Bacterial infections are of great economic importance in the poultry industry worldwide. Particularly in
the chicken layer industry recent reports of colibacillosis, erysipelas and fowl cholera in Western
Europe implicate a resurgence of these diseases (ZANELLA et al. 2000, VANDEKERCHOVE et al.
2004, MAZAHERI et al. 2006). This might be attributed to a change from cage housing systems to
alternative housing systems like aviaries and free-range systems and an accompanied increase of
infectious pressure (MAZAHERI et al. 2006).
To prevent bacterial diseases in the egg production industry, vaccination is of increasing relevance
as it encounters problems due to growing antibiotic resistance and an increased consumer demand
concerning food safety (VAN OIRSCHOT 1994, WHITE et al. 2002). Furthermore, vaccination can
avoid withdrawal periods in marketing eggs following antibiotic treatment. As commercial vaccines
are frequently not efficient due to the high antigenic variability of the disease causing organism
(BRAGG, R.R. 2002, SMART et al. 1993) or as commercial vaccines are not available for certain
diseases like erysipelas in chickens, autogenous vaccines can offer sophisticated alternatives.
Adjuvants used in vaccine preparations for layers are in general either aluminium hydroxide or different
mineral oils in different phase combinations. Each adjuvant has advantages and disadvantages. An
aluminium hydroxide gel adjuvant vaccine usually causes little local reaction and the antibody titre
increases quite rapidly after injection (MOREIN et al. 1996). Unfortunately, this titre also decreases
rapidly in almost the same manner, thus requiring a second injection. A mineral oil adjuvant will release
the antigen slowly and respond directly to antigen-presenting macrophages, resulting in antibodies
being able to have a longer persistence than aluminium hydroxide gel adjuvant vaccine. Thus, often
only a single application is required. On the other hand, the oil responsible for the slow release might
cause local reactions (DEGUCHI et al., 1998, FUKANOK et al., 2000, MOREIN et al., 1996, REID et
Field records from layers have been used to estimate the effect of different adjuvants on the protection
level after vaccinating against E.coli during rearing.
We conducted a retrospective analysis of field records to elucidate the influence of the adjuvant in
the autogenous vaccine on hen-housed production and mortality, based on weekly records from five
farms with up to 9 years of data per farm. Differences due to the autogenous vaccine were estimated
from a total of 81 flocks, using variance analysis with the program package SAS (SAS, 2004).
Material and Methods
Farms, housing systems and genotypes
The data included in the analysis were recorded on five farms of different egg producers. Each farm
had two separate houses with up to nine flocks with complete weekly records for this study. The
number of hens per flock varied between about 8,000 and 43,000 hens with an average of 20,000
hens per flock. Two farms were equipped with aviaries, three used free-range systems. Three different
genotypes, two brown and one white-egg layer strain were included in the study. Since the type of
layer was highly confounded with farm and housing system, the genotype was not considered in
further analysis. Each house was treated as separate experimental unit. The general time trend was
measured as linear regression.
Each flock was vaccinated during rearing with a trivalent autogenous vaccine containing relevant
strains of the bacterial pathogens Escherichia coli, Pasteurella multocida and Eryselpelothrix
The last two flocks per farm were vaccinated intramuscularly once with an autogenous vaccine
produced by Lohmann Tierzucht (LTZ) at an age of 14 weeks, following several flocks on the same farms
vaccinated twice with an autogenous vaccine produced by another company at 14 and 16 weeks of
age. The LTZ vaccine was blended with mineral oil as adjuvant and the other vaccine used aluminium
hydroxide as carrier substance to enhance the immune response.
The distribution of flocks over system, farm, house and vaccination programs in rearing is shown in
Table 1: Number of tested flocks per environment, farm, house and type of vaccine
The weekly records for mortality and rate of lay from 22 to 69 weeks of age were combined into twelve
4-week periods for analysis purposes. To estimate the time trend, consecutive flocks in each house
were assigned a serial number, which was included in the model. In total, up to 12 four-week
observations of 81 flocks were analysed in the model that considers the following fixed and random
effects as well as a linear regression.
1. housing system (HS)
2. house (H) within HS
3. vaccination (V)
4. interaction between HS and V
5. interaction between H and V within HS
6. age (A) in terms of 4-week period
7. interactions between V and A
8. week within four-week period (W)
9. sequence of the flocks (S)
10. flock (f) within HS, V, and
11. random error
As expected, age (A) effects, represented by the four-week observations, were highly significant, but
also the sequence of the flocks influenced both traits, mortality and hen housed rate of lay. Additionally,
highly significant were the considered interaction between vaccination and age as well as the week
within the period of a four week observation (W(A)) on laying performance. The highest F-value was
estimated for the effect age, which therefore has the strongest influence on both traits.
