Caractérisation zootechnique et formule barymétrique de la race zébu Azawak à Ménaka au nord du Mali

Dans le nord du Mali, l’élevage bovin est soumis à des contraintes climatiques (aridité) et socioéconomiques (accès à l’eau potable et abris de nuit, parc de vaccination, marché à bétail) freinant son essor. De plus, dans un milieu aride, où la ressource alimentaire évolue fortement au cours de l’année, les éleveurs sont en quête permanente de satisfaction des besoins alimentaires des animaux (petite transhumance, séjour sur terres salées). En fin de saison sèche, les animaux sont très amaigris, ce qui affecte la viande en termes de flaveur et de qualité. Des études ont été menées au Niger, en Côte d’Ivoire et au Burkina Faso pour évaluer les performances de production de la race Azawak (Dodo et al., 2001 ; Marichatou et al., 2005 ; Sokouri et al., 2010). Cependant, rares sont celles qui ont concerné la race zébu Azawak au Mali.

■ CONCLUSION La modélisation de la croissance par l'équation de Gompertz a permis de choisir des animaux pour le testage. Le périmètre thoracique, paramètre facilement mesurable, a constitué un élément estimant le mieux Summary Touré A., Antoine-Moussiaux N., Kouriba A., Leroy P., Moula N. Zootechnical characterization and barymetric formula of the Azawak zebu breed in Menaka, Northern Mali The objective of this study was to contribute to the preservation and improvement of the performance of the Azawak zebu in Mali by proposing an estimation of the growth and reproduction performances of this cattle breed reared in the north of the country. Monitoring involved 1129 animals in four townships in Menaka Circle, between November 2009 and February 2010. At 30 days, males weighed on average 39 ± 10 kg and females 32 ± 4 kg. No significant differences (p > 0.05) were recorded between the adult weights of bulls (370 ± 88 kg) and cows (327 ± 46 kg). The age at first calving was 50 ± 11 months with a calving interval of 16 ± 4 months. The correlations of weight with the thoracic girdle (r = 0.95) and the height at the withers (r = 0.94) were the highest. Barymetric equations were developed for each animal category (male calf / female calf, heifer / young bull, cow/bull) using polynomial weight regressions (y) on the thoracic girdle (x). These could enable livestock agents to establish a conversion table of the thoracic girdle into live weight or a barymetric measuring tape to facilitate husbandry and sanitary monitoring of animals in the field.

■ INTRODUCTION
Au Burkina Faso, l'élevage de la pintade locale a une grande importance et de grandes opportunités de développement. En revanche, cette volaille n'a pas encore fait l'objet d'études approfondies tant de la part des techniciens de développement rural que de celle de la recherche dans ce pays. L'absence de recommandations techniques adaptées restreint la valorisation de cette production et de nombreuses contraintes limitent les producteurs dans leurs activités.
Hematological and serum biochemical evaluations are important in the diagnosis of diseases because of their predictive value of pathological changes in vital internal organs and deviation from normal caused by invasion of the body by pathogens (Stockham and Scott, 2008). There have been many reports on hematological and serum biochemical changes associated with TB in cattle (Amin et al., 1990;Rao et al., 1992;Kumar et al., 1994;Javed et al., 2006;Olivia et al., 2008), but the present study assessed hematological and serum biochemical alterations in slaughtered cattle with gross tuberculous lesions at Nsukka abattoir.

