspicious rection (swelling without erythem) ws observed in ll of the procedures performed. This rection occurred fter the ppliction of gum mixture (solution ; Tble ) nd ws not confirmed by positive dietry chllenge to gums. All of the food-sensitive cts hd histologicl chnges in t lest region of the bowel. However, the region of the gstrointestinl trct ffected ws not consistent, nd the histologicl chnges were nonspecific nd were similr in

5 Food Sensitivity in Cts frequency nd severity in food-sensitive nd non food-sensitive cts. One food-sensitive ct hd subcute diffuse suppurtive glossitis with focl ulcertion. Gstric mucosl biopsy specimens were bnorml in two thirds of the foodsensitive cts. One ct hd severe lymphocytic gstroenteropthy chrcterized by diffuse infiltrtion of lymphocytes throughout the gstric lmin propri long with clumps of plsm cells. In some res, invsion nd oblitertion of glnds ws observed. Nine other food-sensitive cts hd mild histologicl chnges in their gstric mucos, including smll numbers of eosinophils migrting through the glndulr epithelium, scttered foci of lymphocytes, ptchy subepithelil edem or hemorrhge, or mild fibroplsi. Lymphoid nodules were observed in the gstric mucos of 2 food-sensitive cts. The duodenl mucosl biopsy specimens were bnorml in 50% of the food-sensitive cts. On most occsions, the pthologic dignosis ws mild lymphocytic-plsmcytic enteritis. Two food-sensitive cts hd modertely severe lymphocytic-plsmcytic enteritis, nd 2 hd moderte to severe eosinophilic enteritis. In the most severely ffected ct with eosinophilic enteritis, the villi in ll sections were blunt nd often showed fusion. At the tips of the villi, epithelil cells often were vcuolted. Some of the duodenl biopsy specimens of the food-sensitive cts hd vriety of other fetures. These included smll numbers of neutrophils, occsionl globulr leukocytes, evidence of edem, nd foci of hemorrhge, mild fibrosis, or hyperplsi of glnds in the bsl mucos. The colonic or rectl mucosl biopsy specimens were bnorml in two thirds of the food-sensitive cts. The pthologicl dignosis ws usully mild colitis or proctitis. On most occsions, the findings included mild lymphocytic-plsmcytic infiltrtion. Attenution of the epithelium occsionlly ws reported, s ws the presence of smll numbers of neutrophils, eosinophils or globulr leukocytes, mild fibrosis, or mucus distension of crypts. Discussion The results of this study suggest tht lmost one third of cts presented to referrl hospitl with chronic idiopthic gstrointestinl problems hve food sensitivity. It is importnt to emphsize tht this high prevlence of food sensitivity ws observed in highly selected popultion of cts tht hd undergone extensive dignostic effort to rule out other common cuses of gstrointestinl dysfunction. The popultion studied included mny cts tht stisfied the current definition of inflmmtory bowel disese nd other cts tht did not stisfy this definition, either becuse they were dignosed s food sensitive or becuse they were considered to hve norml numbers of mucosl inflmmtory cells. 2 In the present study, the dignosis of food sensitivity ws bsed on dietry elimintion-chllenge trils. Of note ws the quick resolution of clinicl signs in the food-sensitive cts during the dietry elimintion trils. These results, together with the high prevlence of food sensitivity detected in the present study, suggest tht the durtion of the elimintion diet in cts with gstrointestinl disese need not exceed 4 dys. In contrst, veterinry dermtologists often recommend elimintion diets for minimum durtion of 8 weeks to dignose the dermtologicl mnifesttions of food sensitivity.,,4 Rpid recrudescence of signs fter rechllenge of food-sensitive cts with the responsible llergen ws seen in the present study. In studies of cts with dermtologicl problems, most cts recrudesce within to 5 dys, but some took up to 0 dys.,5 This observtion rises the possibility tht longer period of rechllenge in the present study my hve incresed the number of dignoses of food sensitivity. Countering this possibility, however, is tht ll of the food-sensitive cts in the present study first showed recrudescence within dys, nd none required the full 4 to 7 dy chllenge period llotted. The resons for these pprent differences in behvior between skin nd gstrointestinl food sensitivities re unknown. In ddition to the food-sensitive cts, the clinicl signs of nother 20% of cts entering the study resolved on the selected-protein diet but did not recrudesce on chllenge with their stple diets. If the dignosis of food sensitivity hd not depended on observing recrudescence of signs fter rechllenge with their stple diets, the number of cts dignosed s