- Email: [email protected]
Hemodynamic instability in patients undergoing pulmonary embolectomy: institutional experience
Journal of Clinical Anesthesia (2014) xx, xxx–xxx
Original contribution
Hemodynamic instability in patients undergoing pulmonary embolectomy: institutional experience Jeremy M. Bennett MD (Assistant Professor)a,⁎, Mias Pretorius Mb, Chb, MSCI (Associate Professor)a , Rashid M. Ahmad MD (Assistant Professor)b , Susan S. Eagle MD (Associate Professor)a a
Department of Anesthesiology Vanderbilt University Medical Center, Nashville, TN 37232 Department of Cardiac Surgical Sciences Vanderbilt University Medical Center, Nashville, TN 37232
b
Received 17 April 2014; revised 12 August 2014; accepted 29 October 2014
Keywords: Pulmonary embolectomy; Cardiac surgery; Pericardiotomy; Hemodynamic instability; Right ventricle; Right ventricular failure; Emergent surgery; Emergent cardiac surgery
Abstract Objectives: Acute pulmonary embolism is a major cause of morbidity and mortality in patients presenting for emergent cardiac surgery with overall mortality ranging from 6% to as high as 85%. While the initial focus of treatment is nonsurgical or percutaneous interventions, surgical treatment continues to be a treatment for patients with refractory thrombus burden or cardiogenic shock. Our institution regularly performs surgical pulmonary embolectomy with improved outcomes compared to current reports. We thus performed a retrospective analysis of outcomes of pulmonary embolectomy patients and anesthetic management. Design: A retrospective review of 40 patients undergoing emergent pulmonary embolectomy over a 4 year period (2008-2012) at our institution was performed to assess for a 2nd period of critical instability. Setting: The study was conducted at a tertiary, level 1, trauma university medical center. Participants: The study was performed through chart review of patient hospital records. Interventions: No interventions were performed. MeasurementsAnesthetic records were reviewed along with echocardiographic records and surgical reports to assess cardiac function, need for emergent cardiopulmonary bypass, and degree of patient morbidity. Conclusions: A total of 40 patients were studied. Hemodynamic instability occurred in 12.5% of patients at time of induction requiring emergent cardiopulmonary bypass. Another 17% of patients who remained stable following induction developed subsequent instability requiring emergent cardiopulmonary bypass during pericardial opening or manipulation which has not been previously reported. One patient died during hospitalization. Patients who required emergent bypass following induction of general anesthesia tended to receive higher doses of induction drugs than the other groups. In patients who needed emergent bypass during pericardial manipulation there were no identifiable factors suggesting that these patients remain at risk despite a stable post-induction course. © 2014 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Tel.: +1 615 322 4650. E-mail address: [email protected] (J.M. Bennett). http://dx.doi.org/10.1016/j.jclinane.2014.10.007 0952-8180/© 2014 Elsevier Inc. All rights reserved.
2
1. Introduction Anesthetic management of patients undergoing pulmonary embolectomy is challenging given the emergent nature of the procedure, acute hemodynamic instability, patient comorbidities, and experience in various centers. Surgical management of pulmonary embolism is most often considered in patients with central or branch pulmonary artery thrombus burden with cardiogenic shock. It may also be elected for patients with stable hemodynamics who demonstrate right ventricular failure on echocardiographic examination or have failed medical therapy with thrombolytics, heparin, or percutaneous embolectomy [1,2]. A recent case series of 52 patients demonstrated a 19% incidence of hemodynamic collapse during induction of general anesthesia (GA) necessitating emergent cardiopulmonary bypass (CPB) [3]. Further studies have demonstrated overall mortality from surgical intervention ranging from 6%-85% [1,2]. Given the significant morbidity of these patients, surgical manipulation and stress are likely to impact venous return, pulmonary vascular resistance (PVR), and cardiac output of these critically ill patients. Because of paucity of anesthetic-related outcomes studies for this critical patient population, we performed a retrospective study over a 4-year period of patients presenting for pulmonary embolectomy. The aim of this study is to assess the prevalence of hemodynamic instability during induction of GA as well as at any time before institution of CPB with attention toward anesthetic technique.
2. Methods A retrospective analysis was performed on all patients who underwent urgent/emergent pulmonary embolectomy at Vanderbilt University Medical Center between January 2008 and December 2012. A waiver of consent was obtained from the Vanderbilt Human Research Protection Program for assessment of patient records. Patients who arrived to the operating room (OR) endotracheally intubated or having received active cardiopulmonary resuscitation at any time before arrival in the OR were excluded from the study. Patient's medical records were reviewed for patient demographic information including but not limited to age, sex, right ventricle (RV) and left ventricle (LV) function, comorbidities (chronic lung disease, history of smoking, cardiac disease, renal insufficiency, and anemia), body mass index, prior cardiac surgery, and postoperative morbidity. Anesthetic records were reviewed for anesthetic induction agent (etomidate, fentanyl, ketamine, and propofol), inotropic administration, ventilator parameters (ventilator mode, peak airway pressures (PAWs), and positive end-expiratory pressure), and hemodynamic changes in heart rate (HR) and central venous pressure (CVP). Preoperative radiologic and transthoracic echocardiographic information was reviewed,
J.M. Bennett et al. if available as well as all intraoperative transesophageal echocardiographic (TEE) findings to assess thrombus location and burden, RV function, and RV dilation. Right ventricular function and dilation were determined based on accepted American Society of Echocardiography standards for assessment of the RV [4]. Preoperative transthoracic echocardiographic data were available in all patients with the exception of 2 patients in the stable group and 1 patient in both the induction-unstable and delayed-unstable groups (total of 4 patients). All patients received intraarterial catheters and central venous catheters before induction of GA using local anesthesia, supplemental oxygen as supplied by a facemask, and titration of sedation as dictated by the attending anesthesiologist. The location of these catheters was in the radial artery and right internal jugular vein, respectively, for most patients. If radial arterial catheter placement was difficult or unable to be performed, a femoral arterial catheter was placed. Sedation most commonly consisted of boluses of midazolam and ketamine. After central venous catheter placement, pulmonary artery catheters were placed in all patients and advanced to 20-cm depth until separation from CPB, at which time the pulmonary artery catheter was advanced into the main pulmonary artery. All patients were prepared and surgically draped with a surgeon scrubbed and ready in the OR before induction of GA. The type of anesthetic induction drug was at the discretion of the attending anesthesiologist. Transesophageal echocardiographic was performed in all patients after tracheal intubation, where there was no concern for esophageal pathology. Hemodynamic instability requiring emergent CPB was identified as acute and persistent hypotension occurring after GA induction that did not respond to vasopressor support and required emergent initiation of CPB. If patients remained stable after induction of GA, we assessed for any other period, when hemodynamic instability developed before CPB initiation. Surgeon and anesthesia records were reviewed for all patients if instability was identified for potential description of the event. Data are presented as median with interquartile range unless otherwise indicated. Categorical data were compared between groups using χ2 or Fisher exact tests as appropriate. Continuous baseline data were compared using the KruskalWallis test. Comparisons of CVP and HR among groups were made with the Wilcoxon signed rank test. A 2-tailed P value less than .05 was considered statistically significant. Statistical analyses were performed with the statistical package SPSS for Windows (version 21.0; IBM, New York, NY).
