Anaesthesia 2014, 69, 558–572

doi:10.1111/anae.12604

Original Article Analysis of transthoracic echocardiographic data in major vascular surgery from a prospective randomised trial comparing sevoflurane and fentanyl with propofol and remifentanil anaesthesia* E. E. Lindholm,1 E. Aune,2 G. Frøland,2 K. A. Kirkebøen3,4 and J. E. Otterstad2 1 Consultant, Department of Anaesthesiology, 2 Consultant, Department of Cardiology, Vestfold Hospital Trust, Tønsberg, Norway 3 Clinical Professor, Department of Anaesthesiology, Oslo University Hospital, Oslo, Norway 4 Professor, Faculty of Medicine, University of Oslo, Oslo, Norway

Summary The aim of this study was to define pre-operative echocardiographic data and explore if postoperative indices of cardiac function after open abdominal aortic surgery were affected by the anaesthetic regimen. We hypothesised that volatile anaesthesia would improve indices of cardiac function compared with total intravenous anaesthesia. Transthoracic echocardiography was performed pre-operatively in 78 patients randomly assigned to volatile anaesthesia and 76 to total intravenous anaesthesia, and compared with postoperative data. Pre-operatively, 16 patients (10%) had left ventricular ejection fraction < 46%. In 138 patients with normal left ventricular ejection fraction, 5/8 (62%) with left ventricular dilatation and 41/130 (33%) without left ventricular dilatation had evidence of left ventricular diastolic dysfunction (p < 0.001). Compared with pre-operative findings, significant increases in left ventricular enddiastolic volume, left atrial maximal volume, cardiac output, velocity of early mitral flow and early myocardial relaxation occurred postoperatively (all p < 0.001). The ratio of the velocity of early mitral flow to early myocardial relaxation remained unchanged. There were no significant differences in postoperative echocardiographic findings between patients anaesthetised with volatile anaesthesia or total intravenous anaesthesia. Patients had an iatrogenic surplus of approximately 4.1 l of fluid volume by the first postoperative day. N-terminal prohormone of brain natriuretic peptide increased on the first postoperative day (p < 0.001) and remained elevated after 30 days (p < 0.001) in both groups. Although postoperative echocardiographic alterations were most likely to be related to increased preload due to a substantial iatrogenic surplus of fluid, a component of peri-operative myocardial ischaemia cannot be excluded. Our hypothesis that volatile anaesthesia improved indices of cardiac function compared with total intravenous anaesthesia could not be verified. .................................................................................................................................................................

Correspondence to: E. E. Lindholm Email: [email protected] *Presented in part at the Association of Anaesthetists of Great Britain and Ireland Annual Congress, Dublin, Ireland, September 2013, and the Norwegian Annual Anaesthesiology Congress, Lillestrøm, Norway, October 2013. Accepted: 13 January 2014

558

© 2014 The Association of Anaesthetists of Great Britain and Ireland

Lindholm et al. | Echocardiographic variables in major vascular surgery

Introduction Major vascular surgery implies a high risk for myocardial ischaemia and infarction, due to surgical stress [1] and co-existing coronary and peripheral atherosclerotic disease [2, 3]. Recent meta-analyses have shown that volatile anaesthetics reduce myocardial infarction and mortality in cardiac surgery compared with total intravenous anaesthesia (TIVA) [4, 5]. Although such an effect has not been demonstrated in large randomised trials to date, the American College of Cardiology/ American Heart Association guidelines recommend volatile anaesthesia for non-cardiac major vascular surgery in haemodynamically stable patients at risk for peri-operative myocardial ischaemia [6]. We recently published data from a randomised controlled trial that did not demonstrate a reduction in troponin-T release using volatile anaesthesia compared with TIVA in patients undergoing major vascular surgery [7]. In this study, we analysed echocardiographic data collected during this trial to define baseline pre-operative indices and to explore if echocardiography could demonstrate any changes in cardiac function after open abdominal aortic surgery and whether these were affected by the anaesthetic regimen. We hypothesised that volatile anaesthesia would have favourable effects on postoperative cardiac function compared with TIVA in noncardiac surgery. We also analysed serum concentration of N-terminal prohormone of brain natriuretic peptide (N-terminal pro-BNP), as this is an important determinant in diagnosing heart failure [8]. It is also a marker of major adverse cardiac events and reflects stress-induced myocardial ischaemia in vascular surgical patients [9, 10].

