Ministry of Agriculture & Rural Development

Project Progress Report
A blueprint for sustainable smallholder pig production
in Central Vietnam
CARD Project 001/04VIE

Milestone 8: FINAL REPORT

APRIL 2010

Table of contents
TABLE OF CONTENTS 2
1. INSTITUTE INFORMATION 3
2. PROJECT ABSTRACT 4
3. EXECUTIVE SUMMARY 4
4. INTRODUCTION & BACKGROUND 6
5. PROGRESS TO DATE 6
5.1 IMPLEMENTATION HIGHLIGHTS 6
5.3 SMALLHOLDER BENEFITS 15
5.4 CAPACITY BUILDING 15
5.5 PUBLICITY 16
5.6 PROJECT MANAGEMENT 16
6. REPORT ON CROSS-CUTTING ISSUES 16
6.1 ENVIRONMENT 16
6.2 GENDER AND SOCIAL ISSUES 16
7. IMPLEMENTATION & SUSTAINABILITY ISSUES 16
7.1 ISSUES AND CONSTRAINTS 16
7.2 OPTIONS 17
7.3 SUSTAINABILITY 17

617 336 52985
Position:
Associate Professor of Veterinary
Microbiology
Fax:
617 336 51355
Organisation School of Veterinary Science The
University of Qld
Email:
In Australia: Administrative contact
Name:
Melissa Anderson
Telephone:
61 7 33652651
Position:
Manager Research Projects Office
Fax:
61 7 33651188
Organisation
School of Land and Food The
University of Qld
Email:

In Vietnam
Name:
Dr Cu Huu Phu
Telephone:
84 4 8693923
Position:
Head of Bacteriology Department

control farms for a 12 month period) (Develop a management plan for preweaning diarrhoea using
a continuous improvement model-Logframe reference 2a and 2b).

3. Development of polyclonal sera and/or PCR incl. rapid detection of novel fimbrial antigens
(Improve diagnostics for preweaning diarrhoea-Logframe reference 3).

Whilst this project achieved outputs for all three objectives according to the project logframe, some
significant problems were experienced in trying to identify the novel fimbrial antigen present in
Vietnamese O8 strains (christened F19) and in developing an enteric management plan within a
holistic continuous improvement framework. A final attempt to purify the novel fimbrial antigen
was undertaken with great success in mid-2010 using funds from the University of Adelaide and we
are now awaiting identification of the amino acid and gene sequences for this unusual antigen.

In small scale trials conducted at NIVR, the ETEC vaccine (still encorporating F4, F5 and the new
F19 antigens) was proven to be safe and efficacious when administered to pregnant sows (2 doses at
5 and 2 weeks before farrowing). It is now being supplied to selected piggeries in North Vietnam on
a research only basis, with reports of good efficacy against neonatal E. coli infection and no side-
effects. The vaccine has also been produced for the CARD 004/05VIE project and used in the
selected smallholder farms in central Vietnam in this related AUSAID project as part of a
Continuous Improvement Model to integrate best management practices into a holistic pig
production improvement plan. A small scale field trial showed that the vaccine significantly
reduced the occurrence of diarrhoea in general and in investigations of vaccinated herds that
reported diarrhoea, no enterotoxigenic E. coli was isolated from faecal samples confirming that the
cause of the diarrhoea was not neonatal colibacillosis.

Production data for the five test and five control farms over a 12-month period were analysed and a
statistically significant improvement in preweaning mortality was noted in the test farms (8.6% ±
3.6) over the trial period compared to the controls (15.6 ± 4.3; p<0.05). A bigger improvement may
have been confounded by the small sample size, but problems in the adoption of the Continuous
Improvement Model may also have had an impact. The major problem encountered from the farm

In summary, the NIVR vaccine has been shown to be safe, efficacious and now must be registered
as soon as possible and licensed throughout the country. An ongoing field trial will conclude in
November 2010 and on the basis of this data, partnerships should be sought with local vaccine
companies such as NAVETCO for the mass production and distribution of the vaccine. A large
number of pathogens have been isolated from preweaning pigs with diarrhoea confirming that
greater attention to disease prevention through better husbandry and management, introduction of
the NIVR vaccine, key preventative medications and minimal antimicrobial use will contribute
strongly towards maintaining the profitability of smallholder farmers.

