Livestock Research for Rural Development 30 (7) 2018 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Reproductive performances and abortion etiologies of native Bedouin goats in the arid zones of Algeria

A Kouri, S Charallah, F Kouri, Z Amirat and F Khammar

University of Science and Technology Houari Boumediene, Faculty of Biology, Research Laboratory of Arid Lands, BP.32 El-Alia, Algiers, Algeria


The objectives of this study were to determine the reproductive performances of Bedouin goats raised in arid environments, and to identify abortion etiologies under traditional management. The study consisted of three parts: first, 27 flocks were surveyed to determine reproductive performances and abortion rates; second, a retrospective study was performed on 105 goats to analyze the seroprevalence of some agents implicated in infectious abortions (Brucella, CAEV and bluetongue); third, a follow-up study was carried out on 21 pregnant goats to evaluate the effects of non-infectious factors. (1) Pregnancy, fertility, fecundity, and prolificacy rates were 95.7%, 55.9%, 64.2%, and 114% respectively, and abortion and pre-weaning kid mortality rates were high (13.8 ± 3.3% and 12.2 ± 5.1%, respectively). (2) Brucellosis (23.8%) and bluetongue (66.7%) were widespread, whereas the goats were free of caprine arthritis encephalitis virus and salmonellosis. (3) Mean progesterone levels were higher in individuals exhibiting normal pregnancies than in those for which pregnancy was interrupted (8.17 vs 5.09 ng/mL, p≤0.01), whereas lower levels of total protein (5.01 vs 6.54 g/dL, p≤0.01) and albumin (2.26 vs 3.01 g/dL, p ≤0.05) were observed. Both infectious and non-infectious abortion etiologies were implicated. This study will help to establish a control strategy limiting the repercussions of failed pregnancies.

Key words: bluetongue, Brucella, CAEV, caprine, gestation, management system, reproductive wastage, Sahara desert, Salmonella, seroprevalence


Of all domesticated ruminants, goats are the best adapted to harsh environments and as such represent the primary form of animal production in many marginal areas. In the Sahara Desert, which comprises 84% of Algeria ( FAO 2012), Bedouin goats are especially of great importance for local populations, serving as sources of meat, milk, and hides. Most production systems in Algeria are extensive, but under natural conditions, particularly during dry years, water and feed scarcities, along with extreme climatic conditions, are limiting factors for improving goat production. Bedouin goats have, however, developed numerous physiological adaptations that allow them to proliferate under such conditions, including reduced water turnover and glomerular filtration rates (Silanikove 1984), low metabolic requirements (Silanikove et al 1993), and production of high concentrations of caprine pregnancy-associated glycoproteins (cPAGs) (Charallah et al 2009). Indeed, high fertility, fecundity, and prolificacy rates have been reported for this breed (Charallah et al 2002). However, traditional management practices, characterized by inappropriate sanitary monitoring, increase the susceptibility of this breed to higher rates of fetal abortion and kid mortality that reduce their profitability. Previous studies of different goat breeds raised under different climatic conditions showed that the contribution of infectious agents to reproductive losses varied from minor (2.4%) (Waldeland and Løken 1991) to major (37%) (Moeller 2001). The incidence of infectious abortions is largely dependent on herd size, contact between herds, and exchange of animals among herds ( Al-Majali 2005; Al-Qudah et al 2006 Ghanem et al 2009). Most research has focused on the prevention of infectious abortions to minimize economic losses, whereas only a handful of studies have focused on non-infectious abortions despite their potential significance; for example, Waldeland and Løken (1991),Engeland et al (1998), and Mellado et al (2004) have all reported that infections were not a prime cause of abortion, with age, season of mating, management practices, feeding system type, and hormonal deficiencies constituting the most important risk factors (Hussain et al 1996; Engeland et al 1998; Mellado et al 2006). Studies of the progesterone (P4) and cPAGs profiles of Bedouin goats with interrupted pregnancies under controlled conditions (Charallah et al 2009) determined that non-infectious causes were responsible for most abortions, but failed to explore the implications of these findings and the risk factors involved. Thus, our objective here was to examine the reproductive performances of Bedouin goats living in arid environments, and to investigate the abortion etiologies and some associated factors for this breed when raised under traditional management practices.

Materials and methods

The study was carried out at Beni-Abbes (30°07’ N, 2°10’ W), in the Algerian Sahara. This area is characterized by a hot and dry climate (temperature > 44°C in summer, precipitation < 30 mm/year) and little vegetation.

