Agreement Between Doppler and Invasive Blood Pressure Monitoring in Anesthetized Dogs Weighing <5 kg
The objectives of this study were to determine if Doppler (DOP) blood pressure measurements more closely estimate either invasive systolic or invasive mean arterial blood pressures (ISAP or IMAP, respectively) in small dogs under general anesthesia and to assess the ability of DOP to detect anesthesia-related hypotension in small dogs. Blood pressure measurements (n = 203) were obtained from 10 client-owned dogs. DOP, ISAP, and IMAP were recorded simultaneously, and the data were categorized into two groups: hypotensive (ISAP <90 mm Hg) and normotensive (ISAP ≥90 mm Hg and ≤160 mm Hg). DOP overestimated ISAP and IMAP in both the normotensive and hypotensive groups. The DOP was highly specific (97%) but poorly sensitive (56%) for detecting hypotension. The smallest bias was achieved when using DOP as an estimate of systolic arterial blood pressure in both normotensive and hypotensive dogs, suggesting that DOP measures systolic arterial blood pressure in dogs <5 kg. For dogs with hypotension, DOP met all of the performance criteria for noninvasive blood pressure monitors recommended by the American College of Veterinary Internal Medicine. DOP is an acceptably accurate and highly specific means of detecting hypotension in small dogs under general anesthesia.
Introduction
Vigilant patient monitoring and early detection of complications can reduce the risk of anesthetic related morbidity and mortality. The American College of Veterinary Anesthesia and Analgesia recommends blood pressure monitoring for all small animals under general anesthesia as a means of ensuring adequate circulatory function.1 Hypotension is the most common complication in dogs under general anesthesia, with the incidence of hypotension reportedly ranging from 7 to 38% depending on health status.2,3 It is generally recommended that mean arterial blood pressure (MAP) be maintained >60 mm Hg because either untreated severe or prolonged hypotension can result in cardiac arrest, blindness or other forms of central nervous system dysfunction, and renal compromise or failure.4,5 Accurate blood pressure monitoring is important for detecting hypotension and is critical for assessing the patient's response to treatment and determining whether appropriate blood pressure has been achieved and maintained.
There are several options available when selecting a blood pressure monitor. The gold standard is invasive blood pressure (IBP) monitoring via insertion of an arterial catheter and use of a pressure transducer. The values obtained with IBP monitoring vary depending on the site of arterial catheterization, with the systolic pressure increasing and the diastolic pressure decreasing as the site of measurement moves farther downstream form the aorta.6 IBP monitoring can be expensive if sophisticated multiparameter monitors are used. Arterial catheterization can be challenging (especially in small patients) and carries the risk of thrombosis, hemorrhage, inflammation, or infection; therefore, this method may not be practical in all situations.5 A more accessible option for many practitioners is noninvasive blood pressure (NIBP) monitoring. NIBP monitoring is economical, portable, and requires less technical skill. NIBP monitor options include oscillometric methods and Doppler ultrasonic flow detector (DOP). The size of the patient can affect oscillometric measurements, making them less accurate.7 Therefore, it has been proposed that DOP be used for patients <10 kg.5
Most of the investigations comparing DOP to IBP have used medium- to large-breed dogs and found DOP to be an estimate of invasive systolic arterial pressure (ISAP).7–12 When DOP was compared to IBP in anesthetized cats, DOP was found to be an accurate estimate of invasive mean arterial pressure (IMAP).13 Those investigators concluded that the results may have been due to patient size. Many anesthetists have speculated that the same correlation may hold true for small dogs with a body weight similar to cats; however, a direct comparison of DOP to IBP in small dogs has not been reported. For purposes of intervention during hypotensive events, it is important to know whether DOP provides a number closer to MAP or systolic arterial blood pressure (SAP) in small dogs. For example, a MAP of 75 mm Hg is acceptable, whereas a SAP of 75 mm Hg (for most anesthetists) would result in an intervention. The objectives of this clinical investigation were to (1) determine if DOP is a better estimate of SAP or MAP in anesthetized dogs weighing <5 kg; and (2) assess DOP accuracy to detect spontaneous hypotension in this population.
