Print this page | Close


Archive: 1999 Session - Appendix 8

1999 Session of the Research Group of the Standing Technical Committee of EuFMD


Serological surveillance for FMD in North Africa

D K J Mackay, Institute for Animal Health , Pirbright Laboratory
Ash Road, Pirbright, Woking, Surrey, GU24 ONF, UK


North Africa has experienced two epidemics of foot-and-mouth-disease (FMD) during the 1990Ôs, both due to FMD virus serotype O. The first epidemic occurred between 1989 and 1992 and resulted from the introduction of a Middle Eastern strain of FMD virus type O (Samuel et al. 1999). The epidemic started in Libya in late 1989 and rapidly spread across the border into Tunisia. In Tunisia, the virus entered at the time of lambing and caused devastating losses of up to 100% of the lambs in some flocks. Over 2,000 outbreaks were recorded in December 1989 alone. The epidemic showed a rapid westward progression reaching Algeria in May 1990 and Morocco in December of the same year. In all countries the disease showed a similar epidemiological pattern in that sheep were the major species affected. There were few reports of disease in cattle but, when cattle were affected, they presented with typical and severe lesions of FMD. Control was by vaccination which appeared to be effective at controlling the epidemic as the number of outbreaks progressively declined over the period 1990-92. In all countries affected, the incidence of the disease was seasonal with a rise in the number of outbreaks in the spring due to the increased movement of animals associated with the religious festival of ÍLÔAid el KebirÔ. The last outbreaks associated with the epidemic occurred in the final quarter of 1992.

The next occurrence of FMD in North Africa occurred in 1994 when limited outbreaks were reported in the north and south of Tunisia. Nucleotide sequence analysis showed the virus responsible to be similar to viruses from Egypt, Libya and the Middle East and distinct from the virus responsible for the 1989-92 epidemic (Samuel et al. 1999). This limited incursion probably resulted from illegal importation of animals from a neighbouring country and did not spread as mass annual vaccination had been carried out since 1989.

In February 1999 FMD due to type O was declared in Algeria. The disease rapidly spread both east and west. Only five days after the first declaration in Algeria an outbreak was reported in Oujda in the east of Morocco and only four days later disease was detected in central Tunisia. This epidemic was much less extensive than the 1989-92 epidemic, involving 160 outbreaks in Algeria, 11 in Morocco and 2 in Tunisia. Also striking was the fact that whereas the 1989-92 epidemic affected predominantly sheep, the 1999 epidemic affected almost exclusively cattle. Clinical disease was reported mainly in young steers which had either never been vaccinated or in which vaccinal antibody had waned. Older animals, which had been vaccinated in the past, were largely protected. Clinical signs were reported in sheep in Tunisia and sheep were also seropositive without clinical signs in both Tunisia and Morocco on farms where clinically affected cattle were detected. Nucleotide sequence analysis at the FAO/OIE World Reference Laboratory (WRL) for FMD, Pirbright showed that the virus responsible belonged to a distinct genetic lineage of type O viruses unique to West Africa (N.Knowles, IAH, Pirbright, unpublished). The virus must therefore have been transported from West Africa to North Africa in the spring of 1999 but the exact means of entry remains obscure.

This paper briefly reviews some of the surveillance work carried out in North Africa in the recent past and draws conclusions on the usefulness of serosurveillance for directing FMD control in North Africa and other, similar areas.



During the period 1992-96 a series of surveys were carried out, firstly in Morocco and subsequently in Tunisia. The surveys were collaborative ventures between the WRL for FMD, the veterinary services and the veterinary diagnostic laboratories of the respective countries, and the International Atomic Energy Agency, Vienna, which provided the funding for the technical co-operation. The 1999 survey was a collaboration between the Laboratorio de Sanidad Animal, Spain, and the Direction de lÔÃlevage, Morocco. Details of the individual surveys are summarised in Table 1.
In all surveys, sera were examined for antibody to FMD virus by liquid phase blocking (LPB) ELISA (Hamblin et al. 1986). Where possible, positive sera were further examined by VNT. In some cases, sera were examined for antibody to the non-structural (NS) proteins of FMD virus by agar gel immunodiffusion (AGID) against the virus infection associated antigen (VIAA), by immunoblotting against a range of recombinant NS proteins or by monoclonal antibody trapping (MAT) ELISA for antibody to 3ABC (De Diego et al. 1997).



