The design and basic descriptive epidemiology of the IDEAL project is fully described in Bronsvoort et al.3. To give the IDEAL database a context, we briefly describe the study in relation to the samples taken.
The study population
Between 2007–2009, 548 free-grazing indigenous East African Shorthorn Zebu calves in Western Kenya were recruited. The project was based in Busia, a town in the west of Kenya on the border with Uganda. The town had a veterinary lab that was able to be used and developed by the project.
Calves were recruited using a stratified two-stage random cluster study design (Fig. 1). In the first stage, a weighted random sample of 20 sublocations was selected from across the four agro-ecological zones represented in the project area. This area was roughly 45 × 90 km2. This area was chosen because each area would be drivable from Busia and back to the lab and all visits completed within a day.
In the second stage, approximately 28 3–7 day-old calves were randomly recruited from each sublocation. The sublocations were visited on a rolling 5-week cycle to ensure there was an even distribution of calves recruited across space and season over the study period. Only one calf per dam was recruited and a farmer could only have one calf at a time in the study. Recruited calves were followed for their first year (51 weeks) of life. A calf was selected at random from all those available and eligible on the day of the recruitment visit.
As described in Bronsvoort et al.3, each calf had to meet the following inclusion criteria to be enrolled in the study: (1) it was between 3 and 7 days old at recruitment; (2) it was not born as a result of artificial insemination of the dam; (3) the dam was not managed under zero-grazing conditions (as this is likely to reflect cattle associated with dairy production and potential exotic genetics and thus not be representative of the traditional small holder farming system); (4) it did not have any congenital deformities (the project was interested in infectious causes of disease and the association with host genetics, rather than direct genetic disease). Recruitment was conditional on the farmer allowing access to the calf, willingness to report clinical episodes to the project team, and agreeing not to self-treat their calves. Farmers were compensated at a rate agreed with the staff of the District Veterinary Office; this comprised the estimated cost of raising the calf for one year as calves were nominally owned by the project for that year. Owners were free to refuse to participate3 and consent was obtained in a language in which the participant was confident.
There were two periods during the IDEAL project when sampling and recruitment was suspended. In the first, field work was suspended for 6 weeks in 2008 due to political unrest. This resulted in a small number of calves missing one or two 5 weekly visits. The second was due to an extended holiday period that resulted in staffing problems in 2009/2010.
Data collection
Below, we briefly describe how the IDEAL data was collected. See Bronsvoort et al.3 for more details.
Routine clinical examination of calves
At the recruitment visit, a household questionnaire was completed by interview with the calf owner or head of the household. This questionnaire collected information about the farmer and the farm, such as the type of livestock kept, and animal management practices. Calves were raised according to the farmer’s practices.
Calves received a routine systematic physical examination from a veterinary surgeon at the recruitment visit and these examinations where repeated every 5 weeks until the calf was 51 weeks old. Some routine measurements were taken during the exam (body weight and lymph node width). All abnormalities were noted and all areas were noted as checked. Ectoparasites found were identified. During each visit a standard set of biological samples were taken for further laboratory analysis. In addition, a questionnaire was carried out at each visit to update the IDEAL project about other activities on the farm such as illness or treatment of the other livestock and animal movements. At the final visit, physical phenotype was recorded following a standardised format. The study design, sampling procedure, clinical examinations and sample collection are summarised in Figs. 2 and 3, and can also be read in detail in the supporting documentation online (http://data.ctlgh.org/ideal/).
Clinical episodes and post-mortem examinations of calves
If a calf became unwell, the farmer was asked to contact the IDEAL project veterinary surgeon. The calf was visited and a full clinical examination and history was taken. Both routine and additional appropriate specific samples were collected based on the syndrome observed3. If a calf was deemed to be suffering and if that suffering would be alleviated with treatment, that treatment was given and the calf was no longer visited and was censored from the study from that point. If calves were suffering and that could only be alleviated by euthanasia, this was carried out and the calf was examined post-mortem following the standard protocol. Following the death or euthanasia of a calf, a full gross post-mortem examination was carried out using standard veterinary procedures and clinically appropriate samples collected for further analysis. Cause of death was attributed using a panel of experts with access to all available diagnostic results and the necropsy report. This was carried out for all calves that died at a single timepoint after the close of the project.
