Evaluation of the Accuracy and Precision of the GE Dash 4000 Oscillometric Monitor for Blood Pressure Measurement in Anesthetized Female Dogs

Abstract

Background: Indirect measurement of arterial blood pressure, such as the oscillometric method, is the most commonly used in clinical practice of dogs and cats. This method measures blood pressure values that are estimates of direct (invasive) arterial blood pressure values. Oscillometric devices are easy to use even for non-experienced personnel. However, there is considerable variation in accuracy and precision of blood pressure values measured by different oscillometric monitors. The present study aimed to determine the accuracy and precision of the GE Dash 4000 oscillometric monitor for arterial blood pressure measurement in anesthetized female dogs.

Materials, Methods & Results: Sixteen healthy adult female dogs received 0.3 mg/kg morphine as premedication and were anesthetized with propofol and isoflurane. A 22-gauge catheter was introduced into the dorsal pedal artery and connected to a rigid tubular system and a pressure transducer filled with heparinized solution to allow direct (invasive) measurement of systolic (SAP), mean (MAP) and diastolic arterial pressure (DAP). A blood pressure cuff was positioned proximal to the carpus and connected to the oscillometric device (GE-DASH 4000 monitor) in order to obtain indirect measurements of SAP, MAP and DAP. Cuff width was 40% of limb circumference. During anesthesia, invasive arterial blood pressure values were measured and recorded simultaneously with the oscillometric method. The Bland Altman method was used to evaluate agreement between the methods by calculating the bias (invasive - oscillometric) and limits of agreement. Percentages of differences between the methods with an error ≤ 10 mmHg and ≤ 20 mmHg were calculated. Results were compared with the criteria from the American College of Veterinary Internal Medicine (ACVIM) for validation of noninvasive blood pressure methods. Weight and age of dogs were 7.6 ± 2.2 kg and 20 ± 17 months, respectively. A total of 195 pairs of measurements were obtained from 16 animals. Of these pairs, 146 were classified as normotension (SAP: 90 to 140 mmHg), 28 as hypertension (SAP > 140 mmHg) and 21 as hypotension (SAP < 90 mmHg). Bias values ± SD (95% limits of agreement) were: SAP, 5.0 ± 16.5 mmHg (-27.3 to 37.4 mmHg); MAP, -3.4 ± 14.3 mmHg (-31.4 to 24.6 mmHg); and DAP, 4.2 ± 11.8 mmHg (-18.9 to 27.4 mmHg). According to the ACVIM criteria, maximum values accepted for bias (± SD) are 10 ± 15 mmHg. Percentages of differences ≤ 10 mmHg and ≤ 20 mmHg were: SAP, 41% and 80%; MAP, 54% and 84%; and DAP, 64% and 91%. ACVIM recommendations are ≥ 50% for errors within 10 mmHg and ≥ 80% for errors within 20 mmHg.

Discussion: MAP and DAP values obtained by the GE-DASH 4000 monitor matched the ACVIM criteria for validation of noninvasive methods. Conversely, SAP values did not meet all the criteria, and were not considered reliable. Limitations of the study include: a) most dogs were of low weight; b) the ACVIM criteria refer to SAP measurements, but in the present study, the same criteria were applied to MAP and DAP measurements; c) the majority of observations were obtained during normotension. We conclude that MAP and DAP measurements obtained by the GE Dash 4000 monitor met the ACVIM criteria for validation of noninvasive blood pressure monitors. Therefore, this monitor was considered to have adequate accuracy and precision for MAP and DAP measurements in anesthetized normotensive dogs. Under the conditions of this study, SAP measurements obtained by this monitor were not reliable.