Serological surveillance
The series of surveys carried out in small ruminants in Morocco between 1992 and 1996 showed that the prevalence of animals seropositive for antibody to FMD decreased with time (Fig 1). Frequency distributions of the titres of positive sera likewise showed a decrease in the proportion of sera positive at high titre over time (Fig 1). The surveys were designed to exclude vaccinated animals either by sampling only unvaccinated flocks or by selecting unvaccinated animals on the basis of age. These findings are therefore consistent with a gradual reduction in exposure of small ruminants to infection but suggest that the virus did not stop circulating immediately after the elimination of the last declared outbreaks in 1992. The surveys are not representative of the country as a whole as each successive survey concentrated only on those areas in which seropositive animals were detected in the previous survey. This sequential Ífocussing inÔ on residual areas of positivity gives the impression that the prevalence of seropositive animals remained high even in 1996, but this would not have been the case had the surveys been nationwide in scope throughout.

In all surveys carried out between 1992 and 1996, there was very clear clustering of seropositive animals within flocks and of seropositive flocks within provinces. In general, the geographical distribution of flocks, or villages, containing seropositive animals was consistent with the known distribution of clinical disease in the 1989-92 epidemic. In some cases, seropositive flocks were also detected in provinces adjacent to previously infected areas or in areas along the main east to west trade route across the country. In all surveys, the great majority of sera positive by ELISA were positive by VNT. In the 1992, survey representative sera which were positive by ELISA were examined in the AGID tests and the majority were also positive for antibody to VIAA. In contrast, ELISA-positive sera from the 1994 and 1996 surveys were examined for antibody to NS proteins by immunoblotting and/or ELISA and all were negative.

In 1999, a survey was carried out of sera from small ruminants in provinces in which outbreaks had occurred, in neighbouring provinces, and in the remainder of the country. The overall prevalence of positive sera was low, with only 0.4% of sera being positive in the definitive VNT. As with previous surveys, there was a clear clustering of positive sera within villages, and of positive villages within certain provinces. Almost all seropositive flocks were detected in provinces in which clinical disease had occurred. A small numbers of positive sera were also detected in neighbouring regions and in provinces along the border with Algeria. No positive sera were detected in the regions of the country considered to be free of the disease. All positive sera were examined by AGID and the 3ABC MAT ELISA. Antibody to NS proteins were detected only in sera from regions where outbreaks had occurred.


Virus isolation
Although serology is useful as an indirect indicator of viral activity, the continued presence of FMD virus can only be confirmed by isolation of the virus itself or, more recently, by identification of viral RNA using RT-PCR. In a study in Tunisia, paired serum and probang samples were collected from a herd of cattle following the outbreak of FMD in May 1994. Out of 15 cattle examined, virus was isolated from three in May, five in July, one in September and none thereafter (Table 1: Hammami and Mackay, unpublished). The animals had been vaccinated immediately FMD was recognised in the herd, resulting in high initial antibody levels which fell with time after vaccination. In some animals, titres rose again in November following re-vaccination. As expected, the carrier state was unrelated to the level of circulating antibody. In Morocco between 1992 and 1996, several attempts were made to isolate virus from sheep from flocks located close to declared outbreaks or from flocks in which clusters of seropositive animals had been detected. Paired serum and probang samples were collected. Although some animals were seropositive, FMD virus was never isolated from any animal.