Methods This prospective, randomised, single-centre trial was conducted at a central hospital in Norway according to the Declaration of Helsinki principles. Necessary health authorities, the institutional review board and Local Research Ethics Committee approved the protocol. Patients of ASA physical status 1–4 with abdominal aortic aneurysm or aortic arteriosclerosis obliterans accepted for elective open abdominal aortic surgery who gave written informed consent were included consecutively from February 2008 to February 2012. Exclusion criteria were: age < 18 years; pregnancy or © 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 558–572

breastfeeding; participation in another pharmaceutical study; benzodiazepines, antiepileptic drugs, alcohol or a2-agonist abuse; family history of malignant hyperthermia; known hypersensitivity to opioids, propofol or volatile anaesthetics; cardiac valvular disease requiring surgical repair before non-cardiac surgery; uncontrolled hypertension; unstable angina; myocardial infarction 30 days before inclusion; decompensated heart failure; serious psychiatric disease; and serious arrhythmias, either ventricular in origin or tachycardia > 100 beats.min
1. All patients had echocardiography performed pre- and postoperatively, as part of a predefined substudy of the ABSENT study [7]. Patients were randomly assigned to sevoflurane-based anaesthesia or TIVA (Fig. 1). They were subjected to pre-operative evaluation and accepted for surgery by a cardiologist not participating in the study. Stress tests were used to detect myocardial ischaemia. Three cardiologists, blinded for randomisation, performed echocardiography (JEO, EA, GF). The same cardiologist performed the pre- and postoperative examinations in each patient. N-terminal pro-BNP analysis and postoperative care were also blinded for randomisation. Protocols for medication, monitoring and type of anaesthesia have been reported previously [7]. Intravenous fluid, bleeding and urine output were recorded from induction of anaesthesia until 08.00 on the first postoperative day. Pre-operative transthoracic echocardiography was performed using a M4S 1.5–4.3 MHz probe (Vivid 7; General Electric, Horten, Norway) between 1 and 64 days before surgery. Patients had no evidence of cardiac events between the pre-operative echocardiogram and surgery. Postoperative studies were performed on the first or second day after surgery. All measurements were performed online, and a form was filled out for all echocardiographic variables pre- and postoperatively. In addition, all echocardiograms were stored as digital loops. To reflect daily practice, one heart cycle was used for all variables measured. In patients with atrial fibrillation, measurements were performed over an average of three cardiac cycles. In a previous reproducibility study in patients with sinus rythm, we found deviations from baseline echocardiographic recordings and their repeated video measurements ranged between
5% and +5% when evaluated 559

Anaesthesia 2014, 69, 558–572

Patient not treated

Lindholm et al. | Echocardiographic variables in major vascular surgery

Patients enrolled

Patients failed screening

n = 231

n = 31

Randomised to

Randomised to

sevoflurane and fentanyl

propofol and remifentanil

n = 100

n = 100

Randomised and treated

Randomised and treated

sevoflurane and fentanyl

propofol and remifentanil

n = 97

n = 96

n=3

No available cardiologist

Patient not treated n=4

n = 19

No available cardiologist n = 20

Complete echocardiographic examination pre- and postoperatively

Complete echocardiographic examination pre- and postoperatively

n = 78

n = 76

Figure 1 CONSORT-style flow diagram.