Examination of the dams
A limited clinical examination of the dam was performed at recruitment and in subsequent routine visits until the calf was weaned. At these visits the girth was measured and body condition was scored and the udder was examined for evidence of lesion or mastitis that could affect calf nutritional intake. Phenotypic measurements of coat colour and pattern, horn length and shape, ear shape, size of hump and dewlap was recorded at recruitment3.
Laboratory analysis
Blood samples collected into EDTA tubes were used for differential blood cell counts, performed using the pocH-100iV Diff (Sysmex®, Europe GMBH). The haematological parameters investigated are listed in Online-only Table 1. In addition to the automated blood analysis, EDTA unclotted samples were used to make thin blood smears for manual differential cell counts. These smears were transported to the University of Pretoria, South Africa, where blood smears were stained with Diff Quick (Kyron, South Africa) for differential counts. Packed cell volume (PCV) was measured manually using a Hawksley micro-hematocrit reader4. Total serum protein (TSP) was measured from 100 μL serum using a refractometer (model RHC-200ATC, Westover Scientific).
Peripheral ear vein blood smears were collected and examined for haemoparasites by microscopy. Thin smears were fixed using methanol and stained using Giemsa. Thick smears were directly stained. One hundred fields were examined under an oil immersion lens. Haemoparasites present were identified to genus level.
Reverse line blot (RLB) hybridization assay was performed as previously described5 to detect tick-borne haemoparasites in the blood (Theileria, Anaplasma, Ehrlichia, and Babesia (Online-only Table 1))5.
The p104 nested PCR was carried out on calves where ECF was suspected on clinical grounds to specifically identify T. parva6.
Whole blood collected in EDTA was mixed in sodium EDTA tubes in a 1:1 ratio with ‘magic buffer’ (which acted as an anti-coagulant, anti-fungus, anti-bacterial and preservative; Biogen Diagnostica, Villaviciosa De Odon, Spain) at the recruitment visit in readiness for genomic analysis. DNA was extracted from these samples using the Nucleon Genomic DNA extraction kit (TepnelnLife Sciences, Manchester, UK). The Illumina® BovineSNP50 v. 1 BeadChip (Illumina Inc., San Diego, CA, USA) used to genotype the cattle. Genotyping of the 548 calves was carried out at the USDA-ARS bovine functional (Beltsville, MD, USA) and GeneSeek (https://genomics.neogen.com) laboratories using the genome assembly v3.0. In addition, due to the cost of sequencing at the time, a subset of 114 cattle were genotyped using the Illumina® BovineHD Genotyping BeadChip.
Faecal samples were also collected during the study and these where routinely screened using the standard Baermann and McMasters protocols7. The number of strongyle eggs per gram of faeces was evaluated using the McMasters counting technique. These could be read to the nearest 50 eggs per gram. Sedimentation was carried out for detection of fluke eggs and larval cultures were used to speciate strongyle eggs7. Species where reported at the highest level; strongyle eggs/strongyloides/coccidia/nematodirus. See Online-only Table 1 for more details on the parasites detected.
Serum samples were collected from blood collected into plain vacutainer tubes. These were stored in duplicate for serological analyses. Species-specific antibody response enzyme-linked immunosorbent assays (ELISAs) were performed and analysed according to the manufacturer’s instructions.
The full list of pathogens and viruses for which the cattle have so far been screened can be found in Online-only Table 1. The prevalence of pathogens currently identified across the whole study period and at each visit is presented in Figs. 4 and 5.
Since the database is linked to a biobank, the list of pathogens tested and screened for is continuously updated as new tests are performed or new tools are developed.
Source: Ecology - nature.com
