the VETERINARY RECORD | July 4, 2009 PAPERS Comparison of analyses of respiratory gases made with the K4b 2 portable and Quark laboratory analysers in horses P-M. Leprêtre, N. Metayer, G. Giovagnoli, E. Pagliei, E. Barrey P-M. Leprêtre, PhD, Laboratoire de Recherche ‘Adaptations Physiologiques à l’Exercice et Réadaptation à l’Effort’, Faculté des Sciences du Sport, Université de Picardie Jules Verne, Allée P. Grousset 80025 Amiens Cedex 1, France N. Metayer, MS, E. Barrey, DVM, PhD, Unité de Biologie Intégrative des Adaptations à l’Exercice, INSERM 902, Genopole, Université d’Evry, Val d’Essonne, 3 Impasse Christophe Colomb, 91000 Evry, France G. Giovagnoli, DVM, PhD, Italian Equestrian Federation, Viale Tiziano 74, 00196 Rome, Italy E. Pagliei, MS, COSMED, Via Dei Piani Di Monte Savello 37, Pavona Di Albano, 00196 Rome, Italy Ms Metayer and Professor Barrey are also at Institut National de la Recherche Agronomique, Génétique Animale et Biologie Intégrative, 78352 Jouy-en-Josas, France E-mail for correspondence: pierre-marie.lepretre@u-picardie.fr Veterinary Record (2009) 165, 22-25 Measurements of minute ventilation (V E ) and expired oxygen and carbon dioxide fractions (FeO 2 and FeCO 2 ) were measured at rest and during exercise in seven warmblood horses performing two consecutive standardised incremental treadmill exercise tests at submaximal speed, using the portable K4b 2 telemetric unit and the laboratory Quark metabolic cart in random order. Oxygen consumption (VO 2 ) and carbon dioxide production (VCO 2 ) were estimated using the Haldane equation. There were no significant differences between the measurements made with the two devices. However, V E was overestimated when the horses were cantering at 10 m/s in comparison with values available in the literature. AUTOMATED systems are commonly used to analyse respiratory gases and measure maximal oxygen uptake (VO 2max ) in human beings and ani- mals under laboratory conditions (Bayly and others 1987, Attenburrow and others 1990, Foster and others 1993). Graded exercise methods are often used to elicit physiological responses (Barrey and others 1993), but this pattern of power output is different from that ordinarily used by human athletes during track and field exercises (Foster and others 1993). This led to the development of small, non-invasive portable devices that determine the lactate threshold and oxygen uptake (VO 2 ) during outdoor activities. Several studies have demonstrated the accu- racy of portable K4b 2 telemetric devices (COSMED) that measure VO 2 , carbon dioxide production (VCO 2 ), minute ventilation (V E ) and respi- ratory gas ratio (RER) in human beings (De Lorenzo and others 2001, McLaughlin and others 2001, Pinnington and others 2001, Maiolo and others 2003). The metabolic status of animals has generally been assessed by meas- urements of airflow through chambers into which the animals have been placed (Bedford and others 1979) or that have been placed over their heads (Seeherman and others 1981), methods that could not be applied to horses. In horses, measurements of gas exchange are difficult because of their large V E and peak flow. However, high-speed treadmills have made it possible to study the cardiorespiratory function of athletic horses using the Douglas bag method (Evans and Rose 1988a, b). In studies by Barrey and others (1993) and Couroucé and others (1999) comparing treadmill tests with track and field exercises, factors such as the biomechanics of locomotion and the atmospheric conditions may partly explain the dif- ferences in physiological and locomotor variables measured, despite the fact that the horse was pulling a sulky and a driver on the track, and was unloaded on the treadmill. During field exercise tests, a portable device for measuring respiratory gases could be an advantage for measurements of VO 2 in horses. In a pilot study, Castellano and others (1997) used the K4b 2 telemetric system in two thoroughbreds and three standardbred horses galloping on turf for 2250 m. Art and others (2006) reported that the metabolic values obtained with the K4b 2 were reproducible at any workload, in five untrained saddle-bred horses engaged in a progressive speed treadmill test. They also observed a highly significant relationship between the VO 2 values obtained with the K4b 2 portable gas analyser and a fixed mass spectrometer gas analyser. However, the face mask and sam- pling line were different in the two systems. COSMED has also developed Quark, a standard laboratory metabolic cart compatible with the K4b 2 sampling line. The purpose of this study was to compare the respiratory gas exchange values obtained with the portable K4b 2 and the fixed Quark analyser in horses running at increasing speed, to test the validity of the portable device under laboratory conditions. Materials and methods Seven trained Anglo-Arabian and thoroughbred cross warmblood horses with a mean (sd) age of 8·4 (1·4) years, height of 200 cm and weighing 555 (41·5) kg were studied. They performed two incremental tests, in randomised order (Quark v K4b 2 ), on a high-speed treadmill (Mustang 2200; Kagra) with a 3 per cent incline to reproduce the same exercise intensity as on a track (Barrey and others 1993). In this incremental protocol, after a three-minute warm-up walking at 1·6 m/s, each horse performed four three-minute stages: trotting at 4·0 m/s, and cantering at 8·0, 9·0 and 10·0 m/s, for a total period not exceeding 20 minutes. Before the exercise test, each horse was accustomed five times to running on the treadmill while wearing the mask. The protocol and care of the horses were in compliance with the Declaration of Helsinki.