Abstract
Objective—To determine whether veterinary-specific oscillometric blood pressure units yield measurements that are in good agreement with directly measured blood pressures in cats.
Design—Evaluation study.
Animals—21 cats undergoing routine spaying or neutering.
Procedures—A 24-gauge catheter was inserted in a dorsal pedal artery, and systolic, diastolic, and mean arterial pressures were directly measured with a validated pressure measurement system. Values were compared with indirect blood pressure measurements obtained with 3 veterinary-specific oscillometric blood pressure units.
Results—There was poor agreement between indirectly and directly measured blood pressures. For unit 1, bias between indirectly and directly measured values was −14.9 mm Hg (95% limits of agreement [LOA], −52.2 to 22.4 mm Hg), 4.4 mm Hg (95% LOA, −26.0 to 34.8 mm Hg), and −1.3 mm Hg (95% LOA, −26.7 to 24.1 mm Hg) for systolic, diastolic, and mean arterial pressures, respectively. For unit 2, bias was −10.3 mm Hg (95% LOA, −52.9 to 32.2 mm Hg), 13.0 mm Hg (95% LOA, −32.1 to 58.0 mm Hg), and 9.1 mm Hg (95% LOA, −32.9 to 51.2 mm Hg) for systolic, diastolic, and mean arterial pressures, respectively. For unit 3, bias was −13.4 mm Hg (95% LOA, −51.8 to 25.1 mm Hg), 8.0 mm Hg (95% LOA, −25.5 to 41.6 mm Hg), and −3.6 mm Hg (95% LOA, −31.6 to 24.5 mm Hg) for systolic, diastolic, and mean arterial pressures, respectively.
Conclusions and Clinical Relevance—Results suggested that none of the 3 veterinary-specific oscillometric blood pressure units could be recommended for indirect measurement of blood pressure in cats.
The ability to measure blood pressure in cats has taken on increasing importance not only for monitoring anesthetized patients but also for evaluating critical patients and their responses to treatment.1 In addition, as important epidemiological information becomes available, it has become increasingly clear that certain disease conditions and drugs are associated with secondary hypertension.2–7 This is important clinically because hypertension can cause damage to target organs including the eyes, kidneys, heart, and brain.5,8–10
Direct measurement of blood pressure requires inserting a catheter in a suitable artery and measuring arterial pressures with a transducer.8,11,12 However, this can be technically challenging, uncomfortable for patients, and unsuitable for many clinical situations.13 For this reason, blood pressure is frequently measured indirectly, rather than directly, by means of Doppler ultrasonographic or oscillometric methods.
Although Doppler ultrasonographic methods are commonly used for indirect measurement of blood pressure in cats,13 results may not be entirely accurate. In 1 study,12 for instance, blood pressure measurements obtained by means of Doppler ultrasonography were deemed to have low mean error values when compared with directly measured blood pressures; however, agreement between the 2 methods was poor. Results of another study14 indicated that although there was good correlation between directly measured systolic arterial blood pressures and Doppler ultrasonographic measurements, agreement analysis revealed a large negative bias. Nevertheless, the authors of this study suggested that blood pressure measurements obtained by means of Doppler ultrasonography could be used as a reliable predictor of mean arterial pressure. Other authors15 have determined that use of simple correction factors can improve the reliability of systolic blood pressures obtained by means of Doppler ultrasonography. Nevertheless, results obtained by means of Doppler ultrasonography are somewhat operator dependent, and measuring blood pressures with this procedure can be time-consuming. In addition, determination of diastolic arterial blood pressure may be problematic.12
Oscillometric units offer the promise of automated measurement of systolic, diastolic, and mean arterial pressures and heart rate in animals and are technically easier to use than Doppler ultrasonographic methods. Previous studies11,12 have questioned the use of oscillometric devices in cats, but other studies14,16 have shown that oscillometric devices produce results that are in good agreement with directly measured mean and diastolic arterial pressures.
Recently, 3 veterinary-specific oscillometric devicesa–c optimized for use in cats have been released to the market. The purpose of the study reported here was to determine whether systolic, diastolic, and mean arterial pressures obtained with these units were in good agreement with directly measured pressures in cats. Our hypothesis was that there would be good agreement between indirectly and directly measured blood pressures.
Materials and Methods
The study protocol was approved by the Louisiana State University Clinical Protocol Committee. Cats brought to the Louisiana State University Animal Sterilization Assistance Program for routine spaying or neutering were eligible for enrollment in the study. Some of the cats were owned by private individuals; others were owned by charitable organizations. In general, medical histories were incomplete, but all cats were deemed to be healthy on the basis of a complete physical examination. Owners of cats included in the study provided signed informed consent prior to enrollment.
Cats were sedated with midazolam (0.1 mg/kg [0.045 mg/lb], IM) and ketamine (7 mg/kg [3.2 mg/lb], IM), and anesthesia was induced with isoflurane (3%) administered with oxygen by mask. In all cats, an appropriately sized, cuffed endotracheal tube was inserted, and anesthesia was maintained with isoflurane (1%) administered with a Bain circuit. A multifunction monitorc was used to continuously monitor an ECG and oxygen saturation as determined by use of pulse oximetry. Endtidal partial pressure of carbon dioxide was also monitored, and cats were manually ventilated as needed to maintain an end-tidal partial pressure of carbon dioxide between 35 and 45 mm Hg. Cats were placed in lateral recumbency on a hot-water blanket, and body temperature was monitored. A convective air warmer was available to maintain body temperature if needed.
For direct measurement of arterial blood pressures, a 24-gauge catheter was placed in a dorsal pedal artery. The catheter was connected to a continuous multifunction monitorc via a disposable pressure transducer system.d The transducer's accuracy was checked against a mercury manometer. The system was connected to a pressurized (300 mm Hg) bag containing saline (0.9% NaCl) solution, and the transducer was placed at the level of the heart and zeroed to atmospheric pressure. The direct blood pressure monitoring system was periodically flushed to prevent clots and to remove air bubbles that could have changed the damping coefficient of the system. The system was periodically checked for air bubbles, and bubbles were cleared as needed. Direct pressure measurements were checked for stability and consistency and the waveform was analyzed before study recordings commenced.
Three oscillometric unitsa–c were set up and calibrated in accordance with the manufacturers' instructions and were used to obtain indirect arterial blood pressure measurements. Four paired measurements of systolic, diastolic, and mean arterial pressure were obtained with each device, with a 1-minute interval between pairs of measurements. For unit 1,a indirect blood pressure measurements were obtained with a cuff placed on the left antebrachium. An appropriately sized cuff was selected by use of the integrated cuff selection system. For unit 2,b measurements were obtained with a cuff placed on the tail; the tail cuff provided by the manufacturer was used. For unit 3,c measurements were obtained with a cuff placed on the left antebrachium. A cuff with a width that was 30% to 35% of the limb circumference was used.
In general, oscillometric units measure mean arterial pressure and then calculate systolic and diastolic arterial pressures by use of proprietary algorithms.17–19 Because units 1 and 2 displayed cuff pressure as the cuff was deflated, the point at which the cuff pressure matched the directly measured mean arterial pressure was used to record directly measured systolic, diastolic, and mean arterial pressures for comparison with indirectly measured blood pressures. For unit 3, directly measured systolic, diastolic, and mean arterial pressures measured at approximately 50% of the total time needed for cuff deflation were used for comparison with indirectly measured blood pressures.
The dynamic response of the direct blood pressure monitoring system was tested by means of the fast flush test. This test allows the natural frequency and damping coefficient of the entire blood pressure system, from the tip of the needle to the pressure transducer, to be determined.20 Ideally, this would be performed in an actual patient. However, cat heart rates are typically too fast to allow this test to be performed without pharmacological manipulation, and because all cats used in the study were privately owned, this was not possible. Therefore, because the fast flush test measures only the response of the instrumentation and is not influenced by patient factors,20,21 the direct blood pressure monitoring system was tested in a dog undergoing surgery for other purposes. Briefly, a 24-gauge catheter was placed in a dorsal pedal artery of the dog, the catheter was connected to the same pressure transducer and pressure line set used for cats in the study, and 10 fast flush tests were performed. The natural frequency of the direct blood pressure monitoring system was consistently 16.6 Hz. Mean fundamental frequency for patients in the study was 2.5 Hz. Mean amplitude ratio for the 10 fast flush tests was 0.21 (SD, 0.035), which corresponded to a damping coefficient of 0.48.21 On the basis of previously established graphic techniques,21 the allowable damping coefficient for this system was determined to range from 0.30 to 0.88. Thus, the direct blood pressure monitoring system appeared adequate for patients in the study.
Statistical analysis—The Shapiro-Wilk test was used to determine whether body weights of cats included in the study were normally distributed. Agreement between direct and indirect blood pressure measurements was determined by use of the Bland-Altman method.22 Because the sampling strategy involved a repeated-measures approach, the mean for each of the pressure measurements was calculated and used for comparison purposes. Bias was defined as the mean difference between the 2 methods; 95% LOA were calculated as bias ± (1.96 × SD). Because of the potential for underestimating the SD of the differences when using a repeated-measures approach, calculated SDs were corrected as described.23 Agreement was defined as good if the absolute value of the bias was ≤ 15 mm Hg and the value for 1.96 × SD (LOA) was ≤ 15 mm Hg; otherwise, agreement was considered poor. This definition was more generous than the definition adopted by the Association for the Advancement of Medical Instruments, which states that a paired reading shall have a mean difference ≤ 5 mm Hg and an SD < 8 mm Hg.24 All statistical analyses were performed with commercially available statistical programs.e
Results
Twenty-one cats were included in the study, although indirect blood pressure measurements were obtained with unit 3 from only 18 cats. Body weight (mean ± SD, 4.6 ± 1.0 kg [10.1 ± 2.2 lb]) of the cats was normally distributed. There were 7 females and 14 males.
For all 3 oscillometric units, agreement between indirectly and directly measured blood pressures was poor. For unit 1,a bias between indirectly and directly measured values was −14.9 mm Hg (95% LOA, −52.2 to 22.4 mm Hg), 4.4 mm Hg (95% LOA, −26.0 to 34.8 mm Hg), and −1.3 mm Hg (95% LOA, −26.7 to 24.1 mm Hg) for systolic, diastolic, and mean arterial pressures, respectively (Figures 1–3).
Bland-Altman plot of agreement between direct systolic arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 1a) in 21 anesthetized cats. The solid horizontal line represents mean bias; the dotted horizontal lines represent 95% LOA.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Bland-Altman plot of agreement between direct diastolic arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 1a) in 21 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Bland-Altman plot of agreement between direct mean arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 1a) in 21 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
For unit 2,b bias was −10.3 mm Hg (95% LOA, −52.9 to 32.2 mm Hg), 13.0 mm Hg (95% LOA, −32.1 to 58.0 mm Hg), and 9.1 mm Hg (95% LOA, −32.9 to 51.2 mm Hg) for systolic, diastolic, and mean arterial pressures, respectively (Figures 4–6). There was a tendency for this unit to overestimate low blood pressure values and underestimate high diastolic and mean arterial blood pressure values.
Bland-Altman plot of agreement between direct systolic arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 2b) in 21 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Bland-Altman plot of agreement between direct diastolic arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 2b) in 21 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Bland-Altman plot of agreement between direct mean arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 2b) in 21 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
For unit 3, bias was −13.4 mm Hg (95% LOA, −51.8 to 25.1 mm Hg), 8.0 mm Hg (95% LOA, −25.5 to 41.6 mm Hg), and −3.6 mm Hg (95% LOA, −31.6 to 24.5 mm Hg) for systolic, diastolic, and mean arterial pressures, respectively (Figures 7–9).
Bland-Altman plot of agreement between direct systolic arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 3c) in 18 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Bland-Altman plot of agreement between direct diastolic arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 3c) in 18 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Bland-Altman plot of agreement between direct mean arterial blood pressure measurements and indirect measurements obtained with a veterinary-specific oscillometric unit (unit 3c) in 18 anesthetized cats. See Figure 1 for key.
Citation: Journal of the American Veterinary Medical Association 237, 4; 10.2460/javma.237.4.402
Discussion
For all 3 veterinary-specific oscillometric units examined in the present study, agreement between indirectly and directly measured blood pressures was poor. These findings suggested that none of the units could be recommended for indirect measurement of blood pressure in cats.
Oscillometric units work by inflating a cuff around a limb so as to impede arterial blood flow and then monitoring arterial oscillations as the cuff is slowly deflated. Mean arterial pressure is estimated by determining the peak amplitude of arterial oscillations, and proprietary algorithms are then used to calculate systolic and diastolic blood pressures.17–19 If these units had produced mean arterial pressures that were in good agreement with directly measured mean arterial blood pressure, then poor agreement between indirectly and directly measured systolic or diastolic pressures could have been a result of faulty algorithms. However, indirectly measured mean arterial pressures agreed poorly with directly measured values, suggesting that all 3 units had difficulty accurately determining this key parameter. Visual inspection of the Bland-Altman plots suggested that unit 3 came closer to approximating this value than the other units.
Probably the most important user-specific factors affecting the performance of oscillometric blood pressure units are selection of cuff size and placement of the cuff. Unit 1 in the present study had an integrated selection system that was used to pick the best cuff for each patient. This unit also had a specific setting for pressure measurements obtained when the cuff was placed on the antebrachium, and this setting was selected for all pressure measurements. Unit 2 came with a tail cuff that was specifically designed for use in cats, and this cuff was used for all measurements in the present study. Unit 3 required that the user select the correct cuff size, and in keeping with a previous recommendation,12 a cuff with a width that was 30% to 35% of the limb circumference was used. One possible explanation for the lack of agreement between indirect and direct blood pressure measurements in the present study was that the cuffs we used were an inappropriate size. Improper cuff size has been shown to adversely affect blood pressure measurement in cats; however, these errors have been shown to be consistent.12 That is, use of cuffs that are too wide results in pressure measurements that are consistently too low, whereas use of cuffs that are too narrow results in pressure measurements that are consistently too high. In the present study, however, indirect blood pressure measurements were not consistently higher or lower than directly measured values. Thus, improper cuff size was not a likely cause of the disagreement.
A possible shortcoming of the present study was that validation of the direct blood pressure monitoring system with the fast flush test was performed in a dog and not in any of the cats used in the study. Ideally, the direct blood pressure monitoring system should be validated before use in each patient. The fast flush test allows the natural frequency and damping coefficient of the entire invasive blood pressure system to be calculated. In brief, the transducer flush valve is opened briefly and then abruptly closed, and the oscillations resulting from the closing of the valve are used to calculate natural frequency and damping coefficient. However, heart rates of cats used in the present study were too fast for oscillations between pulses to be measured; therefore, a dog was used instead. For each patient in the present study, a new catheter, pressure transducer, and pressure line set were used. Although all of these instruments were the same model and met standards established by the Association for the Advancement of Medical Instruments, there could, in theory, have been small differences between sets. Nevertheless, the natural frequency of the transducer set provided such a wide range of allowable damping coefficients that it seems unlikely that differences between sets would have had a substantial impact on our results. In addition, we repeated the fast flush test in 2 different dogs and received identical results, and the authors examined pressure waveforms for cats used in the study and saw no evidence of over or under damping.
Previous studies12,16 used pharmacological manipulation to ensure that recorded blood pressures were distributed in low, normal, and high ranges. In the present study, this was not possible because all cats were privately owned clinical patients. Nevertheless, a wide range of blood pressures were obtained, including values in the normotensive and hypotensive ranges. Because none of the 3 units were able to produce measurements that were in good agreement with directly measured blood pressure in these ranges, there seemed to be little benefit of further exploring agreement for pressures in the hypertensive range.
ABBREVIATION
| LOA | Limits of agreement |
PetMap, Ramsey Medical, Tampa, Fla.
VET HDO, Vetline LLC, Saint Kitts and Nevis, West Indies.
Cardell Max-1, Sharn Veterinary Inc, Tampa, Fla.
DTX Plus, Becton-Dickinson, Sandy, Utah.
Graphpad Prism, version 5.0a for Macintosh, Graphpad Software Inc, San Diego, Calf.
References
- 1.↑
Rozanski EA, Rush JE. Monitoring critical care patients. In: Small animal emergency and critical care medicine. London: Manson Publishing Ltd, 2007;213–219.
- 2.
Kobayashi DL, Peterson ME, Graves TK, et al. Hypertension in cats with chronic renal failure or hyperthyroidism. J Vet Intern Med 1990;4:58–62.
- 3.
Stiles J, Polzin D, Bistner SL. The prevalence of retinopathy in cats with systemic hypertension and chronic renal failure or hypertension. J Am Anim Hosp Assoc 1994;30:564–572.
- 4.
Syme HM, Barber PJ, Markwell PJ, et al. Prevalence of systolic hypertension in cats with chronic renal failure at initial evaluation. J Am Vet Med Assoc 2002;220:1799–1804.
- 5.
Bartges JW, Willis AM, Polzin DJ. Hypertension and renal disease. Vet Clin North Am Small Anim Pract 1996;26:1331–1345.
- 6.
Panciera DL. Cardiovascular complications of thyroid disease. In: Bonagura JD, ed. Kirk's current veterinary therapy XII: small animal practice. 13th ed. Philadelphia: WB Saunders Co, 2000;716–719.
- 7.
Cowgill LD, James KM, Levy JK, et al. Use of recombinant human erythropoietin for management of anemia in dogs and cats with renal failure. J Am Vet Med Assoc 1998;212:521–528.
- 8.
Brown SA, Henik RA. Diagnosis and treatment of systemic hypertension. Vet Clin North Am Small Anim Pract 1998;28:1481–1494.
- 9.
Chetboul V, Lefebvre HP, Pinhas C, et al. Spontaneous feline hypertension: clinical and echocardiographic abnormalities and survival rate. J Vet Intern Med 2003;17:89–95.
- 10.
Litman MP. Spontaneous systemic hypertension in 24 cats. J Vet Med Intern Med 1994;8:79–86.
- 11.
Branson KR, Wagner-Mann CC, Mann FA. Evaluation of an oscillometric blood pressure monitor on anesthetized cats and the effect of cuff placement and fur on accuracy. Vet Surg 1997;26:347–353.
- 12.↑
Binns SH, Sisson DD, Buoscio DA, et al. Doppler ultrasonographic, oscillometric sphygmomanometric, and photoplethysmographic techniques for noninvasive blood pressure measurement in anesthetized cats. J Vet Intern Med 1995;9:405–414.
- 13.↑
Acierno MJ, Labato MA. Hypertension in renal disease: diagnosis and treatment. Clin Tech Small Anim Pract 2005;20:23–30.
- 14.↑
Caulkett NA, Cantwell SL, Houston DM. A comparison of indirect blood pressure monitoring techniques in the anesthetized cat. Vet Surg 1998;27:370–377.
- 15.↑
Grandy JL, Dunlop CI, Hodgson DS, et al. Evaluation of the Doppler ultrasonic method of measuring systolic arterial blood pressure in cats. Am J Vet Res 1992;53:1166–1169.
- 16.
Pedersen KM, Butler MA, Ersboll AK, et al. Evaluation of an oscillometric blood pressure monitor for use in anesthetized cats. J Am Vet Med Assoc 2002;221:646–650.
- 17.
Ramsey M III. Noninvasive automatic determination of mean arterial pressure. Med Biol Eng Comput 1979;17:11–18.
- 18.
Mauck GW, Smith CR, Geddes LA, et al. The meaning of the point of maximum oscillations in cuff pressure in the indirect measurement of blood pressure—part ii. J Biomech Eng 1980;102:28–33.
- 19.
Lee TK, Westenskow DR. Comparison of blood pressure measured by osscilimetry from the supraorbital artery and invasively from the radial artery. J Clin Monit Comput 1998;14:113–117.
- 20.↑
Kleinman B, Powell S, Kumar P, et al. The fast flush test measures the dynamic response of the entire blood pressure monitoring system. Anesthesiology 1992;77:1215–1220.
- 21.↑
Gardner RM. Direct blood pressure measurement—dynamic response requirements. Anesthesiology 1981;54:227–236.
- 22.↑
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–310.
- 23.↑
Bland JM, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat 2007;17:571–582.
- 24.↑
ANSI/AAMI SP10. 2002: manual, electronic, or automated sphygmomanometers. Arlington, Va: Association for the Advancement of Medical Instruments, 2008.
