SUMMER INFERTILITY IN SOWS IN FRANCE: DOES THE SUMMER

TEMPERATURE REALLY PLAY A PART?

R. Krejcí1, V. Auvigne2, P. Leneveu3, C. Jehannin4, E. Sallé1

1
Ceva Santé Animale, ZI La Ballastière, BP 126, 33501 Libourne Cedex France ; 2 Ekipaj, C/
Valencia 3, 28223 Pozuelo de Alarcón, Madrid, Spain; 3 Ispaia, Zoopole les Croix, BP 7, 22440
Ploufragan, France ; 4 Gènes Diffusion, 3595 route de Tournai, BP 70023, 59501 Douai Cedex,
France ; E-mail : vincent.auvigne@ekipaj.com

Summer infertility is described in many countries, from Australia to Finland. The relative
importance of parameters such as high temperatures, lactational catabolism or photoperiod is
widely discussed.
The results of five years (2003 to 2007) of ultrasound pregnancy diagnosis carried out in
266 indoor farms were analyzed. For all farms the data cover the entire study period. Farms were
situated in 4 areas, with a marked difference in average summer temperatures, in the west and
north of France. The data of 22,773 batches and 610,117 sows were included. Summer infertility
was defined as the difference between the fertility rate in “summer” (inseminations during weeks
25 to 42) and “winter” (inseminations during weeks 1 to 18 of the same year). For each farm,
baseline fertility was defined as the average winter fertility over the past five years. In each area,
two meteorological variables were defined on a weekly basis, using data from a reference weather
station: the number of warm days (maximum temperature ≥ 25 °C) and the number of tropical
days (maximum temperature ≥ 32 °C and minimum temperature ≥ 18 °C). Statistical unit was
“farm year” (n=1330) using Mann-Whitney/Wilcoxon test (EpiInfo 3.4.3 – November 2007).
The mean fertility was 85%. The median summer infertility was 2.8% and more than 7.1%
for a quarter of farms. Summer infertility was observed every year in every area. Summer
infertility did not vary between areas and was independent of the baseline fertility of the farms
(Table 1, p=0.38). In the four areas, the summer of 2003 was exceptionally hot and the summer of
2007 exceptionally cold. Summer infertility during 2003 was significantly higher (p< 0.01) than
the other four years, which did not differ among each other. Thus infertility had not improved
during the summer of 2007 even if this summer was exceptionally cold. Lastly, the correlation
between summer infertility of two subsequent years on the same farm is significant but weak.
It can be concluded that peaks of high temperatures may exacerbate summer infertility.
However, other yearly recurrent factors (such as photoperiod) probably play an essential role in
this phenomenon.

Summer infertility (%)

Baseline fertility n 1st quart. Median 3rd quart.
<=82.6 % 335 -8.8 -3.5 1.6
>82.6 – 86.2 % 345 -6.9 -2.3 1.3
>86 – 89.4 % 315 -7.2 -3 0.1
>89.4 % 335 -5.8 -2.5 0.3
Table 1: Summer infertility is not linked to baseline fertility (p=0.38)