Studies on the efficacy of vaccination
Although vaccination was the main method of disease control, few systematic studies were carried out initially to examine the efficacy of the vaccines used under field or experimental conditions. A pilot study to determine the prevalence of cattle with protective titres against FMD virus types O and A was performed in Morocco in 1994. Taking an LPB ELISA titre of 1:100 as protective (Hamblin et al. 1987), only 25% of cattle had protective titres to type O and only 31% to type A. There was considerable variation between farms in the proportion of animals with protective titres. Surprisingly, an increase in mean titre with age was not observed, despite the fact that older animals had received more vaccinations than young ones. In 1995, a more detailed study was undertaken to examine the efficacy of vaccination of sheep in Tunisia (Hammami et al. 1998). The study was limited to sheep on state farms as the management and record keeping on these premises was good. The results showed that the overall level of immunity was adequate, with 66% of animals still having protective titres six months after vaccination. The most important factor in the level of immunity obtained was the age of the animal at vaccination. It was clear that a primary vaccination course of two injections would be beneficial compared to the single injection carried out at the time.

In all studies of the efficacy of vaccination, there was very wide variation in the proportion of animals with protective titres between flocks. In the case of the Tunisian survey the proportion varied from 0 to 100%. In some flocks there was a high prevalence of animals with high titres before vaccination and in other herds rises in titre were detected which were not associated with vaccination. Representative sera from these herds were examined for antibody to 3ABC by MAT-ELISA and all were negative. As a result of this survey a more detailed study is currently underway in Tunisia to determine the kinetics of the antibody response to primary and secondary immunisations and to compare vaccine obtained directly from the manufacturer with that obtained from the Veterinary Service at the point of use. It is also planned to examine antibody titres in privately owned flocks where the level of management is more variable.



These surveys demonstrate both the usefulness and the limitations of serosurveillance in FMD control. It is clear that vaccination programs should always be accompanied by serological surveys to investigate their effectiveness. In all the surveys of vaccine efficacy undertaken there was wide variation between herds in the proportion of animals with protective titres. The fact that in each survey some herds had high levels of protection suggests that the vaccines themselves were effective. It appears therefore that deficiencies arose at some stage in the chain between the arrival of the vaccine in the country and its administration to animals. It was apparent in Morocco in 1994 that once yearly vaccination was not sufficient to maintain high levels of immunity all year round and, in Tunisia, it was clear that revaccination of primo-vaccinates would be highly beneficial. This kind of information is extremely useful in practical terms and can only be obtained by serological follow-up of vaccination.

The surveys between 1992 and 1996 in Morocco clearly showed that FMD was widespread in the sheep population. It was also clear from the decrease in prevalence of seropositive animals in successive surveys that the extent of exposure of animals to FMD virus decreased with time after the last epidemic. What was much more difficult to interpret was the continued detection of seropositive animals up to four years after the last outbreak. Animals could have been seropositive due to residual maternally derived antibody (MDA), due to vaccination, or due to the continued circulation of FMD virus in a sub-clinical form. MDA was unlikely, as only animals over 6 months were included in the survey. Previous vaccination was a possible cause. Vaccinated animals could have been sampled if either the age restrictions were ignored or unreported vaccination had taken place, both of which were possible. The possibility of subclinical infection was the most important and the most difficult to confirm or refute. In the 1992 survey, most ELISA positive animals were also positive for antibody to NS proteins, confirming previous exposure to replicating virus. In the surveys between 1994 and 1996, although sera were positive by ELISA and VNT they were negative for antibody to NS proteins. Unfortunately, not all sheep infected with FMD virus seroconvert to NS proteins, particularly if infection is subclinical, as is often the case in small ruminants (Brocchi et al. 1998; Mackay, unpublished). Currently, the persistence of FMD virus can still only be definitively confirmed by virus isolation from samples of oesophagepharyngeal (probang) fluid. This is extremely expensive and difficult to organise in countries with limited facilities to isolate FMD virus, but Table 2 shows that it can be successfully achieved. In the absence of isolation of FMD virus from Morocco between 1992 and 1996, it was not possible to confirm that the virus was present. However, following the epidemic of FMD type O in Greece in 1994, it appeared that virus continued to circulate for some time after clinical disease had ceased and that the disease was ultimately self-limiting (Mackay et al. 1995). It is possible that a similar phenomenon occurred in Morocco. The epidemiology of FMD in sheep is currently under experimental investigation at the IAH, Pirbright.