Summary
This study assessed hematological and serum biochemical alterations associated with the occurrence of tuberculous lesions in slaughtered White Fulani cattle at Nsukka abattoir, Enugu State, Nigeria. Diagnosis was confirmed by the immunochromatographic technique and histopathology. Out of 567 cattle examined, ten (1.76%) had tuberculous lesions. The tuberculous cattle had normocytic normochromic anemia, leukocytosis, lymphocytosis, eosinophilia and an increased erythrocyte sedimentation rate. They also had significantly lower (p < 0.05) serum alanine aminotransferase, albumin and urea, and significantly higher (p < 0.05) serum globulin than the apparently healthy control animals. We concluded that the occurrence of tubercles in cattle was associated with alterations in hematological and serum biochemical parameters, which may be relevant to the establishment of an antemortem diagnosis of tuberculosis. Blood samples were collected from the jugular vein. They were anticoagulated with sodium ethylene diamine tetra-acetic acid (EDTA) for hematology, whereas those for serum biochemistry were collected in plain glass test tubes and allowed to clot. A postmortem examination was performed on the carcasses for presence of tubercles , and blood samples from cattle with tubercles (positive cases) were retained. For each positive case, blood samples from four apparently healthy (non-tuberculous) cattle were also collected as control.
The diagnosis of bovine TB was further confirmed by the histopathological evaluation of organs with tubercles (Varello et al., 2008) and the immunochromatographic technique using Anigen Rapid Bovine TB Antibody Test Kit (Bionote, South Korea), as its sensitivity and specificity are very high .
The packed cell volume (PCV) was determined by the microhematocrit method (Thrall and Weiser, 2002). Hemoglobin (Hb) concentration was measured by the cyanmethemoglobin method (Burtis et al., 2008). Red blood cell (RBC) and total leukocyte counts were conducted with the hemocytometer method. Differential leukocyte counts were performed on air-dried thin blood smears stained by the Leishman technique and enumerated with the battlement method (Thrall and Weiser, 2002). The mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH) and mean corpuscular hemoglobin concentration (MCHC) of the erythrocytes were calculated with the standard formula (Thrall and Weiser, 2002). The erythrocyte sedimentation rate (ESR) was determined with the modified Wintrobe method (Thrall and Weiser, 2002). All serum biochemical determinations were carried out with the standard colorimetric method using commercial test kits (Quimica Clinica Applicada, Spain). Serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activities were determined by the Reitman-Frankel colorimetric method, and the alkaline phosphatase (ALP) activity by the phenolphthalein monophosphate method (Burtis et al., 2008). Serum total bilirubin was determined by the modified Jendrassik-Grof method, and total protein by the direct Biuret method (Burtis et al., 2008). Serum albumin was determined by the bromocresol green method, and globulin was calculated by subtraction of the value of albumin from that of total serum protein (Burtis et al., 2008). Serum total cholesterol was determined by the enzymatic colorimetric method, serum urea by the modified Berthelot-Searcy method, and creatinine by the modified Jaffe method (Burtis et al., 2008).
Data analysis was performed with the statistical Software Package for Social Sciences (SPSS) version 16. The hematological and serum biochemical parameters of control and tuberculous cattle were compared with Student's t-test (significance level of p < 0.05).

■ RESULTS AND DISCUSSION
Out of the 567 cattle, ten (1.76%) had tubercles on the lungs, liver, spleen, gall bladder and/or lymph nodes which were confirmed by the histopathology and immunochromatographic technique. The positive cases were six male and four female adult White Fulani cattle. The 1.76% prevalence of tuberculosis in the present study was comparable to the reported prevalences of 1.1% in Maiduguri (Raufu and Ameh, 2010) and 1.9% in Makurdi (Ejeh et al., 2014a), in Nigeria. It was, however, relatively lower than those of 2.8% (Igbokwe et al., 2001), 4.05% (Aliyu et al., 2009) and 2.9-6.5% (Ejeh et al., 2014b) reported in areas of Northern Nigeria. These differences may be associated, on one hand, with climatic conditions that facilitate bovine tuberculosis persistence and transmission, as they vary from north to south of Nigeria depending on the geographical location, and, on the other hand, with the methods of diagnosis of positive cases.
The hematological parameters are summarized in Table I. Mean PCV, RBC count and Hb concentration were significantly lower (p < 0.05) in tuberculous cattle than in apparently healthy control animals, whereas mean ESR was significantly higher (p < 0.05) in tuberculous cattle than in control animals. Mean total leukocyte, lymphocyte and eosinophil counts were significantly higher (p < 0.05) in tuberculous cattle than in control, but the basophil count of tuberculous cattle was significantly lower (p < 0.05) than that of control. There were no significant (p > 0.05) differences between means of MCV, MCH, MCHC, neutrophil and monocyte counts in tuberculous cattle and in control cattle.
Erythrocytic alterations in tuberculous cattle showed normocytic normochromic anemia, indicated by significantly lower means of PCV, RBC count, and Hb concentration without any significant changes in their MCV, MCH and MCHC (Stockham and Scott, 2008). The anemia could have been caused by the chronic disease condition and the suppression of erythropoiesis by inflammatory mediators (Weiss, 2002;Lee et al., 2006). The finding of normocytic normochromic anemia is in agreement with other reports in cattle (Rao et al., 1992;Kumar et al., 1994) and humans with tuberculosis (Lee et al., 2006). The higher ESR could have been caused by tissue destruction associated with the formation of granulomas in the parenchyma of the lungs, liver, spleen and lymph nodes (Stockham and Scott, 2008). It agreed with reports on buffaloes (Amin et al., 1990) and humans (Olaniyi and Aken'ova, 2003;Olivia et al., 2008), but disagreed with another report on buffaloes (Javed et al., 2006). Leukocytosis associated with lymphocytosis and eosinophilia in the present study could have been due to an antigenic  stimulation caused by active chronic Mycobacterium bovis infection (Stockham and Scott, 2008). This finding agreed with earlier reports in humans (Morris et al., 1989) and cattle (Javed et al., 2006) with tuberculosis.
Serum biochemical parameters are summarized in Table II. Means of serum ALT activity, albumin and urea levels of tuberculous cattle were significantly lower (p < 0.05) than those of control, but serum globulin levels were significantly (p < 0.05) higher in tuberculous cattle than in control. Means of serum AST and ALP activities, total protein, total cholesterol, creatinine and total bilirubin levels showed no significant (p > 0.05) differences between tuberculous and control cattle. The alterations in serum ALT activity, albumin and urea levels could have been caused by lesions in the liver and loss of its synthetic capacity (Stockham and Scott, 2008). Hypoalbuminemia has been reported in humans with tuberculosis (Morris et al., 1989). Increased serum globulin levels could have been caused by high levels of immunoglobulin production stimulated by chronic antigenic challenge by tubercle bacilli, similarly to what has been reported in humans with tuberculosis (Damburam et al., 2012).

