Effect of Increasing Heart Rate and Tidal Volume on Stroke Volume Variability in Vascular Surgery Patients Nathan A. Roeth, MD,* Timothy R. Ball, MD,* William C. Culp Jr, MD,* W. Todd Bohannon, MD,† Marvin D. Atkins, MD,† and William E. Johnston, MD* Objective: Because heart rate affects ventricular filling, the aim of the present study was to assess the effects of increasing heart rate and tidal volume on stroke volume variability to determine whether this dynamic index is heart-rate dependent. Design: Prospective, randomized study. Setting: Single university hospital. Participants: Eighteen vascular surgery patients having general anesthesia and endotracheal intubation with an arterial catheter connected to the Vigileo FloTrac system (Edwards Lifesciences, Irvine, CA) and a transesophageal atrial pacemaker (CardioComman Inc, Tampa, FL). Intervention: A 2
2 factorial study of changes in heart rate (80 bpm and 110 bpm) and tidal volume (6 mL/kg and 10 mL/kg). Measurements and Main Results: With tidal volume at 6 mL/kg, increasing heart rate from 80 mL/kg to 110 bpm caused stroke volume variability to increase from 12.2% ⫾ 5.7% to 13.2% ⫾ 5.3% (p o 0.05), and with tidal volume at 10 mL/kg,

increasing heart rate from 80 mL/kg to 110 bpm caused stroke volume variability to increase from 19.7% ⫾ 7.9% to 22.0% ⫾ 8.6% (p o 0.05). In comparison, increasing tidal volume from 6 mL/kg to 10 mL/kg produced a significantly greater effect on stroke volume variability than increasing heart rate. Conclusions: Stroke volume variability is sensitive to increases in heart rate in addition to tidal volume. Increasing heart rate caused stroke volume variability to increase significantly, although not to the same magnitude as increasing tidal volume. When using dynamic volume indices, clinicians should be aware of increases in heart rate, although its clinical impact may be relatively minor compared with changes in tidal volume. & 2014 Elsevier Inc. All rights reserved.

A

hypovolemic patients, the overall reduction in cardiac filling may be compounded by a reflex tachycardia. The effect of increasing HR on a dynamic index such as SVV has not been assessed. The authors hypothesized that increasing HR would increase the amount of SVV in anesthetized patients receiving positive-pressure ventilation. If so, the clinical use of these dynamic criteria would need to consider potential confounding effects of HR changes independent of the patient’s volume status. Because Vt already has been shown to be a significant factor affecting SVV,4 the effect of increasing HR by atrial pacing was compared with that of increasing Vt in perioperative vascular surgical patients to assess their relative magnitudes of effect.

PRIMARY GOAL of hemodynamic management in critically ill patients is adequate volume replacement to achieve optimal cardiac performance. Dynamic volume indices such as stroke volume variation (SVV) supplement the use of static indices of filling pressures for predicting patient responsiveness to fluid resuscitation.1,2 These dynamic indices relate to the cyclical changes in arterial blood pressure resulting from acute changes in right ventricular (RV) and left ventricular (LV) loading conditions involving both preload and afterload caused by positive-pressure ventilation. Any reduction in RV filling, such as occurring with hemorrhage, hypovolemia, an increase in tidal volume (Vt), or a reduction in chest wall compliance can further decrease RV stroke volume (SV) and linearly increase the variability in dynamic indices.3–5 Another potentially relevant factor that can alter venous return and RV preload is heart rate (HR). It long has been known that increasing HR by atrial pacing decreases LV filling pressures and SV6 so that cardiac output remains unchanged.7,8 More recent studies have shown that tachycardia by atrial pacing causes linear reductions in right-sided8 and, consequently, left-sided9 preload, thereby reducing SV by the Frank-Starling relationship. In the clinical management of

