Article — Artikel

The effect of high broiler litter diets as survival ration on the health of sheep

D T Mavimbelaa, J B J van Ryssena* and R Lastb

INTRODUCTIONDrought occurs frequently in southern

Africa and devastating veld fires oftendestroy vast areas of grassland during thedry winter periods. Consequently, live-stock farmers often have to resort to emer-gency feeding measures to sustain theiranimals, utilising whatever feed is avail-able. Such conditions have promptedfarmers to resort to feeding poultrymanure/litter10 as the sole feed for theirlivestock.

Although poultry manure/litter hasbeen used successfully as a feed supple-ment in ruminant diets6,14, many potentialproblems exist. Problems of nutrientimbalances and the effects of potentiallytoxic components such as pathogens,dietary copper (Cu), medicinal com-pounds, mycotoxins, ionophores andbotulinum toxin9,10 will be accentuatedwhen the product is fed as a sole feed.Silanikove and Tiomkin18 reported liverdamage in beef cows when overwinteredon high poultry-litter diets in Israel. The

authors attributed the damage to highruminal ammonia concentrations due tofermentation of the litter in the rumen. InSouth Africa, this report of liver damageseems to have prompted warningsagainst the use of poultry litter as aruminant feed at any rate of inclusion.However, liver damage is not usuallyobserved when poultry litter is fed, exceptwhen it contains high concentrations ofCu17. The objective of the present studywas to investigate the effect of high levelsof broiler litter on the health of sheep.Since broiler litter contains excessivelevels of non-protein nitrogen, improvingthe dietary nitrogen (N) to energy balanceto improve the product as a survival feedwas evaluated as well.

MATERIALS AND METHODS

Experimental procedureEighteen South African mutton Merino

wethers (ca. 2 years old and ca. 42 kg bodymass) were assigned randomly to 3 treat-ments, namely pure broiler litter andbroiler litter mixed with 7.5 or 15.0 %molasses. Sun-dried broiler litter withwood shavings as bedding material wasused. The litter was sifted to removelumps and foreign material. Before theonset of the trial the sheep were vacci-nated twice against botulism and dosedwith a broad-spectrum anthelmintic.

During a 14-day adaptation period thesheep received broiler litter and grass hay.The sheep were fed individually inslatted-floor feeding pens with free accessto water. Feed refusal was measuredweekly. The sheep were weighed once amonth. Feed samples were collectedthroughout the trial for chemical analysis.Blood samples collected by venipuncturein heparinised vacutainers and spot urinesamples were obtained 4 times during theexperimental period. A spot urine samplewas collected in a clean plastic bag tied tothe stomach of the wether. The urine waspreserved in 10 % HCl.

After an average of 83 days the wetherswere slaughtered over a period of 10 days.The warm carcasses, livers and kidneyswere weighed. Liver, heart and kidneysamples were taken and preserved in10 % buffered formalin for histopatho-logical examination. Liver samples werealso dried at 80 °C for Cu analysis. Backfatthickness at the 13th rib was measured onthe cold carcass.

Laboratory analysisThe concentrations of haemoglobin

(Hb) in whole blood, glucose and bilirubinin plasma, and the activities of thefollowing plasma enzymes: aspartateaminotransferase (AST, EC 2.6.1.1),creatine kinase (CK, EC 2.7.3.2), alanineaminotransferase (ALT, EC 2.6.1.2),δ-glutamyltransferase (δGT, EC 2.3.2.2)and sorbitol dehydrogenase (SDH, EC1.1.1.14) were measured using BoehringerMannheim analytical kits (BoehringerMannheim GmbH Diagnostica, Germany).An analytical kit was used to determinethe concentration of β-hydroxybutyratein plasma (Sigma Diagnostics). Packedcell volume (PCV) in whole blood wasdetermined by microhaematocrit. Themethod described by De Villiers et al.8

was used to determine free fatty acidconcentration in plasma, that of Tietz19 forplasma albumin and the Bertelot method2

for plasma urea N concentrations. Totalallantoin concentration in urine wasmeasured5 to estimate urinary purineconcentration. The creatinine concentra-tion in urine was read on an auto-analyserusing the alkaline picrate method.