■ CONCLUSION
Bovine tuberculosis was associated with normocytic normochromic anemia, increased ESR, total leukocyte, lymphocyte and eosinophil counts, and serum globulin levels, as well as decreased serum ALT activity, albumin and urea levels. Therefore, blood tests may be useful to establish the diagnosis of bovine tuberculosis in endemic areas.

Summary
Tuberculosis (TB) is a chronic bacterial disease of humans and animals. It is characterized by the progressive development of specific granulomatous lesions in affected organs. Human TB is endemic in EasternSudan. However, knowledge on the epidemiology of TB in ruminants is scarce. In a six-month study from June to November 2014, a total of 2304 carcasses of cattle, sheep, goats and camels slaughtered at the East and West Gaash slaughterhouses of Kassala were inspected to investigate TB prevalence. Only 0.1% (n = 2) of the carcasses had suspicious TB lesions. These lesions were solely found in carcasses of sheep, in the liver, lungs, and peritoneal cavity. The samples collected from the lesions were investigated for the presence of mycobacteria, which were found in one of the two carcasses. The grown bacteria were subjected to a line probe assay (GenoType Mycobacterium CM), and to 16S rDNA and ITS gene sequencing, and whole genome sequencing (WGS). However, none of these methods identified this isolate as a valid Mycobacterium species. Nevertheless, 16S sequence allocated this isolate to slow growing mycobacteria. Neither Mycobacterium bovis nor M. caprae nor M. tuberculosis were found in the collected granulomatous lesions. In conclusion, the overall prevalence of TB-suggestive lesions in ruminants in Kassala was very low. Extended studies combining the use of a tuberculin skin test and slaughterhouse-based investigations should be adopted for TB surveillance in ruminants in Sudan.
Worldwide prevalence of TB in cattle and other ruminants varies greatly by region and country. The disease has been effectively controlled in many countries but it is still prevalent in many others (Schiller et al., 2010). In Africa, around 85% of cattle and 82% of human populations live in areas where bovine TB is poorly controlled. In these areas, zoonotic TB caused by M. bovis is common Smith et al., 2006;Kulchavenya, 2014). Human TB is endemic in Eastern Sudan, however, knowledge on the prevalence of TB in ruminants is scarce and its epidemiology is not very well understood. This is attributed to the lack of systemic surveys and monitoring programs (Asil et al., 2013). Inadequate animal health infrastructures, traditional extensive grazing system for rearing animals, and a lesser interest in TB are other significant reasons (Asil et al., 2013;Ehsan and Nganwa, 2014).
Several diagnostic techniques can be used to investigate TB prevalence in livestock and to understand better its epidemiology and dynamics (Ramos et al., 2015). Tuberculin skin test and postmortem inspection remain the most important and economically affordable techniques, especially in resource-limited countries (Schiller et al., 2010;Ramos et al., 2015). Recently, few studies have investigated the prevalence of TB-suggestive lesions in cattle carcasses in the central and western parts of Sudan (Osman, 2007;Manal et al., 2010;Asil et al., 2013;Ehsan and Nganwa, 2014;Aljameel et al., 2014). The aim of the present study was to provide a better understanding of the epidemiology of TB and other mycobacterial infections in carcasses of ruminants slaughtered in Kassala in Eastern Sudan, as well as to identify the Mycobacterium species causing the infections.

