A sero-survey of rinderpest in nomadic pastoral systems in central and southern Somalia from 2002 to 2003, using a spatially integrated random sampling approach.

A cross-sectional sero-survey, using a two-stage cluster sampling design, was conducted between 2002 and 2003 in ten administrative regions of central and southern Somalia, to estimate the seroprevalence and geographic distribution of rinderpest (RP) in the study area, as well as to identify potential risk factors for the observed seroprevalence distribution. The study was also used to test the feasibility of the spatially integrated investigation technique in nomadic and semi-nomadic pastoral systems. In the absence of a systematic list of livestock holdings, the primary sampling units were selected by generating random map coordinates. A total of 9,216 serum samples were collected from cattle aged 12 to 36 months at 562 sampling sites. Two apparent clusters of RP seroprevalence were detected. Four potential risk factors associated with the observed seroprevalence were identified: the mobility of cattle herds, the cattle population density, the proximity of cattle herds to cattle trade routes and cattle herd size. Risk maps were then generated to assist in designing more targeted surveillance strategies. The observed seroprevalence in these areas declined over time. In subsequent years, similar seroprevalence studies in neighbouring areas of Kenya and Ethiopia also showed a very low seroprevalence of RP or the absence of antibodies against RP. The progressive decline in RP antibody prevalence is consistent with virus extinction. Verification of freedom from RP infection in the Somali ecosystem is currently in progress.

Epidemiological assessment of rinderpest surveillance and control in Uganda between 1990 and 1998.

Based on passive and active data, we report on an epidemiological assessment of surveillance and control of rinderpest (RP) in Uganda between 1990 and 1998. Active data were collected by administration of questionnaires to animal health personnel and their auxiliaries and to stockowners in six selected districts of eastern and northeastern Uganda. Passive data were extracted from vaccination and seromonitoring reports, and from field and laboratory reports. RP events were classified as "confirmed outbreaks", "suspected outbreaks" and "rumours". The classification of 56% of the RP events as "suspected outbreaks" indicates the difficulty in investigating disease outbreaks in Uganda. Although vaccination coverage and seroprevalence were <85% (the recommended target), they nevertheless corresponded well-reflecting effective vaccination. However, because of the low seroprevalence, a sizable population of cattle in Uganda remained at risk of RP. The agreement between the local and national disease reporting systems was low-to-moderate (kappa=0.39); this indicates inefficiency in disease reporting. Risk factors for RP outbreaks were cattle raids and communal grazing. Based on overlaid thematic maps of seroprevalence, vaccination coverage and RP events, close spatial and temporal associations were observed between cattle raids, transhumance and outbreaks and rumours. The high-risk areas were in the eastern and northeastern parts of the country. The results of this study support a phase approach of following the OIE pathway.

The control of rinderpest in Tanzania between 1997 and 1998.

In January 1997, Tanzania requested international assistance against rinderpest on the grounds that the virus had probably entered the country from southern Kenya. Over the next few months, a variety of attempts were made to determine the extent of the incursion by searching for serological and clinical evidence of the whereabouts of the virus. At the clinical level, these attempts were hampered by the low virulence of the strain, and at the serological level by the lack of a baseline against which contemporary interpretations could be made. Once it became apparent that neither surveillance tool was likely to produce a rapid result, an infected area was declared on common-sense grounds and emergency vaccination was initiated. The vaccination programme had two objectives, firstly to prevent any further entry across the international border, and secondly to contain and if possible eliminate rinderpest from those districts into which it had already entered. On the few occasions that clinical rinderpest was subsequently found, it was always within this provisional infected area. Emergency vaccination campaigns within the infected area ran from January to the end of March 1997 but were halted by the onset of the long rains. At this time, seromonitoring in two districts showed that viral persistence was still theoretically possible and therefore a second round of emergency vaccination was immediately organized. Further seromonitoring then indicated a large number of villages with population antibody prevalences of over 85%. These populations were considered to have been 'immunosterilized'. Although no clinical disease had been observed in them, it was decided to undertake additional vaccination in a group of districts to the south of the infected area. Serosurveillance indicated that rinderpest could have been present in a number of these districts prior to vaccination. Serosurveillance in 1998 suggested that numerous vaccinated animals had probably moved into districts outside the infected and additional vaccination areas, but did not rule out the continued presence of field infection.

Use of dot-immunobinding assay for visual detection of rinderpest antibodies in vaccinated cattle.

A dot-immunobinding (DIB) assay was used to detect rinderpest antibodies in cattle vaccinated with Kabete 'O' strain vaccine, using purified rinderpest virus. Of 120 serum samples from vaccinated and non-vaccinated animals, rinderpest antibodies were detected in 80%, 88.4% and 91.6% of samples at 2, 3 and 4 weeks postvaccination respectively. All the serum samples from non-vaccinated animals were negative. The DIB results had a good correlation with those of the micro neutralisation test. The technique is simple, easy to perform and suitable for routine use in detecting rinderpest antibodies following vaccination.

Fluid, electrolyte and haematological changes in experimental peste des petits ruminants virus infection in West African dwarf goats.

The changes in plasma volume, blood volume, erythrocyte indices, leucocyte indices and serum biochemical values during experimental pestes des petits ruminants (PPR) virus infection of West African dwarf goats (Fouta djallon) were investigated. There was a progressive decline of the plasma volume and serum sodium and potassium values in all the infected goats as from day 5 post-infection and this became quite marked between days 10 and 23 post-infection. The fall was severest in those animals which had diarrhoea. There were no significant changes in the chloride, bicarbonate, phosphate, urea, total protein and albumin values of the goats during the course of the diseases. Haemoconcentration and apparent polycythaemia in the early and terminal stages of the disease were associated with severe dehydration which characterized the infection. There was initial neutrophilic leucocytosis during the phase of fever followed by marked lymphopaenic leucopaenia which progressed terminally in most of the infected goats. The destructive effect of the virus on lymphocytes is suggested as a cause of lymphopaenia.

The differential growth of virulent and avirulent strains of rinderpest virus in bovine lymphocytes and macrophages.

Rinderpest virus (RV) grew readily in cultures of purified bovine peripheral blood lymphocytes and udder macrophages. The growth of three strains of RV was compared and there appeared to be a relationship between increasing virulence and increased ability to infect lymphocytes and macrophages. The proportion of infected cells as determined by the presence of virus antigens was a better indicator of affinity between a strain and cell type than production of new infectious virus. RV grew better in populations of predominantly T lymphocytes than in T-depleted cultures. Although RV could infect 100% of cells in macrophage monolayers, it did not appear to infect more than about 30% of cells in lymphocyte cultures. Virulent RV grew more readily in bovine than caprine or ovine lymphocytes, whereas virulent peste des petits ruminants virus (PPRV) grew better in lymphocytes from sheep and goats. There was no marked difference in the growth of either virus in lymphocytes from uninfected or recently convalescent animals.

Serological studies with the virus of peste des petits ruminants in Nigeria.

It was possible to distinguish separate serological responses when experimental goats were inoculated with either rinderpest virus or peste des petits ruminants virus. Examination of field samples established that peste des petits ruminants occur commonly in Nigerian sheep and goats although some villages have escaped recent infection. There was no evidence of infection with rinderpest in these animals.