The use of lingual venous blood to determine the acid-base and blood-gas status of dogs under anesthesia.

Objective: To assess the suitability of lingual venous blood (LBG) as an alternative to arterial blood (ABG) samples in determining acid-base balance and blood-gas status in dogs anesthetized for elective procedures and with medetomidine and isoflurane administration under experimental conditions.

Study Design: Prospective, randomized clinical and experimental study.

Animals: Clinical population of 18 ASA I/II dogs for elective surgery and five healthy Beagles (3 females and 2 males) for the experimental study.

Methods: Blood sampling was simultaneously performed at dorsal pedal arterial and lingual venous sites, generating paired data. Two paired samples were collected from each dog in the clinical part and four from each dog in the experimental part (two during isoflurane anesthesia and two during isoflurane plus medetomidine). A modified Bland and Altman method was used to examine data from the clinical part and the experimental data were subjected to a paired sign's test following transformation where appropriate.

Results: The pH of LBG overestimated ABG, with limits of agreement of (-0.01, 0.02). The partial pressure of carbon dioxide (PCO(2)) of LBG overestimated ABG by 0.6 mmHg [0.1 kPa], with limits of agreement of (-3.5, 4.6) mmHg [-0.5, 0.6 kPa]. The partial pressure of oxygen (PO(2)) of LBG underestimated ABG by 86.3 mmHg [-11.5 kPa], with limits of agreement of (-199.8, 27.3) mmHg [-26.6, 3.6 kPa]. During medetomidine administration values for PO(2) (p = 0.03) and lactate (p = 0.03) were lower for LBG when compared with ABG. The LBG value of PO(2) was lower (p = 0.03) during medetomidine and isoflurane administration versus isoflurane alone.

Conclusions And Clinical Relevance: The pH and PCO(2) of LBG samples provide clinically acceptable substitutes of ABG samples in the dog population studied. The wider limits of agreement for PO(2) render it less reliable as a substitute for ABG. The difference in PO(2) identified between LBG and ABG during medetomidine administration may not preclude the use of LBG as substitutes for ABG samples.