Study area
Kassala State is located in the eastern part of Sudan at latitude 14°-17° N, and longitude 34°-37° E (Figure 1). It is delimited by a border with Eritrea (235 kilometers) in the east, and by national borders with the Red Sea State in the north, Khartoum State and the River Nile State in the west, and Al-Gadarif State in the southwest. It covers an area of 42,282 square kilometers. It falls within the Sahelian climate zone of Africa. Soils are dark, heavy, deep cracking Vertisol. Rainfall is concentrated in a single relatively short autumn season from June to September and reaches about 680 millimeters per year. Temperature ranges from a mean minimum of 17°C in January to a mean maximum of 47°C in April and May (Sulieman and Buchroithner, 2006). During the rainfall season, Kassala State is rich in fodder, grazing areas, by-products of sorghum and sesame. Moreover, the State has abundant amounts of water. Gaash, Siteet and Atbrah seasonal rivers, and Halfa Al-Gadeeydah canal are full of water during the rainfall season. Port Sudan, the main port for exportation of livestock and livestock products, is located 575 kilometers to the northeast of Kassala. Mixed crop-livestock, nomadic and semi-nomadic production systems predominate in the State. Animals are bred in Kassala for domestic consumption and for exportation to international markets (Omar et al., 2009). The State has an estimated livestock population of about 4.5 million head (MARF, 2008).

Study design, sample size and sampling
A cross-sectional study was conducted from June to November 2014 in Kassala State. This State was selected as it is one of the animal production sites in Sudan. The main town of the State has two slaughterhouses: East and West Gaash. On average, around 7500 head of animals are slaughtered per month in these slaughterhouses, among which about 2500 cattle, 4000 sheep, 950 goats, and 50 camels. The two slaughterhouses were conveniently included in this survey.
The sample size (n) was calculated based on the standard formula of Thrusfield (2007). The parameters of the formula were as follows: 95% confidence level, ± 5% desired level of precision, an assumed expected prevalence of 50% as the prevalence of TB in ruminants was unknown in Kassala. We used: where n was the required sample size, (1.96) 2 the constant, P exp the expected prevalence rate of 50%, and d the desired precision level of ± 5%.
The required sample size n was determined to be 384 animals from each of the two slaughterhouses. This number was inflated threefold to account for the effect of randomness and representativeness (Thrusfield, 2007). Thus, total n was 1152 animals from each slaughterhouse.
Convenient stratification according to species and systematic random sampling were used to select animals (Thrusfield, 2007). Selected animals were identified using permanent markers, kept in a separate place and released for slaughter one after another. The sampling interval was obtained by dividing the total number of slaughtered animals on that day by the estimated daily sample size (Thrusfield, 2007). Four animals were selected each study day from each of the four species. Each of the two slaughterhouses was visited 72 times (three visits per week). Therefore, the total number of animals slaughtered on a particular day from a particular species was divided by four and every n th animal was selected after random selection of the first animal. Accordingly, n comprised a total of 710 cattle, 729 sheep, 700 goats, and 165 camels. Fewer than four camels were slaughtered per day in each slaughterhouse. To account for this, the required n per day was achieved by selecting more animals from other species.

Postmortem inspection
Postmortem inspection was performed as described by Corner et al. (1990), and. This included visual inspection, palpation and in situ slicing of lymph nodes, organs of the thoracic and abdominal cavities, and the head. Additionally, lymph nodes of the fore and hind limbs and the inguinal region were also inspected.