Survey of reproductive performances

In order to determine reproductive performance and abortion frequency, 27 randomly selected flocks were surveyed over two kidding seasons. We calculated pregnancy, fertility, fecundity, prolificacy, and abortion rates, as well as the mortality rates of pre-weaning kids, pregnant females, and parturient females.

Retrospective study on infectious abortions

To elucidate whether infectious agents are key factors in fetal loss, blood samples were collected for serological analysis from the jugular veins of 105 goats (86 aborted and 19 kidded) belonging to 23 randomly selected flocks managed under different breeding systems. Forty-three goats were also milked into sterile plastic flasks for bacteriological analysis. Samples were transported from the field to the laboratory in an icebox. Blood samples were centrifuged to collect serum, and both the serum and milk samples were kept frozen at -20°C until used in the analyses.

Follow-up of pregnant goats
Flocks and management

The experiment was performed on 30 Bedouin goats, identified by ear tags and in apparent good health, chosen at random from two flocks managed under extensive systems. Twelve goats were selected from the first flock (1-9 yr old, average weight of 23.8 ± 1.9 kg) and 18 from the second (2-7 yr old, average weight of 26.9 ± 1.1 kg). Of the 30 goats, two females were nulliparous, 12 were primiparous, and 16 were multiparous; four of the goats had previously aborted.

The first flock was managed with a semi-sedentary system, whereas the second flock was managed under a nomadic system. The two flocks were mixed; both goats and sheep were kept together for the whole of the experimental period. The second flock also included camels, which were kept with the other animals during summer. Goats in the first flock were kept in a fenced area close to the village for the summer and fall, and fed twice daily with forage cereals, barley, bran, dry bread, and Drinn ( Aristida pungens). In winter, the flock was moved to the Hamada (rocky plateaus) and reared in pasture without concentrate supplementation, except when temperatures were unusually low or for parturient females. The second flock spent the entire summer and fall in the Erg (sand dune), then moved to Hamada pastures for the winter. These goats were allowed to graze on natural rangelands for ~ 10 h per day and their diet was supplemented with dates, barley, and dry bread upon return from the pasture. Salt stones were also made available. At night, the goats were kept in a fenced area to protect them from predators. Water for both flocks was provided by a well (Hassi) that was available ad libitum. No goats in either flock were vaccinated.

During the breeding season, a fertile buck was allowed to mate with the females of the first flock and kept permanently with the flock until late fall. In the second flock, females were exposed to three bucks, but these were separated from the flock in late summer. One male was reintroduced into the flock for two weeks in fall after the breeder became aware that some females were not pregnant. The flocks were visited weekly in the field for sampling, at which times all health parameters and reproductive events were recorded.

Sampling and processing

Blood samples were collected weekly during the gestation period and for one month after kidding or reproductive wastage, for biochemical and hormonal analyses. For the serological examination, blood samples were collected in the beginning and after the outcome of pregnancy. The goats were also milked for bacteriological examination.

Sample analyses
Progesterone assay

Serum P4 levels were determined using an enzyme immunoassay kit (Vitros Immunodiagnostic Products, Ortho-Clinical Diagnostics, Inc., UK) and a Vitros ECiQ auto-analyzer. The assay is based on a competitive reaction in which the horseradish peroxidase (HRP)-labeled P4 competes with P4 in the serum for a limited number of binding sites on a biotinylated rabbit anti-P4 antibody. Assay sensitivity was 0.08 ng/mL. The intra-assay CVs were 0.2%, 1.1%, and 2.5% at mean values of 21.40, 10.03, and 0.38 ng/mL, respectively. The inter-assay CVs for the same samples were 1.2% (21.09 ng/mL), 4.9% (9.93 ng/mL), and 3.3% (0.36 ng/mL).

Biochemical analysis

Total serum proteins (TP) and albumin (Alb) were measured spectrophotometrically using biuret and bromocresol green colorimetric methods, respectively. Gamma-glutamyl transferase (γ-GT) catalytic activity was determined via kinetic reaction using a Cobas Mira auto-analyzer at 37°C. Analyses were carried out with Spinreact kits (Spain). Sera from goats that aborted were subjected to further examination for quantification of the six major protein fractions (alb, α1, α2, β1, β2, and γ-globulins) by capillary electrophoresis with a Minicap Sebia System and a Minicap Protein(e)6 kit (Sebia, France).