Atomic absorption spectrophotometry

ABSTRACTThe use of broiler litter as an emergency feed during droughts and other periods of feedshortages was evaluated in terms of its effect on the health of sheep. Pure broiler litter(i.e. excreta plus wood shavings), and litter mixed with 7.5 or 15 % molasses were fed for 83days to 2-year-old wethers. The addition of molasses to the litter caused a significantincrease (p < 0.01) in feed intake and final body mass. Various parameters such as plasmaenzyme activity, plasma metabolite concentrations and urine mineral and purine excretiondid not differ among treatments. Histological evaluation revealed no liver or kidneypathology. Mild myocardial pathology was observed in all 3 treatment groups. This seemsto be related to the presence of the ionophore-based coccidiostat, narasin, which waspresent in the litter at a concentration of 10 mg/kg. It is suggested that these histologicallesions are of little practical significance and would not affect the health of sheep being fedthe broiler litter as a survival feed.

Key words: broiler litter, drought feeding, ionophore, myocardial pathology, sheep.

Mavimbela D T, Van Ryssen J B J, Last R The effect of high broiler litter diets as survivalration on the health of sheep. Journal of the South African Veterinary Association (1997) 68(4):121–124 (En.). Department of Animal Science and Poultry Science, University of Natal,Pietermaritzburg, 3209 South Africa.

aDepartment of Animal Science and Poultry Science,University of Natal, Pietermaritzburg, 3209 South Africa.

bVeterinary Pathology Services, PO Box 13624,Cascades, Pietermaritzburg, 3202 South Africa.

*Present address: Department of Animal and WildlifeSciences, University of Pretoria, Pretoria, 0002 SouthAfrica.

Received: February 1996. Accepted: September 1997.

0038-2809 Jl S.Afr.vet.Ass. (1997) 68(4): 121–124 121

was used to determine the calcium (Ca),Cu, magnesium (Mg), sodium (Na),potassium (K), manganese (Mn) and zinc(Zn) concentrations in the respectivesamples. The fluorometric method of Kohand Benson15 was used to determineselenium concentrations. Crude protein,ash, phosphorus (P), ether extract, neutraldetergent fibre and acid detergent fibreconcentrations in each sample wereobtained using standard laboratorytechniques. The litter was screened withthin-layer chromatography for thepresence of ionophore antibiotics andquantified colorimetrically11.

PathologyTissues were fixed in 10 % formalin,

sectioned and stained with haematoxylinand eosin according to standard methods.Myocardial lesions were scored accordingto a histological classification systemapplied by Bastianello et al.3.

Statistical analysisData were subjected to analysis of

variance and correlation coefficientswere calculated using Minitab StatisticalSoftware (Minitab, State College,Pennsylvania).

RESULTSThe dry matter (DM) content of the pure

litter was 923 g/kg, and the addition ofmolasses decreased it to 917 and 870 g/kgfor the 92.5 and 85 % litter treatmentsrespectively. The chemical composition ofthe experimental diets is presented inTable 1. The crude protein concentrationof the pure litter was 182 g/kg DM. Theinclusion of molasses decreased the crudeprotein concentration to 169 and 168 g/kgDM for the 92.5 % and 85 % broiler litterdiets respectively. The Cu concentrationin the diets varied between 22 and27 mg/kg DM. The Ca:P ratio in all thediets was approximately 2:1.

The sheep in the 15 % molasses groupconsumed significantly more feed (p <0.01) than the sheep on the 7.5 % molassesand the pure litter groups (Table 2). Thiswas associated with significantly higher(p < 0.01) final body mass of the sheepin the high molasses treatment. In the15 % molasses group, intakes reached anoptimum after 3 weeks on the diet whilethose on the other 2 diets reachedoptimum intake at about 6–7 weeks.Carcass mass and fat cover at the last ribarea of the carcass tended to be higher inthe molasses-supplemented groups thanin the control, but differences were notsignificant (Table 2).