Materials and Methods
Animals
This clinical study was performed at the University of Minnesota Veterinary Medical Center. Canine patients undergoing general anesthesia for diagnostic medical imaging and necessary surgical treatment were considered for enrollment. Patients were included in the study if they had an arterial catheter already in place as part of their monitoring plan for anesthesia and a body weight of approximately ≤5 kg. All dogs were anesthetized under the supervision of a board-certified anesthesiologist (either the American or European College of Veterinary Anesthesia and Analgesia) and received a high standard of care that met or exceeded the regulations and standards for animal care and use set by the National Institutes of Health.14
Procedure
Each patient was premedicated, an IV catheter was placed, and anesthesia was induced according to the plan approved by the attending anesthesiologist. Each dog was orotracheally intubated and general anesthesia was maintained with either isofluranea or sevofluraneb vaporized in 100% O2. While under general anesthesia, each patient received IV fluids at a rate of 5–10 mL/kg/hr according to the plan approved by the anesthesiologist. Body temperature, heart rate, and respiration rate were continuously monitored and recorded q 5 min. Each animal was instrumented with a pulse oximeter and electrocardiogram for continuous monitoring of arterial hemoglobin saturation, heart rate, and rhythm. Over-the-needle cathetersc (22- or 24-gauge) were placed in either the right or left dorsal pedal artery by a licensed veterinarian or certified veterinary technician. The arterial catheter was connected to a previously calibrated disposable pressure transducerd by low-compliance pressure tubing filled with heparinized saline from a pressurized bag (inflated to 300 mm Hg), and heparinized saline was administered at a rate of 3 mL/hr through the catheter to prevent clot formation. Patients were placed in either lateral or dorsal recumbency according to the procedure to be performed. The pressure transducer was then positioned at approximately the level of the right atrium. The transducer was connected to a blood pressure monitore and zeroed as instructed by the manufacturer. Blood pressure was continuously monitored, and the waveform assessed for dampening. ISAP and IMAP were recorded simultaneously and at the same time points as DOP was recorded.
For DOPf blood pressure measurement, either a forelimb or the hind limb contralateral to the arterial catheter was fitted with a blood pressure cuff proximal to the carpus or tarsus. Forelimb versus hind limb cuff position was determined by accessibility during the given procedure to be performed. A cuff width of approximately 40% the limb circumference was used. The hair was clipped just proximal to the palmar metacarpal pad or the plantar metatarsal pad. Ultrasonic gelg was applied to the concave surface of the piezoelectric crystal, and the crystal was then applied to the previously shaved area and secured in place with adhesive tape. The cuff was inflated then gradually deflated, and the pressure was recorded when arterial flow was first audibly detected. Immediately after the DOP measurement was obtained the displayed IBP measurements were recorded, thus giving a pair of blood pressure measurements for each time point. This procedure was repeated q 5 min by an experienced anesthetist.
Previous investigations in anesthetized dogs found that the accuracy of indirect blood pressure monitoring devices varies with hypotension versus normotension.7,9 Thus, for the purpose of this study, normotension was defined as ISAP ≥90 mm Hg and ≤160 mm Hg. Hypotension was defined as ISAP <90 mm Hg. The ranges of blood pressures observed in this study were spontaneously occurring and were not intentionally induced by the investigators. Every effort was made to ensure that each patient's IMAP remained at least 60 mm Hg during the anesthetic period. Hypotension was addressed by decreasing the level of inhalant, administering IV fluid boluses, and administering dobutamineh and/or dopaminei as directed by the attending anesthesiologist.
Statistical Analysis
Data were categorized and analyzed separately as either hypotensive or normotensive. Agreement between DOP and IBP measurements was assessed using the Bland-Altman method for repeated measures on an individual when the true value is changing.15,16 When comparing DOP and ISAP, the mean bias and standard deviation of the difference were calculated for each set of measurements by subtracting the DOP pressure from the simultaneous ISAP. A positive bias would indicate that DOP underestimates ISAP, and a negative bias would indicate that DOP overestimates the corresponding ISAP. DOP and IMAP were compared in a similar fashion with bias calculated by subtracting DOP from IMAP. Additionally, the ability of DOP to detect hypotension was assessed using a standard 2 × 2 table with a binary outcome to calculate sensitivity, specificity, negative predictive value, and positive predictive value. All analyses were performed using a commercially available softwarej.