3. Results The study population consisted of 46 patients of whom 6 patients were excluded for preoperative cardiopulmonary resuscitation and/or arriving tracheally intubated before the OR. A total of 40 patients were included in the study and
Hemodynamic instability in patients Table 1
3
Preoperative patient characteristics
Characteristic
Induction unstable (n = 5)
Delayed unstable (n = 6)
Stable (n = 29)
P
Age, y BMI, kg/m2 Female sex, n (%) Diabetes, n (%) Current smoker, n (%) History of COPD, n (%) History of MI, n (%) Hypertension, n (%) Creatinine (mg/dL) Hematocrit (%) Preoperative LVEF, % Preoperative RV function, n (%) Normal-to-mild depression Moderate depression Severe depression RV dilation, n (%) Normal-to-mild dilation Moderate dilation Severe dilation Medications, n (%) Inotropes before induction ACEi or ARB β-blockers Coumadin Heparin
45 (37-60) 27 (21-39) 4 (80) 2 (40) 2 (40) 0 1 (20) 2 (40) 0.88 (0.72-2.05) 39 (34-42) 55 (43-55)
48 (23-75) 34 (26-41) 3 (50) 1 (17) 0 0 1 (17) 3 (50) 1.14 (0.90-1.59) 42 (35-44) 55 (54-58)
58 30 9 8 2 2 4 13 1.14 41 55
.19 .45 .10 .69 .05 .67 .93 .95 .44 .64 .29 .85
(52-68) (28-38) (31) (28) (7) (7) (14) (45) (0.97-1.37) (36-43) (55-55)
2 (40) 2 (40) 0
2 (33) 2 (33) 1 (17)
8 (28) 16 (55) 3 (10)
1 (20) 1 (20) 2 (40)
3 (50) 1 (17) 1 (17)
7 (24) 10 (35) 10 (35)
2 (40) 1 (20) 1 (20) 0 1 (20)
0 1 (17) 1 (17) 0 2 (33)
3 6 1 1 10
.76
(10) (21) (3.4) (3) (35)
.11 .98 .28 .82 .82
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; LVEF, left ventricular ejection fraction; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker.
divided into 3 groups based on hemodynamic status: stable group, unstable at induction group (induction unstable), and the delayed-unstable group. Five patients (12.5%) developed refractory hemodynamic collapse requiring emergent CPB immediately after anesthesia induction and intubation (induction-unstable group). Six patients (15%) developed hypotension that was refractory to vasopressor or inotropic support after a stable induction of GA (delayed-unstable group). One patient died in the induction-unstable group during their hospitalization. No other deaths occurred in any of the other groups. The surgeon operative report and/or anesthetic records for the 6 patients in the delayed-unstable group indicated that instability occurred during pericardial sac manipulation (either pericardiotomy or during retention suture placement). Twenty-nine patients (72.5%) remained hemodynamically stable throughout induction and up to initiation of CPB (stable group). Preoperative patient characteristics for the 3 groups are presented in Table 1. Female sex was more frequent in the unstable postinduction group, although this was not significant (P = .10). Preoperative RV function and dilation did not differ significantly among the groups. Table 2 details intraoperative patient characteristics. There were no significant differences noted in the choice
of induction drug or dosage administered, although there was a trend toward lower induction drug dosages in the stable and delayed-unstable groups compared with the induction-unstable group. As expected, the lowest mean arterial pressure (MAP) and decrease in systolic blood pressure greater than 40 mm Hg were significantly different among the groups with the lowest MAP seen in the induction-unstable group. Severe intraoperative RV dysfunction trended more frequently in the induction-unstable (60%) and delayed-unstable (50%) groups but was not significantly different from the stable group (28%). Peak airway pressure, tidal volume, and mode of ventilation were not significantly different among the 3 groups, although higher PAW tended in the induction-unstable group. Central venous pressure was significantly higher in the delayed-unstable group compared with the stable group after pericardial opening. The Figure details the change in HR and CVP from poststernotomy measurements to the time of pericardial opening in the stable and delayed-unstable groups. Neither HR (from a median baseline of 105-100 beats per minute, P = .92) nor CVP (from a median baseline of 22-22 mm Hg, P = .87) changed significantly in the stable group. In contrast, HR decreased (from a median baseline of 108-95 beats per minute, P = .039), and CVP increased (from a median baseline of 22-28
4
J.M. Bennett et al. Table 2
Intraoperative patient characteristics
Ketamine, n (%) Dose, mg/kg Etomidate, n (%) Dose, mg/kg Fentanyl, n (%) Dose, μg/kg Lowest MAP postinduction (mm Hg) SBP decrease N 40 mm Hg postinduction, n (%) HR at baseline, beats per minute HR after pericardial opening, beats per minute CVP baseline, mm Hg CVP after pericardial opening, mm Hg Severe RV dysfunction, n (%) Intraoperative pH Ventilator PAW, mm Hg Tidal volume, mL/kg Volume control ventilation, n (%)
Induction unstable (n = 5)
Delayed unstable (n = 6)
3 (60) 1.4 (0-2.4) 0 NA 4 (80) 4.3 (2.0-7.7) 47 (43-58)
3 (50) 0.4 (0-2.3) 3 (50) 0.04 (0-0.07) 5 (83) 3.5 (1.6-8.5) 66 (49-70)
5 (100) 125 (105-135) NA 25 (16-27) NA 3 (60) NA 43 (34-44) 6.9 (4.2-10.2) 3 (60)
2 (33)
Stable (n = 29)
P
18 (62) 0.6 (0-1.4) 9 (31) 0.1 (0.1-0.18) 22 (76) 2.9 (1.3-4.5) 72 (63-83)
.86 .83 .19 .31 .91 .52 .004
10 (35)
.02
108 (99-113) 95 (84-103)
105 (95-120) 100 (95-113)
.17 .13
22 (21-30) 28 (24-34)
22 (20-24) 22 (20-24)
.75 .01
3 (50) 7.33 (7.30-7.39) 33 (29-41) 5.9 (4.6-7.8) 5 (83)
8 (28) 7.36 (7.32-7.40) 34 (29-39) 5.8 (4.8-6.9) 27 (93)
.26 .36 .18 .72 .11
Abbreviation: SBP, systolic blood pressure; NA, not available.
mm Hg, P = .045) significantly in the delayed-unstable group, although 1 patient had a decrease in CVP and increase in HR in the delayed-unstable group.
