Evaluation of high-definition and conventional oscillometric blood pressure measurement in anaesthetised dogs using ACVIM guidelines

Transcription

1 ttp:// PAPER Evaluation of high-definition and conventional oscillometric blood pressure measurement in anaesthetised dogs using ACVIM guidelines OBJECTIVES: To evaluate high-definition and conventional oscillometry in comparison with direct blood pressure measurements in anaesthetised dogs. METHODS: Eight simultaneous readings for systolic, diastolic and mean pressure were obtained directly and with each of two devices in nine anaesthetised dogs. Measurement procedure and validation were based on the 7 ACVIM guidelines. RESULTS: Sixty-three simultaneous readings were evaluated for each device and direct measurements. The mean differences (bias) to direct values were within 1 mmhg for both devices although bias for systolic and diastolic blood pressures was higher for Memodiagnostic. The standard deviations of differences (precision) were within 15 mmhg for Dinamap but exceeded for Memodiagnostic. Correlation coefficients were higher for Dinamap than Memodiagnostic but both failed to reach a correlation of 9. Over 5% of values lay within 1 mmhg of direct measures for both devices, but this percentage was greater for Dinamap than Memodiagnostic. Over 8% of values lay within mmhg of direct measures for Dinamap but not for Memodiagnostic. CLINICAL SIGNIFICANCE: Both devices failed to meet ACVIM guideline validation. However, Dinamap only failed with regards to correlation. Memodiagnostic failed on several requirements, and based on poor correlation, accuracy and precision, this device cannot be currently recommended for dogs under anaesthesia. M. WERNICK, M. DOHERR*, J. HOWARD, AND T. FRANCEY Journal of Small Animal Practice (1) 51, DOI: /j x Accepted: 26 January 1; Published online: 11 May 1 Small Animal Clinic, Department of Small Animal Internal Medicine, *Department of Clinical Research and Veterinary Public Health and Clinical Laboratory, Department of Veterinary Medicine, Vetsuisse Faculty, University of Bern, Bern, Switzerland M. Wernick s current address is the Clinic for Zoo Animals, Exotic Pets and Wildlife, Vetsuisse Faculty, University of Zurich, Bern, Switzerland INTRODUCTION The importance of detecting alterations in systemic blood pressure (BP) is increasingly recognised in veterinary medicine. As early detection and treatment of hypertension may avert serious and potentially fatal sequela, measurement of BP in dogs and cats affected by a variety of disorders is frequently performed in clinical practice (Haberman and others 6, Brown and others 7). In addition, continuous BP monitoring of dogs and cats under general anaesthesia has long since become routine, in particular because of the importance of recognising hypotension (Sawyer and others 4). Direct intra-arterial BP measurement is the accepted gold standard but is not commonly used in clinical practice because it is technically demanding and invasive. Instead, a variety of non-invasive devices are in widespread use (Grandy and others 1992, Stepien and others 3, Sawyer and others 4, Mandigers 5, Habermann and others 6, Brown and others 7, Deflandre and Hellebrekers 8). Clinicians expect that comparative studies showing that measurements are meaningful are performed when new non-invasive BP devices are introduced. Before evaluating devices in conscious clinical patients, initial studies are often performed comparing results with direct BP measurements in animals under anaesthesia (Geddes and others 198, Sawyer and others 1991, Bodey and others 1994, Gains and others 1995, Meurs and others 1996, Sawyer and others 4), but no specific requirements for validation of veterinary BP devices exist. As a result, the manner in which devices are assessed varies greatly from study to study, making direct comparisons difficult between assessments of different devices in different reports. In an attempt to offer veterinary practitioners guidelines to standardise BP assessment, the ACVIM issued a consensus 318 Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association