across recordings and both investigators [11]. Reproducibility using similar methodology has previously been shown to be acceptable [12, 13]. Left ventricular mass index was calculated according to the American Society of Echocardiography recommendations, using M-mode from the parasternal long axis view [14]. Two-dimensional recordings included measurements of left ventricular volumes and ejection fraction (biplane Simpson method) [11]. Cardiac output was calculated as the difference between left ventricular end-diastolic volume and left ventricular end-systolic volume (left ventricular stroke volume), multiplied by heart rate. Left atrial volumes were measured from the apical view with the biplane area-length method. A diagnosis of left atrial dilatation 560

was based on increased maximal volume. Normal ranges [12, 13, 15] for echocardiographic indices are given in Appendix S1. In accordance with recommendations for evaluation of left ventricular diastolic function [16], left atrial volume often reflects the cumulative effects of filling pressure over time. To reflect filling pressure, early left ventricular relaxation was assessed from the velocity of both passive mitral flow (E) derived from pulsedDoppler interrogation of flow at the level of the mitral  from valve tip and early left ventricular relaxation (E) tissue Doppler measurements of left ventricular early relaxation (average of septal and lateral basal left ven ratio. Left tricular wall values), expressed by the E/E ventricle filling pressures were defined as shown in © 2014 The Association of Anaesthetists of Great Britain and Ireland

Lindholm et al. | Echocardiographic variables in major vascular surgery

Appendix S2 according to consensus statements [17]. Mitral inflow pattern was also identified from the deceleration time and A-wave velocity driven by atrial contraction, including the E/A ratio [16]. To assess diastolic dysfunction, we used modified criteria from Paulus et al. [17] and our own reference data [12, 13, 15]. When assessing diastolic dysfunction, we did not study patients with left ventricular ejection fraction < 46% and aimed to assess diastolic function in patients with preserved left ventricular ejection fraction, comparing those with normal and dilated left ventricles. We anticipated technical difficulties with regard to adequate recordings of pulmonary venous flow on the first postoperative day [16]. Thus, we did not incorporate differences between duration of reverse pulmonary vein atrial flow and duration of mitral atrial wave flow in our measurements. Valvular disease was based on combined colour Doppler and Doppler recordings. Only moderate or greater stenosis and/or regurgitation were categorised as significant. Patients with severe valvular heart disease were evaluated for cardiac surgery and not included. Blood samples for analysis of N-terminal pro-BNP were obtained pre-operatively, on the first postoperative day at 08.00 h and 30 days after surgery. Heart failure based on N-terminal pro-BNP was diagnosed according to the 2008 European Society of Cardiology guidelines [8], as shown in Appendix S2. N-terminal pro-BNP was analysed using the immunometric technique (Vitros 5600; Ortho Clinical Diagnostics, Rochester, NY, USA) with a coefficient of variance of 7%. Diagnostic criteria and definitions for pre-operative coronary artery disease, stroke/transient ischaemic attack, peripheral artery disease, carotid artery stenosis, atherosclerotic disease, systolic and diastolic left ventricular dysfunction at baseline and postoperative myocardial infarction are given in Appendix S2 [17–20]. Normally distributed data were analysed using independent and paired sample t-tests. The Kolmogorov–Smirnov test was used to test normality. Non-continuous data were analysed using the Mann–Whitney U-test or the Wilcoxon signed rank test. Categorical data were analysed using Fisher’s exact test or Friedman’s test. Statistical analyses were performed in SPSS 17 (SPSS Inc., Chicago, IL, USA). Sample size calculations to detect differences in troponin-T, the primary © 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 558–572

endpoint in the original study, have been previously reported [7]. A p value of < 0.05 was considered to be statistically significant. A post-hoc power calculation for intergroup differences in left ventricular volumes and ejection fraction was performed to evaluate if this study was underpowered. As a basis for this post-hoc power analysis, we used a previous reproducibility study [11]. In that study, reproducibility for left ventricular volumes and ejection fraction were performed by the three-way ANOVA model using estimates of within-subject variance components for each factor, and coefficients of variance were used to estimate reproducibility. The total variability (standard deviation) within subjects for each variable was calculated as the square root of the sum of the within-subject variance components across the three factors (repeated echo recording, investigator and repeated video measurement). This standard deviation was used to calculate the least detectable individual increase/decrease by multiplying it by 1.28, corresponding to an upper 10% one-sided error of classification. Based on these calculations, differences in average changes in echocardiographic variables that could be detected with 80% statistical power during a randomised trial between two groups with 250 patients were 8.5 ml for left ventricular end-diastolic volume, 5.5 ml for left ventricular end-systolic volume and 2.3% for left ventricular ejection fraction [11]. These values are categorised as delta 1. Presuming identical standard deviations for changes in each variable in this study, delta 2 values were calculated as follows (Ingar Holme, personal communication): delta 2 ¼ delta 1 