4. Discussion Pulmonary embolectomy is a challenging operative procedure performed for patients with severe, massive PE. In-hospital morbidity and mortality levels can be high and may be significantly affected by anesthetic management with hemodynamic instability after GA induction and need for emergent CPB [3]. In addition, sternotomy, pericardial opening, and cardiac manipulation along with poor respiratory management (high PAWs, use of positive end-expiratory pressure, etc) may result in hemodynamic instability after GA induction. In our study, hemodynamic collapse necessitating emergent institution of CPB occurred in 13% of patients after induction as has been previously described [3]. The PAW in these patients tended to be higher than the other 2 groups. Although we are not able to elaborate on why the PAW tended toward higher values, we suspect this may reflect the acute decompensation during GA induction resulting in reduced ventilator vigilance. Our current ventilators have preset 600 mL tidal volumes settings with a respiratory rate of 8 and zero positive end-expiratory pressure. If the tidal volumes were not adjusted while emergently instituting CPB, this may cause substantial
increase in PAW, when not adjusted for patient ideal body weight and could reflect the observed trend. In patients who remained stable after anesthetic induction, an additional 15% of patients developed acute hemodynamic decompensation that did not seem to be related to the induction period. In particular, review of surgeon and anesthesiologist documentation and anesthetic records seemed to indicate that acute decompensation occurred during pericardial manipulation/opening. Of interest is the finding that, in 5 of the 6 patients who developed acute decompensation during pericardial manipulation, a significant increase in CVP and decrease in HR occurred. This likely reflects acute RV failure, possibly from ischemia as has been demonstrated recently in animal studies [5]. Unfortunately, we did not have any recorded intraoperative TEE images to assess RV function and size during this time, nor was there a report detailing observed changes in ventricular parameters during pericardial manipulation. The lack of records likely reflects sudden and dramatic cardiac collapse that necessitated emergent cannulation and CPB. To our knowledge, instability after a stable induction period has not been previously reported and may represent an additional time of heightened awareness, aggressive management of inotropic support, volume status, and vasopressor support beyond the initial induction GA. Acute hemodynamic collapse during pulmonary embolectomy due to occlusion of cannulas by intracardiac thrombi has been reported [6], but this was during initiation of CPB and not during pericardial manipulation. There have been reports of patients
Hemodynamic instability in patients
5 Stable after Pericardiotomy
Stable after Pericardiotomy 150 140 35
CVP (mmHg)
HR (bpm)
130 120 110 100
25
90 80 70
15 Initial HR
Initial CVP
Pericardiotomy
Pericardiotomy
Unstable after Pericardiotomy
Unstable after Pericardiotomy 120
110
CVP (mmHg)
HR (bpm)
35
100
25
90
15
80 Initial HR
Pericardiotomy
Initial CVP
Pericardiotomy
Figure Heart rate and CVP changes in the stable and delayed-unstable group at baseline (poststernotomy) and after pericardiotomy. Heart rate decreased, and CVP increased significantly in the delayed-unstable group but not in the stable group.
developing acute dilation of the RV with resultant cardiogenic shock during pericardiocentesis [7,8] as well as pericardiectomy for chronic pericarditis [9]. The authors attributed the hemodynamic decompensation to a sudden increase in venous return resulting in myocardial ischemia and RV dilation. Furthermore, an animal study by Page et al [10] described a decreased ability of the RV to handle increasing pressure work during pericardial opening despite maintenance of right coronary perfusion pressure. The authors theorized that potential alterations in LV-RV interactions with an open pericardium could lead to acute RV failure. Alterations in septal geometry related to decreased LV filling have been demonstrated to worsen RV performance [11], which may occur due to the fixed right-sided output in patients with large thrombus burden. In our study, GA induction was performed with fentanyl, ketamine, and etomidate being preferred. Although significant hemodynamic instability could not be associated with 1 particular induction agent, there was a trend toward larger induction dosages of all drugs in patients who developed hemodynamic collapse after induction. The anesthesiologist should carefully titrate medications during induction to avoid abrupt changes in compensatory measures, and the induction-unstable group may reflect an overly aggressive induction
strategy. Etomidate and ketamine are frequently chosen in our institution for induction of anesthesia. Despite concerns with etomidate in critically ill, septic patients [12,13], a recent publication of cardiac surgical patients did not demonstrate worse outcomes [14] and is often considered the “drug of choice” for anesthetic induction in unstable patients. In our institution, ketamine has gained popularity with increasing experience with pulmonary embolectomy surgery and is often preferred over etomidate. Despite the theoretical concerns for elevation in PVR with ketamine, several studies have failed to demonstrate an increases in PVR with ketamine, [15,16] and, in an animal model, ketamine was shown to actually reduce PVR, when sympathetic tone was increased [17]. Patients undergoing pulmonary embolectomy are at an increased risk for acute RV failure during induction of anesthesia as well as pericardial manipulation/opening as seen in our study. Because of the increased RV pressure work combined with a relatively fixed cardiac output from pulmonary artery obstruction, RV ischemia from hypotension or sudden increases in venous return can lead to hemodynamic collapse. Regardless of the mechanism, these periods of instability indicate that the anesthesiologist needs to remain acutely vigilant with judicious fluid
6 administration, regular assessment of RV function on TEE and evaluation of filling pressures, and maintenance of coronary perfusion pressures to reduce RV ischemia. Use of inotropic support may be indicated to support these goals as has been previously reported [18-20], although we did not see any benefit in our review. Our institution continues to perform pulmonary embolectomy on a semiroutine basis, when thrombolytics or other therapies are not considered beneficial (~ 10 surgeries per year). Patients in both unstable groups occurred mostly during the first 2 years, when we started performing pulmonary embolectomy. This likely indicates a change in anesthetic as well as surgical practice, when approaching these patients. Ketamine has become more regularly used for induction with additional benefits of analgesia and amnesia. This may help reduce the fentanyl dose delivered, which reduces potential bradycardia that can lead to decompensation. Additional changes in anesthetic management are provider specific and hard to detail in this retrospective review. Pulmonary vasodilator therapy with inhaled epoprostenol has become more frequent but only recently and did not represent a change in practice during this study period. Although we did not have extracorporeal life support (ECLS) available in our institution during most of the study period (limited to pediatric patients with rare adult use for acute respiratory dysfunction), there are recent studies using ECLS in patients with cardiac arrest or severe cardiogenic shock [21-23]. Although the in-hospital mortality in the patients receiving ECLS ranged from 30%-50% this may be a potential life-saving treatment for patients who otherwise may not survive pulmonary embolectomy. Limitations of this study include the retrospective nature, single institution, limited records for echocardiographic parameters, and small sample size. The intraoperative echocardiographic reports were not fully inclusive, and assessment of RV function was made semiquantitatively. In addition, the anesthetic records, especially in acutely ill patients who are rapidly decompensating, often do not correctly reflect the drug(s) used or infusions being administered at critical time points.