Laboratory procedures
Samples that were collected from the suspected granulomatous lesions were frozen at -20°C until shipping to the National Reference Laboratory (NRL) for mycobacteria, Research Center Borstel (FZB), Germany. At NRL, samples were minced, then homogenized and decontaminated using sodium hydroxide, and the mucolytic agent N-acetyl-L-cysteine (NALC-NaOH) according DIN's guidelines (2011). Smears were prepared from the decontaminated and resuspended specimens according to Kinyoun's method in conjunction with an automated staining system (ZN Aerospray TB Slide Stainer/ Cytocentrifuge, Wescor, Logan, USA). Moreover, cultures were made by inoculation into a mycobacteria-growth indication tube (MGIT) liquid medium (Bactec MGIT 960 system) (Becton-Dickinson, Heidelberg, Germany) and onto solid Löwenstein-Jensen and Stonebrink media (own production). Cultures were incubated at 37°C and 31°C for six (MGIT) or eight (solid media) weeks, respectively.
For molecular characterization of the isolate of the positive cultures, DNA was extracted with Tris-EDTA (TE) boiling/sonication extraction technique or the cetyltrimethylammonium bromide (CTAB) method, respectively. Commercially available line probe assay (LPA) (GenoType Mycobacterium CM, HAIN Lifescience GmbH, Nehren, Germany) and 16S rDNA and ITS gene sequencing were conducted from TE buffer prepared samples. In addition, whole genome sequencing (WGS) was carried out with Illumina MiSeq sequencer, Nextera XT library preparation kits, and MiSeq reagent kits according to the manufacturer's instructions (Illumina, San Diego, CA, USA). Classification of the reads was done by Kraken and reads were mapped to the M. tuberculosis H37Rv genome and M. intracellulare MOTT-64 with the exact alignment program Saruman.

Prediction of the complete sequence of 16S gene by WGS
To predict the complete sequence of 16S gene, de novo genome assembly was carried out. Reads were trimmed with Trimmomatic version 0.35 for Q30 in a sliding window of 20 bp and nextera adapter contamination discarding all reads shorter than 100 bp (Bolger et al., 2014). Trimmed paired and unpaired reads were assembled with SPAdes version 3.6.2 using the built-in read error correction based on k-mer frequencies and the 'careful' option (Bankevich et al., 2012;Caverly et al., 2016). The 16S gene sequence was predicted by the RNAmmer Prediction Server from the assembled contigs which were filtered for containing ribosomal sequences (Lagesen et al., 2007).

■ RESULTS
In total 40,089 animals were slaughtered in the two surveyed slaughterhouses during the study period. More than three quarters (81.0%) were slaughtered at East Gaash slaughterhouse and 19.0% at West Gaash. The male to female ratio was 1:1 for cattle, goats and camel, whereas it was 1:2 for sheep. Only animals older than one year in the case of cattle, sheep and goats, and older than five years in the case of camels had been slaughtered. The animals were mostly of indigenous breeds, e.g. Gaash cattle, and Garrage, Dubassy and Gaash sheep, Baladi, Garrage and Nubi goats, and Bushari and Anaafy camels. A few were exogenous from Eritrea and Ethiopia, e.g. Karour and Barka cattle, and Arrit and Horro sheep.
The overall prevalence of granulomatous lesions in all inspected carcasses was 0.1% (2/2304). The two carcasses that had suggestive TB lesions were carcasses of sheep. No TB-suggestive lesions were detected in carcasses of cattle (0/710), goats (0/700), and camels (0/165). The detected granulomatous lesions were typical caseous necrosis, with a whitish or yellowish color. They were of various sizes congregating together, or scattered and enclosed in light congested grey fibrous tissue. They were found in the liver and lungs, and in the peritoneal cavity covering the whole abdomen and viscera ( Figure 2).
Cultures revealed growth of mycobacteria from samples collected from one carcass. Samples collected from the other carcass were extensively contaminated with bacteria and yeast, and no mycobacterial growth could be found. Furthermore, MTBC isolates including M. bovis, M. caprae, and M. tuberculosis could not be grown from any of the samples.
LPA revealed a banding pattern (bands 1, 2, 3, and 10) that was not specific for any of the mycobacterium species identified by the assay (Figure 3). Moreover, 16S rDNA and ITS sequences of the isolated mycobacteria did not align with any known validly described Mycobacterium species in NCBI Blast database.
The fully sequenced genome had 4.8 mbp. Metagenomics classification by Kraken showed that 71.6% of the reads were unclassified and the closest relative was Mycobacterium sinense JDM601. Moreover, Kraken showed that only 0.2% and 0.1% of the sequenced genome was identical to the genome of M. tuberculosis H37Rv and M. intracellulare MOTT-64, respectively.
The complete sequence of 16S allocated this isolate to slow growing mycobacteria with the closest similarity to members of the M. terrae group. The alignment of the sequence of 16S in the NCBI Blast database is shown in Figure 4.