Table 2: Test of significance for the considered effects on laying performance and mortality
per hen housed
(ADH) is released. This causes the kidneys to retain more water and excrete less. In the event of
water excess the release of the hormone ADH and the feeling of thirst are inhibited and thus the
amount of water excreted by the kidneys increased. The concentration of sodium in the blood plasma
in connection with the renin-angiotensin-aldosterone system is also involved in the regulation of water
supply. A reduction in sodium concentration leads to a reduction in the plasma volume. The reninangiotensin-aldosterone system is thus set into operation. It results in an increased resorption of
sodium in the kidneys and an increase in plasma volume.
The water taken up is resorbed in the small intestine via osmotic activity. 70% of the body mass of
broiler chickens is water (Wilson, 2009). Of this around 50 % is located in the intracellular space and
20 % in the extracellular space which is subdivided into intravasal space and extravasal space (fig. 2).
The exchange of water between the fluid spaces is caused by differences in concentration with the aim
of achieving osmotic balance.
Figure 2: Fluid spaces
The estimated regression coefficients for the sequence of the flocks are for both traits significant.
From flock to flock, rate of lay per hen housed increased by 1.7 %, and mortality decreased by 0.03 %
per week. In a full laying year of 52 weeks this is an increase of about 5 eggs per year and 1.5% less
As shown in table 1, two flocks per house were vaccinated with an oil based autogenous vaccine of
Lohmann Tierzucht (oil) and a variable number of flocks vaccinated with an autogenous vaccine
using aluminium hydroxide (Al(OH)3) as carrier from another company. Although the two types of
vaccines were sequentially tested, the Least-Squares estimates for vaccine effects shown in table 2
comparing the two vaccines should be free of the general time trend, which was included as linear
regression in the model. Under ideal experimental conditions, the comparisons would have been
simultaneous and with untreated controls.
Significant differences between flocks due to the vaccine used can also be seen when looking at the
interaction effect between housing system and vaccine type. The Least Squares mean rate of lay
was lower when an Al(OH)3 based vaccine (76.3 %) was used than an oil-based vaccine (77.5 %)
was inoculated. The same positive tendency was estimated for mortality and can be read in table 3.
Weekly mortality is relatively high, with 0.30 % and 0.27 % for the two types of vaccines, but favours
the oil-based LTZ vaccine consistently in both types of housing by a margin of .03 % or 1.56 % per year.
The different standard errors of the estimates reflect the fact that more flocks were vaccinated with the
Al(OH)3 based vaccine before changing to the LTZ vaccine with oil suspension.
Table 3: LS-Means for hen housed rate of lay and mortality
The LS-Means for the two vaccination groups in four-week periods (not shown in detail here) indicate
that the main benefit of the oil-based LTZ vaccine is expressed after peak production: insignificant
+0.1 % per week from 22-33 weeks of age, but a significant difference of + 1.7 % per week from 34-
69 weeks of age. During the last four-week period, average rate of lay per hen housed was 57.6 %
versus 62.0 %. Similar results were found for cumulative mortality, which is included in the trait henhoused rate of lay.
Significant differences between housing systems and farms with the same housing system suggest
considerable potential for further improvement of farm management irrespective of the housing system
and vaccination program. Farm 5 achieved the best performance of all farms, averaging nearly 10 %
above the results of farm 2.
Consumer demand for safe food and awareness of potential problems due to antibiotic resistance
on the one hand and the preference for eggs from non-cage systems on the other hand present a
challenge for egg producers and poultry veterinarians to control bacterial infections by vaccination
(VAN OIRSCHOT 1994, WHITE et al. 2002). Prevention and control of bacterial infections in laying hens
by means of commercial vaccines is often hampered by a lack of efficiency due to the high antigenic
variability of the pathogens. Therefore, autogenous vaccines are used instead (BRAGG 2002,
WAMBURA 2010). They offer the unique possibility to cover bacterial strains currently responsible
for disease on a specific farm. The efficiency of autogenous products is highly influenced by a large
number of factors like pathogen detection, typing and selection, regular “update” of the vaccine with
the strains currently causing disease as well as infectious pressure in the flock.
In this study we tried to elucidate the influence of the adjuvant on the efficiency of autogenous vaccines
for laying hens in the field. We therefore conducted a retrospective analysis of performance data
provided by five different egg producers with two comparable houses each. In all 10 houses, the last
two flocks of layers were vaccinated with an autogenous trivalent vaccine produced by Lohmann
Tierzucht and blended using mineral oil as adjuvant (oil), following previous vaccination of several
flocks with a comparable vaccine of a competitor using aluminium hydroxide as adjuvant (Al(OH)3).