2 Evaluation of oscillometric devices statement for the identification, evaluation and management of systemic hypertension in dogs and cats in 7 (Brown and others 7). These guidelines include recommendations for device calibration, patient restraint, cuff size and position, and patient positioning for BP measurement. Instructions for the validation of BP devices in anaesthetised and conscious animals, based on guidelines of the Association for the Advancement of Medical Instrumentation (AAMI), are listed in the Consensus Statement appendix and include specific conditions under which systemic efficacy of a device is satisfied. Oscillometric BP devices have been in widespread clinical use for many years in small animals. Recently, a new high- definition oscillometric device, the Memodiagnostic (S+B MedVET GmbH, Germany) was introduced and is marketed in Europe and USA. This device holds promise for delivering accurate and reliable measures of BP in dogs and cats; however, no reports on the validity of BP readings with this device have been published. If clinicians are to opt for use of this device, they would expect to know how measurements compare with a gold standard as well as with conventional oscillometry. One of the most widely assessed conventional oscillometric devices has been the Dinamap (Critikon, Tampa, FL, USA). A variety of studies evaluating the Dinamap 1846sx, Dinamap 81 and the Dinamap 83, specifically developed for clinical veterinary patients in 1993, have been published (Bodey and others 1994, 1996, Binns and others 1995, Gains and others 1995, Meurs and others 1996, Bodey and Mitchell 1997, Grosenbaugh and Muir 1998, Haberman and others 6, Sawyer and others 4), but no study has evaluated this device in accordance with the 7 ACVIM guidelines and so direct comparison with future devices is difficult. The aim of this study was to evaluate the high-definition oscillometric device, Memodiagnostic MD_15/9 Pro, and the conventional oscillometric device, Dinamap Model 83, compared with direct arterial BP measurement in anesthetised dogs based on the guidelines of the 7 ACVIM Consensus Statement. MATERIALS AND METHODS Nine client-owned dogs were recruited from patients requiring anaesthesia for reasons unrelated to the study in the Small Animal Clinic, Department of Clinical Veterinary Medicine, Vetsuisse Faculty, University of Bern. The study protocol was evaluated and approved by the institution s Ethics Committee. Dogs were recruited by requesting owner consent for participation in the study of dogs scheduled for a surgical procedure. The measurements were performed after reaching a stable plane of anaesthesia, but before any other procedures, in the surgery preparation room with the dogs in dorsal recumbency. For oscillometric devices, the BP cuff was placed around the mid left or right antebrachium overlying the median artery. For conventional oscillometric measurements, cuff size was selected to account for approximately 4% of the circumference of the limb at placement site. For high-definition oscillometric measurement, the cuff used was selected from one of those provided with the device according to the manufacturer s guidelines. Care was taken to ensure that measurements were taken at the level of the heart base and that ECG readings showed a sinus rhythm. For direct arterial pressure measurement, the skin over the left or right dorsal hind paw was surgically prepared, and a - or 22-gauge, 1-inch fluoro-ethylene-propylene catheter (Jelco 2 i.v., Smiths Medical International Ltd., Kent, UK) was placed in the dorsal pedal artery. The arterial catheter was connected to a disposable transducer via a non-compliant saline-filled tubing. The transducer was connected to a monitor and attached to a pressurised (3 mmhg) bag of saline ( 9% NaCl) solution. The catheter was flushed intermittently to avoid blood clots, and the system was cleared of air bubbles. The transducer was placed at the level of the heart and zeroed to atmospheric pressure before starting the procedure. Before all measurements, the direct BP monitor and the oscillometric devices were calibrated from to mmhg using a standard aneroid manometer that met the standards for non-automated sphygmomanometers established by the AAMI. At the end of the study, the calibration was rechecked to ensure that their accuracy was maintained. A total of eight consecutive readings were taken simultaneously by direct BP measurement and with each of the oscillometric devices (Dinamap Model 83 and Memodiagnostic MD_15/9 Pro). For each single measurement, the systolic (SAP), diastolic (DAP) and mean (MAP) BPs and the heart rate were recorded. Other data recorded included date, start time of the measurement session, the circumference of the dog s limb, the size of the BP cuff, and drugs and intravenous infusions administered to the patient. At least 5 seconds elapsed between consecutive measurements. Throughout the entire study, all measurements were taken by a single investigator, familiarised with all devices and the protocol for BP measurements conformed to the ACVIM consensus statement (see Table 2 in Brown and others 7). Based on these guidelines, the first in the series of eight consecutive BP measurements was discarded and not used for data