231
mixed blood was sampled at two occasions (April 2016 and April 2017) and blood
smears were prepared on site.
2.3 Laboratory Analyses
All sera were analysed for antibodies against BVDV, Brucella spp. and N. caninum
using enzyme-linked immunosorbent assay (ELISA) from Boehringer Ingelheim
Svanova in Uppsala, Sweden. The optical density (OD) was measured in a
LabSystems Multiskan EX spectrophotometer (LabX, Ontario, Canada). All OD
values were correlated to positive control sera referred to as corrected OD values,
and percent positivity (PP) of each sample was calculated following the manufacturer’s instructions. Anti-BVDV antibodies were detected using an indirect ELISA
commercial kit (SVANOVIR® BVDV-Ab I-ELISA) following the manufacturer’s
instructions. Serum samples with PP ≥10% were considered positive and those with
PP <10% were negative. The presence of antibodies to Brucella spp. was analysed
using the SVANOVIR® Brucella-Ab-I-ELISA antibody test following the manufacturer’s instructions. Serum samples with PP ≥40% were considered positive, and
samples with PP <40% were considered negative. N. caninum specific antibodies
were detected using an indirect SVANOVIR® N. caninum ISCOM ELISA following manufacturer instructions. Sera were considered positive when PP ≥20%, while
samples with PP <20% were considered negative.
The air-dried blood smears were fixed in methanol, stained with Giemsa stain
and examined for evaluation of Theileria spp., Babesia spp. and Trypanosoma spp.
in microscopy at 100 × magnification by a trained technician from Central Veterinary
Laboratory, Lilongwe, Malawi.
3 Statistical Analyses
The statistical software package STATA 14 (StataCorp LLC, Texas, USA) was used
in all statistical analyses. Statistical significance was considered with a P-value less
than 0.05 in all models. Variance components mixed model analyses using the
‘xtmixed’ procedure were performed allowing REML (restricted maximum likelihood) estimation to discriminate between variance of calving interval at cow, farm
and section level. Agricultural sections and farm were included as random effect
variables both separately and jointly in three models.
Univariable Kaplan–Meier (K-M) estimators were used to assess relationships
between time from calving to pregnancy for the explanatory variables; presence of
a breeding bull in the herd (1/0) and first parity versus multiparous cows (1/0).
The time variable describing days from calving to pregnancy was calculated for
cows that delivered a live calf within the period June 5, 2016, to October 6, 2017.
All other cows were censored at the end of the study, at culling or death or when a
A Cohort Study of Reproductive Performance, Associated Infections and Management…
mixed blood was sampled at two occasions (April 2016 and April 2017) and blood
smears were prepared on site.
2.3 Laboratory Analyses
All sera were analysed for antibodies against BVDV, Brucella spp. and N. caninum
using enzyme-linked immunosorbent assay (ELISA) from Boehringer Ingelheim
Svanova in Uppsala, Sweden. The optical density (OD) was measured in a
LabSystems Multiskan EX spectrophotometer (LabX, Ontario, Canada). All OD
values were correlated to positive control sera referred to as corrected OD values,
and percent positivity (PP) of each sample was calculated following the manufacturer’s instructions. Anti-BVDV antibodies were detected using an indirect ELISA
commercial kit (SVANOVIR® BVDV-Ab I-ELISA) following the manufacturer’s
instructions. Serum samples with PP ≥10% were considered positive and those with
PP <10% were negative. The presence of antibodies to Brucella spp. was analysed
using the SVANOVIR® Brucella-Ab-I-ELISA antibody test following the manufacturer’s instructions. Serum samples with PP ≥40% were considered positive, and
samples with PP <40% were considered negative. N. caninum specific antibodies
were detected using an indirect SVANOVIR® N. caninum ISCOM ELISA following manufacturer instructions. Sera were considered positive when PP ≥20%, while
samples with PP <20% were considered negative.
The air-dried blood smears were fixed in methanol, stained with Giemsa stain
and examined for evaluation of Theileria spp., Babesia spp. and Trypanosoma spp.
in microscopy at 100 × magnification by a trained technician from Central Veterinary
Laboratory, Lilongwe, Malawi.
3 Statistical Analyses
The statistical software package STATA 14 (StataCorp LLC, Texas, USA) was used
in all statistical analyses. Statistical significance was considered with a P-value less
than 0.05 in all models. Variance components mixed model analyses using the
‘xtmixed’ procedure were performed allowing REML (restricted maximum likelihood) estimation to discriminate between variance of calving interval at cow, farm
and section level. Agricultural sections and farm were included as random effect
variables both separately and jointly in three models.
Univariable Kaplan–Meier (K-M) estimators were used to assess relationships
between time from calving to pregnancy for the explanatory variables; presence of
a breeding bull in the herd (1/0) and first parity versus multiparous cows (1/0).
The time variable describing days from calving to pregnancy was calculated for
cows that delivered a live calf within the period June 5, 2016, to October 6, 2017.
All other cows were censored at the end of the study, at culling or death or when a
A Cohort Study of Reproductive Performance, Associated Infections and Management…