p

ð250=75Þ

Results The results regarding outcomes have previously been published. In summary, there were no differences in troponin-T between patients who received volatile anaesthesia or TIVA [7]. Pre- and postoperative echocardiograms were performed in 154 of the total of 193 patients in the main study (78 randomised to volatile anaesthesia and 76 to TIVA, Fig. 1). The second echocardiographic examination was performed on the first postoperative day in 136 (88%) patients, and the second 561

Anaesthesia 2014, 69, 558–572

Lindholm et al. | Echocardiographic variables in major vascular surgery

postoperative day in 18 (12%) patients. Baseline characteristics were similar between patients with or without echocardiography performed (Table 1), except there Table 1 Characteristics of patients with pre- and postoperative echocardiograms (n = 154) compared with patients who had no such examinations (n = 39). Data values are number (proportion), mean (SD) or median (IQR [range]). Echocardiography (n = 154) Men Age; years Body surface area; m2 Statins Betablockers ACEI / A2RB Aspirin Warfarin Arteriosclerosis Coronary artery disease Angina pectoris History of acute myocardial infarction Previous CABG Previous PCI Cardiac failure Prior LVEF < 40% Combined TIA and stroke Carotid artery stenosis Intermittent claudication Diabetes mellitus Hypertension History of SVT or AF History of ventricular arrhythmia Creatinine; lmol.l
1 Systolic BP; mmHg Diastolic BP; mmHg Heart rate; beats.min
1

No echocardiography (n = 39)

110 (71%) 68 (9) 1.9 (0.2)

35 (90%) 68 (8) 2.0 (0.2)

110 70 62 103 9 103 57

27 14 10 22 3 21 13

(71%) (46%) (40%) (67%) (6%) (67%) (37%)

25 (16%) 44 (29%)

23 24 6 7 16

(15%) (16%) (4%) (5%) (10%)

(69%) (36%) (26%) (56%) (8%) (54%) (33%)

8 (21%) 11 (28%)

4 4 3 0 3

(10%) (10%) (8%) (0%) (8%)

4 (3%)

2 (5%)

55 (36%)

8 (21%)

15 (10%) 90 (58%) 20 (13%)

2 (5%) 21 (54%) 6 (15%)

6 (4%)

1 (3%)

76 (66–90 [43–218]) 153 (26)

84 (70–102 [56–538]) 157 (22)

83 (14)

84 (12)

69 (12)

67 (14)

ACEI, angiotensin-converting enzyme inhibitor; A2RB, angiotensin-2 receptor blocker; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention; LVEF, left ventricular ejection fraction; TIA, transient ischaemic attack, SVT, supraventricular tachycardia; AF, atrial fibrillation or flutter; BP, blood pressure. 562