comparison. STATISTICS AND VALIDATION Simple linear regression analysis was used to evaluate the relationships between the direct values and the indirect values obtained with each of the oscillometric devices. The agreement between oscillometric and simultaneous direct BP measurements for SAP, DAP and MAP were evaluated using Bland and Altman plots (Bland and Altman 1986, 1995) in which the differences between the two measurements were plotted against the mean of the two measurements for each set of simultaneous measurements. The bias (mean difference from direct BP values), precision (standard deviation of differences) and the 95% limits of agreement (mean±2 sd) were calculated. The percentages of measurements within 5, 1 and mmhg of the direct measures for both indirect devices were calculated. The indirect devices were considered to pass validation if they met all of the following conditions (Brown and others 7): the mean difference of paired measurements for SAP and DAP treated separately is ±1 mmhg or less with a standard deviation of 15 mmhg or Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association 319

3 M. Wernick and others less; the correlation between paired measures for SAP and DAP treated separately is 9 across the range of measured values of BP; 5% of all measurements for SAP and DAP treated separately lie within 1 mmhg of the direct values; 8% of all measurements for SAP and DAP treated separately lie within mmhg of direct values; data are obtained for no fewer than eight animals. In addition, the devices were considered to give reliable readings only if the coefficient of variation for each set of seven measurements for SAP was below % (Brown and others 7). Analyses were performed using statistical software (Medcalc v11..1, Medcalc Software bvba, Belgium). RESULTS Direct SAP - Dinamap SAP Average SAP (mmhg) Dogs recruited were between 3 and 11 years old (mean age 6 5±2 87 years) and weighed between 11 8 and 45 kg. Readings were obtained for all 72 measurements with both the Dinamap and the Memodiagnostic devices. All sets of values for SAP had a coefficient of variation of less than % for all devices; the ranges of coefficients of variation for SAP were 1 9 to 12 11% (median 2 3%) for direct, 1 15 to 13 22% (median 3 7%) for Dinamap and 1 87 to 9 24% (median 4 89%) for Memodiagnostic. Comparative studies of values obtained with oscillometric and direct measurements were performed after discarding the first value (Brown and others 7), giving a total of 63 simultaneous data for each of the two oscillometric devices. Bland and Altman plots revealed a slight underestimation of SAP and DAP and overestimation of MAP with the Dinamap (Fig 1, Table 1). Values for SAP, DAP and MAP were all overestimated with the Memodiagnostic and overestimation increased for SAP and DAP with increasing pressures (Fig 2, Table 1). A positive correlation was found between all indirect and simultaneously measured direct values (Figs 3 and 4). The correlation was poor with the Memodiagnostic (Table 1, Fig 4). Although a strong correlation was found with the Dinamap, this was slightly below 9, required for validation (Table 1, Fig 3). The mean Direct DAP - Dinamap DAP (mmhg) Direct MAP - Dinamap MAP (mmhg) Average DAP (mmhg) Average MAP (mmhg) FIG 1. Bland and Altman plots for sytolic (SAP), diastolic (DAP) and mean (MAP) arterial pressures using the Dinamap device in nine anaesthetised dogs with seven readings per dog (63 comparisons). Each data point corresponds to the difference between the direct and the Dinamap measurements plotted against the average of the two simultaneously obtained results. Also depicted are the mean overall difference (solid line) and the limits of agreement (dashed lines) 3 Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association

4 Evaluation of oscillometric devices Table 1. Agreement and correlation of blood pressures measured by use of two oscillometric devices and those measured directly in nine anaesthetised dogs Parameter ACVIM recommended limits for systolic and diastolic values (Brown and others 7) Dinamap Memodiagnostic Systolic Diastolic Systolic Diastolic Correlation (R) 9 885* 8734* 865* 5546* 7186* 5671* Bias (mean difference, ± mmhg) 1 Precision (Standard deviation, mmhg) Limits of agreement (mmhg) none 13 to to to to to to mmhg none 57 1% 47 6% 49 2% 22 2% 42 8% 26 9% 1 mmhg 5% 76 2% 88 9% 76 2% 52 4% 58 7% 61 9% mmhg 8% 1% 98 4% 98 4% 82 5% 76 2% 96 8% Range (mmhg) 76 to to to to to to 96 *Significant linear relationship (P< 1). Direct-indirect method. Percentage of indirectly obtained values within 5, 1 or mmhg of the direct values. Range of values obtained by direct arterial measurements. differences (bias) between direct and indirect measures were greater for the Memodiagnostic than the Dinamap (Table 1) although all were within 1 mmhg required for validation. The precision (standard deviation of the differences) was less than 1 mmhg for the Dinamap but exceeded the 15 mmhg threshold required for validation with the Memodiagnostic for both SAP and DAP (Table 1). The limits of agreement were likewise very large for the Memodiagnostic (Table 1). Greater than 75% of oscillometric readings were within 1 mmhg and the vast majority were within mmhg of direct values for the Dinamap (Table 1). However, only between 5 and 6% were within 1 mmhg for the Memodiagnostic (Table 1). Moreover, the Memodiagnostic did not fulfil the requirement of over 8% of values within mmhg for DAP. DISCUSSION This study compared BP between two oscillometric devices and direct measurements in anaesthetised dogs based on the ACVIM guidelines. This included a standardised approach to measurements (same position of animals, same cuff site and size, serial measurements with discard of the first measurement, a single examiner) with validation according to the suggested guideline. Although disease state and anaesthetic protocol might significantly affect BP, this was considered irrelevant to the aim of the study, and no study inclusion limitations were enforced in this respect. A shortcoming of this study is that dogs were evaluated under anaesthesia, and the range of BPs was relatively narrow (direct SAP between 76 and 137 mmhg). Results of this study are therefore only valid for anaesthetised dogs with pressures within this range. The Memodiagnostic oscillometric device is a relatively new device using high-definition oscillometry. There are currently no publications evaluating this device in dogs but the developer claims many advantages to conventional indirect BP devices (S+B Med Vet 1998, Egner 6a, 6b). Contrary to conventional oscillometry, in which the MAP is measured, and SAP and DAP are calculated using an algorithm based on the oscillation curve, the Memodiagnostic performs real-time analysis of artery oscillations to obtain pulse amplitudes. Other purported benefits are the electronically controlled values that adapt to maintain linearity during deflation, permitting readings from 5 to 3 mmhg, high sensitivity that allows measurements from minimal signals and heart frequencies of up to 5 beats per minute, and high-speed analysis that allows the detection of arrhythmias (S+B Med Vet 1998). In a previously published study, this device was tested for its application in the cynomolgus monkey (Schmelting and others 9), both under general anaesthesia in comparison with telemetry and in conscious animals. In this study, MAP was lower with the Memodiagnostic than with telemetry. However, this study did not evaluate simultaneously obtained paired readings, and the accuracy of the readings was therefore not assessed. The Dinamap has been previously evaluated in anesthetised dogs in several studies, most of which found favourable results compared with direct measurements (Hamlin and others 1982, Bodey and others 1994, Gains and others 1995, Meurs and others 1996, Grosenbaugh and Muir 1998). However, these studies were performed before the ACVIM consensus, and evaluation procedure varied widely between studies. In this study, the Dinamap underestimated SAP and DAP, as was found in previous reports (Bodey and others 1996, Grosenbaugh and Muir 1998, Habermann and others 6). In contrast, the Memodiagnostic overestimated SAP and DAP. The coefficient of variation of consecutive values for SAPs in each dog was below % with both devices, showing adequate consistency in readings with each device. The correlation between Dinamap and direct measures was not calculated in some previous studies and varied so greatly in others (R for SAP between 24 and 98), not least because of variation in cuff site, number of consecutive measurements, position of dogs, and BP ranges, that Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association 321