■ DISCUSSION
TB and other mycobacterial infections in animals are economically significant. The consequential production losses from these infections include lower milk production, loss of body weight, and condemnation of infected organs and carcasses. Animal reproduction and breeding as well as livestock trade at national and international levels are directly and adversely affected by mycobacterial infections (Radostits et al., 2007;Ehsan and Nganwa, 2014;Ramos et al., 2015). TB and other mycobacterial infections are quite prevalent in dairy growing mycobacterium species have also been detected (Sulieman and Hamid, 2002;Oloya et al., 2007;Osman, 2007;Manal et al., 2010;Asil et al., 2013;Aljameel et al., 2014;Nalapa et al., 2017). Environmental risk factors, malnutrition, presence of other diseases, and heavy infestation with endoparasites are important factors that increase the susceptibility of animals to TB (Radostits et al., 2007;Ehsan and Nganwa, 2014). Environmental risk factors have not been investigated in this study; therefore, any solid conclusion regarding their effect in the low observed prevalence of granulomatous lesions could not be made. On the other hand, emaciated animals that are likely to have TB-suggestive lesions in their organs had been excluded during antemortem examination and were not slaughtered for human consumption, in accordance with public health and animal welfare regulations. Nonetheless, some of these excluded emaciated animals might be illegally slaughtered and prepared for human consumption. This practice is common in Sudan and is locally called 'Keery slaughtering'. TB-suggestive lesions might be more prevalent in carcasses of ruminants that are illegally slaughtered for human consumption.
Sensitivity of postmortem inspection for detection of TB-suggestive lesions in animal carcasses is between 28% and 64% (Biffa et al., 2010;Schiller et al., 2010;Ramos et al., 2015;Worku and Abreham, 2016). However, this technique is cost-effective for passive surveillance and TB monitoring in animals. Besides, it is affordable in resource-limited countries (Schiller et al., 2010). The very low prevalence of TB-suggestive lesions in animal carcasses reported in this study could be attributed to the limited sensitivity of postmortem inspection. Furthermore, this study was conducted over a period of six months. Such period is short, especially in case of seasonal variations in the prevalence of animal TB. The sensitivity of the postmortem inspection technique is improved when animals are tested using the tuberculin skin test before slaughtering. Granulomatous lesions are likely to be detected in internal organs of animals that are positive by the tuberculin skin test. Moreover, extended slaughterhouse-based surveys should explore whether there are seasonal variations in the prevalence of granulomatous lesions suggestive of TB or not.
In this study, neither M. bovis nor M. caprae nor M. tuberculosis have been isolated from the detected granulomatous lesions. It suggests that zoonotic transmission of these mycobacteria from animals to humans via the consumption of contaminated meat is unlikely, although it cannot be completely ruled out, especially when the meat is prepared through unsupervised Keery slaughtering. A very slight potential risk of zoonotic transmission of unknown species of mycobacteria to humans via consumption of contaminated meat exists. It is not likely that local and international trade in livestock from Eastern Sudan is going to be restricted or hindered because of animal TB. Nevertheless, occurrence of this disease in animals that are reared on farms and in animals that are grazing in pastures should be monitored.
There is no control program directed against TB in animals in Kassala. Conversely, none of the carcasses inspected at the two slaughterhouses of Kassala had been infected with any of the MTBC bacteria. Radostits et al. (2007) indicate that the overall incidence and prevalence of TB are low in indigenous animals such as zebu type cattle and this might be due to natural resistance. Furthermore, these authors specify that low overall incidences of TB have also been observed in countries where cattle are at pasture all year round. These might probably explain the findings of this study. In Sudan, most of the slaughtered ruminants are indigenous breeds, and animals are at pasture for the whole year in search of fodder and water.
One of the strengths of this study was the investigation of a large number of different animal species at the two slaughterhouses of Kassala. Therefore, the findings might be generalized to the same animal species slaughtered and prepared for human consumption in other slaughterhouses in Eastern Sudan. Nonetheless, the study had limitations. No dairy farms have been investigated, hence, TB prevalence could and meat animals in countries where control and management measures are not in place (Cousins, 2001;Humblet et al., 2009).
This study revealed a very low prevalence of granulomatous lesions (0.1%) in carcasses of cattle, sheep, goats, and camels slaughtered at the two slaughterhouses of Kassala in Eastern Sudan. Overall in Sudan, the prevalence of granulomatous lesions suggestive of TB in slaughtered ruminant carcasses has been reported to range from 0.01% to 8.0% in different parts of the country. Lesions have been found in organs or lymph nodes, or generalized in the whole carcass (Manal et al., 2010;Asil et al., 2013;Aljameel et al., 2014;Ehsan and Nganwa, 2014). With the exception of one study that reported a small number of dairy cattle (11/569) that were positive by the tuberculin skin test in Eastern Sudan (Ayman et al., 2014), prevalence of TB in dairy and fattening farms has not been investigated in Sudan. Moreover, in neighboring and other African countries, prevalence of granulomatous lesions in slaughtered food animal carcasses has been up to 16% (Mohamed et al., 2009;Biffa et al., 2010;Manal et al., 2010;Kassa et al., 2012;Youssef and Ahmed, 2014;Worku and Abreham, 2016;Nalapa et al., 2017).  not be compared between animals raised for meat and animals raised for milk. Additionally, the study was conducted in a short period. Perhaps TB prevalence was underestimated because of seasonal variations. Furthermore, only postmortem inspection was used as a mean to investigate TB, but granulomatous lesions are small during early stages of the disease and might not be visible with naked eyes.