5. Conclusion In patients undergoing surgery for pulmonary embolism, the anesthesiologist needs to remain acutely vigilant for hemodynamic collapse, which may occur beyond induction of GA. Specifically, management of cardiac function and blood pressure with judicious use of inotropic and vasopressor infusions while optimizing ventilation to reduce intrathoracic pressures is critical. Patients may develop acute hemodynamic compromise at any time. In the current study, patients had increased instability during 2 periods: at the time of GA induction and during pericardial opening. Because of the acute nature and severity of these patients, a plan for emergent
J.M. Bennett et al. cannulation for initiation of CPB should be formulated between surgical, anesthetic, and perfusion teams. Having the patient prepared, draped, and a surgeon scrubbed in with CPB circuitry immediately available is highly advisable. Infusions of inotropes should be considered for all patients, particularly if the RV function is depressed.
References [1] Leacche M, Unic D, Goldhaber SZ, et al. Modern surgical treatment of massive pulmonary embolism: results in 47 consecutive patients after rapid diagnosis and aggressive surgical approach. J Thorac Cardiovasc Surg 2005;129:1018-23. [2] Carvalho EM, Macedo FI, Panos AL, Ricci M, Salerno TA. Pulmonary embolectomy: recommendation for early surgical intervention. J Card Surg 2010;25:261-6. [3] Rosenberger P, Shernan SK, Shekar PS, et al. Acute hemodynamic collapse after induction of general anesthesia for emergent pulmonary embolectomy. Anesth Analg 2006;102:1311-5. [4] Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18: 1440-63. [5] Haraldsen P, Lindstedt S, Metzsch C, Algotsson L, Ingemansson R. A porcine model for acute ischaemic right ventricular dysfunction. Interact Cardiovasc Thorac Surg 2014;18:43-8. [6] Augoustides JG, Plappert T, Bavaria JE. Hemodynamic collapse during pulmonary embolectomy due to loss of venous return from acute occlusion of the cardiopulmonary venous cannula with thromboembolus. Interact Cardiovasc Thorac Surg 2008;7:661-2. [7] Armstrong WF, Feigenbaum H, Dillon JC. Acute right ventricular dilation and echocardiographic volume overload following pericardiocentesis for relief of cardiac tamponade. Am Heart J 1984;107: 1266-70. [8] Anguera I, Pare C, Perez-Villa F. Severe right ventricular dysfunction following pericardiocentesis for cardiac tamponade. Int J Cardiol 1997;59:212-4. [9] Yu HT, Ha JW, Lee S, et al. Transient right ventricular dysfunction after pericardiectomy in patients with constrictive pericarditis. Korean Circ J 2011;41:283-6. [10] Page RD, Harringer W, Hodakowski GT, et al. Determinants of maximal right ventricular function. J Heart Lung Transplant 1992;11: 90-8. [11] Klima UP, Lee MY, Guerrero JL, Laraia PJ, Levine RA, Vlahakes GJ. Determinants of maximal right ventricular function: role of septal shift. J Thorac Cardiovasc Surg 2002;123:72-80. [12] Chan CM, Mitchell AL, Shorr AF. Etomidate is associated with mortality and adrenal insufficiency in sepsis: a meta-analysis*. Crit Care Med 2012;40:2945-53. [13] Komatsu R, You J, Mascha EJ, Sessler DI, Kasuya Y, Turan A. Anesthetic induction with etomidate, rather than propofol, is associated with increased 30-day mortality and cardiovascular morbidity after noncardiac surgery. Anesth Analg 2013;117:1329-37. [14] Wagner CE, Bick JS, Johnson D, et al. Etomidate use and postoperative outcomes among cardiac surgery patients. Anesthesiology 2013. [15] Basagan-Mogol E, Goren S, Korfali G, Turker G, Kaya FN. Induction of anesthesia in coronary artery bypass graft surgery: the hemodynamic and analgesic effects of ketamine. Clinics (Sao Paulo, Brazil) 2010;65:133-8.
Hemodynamic instability in patients [16] Oklu E, Bulutcu FS, Yalcin Y, Ozbek U, Cakali E, Bayindir O. Which anesthetic agent alters the hemodynamic status during pediatric catheterization? Comparison of propofol versus ketamine. J Cardiothorac Vasc Anesth 2003;17:686-90. [17] Kaye AD, Banister RE, Fox CJ, Ibrahim IN, Nossaman BD. Analysis of ketamine responses in the pulmonary vascular bed of the cat. Crit Care Med 2000;28:1077-82. [18] Hawn MT, Graham LA, Richman JS, Itani KM, Henderson WG, Maddox TM. Risk of major adverse cardiac events following noncardiac surgery in patients with coronary stents. JAMA 2013; 310:1462-72. [19] Arrowsmith JE, Dunning JJ, Gray SJ, Large SR, Tsui SL. Anesthesia for ventricular assist device placement. J Cardiothorac Vasc Anesth 2001;15:274-5.
7 [20] Kocabas S, Askar FZ, Yagdi T, Engin C, Ozbaran M. Anesthesia for ventricular assist device placement: experience from a single center. Transplant Proc 2013;45:1005-8. [21] Cho YH, Kim WS, Sung K, et al. Management of cardiac arrest caused by acute massive pulmonary thromboembolism: importance of percutaneous cardiopulmonary support. ASAIO J 2014;60:280-3. [22] Wu MY, Liu YC, Tseng YH, Chang YS, Lin PJ, Wu TI. Pulmonary embolectomy in high-risk acute pulmonary embolism: the effectiveness of a comprehensive therapeutic algorithm including extracorporeal life support. Resuscitation 2013;84:1365-70. [23] Taniguchi S, Fukuda W, Fukuda I, et al. Outcome of pulmonary embolectomy for acute pulmonary thromboembolism: analysis of 32 patients from a multicentre registry in Japan. Interact Cardiovasc Thorac Surg 2012;14:64-7.