Serological analysis

All sera collected were tested for presence of Brucella, Caprine arthritis encephalitis virus (CAEV), and Bluetongue antibodies. They are known here for their high incidence in goats.

Brucellosis (Br) diagnosis was assessed using the Rose Bengal test, a qualitative approach that produces a rapid agglutination reaction when a stained bacterial suspension (B. abortus antigen, strain 99; Lillidale Diagnostics, Ltd., UK) is mixed with samples containing specific Brucella antibodies; for confirmation, positive serums were further tested using the complement fixation test. Brucella abortus antigens (strain 99, Pourquier, Montpellier, France) and Guinea-pig complement (bioMérieux, France) were used in this analysis. The indicator system consisted of fresh sheep red blood cells (SRBC) and rabbit anti-SRBC serum (Hymolysin). A calcium-magnesium-veronal buffer (Pourquier, Montpellier, France) was used for dilutions.

Sera were evaluated for anti-CAEV and anti-Bluetongue surface envelope antibodies (anti-P28 and anti-VP7, respectively) using indirect ELISA and competitive ELISA kits, respectively (Pourquier, Montpellier, France).

Milk bacteriological analysis

Milk samples were tested for the presence of Brucella antibodies with the milk ring test, in which B. abortus antigens were stained with a hematoxylin solution (Pourquier, Montpellier, France), and positive reactions are indicated by the formation of a dark blue ring in the tube above the milk column. Milk samples were also tested for Salmonella using CONDA media (Madrid, Spain). A pre-enrichment was performed by incubation in buffered peptone water (37°C for 24 h), after which the aliquot was transferred into a Rappaport-Vassiliadis broth and incubated at 37°C. Thereafter, growth from each sample was streaked on Hektoen agar medium and incubated at 37°C for 24 h.

Statistical analysis

Data were analyzed using the GraphPad Prism program (version 5.00 for Windows, GraphPad Software, San Diego California USA). Shapiro-Wilk analysis was used to test variables for normal distributions and Levene’s test was used to assess homogeneity of variances. Statistical significance was set at p≤0.05, and results are expressed as mean ± SEM. The effect of management system on abortion rate was assessed using the Kruskal-Wallis test and Dunn’s multiple comparison post-test; evaluation of the effect of year on abortion and kid mortality rates were conducted with a Mann-Whitney test; data from the serological and milk bacteriological examination were analyzed with a Chi-square test. A Mann-Whitney test was also used to compare normal and failed pregnancies, and to assess the effect of management system on blood parameters.


Management system, flock structure, and reproductive performance

Of the 27 flocks we examined, 18.5% were managed using a sedentary system, and 81.5% were equally maintained under semi-sedentary and nomadic systems. The majority of the flocks consisted of a mix of goats, sheep, and camels (44.4%) or goats and sheep (37%). The remaining flocks were composed almost exclusively of goats (14.8%), with one breeder keeping camels and goats (3.7%). Considering only goats, the average flock size was 67 ± 17, and ranged from 3 to 350 goats. The flocks included mostly breeding females (71.9 ± 3.1%); the ratio of breeding male to breeding female averaged 1:15, but varied greatly among flocks (1:2 to 1:61).

Overall pregnancy, fertility, fecundity, and prolificacy rates obtained were 95.7 ± 1.9%, 55.9 ± 5.5%, 64.2 ± 6.6% and 114 ± 2.6% respectively (Table 1). Management system and year of kidding did not have effects on these rates. With the exception of one set of triplets, does typically gave birth to either single kids (86.5 ± 2.5%) or twins (12.9 ± 2.6%). The number of does that aborted varied between 0-70, with a mean incidence of 13.8 ± 3.3%. The incidence of fetal loss recorded in flocks managed with sedentary, semi-sedentary, and nomadic systems was 22.5 ± 12.6%, 3.5 ± 1.9%, and 20.2 ± 4.2%, respectively, with a difference observed only between the latter two systems (p<0.01). Abortion rates were higher in the first year than in the second (18.8 ± 5.2% vs 9.19 ± 3.8%) although the difference was not statistically significant. Likewise, pre-weaning kid mortality was higher in the first year than in the second (25.3 ± 9.7% vs 0.98 ± 0.7%; p< 0.01) with a mean of 12.2 ± 5.1%. Mortality rates of pregnant females (five flocks) and parturient females (nine flocks) averaged 1.16 ± 0.9% and 5.73 ± 2.8%, respectively.