were proportionally fewer men in the echocardiographic group compared with those who did not have echocardiography. Three patients in the latter group died before the second echocardiographic examination could be undertaken and were excluded from the present study. One patient in the volatile anaesthesia group had moderate aortic stenosis and four in the TIVA group had moderate mitral regurgitation, combined with moderate aortic regurgitation in one patient. There was a median (IQR [range]) iatrogenic fluid surplus of 4.1 (2.9–5.9 [0.3–13.1]) l by 08.00 on the first postoperative day in the total population. Fluid surplus was 4.3 (3.0–5.5 [1.3–12.4]) l in the volatile anaesthesia group vs 3.7 (2.8–5.4 [0.3–13.1]) l in the TIVA group (p = 0.625). Diagnoses based on maximum/minimum values of echocardiographic data before surgery are presented in Table 2. For assessment of left ventricular diastolic dysfunction, 16 (10%) patients with reduced left ventricular ejection fraction were not studied from the analysis; nine (6%) of these also had left ventricular dilatation. Among the remaining 138 patients with normal left ventricular ejection fraction, 5/8 (62%) patients with a dilated left ventricle had diastolic dysfunction, compared with 41/130 (32%) with a normal left ventricular end-diastolic volume index (p < 0.001). The respective median (IQR [range]) left ventricular end-diastolic volume index in these two subgroups were 121 (116–150 [115–158]) ml.min
2 and 77 (66– 88 [47–102]) ml.m
2, respectively. Pre- and postoperative echocardiographic data for the total population are shown in Table 3. Postoperatively, a significant increase in heart rate was accompanied by increased cardiac output and increases in left ventricular end-diastolic volume, left ventricular ejection fraction and left atrial volumes, with increased velocities of passive mitral flow and left ventricular  ratio. A-wave relaxation resulting in an unchanged E/E velocity remained unchanged and deceleration time decreased. The changes in echocardiographic variables observed were not significantly different between the two anaesthetic groups (Table 4). Compared with pre-operatively, N-terminal proBNP peptide levels increased by the first postoperative day and were still elevated 30 days after surgery (p < 0.001 vs before surgery, Fig. 2). Before surgery, © 2014 The Association of Anaesthetists of Great Britain and Ireland

Lindholm et al. | Echocardiographic variables in major vascular surgery

Anaesthesia 2014, 69, 558–572

Table 2 Pre-operative prevalence of pathological echocardiographic variables and N-terminal proBNP in men, women and the total study group. Values are number (proportion).

LV dilatation (n = 154) Reduced LVEF (n = 154) LA dilatation (n = 153) Reduced LAEF (n = 144) Increased LVMI (n = 146)  ratio 8–15 (n = 147) E/E  ratio > 15 (n = 147) E/E LV diastolic dysfunction fulfilling our criteria modified according to Paulus et al. [17] (n = 138) LV diastolic dysfunction with normal LVEF and normal LVEDVI (n = 130) LV diastolic dysfunction with normal LVEF and dilated LVEDVI (n = 8) E/A ratio < 0.5 + DT > 280 ms (n = 139) Atrial fibrillation (n = 154) N-terminal pro-BNP normal (< 47 pmol.l
1) N-terminal pro-BNP uncertain diagnosis (47–237 pmol.l
1) N-terminal pro-BNP elevated (> 237 pmol.l
1)

Men

Women

15 14 47 10 43 55 4 34

2 2 10 7 9 25 3 12

(14%) (13%) (43%) (10%) (42%) (52%) (4%) (35%)

30 (34%) 4 (57%) 3 9 78 23

(3%) (8%) (74%) (22%)

5 (5%)

(5%) (5%) (23%) (17%) (21%) (60%) (7%) (29%)

11 (27%) 1 (100%) 1 2 33 8

(3%) (5%) (75%) (18%)

3 (7%)

Total

p value

17 16 57 17 52 80 7 46

(11%) (10%) (37%) (12%) (36%) (54%) (5%) (33%)

0.154 0.156 0.026 0.255 0.022 0.468 0.408 0.556

41 (32%)

0.543

5 (62%)

0.999

4 11 111 31

(3%) (7%) (74%) (21%)

0.999 0.730 0.999 0.825

8 (5%)

0.693

 LV, left ventricular; EF, ejection fraction; LA, left atrium; LVMI, left ventricular mass index; E, velocity of passive mitral flow; E, velocity of left ventricular relaxation; EDVI, end-diastolic volume index; proBNP, prohormone brain natriuretic peptide. A, velocity of mitral flow driven by atrial contraction; DT, deceleration time.