Original contribution
Hemodynamic instability in patients undergoing pulmonary embolectomy: institutional experience Jeremy M. Bennett MD (Assistant Professor)a,⁎, Mias Pretorius Mb, Chb, MSCI (Associate Professor)a , Rashid M. Ahmad MD (Assistant Professor)b , Susan S. Eagle MD (Associate Professor)a a
Department of Anesthesiology Vanderbilt University Medical Center, Nashville, TN 37232 Department of Cardiac Surgical Sciences Vanderbilt University Medical Center, Nashville, TN 37232
b
Received 17 April 2014; revised 12 August 2014; accepted 29 October 2014
Keywords: Pulmonary embolectomy; Cardiac surgery; Pericardiotomy; Hemodynamic instability; Right ventricle; Right ventricular failure; Emergent surgery; Emergent cardiac surgery
Abstract Objectives: Acute pulmonary embolism is a major cause of morbidity and mortality in patients presenting for emergent cardiac surgery with overall mortality ranging from 6% to as high as 85%. While the initial focus of treatment is nonsurgical or percutaneous interventions, surgical treatment continues to be a treatment for patients with refractory thrombus burden or cardiogenic shock. Our institution regularly performs surgical pulmonary embolectomy with improved outcomes compared to current reports. We thus performed a retrospective analysis of outcomes of pulmonary embolectomy patients and anesthetic management. Design: A retrospective review of 40 patients undergoing emergent pulmonary embolectomy over a 4 year period (2008-2012) at our institution was performed to assess for a 2nd period of critical instability. Setting: The study was conducted at a tertiary, level 1, trauma university medical center. Participants: The study was performed through chart review of patient hospital records. Interventions: No interventions were performed. MeasurementsAnesthetic records were reviewed along with echocardiographic records and surgical reports to assess cardiac function, need for emergent cardiopulmonary bypass, and degree of patient morbidity. Conclusions: A total of 40 patients were studied. Hemodynamic instability occurred in 12.5% of patients at time of induction requiring emergent cardiopulmonary bypass. Another 17% of patients who remained stable following induction developed subsequent instability requiring emergent cardiopulmonary bypass during pericardial opening or manipulation which has not been previously reported. One patient died during hospitalization. Patients who required emergent bypass following induction of general anesthesia tended to receive higher doses of induction drugs than the other groups. In patients who needed emergent bypass during pericardial manipulation there were no identifiable factors suggesting that these patients remain at risk despite a stable post-induction course. © 2014 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Tel.: +1 615 322 4650. E-mail address: [email protected] (J.M. Bennett). http://dx.doi.org/10.1016/j.jclinane.2014.10.007 0952-8180/© 2014 Elsevier Inc. All rights reserved.
2
1. Introduction Anesthetic management of patients undergoing pulmonary embolectomy is challenging given the emergent nature of the procedure, acute hemodynamic instability, patient comorbidities, and experience in various centers. Surgical management of pulmonary embolism is most often considered in patients with central or branch pulmonary artery thrombus burden with cardiogenic shock. It may also be elected for patients with stable hemodynamics who demonstrate right ventricular failure on echocardiographic examination or have failed medical therapy with thrombolytics, heparin, or percutaneous embolectomy [1,2]. A recent case series of 52 patients demonstrated a 19% incidence of hemodynamic collapse during induction of general anesthesia (GA) necessitating emergent cardiopulmonary bypass (CPB) [3]. Further studies have demonstrated overall mortality from surgical intervention ranging from 6%-85% [1,2]. Given the significant morbidity of these patients, surgical manipulation and stress are likely to impact venous return, pulmonary vascular resistance (PVR), and cardiac output of these critically ill patients. Because of paucity of anesthetic-related outcomes studies for this critical patient population, we performed a retrospective study over a 4-year period of patients presenting for pulmonary embolectomy. The aim of this study is to assess the prevalence of hemodynamic instability during induction of GA as well as at any time before institution of CPB with attention toward anesthetic technique.
2. Methods A retrospective analysis was performed on all patients who underwent urgent/emergent pulmonary embolectomy at Vanderbilt University Medical Center between January 2008 and December 2012. A waiver of consent was obtained from the Vanderbilt Human Research Protection Program for assessment of patient records. Patients who arrived to the operating room (OR) endotracheally intubated or having received active cardiopulmonary resuscitation at any time before arrival in the OR were excluded from the study. Patient's medical records were reviewed for patient demographic information including but not limited to age, sex, right ventricle (RV) and left ventricle (LV) function, comorbidities (chronic lung disease, history of smoking, cardiac disease, renal insufficiency, and anemia), body mass index, prior cardiac surgery, and postoperative morbidity. Anesthetic records were reviewed for anesthetic induction agent (etomidate, fentanyl, ketamine, and propofol), inotropic administration, ventilator parameters (ventilator mode, peak airway pressures (PAWs), and positive end-expiratory pressure), and hemodynamic changes in heart rate (HR) and central venous pressure (CVP). Preoperative radiologic and transthoracic echocardiographic information was reviewed,
J.M. Bennett et al. if available as well as all intraoperative transesophageal echocardiographic (TEE) findings to assess thrombus location and burden, RV function, and RV dilation. Right ventricular function and dilation were determined based on accepted American Society of Echocardiography standards for assessment of the RV [4]. Preoperative transthoracic echocardiographic data were available in all patients with the exception of 2 patients in the stable group and 1 patient in both the induction-unstable and delayed-unstable groups (total of 4 patients). All patients received intraarterial catheters and central venous catheters before induction of GA using local anesthesia, supplemental oxygen as supplied by a facemask, and titration of sedation as dictated by the attending anesthesiologist. The location of these catheters was in the radial artery and right internal jugular vein, respectively, for most patients. If radial arterial catheter placement was difficult or unable to be performed, a femoral arterial catheter was placed. Sedation most commonly consisted of boluses of midazolam and ketamine. After central venous catheter placement, pulmonary artery catheters were placed in all patients and advanced to 20-cm depth until separation from CPB, at which time the pulmonary artery catheter was advanced into the main pulmonary artery. All patients were prepared and surgically draped with a surgeon scrubbed and ready in the OR before induction of GA. The type of anesthetic induction drug was at the discretion of the attending anesthesiologist. Transesophageal echocardiographic was performed in all patients after tracheal intubation, where there was no concern for esophageal pathology. Hemodynamic instability requiring emergent CPB was identified as acute and persistent hypotension occurring after GA induction that did not respond to vasopressor support and required emergent initiation of CPB. If patients remained stable after induction of GA, we assessed for any other period, when hemodynamic instability developed before CPB initiation. Surgeon and anesthesia records were reviewed for all patients if instability was identified for potential description of the event. Data are presented as median with interquartile range unless otherwise indicated. Categorical data were compared between groups using χ2 or Fisher exact tests as appropriate. Continuous baseline data were compared using the KruskalWallis test. Comparisons of CVP and HR among groups were made with the Wilcoxon signed rank test. A 2-tailed P value less than .05 was considered statistically significant. Statistical analyses were performed with the statistical package SPSS for Windows (version 21.0; IBM, New York, NY).