Weekly rate of lay and mortality was expressed in twelve 4-week periods from 22 to 69 weeks of age
for statistical analysis.
To get a reliable estimate of differences due to the vaccine used, we ignored strain of layers (confounded
with house within farm and housing system) in the model; other unknown differences due to farm
management and nutrition were minimized by using both vaccines on each farm in consecutive flocks.
Ideally, both types of vaccines should have been used simultaneously in the two houses per farm
and repeated over two years to get unbiased estimates of the vaccination effect. To estimate the
effects of vaccination as free as possible from the general time trend, we included the test year in
the model as linear regression over the whole observation period. The regressions of hen-housed
rate of lay and weekly mortality on years estimate the combined effects of selection for improved
adaptability to non-cage systems and improved environment due to cumulative experience on each
farm with a given housing system. The estimated difference of the performance following the oilbased vaccination compared to the alternative vaccine previously used should be free of the general
time trend and understood as additional effect on all farms in the last two flocks.
The results found under these conditions indicate that commercial layers vaccinated once during
rearing with an autogenous mineral oil based bacterin outperform layers vaccinated twice with an
aluminium hydroxide based autogenous vaccine. This implies that an oil-suspension vaccine provides
better protection than the aluminium hydroxide based vaccine, which is in accordance with previous
studies (MURTHY et al., 2007, BENNECKE 2008, JACOBS et al. 1992). One reason for this result might
be the longer duration of the immune response. Duration of immunity is one of the key prerequisites
for vaccines used in layers, as laying periods last up to one year after immunisation and a booster
vaccination during the laying period is technically more than difficult. Mineral oil adjuvants lead to a
prolonged release of antigens, respond directly to the macrophages and thereby display a longer
lasting antibody titre level than aluminium hydroxide gel adjuvant vaccines (FUKANOKI et al. 2000).
Several other factors may affect the efficiency of a vaccine like the antigen quantity per dose as well
as the production procedure of the antigen. As the vaccine with the aluminium hydroxide as adjuvant
was produced by a different producer and their production scheme is not known, our conclusions are
limited to the adjuvants, which is the only objective distinguishing feature between the two vaccines.
In conclusion, in this retrospective study the mineral oil based autogenous vaccine produced by
Lohmann Tierzucht lead to better performance of commercial layers in comparison to vaccination
with an aluminium hydroxide based autogenous vaccine. Furthermore, better performance was
achieved with a single immunisation compared to the labour- and cost-intensive double immunisation required by the aluminium hydroxide based vaccine.
Praxisergebnisse mit dem Einsatz autogener Impfstoffe
zur Verringerung von E. Coli bei Legehennen in Bodenhaltung
Bei der Aufbereitung von bestandsspezifischen Impfstoffen setzen verschiedene Hersteller
unterschiedliche Adjuvantien (Trägerstoffe) zur Verstärkung der Immunantwort ein. In der Regel sind
dies verschiedene Mineralöle oder Aluminiumhydroxid (Al(OH)3), deren Einfluss auf die Wirkung des
Impfstoffes in diesem Beitrag untersucht wurde. Die hierzu ausgewerteten Felddaten berücksichtigen
Leistungsdaten von 81 Legehennenherden, die sich auf fünf Betriebe mit jeweils zwei Häusern – drei
Freilandbetriebe und zwei geschlossene Bodenhaltungen mit integrierter Voliere – aufteilen. Die letzten
beiden Herden eines Hauses wurden einmalig mit einem bestandsspezifischen Impfstoff der Firma
Lohmann Tierzucht geimpft, dessen Adjuvanz eine Ölsuspension ist. In allen anderen 61 Herden
wurde den Legehennen hingegen zweimalig ein bestandsspezifischer Impfstoff mit Al(OH)3 als
Unter Berücksichtigung verschiedener Einflussfaktoren zeigen die einmalig geimpften Legehennen im
Vergleich zu den mit Al(OH)3 als Trägerstoff geimpften Hennen eine bessere Legeleistung je
Anfangshenne (77,5 % vs. 76,3 %) sowie eine niedrigere Verlustrate (0,27 % vs. 0,30 % pro Woche).