4. Introduction & Background
Diarrhoea during the suckling period has been recognised as the principle health problem affecting
both smallholder and commercial pig production in Vietnam. Previous research has confirmed the
presence of a new fimbrial type in E. coli strains causing colibacillosis in Vietnam that would not be
controlled by existing vaccines. Existing vaccines are currently imported into Vietnam at
considerable cost. In addition, there are many other causes of suckling diarrhoea, the significance of
which is currently unknown in Vietnam, which are all affected by husbandry and management
during farrowing and lactation. Project 001/04VIE (Diagnosis and control of diarrhoea in suckling
pigs) began with three objectives to solve this problem:

1. Production and testing of locally-produced E. coli vaccines
2. Development of a management plan for preweaning diarrhoea using a continuous improvement
(CIP) model
3. Improved field and laboratory diagnosis of preweaning diarrhoea

5. Progress to Date
5.1 Implementation Highlights
Objective 1: Production and testing of local produced vaccine
Output 1.1: Identification and confirmation of components, including novel strain.

The vaccine Master Seed (50 x 1ml vials of each of the three vaccine strains in Brain Heart Infusion

(CARD-VN2)
NVP1372
(CARD-VN3)
O64 F5 STa

* Negative for all five recognized fimbriae associated with porcine enterotoxigenic E. coli (F4, F5,
F6, F18 and F41). May therefore possess a novel fimbrial antigen.

Output 1.2: Characterization of the novel fimbrial antigen
The two 5F- ETEC strains were examined for mannose-resistant haemagglutinating activity using
Sheep Red Blood Cells. Mannose-resistant haemagglutination was observed at 37
o
C, but not at
18
o
C for both strains, confirming the production of adhesins (ie fimbriae) at 37
o
C (Table 1).

Table 3
: Haemagglutination results of two 5F- ETEC strains
Cultures grown at:
37
o
C 18
o
C

Strain
NaCl 0.85% 1.5% D-Mannose NaCl 0.85% 1.5% D-Mannose

Output 1.4: Efficacy testing of vaccine
The NIVR prepared the vaccine for small scale protection, safety and efficacy trials. In summary,
the vaccine produced no unacceptable side effects in vaccinated gilts and their progeny. When
compared to Littergard and Ecovac, two commercially available vaccines from Pfizer and Intervet,
respectively, the NIVR vaccine produced statistically similar specific antibody titres to an E. coli F4
fimbriae strain. This confirms that under experimental conditions, the vaccine is both safe and
efficacious in generating anti-F4 agglutinating antibodies. Small amounts of the vaccine were
supplied to selected herds in the North of Vietnam and to smallholder farmers in Central Vietnam as
part of the 004/05VIE project. No side effects or vaccine reactions were reported and anecdotal

Appropriate culture
media (37
o
C,
overnight)
Purity testing
10% (v/v) bufferred
formaldehyde to a final
concentration of 0.3%
Mix with equal colume of
each bacterin
Dispense into sterile
bottles and label

Sterility testing
Sterility testing
Output 1.5: Field testing of vaccine
Field trials were conducted at two communes in Thua Thien Hue and three communes in
Quang Tri in 2009/2010. Pregnant sows each received 2 ml of vaccine (approximately 1.5 x
10
9
bacteria) at 9 and 12 weeks of gestation compared to the control group which were not
vaccinated. No local or systemic reaction to the vaccine was observed and all sows gave
birth at the correct stage of gestation to an average of 9.3 healthy piglets per sow. The
prevalence of pre-weaning diarrhoea in piglets born from vaccinated sows at 1, 2 or 3 weeks
of age were: 16.1; 22.7 and 26.5%, compared with those of 48.1; 33.8 and 37.5%,
respectively from control group (P<0.005).

Random faecal samples (n=37) taken from piglets with diarrhoea were assayed for the
presence of the six most common enteric pathogens, causing pre-weaning diarrhoea.

continuous improvement plan.
Output 2.1 Field data collected at test and control farms.
An analysis of preweaning mortality reported over a 14-month observation period established that
the test farms, which were subject to a number of recommendations during the life of the project,
had a significantly lower average pre-weaning mortality compared to the control farms (8.6% ± 3.6
vs 15.6 ± 4.3; p<0.05). One of the control farms was removed from the trial due to an outbreak of
hog cholera. For the majority of test farms, consistently lower pre-weaning mortalities were
sustained over the trial period, however for Dong May farm in Thai Binh, pre-mortalities of close to
20% were reduced to 10% towards the end of the observation period. It is difficult to determine
whether this reduction in preweaning mortality was associated with uptake of any of the previous
visit’s recommendations as the same problems were still observed on the second visit.
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
45.00
50.00
Apr-05
May-05
Jun-05
Jul-05
Aug-05
Sep-05
Oct-05
Nov-05