The prevalence of seropositive small ruminants was much lower following the epidemic of 1999 than the epidemic of 1990-92 and the geographical distribution of positive flocks was more restricted. These findings are consistent with the fact that clinical disease was observed only in cattle in 1999, but sheep were the major species affected in 1990-92. The 1999 survey also showed that, although some subclinical infection of small ruminants had occurred, the disease was not widespread in these species. A follow-up survey is planned for the autumn of 1999 to study the evolution of the disease with time. In the 1992 and 1999 surveys, tests for antibody to NS proteins were extremely useful in confirming exposure of small ruminants to infection. Tests for antibody to NS proteins will be increasingly applied to detect evidence of virus circulation following outbreaks (Brocchi 1998, Sorenson, 1998). However, further field and experimental studies are required to determine the sampling frame that is necessary to detect animals, particularly sheep, positive for antibody to NS proteins with the required degree of statistical certainty. It is likely that different, more rigorous sampling regimes are required when carrying out surveillance for antibody to NS proteins than for antibody to structural proteins. Also, there is still a need for techniques which identify the agent itself, such as virus isolation or PCR, as these techniques remain the only definitive way of detecting persistently infected animals..



The work of the following organisations in organising, executing, analysing and funding the serological surveys is acknowledged: Direction de lÔElevage, Morocco; Laboratoire dÔAnalyses et de Recherches Veterinaires, Casablanca; Institut de la Recherche Veterinaire de Tunisie, Tunis; Laboratorio de Sanidad Animal, Algete, Spain; International Atomic Energy Agency, Vienna and the Veterinary Services of Morocco and Tunisia.



  • Brocchi, E., De Diego, M., Berlinzani, A., Gamba, D., De Simone, F. Í Diagnostic potential of Mab-based ELISAÔs for antibodies to non-structural proteins of foot-and-mouth disease virus to differentiate infection from vaccinationÔ. Veterinary Quarterly (1998) Vol 20/S2/S20.
  • De Diego, M., Brocchi, E., Mackay, D., De Simone, F. (1997). ÍThe use of the non-structural polyprotein 3ABC of FMD virus as a diagnostic antigen in ELISA to differentiate infected from vaccinated cattleÔ. Archives of Virology (1997) 142; 2021-2033.
  • Hamblin, C., Barnett, L.T.R., and Hedger, R.S. A new enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against foot-and-mouth disease. Journal of Immunological Methods (1986) 93; 115-121.
  • Hamblin C., Kitching R. P., Donaldson A. I., Crowther J. R., Barnett I. T. R., Enzyme-linked immunosorbent assay (ELISA) for the detection of antibodies against foot-and-mouth disease virus. III. Evaluation of antibodies after infection and vaccination. Epidemiology and Infection (1987) 99; 733-744.
  • Hammami S, Mackay D K J, Seghaier C, Aouina Th, Tlatli A, Haouala K, Sanaa M. ÍEvaluation of the vaccination campaign against foot-and-mouth disease in sheep in TunisiaÔ. Proceedings of the Session of the Research Group of the Standing Technical Committee of the European Commission for the Control of Foot-and-Mouth Disease, UK, 14-18th September 1998. Appendix 26, p196.
  • Mackay, D., Newman, B., Sachpatzidis, A. ÍEpidemiological analysis of the serological survey for antibody to FMD virus, Greece 1994'. Report to FAO 1995.
  • Sorensen, K.J., Madsen, K.G., Madsen, E.S., Salt, J.S., Nqindi, J., Mackay, D.K.J. ÍFoot-and-mouth disease: Detection of antibodies against the non-structural proteins 3D, 3AB and 3ABC by ELISA. Differentiation between antibodies induced by infection and by vacccinationÔ. Archives of Virology (1998) 143 (8); 1461-1476.
  • Samuel, A.R., Knowles, N.J., Mackay, D.K.J. Genetic analysis of type O viruses responsible for epidemics of foot-and-mouth disease in North Africa. Epidemiology and Infection (1999) 122; 529-538.