Results
Over the 8 mo investigation period, 10 client-owned dogs were enrolled. The group included 5 males and 5 females between the ages of 8 mo and 12 yr with a mean body weight of 3.85 ± 0.9 kg. The hypotensive measurements included 103 paired blood pressure readings. The normotensive measurements had 100 paired blood pressure readings. Table 1 shows the calculated bias for each group of measurements when comparing DOP to both ISAP and IMAP. DOP overestimated SAP in both the normotensive and hypotensive measurement groups. When comparing DOP to MAP, DOP overestimated MAP in both the normotensive and hypotensive groups. The smallest bias (i.e., best accuracy) was achieved when using DOP as an estimate of ISAP versus IMAP in both normotensive (−17.3 ± 14.9 mm Hg) and hypotensive (−5.7 ± 12.3 mm Hg) dogs. Bland-Altman plots, shown in Figures 1 and 2, illustrate the smaller bias observed when using DOP as an estimate of direct SAP in both groups. For both groups of data, DOP tended to overestimate IBP to a greater extent as the IBP increased. For hypotensive measurements, 61.2% of DOP measurements were within 10 mm Hg of ISAP, whereas 60% of DOP measurements were within 20 mm Hg of ISAP for normotensive measurements (Table 1). Using DOP as an estimate of SAP, DOP was 97% specific for detecting hypotension with a positive predictive value of 95% and 56% sensitive with a negative predictive value of 68%.
Citation: Journal of the American Animal Hospital Association 51, 5; 10.5326/JAAHA-MS-6163
Citation: Journal of the American Animal Hospital Association 51, 5; 10.5326/JAAHA-MS-6163
Discussion
For both hypotensive and normotensive small dogs, the best accuracy was observed when using DOP as an estimate of ISAP. This finding suggests that blood pressures measured using DOP are most representative of SAP in small dogs. When Caulkett et al. (1998) reported DOP to be an accurate estimation of MAP in cats, the mean body weight of animals in that study was 3.5 kg (±0.8 kg), which is comparable to the group of dogs used in this investigation with a mean bodyweight of 3.85 kg (±0.9 kg).13 The similar body weights of the two groups and the different findings in this study compared to Caulkett et al. (1998) suggest that the DOP's accurate prediction of MAP in the cat is due to species differences rather than patient size; however, small differences in methodology cannot be completely ruled out. In a previous study comparing DOP to ISAP and IMAP in normotensive and hypotensive hospitalized dogs weighing 26.2 kg (±11 kg), DOP was a more accurate predictor of ISAP rather than IMAP, which agrees with the findings of the current study despite the large difference in mean patient body weight.8
In the present investigation, DOP overestimated SAP in hypotensive dogs by a mean of 5.7 mm Hg (±12.3 mm Hg), which is similar to the results of previous investigations despite a difference in average patient body weight.8,10 In the current investigation, DOP measurements of SAP were more accurate under conditions of hypotension compared to conditions of normotension. Previous investigations have also found a greater magnitude of bias at normal blood pressures.9,10 The American College of Veterinary Internal Medicine's (ACVIM's) recommendations for evaluating the performance of NIBP monitors state that the absolute bias should be ≤10 mm Hg, that 50% of all measurements lie within 10 mm Hg of the reference method, and that 80% of all measurements lie within 20 mm Hg of the reference method.17 In this investigation, DOP met all of those criteria for small dogs with hypotension but failed to meet any of those criteria for normotensive small dogs. When Dyson et al. (2007) compared DOP to the invasive reference method in large dogs, they were able to meet all the ACVIM's criteria in both hypotensive and normotensive states.9 Garofalo et al. (2012) found that the ability of DOP to meet the criteria depended on cuff placement, with the cuff being proximal to the tarsus producing the most accurate results.10
From the investigation reported herein, looking strictly at the performance of DOP in detecting hypotension in small dogs under general anesthesia, its high specificity and high positive predictive value make DOP a reliable blood pressure monitoring tool for ruling in hypotension. However, hypotension may also be missed by DOP given its low sensitivity and low negative predictive value. The ideal blood pressure monitor should have high sensitivity because veterinarians are typically monitoring in order to rule out hypotension, especially in light of the relative frequency of hypotension in anesthetized dogs. Thus, IBP monitoring would still be indicated in cases where one may expect significant hemorrhage or difficult to manage hypotension (i.e., septic patients).
This study has several limitations. Only 10 dogs were enrolled during the study period for a total of 203 paired blood pressure readings. There was more than one anesthetist performing the DOP measurements, and slight differences in technique may have introduced bias. Additionally the anesthetist was not blinded to the IBP readings; therefore, with the IBP readings being continuously displayed, the anesthetist's knowledge of the IBP readings could have influenced the next DOP measurement and resulted in DOP measurements that were falsely close to the IBP measurements. Ideally, the DOP and IBP measurements should have been performed at the exact same moment by two separate individuals without any knowledge of what values the other was recording. The dogs were not all placed in the same position at the time of blood pressure measurement. Instead, they were positioned according to the requirements of the procedure being performed. The blood pressure cuffs were placed on either the pelvic limb or thoracic limb depending on procedure instead of being placed in the same location. The patient position required by the procedure also limited the investigators' abilities to keep the DOP cuff at the level of the heart, and any dog with a DOP cuff positioned at a level below the heart would have contributed to the DOP overestimating the ISAP due to hydrostatic differences. However, given the small size of the dogs in this study and the need for the cuff to be 10 cm below the heart to overestimate the blood pressure by 7mm Hg, the bias introduced by cuff position is likely small.18 While these limitations may have introduced more variability into the data, the results of the investigation are relevant to the clinical setting, where variations in patient positioning, cuff placement, and operator are to be expected.
Conclusion
The results of this study indicate that DOP measurements are a close estimate of SAP in small dogs under general anesthesia. Though DOP did not fulfill any of the ACVIM's criteria for comparison to IBP at normal SAP, it fulfilled all criteria to be used as an estimate of SAP in hypotensive dogs. DOP is an acceptably accurate and highly specific means of detecting hypotension (SAP <90 mm Hg) in small dogs under general anesthesia; therefore, any hypotension indicated by DOP should be treated accordingly. However, hypotension may also be missed by DOP given its low sensitivity, and direct blood pressure monitoring is still indicated in circumstances where significant hypotension is likely or especially detrimental to patient outcome.
Bland-Altman plots of DOP versus IBP during normotension. A: DOP versus IMAP. B: DOP versus ISAP. For each pair of simultaneous measurements, the bias is plotted against the average of DOP and the corresponding IBP technique. A positive bias indicates that DOP underestimates the corresponding IBP, and a negative bias indicates that DOP overestimates the corresponding IBP. The solid lines represent mean bias, dashed lines represent the limits of agreement (means bias ± 1.96 standard deviations). Normotension was defined as an ISAP ≥90 mm Hg and ≤160 mm Hg. DOP, Doppler; IBP, invasive blood pressure; IMAP, invasive mean arterial blood pressure; ISAP, invasive systolic arterial blood pressure.
Bland-Altman plots of DOP versus IBP during hypotension. A: DOP versus IMAP. B: DOP versus ISAP. For each pair of simultaneous measurements, the bias is plotted against the average of DOP and the corresponding IBP. A positive bias indicates that DOP underestimates the corresponding IBP, and a negative bias indicates that DOP overestimates the corresponding IBP. Solid lines represent mean bias, and dashed lines represent the limits of agreement (means bias ± 1.96 standard deviations). Hypotension was defined as an ISAP <90 mm Hg. DOP, Doppler; IBP, invasive blood pressure; IMAP, invasive mean arterial blood pressure; ISAP, invasive systolic arterial blood pressure.
Contributor Notes