Table 3 Echocardiographic findings and heart rate pre-operatively and 1st/2nd postoperative day in 154 patients undergoing vascular surgery. Values are mean (SD) or median (IQR [range]). Pre-operative TR maximum velocity; m.s
1 TR gradient; mmHg LVEDV; ml LVESV; ml LVSV; ml HR; beats.min
1 CO; l.min
1 LVEF; % LA max volume; ml LA min volume; ml LAEF; % Mitralflow E; m.s
1 Mitralflow A; m.s
1 E/A ratio DT; ms  Tissue doppler Eave m.s
1  ratio E/E

2.4 23 150 65 83 70 5.5 54 76 30 59 0.6 0.7 0.8 262 7.2 8.4

(2.3–2.6 [1.8–3.1]) (21–26 [14–39]) (120–180 [57–354]) (52–85 [19–208]) (65–95 [35–209]) (12) (4.4–6.8 [2.3–13.8]) (51–59 [27–68]) (54–104 [17–272]) (19–42 [4–99]) (53–68 [20–81]) (0.5–0.8 [0.3–1.2]) (0.6–0.9 [0.1–1.4]) (0.7–1.0 [0.4–2.3]) (222–329 [137–552]) (6.1–8.8 [3.6–14.2]) (6.6–10.4 [2.3–20.0])

Postoperative 2.6 28 154 67 86 80 6.8 58 86 35 61 0.8 0.8 1.0 224 9.8 8.4

(2.4–2.8 [2.1–3.6]) (23–32 [17–53]) (126–190 [77–341]) (50–87 [25–203]) (69–107 [45–180]) (14) (5.8–8.3 [2.9–14.3]) (51–62 [31–73]) (63–111 [27–280]) (21–48 [3–124]) (55–68 [16–86]) (0.7–1.0 [0.5–1.5]) (0.7–1.0 [0.4–1.5]) (0.8–1.2 [0.3–2.7]) (188–265 [111–495]) (8.1–11.5 [4.6–18.2]) (7.1–10.7 [3.8–20.5])

p value


15, and a substantial numpatients had an E/E ber were in the ‘grey zone’. We cannot exclude that some patients may have elevated left ventricular filling pressures not detected by the methods used. In addition, to compensate for peri-operative bleeding, inflammation and fluid loss, a positive fluid balance could be triggered by epidural analgesia causing vasodilatation due to sympathetic blockade and thereby pooling of venous blood and reduced preload. We found no differences between the two anaesthetic groups in epidural use on the first postoperative day. Thus, epidural analgesia was not a confounder in the analysis comparing the two anaesthetic groups.  increased postoperatively, parallel to Both E and E  ratio and hence left the increased heart rate. The E/E ventricular filling pressure remained unchanged, compatible with preserved left ventricular diastolic function. Consequently, the increased left ventricular flow  along with the increase in (E) and tissue velocities (E) left atrial volumes observed are probably related to the observed fluid surplus, and not to the development of diastolic dysfunction. Also, the increase in left ventricular end-diastolic volume was not associated with any increase in left ventricular systolic volume or any deterioration of left ventricular ejection fraction. Thus, these alterations most likely represent myocardial stretch, which may also account for the increased post© 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 558–572

operative N-terminal pro-BNP level. We cannot exclude that some echocardiographic changes may be related to myocardial ischaemia, even in the absence of detectable myocardial infarction [23]. The rise in Nterminal pro-BNP seen on the first postoperative day had not fully returned to normal after 30 days. Chronic elevations of N-terminal pro-BNP may be present in patients with or without ischaemic heart disease, which implies significantly worse prognosis [8, 10, 24, 25]. Bicard et al. [9] evaluated pre-operative B-type natriuretic peptides in patients undergoing vascular surgery. None of the revised cardiac index risk factors were independent predictors of major adverse events. The only independent predictor was B-type natriuretic peptide stratification. Pre-operative N-terminal pro-BNP levels in this study did not influence anaesthesia management. This important risk predictor was not significantly different between the two anaesthetic groups pre- and postoperatively, parallel to the echocardiographic findings. The prognostic impact of elevated markers in a minority of patients 30 days after surgery must be weighed against the benefit of surgical repair of serious abdominal aortic diseases, including large aneurysms [26]. Bitoh et al. [27] used transoesophageal echocardiography in eight patients undergoing infrarenal abdominal aortic aneurysm repair with left ventricular ejection fraction ≥ 40% and normal dobutamine stress echocardiography. They reported a reduction in aver velocity (termed Ea and taken as the average of age E early left ventricular relaxation velocity measured from the basal left ventricular septal and lateral wall), result ratio. These findings might ing in an increased E/E appear contradictory to ours. However, our results were obtained in the first postoperative day(s) when patients had received an excess of intravenous fluid volume, whereas their data were obtained during surgery. Thus, the two studies may be complementary, as  had conversely been our results indicate that E increased during the initial postoperative period. The indices of intra-operative diastolic dysfunction in their study are probably reversible and related to aortic clamping and declamping. This is further supported by reduced left ventricular ejection fraction and stroke volume 30 min after declamping followed by an increase to above pre-operative values at the end of 567