Table 1. Reproductive performances (% ± SEM) of the Bedouin goats included in this study (n = 27 flocks)

First year of survey
(n = 13)

Second year of survey
(n = 14)

(n = 27)


94.1 ± 2.6

97.1 ± 2.9

95.7 ± 1.9


50.8 ± 8.8

60.5 ± 7.0

55.9 ± 5.5


59.8 ± 10.7

68.2 ± 8.1

64.2 ± 6.6


117 ± 4.3

112 ± 2.9

114 ± 2.6


18.8 ± 5.2

9.19 ± 3.8

13.8 ± 3.3


Pre-weaning kids

25.3 ± 9.7

0.98 ± 0.7

12.2 ± 5.1

Pregnant females

2.28 ± 1.8

0.12 ± 0.1

1.16 ± 0.9

Parturient females

9.91 ± 5.6

2.15 ± 1.2

5.73 ± 2.8

Seroprevalence of infectious agents from the retrospective study

Overall pathogen prevalence is shown in Table 2. Twenty-eight goats (26.7%) were free of all pathogens for which we tested, whereas the remaining 77 goats were infected by either Br or Bl or by both. All sera tested for CAEV were negative, and all milk samples were found to be free of salmonellosis.

Table 2. Serological analyses of serum (n = 105) and milk (n = 43) samples collected during the retrospective study

Examined goats


Positive goats (%)

No of positive
flocks (%)










2 (10.5)

23 (26.7)

9 (39.1)












9 (47.4)

61 (70.9)

17 (73.9)







6 (15.8)

3 (30)








CAEV = caprine arthritis encephalitis virus

Twenty-five (23.8%) of the 105 serum samples and six (14.0%) of the 43 milk samples tested for anti-Brucella agglutinins were found to be positive. At the flock level, nine (39.1%) flocks contained at least one seroreactive individual. Serological analysis also revealed that 70 goats (66.7%) were infected by Bl, and 17 (73.9%) of the tested flocks were seropositive. Mean within-flock prevalence was 19.5 ± 6.6% and 8.6 ± 5.3% for Br in serum and milk, respectively, and averaged 64.4 ± 9.1% for Bl. Eighteen goats (17.1%) were found to be seropositive for both Br and Bl infection.

Does that aborted and those that kidded differed in Bl prevalence (p<0.05) but not in Br prevalence (p>0.05). However, pregnancy outcome was still affected by overall seroreactivity (p <0.01), as goats infected by at least by one pathogen had a higher incidence of abortion than did those free of any of the pathogens included in the analysis (64.8% vs 17.1%). As a risk factor, the type of management system was found to have an effect on serological status; in fact, the Br infection rate was higher in nomadic goats than in goats reared under sedentary systems (21% vs 2.9%; p<0.05), and Bl infection was more common in nomadic goats than in goats raised under sedentary and semi-sedentary systems (53.3% vs 1.9% and 11.4%, respectively; p<0.05). Age, gestation rank, parity and previous incidence of abortion were not associated with either Br or Bl seropositivity (p>0.05).

Hormonal and biochemical analyses
Outcome of pregnancy

Nine of the 30 goats (six from the first flock, three from the second) were omitted from the study: one died from non-identified causes, two were sold, four failed to conceive, and two kidded at the beginning of the follow-up. Of the remaining 21 goats, 17 (81%) kidded and four (19%) experienced fetal loss. One goat (G1) in the first flock exhibited bloody vaginal discharges in the fall; in the second flock, one goat (G2) aborted in the fall and two (G3 and G4) in winter. Table 3 shows data pertaining to the goats that experienced fetal loss.

Table 3. Summary of the serological status, and serum hormonal and biochemical parameters
during pregnancy (mean ± SEM) in goats that aborted from the follow-up

Goat 1

Goat 2

Goat 3

Goat 4

Serological status
















Serum parameters

Progesterone (ng/mL)

3.58 ± 1.22

7.16 ± 0.45

3.70 ± 0.25

5.92 ± 0.49

Total proteins (g/dL)

5.48 ± 0.40

7.00 ± 0.27

7.53 ± 0.18

6.16 ± 0.14

Albumin (g/dL)

2.19 ± 0.27

3.67 ± 0.11

3.05 ± 0.08

3.14 ± 0.11

α1 globulins (g/dL)