3. Results The study population consisted of 46 patients of whom 6 patients were excluded for preoperative cardiopulmonary resuscitation and/or arriving tracheally intubated before the OR. A total of 40 patients were included in the study and
Hemodynamic instability in patients Table 1
3
Preoperative patient characteristics
Characteristic
Induction unstable (n = 5)
Delayed unstable (n = 6)
Stable (n = 29)
P
Age, y BMI, kg/m2 Female sex, n (%) Diabetes, n (%) Current smoker, n (%) History of COPD, n (%) History of MI, n (%) Hypertension, n (%) Creatinine (mg/dL) Hematocrit (%) Preoperative LVEF, % Preoperative RV function, n (%) Normal-to-mild depression Moderate depression Severe depression RV dilation, n (%) Normal-to-mild dilation Moderate dilation Severe dilation Medications, n (%) Inotropes before induction ACEi or ARB β-blockers Coumadin Heparin
45 (37-60) 27 (21-39) 4 (80) 2 (40) 2 (40) 0 1 (20) 2 (40) 0.88 (0.72-2.05) 39 (34-42) 55 (43-55)
48 (23-75) 34 (26-41) 3 (50) 1 (17) 0 0 1 (17) 3 (50) 1.14 (0.90-1.59) 42 (35-44) 55 (54-58)
58 30 9 8 2 2 4 13 1.14 41 55
.19 .45 .10 .69 .05 .67 .93 .95 .44 .64 .29 .85
(52-68) (28-38) (31) (28) (7) (7) (14) (45) (0.97-1.37) (36-43) (55-55)
2 (40) 2 (40) 0
2 (33) 2 (33) 1 (17)
8 (28) 16 (55) 3 (10)
1 (20) 1 (20) 2 (40)
3 (50) 1 (17) 1 (17)
7 (24) 10 (35) 10 (35)
2 (40) 1 (20) 1 (20) 0 1 (20)
0 1 (17) 1 (17) 0 2 (33)
3 6 1 1 10
.76
(10) (21) (3.4) (3) (35)
.11 .98 .28 .82 .82
Abbreviations: BMI, body mass index; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; LVEF, left ventricular ejection fraction; ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker.
divided into 3 groups based on hemodynamic status: stable group, unstable at induction group (induction unstable), and the delayed-unstable group. Five patients (12.5%) developed refractory hemodynamic collapse requiring emergent CPB immediately after anesthesia induction and intubation (induction-unstable group). Six patients (15%) developed hypotension that was refractory to vasopressor or inotropic support after a stable induction of GA (delayed-unstable group). One patient died in the induction-unstable group during their hospitalization. No other deaths occurred in any of the other groups. The surgeon operative report and/or anesthetic records for the 6 patients in the delayed-unstable group indicated that instability occurred during pericardial sac manipulation (either pericardiotomy or during retention suture placement). Twenty-nine patients (72.5%) remained hemodynamically stable throughout induction and up to initiation of CPB (stable group). Preoperative patient characteristics for the 3 groups are presented in Table 1. Female sex was more frequent in the unstable postinduction group, although this was not significant (P = .10). Preoperative RV function and dilation did not differ significantly among the groups. Table 2 details intraoperative patient characteristics. There were no significant differences noted in the choice
of induction drug or dosage administered, although there was a trend toward lower induction drug dosages in the stable and delayed-unstable groups compared with the induction-unstable group. As expected, the lowest mean arterial pressure (MAP) and decrease in systolic blood pressure greater than 40 mm Hg were significantly different among the groups with the lowest MAP seen in the induction-unstable group. Severe intraoperative RV dysfunction trended more frequently in the induction-unstable (60%) and delayed-unstable (50%) groups but was not significantly different from the stable group (28%). Peak airway pressure, tidal volume, and mode of ventilation were not significantly different among the 3 groups, although higher PAW tended in the induction-unstable group. Central venous pressure was significantly higher in the delayed-unstable group compared with the stable group after pericardial opening. The Figure details the change in HR and CVP from poststernotomy measurements to the time of pericardial opening in the stable and delayed-unstable groups. Neither HR (from a median baseline of 105-100 beats per minute, P = .92) nor CVP (from a median baseline of 22-22 mm Hg, P = .87) changed significantly in the stable group. In contrast, HR decreased (from a median baseline of 108-95 beats per minute, P = .039), and CVP increased (from a median baseline of 22-28
4
J.M. Bennett et al. Table 2
Intraoperative patient characteristics
Ketamine, n (%) Dose, mg/kg Etomidate, n (%) Dose, mg/kg Fentanyl, n (%) Dose, μg/kg Lowest MAP postinduction (mm Hg) SBP decrease N 40 mm Hg postinduction, n (%) HR at baseline, beats per minute HR after pericardial opening, beats per minute CVP baseline, mm Hg CVP after pericardial opening, mm Hg Severe RV dysfunction, n (%) Intraoperative pH Ventilator PAW, mm Hg Tidal volume, mL/kg Volume control ventilation, n (%)
Induction unstable (n = 5)
Delayed unstable (n = 6)
3 (60) 1.4 (0-2.4) 0 NA 4 (80) 4.3 (2.0-7.7) 47 (43-58)
3 (50) 0.4 (0-2.3) 3 (50) 0.04 (0-0.07) 5 (83) 3.5 (1.6-8.5) 66 (49-70)
5 (100) 125 (105-135) NA 25 (16-27) NA 3 (60) NA 43 (34-44) 6.9 (4.2-10.2) 3 (60)
2 (33)
Stable (n = 29)
P
18 (62) 0.6 (0-1.4) 9 (31) 0.1 (0.1-0.18) 22 (76) 2.9 (1.3-4.5) 72 (63-83)
.86 .83 .19 .31 .91 .52 .004
10 (35)
.02
108 (99-113) 95 (84-103)
105 (95-120) 100 (95-113)
.17 .13
22 (21-30) 28 (24-34)
22 (20-24) 22 (20-24)
.75 .01
3 (50) 7.33 (7.30-7.39) 33 (29-41) 5.9 (4.6-7.8) 5 (83)
8 (28) 7.36 (7.32-7.40) 34 (29-39) 5.8 (4.8-6.9) 27 (93)
.26 .36 .18 .72 .11
Abbreviation: SBP, systolic blood pressure; NA, not available.
mm Hg, P = .045) significantly in the delayed-unstable group, although 1 patient had a decrease in CVP and increase in HR in the delayed-unstable group.