From the Departments of *Anesthesiology and †Surgery (Division of Vascular Surgery), Texas A&M University System Health Science Center College of Medicine, Temple, TX. This study was funded in part by a resident research grant from S&W Healthcare. Presented as an abstract at the 42nd Annual Congress of the Society of Critical Care Medicine, January 19-23, 2013, San Juan, Puerto Rico. Address reprint requests to William E. Johnston, MD, Department of Anesthesiology, Scott & White Healthcare, 2401 South 31st Street, Temple, TX 76508. E-mail: [email protected] © 2014 Elsevier Inc. All rights reserved. 1053-0770/2601-0001$36.00/0 http://dx.doi.org/10.1053/j.jvca.2014.05.014 1516

KEY WORDS: heart rate, tidal volume, stroke volume variability, surgical patients, general anesthesia, positivepressure ventilation

MATERIALS AND METHODS This research study was approved by the institutional review board, and written informed consent was obtained from each patient before study enrollment. Patients undergoing elective vascular surgery were enrolled and consented for study. Exclusion criteria included rhythm other than sinus, ejection fraction less than 50%, significant valvular heart disease, or patient refusal. Patients had nothing by mouth for at least 8 hours before surgery and were hydrated with 500 mL to 750 mL of 0.9% normal saline in the preoperative holding area. Twenty-two nonemergent American Society of Anesthesiology 3 and 4 patients in normal sinus rhythm were studied after induction of general anesthesia before initiation of various vascular surgeries. Each enrolled subject received a 20-gauge radial arterial catheter connected to a FloTrac/ Vigileo algorithm software system (version 3.02, Edwards LifeSciences, Irvine, CA). Upon arrival in the operating room, standard monitors were applied to the patient, along with a bispectral index monitor (BIS Brain Monitor, Covidien, Mansfield, MA). General anesthesia was induced with intravenous propofol, 0.5 mg/kg to 1.0 mg/kg, sufentanil, 10 μg to 25 μg, and rocuronium, 0.6 mg/kg to 0.8 mg/kg, to facilitate endotracheal intubation. After intubation, a transesophageal atrial pacer (CardioComman, Inc., Tampa, FL) was inserted in the esophagus and atrial capture was confirmed with corresponding heart rate changes on the electrocardiogram and arterial waveform tracings. Patients with

Journal of Cardiothoracic and Vascular Anesthesia, Vol 28, No 6 (December), 2014: pp 1516–1520

1517

EFFECT OF HEART RATE AND TIDAL VOLUME ON DYNAMIC VOLUME INDICES

arrhythmias during pacing or without pacemaker capture were eliminated from the study protocol. General anesthesia was maintained with 0.5 MAC of sevoflurane and 100% oxygen. Bispectral index monitor readings were measured repeatedly to ensure an adequate depth of anesthesia throughout data collection. The respiratory rate initially was adjusted to maintain end-tidal carbon dioxide tension between 32 mmHg and 35 mmHg, with a baseline Vt of 6 mL/kg body weight and remained constant throughout the experimental protocol. Each subject was assigned randomly to a 2
2 factorial design of changes in HR (80 bpm or 110 bpm) and Vt (6 mL/kg body weight or 10 mL/kg body weight). After each change in HR and Vt, a period of 3 minutes elapsed to allow for physiologic stabilization. At the end of each period, various hemodynamic parameters were recorded and the next randomized physiologic change was initiated. After the fourth and final physiologic change, Vt and HR were adjusted at the attending anesthesiologist’s discretion, and the surgical case proceeded. If during any of the 4 study periods hemodynamic instability (defined as a decrease in mean arterial blood pressure below 50 mmHg) ensued, the attending anesthesiologist stopped the study and treated the patient as clinically indicated and that subject was eliminated from the data analysis. At each measurement period, the various hemodynamic parameters were acquired. From the Vigileo FloTrac system, measurements included cardiac output (CO), cardiac index (CI), SV, stroke volume index (SVI), and SVV. The invasive blood pressures also were recorded at each interval, including systolic arterial pressure, diastolic arterial pressure, and mean arterial pressure. Recorded pulmonary parameters from each period included peak airway pressure as well as Vt. On the basis of data from a previous study that examined the effects of increasing HR on RV filling and SV,8 it was decided to study 16 patients. Accordingly, to account for potential dropout during the study, 25% more patients were enrolled for a final sample size of 22 patients. The research study was a 2
2 randomized factorial design, with the 2 factors being HR and Vt. Data, presented as mean ⫾ standard deviation, were analyzed by ANOVA, which included a combined group effect if there was no interaction between groups. Statistical significance was considered at p o 0.05. RESULTS