0.046 ± 0.003

0.081 ± 0.003

0.074 ± 0.010

0.055 ± 0.010

α2 globulins (g/dL)

0.59 ± 0.03

0.57 ± 0.04

0.52 ± 0.02

0.55 ± 0.02

β1 globulins (g/dL)

0.75 ± 0.07

0.88 ± 0.07

0.84 ± 0.03

0.90 ± 0.04

β2 globulins (g/dL)

0.18 ± 0.02

0.27 ± 0.02

0.30 ± 0.02

0.19 ± 0.01

γ globulins (g/dL)

1.38 ± 0.06

1.53 ± 0.09

2.75 ± 0.08

1.32 ± 0.04

γ-GT (U/dL)

13.0 ± 0.71

19.2 ± 1.16

15.3 ± 0.59

18.0 ± 0.49

CAEV = caprine arthritis encephalitis virus; γ-GT = gamma-glutamyl transferase

Progesterone and serum biochemical profiles

Variations in the blood parameters of the 17 goats that experienced normal pregnancies and the individual rates of the goats that experienced fetal loss are shown in Fig.1-4.

In females that experienced normal pregnancies, the mean P4 value was 0.53 ± 0.17 ng/mL in the first week, but then climbed to 3.31 ± 0.37 ng/mL during the second week and continued to increase progressively until the fourth month. The highest mean P4 level, recorded at week 15, was 12.23 ± 1.23 ng/mL (maximum individual rate was 29.8 ng/mL). After peaking, P4 levels declined gradually between weeks 15 and 20, then dropped sharply over the two weeks immediately prior to kidding. The post-partum mean P4 rate was 0.38 ± 0.03 ng/mL. Mean P4 levels were higher in does exhibiting normal gestation than in does in which pregnancies were interrupted (8.17 ± 0.4 vs 5.09 ± 0.88 ng/mL, respectively; p<0.05).

Total protein, Alb, and γ-GT levels fluctuated and displayed no distinct patterns in normal pregnancy (means of 5.01 ± 0.09 g/dL, 2.26 ± 0.06 g/dL and 19.8 ± 0.76 U/dL, respectively). Mean post-partum levels of TP, Alb, and γ-GT were 6.66 ± 0.19 g/dL, 2.26 ± 0.08 g/dL, and 24.1 ± 1.02 U/dL, respectively, and mean TP and Alb values for goats that experienced fetal loss were 6.54 ± 0.45 and 3.01 ± 0.31 g/dL, respectively; these levels were higher (p<0.05) than those found in does experiencing normal pregnancies. Other protein fractions (α1, α2, β1, β2, and γ-globulins) were relatively stable (Table 3 and Fig. 4). The mean level of γ-GT in goats that aborted (16.4 ± 1.39 U/dL) did not differ from goats displaying normal pregnancies (p>0.05). Albumin levels during gestation were higher in goats in the second flock than in goats in the first flock (2.55 ± 0.11 vs 2.03 ± 0.08 g/dL; p<0.01), whereas P4, TP, and γ-GT levels were unaffected by management system (p>0.05).

Figure 1. Mean progesterone levels (P4) during normal pregnancy in Bedouin goats. P4 levels were normal for the gestation
period, peaking at week 15 (12.2 ± 1.23 ng/mL).Post-partum P4 levels declined to a low of 0.38 ± 0.03 ng/mL.

Figure 2. Progesterone levels (P4) during interrupted pregnancy in 4 individual Bedouin goats. // = samples were not collected. Goat 1 and 3
presented disturbed profiles of P4. Goat 1: P4 was high in the second week of pregnancy and then decreased drastically and
remained low until embryo-loss. Goat 3: from week 7, P4 declined and remained around 3 ng/mL prior to abortion.

Figure 3. Mean levels of serum total protein, albumin, and γ-glutamyl
transferase (γ-GT) during normal pregnancy in Bedouin goats.

Figure 4. Levels of serum total protein, albumin, γ-globulin and γ-glutamyl transferase (γ-GT) during interrupted pregnancy in
4 Bedouin goats. // = samples were not collected. Goat 3: γ-globulin levels were high (2.75 ± 0.08 g/dL).


Management system and reproductive performance

The ratio between breeding male and breeding female was favorable, given that a ratio of 1:25 is recommended for goats managed under traditional production systems (Wilson and Durkin 1988). In Bedouin flocks, females are typically retained for breeding purpose, whereas males are often sold, especially to finance social events. Thus, some breeders borrow or purchase fertile bucks during the breeding season.