Throughout the trial the concentrationsof free fatty acid, β-hydroxybutyrate and

glucose in plasma showed no differencesamong treatments (Table 3). Plasma ureaN concentrations at the last collection ofthe trial were 298, 286, and 276 m/l for the100, 92.5 and 85 % broiler litter treatmentsrespectively, versus a quoted normalrange of 80–200 mg/l14 (Table 3). However,high concentrations were measured fromthe onset of the trial. The average ± stand-ard deviation (SD) plasma albumin (35.3 ±4.14 g/l), plasma globulin (24.9 ± 4.79 g/l),bilirubin (30.4 ± 6.2 g/l), PCV (0.78 ± 0.048)and haemoglobin (164 ± 13.9 g/l) concen-trations remained relatively constantthroughout the trial, showed no treat-ment effects and were within the normalrange for sheep13. Copper concentrationsin the livers did not differ significantlyamong treatment groups and were 387,383 and 338 mg/kg DM for the 100, 92.5and 85 % broiler litter treatmentsrespectively.

No significant differences in the activityof any of the plasma enzymes wereobserved between the different treatmentgroups (Table 3). Activities were withinthe normal ranges for sheep with theexception of plasma CK, where theactivity was higher than the expectednormal values. At the final blood collec-tion the CK activities were recorded at62.9 ± 10.6, 56.1 ± 22.6 and 40.0 ± 3.6 U/lfor the pure litter, 92.5 and 85 % litterdiets, respectively. However, these valueswere within a similar range throughoutthe trial. The activity of plasma AST wasrandomly distributed and did not showany correlation with pathological lesionsin the liver or heart.

The purine:creatinine (W0.75)–1 (PD:C)ratios during the trial are presented inTable 4. At 28 d after the onset of the trialthese ratios were lower than at otherstages, although differences were not

Table 1: Chemical composition of broiler litter and litter with molasses.

Broiler litter (%): 100 92.5 85Molasses (%): 0 7.5 15

Dry matter (g/kg) 923 917 870Ash (g/kg DM) 97 110 124Crude protein (g/kg DM) 182 169 168Ether extract (g/kg DM) 11.2 9.4 8.2NDFa (g/kg DM) 540 520 460ADFb (g/kg DM) 395 355 334Calcium (g/kg DM) 15.0 16.0 16.0Phosphorus (g/kg DM) 8.1 8.4 7.4Sodium (g/kg DM) 2.90 2.75 3.35Potassium (g/kg DM) 9.9 11.9 12.7Magnesium (g/kg DM) 4.71 4.79 4.77Copper (mg/kg DM) 22 27 22Manganese (mg/kg DM) 214 241 231Selenium (mg/kg DM) 0.68 0.66 0.66

aNDF = neutral detergent fibre.bADF = Acid detergent fibre.

Table 2: Average feed intake, body mass gain, body mass at slaughter, fresh liver, kidneyand carcass mass and fat thickness at the last rib area of the carcass (mean ± SD).

Broiler litter (%): 100 92.5 85Molasses (%): 0 7.5 15

Feed intake (kg/d) 1.3a* ± 0.08 1.5a ± 0.15 2.1c ± 0.07Final body mass (kg) 44a ± 3.6 48ab ± 3.6 52b ± 2.4Gain/d (g) 26 ± 23 63 ± 32 101 ± 10 Carcass mass (kg) 20 ± 0.9 22 ± 1.3 23 ± 0.9Liver: mass (fresh) (g) 610 ± 40 592 ± 41 668 ± 23 as % body mass 1.4 ± 0.19 1.2 ± 0.18 1.3 ± 0.05 as % carcass mass 3.0 ± 0.33 2.8 ± 0.15 2.9 ± 0.12Kidney mass (g) 108 ± 6.0 134 ± 5.4 134 ± 6.5Fat thickness (mm) 1.8 ± 0.21 2.6 ± 0.52 2.8 ± 0.68

*Values within a row with different superscripts differ significantly at p < 0.01.

122 0038-2809 Tydskr.S.Afr.vet.Ver. (1997) 68(4): 121–124

statistically significant.Mineral concentrations in the plasma

did not reflect any treatment effect.Average concentrations ± SD (mmol/l)over the experimental period were: 2.15 ±0.12 Ca; 2.48 ± 0.394 inorganic P; 12.6 ±2.49 Cu; 0.987 ± 0.076 Mg; 145 ± 3.6 Na;5.2 ± 0.29 K; 16.8 ± 1.66 Zn. The Ca:P ratioin urine varied widely and did not showany consistent pattern. However, in many

of the sheep, irrespective of treatment, theconcentration of P in urine was higherthan its respective Ca concentrations.