Twenty-two subjects were screened and enrolled in the study. Three subjects were excluded from the study secondary to hypotension during the study protocol and intervention by the attending anesthesiologist; one other patient was excluded due to proper capture failure by the transesophageal atrial pacer. Eighteen patients successfully completed the study protocol, and their surgeries consisted of femoral-popliteal artery bypass, carotid endarterectomy, and femoral-femoral artery bypass. Table 1 displays the demographics for the patients included in the data analysis. Table 1. Patient Demographic Characteristics (n ¼ 18 patients) Variable

Gender Male Female Age (years) Weight (kg) Height (cm) Body mass index

11 (61%) 7 (39%) 67 ⫾ 15 76 ⫾ 14 172 ⫾ 10 26 ⫾ 4

NOTE. Data are expressed as mean ⫾ standard deviation or frequency (%).

Table 2. Hemodynamic Measurements at the 4 Different Measurement Intervals Heart Rate 80 bpm Measurement

Vt 6 mL/kg

Vt 10 mL/kg

Heart Rate 110 bpm Vt 6 mL/kg

Vt 10 mL/kg

Systolic arterial 111 ⫾ 23 103 ⫾ 21† 113 ⫾ 24 101 ⫾ 22† pressure (mmHg) Diastolic arterial 54 ⫾ 8 52 ⫾ 8† 61 ⫾ 8* 55 ⫾ 8* † pressure (mmHg) Mean arterial 75 ⫾ 14 68 ⫾ 11† 77 ⫾ 13 71 ⫾ 12† pressure (mmHg) Cardiac output 5.7 ⫾ 1.3 5.1 ⫾ 1.4† 6.3 ⫾ 1.4* 5.6 ⫾ 1.5*† (L/min) Stroke volume 72 ⫾ 15 63 ⫾ 17† 57 ⫾ 12* 51 ⫾ 14*† (mL/beat) Tidal volume (mL) 451 ⫾ 85 757 ⫾ 143† 451 ⫾ 85 757 ⫾ 143† Peak airway pressure 12 ⫾ 4 18 ⫾ 5 12 ⫾ 4* 17 ⫾ 6* (mmHg) NOTE. Data expressed as mean ⫾ standard deviation. *p o 0.05 compared with corresponding heart rate 80 bpm value. †p o 0.05 compared with corresponding Vt 6 mL/kg value.

Hemodynamic variables for the 4 measurement intervals during the experimental protocol are shown in Table 2. Increasing Vt from 6 mL/kg to 10 mL/kg produced significant and expected decreases in SV, cardiac output, and arterial blood pressures. Increasing HR from 80 bpm to 110 bpm caused a significant 20% reduction in SV at each Vt setting, although cardiac output increased overall. Diastolic arterial blood pressure increased significantly with an increase in HR without changes in the systolic arterial blood pressure. There were significant and independent HR and Vt effects on SVV. Fig 1 summarizes the data from the 4 measurement periods, with combined group effects because no interactions between HR and Vt were found. Combining data from both Vt 6 mL/kg and 10 mL/kg groups in the 2
2 model, atrial pacing at 110 bpm HR produced greater SVV (17.6% ⫾ 8.3%) than 80 bpm HR (15.9% ⫾ 7.8%; p ¼ 0.027). Similarly, when data from both HR 80 bpm and 110 bpm groups were combined, increasing Vt to 10 mL/kg produced significantly greater SVV (20.8 ⫾ 8.2%) than ventilating with Vt 6 mL/kg (12.7% ⫾ 5.5%; p o 0.0001). Although the present data show a statistically significant effect from increasing HR (Fig 1), the overall clinical effect may be less significant. An SVV cutoff value of 12% as suggested by other authors1,10,11 may be applied to our data to predict responders and non-responders to a fluid challenge (Fig 2). At Vt 6 mL/kg, increasing HR from 80 to 110 bpm would cause 2 nonresponders to become responders. In contrast, at Vt 10 mL/kg, nearly all patients become responders without any group effect from increasing HR. The pronounced effect from increasing Vt from 6 mL/kg to 10 mL/kg for the same HR on the number of responders is shown in in Fig 2. DISCUSSION