■ CONCLUSION
TB was not detected in carcasses of slaughtered ruminants in Kassala. This indicates that TB incidence in animals is very low in Kassala, and zoonotic transmission of MTBC bacteria is unlikely. One of the carcasses that was found with typical granulomatous lesions was infected with an unknown Mycobacterium species. This isolate could be pathogenic to humans. Slaughterhouse-based surveys combined with pre-slaughtering testing using the tuberculin skin test would help to explore TB epidemiology in animals in Sudan. Palabras clave: rumiante, Mycobacterium, tuberculosis, matadero, Sudan management, a long feeding gap during the dry season, and the high prevalence of diseases (mainly viral, bacterial and parasitic). In several countries and production systems, lungworms are responsible for huge economic losses in meat and milk production and cause mortality in sheep, goats and cattle (Soulsby, 1982), especially because the lesions caused by the worms may be complicated by bacterial bronchopneumonia. The life cycle of these parasites is influenced by geographic and climatic factors (Panayotova-Pencheva, 2011) leading to specific regional dynamics (Phillips et al., 2010). Infestation of small ruminants occurs through ingestion of grass contaminated by infesting larvae or hosting the intermediate hosts of the larvae of parasites with indirect life cycle (Snyder et al., 2015). The infestation impact depends on the host susceptibility, the host-parasite interaction and the number of larvae ingested.

Résumé
In Tunisia, several studies have been carried out on gastrointestinal nematodes (Lahmar et al., 1990;Akkari et al., 2011;Rouatbi et al., 2016), but as far as it could be ascertained, there has been no longitudinal survey regarding respiratory lungworms, except a few veterinary theses studying In addition there was no significant difference in infestation between fat-tailed Barbarine (10.8 ± 3.6%) and Queue fine de l'Ouest (Western Thin Tail) (8.5 ± 2.6%; p = 0.31) sheep breeds. The lowest prevalence was recorded in spring (6.7 ± 3.6%) and the highest in winter (11.7 ± 4.8%), but the difference between seasons was not significant (p = 0.32). This study concerning the activity dynamics of lungworms should be completed by a monitoring project to determine the economic importance of these infestations and provide a solid basis for the establishment of specific control programs against these parasites in Tunisia. the histological effects of these parasites (Aloui, 2005;Temri, 2007;Ben Said, 2008). Lungworm importance still needs to be assessed. Therefore, the aim of this study was to investigate the seasonal variation of lungworm infestation in sheep in Sidi Bouzid region.