food sensitive would hve lmost doubled in the present study. It is uncler why the clinicl signs of so mny cts resolved on the selected-protein diet nd then did not recrudesce fter rechllenge. The dietry histories of some of these cts my hve been incomplete, nd they my not hve been chllenged with the food to which they were sensitive. Alterntively, some of the cts my hve spontneously recovered from their gstrointestinl disorder, nd the recovery of others my hve been fcilitted by other therpeutic spects of the selected-protein diet, such s high digestibility. Collectively, 50% of the cts fed the selected-protein diets hd resolution of their clinicl signs. This observtion suggests tht selected-protein diets should be considered n importnt prt of the mngement of cts with idiopthic gstrointestinl problems. The foods or food ingredients responsible for the clinicl signs usully were dietry stples of the ffected cts. Beef, whet, nd corn were common llergens. Beef lso ws common llergen in nother recent study of food-sensitive cts nd frequently is incriminted in dogs. 4 In other studies of food-sensitive cts, the foods most often incriminted were fish nd diry products. 8,6 To our knowledge, the dverse rections to gluten, viscer, nd the mixture of dditives observed in the present study re the first documented reports of these food sensitivities in cts. The high percentge of cts in the present study tht were sensitive to more thn food ingredient contrsts with reports of nimls with dermtologicl problems, in which rections to multiple llergens re considered unusul. 4 No clinicl fetures llowed differentition of food-sensitive from non food-sensitive cts without the use of the elimintion-chllenge tril. The signlment ws diverse, the vomiting pttern not helpful, nd the chrcteristics of the dirrhe were vrible. Of note ws tht history of lrgebowel dirrhe ws more common thn history of smllbowel dirrhe in the food-sensitive cts. Lymphocyticplsmcytic colitis in cts previously hs been observed to be food responsive, 7 nd lrge-bowel dirrhe lso hs been described s n importnt feture of dogs with food sensitivity. 8 This predilection for lrge-bowel disese contrsts with the typicl clinicl picture in gluten enteropthy

6 2 Guilford et l of humns, in which the upper smll intestine is most severely ffected nd suggests tht the concentrtion of ntigen is not the primry determinnt of the severity of the mucosl histologicl chnges in ct food sensitivity. The high frequency of weight loss in the food-sensitive cts ws unexpected nd without cler explntion in some of the cts. The ppetite of some of the cts with weight loss ws reduced, but most te well. The smll-bowel dirrhe nd fltulence observed in some cts indicted mlbsorption, but most cts with weight loss did not hve these signs, nd mny hd few histologicl bnormlities in their smll intestinl mucos. Weight loss is lso prominent clinicl feture of mild gluten enteropthy in Irish Setters. 9 Enteric protein loss hs been recognized during intestinl nphylxis to foods in lbortory nimls. 5 Chronic enteric protein loss is energeticlly demnding nd my explin the weight loss observed in the food-sensitive cts nd dogs. The clinicl feture most suggestive of food sensitivity ws the concurrent occurrence of gstrointestinl nd dermtologicl signs. Others lso hve noted n ssocition between dermtologicl nd gstroenteric diseses in cts with food sensitivity. Twenty-nine percent of the food-sensitive cts in recent study hd both pruritus nd gstrointestinl signs. The concomitnt ppernce of cutneous nd gstrointestinl signs in food-sensitive ptients lso ws noted by Wlton. 6 A high frequency of simultneous cutneous nd gstrointestinl signs hs been reported in one study of dogs with food sensitivity, 8 but other studies hve not described this ssocition. 4 White hs pointed out tht this discrepncy my be prtilly cused by filure of the owner to detect or report the gstrointestinl signs nd lso emphsizes the importnce of thorough medicl history with respect to both gstrointestinl nd dermtologicl signs. The concurrent ppernce of cutneous nd gstrointestinl signs is not pthognomic, however, for food sensitivity. In the present study, some non food-sensitive cts hd gstrointestinl disese in ssocition with fle llergic or idiopthic pruritus. The routine lbortory work did not llow differentition of food-sensitive nd non food-sensitive cts. Peripherl eosinophili ws present in fewer thn one third of the food-sensitive cts nd lso ws observed in non food-sensitive cts. The dignostic vlue of in vitro tests for food-specific IgE ntibodies vries widely in studies of humns 0 nd hs been disppointing in dogs. 4 Severl observtions in the present study cll into