Increasing HR from 80 bpm to 110 bpm caused a reduction in SV and a slight but statistically significant increase in SVV in normovolemic, anesthetized patients receiving positivepressure ventilation. These findings indicated that HR should be considered when using dynamic indices to guide fluid

1518

Fig 1. Stroke volume variability (SVV) data are reported as mean ⫾ standard deviation. The inner cells (solid lines) contain SVV data from each of the 4 timepoints: Tidal volume (Vt ) 6 mL/ kg and heart rate (HR) 80 bpm; Vt 6 mL/kg and HR 110 bpm; Vt 10 mL/kg and HR 80 bpm; Vt 10 mL/kg and HR 110 bpm. Because there was no interaction between HR and Vt , the independent effects of each intervention could be examined and are reported in the outer cells (dashed lines). Increasing HR caused an increase in SVV, although the effect from augmenting Vt was greater.

therapy in the operating room. In this study, increasing HR by nearly 40% caused SVV to increase by only 10% so that its overall clinical impact may not be consequential. In contrast, as seen in Fig 2, the effect of increasing Vt clearly was greater than the effect of increasing HR. Increasing Vt by nearly 67%

ROETH ET AL

caused a similar (64%) percentage increase in SVV. The effect of Vt on dynamic indices was consistent with the findings of other studies.3–5 Although the change in SVV from increasing HR was less than that from increasing Vt, there are several potential explanations. First, the magnitude of the hemodynamic impact from increasing HR on RV filling may not have been as severe as the magnitude of the impact from increasing Vt. Several studies have shown that dynamic indices are related directly to Vt,3–5 which would be expected because positive-pressure ventilation adversely affects both RV preload and afterload. Normally, afterload and preload are matched in which an acute increase in afterload causes a similar increase in preload so that SV can be maintained by the Frank-Starling relationship.12–14 However, positive intrathoracic pressure limits venous return and, consequently, any compensatory preload response to an increase in RV afterload. This, in turn, causes RV preload-afterload mismatch and an acute reduction in SV. Although this physiology has been described primarily for the left ventricle,12–13 a similar contractile response would occur during positive-pressure ventilation on the right side of the heart as well. Because the change in intrathoracic pressure is related directly to the magnitude of Vt, greater values of SVV with higher Vt would be expected. Several laboratory and clinical studies have found that increasing HR reduces the time for diastolic filling and, consequently, right- and left-sided end-diastolic volumes, causing a proportionate reduction in SV.7–9,15 In patients with atrial fibrillation, increasing HR caused greater variability in the aortic velocity-time integral, which suggested that SVV would increase with tachycardia.16 In addition, increasing HR by atrial

Fig 2. Bar graph of number of patients when grouped as predicted responders or nonresponders to a fluid challenge according to an arbitrary stroke volume variability cutoff value of 12% where Vt ¼ tidal volume in mL/kg and HR ¼ heart rate in bpm. With Vt 6 mL/kg, increasing heart rate from 80 bpm to 110 bpm would cause 2 nonresponders to become predicted responders. This effect from increasing heart rate was not seen at Vt 10/mL/kg.