5 M. Wernick and others direct comparison of data between studies is difficult (Hamlin and others 1982, Bodey and others 1994, Gains and others 1995, Meurs and others 1996, Grosenbaugh and Muir 1998). As this study used the same procedure for both devices, correlation can be compared between the two devices. Neither device fulfilled the validation requirement of R 9. However, contrary to the Memodiagnostic, the Dinamap achieved high correlations and fell just short of this validation criterion. The mean differences (bias) between SAP and DAP values obtained with the Dinamap and direct measurements was greater than the 1 mmhg requirement in two previous studies (Bodey and others 1994, Grosenbaugh and Muir 1998). However, this was only the case for hypertensive dogs but not for normo- or hypotensive dogs in one of these studies (Grosenbaugh and Muir 1998), and only for tail and distal hindlimb but not for proximal hindlimb cuff sites in the other study (Bodey and others 1994). Bias (mean difference) for DAP ranged to values up to 13 mmhg in these studies. We found a very low bias for the Dinamap and somewhat higher values for the Memodiagnostic, indicating better accuracy with the Dinamap. The standard deviations of the differences (precision) for SAP and DAP were within 15 mmhg for the Dinamap, as was the case in a previous study evaluating BP over a wide pressure range in anaesthetised dogs (Grosenbaugh and Muir 1998). However, standard deviations for the Memodiagnostic device were greater than 15 mmhg, indicating a relatively poor precision. The validation criteria that greater than 5% and greater than 8% of SAP and DAP values must lie within 1 and mmhg of direct measurements, respectively, has not been previously evaluated for the Dinamap in dogs under anaesthesia. In one study with conscious beagles, neither the Dinamap nor Doppler met this standard for 1 mmhg (Habermann and others 6) despite the fact that this criterion is less stringent than AAMI requirements of human indirect BP devices. In this study, the Dinamap far surpassed this requirement but the Memodiagnostic fell short. Direct SAP - MemoDiagnostic SAP (mmhg) Direct DAP - Memodiagnostic DAP (mmhg) Direct MAP - Memodiagnostic MAP (mmhg) Average SAP (mmhg) Average DAP (mmhg) Average MAP (mmhg) FIG 2. Bland and Altman plots for sytolic (SAP), diastolic (DAP) and mean (MAP) arterial pressures using the Memodiagnostic device in nine anaesthetised dogs with seven readings per dog (63 comparisons). Each data point corresponds to the difference between the direct and the Memodiagnostic measurements plotted against the average of the two simultaneously obtained results. Also depicted are the mean overall difference (solid line) and the limits of agreement (dashed lines) 322 Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association

6 Evaluation of oscillometric devices Dinamap SAP (mmhg) Dinamap DAP (mmhg) Direct SAP (mmhg) Direct DAP (mmhg) Despite the purported advantages of the Memodiagnostic, this device did not meet the ACVIM requirements for BP device validation in this study and demonstrated poor correlation with direct measures in anaesthetised dogs. Moreover, accuracy and precision were poorer than with Dinamap, leading to limits of agreement that may be clinically unacceptable. Furthermore, the percentages of measured values within 5, 1 or mmhg of those obtained directly were very low compared with those achieved with the Dinamap. Although the Dinamap met almost all validation criteria, both devices are likely to fair less well in satisfying criteria when measurements are taken in conscious dogs and across a greater range of BP values. The ACVIM Consensus Statement panel recognises that the guidelines may be so rigorous as to exclude any device presently in use when applied to conscious dogs and cats (Brown and others 7). The results of this study suggest that the criterion for correlation may indeed be too high as it could not be satisfied under the narrow conditions of controlled measurements in anaesthetised dogs in this study. However, the criteria for the percentage of values lying within 1 and mmhg of direct measurements may be too lenient, in particular because a difference of mmhg may significantly affect the clinical interpretation of results in patients. Dinamap MAP (mmhg) Direct MAP (mmhg) FIG 3. Results of linear regression analysis for systolic (SAP), diastolic (DAP) and mean (MAP) arterial blood pressures for the Dinamap device in nine dogs with seven readings per dog (63 comparisons). Each data point represents data obtained by the direct simultaneous measures. The dashed lines represent 95% confidence intervals. The linear regression equation was y = x (SAP); y = x (DAP) and y = x (MAP) Acknowledgements The authors would like to thank Dr Marcus Clauss for his assistance. References BINNS, S. H., SISSON, D. D., BUOSCIO, D. A. & SCHAEFFER, D. J. (1995) Doppler ultrasonographic, oscillometric sphygmomanometric, and photoplethysmographic techniques for noninvasive blood pressure measurement in anesthetized cats. Journal of Veterinary Internal Medicine 9, BLAND, J. M. & ALTMAN, D. G. (1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1, BLAND, J. M. & ALTMANN, D. G. (1995) Comparing methods of measurement: why plotting difference against standard method is misleading. Lancet 346, BODEY, A. R. & MICHELL, A. R. (1997) Longitudinal studies of reproducibility and variability of indirect (oscillometric) blood pressure measurements in dogs: evidence for tracking. Research in Veterinary Science 63, BODEY, A. R., MICHELL, A. R., BOVEE, K. C., BURANAKURL, C. & GARG, T. (1996) Comparison of direct and indirect (oscillometric) measurements of arterial blood pressure in conscious dogs. Research in Veterinary Science 61, Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association 323