Myocardial pathology was mild andsimilar for all the treatment groups (Table5). The hepatic pathology was mild andconsisted of hyaline degeneration andsingle-cell necrosis, considered to benon-specific, and triaditis, an incidentalfinding usually associated with parasites.

Renal pathology was restricted to 1 case oftubular degeneration and 11 of intertubu-lar congestion.

When myocardial pathology wasobserved, the litter was screened forionophores. Narasin, at a concentration of10 mg/kg, was measured in the broilerlitter.

DISCUSSIONThe crude protein concentration of

180 g/kg DM, although less than theaverage South African value, was wellwithin the levels of crude protein in litterfed to animals. However, this was sub-stantially lower than the 300 g dietaryprotein/kg feed that reportedly causedthe liver damage in cows18. A directcomparison of results is therefore notpossible. The plasma urea N in the presenttrial is well above the normal range of80–200 mg/l, indicating that excessammonia must have been absorbed fromthe rumen. The normal concentrations ofplasma albumin, bilirubin and hepaticenzyme activity in plasma indicatenegligible hepatic pathology. Rankinset al.17 questioned the observation ofSilanikove and Tiomkin18, and suggestedthat the observed liver damage mighthave been due to Cu toxicity. Unfortu-nately Silanikove and Tiomkin18 did notreport the Cu concentrations in the broilerlitter used or in the livers of their cows. Inthe present study the Cu concentration inthe litter was 22 mg/kg DM. Copperconcentration in the livers of our sheep(<400 mg/kg) was well below toxiclevels12. Van Ryssen et al.20 concluded fromthe survey of mineral concentrations inpoultry manure that copper sulphate is atpresent not included in broiler rations inSouth Africa. Consequently, in SouthAfrica, Cu toxicosis due to ingestion oflitter is unlikely, contrary to the situationin the USA17. Angus et al.1 reported seriousliver damage in sheep consuming dietscontaining up to 60 % manure fromlaying hens. Their findings included anaccumulation of fluid in the abdomen andmarked histological changes in the liver.No explanation could be given and thepresence of a toxic substance was sug-gested. They also described kidneydamage in sheep on high intakes ofbroiler litter, again of unknown cause1.The possibility of Cu toxicity was consid-ered and ruled out. The causes of theabove lesions were obviously not presentin our experimental diets.

The elevated plasma CK activity mayindicate muscle damage. However,elevated activity was measured through-out the trial and may be unrelated to thebroiler litter treatments. Although the

Table 3: Concentration of various plasma metabolites and activity of enzymes in plasmaat the last collection (mean ± SD).

Broiler litter (%): 100 92.5 85Molasses (%): 0 7.5 15 Norm13

Free fatty acids (mmol/l) 0.6 ± 0.16 0.35 ± 0.05 0.47 ± 0.08 0.558

β-hydroxybutyrate (mmol/l) 0.3 ± 0.03 0.19 ± 0.02 0.21 ± 0.02 0.55Glucose (mmol/l) 3.55 ± 0.1 3.35 ± 0.1 3.47 ± 0.12 2.8–4.4Urea nitrogen (mg/l) 293 ± 17 267 ± 14 257 ± 16 80–200ASTa (U/l) 58 ± 3.9 56 ± 2.5 61 ± 8.5 61δ-GTb (U/l) 39 ± 1.9 37 ± 4.1 39 ± 3.5 20–50CKc (U/l) 63 ± 10.6 56 ± 22.6 40 ± 3.6 8–13SDHd (U/l) 1.6 ± 0.2 1.8 ± 0.16 1.4 ± 0.19 8–16ALTe (U/l) 16.5 ± 1.94 13.3 ± 3.15 16.5 ± 2.3 30

aAST = aspartate aminotransferase.bδGT = gamma glutamyltransferase.cCK = creatine kinase.dSDH = sorbitol dehydrogenase.eALT = alanine aminotransferase.Differences between treatments were not significant.

Table 4: Purine:creatinine ratio (W0.75)–1 in urine from spot samples taken during the trial(mean ± SD).