1519

EFFECT OF HEART RATE AND TIDAL VOLUME ON DYNAMIC VOLUME INDICES

pacing may impair LV diastolic function, which further could reduce filling.17 An earlier study of anesthetized patients undergoing coronary revascularization found a linear, inverse relationship between HR and venous return.8 Increasing HR caused a reduction in RV end-diastolic volume while end-systolic volume remained relatively constant so that SV progressively decreased. De Backer et al established a linear relationship between pulse pressure variation and the ratio of HR/respiratory rate in critically ill, mechanically ventilated patients, in whom the amount of variation increased at higher ratio values exceeding 3.6.18 Although only the respiratory rate was varied in that study, the present data support this finding because a similar SV response was found with increasing HR by atrial pacing. Another potential factor that could influence results is the myocardial force-frequency relationship, or Treppe effect, of the heart, which relates to an intrinsic property of the myocardium to increase the force of contraction as HR increases.19–21 Originally described by Bowditch in 1871,22 this contractility property has been described in patients with normal cardiac function23 as well as in patients after cardiac surgery9 and correlates directly with the magnitude of increase in HR. On a molecular level, the force-frequency relationship is associated with an increase of ionic calcium (Caþþ) transits as the stimulation frequency increases.21 Intracellular Caþþ concentration is determined by the sarcoplasmic reticulum Caþþ release and the Caþþ flux through the sarcolemma via L-type Caþþ channels and the sodium (Naþ)–Caþþ exchange pump. It is postulated that when the frequency of contraction is elevated, the uptake and storage of Caþþ by the sarcoplasmic reticulum are enhanced. Consequently, greater Caþþ is released causing greater contractility. Potentially, by causing the heart to eject to a lower end-systolic volume, this acute increase in contractility could mask the reduction in preload caused by increasing HR, thereby attenuating the degree of SV reduction and SVV. Schaefer et al found that the preload response to an

increase in HR determined, in part, the inotropic effect of tachycardia.23 Several potential limitations of the present study should be discussed. First, patients having normal cardiac function without known valvular abnormalities were studied. The response of increasing HR in patients with congestive failure and dilated cardiomyopathy is unknown and potentially could be altered by their blood volume and responsiveness to preload. There were constraints on how fast to safely pace the right atrium in vascular surgery patients with probable co-existing coronary disease, and any effect from a faster HR was not known. Second, patients with documented peripheral vascular disease were studied. Although it has been suggested that this pathology could interfere with proper Vigileo readings, inconsistencies in the recording during the experimental protocol were not detected. Third, the respiratory rate was maintained constant as Vt changed, which could alter minute ventilation, although each measurement interval was randomized and short-lived. Lastly, although subjects were classified arbitrarily as responders or nonresponders based on reported SVV values in the literature,1,10,11 a fluid bolus was not administered to patients. However, Fig 2 illustrates the relative significance of increases in Vt and in HR in anesthetized, normovolemic, ventilated patients. It is interesting to note that increasing HR had a greater effect on the predicted number of responders at the lower Vt than the higher Vt tested. CONCLUSION

Increases in both Vt and HR independently affect SVV. An increase in Vt has a greater influence on the magnitude of SVV increase when compared with increasing HR, but the potential influence of a faster HR on SVV should be considered when using dynamic indices to guide intraoperative fluid management particularly at lower Vt values.

