Effects of anesthesia and blood sampling techniques on plasma metabolites and corticosterone in the rat.

Blood is routinely sampled from laboratory animals in biomedical research, and many of the commonly applied sampling techniques require anesthesia. Acute effects of many sampling and anesthesia procedures may confound the results, but those effects are incompletely characterized. We here compare the effects of four common anesthesia procedures (inhalation anesthesia with ether (EA) or isoflurane (IA) and intraperitoneal injection anesthesia with xylazin/ketamine (XKA) or medetomidine/midazolam/fentanyl (MMFA)) on plasma concentrations of glucose, lactate, non-esterified fatty acids (NEFAs), and corticosterone in blood obtained from a previously implanted jugular vein (JV) catheter with the effect of JV blood sampling from non-anesthetized, freely-moving rats (JV-NA). Also, we included in the comparison two other blood sampling procedures usually performed without anesthesia (NA), i.e., puncture of the saphenic vein (SV) and tail incision (TI). Whereas the control procedure (JV-NA) did not significantly affect any of the target parameters, plasma glucose increased from 14 (JV-IA) to 44 (JV-MMFA) % (all Ps=0.05 when compared with the control procedure) in all blood samples collected in anesthesia and was 12 and 14% lower (both Ps<0.05) in SV-NA and TI-NA samples, respectively. Plasma lactate increased from 74 (JV-IA) to 226% (SV-NA) (all Ps<0.05) with all sampling and anesthesia procedures except for JV-XKA and JV-MMF. Plasma NEFAs increased to 52% (P<0.05) with the TI-NA procedure and appeared to decrease with the JV-IA and JV-MMFA procedures (both Ps>0.05). Finally, only the JV-EA and the JV-MMFA procedures increased plasma corticosterone (+525 and +353%, respectively, both Ps< 0.05). The JV-IA and JV-XKA procedures appeared to increase it as well, but these differences did not reach statistical significance. Thus, anesthesia and blood sampling procedures can have profound acute effects on plasma metabolite and hormone concentrations. This must be considered for the design and interpretation of blood sampling experiments in laboratory animals.