question the vlue of the serum ntigen-specific IgE ssy s screening test for food sensitivity in cts. These include the observtions tht s mny s 25% of the non food-sensitive cts hd positive tests, only 58% of the food-sensitive cts hd positive tests, nd only 25% of the food-sensitive cts showed concordnce between positive ntigen-specific IgE nd positive orl chllenge test. The lck of vlue of ntigen-specific IgE s screening test for gstrointestinl food sensitivity is not surprising. It is unresonble to expect mesurement of serum ntigen-specific IgE concentrtion to be effective for this purpose when there is strong evidence tht mny gstrointestinl food sensitivities re food intolernces tht re not medited by IgE. 5,20,2 It is perhps more pproprite for clinicins to use serum ntigen-specific IgE tests to ssist them in determining whether nimls suffering from dverse rections to foods re ffected by type I hypersensitivity rections. The results of the present study suggest tht type I hypersensitivities ccount for only 25% of gstrointestinl food sensitivities in cts. Cution must be exercised, however, in interpreting the serum IgE results of the present study becuse, to the uthors knowledge, vlidtion of this commercil ssy hs not been reported. An erlier publiction, compring the results of intrderml skin tests nd serum ntigen-specific IgE in cts showed poor correltion. 22 Other thn the observtion tht the positive serum ntigen-specific IgE tests were twice s common in the food-sensitive cts thn the non food-sensitive cts, there ws little evidence to support the dignostic ccurcy of the ssy. The observtion tht 7% of the food-sensitive cts nd 25% of the non foodsensitive cts hd positive serum IgE tests to ntigens to which they showed no cliniclly evident dverse rections indictes the presence of subclinicl food sensitivity, symptomtic sensitiztion, or flse-positive test results. 0 Gstroscopic food sensitivity testing hs been successfully used to detect food sensitivity in dogs,2 nd humns 9 but ws not helpful in detecting food sensitivity in cts in this study. The reson for this pprent species difference is uncler. Possibilities include difference in gstric mucosl permebility to mcromolecules, differences in the behvior of mucosl mst cells, or difference in the pthogenesis of food sensitivity mong species. The modifiction of GFST proposed by Ermel et l, 24 in which the food llergen is injected into the mucos rther thn dripped onto it, my provide better results in cts. Unfortuntely, however, in one uthor s experience this method in dogs cretes difficulties in seprting the cute phse rection to mucosl injury from tht induced by hypersensitivity (Guilford, personl communiction). No histologicl fetures distinguished food-sensitive from non food-sensitive cts in the present study. The stomch nd rectum were the gstrointestinl sites tht were most frequently considered bnorml in the food-sensitive cts but, s with humns nd lbortory nimls, 20,25 27 ll levels of the gstrointestinl trct cn be ffected by food sensitivity. Eosinophilic infiltrtes were observed in the mucos of some food-sensitive cts, but the mjority of cts did not hve excessive mucosl eosinophilic inflmmtion. The severity of the mucosl chnges ws lso not helpful in differentiting food-sensitive from non food-sensitive cts. Some food-sensitive cts hd no detectble mucosl bnormlities, nd others hd severe chnges. Idiopthic chronic gstritis, lymphocytic-plsmcytic enteritis, nd chronic colitis currently cnnot be differentited from food sensitivity on the bsis of histologicl exmintion of single set of biopsy specimens. At times, the lymphocytic infiltrte in the intestinl mucos of food-sensitive cts cn be so intense s to mimic lymphosrcom. 28 The diverse histologicl findings in the gstrointestinl trct of food-sensitive cts observed in the present study hve been described previously. Eosinophils hve been found in rectl scrpings from food-sensitive cts. Lymphocytic-plsmcytic colitis hs been ttributed previously to food sensitivity in cts. 7 The smll intestine of ct llergic to milk showed congestion, edem, villous degen-