REFERENCES 1. Marik PE, Monnet X, Teboul JL: Hemodynamic parameters to guide fluid therapy. Ann Intensive Care 1:1-9, 2011 2. Enomoto TM, Harder L: Dynamic indices of preload. Crit Care Clin 26:307-321, 2010 3. Mesquida J, Kim HK, Pinsky MR: Effect of tidal volume, intrathoracic pressure, and cardiac contractility on variations in pulse pressure, stroke volume, and intrathoracic blood volume. Intensive Care Med 37:1672-1679, 2011 4. Reuter DA, Bayerlein J, Goepfert MS, et al: Influence of tidal volume on left ventricular stroke volume variation measured by pulse contour analysis in mechanically ventilated patients. Intensive Care Med 29:476-480, 2003 5. Charron C, Fessenmeyer C, Cosson C, et al: The influence of tidal volume on the dynamic variables of fluid responsiveness in critically ill patients. Anesth Analg 102:1511-1517, 2006 6. Linhart JW: Pacing-induced changes in stroke volume in the evaluation of myocardial function. Circulation 43:253-261, 1971 7. Linhart JW: Myocardial function in coronary artery disease determined by atrial pacing. Circulation 44:203-212, 1972 8. Johnston WE, Robertie PG, Dudas LM, et al: Heart rate–right ventricular stroke volume relation with myocardial revascularization. Ann Thor Surg 52:797-804, 1991

9. Eichhorn EJ, Diehl JT, Konstam MA, et al: Left ventricular inotropic effect of atrial pacing after coronary artery bypass grafting. Am J Cardiol 63:687-692, 1989 10. Zimmermann M, Feibicke T, Keyl C, et al: Accuracy of stroke volume variation compared with pleth variability index to predict fluid responsiveness in mechanically ventilated patients undergoing major surgery. Eur J Anaesthesiol 27:555-561, 2010 11. Benes J, Chytra I, Altmann P, et al: Intraoperative fluid optimization using stroke volume variation in high risk surgical patients: Results of prospective randomized study. Critical Care 14:118-132, 2010 12. Ross J Jr: Afterload mismatch and preload reserve: A conceptual framework for the analysis of ventricular function. Prog Cardiovasc Dis 18: 255-264, 1976 13. Ross J, Franklin D, Sasayama S: Preload, afterload, and the role of afterload mismatch in the descending limb of cardiac function. Eur J Cardiol 4:77-86, 1976 14. Ribaric S, Kordas M: Simulation of the Frank-Starling law of the heart. Comput Math Methods Med 2012:267834, 2012 15. Cowley AW Jr, Guyton AC: Heart rate as a determinant of cardiac output in dogs with arteriovenous fistula. Am J Cardiol 28:321-325, 1971 16. Kerr AJ, Simmonds MB, Stewart RAH: Influence of heart rate on stroke volume variability in atrial fibrillation in patients with normal and impaired left ventricular function. Am J Cardiol 82:1496-1500, 1998

1520

17. Royse CF, Royse AG, Wong CT, et al: The effect of pericardial restraint, atrial pacing, and increased heart rate on left ventricular systolic and diastolic function in patients undergoing cardiac surgery. Anesth Analg 96:1274-1279, 2003 18. De Backer D, Taccone FS, Holsten R, et al: Influence of respiratory rate on stroke volume variation in mechanically ventilated patients. Anesthesiology 110:1092-1097, 2009 19. Alpert NR, Leavitt BJ, Ittleman FP, et al: A mechanistic analysis of the force-frequency relation in non-failing and progressively failing human myocardium. Basic Res Cardiol 93:23-32, 1998 20. Antoons G, Mubagwa K, Nevelsteen I, et al: Mechanisms underlying the frequency dependence of contraction and [Ca(2þ)](i)

ROETH ET AL

transients in mouse ventricular myocytes. J Physiol 543: 889-898, 2002 21. Endoh M: Force-frequency relationship in intact mammalian ventricular myocardium: physiological and pathophysiological relevance. Eur J Pharmacol 500:73-86, 2004 22. Bowditch HP: Über die Eigenthüemlichkeiten der Reizbarkeit, welche die Muskelfasern des Herzens zeigen. Verh. K Sachs Gen Wochenschr Leipzig Math Phys Cl 23:652-669, 1871 23. Schaefer S, Taylor AL, Lee HR, et al: Effect of increasing heart rate on left ventricular performance in patients with normal cardiac function. Am J Cardiol 61:617-620, 1988