Ähnlich positiv zeigt sich die Wirkung des Öladjuvanz bei einer Betrachtung innerhalb der beiden
Haltungssysteme. Eine Verknüpfung der Effekte Betrieb und Haltungssystem macht jedoch deutlich,
dass die besseren Leistungsergebnisse in der Bodenhaltung in erster Linie auf den Betriebseffekt
zurückzuführen sind. Die Unterschiede in der Legeleistung betragen von Betrieb zu Betrieb bis zu
10 %. Ferner verstärkt die Betriebsentwicklung im Verlauf der neun Beobachtungsjahre die
Überlegenheit des Öladjuvanz gegenüber Al(OH)3, durch dessen Einsatz in den letzten beiden
Beobachtungsjahren. Der geschätzte Regressionskoeffizient zeigt von Herde zu Herde eine Steigerung
in der Legeleistung um 1,7 % und eine Abnahme der Verluste um 0,03 %. Das entspricht einer
jährlichen Steigerung der Legeleistung je Anfangshenne um 5 Eier und einer Senkung der Verlustrate
um 1,5 % als kombinierter Effekt genetisch verbesserter Anpassung an Bodenhaltungsbedingungen,
Erfahrung der Betriebsleiter mit den besonderen Haltungsbedingungen und verbesserter Impftechnik.
Die verbesserten Leistungsergebnisse in Kombination mit nur einmaliger und dadurch weniger zeitund kostenaufwendiger Immunisierung sprechen für einen bestandsspezifischen Impfstoff mit
Ölsuspension als Adjuvanz.
BENNECKE, K., 2008: Vergleichende Untersuchungen zur protektiven Wirkung einer Streptococcus suis Ganzzell- und
einer Subunitvakzine. Diss., Tierärztliche Hochschule Hannover.
BRAGG, R.R., 2002: Isolation of serovar C-3 Haemophilus paragallinarum from Zimbabwe: a further
indication of the need for the production of vaccines against infectious coryza containing local isolates of H. paragallinarum.
Onderstepoort. J. Vet. Res. 69: 129- 132.
DEGUCHI, K., T. HONDA, K. MATSUO, H. FUJIKAWA, T. IWAMOTO and Y. SAKANOUE, 1998: Influence of inoculation
site of combined oil-adjuvanted vaccine on the antibody response in chickens. J. Vet. Med. Sci. 60: 831-835.
FUKANOKI, S., K. MATSUMOTO, H. MORI and R. TAKEDA, 2000: Relation between antigen release and immune response
of oil adjuvanted vaccines in chickens. J. Vet. Med. Sci. 62: 571-574.
JACOBS. A.A.C., W. CUENEN, and P.K. STROM, 1992: Efficacy of a trivalent Haemophilus paragallinarum vaccine
compared to bivalent vaccines. Vet. Microbiol. 32: 43-49.
MAZAHERI, A., H.C. PHILIPP, H. BONSACK AND M. VOSS, 2006: Investigations of the vertical transmission of Erysipelothrix
rhusiopathiae in laying hens. Avian Diseases 50(2): 306-308.
MOREIN, M., M.V. ERIKSSON, A. SJOLANDER and K.L. BENGTSSON, 1996: Novel adjuvants and vaccine delivery
systems. Vet. Immunol. Immunopathol. 54: 373-384.
MURTHY, A.K., J.P. CHAMBERS, P.A. MEIER, G. ZHONG, B.P. ARULANANDAM, 2007: Intranasal vaccination with a
secreted chlamydial protein enhances resolution of genital Chlamydia muridarum infection, protects against oviduct
pathology, and is highly dependent upon endogenous gamma interferon production. Infection and Immunity, 75 (2): 666-
REID, G.G., and P.J. BLACKALL, 1987: Comparison of adjuvants for an inactivated infectious coryza vaccine. Avian Dis.
SAS INSTITUTE, 2004: SAS/STAT User’s Guide (Release 8), Cary NC, USA.
SMART N.L., HURNIK D., MACINNES J.I., 1993: An investigation of enzootic Glasser’s disease in a specific-pathogenfree grower-finisher facility using restriction endonuclease analysis. Can Vet J. 34(8): 487-490.
VANDEKERCHOVE, D., P. DE HERDT, H. LAEVENS, F. PASMANS, 2004: Risk factors associated with colibacillosis
outbreaks in caged layer flocks. Avian Pathology, 33(3): 337-342.
VAN OIRSCHOT, J.T. 1994: Vaccination in food animal populations. Vaccine 12: 415-418.
WAMBURA P.M., 2010: Preparation and use of autogenous vaccine from Avibacterium paragallinarum (strain Tan 1-05)
in layer chickens. Trop. Anim. Health Prod. 42(3): 483-486.
WHITE, D.G., S. ZHAO, S. SIMJEE, D.D. WAGNER and P.F. MCDERMOTT, 2002: Antimicrobial resistance in foodborne
pathogens. Microbes Infect. 4: 405-412.
ZANELLA, A., G.L. ALBORALI, M. BARDOTTI, P. CANDOTTI, P.F. GUADAGNINI, P.A. MARTINO
AND M. STONFER, 2000: Severe Escherichia coli O111 septicaemia and polyserositis in hens at the start of lay. Avian
Pathology, 29(4): 311-317.