A previous study (Charallah et al 2002) reported higher rates of fertility (93.5%) and fecundity (86.9%) but a similar prolificacy rate (110.7%) among Bedouin goats than was observed here. Pregnancy rates were higher among the goats included in our study than for Damascus goats reared under semi-arid conditions (64.2%), whereas higher fertility rates were recorded for this breed (70%) (Shalaby et al 2000). The prolificacy rate was similar to that of local Tunisian goats (110%) (Ben Saïd 1992) but lower than that of Draa goats (Boujenane et al 2010) and Moroccan indigenous goats (Chentouf et al 2011).

The abortion rate we observed was high, as a rate of 2-5% is considered normal among small ruminants (Dubreuil and Arsenault 2003). However, the abortion rate recorded here was similar to that previously reported for the same breed (11.7%) (Charallah et al 2002), in Norwegian dairy goats (15.6% and 11.1%) (Waldeland and Løken 1991; Engeland et al 1998), and among native Indian breeds (13.3%) (Singh et al 1994). It should be pointed out that the abortion rate is an estimate and is based on the word of the breeders; when infectious abortions are suspected, breeders may be less forthcoming in order to avoid potential economic losses. According to the breeders, most abortions occurred during the final two weeks of gestation, but many also occurred at earlier stages; such losses are often unperceived by inexperienced shepherds. The abortion rate was affected by management system, as was previously found for Bedouin goats (Charallah et al 2002).

The pre-weaning mortality rate was higher than that previously reported for Bedouin kids (8.4 ± 2.7%) (Charallah et al 2002) ; such mortalities were primarily due to inadequate management practices, undernutrition, and exposure to very high temperatures following suckling. Deaths of pregnant and post-partum females were sporadic, with rates in accordance with those previously reported for Bedouin goats (Charallah et al 2002). These mortalities were generally the result of malnutrition, diseases, and, in few cases, predators and accidents.

Abortion, pre-weaning kid mortality, pregnant female mortality, and parturient female mortality declined by 51.1%, 96.1%, 94.7% and 78.3%, respectively, from the first year to the second; however, such temporal discrepancy was not unexpected given the greater amount of precipitation that fell during the second years, leading to much improved grazing conditions. Moreover, according to the breeders, very low temperatures during winter nights were also a major cause of abortions and mortalities.

Infectious abortions

Several other possible infectious agents were not looked for such as chlamydiosis, leptospirosis, listeriosis, mycosis, Q fever and Rift valley fever.

However, serological analysis revealed that goats infected by at least one pathogen were more likely to abort than were uninfected goats, and infectious agents were diagnosed in 79.1% of goats that aborted. Based on an indirect diagnosis, it is unlikely that this reflects the full impact of infectious agents on reproductive losses, as recently infected animals may be confused with those infected during previous pregnancies and with goats that have been vaccinated (Menzies 2012).

The results of the follow-up survey indicated that γ-GT did not differ between goats experiencing normal and interrupted pregnancies, suggesting therefore that infections that interfere with liver functioning are not involved in the abortion process. However, the high level of γ-globulins in the blood of G3 indicated an infectious condition in this individual, which was supported by analysis of its serological status.

Bedouin goats were mostly free of CAEV and salmonellosis, but Brucella and bluetongue infection was widespread. A similar result was reported for Br in a prevalence survey conducted in North Africa under the auspices of a FAO/RADISCON program (Benkirane 2006), in which Br was found to be highly prevalent in Algeria (herd prevalence of 42%, 1986-89) and Tunisia (61%, 1991), whereas Br prevalence was much lower in Morocco (2.4%, 1996).

Brucellosis infection may persist for years, but infected goats are capable of delivering normal births following subsequent pregnancies, which may explain the occurrence of seropositivity among parturient goats (2/19 goats). In addition, natural resistance can be developed in countries where Bl is endemic (Ganter 2014). Thus, seropositive goats may produce normal kids, as we observed for nine parturient goats.

As a risk factor, the type of management system had an effect on the seroprevalence of Br and Bl, which were more common in nomadic flocks. Animals managed under this system often come into contact with goats from other herds while grazing and when watering at traditional shared wells (Hassi), which increases the risk of cross-herd contamination. Goats may also be exposed to contact with wild animals that serve as natural reservoirs of pathogens, especially for herds that are unfenced at night. Rearing in close association with sheep (81.5% of the flocks), and the common practice of borrowing and purchasing bucks for fertilization may also enhance disease transmission. Another factor contributing to the comparatively high rates of seropositivity in nomadic flocks is their access to remote areas, which reduces hygiene and limits their availability to veterinarians for vaccination. We determined that age, gestation rank, and parity were key risk factors for infection. In a previous study, age has been linked to prolonged contact with infected animals (Ghanem et al 2009).