Study area
Sidi Bouzid is located in the center of Tunisia (35° 02' N; 9° 29' E) ( Figure 1). It covers 7405 square kilometers representing 5% of the total area of the country. The climate of Sidi Bouzid is Mediterranean continental: it is arid in the southern part of the district and semiarid in its northern part. Winter is cold and dry (mean temperature of 11.8°C) whereas summer is hot (28.2°C). Mean annual rainfall is low (287 millimeters) with high interannual and interseasonal variations.

Animals and coprologic samples
Monthly visits were implemented between May 2014 and April 2015 at Sidi Bouzid regional slaughterhouse. Feces were collected from 720 female sheep (60 per month) belonging to two breeds, fat-tailed Barbarine (287) and Queue fine de l'Ouest (Western Thin Tail) (433). The age of sheep was estimated by dental examination (Pavaux, 1975). Fecal samples were brought to the laboratory and stored at +4°C until analysis. For the coprological test we placed 5 g of feces in Baermann's apparatus and left them to incubate at 25°C for 24 h. Then, 10-15 ml of the sediment were collected in a tube and centrifuged for 10 min at 1500 rpm. The pellet was examined for the presence of lungworm after adding one drop of Lugol to uncoil the larvae, which were determined under a microscope, based on morphological characteristics.

Parasitological indicators and statistical analysis
The infestation prevalence was calculated as follows: Prevalence = (100 × num. of infested sheep) / (num. of examined sheep) To study variations in the infestation prevalence according to the age group, breed and month, a chi-square test was performed with Epi Info 6 software (Dean et al., 2011). The threshold value was p = 0.05.

Overall infestation indicators
Among the 720 samples of female sheep feces, 68 were infested, corresponding to a prevalence of 9.4 ± 2.1%. There was no significant difference between the infestation prevalence in the two breeds (p = 0.31) nor between age groups (p = 0.64) ( Table I). The infestation prevalence was higher in winter (December to February: 11.7 ± 4.7%) and lower in spring (March to May: 6.7 ± 3.6%). The highest infestation prevalence was reported in December (25.0 ± 11.0 %) (p = 0.004; Figure 2).

Infestation prevalence according to lungworm species
The most frequent parasite was Protostrongylus rufescens (4.4 ± 1.5%) followed by Dictyocaulus filaria (2.6 ± 1.2%), Cystocaulus ocreatus (1.3 ± 0.8%) and Neostrongylus linearis (1.3 ± 0.8%), and finally Muellerius capillaris (0.3 ± 0.4%). The co-infestation prevalence was very low (0.4 ± 0.5%) and observed with i) P. rufescens and C. ocreatus, and ii) P. rufescens and D. filaria. P. rufescens was the most prevalent parasite in both Queue fine de l'Ouest and Barbarine sheep (4.6% and 4.2%, respectively) (p = 0.78) ( Table I). P. rufescens was also the dominant parasite during all seasons except spring (Table I; Figure 3). This species was present during the whole year except in March and April, with a peak in December (16.7 ± 9.4%) (Figure 4).     The aim of the study was to assess the prevalence of lungworm infestation in slaughtered female sheep in Sidi Bouzid District (Central Tunisia). In this region, male and female fattened lambs are very popular and many animals are slaughtered when they are young. Because of drought and illicit import of sheep from neighboring countries in the southern area, sheep meat was in overproduction at the time of the study, and the value of female sheep decreased, prompting farmers to sell them because of high feeding costs. The choice of examining female animals was probably due to the fact that young animals did not graze and were not regularly infested by worms, preventing the acquisition of immunity.
This survey provided information about seasonal infestation dynamics by several lungworm species; sheep were more infested in winter. There is however a gap in the knowledge of the economic impact of infestation by these parasites. A cost-benefit analysis should be carried out to decide whether a control program should be implemented in Tunisia. Both cost-benefit analysis and preventive anthelminthic treatments should be based on the epidemiological data reported in this study.