Is the relationship between abdominal and rectal temperature constant in healthy and lipopolysaccharide-stimulated, deoxynivalenol-fed pigs?

Body core temperature is an important clinical parameter for monitoring infections in mammals and can be measured with different devices. In practice rectal measurement is the standard method and equated to the body core. In order to investigate the constancy of the relationship between practical body core temperature (rectal) and abdominal temperature under physiological and pathophysiological conditions, we compared both devices in healthy pigs as well as in pigs exposed to LPS and/or DON.

A total of 44 barrows were exposed for 4 weeks either to a DON-contaminated (4.59mg DON/kg feed) or a control (CON) diet. They were surgically equipped with an intraabdominal temperature logger (measurement every 5min) and permanent catheters to facilitate infusion with 0.9%NaCl (CON) or LPS (7.5µg/kg BW) either into portal or jugular vein for 60 min. Rectal body temperature was taken every 15min, from 30min before until 180 min after start of infusion. The combination of diet and infusion created six groups: CON_CON_jug.-CON_por., CON_CON_jug.-LPS_por., CON_LPS_jug.- CON_por., DON_CON_jug.-CON_por., DON_CON_jug.-LPS_por., DON_LPS_jug.-CON_por.. Data were evaluated by PROC MIXED with group and time as main factors and their interaction and with PROC CORR for method relationship (SAS Enterprise Guide 6.1). Linear regression for both temperature methods was calculated with Statistica (StatSoft, Inc. 1984-2014).

Both methods were significantly correlated in all experimental groups (p< 0.001). Under physiological conditions both temperatures remained stable during observation period, whereby rectal temperature (38.8°C) was constantly ~1°C lower than intraabdominal measurement (39.7°C). The linear regression showed an increase of 0.68°C intraabdominal per 1°C rectal temperature. Under pathophysiological conditions LPS induced an increase in body temperature (p< 0.01), starting at 15 min p.i. and reaching its plateau at 60min p.i., also with a constant difference of ~1°C. Due to differences in the degree of the rise of the slope the linear regression line is significantly steeper (p=0.02) in CON_CON_jug.-LPS_por. (0.97°C) compared to CON_LPS_jug.-CON_por. (0.69°C). Data showed also an impact of DON, with DON_LPS_jug.-CON_por. resulting in ~0.5°C lower intraabdominal temperature compared to CON_LPS_jug.- CON_por (p=0.08). Also, the linear regression between both methods showed a steeper slope (p< 0.001) for DON_LPS_jug.-CON_por. (0.87°C) compared to CON_LPS_jug.-CON_por. (0.69°C).

Rectal temperature does not reflect the actual body core, being constantly ~1°C lower. Also, uniformity of the relationship between both methods cannot be equalized for both, physiological and pathophysiological conditions.

Disclosure of Interest: None Declared.