Day of collection

Litter (%) 1 28 63 81

100 27 ± 10.7 16 ± 4.8 24 ± 6.5 24 ± 7.292.5 28 ± 24.6 13 ± 6.1 27 ± 6.0 25 ± 5.785 31 ± 29.0 24 ± 9.5 32 ± 4.3 26 ± 3.2

Differences among means not significant.

Table 5: Number of sheep fed broiler litter with or without molasseswith histopathological lesions in the myocardium (n = 6/treatment)a.

Broiler litter (%): 100 92.5 85Molasses (%): 0 7.5 15

Myofibre hypertrophy 5 3 4Myofibre atrophy 5 3 4Hyaline degeneration 3 3 4Hyaline necrosis 3 2 4Attempted regeneration 3 3 4Interstitial fibrosis 1 2 4

Incidental lesionsSarcocysts 4 1 3Congestion 2 2 5

aDistribution of the lesions was random and severity was classified as mild.

0038-2809 Jl S.Afr.vet.Ass. (1997) 68(4): 121–124 123

myocardial lesions resembled those ofionophore-associated poultry littertoxicity, as described by Bastianello et al.4,they were considered very mild in thisbatch of sheep. The ionophore antibiotic,narasin, which was present in the broilerlitter, is not registered in South Africa asan ionophore for ruminants. However,our findings suggest that the level ofnarasin in the feed (10 mg/kg) was nothigh enough to induce severe myocardialpathology or influence the animals’ abilityto survive on the particular diet. Since thepurpose of feeding high levels of litter wasto keep the animals alive, i.e. at a mainte-nance or even submaintenance level ofnutrition, these histological findings wereprobably of minor significance.

The inclusion of molasses in the dietsresulted in a significant increase in feedintake. This increased intake was reflectedin improved daily mass gains and ahigher final body mass in the respectivegroups. The higher body mass in themolasses-fed sheep must have been duepartly to higher gut content in thesegroups, because differences in carcassmass between treatments were lesspronounced. If the objective is to keep thesheep alive, the inclusion of molasses maybe unnecessary, except if the intake of thepure litter is unsatisfactory. In fact, lowerlitter intakes than recorded in the presenttrial should be sufficient for the survivalof the sheep. Where litter with a crudeprotein content higher than that in thepresent study is used, the addition ofmolasses may have a more pronouncedeffect, e.g. in detoxifying ammonia.

The ratio of PD:C in urine can be usedas an indication of relative rumenmicrobial protein synthesis in the animal7.Although this ratio increased slightly withan increase in molasses inclusion in thediet, variations within treatments werewide, and differences were not signifi-cant. From urinary PD:C ratios, Chenet al.7 estimated rumen microbial nitrogenproductions in sheep of less than 5 tomore than 20 g N/d. At PD:C ratios ofbetween 15 and 30, Chen et al.7 estimateda rumen microbial nitrogen production of6–12 g/d. In the current trial a PD:C ratioof ca. 23 was measured, indicating arelatively low microbial production of lessthan 12 g/d. The low PD:C ratios observedat 28 d may reflect incomplete adaptationof the microbes to the diets at that stage.This was evident in relatively low initialDM intakes which showed an increase for

the first 6 weeks in the 100 % litter diet.Further studies are required to confirmthis apparent long adaptation periodrequired by the sheep or the rumenmicroorganisms before they can utilisebroiler litter effectively.

The concentrations of minerals in thebroiler litter were lower than the meansfor South African samples, but well withinthe range of variation measured20. Thecows used by Silanikove and Tiomkin18

were hypocalcaemic and hyperphos-phataemic when they consumed broilerlitter and reacted positively to calciumborogluconate injections. The authorssuggested that the apparent imbalancesin Ca and P metabolism were due to thesevere l iver damage in the cows.Although the dietary Ca:P ratio in thepresent study was normal at 2:1, the Pintake was well above requirements forsheep16. The high concentration of P inurine relative to that of Ca signifies thatthe high P intake through litter couldcause urinary calculi in wethers.

It is concluded that the feeding of drybroiler litter of the quality used in thepresent investigation did not affect thehealth of the sheep when fed for a periodof ca. 80 days. Therefore, in an emergencyfeeding situation of limited duration, itseems feasible to feed broiler litter toruminants, even at restricted levels ofintake. However, the necessary precau-tions must be taken because broiler litterhas many potential problems as an animalfeed9,10.