Non-infectious abortions

Despite the abortion rate being highest in sedentary flocks, the serological data revealed that the lowest seropositive rate occurred in flocks managed under this system, suggesting that abortion is associated to a greater degree with non-infectious causes.

The survey data revealed that there was an effect of management system on abortion rate, which can be related to nutrition. For nomadic flocks, nutrition is dependent on natural sources of food, the abundance and quality of which vary from season to season and from year to year; thus, a lack of food or poor feed quality can be serious problems during dry periods if the diet is not supplemented. Among the sedentary flocks, we observed that food resources consisted mainly of poor quality forage, and supplementation was typically reserved exclusively for parturient does. This was supported by the results of the biochemical analysis (follow-up data), which showed that goats in the first flock had lower levels of serum Alb than goats in the second flock. Nutritional deficit elevates levels of serum urea, which increases the likelihood of abortion (Mellado et al 2004) by increasing uterine luminal PGF2α ( Gilbert et al 1996).

Based on data obtained from the follow-up, it would appear that a variety of explanatory factors are implicated in fetal loss:

Age and parity of does was associated with the incidence of abortion in three goats: G1 was one-year old, whereas G2 and G3 were of advanced ages. Younger primiparous goats have greater nutritional demands in order to achieve mature body size and to maintain gestation, leading to underfeeding situations (Ben Saïd 1992). On the other hand, older multiparous goats often suffered from decreasing performance and offspring quality (Mellado et al 2006), which increase the risk of abortion.

Stress may have caused G4 to abort on the day the flock was moved, as an increase in cortisol level may inhibit luteal functioning (Engeland et al 1998).

Goat 3 and goat 4 did not conceive during summer but rather were fertilized in the fall. Therefore, abortion may have resulted from improper timing, given that high temperatures are more favorable to reproductive success of goats living in hot and arid environments (Mellado and Meza-Herrera 2002). Alternatively, low fertility expressed by conception difficulties may be another possibility; Engeland et al (1997) attributed such problems to malformations and inadequate luteal functioning.

Progesterone levels declined sharply in G1 and G3 3 and 6 weeks prior to abortion, respectively. Moreover, their average P4 concentration was lower than that in goats experiencing normal pregnancy. Thus, malfunctioning corpus luteum can be considered to be one of the causes of abortion.



This work was supported by the Ministry of Higher Education and Scientific Research of Algeria [project number F 002 2012 0046]. The authors are deeply grateful to M. Yaïch and H. Salmi for their assistance with the experiments. We are thankful to the staff of the Regional Veterinary Laboratory of Tlemcen, and the Regional milk quality laboratory of Igli, for providing the facilities necessary for the serological analyses, as well as for technical assistance. We also thank F. Z. Boucenna for providing the equipment necessary for γ-GT analysis, and M. Hemmel for his help in the serological survey.


Al-Majali A M 2005 Seroepidemiology of caprine brucellosis in Jordan.Small Ruminant Research, 58, 13–18.

Al-Qudah K, Al-Majali A M and Ismail Z B 2006 Epidemiological studies on caprine arthritis-encephalitis virus infection in Jordan. Small Ruminant Research, 66, 181–186.

Ben Saïd M S 1992 L'élevage traditionnel dans les zones montagneuses du Nord de la Tunisie, Contraintes et possibilités d'amélioration. GTZ-OEP-Ministère de l'Agriculture, Tunis, 246p.

Benkirane A 2006 Ovine and caprine brucellosis: World distribution and control/eradication strategies in West Asia/North Africa region. Small Ruminant Research, 62, 19–25.

Boujenane I, Lichir N et El Hazzab A 2010 Performances de reproduction et de production laitière des chèvres Draa au Maroc. Revue d’Elevage et de Médecine vétérinaire des Pays tropicaux, 63, 83–88.

Charallah S, Amirat Z et Khammar F 2002 Systèmes d'élevage et caractéristiques de la reproduction chez la chèvre bédouine (Capra hircus). Proceedings of the 9èmes Rencontres Recherches Ruminants. Institut de l'Elevage-INRA, Paris, France. pp 154.