Some of the disease problems were clearly linked to the unacceptably high heat index recorded in
some of the sheds, restricted feed intake and the large number of sows with low condition scores
and poor ventilation. Anh Hiep Farm (Hung Yen Province) perhaps showed the greatest
improvements over the life of the project, but this farm achieved consistently low rates of
preweaning mortality throughout the year.
The overall objective of the continuous improvement model was, through the farm visits, to provide
Vietnamese scientists with training in herd health monitoring (focused on preweaning mortality)
whilst creating demonstration farms that could be utilized for smallholder training workshops.
However, we soon realised that this model was unworkable and that the resources allocated were
inadequate. With advice and assistance from the CARD Programme Management Unit, Project
number 004/05VIE (A blueprint for smallholder pig production in Central Vietnam) was developed
as a holistic plan for capacity building, focused on smallholder farmers in Quang Tri and Thua
Thien Hue. This project was extremely successful and details are provided in the final report.
However, it must be stressed that without the experience gained from 001/04VIE, we would not
have achieved such a good outcome. There was considerable crossover between the two projects,
particularly in that the E. coli vaccine produced by NIVR was provided free to smallholder farmers
selected in the 004/05VIE project for further training and capital improvement.

Objective 3: Improved diagnostics for preweaning diarrhoea
Outputs 3.1: Prevalence of major causes of pre-weaning diarrhoea on large piggeries and
smallholder farms
Dr Thuy’s investigation of the causes of preweaning mortality in samples from commercial vs
village based piggeries provided some interesting results. Firstly, single disease agents were only
ever identified in commercial piggeries, but these only constituted 21.2% of total samples. By
contrast, multiple agents were always detected in enteric disease samples from village-based pigs.
In commercial piggeries, rotavirus and TGEV, or rotavirus, TGEV and enterotoxigenic E. coli
(ETEC) were identified in 26.3% of samples, indicating that these diseases are most certainly


3 (2.5) RV

3 (2.5) TGEV

11 (9.3) ETEC

4 (3.4) C. per. 2 (1.7)

Total single infections 25 (21.2)

RV ETEC 6 (5.1) RV TGEV

17 (14.4) 3 (6.7)
Cocci

Cocci RV ETEC 2 (1.7) 3 (6.7)
Cocci

RV TGEV

4 (3.4) 3 (6.7)
Cocci

TGEV ETEC

2 (1.7) 2 (4.4)
Cocci Crypto RV

1 (0.8) 1 (2.2)

Crypto

TGEV ETEC

5 (4.2) 4 (4.4)

Crypto RV

ETEC

1 (0.8) 3 (6.7)

Crypto

TGEV


TGEV

C. per. 1 (0.8)

Cocci RV TGEV ETEC 4 (3.4) 1 (2.2)
Cocci Crypto RV TGEV

C. per. 1 (0.8) 1 (2.2)
Crypto RV TGEV ETEC C. per. 1 (0.8)

(36) (50) (97) (111) (76) (23)

Total multiple infections 92 (78.8) 45 (100.0)

Output 3.2: Characterization of ETEC virulence factors.
Dr Thuy’s analysis of virulence factors in ETEC isolates obtained from both preweaning and
postweaning piglets also provided some interesting findings and comparisons between commercial
and village pigs. Firstly, given the information from Prof John Fairbrother’s laboratory on the
typical virulence gene profile possessed by the O8 strains expressing the new fimbrial antigen, Dr
Thuy was able to demonstrate that in the case of preweaning diarrhoea samples from commercial
pigs, the F19 strains were the second most common virulence profile identified after the typical F4
strains. These isolates were only identified in samples from commercial piggeries. In the case of the
postweaning diarrhoea samples, a large number of pathotypes were identified, but 73.2% of the
isolates possessed F18 fimbriae (and are likely to be serotype O141) whereas only 14.6% of the
isolates carried the more common F4 (and are likely to belong to serotype O149). F4 strains are
only associated with postweaning diarrhoea, whereas F18 is often associated with both postweaning
diarrhoea and oedema disease. As an indication of this, stx2, the toxin mostly associated with
oedema disease, was identified in 76.7% of the F18-positive isolates (63.4% of total isolates). In
most pig-producing countries, oedema disease has become quite rare, but it is obviously still a

F18/AIDA-I/STa/STb

3
F18/Paa/AIDA-I/STa/Stx2

2
F18/AIDA-I/STb/Stx2

1
F18/LT/Stx2

1
F18/AIDA-I/STa/STb/Stx2

4
F18/Paa/AIDA-I/STa/STb/Stx2

3
F18/Paa/STa/LT/Stx2

13
Paa/STa/LT/Stx2

2
Paa/STa/STb/LT/EAST1 4

AIDA-I/STb/EAST1 1

AIDA-I/STb/LT/EAST1 1


In 2008 and 2010, Dr Thuy undertook further laboratory training at The E. coli OIE reference
laboratory in Montreal under the direction of Prof John Fairbrother. Ten new gene primers were
introduced, and Dr Thuy has identified new E. coli virulence gene profiles in Vietnam and shown
why oedema disease is so prevalent in this country (discussed in 3.2 above).
5.3 Smallholder Benefits
Smallholder farmers have been the direct recipients of the following major project interventions:

1) Characterization, efficacy and safety testing of the NIVR E. coli vaccine. Currently, the majority
of commercial piggeries in Vietnam use Pfizer Littergard at a price of approximately $0.70 USD
per dose. The majority of smallholder farmers currently do not practice vaccination and could not
afford to routinely use the commercial vaccine, unless it was purchased by a co-operative of
farmers. The Vietnamese vaccine can be produced at $0.15 USD per dose. Several thousand doses
of this vaccine were provided free of charge to the smallholder farmers selected in 004/05VIE with
anecdotal reports confirming that it was safe and efficacious in preventing neonatal diarrhoea in the
first week of life.
2) Identification of the causes of pre-weaning diarrhoea in smallholder farms. The completed study
by Dr Do Ngoc Thuy confirmed that in smallholder piggeries, single aetiological agents of
preweaning diarrhoea were never identified and the most common, multifactorial diseases included
TGEV, rotavirus and enterotoxigenic E. coli. Coccidiosis, a completely preventable disease, was
also detected in over 30% of diarrhoea samples. This confirms that before any improvement in
piglet health can be attained, we must start back at the basics and teach farmers about sow
microclimate (cool, dry) and piglet microclimate (warm, dry). Stressed, wet piglets subjected to
drafts and poor air quality succumb to enteric diseases. This goal was attained in project 004/05VIE
encorporating knowledge and experience gained from 001/04VIE
3) Identification of ETEC pathotypes in smallholder farms. Characterization of the ETEC isolates
obtained from smallholder farmers confirmed that the major ETEC pathotype in smallholder farm
enterprises is F4:Paa:STa:STb:LT:EAST-1. This pathotype causes neonatal, 2-3 week old and
postweaning scour and can be controlled by appropriate antimicrobial treatment or a combination of
sow and piglet killed and live vaccines. Interestingly, the unusual F19 strains, so far, have not been
identified in smallholder farms, which may be due to the overwhelming presence of


Other opportunities for publicity are detailed in the 004/05VIE report.

5.6 Project Management
Operational project management was shared between The University of Queensland, Victorian
Department of Primary Industry (VicDPI), and the National Institute of Veterinary Research. The
University of Queensland was responsible for overall management of the project, with VicDPI
responsible for the delivery of training material, organisation of farm visits and data analysis. NIVR
was responsible for the production and testing of E. coli vaccines for the project and providing
advice and assistance on disease investigation and surveillance through laboratory diagnosis.
6. Report on Cross-Cutting Issues
6.1 Environment
Antibiotic usage and antibiotic resistance: Further development of the NIVR E. coli vaccine and
accurate identification of the causes of preweaning diarrhoea will lead to reduced reliance on
antimicrobials for the treatment and prevention of disease. The majority of smallholder farmers use
enrofloxacin, a top shelf antimicrobial for the treatment of diarrhoea, regardless of the cause. The
widespread availability of the NIVR vaccine would lead to a dramatic reduction in antimicrobial
usage with significant public health benefits.
Other major issues of biosecurity and processing of piggery effluent were explored in the
004/05VIE project final report.
6.2 Gender and Social Issues
These are more completely explored in the 004/05VIE project final report.
7. Implementation & Sustainability Issues
7.1 Issues and Constraints
Issue 1: Commercialisation of the NIVR vaccine.
Constraints: There is considerable investment in time and money involved in the path to
commercialisation of the vaccine. This could result in many delays until the eventually availability
of the vaccine for smallholder farmers.
Issue 2: Preweaning enteric management plan and creation of demonstration piggeries
Constraints: The focus on large piggeries in 001/04VIE was designed to identify herds with a

Philanthropies to be made in 2011 (budget ~$5.0 million) to continue the model in other provinces
and possibly into Laos and Cambodia, employing project farmers as trainers.
3) The CARD conclusion Workshop in November represents a critical phase to galvanise
stakeholders to take this project to the next level.
9. Conclusion
This final report details the considerable research and effort by NIVR and Australian scientists to
achieve project success within the logframe. The considerable time between project commencement
and this final report should take into account that the two projects (001/04VIE and 004/05VIE)
actually blended into a single entity that achieved great success for smallholder farmers in Central
Vietnam and developed a model for pig production and further expansion into other provinces of
Vietnam as well as neighbouring countries in South East Asia. Apart from the final characterization
of the F19 antigen, all objectives have been fulfilled. Strategies for commercialization of the
vaccine will provide a technology platform for taking additional locally produced vaccines through
to final licensing and will provide a continued funding stream for agencies such as NIVR to
continue their research. Persistent application of the CIP model on the selected test commercial
farms did show a difference in preweaning mortality compared to the control commercial farms and
the lessons learnt in technology transfer have been applied to 004/05VIE to work more specifically
with smallholder farmers.