7 M. Wernick and others MemoDiagnostic SAP (mmhg) Memodiagnostic DAP (mmhg) Memodiagnostic MAP (mmhg) Direct SAP (mmhg) Direct DAP (mmhg) Direct MAP (mmhg) FIG 4. Results of linear regression analysis for systolic (SAP), diastolic (DAP) and mean (MAP) arterial blood pressures for the Memodiagnostic device in nine dogs with seven readings per dog (63 comparisons). Each data point represents data obtained by the direct simultaneous measures. The dashed lines represent 95% confidence intervals. The linear regression equation was y = x (SAP); y = x (DAP) and y = x (MAP) BODEY, A. R., YOUNG, L. E., BARTRAM, D. H., DIAMOND, M. J. & MICHELL, A. R. (1994) A comparison of direct and indirect (oscillometric) measurements of arterial blood pressure in anaesthetised dogs, using tail and limb cuffs. Research in Veterinary Science 57, BROWN, S., ATKINS, C., BAGLEY, R., CARR, A., COWGILL, L., DAVIDSON, M., EGNER, B., ELLIOTT, J., HENK, R., LABATO, M., LITTMANN, M., POLZIN, D., ROSS, L., SNYDER, P. & STEPIEN, R. (7) Guidelines for the identification, evaluation and management of systemic hypertension in dogs and cats. Journal of Veterinary Internal Medicine 21, DEFLANDRE, C. J. A. & HELLEBREKERS, L. J. (8) Clinical evaluation of the Surgivet V646, a non invasive blood pressure monitor in anaesthetized dogs. Veterinary Anaesthesia and Analgesia 35, EGNER, B. (6a) Teil 1: technologie und vermeidung von messfehlern. Kleintier Konkret 4, EGNER, B. (6b) Teil 2: interpretation der messergebnisse und behandlungsstrategien. Kleintier Konkret 6, GAINS, M. J., GRODECKI, K. M., JACOBS, R. M., DYSON, D. & FOSTER, R. (1995) Comparison of direct and indirect blood pressure measurements in anesthetized dogs. Canadian Journal of Veterinary Research 59, GEDDES, L. A., COMBS, W., DENTON, W., WHISTLER, S. J. & BOURLAND, J. D. (198) Indirect mean arterial pressure in the anesthetized dog. American Journal of Physiology 238, H GRANDY, J. L., DUNLOP, C. I., HODGSON, D. S., CURTIS, C. R. & CHAPMAN, P. L. (1992) Evaluation of the Doppler ultrasonic method of measuring systolic arterial blood pressure in cats. American Journal of Veterinary Research 53, GROSENBAUGH, D. A., MUIR III, W. W. (1998) Accuracy of noninvasive oxyhemoglobin saturation, end-tidal carbon dioxide concentration, and blood pressure monitoring during experimentally induced hypoxemia, hypotension, or hypertension in anesthetized dogs. American Journal of Veterinary Research 59, HABERMAN, C. E., KANG, C. W., MORGAN, J. D. & BROWN, S. A. (6) Evaluation of oscillometric and Doppler ultrasonic methods of indirect blood pressure estimation in conscious dogs. Canadian Journal of Veterinary Research 7, HAMLIN, R. L., KITTLESON, M. D., RICE, D., KNOWLEN, G. & SEYFFERT, R. (1982) Noninvasive measurement of systemic arterial pressure in dogs by automatic sphygmomanometry. American Journal of Veterinary Research 43, MANDIGERS, P. (5) Non invasive blood pressure measurements in dogs and cats. Tijdschrift voor Diergeneeskunde 13, MEURS, K. M., MILLER, M. W. & SLATER, M. R. (1996) Comparison of the indirect oscillometric and direct arterial methods for blood pressure measurements in anesthetized dogs. Journal of the American Animal Hospital Association 32, S+B MED VET GMBH (1998). Available from: (accessed 1 October 9) SAWYER, D. C., BROWN, M. & STRILER, E. L. (1991) Comparison of direct and indirect blood pressure measurement in anesthetized dogs. Laboratory Animal Science 41, SAWYER, D. C., GUIKEMA, A. H. & SIEGEL, E. M. (4) Evaluation of a new oscillometric blood pressure monitor in isoflurane-anesthetized dogs. Veterinary Anaesthesia and Analgesia 31, SCHMELTING, B., NIEHOFF, M., EGNER, B., KORTE, S. H. & WEINBAUER, G. F. (9) High definition oscillometry: a novel technique for non-invasive blood pressure monitoring in the cynomolgus monkey (Macaca fascicularis). Journal of Medical Primatology 38, STEPIEN, R. L., RAPOPORT, G. S., HENIK, R. A., WENHOLZ, L. & THOMAS, C. B. (3) Comparative diagnostic test characteristics of oscillometric and Doppler ultrasonographic methods in the detection of systolic hypertension in dogs. Journal of Veterinary Internal Medicine 17, Journal of Small Animal Practice Vol 51 June 1 1 British Small Animal Veterinary Association