Charallah S, Amirat Z, Sulon J, Khammar F and Beckers J-F 2009 Pregnancy-Associated Glycoprotein and Progesterone Concentrations during Pregnancy Failure in Bedouin Goat from the Southwest of Algeria. Reproduction in Domestic Animals, 45, 231–238.

Chentouf M, Bister J L and Boulanouar B 2011 Reproduction characteristics of North Moroccan indigenous goats. Small Ruminant Research, 98, 185–188.

Dubreuil P et Arsenault J 2003 Les avortements chez les petits ruminants. Le médecin vétérinaire du Québec, 33, 6–12.

Engeland I V, Waldeland H, Andresen Ø, Løken T, Björkman C and Bjerkås I 1998 Foetal loss in dairy goats: an epidemiological study in 22 herds. Small Ruminant Research, 30, 37–48.

Engeland I V, Waldeland H, Andresen Ø and Tverdal A 1997 Foetal loss in dairy goats: an epidemiological study in 515 individual goats. Animal reproduction science, 49, 45–53.

FAO 2012 Etat des Ressources Génétiques Forestières dans le Monde, Rapport National, Algérie, 1–58.

Ganter M 2014 Bluetongue disease-Global overview and future risks. Small Ruminant Research, 118, 79–85.

Ghanem Y M, El-Khodery S A, Saad A A, Elragaby S A, Abdelkader A H and Heybe A 2009 Prevalence and risk factors of caprine arthritis encephalitis virus infection (CAEV) in Northern Somalia. Small Ruminant Research, 85, 142–148.

Gilbert R O, Shin S T, Rabuffo T S and Chandler S K 1996 An in vitro model for the study of bovine endometrial physiology and pathophysiology. Proceedings of the 13th International Congress on Animal Reproduction and Artificial Insemination. Sydney, Australia. pp II–1.

Hussain Q, Waldeland H, Havrevoll Ø, Eik L O, Andresen Ø and Engeland IV 1996 Effect of type of roughage and energy level on reproductive performance of pregnant goats. Small Ruminant Research, 21, 97–103.

Mellado M and Meza-Herrera CA 2002 Influence of season and environment on fertility of goats in a hot-arid environment. The Journal of Agricultural Science, 138, 97–102.

Mellado M, Valdez R, Garcia J E, López R and Rodriguez A 2006 Factors affecting the reproductive performance of goats under intensive conditions in a hot arid environment. Small Ruminant Research, 63, 110–118.

Mellado M, Valdez R, Lara L M and Garcia JE 2004 Risk factors involved in conception, abortion, and kidding rates of goats under extensive conditions. Small Ruminant Research, 55, 191–198.

Menzies P I 2012 Vaccination programs for reproductive disorders of small ruminants. Animal reproduction science, 130, 162–172.

Moeller R B 2001 Causes of caprine abortion: diagnostic assessment of 211 cases (1991–1998). Journal of Veterinary Diagnostic Investigation, 13, 265–270.

Shalaby A S, Sharawy S M, Saleh N H and Medan M S 2000 Reproductive performance of Damascus goats in semi-arid areas in Egypt. Proceedings of the 7th International Conference on Goats. Institut de l'élevage, France. pp 424–425.

Silanikove N 1984 Renal excretion of urea in response to changes in nitrogen intake in desert (black Bedouin) and non-desert (Swiss Saanen) goats.Comparative Biochemistry and Physiology Part A: Physiology, 79, 651-654.

Silanikove N, Tagari H and Shkolnik A 1993 Comparison of rate of passage, fermentation rate and efficiency of digestion of high fiber diet in desert Bedouin goats compared to Swiss Saanen Goats. Small Ruminant Research, 12, 45-60.

Singh S V, Singh N, Singh M P, Shankar H and Lalwani D D 1994 Occurrence of abortions and seroprevalence of brucellosis in goats and sheep. Small Ruminant Research, 14, 161–165.

Waldeland H and Løken T 1991 Reproductive failure in goats in Norway: an investigation in 24 herds. Acta veterinaria Scandinavica, 32, 535.

Wilson R T and Durkin J W 1988 Small ruminant production in central Mali: reproductive performance in traditionally managed goats and sheep. Livestock Production Science, 19, 523–529.

Received 16 May 2018; Accepted 12 June 2018; Published 3 July 2018

Go to top