ACKNOWLEDGEMENTWe thank Mrs S S M van Malsen for

laboratory assistance.

REFERENCES1. Angus K W, Suttle N F, Munro C S, Field A C

1978 Adverse effects on health of includinghigh levels of dried poultry waste in thediets of lambs. Journal of ComparativePathology 88: 449–466

2. Anonymous 1974 Clinical laboratory. Medico-chemical investigation methods (11th edn).E Merck, Darmstadt

3. Bastianello S S, Fourie N, Prozesky L, NelP W, Kellermann T S 1995 Cardiomyopathyof ruminants induced by the litter of poultryfed on rations containing the ionophoreantibiotic, maduramicin. II. Macropathol-ogy and histopathology. OnderstepoortJournal of Veterinary Research 62: 5–18

4. Bastianello S S, McGregor H L, Penrith M-L,Fourie N 1996 A chronic cardiomyopathy infeedlot cattle attributed to toxic levels ofsalinomycin in the feed. Journal of the SouthAfrican Veterinary Association 67: 38–41

5. Borchers R 1977 Allantoin determination.

Analytical Biochemistry 79: 612–6136. Bosman S W 1973 Chicken litter in fattening

rations for cattle and sheep. South AfricanJournal of Animal Science 3: 57–61

7. Chen X B, Mejia A T, Kyle D J, Orskov E R1995 Evaluation of the use of the purinederivative:creatinine ratio in spot urine andplasma samples as an index of microbialprotein supply in ruminants: studies insheep. Journal of Agricultural Science,Cambridge 125: 137–143

8. De Villiers S, Van der Walt J G, Procos J 1977An accurate, sensitive and reproduciblemethod for the colorimetric estimation offree fatty acids. Onderstepoort Journal ofVeterinary Research 44: 169–172

9. Fontenot J P 1991 Recycling animal waste byfeeding to enhance environmental quality.The Professional Animal Scientist 7: 1–8

10. Fourie N, Bastianello S S, Prozesky L, NelP W, Kellerman T S 1991 Cardiomyopathyof ruminants induced by the litter of poultryfed on rations containing the ionophoreantibiotic, maduramicin. I. Epidemiology,clinical signs and clinical pathology. Onder-stepoort Journal of Veterinary Science 58: 291–296

11. Golab T, Barton S J, Scroggs R E 1973Colorimetric method for monensin. Journalof the Association of Official AnalyticalChemists 56: 171–173

12. Howell J McC, Gooneratne S R 1987 Thepathology of copper toxicity in animals. InHowell J McC, Path F R C, Gawthorne J M(eds) Copper in animals and man Vol 2. CRCPress, Boca Raton, Florida: 53–78

13. Kaneko J J 1989 Clinical biochemistry ofdomestic animals (4th edn). Academic Press,New York

14. Kargaard J, Van Niekerk B D H 1977 Incor-poration of DPW, urea and fish meal withvarying molasses levels in cattle feedlotrations. South African Journal of AnimalScience 7: 117–125

15. Koh T-S, Benson T H 1983 Crit icalre-appraisal of fluorometric method fordetermination of selenium in biologicalmaterials. Journal of the Association of OfficialAnalytical Chemists 66: 918–926.

16. National Research Council 1985 Nutrientrequirements of sheep (6th edn). NationalAcademy of Science, National AcademyPress, Washington DC

17. Rankins D L, Eason J T, McCaskey T A,Stephenson A H, Floyd J G 1993 Nutritionaland toxicological evaluation of three deep-stacking methods for processing of broilerlitter as a foodstuff for beef cattle. AnimalProduction 56: 321–326

18. Silanikove N, Tiomkin D 1992 Toxicityinduced by poultry litter consumption:effect on measurements reflecting liverfunction in beef cows. Animal Production 54:203–209

19. Tietz N W 1976 Fundamentals of clinicalchemistry (2nd edn). W B Saunders, Phila-delphia

20. Van Ryssen J B J, Van Malsen S, Verbeek A A1993 Mineral composition of poultrymanure in South Africa with reference tothe Farm Feed Act. South African Journal ofAnimal Science 23: 54–57

124 0038-2809 Tydskr.S.Afr.vet.Ver. (1997) 68(4): 121–124