4. Discussion Pulmonary embolectomy is a challenging operative procedure performed for patients with severe, massive PE. In-hospital morbidity and mortality levels can be high and may be significantly affected by anesthetic management with hemodynamic instability after GA induction and need for emergent CPB [3]. In addition, sternotomy, pericardial opening, and cardiac manipulation along with poor respiratory management (high PAWs, use of positive end-expiratory pressure, etc) may result in hemodynamic instability after GA induction. In our study, hemodynamic collapse necessitating emergent institution of CPB occurred in 13% of patients after induction as has been previously described [3]. The PAW in these patients tended to be higher than the other 2 groups. Although we are not able to elaborate on why the PAW tended toward higher values, we suspect this may reflect the acute decompensation during GA induction resulting in reduced ventilator vigilance. Our current ventilators have preset 600 mL tidal volumes settings with a respiratory rate of 8 and zero positive end-expiratory pressure. If the tidal volumes were not adjusted while emergently instituting CPB, this may cause substantial
increase in PAW, when not adjusted for patient ideal body weight and could reflect the observed trend. In patients who remained stable after anesthetic induction, an additional 15% of patients developed acute hemodynamic decompensation that did not seem to be related to the induction period. In particular, review of surgeon and anesthesiologist documentation and anesthetic records seemed to indicate that acute decompensation occurred during pericardial manipulation/opening. Of interest is the finding that, in 5 of the 6 patients who developed acute decompensation during pericardial manipulation, a significant increase in CVP and decrease in HR occurred. This likely reflects acute RV failure, possibly from ischemia as has been demonstrated recently in animal studies [5]. Unfortunately, we did not have any recorded intraoperative TEE images to assess RV function and size during this time, nor was there a report detailing observed changes in ventricular parameters during pericardial manipulation. The lack of records likely reflects sudden and dramatic cardiac collapse that necessitated emergent cannulation and CPB. To our knowledge, instability after a stable induction period has not been previously reported and may represent an additional time of heightened awareness, aggressive management of inotropic support, volume status, and vasopressor support beyond the initial induction GA. Acute hemodynamic collapse during pulmonary embolectomy due to occlusion of cannulas by intracardiac thrombi has been reported [6], but this was during initiation of CPB and not during pericardial manipulation. There have been reports of patients
Hemodynamic instability in patients
5 Stable after Pericardiotomy
Stable after Pericardiotomy 150 140 35
CVP (mmHg)
HR (bpm)
130 120 110 100
25
90 80 70
15 Initial HR
Initial CVP
Pericardiotomy
Pericardiotomy
Unstable after Pericardiotomy
Unstable after Pericardiotomy 120
110
CVP (mmHg)
HR (bpm)
35
100
25
90
15
80 Initial HR
Pericardiotomy
Initial CVP
Pericardiotomy
Figure Heart rate and CVP changes in the stable and delayed-unstable group at baseline (poststernotomy) and after pericardiotomy. Heart rate decreased, and CVP increased significantly in the delayed-unstable group but not in the stable group.
developing acute dilation of the RV with resultant cardiogenic shock during pericardiocentesis [7,8] as well as pericardiectomy for chronic pericarditis [9]. The authors attributed the hemodynamic decompensation to a sudden increase in venous return resulting in myocardial ischemia and RV dilation. Furthermore, an animal study by Page et al [10] described a decreased ability of the RV to handle increasing pressure work during pericardial opening despite maintenance of right coronary perfusion pressure. The authors theorized that potential alterations in LV-RV interactions with an open pericardium could lead to acute RV failure. Alterations in septal geometry related to decreased LV filling have been demonstrated to worsen RV performance [11], which may occur due to the fixed right-sided output in patients with large thrombus burden. In our study, GA induction was performed with fentanyl, ketamine, and etomidate being preferred. Although significant hemodynamic instability could not be associated with 1 particular induction agent, there was a trend toward larger induction dosages of all drugs in patients who developed hemodynamic collapse after induction. The anesthesiologist should carefully titrate medications during induction to avoid abrupt changes in compensatory measures, and the induction-unstable group may reflect an overly aggressive induction
strategy. Etomidate and ketamine are frequently chosen in our institution for induction of anesthesia. Despite concerns with etomidate in critically ill, septic patients [12,13], a recent publication of cardiac surgical patients did not demonstrate worse outcomes [14] and is often considered the “drug of choice” for anesthetic induction in unstable patients. In our institution, ketamine has gained popularity with increasing experience with pulmonary embolectomy surgery and is often preferred over etomidate. Despite the theoretical concerns for elevation in PVR with ketamine, several studies have failed to demonstrate an increases in PVR with ketamine, [15,16] and, in an animal model, ketamine was shown to actually reduce PVR, when sympathetic tone was increased [17]. Patients undergoing pulmonary embolectomy are at an increased risk for acute RV failure during induction of anesthesia as well as pericardial manipulation/opening as seen in our study. Because of the increased RV pressure work combined with a relatively fixed cardiac output from pulmonary artery obstruction, RV ischemia from hypotension or sudden increases in venous return can lead to hemodynamic collapse. Regardless of the mechanism, these periods of instability indicate that the anesthesiologist needs to remain acutely vigilant with judicious fluid
6 administration, regular assessment of RV function on TEE and evaluation of filling pressures, and maintenance of coronary perfusion pressures to reduce RV ischemia. Use of inotropic support may be indicated to support these goals as has been previously reported [18-20], although we did not see any benefit in our review. Our institution continues to perform pulmonary embolectomy on a semiroutine basis, when thrombolytics or other therapies are not considered beneficial (~ 10 surgeries per year). Patients in both unstable groups occurred mostly during the first 2 years, when we started performing pulmonary embolectomy. This likely indicates a change in anesthetic as well as surgical practice, when approaching these patients. Ketamine has become more regularly used for induction with additional benefits of analgesia and amnesia. This may help reduce the fentanyl dose delivered, which reduces potential bradycardia that can lead to decompensation. Additional changes in anesthetic management are provider specific and hard to detail in this retrospective review. Pulmonary vasodilator therapy with inhaled epoprostenol has become more frequent but only recently and did not represent a change in practice during this study period. Although we did not have extracorporeal life support (ECLS) available in our institution during most of the study period (limited to pediatric patients with rare adult use for acute respiratory dysfunction), there are recent studies using ECLS in patients with cardiac arrest or severe cardiogenic shock [21-23]. Although the in-hospital mortality in the patients receiving ECLS ranged from 30%-50% this may be a potential life-saving treatment for patients who otherwise may not survive pulmonary embolectomy. Limitations of this study include the retrospective nature, single institution, limited records for echocardiographic parameters, and small sample size. The intraoperative echocardiographic reports were not fully inclusive, and assessment of RV function was made semiquantitatively. In addition, the anesthetic records, especially in acutely ill patients who are rapidly decompensating, often do not correctly reflect the drug(s) used or infusions being administered at critical time points.