7 Food Sensitivity in Cts ertion, hemorrhge, nd n increse in the number of plsm cells fter 4 dys of milk chllenge. 29 The cute phse rections observed in the stomchs of food-sensitive dogs re chrcterized by congestion nd edem. 24 In food-sensitive humns nd lbortory nimls, the mucos cn be microscopiclly norml or cn show mucosl edem, seprtion of the epithelium from the lmin propri, ft ccumultion in the epithelium, sloughing of villus tip epithelium, incresed mucus secretion, incresed intrepithelil lymphocytes, infiltrtion of the lmin propri with lymphocytes nd plsm cells, nd villus trophy. 20,25 27,0, The histologicl fetures of food protein induced enteropthies usully re ptchy nd the villus trophy prtil rther thn complete. 26 In conclusion, dverse rections to dietry stples re common in cts with chronic gstrointestinl problems nd cn be successfully mnged by feeding selected-protein diets. Dignosis requires dietry elimintion-chllenge trils nd cnnot be mde on the bsis of clinicl signs, routine lbortory work, serum ntigen-specific IgE ssy, gstroscopic food sensitivity testing, or histologicl exmintion of single set of gstrointestinl biopsy specimens. Footnotes Bio-Medicl Services, Austin, TX b BIPS, Med ID, Grnd Rpids, MI c Zoletil, TechVet, Aucklnd, New Zelnd d Greer Lbortories, Lenoir, NC e WHISKAS Feline Selected Protein Diet (Wlthm), Msterfoods, Bruck, Austri (chicken nd rice) f WALTHAM Veterinry Diet Feline Selected Protein Diet, Effem Foods, Bolton, Cnd (venison nd rice) Acknowledgments The uthors would like to thnk the mny veterinrins who referred cts for the study, in prticulr Drs Pru Gllowy nd Sturt Burroughs. We would lso like to cknowledge the contribution of Jo Wills of the Wlthm Centre, pthologists Drs Murice Alley nd Michelle Cooke, nd the technicl ssistnce of the clinicl stff of the Mssey University Veterinry Teching Hospitl. The work ws performed t Mssey University nd supported by grnt from the Wlthm Centre for Pet Nutrition, Wlthm-on-the-Wolds, UK. References. Anderson JA, Sogu DD, eds. Adverse Rections to Foods. Ntionl Institute for Helth Publiction No , July Denis S, Prdis M. Food llergies in dogs nd cts. Prt 2: Retrospective study. Med Vet Quebec 994;24: Guguere E. Food intolernce in cts with cutneous mnifesttions: A review of 7 cses. Eur J Compnion Anim Prct 995;5: Hll EJ. Gstrointestinl spects of food llergy: A review. J Smll Anim Prct 994;5: Guilford WG. Adverse rections to foods. In: Strombeck DR, Guilford WG, eds. Smll Animl Gstroenterology, rd ed. Phildelphi, PA: WB Sunders; 996: Anderson JA. Non-immunologiclly-medited food sensitivity. Nutr Rev 984;42: Stogdle L, Bomzon L, vn den Berg PB. Food llergy in cts. J Am Anim Hosp Assoc 982;8: White SD, Sequoi D. Food hypersensitivity in cts: 4 cses ( ). J Am Vet Med Assoc 989;94: Reimnn HJ, Ring J, Ultsch B, Wendt P. Intrgstrl provoction under endoscopic control (IPEC) in food llergy: Mst cell nd histmine chnges in gstric mucos. Clin Allergy 985;5: Ownby DR. In vitro ssys for the evlution of immunologic rections to foods. Immunol Allergy Clin North Am 99;: Guilford WG, Strombeck DR, Rogers Q, et l. Development of gstroscopic food sensitivity testing in dogs. J Vet Intern Med 994; 8: Guilford WG. Idiopthic inflmmtory bowel diseses. In: Strombeck DR, Guilford WG, eds. Smll Animl Gstroenterology, rd ed. Phildelphi, PA: WB Sunders; 996: White SD. Food llergy in dogs. Compend Contin Educ Prct Vet 998;20: Rosser EJ. Dignosis of food llergy in dogs. J Am Vet Med Assoc 99;20: Sous CA. Exudtive, crusting nd scling dermtoses. Vet Clin North Am 995;25: Wlton GS. Skin responses in the dog nd ct to ingested llergens: Observtions on one hundred confirmed cses. Vet Rec 967; 8: Nelson RW, Dimperio ME, Long GG. Lymphocytic-plsmcytic colitis in the ct. J Am Vet Med Assoc 984;84: Pterson S. Food sensitivity in 20 dogs with skin nd gstrointestinl signs. J Smll Anim Prct 995;6: Btt RM, Hll EJ. Chronic enteropthies in the dog. J Smll Anim Prct 989;0: Gryboski JD. Gstrointestinl spects of cow s milk protein intolernce nd llergy. Immunol Allergy Clin North Am 99;: Heymn MB. Food sensitivity nd eosinophilic gstroenteropthies. In: Sleisinger MH, Fordtrn JS, eds. Gstrointestinl Disese, 4th ed. Phildelphi, PA: WB Sunders; 989: Foster AP, O Dir H. Allergy testing for skin disese in the ct in vivo versus in vitro tests. Vet Dermtol 99;4: Elwood CM, Rutgers HC, Btt RM. Gstroscopic food sensitivity testing in 7 dogs. J Smll Anim Prct 994;5: Ermel RW, Kock M, Griffey SM, et l. The topic dog: A model of food llergy. Lb Anim Sci 997;47: Proujnsky R, Winter HS, Wlker WA. Gstrointestinl syndromes ssocited with food sensitivity. Adv Peditr 988;5: Ptrick MK, Gll DG. Protein intolernce nd immunocyte nd enterocyte interction. Peditr Clin North Am 988;5: Smpson HA. Immunologic mechnisms in dverse rections to foods. Immunol Allergy Clin North Am 99;: Wsmer ML, Willrd MD, Helmn RG, Edwrds JF. Food intolernce mimicking limentry lymphosrcom. J Am Anim Hosp Assoc 995;: Wlton GS, Prish WE, Coombs RAA. Spontneous llergic dermtitis nd enteritis in ct. Vet Rec 968;8: Smpson HA, Buckley RH, Metclfe DD. Food llergy. J Am Med Assoc 987;258: Curtis GH, Ptrick MK, Ctto-Smith AG, Gll DG. Intestinl nphylxis in the rt. Effect of chronic ntigen exposure. Gstroenterology 990;98: