History

A 9-year-old 21.3-kg (46.9-lb) spayed female English Bulldog was presented to the University of Florida College of Veterinary Medicine Oncology Service for surgical evaluation after a right adrenal mass was incidentally identified by the referring veterinarian during abdominal ultrasonography performed for evaluation of recurrent urinary tract infections. There were no other changes in the dog's health or behavior reported by the owner. Physical examination revealed moderate pelvic limb ataxia, a recessed vulva, and atopic skin disease. Results of a CBC and serum biochemical analyses were unremarkable, except for mildly high serum alkaline phosphatase activity (162 U/L; reference range, 7 to 116 U/L). Results of a low-dose dexamethasone suppression test were within reference limits, and findings on urine metanephrine testing were inconclusive. Computed tomography revealed a well-defined, soft tissue–attenuating, heterogeneously contrast-enhancing mass (approx 3.5 × 3.7 × 4.0 cm) in the area of the right adrenal gland that caused marked rightward compression of the caudal vena cava (CVC). Exploratory laparotomy and surgical excision of the adrenal mass were scheduled.

The night before surgery, a 20-gauge, 1-inch catheter was aseptically placed in the right cephalic vein, and food but not water was withheld from the dog 8 hours before the surgical procedure. The day of surgery, maropitant (1 mg/kg [0.45 mg/lb], IV) and methadone (0.2 mg/kg [0.1 mg/lb], IV) were administered, and the dog was preoxygenated by face mask for 5 minutes before anesthetic induction. General anesthesia was induced with propofol (2.7 mg/kg [1.2 mg/lb], IV) and diazepam (0.2 mg/kg, IV). Intubation was achieved with a 5.5-mm-internal-diameter cuffed endotracheal tube despite difficulties owing to a severely elongated soft palate and hypoplastic trachea. General anesthesia was maintained with 1% to 1.5% sevoflurane (vaporizer setting) in oxygen (1 L/min), and constant rate infusions of fentanyl (0.1 to 0.2 μg/kg/min [0.05 to 0.09 μg/lb/min], IV), dexmedetomidine (1 μg/kg/h, IV), magnesium sulfate (15 mg/kg/h [6.8 mg/lb/h], IV), and metoclopramide (1 mg/kg/h, IV) were started. Lactated Ringer solution (5 mL/kg/h [2.3 mL/lb/h], IV) was administered throughout the procedure. The dog was instrumented for lead II ECG, sidestream airway gas monitoring, pulse oximetry, esophageal temperature monitoring, and airway spirometry with a multiparametric monitor,^a and direct blood pressure monitoring was performed with a 20-gauge, 1-inch arterial catheter placed in the right dorsal pedal artery. A transversus abdominis plane block was performed with bupivacaine (2.5 mg/kg [1.1 mg/lb]), and the dog was positioned in dorsal recumbency. Intermittent positive-pressure ventilation^a was started at a tidal volume of 12 mL/kg (5.5 mL/lb), respiratory rate of 8 breaths/min, inspiratory-to-expiratory ratio of 1:2.5, inspiratory pause of 25%, and peak inspiratory pressure of 12 cm H₂O.

During the laparotomy (ventral midline approach), the dog's heart rate (HR) varied between 60 and 75 beats/min (reference range, 60 to 120 beats/min), the mean arterial pressure (MAP) remained steady at approximately 75 mm Hg (reference range, 60 to 100 mm Hg), and other vital signs were unremarkable. Hematologic analyses performed on a sample of arterial blood 30 minutes after the start of surgery revealed that the dog's partial pressure of oxygen was 487 mm Hg (reference range, 450 to 500 mm Hg when receiving 100% oxygen), partial pressure of carbon dioxide was 50 mm Hg (reference range, 34 to 40 mm Hg), and blood lactate concentration was 1.75 mmol/L (reference range, 0.6 to 2.9 mmol/L). The right adrenal mass was identified, and dissection of the adhesions around it was performed without anesthetic or surgical complications. The adrenal mass appeared to have invaded through the wall of the CVC, despite the initial diagnosis of tumoral compression of the vessel on the basis of preoperative diagnostic imaging.

A decision was made to partially occlude (approx 40%) the CVC at the level of the mass to facilitate complete resection (Figure 1). During adrenal mass manipulation, the HR on ECG increased to 160 beats/min; however, the monitor^a displayed a 2:1 discordance between the ECG-derived HR and pulse rate calculated from the arterial pressure waveform (80 beats/min; Figure 2). Approximately 150 to 200 mL of blood loss occurred when the CVC was incised. The dog's MAP steadily decreased to 48 mm Hg, and fluid resuscitation was initiated with boluses of lactated Ringer solution (10 mL/kg [4.5 mL/lb]) and hypertonic saline (7.2% NaCl) solution (4 mL/kg [1.8 mL/lb]). The dog's hypotension improved (MAP, 60 mm Hg), and the discordance between the ECG-derived HR and the pulse rate evident from the arterial pressure waveform began to resolve, matching again at an HR of approximately 120 beats/min. Arterial blood gas analyses^b were repeated at this point to evaluate the dog for tissue hypoperfusion following partial occlusion of the CVC. Results indicated that the dog's partial pressure of oxygen was 382 mm Hg, partial pressure of carbon dioxide was 58 mm Hg, and blood lactate concentration was 2.6 mmol/L. As the tumor and the CVC were manipulated over the following 30 minutes, the dog's HR varied between 65 and 160 beats/min. Discordance between the ECG-derived HR and the pulse rate evident from the arterial pressure waveform was again noticed when the HR exceeded 140 beats/min, but the discordance resolved once the vascular clamp was removed from the CVC. Surgery was completed without further complications, and the dog recovered uneventfully in the intensive care unit. Histologic examination of the removed mass confirmed a diagnosis of pheochromocytoma.

View Full Size

Intraoperative images of a right adrenal mass (star) in a 9-year-old 21.3-kg (46.9-lb) spayed female English Bulldog. The dog's head is toward the top in each image. A—The mass (star), retracted with silk suture and a curved hemostat, appears to have invaded the intima of the caudal vena cava (CVC; black arrow). B—Later in surgery, the mass (star) is positioned between the surgeon's fingers, and the CVC (black arrow) is partially occluded with an atraumatic vascular clamp (white arrow).

Citation: Journal of the American Veterinary Medical Association 257, 7; 10.2460/javma.257.7.710

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Intraoperative image of the patient monitor^a used to monitor vital signs in the dog described in Figure 1 depicting a mismatched heart rate (162 beats/min) and pulse rate (80/min) detected by ECG and arterial blood pressure, respectively. On the monitor's display, the ECG tracing (green) is at the top of the screen, measurements of heart rate (black numbers in a yellow box) and pulse rate (red numbers) are listed to the right of the ECG tracing, and the arterial pressure waveform tracing (red) is near the center of the screen.

Citation: Journal of the American Veterinary Medical Association 257, 7; 10.2460/javma.257.7.710

• Download Figure
• Download figure as PowerPoint slide

Question

What caused the apparent mismatch between the ECG-derived HR and the monitor-calculated pulse rate during periods of tachycardia in this patient?

Answer

Closer inspection of the arterial pressure waveform revealed alternating low- and high-amplitude pulse waves, a phenomenon known as pulsus alternans. In calculating the pulse rate, the monitor only counted the high-amplitude waves. The true pulse rate when the low-amplitude waves were included matched the ECG-derived HR.

Discussion

Pulsus alternans describes an arterial pulse waveform that alternates between 2 amplitudes with each heartbeat. Clinically, this results in alternating strong and weak pulses. If ventricular systolic function is sufficiently impaired, isovolumetric contraction may not generate sufficient pressure to open the aortic valve. This results in true pulse deficits, termed total pulsus alternans.¹ Unlike pulsus bigeminus, in which alternating pulse strength is the result of ventricular bigeminy, pulsus alternans can be observed with a sinus rhythm.

Two primary factors have been proposed in the etiology of pulsus alternans: hemodynamic and inotropic alterations.² Either can result in oscillating end-diastolic ventricular volume, which directly affects contractility owing to the Frank-Starling mechanism. Myocardial contractility depends on sarcomere length at the beginning of systole, which in turn depends on preload. With systolic dysfunction, ejection fraction is low. The relatively high end-systolic left ventricular blood volume left behind from the reduced ejection fraction contributes to the preload for the next cardiac cycle, stretching myocardial sarcomeres and increasing their force of contraction. However, oscillating end-diastolic volume can also result from impaired ventricular filling, as in the case of cardiac tamponade and other forms of severe obstructive shock.

Despite being described as far back as 1872,³ modern reports of pulsus alternans with underlying causes other than pericardial effusion and subsequent cardiac tamponade are sparse in both the human and veterinary literature. In a human study,¹ 5 of 11 patients subjected to balloon occlusion of the inferior vena cava developed sustained pulsus alternans. In human medicine, a single case report⁴ attributes pulsus alternans to ineffective diastolic filling because of supraventricular tachycardia in a man, and the supraventricular tachycardia and pulsus alternans resolved with radiofrequency catheter ablation of the arrhythmogenic focus.⁴ In dogs, pulsus alternans has been described during experimentally induced tachycardia (200 to 220 beats/min) and CVC occlusion.⁵ It was attributed to oscillating left ventricular end-diastolic volume and impaired myocardial relaxation, possibly because of altered intracellular calcium reuptake in cardiac myocytes.⁵ In our experience, gastric dilatation-volvulus can precipitate pulsus alternans, likely because of a mechanism similar to obstructive shock. Pulsus alternans has also been reported in dogs given a combination of a vasopressor and an antimuscarinic drug, presumably as a result of tachycardia in the face of severely increased afterload.^6,7 A 2007 case series⁸ reported 10 Cocker Spaniels with pulsus alternans as a result of congestive heart failure from dilated cardiomyopathy. However, pulsus alternans has not been previously reported as a reversible intraoperative complication in clinical patients, as occurred in the dog of the present report.

Variations in pulse intensity may result in erroneous pulse rate calculation by anesthesia monitors. The exact method for calculating pulse rate varies by manufacturer and is the result of proprietary frequency response and filtering algorithms applied to the arterial waveform.⁹ In general, pulse rate is calculated by the frequency with which the arterial waveform varies from its baseline enough to cross a set threshold. The threshold is determined by algorithms, which are designed to capture pulse waves and minimize the contribution of artifact. Therefore, low-amplitude pulse waves may not cross the threshold required for pulse rate calculation, leading to an erroneously low calculated pulse rate. Conversely, pulse waves generated by patient or transducer movement may generate waveform peaks that aberrantly cross the threshold, leading to an erroneously high calculated pulse rate. In the face of substantial arterial pressure waveform variability, alternate methods of pulse rate calculation, including manual monitoring, may be more accurate. In the dog of the present report, close examination of the arterial waveform revealed alternating pulse amplitudes. However, for patients in which direct blood pressure monitoring is not performed, digital pulse palpation along with thoracic auscultation may aid diagnosis.

The relative contribution of hemodynamic and inotropic factors should be taken into consideration when treatment is determined. In dogs with primary cardiac disease that impairs systolic function (eg, dilated cardiomyopathy), impaired contractility of diseased myocardium must be considered as a primary differential diagnosis. Positive inotropic agents and afterload reduction are indicated to improve cardiac output. In the dog of the present report, such a primary cardiac disease was an unlikely underlying cause because the dog had no evidence of preexisting heart disease, whereas pulsus alternans was likely the result of impaired diastolic filling exacerbated by tachycardia. Decreased cardiac output from reduced venous return caused by the partial CVC occlusion, catecholamine release from the pheochromocytoma, and mild amount of blood loss during tumor dissection may have contributed to the intermittent development of tachycardia in this dog. Therefore, crystalloid boluses were administered to increase circulating blood volume, improve venous return, and reduce the HR. If pulsus alternans is observed with supraventricular tachycardia without CVC occlusion, HR control with either pharmacological intervention or catheter ablation of arrhythmogenic foci should be considered.⁴ In the dog of the present report, tachycardia could have been pathological or compensatory; therefore, it was not treated directly.

Pulsus alternans is an uncommon and potentially perplexing manifestation of serious hemodynamic instability. The present report highlighted the importance of corroborating digitally monitored patient parameters with clinical findings. Careful consideration of the patient's health and history and possible contributing factors is essential for guiding successful diagnosis and treatment.