5. Conclusion In patients undergoing surgery for pulmonary embolism, the anesthesiologist needs to remain acutely vigilant for hemodynamic collapse, which may occur beyond induction of GA. Specifically, management of cardiac function and blood pressure with judicious use of inotropic and vasopressor infusions while optimizing ventilation to reduce intrathoracic pressures is critical. Patients may develop acute hemodynamic compromise at any time. In the current study, patients had increased instability during 2 periods: at the time of GA induction and during pericardial opening. Because of the acute nature and severity of these patients, a plan for emergent
J.M. Bennett et al. cannulation for initiation of CPB should be formulated between surgical, anesthetic, and perfusion teams. Having the patient prepared, draped, and a surgeon scrubbed in with CPB circuitry immediately available is highly advisable. Infusions of inotropes should be considered for all patients, particularly if the RV function is depressed.
References [1] Leacche M, Unic D, Goldhaber SZ, et al. Modern surgical treatment of massive pulmonary embolism: results in 47 consecutive patients after rapid diagnosis and aggressive surgical approach. J Thorac Cardiovasc Surg 2005;129:1018-23. [2] Carvalho EM, Macedo FI, Panos AL, Ricci M, Salerno TA. Pulmonary embolectomy: recommendation for early surgical intervention. J Card Surg 2010;25:261-6. [3] Rosenberger P, Shernan SK, Shekar PS, et al. Acute hemodynamic collapse after induction of general anesthesia for emergent pulmonary embolectomy. Anesth Analg 2006;102:1311-5. [4] Lang RM, Bierig M, Devereux RB, et al. Recommendations for chamber quantification: a report from the American Society of Echocardiography's Guidelines and Standards Committee and the Chamber Quantification Writing Group, developed in conjunction with the European Association of Echocardiography, a branch of the European Society of Cardiology. J Am Soc Echocardiogr 2005;18: 1440-63. [5] Haraldsen P, Lindstedt S, Metzsch C, Algotsson L, Ingemansson R. A porcine model for acute ischaemic right ventricular dysfunction. Interact Cardiovasc Thorac Surg 2014;18:43-8. [6] Augoustides JG, Plappert T, Bavaria JE. Hemodynamic collapse during pulmonary embolectomy due to loss of venous return from acute occlusion of the cardiopulmonary venous cannula with thromboembolus. Interact Cardiovasc Thorac Surg 2008;7:661-2. [7] Armstrong WF, Feigenbaum H, Dillon JC. Acute right ventricular dilation and echocardiographic volume overload following pericardiocentesis for relief of cardiac tamponade. Am Heart J 1984;107: 1266-70. [8] Anguera I, Pare C, Perez-Villa F. Severe right ventricular dysfunction following pericardiocentesis for cardiac tamponade. Int J Cardiol 1997;59:212-4. [9] Yu HT, Ha JW, Lee S, et al. Transient right ventricular dysfunction after pericardiectomy in patients with constrictive pericarditis. Korean Circ J 2011;41:283-6. [10] Page RD, Harringer W, Hodakowski GT, et al. Determinants of maximal right ventricular function. J Heart Lung Transplant 1992;11: 90-8. [11] Klima UP, Lee MY, Guerrero JL, Laraia PJ, Levine RA, Vlahakes GJ. Determinants of maximal right ventricular function: role of septal shift. J Thorac Cardiovasc Surg 2002;123:72-80. [12] Chan CM, Mitchell AL, Shorr AF. Etomidate is associated with mortality and adrenal insufficiency in sepsis: a meta-analysis*. Crit Care Med 2012;40:2945-53. [13] Komatsu R, You J, Mascha EJ, Sessler DI, Kasuya Y, Turan A. Anesthetic induction with etomidate, rather than propofol, is associated with increased 30-day mortality and cardiovascular morbidity after noncardiac surgery. Anesth Analg 2013;117:1329-37. [14] Wagner CE, Bick JS, Johnson D, et al. Etomidate use and postoperative outcomes among cardiac surgery patients. Anesthesiology 2013. [15] Basagan-Mogol E, Goren S, Korfali G, Turker G, Kaya FN. Induction of anesthesia in coronary artery bypass graft surgery: the hemodynamic and analgesic effects of ketamine. Clinics (Sao Paulo, Brazil) 2010;65:133-8.
Hemodynamic instability in patients [16] Oklu E, Bulutcu FS, Yalcin Y, Ozbek U, Cakali E, Bayindir O. Which anesthetic agent alters the hemodynamic status during pediatric catheterization? Comparison of propofol versus ketamine. J Cardiothorac Vasc Anesth 2003;17:686-90. [17] Kaye AD, Banister RE, Fox CJ, Ibrahim IN, Nossaman BD. Analysis of ketamine responses in the pulmonary vascular bed of the cat. Crit Care Med 2000;28:1077-82. [18] Hawn MT, Graham LA, Richman JS, Itani KM, Henderson WG, Maddox TM. Risk of major adverse cardiac events following noncardiac surgery in patients with coronary stents. JAMA 2013; 310:1462-72. [19] Arrowsmith JE, Dunning JJ, Gray SJ, Large SR, Tsui SL. Anesthesia for ventricular assist device placement. J Cardiothorac Vasc Anesth 2001;15:274-5.
7 [20] Kocabas S, Askar FZ, Yagdi T, Engin C, Ozbaran M. Anesthesia for ventricular assist device placement: experience from a single center. Transplant Proc 2013;45:1005-8. [21] Cho YH, Kim WS, Sung K, et al. Management of cardiac arrest caused by acute massive pulmonary thromboembolism: importance of percutaneous cardiopulmonary support. ASAIO J 2014;60:280-3. [22] Wu MY, Liu YC, Tseng YH, Chang YS, Lin PJ, Wu TI. Pulmonary embolectomy in high-risk acute pulmonary embolism: the effectiveness of a comprehensive therapeutic algorithm including extracorporeal life support. Resuscitation 2013;84:1365-70. [23] Taniguchi S, Fukuda W, Fukuda I, et al. Outcome of pulmonary embolectomy for acute pulmonary thromboembolism: analysis of 32 patients from a multicentre registry in Japan. Interact Cardiovasc Thorac Surg 2012;14:64-7.