Anaesthesia 2014, 69, 558–572

Lindholm et al. | Echocardiographic variables in major vascular surgery

surgery. Cardiac dysfunction during aortic clamping and declamping may be due to changes in coronary blood flow [28] and release of myocardial-depressant metabolites and cytokines [29]. Sarkar et al. [30] showed that volatile anaesthesia improved left ventricular relaxation and diastolic dysfunction, probably by reducing afterload. Bolliger et al. [31] found reduced early diastolic relaxation and impaired late diastolic left ventricular filling during atrial contraction in general anaesthesia with sevoflurane. There are few data on how propofol affects diastolic function. In animal studies [32, 33], propofol has not been found to affect diastolic function. Filipovic et al. [34] randomised 20 patients to sevoflurane and 20 to propofol undergoing minor surgery. Echocardiography was performed pre-operatively, during induction of anaesthesia and following mechanical ventilation of the lungs. Diastolic peak velocity of the mitral annulus was increased by sevoflurane compared with propofol during induction of anaesthesia in spontaneously breathing patients, but no differences were observed during mechanical ventilation. In this study, patients were not examined during anaesthesia, making these two studies complementary. We aimed to compare two different types of anaesthetic regimen typically used in Norway and not only the agents themselves (sevoflurane and propofol). Thereby, the two anaesthetic groups received different opioids and induction agents. Opioids have been shown to have cardioprotective effects, especially remifentanil [35], whereas the potential cardioprotective effects of thiopental are controversial [36, 37]. Different medications may potentially affect cardioprotection and potential differences could be masked by any differences between anaesthetic agents and opioids. In this study, 36 (23%) patients had ongoing infusions of noradrenaline or dopamine during the postoperative echocardiographic examination. Choice and dosage of these drugs may affect contractility and vasoconstriction and thereby influence echocardiographic data. However, there were no significant differences between the two anaesthetic groups in terms of number of patients receiving these drugs or of the doses used. No significant differences in cardiac output were observed in patients receiving noradrenaline or dopamine compared with patients who did not. This 568

may be explained by the use of relative low doses of both drugs. A major concern is poor reproducibility of echocardiographic measurements, as reported in a previous study from our laboratory [11]. In this study, both intra- and inter-observer variability might be even higher as we used only one cardiac cycle per variable. Similar patterns of echocardiographic changes were found in both anaesthetic groups. However, we cannot exclude small differences, which might be detected by averaging several beats or using a more sensitive method. The decision to use only one beat was related to the limitation that the three cardiologists had to fit many of the echocardiographic examinations and measurements into an already busy clinical schedule. It was therefore not feasible to incorporate the relatively long time required for averaging three beats that we have applied in previous studies. Power calculations in the previously published study [7] were based on troponin-T release and not echocardiographic variables. Importantly, our serial measurements detected expected changes related to volume loading. Our posthoc power calculations for left ventricular volumes were based on a previous reproducibility study [11] with a Hewlett-Packard Sonos 1000 machine, as opposed to modern equipment in this study. The differences in average changes pre- and postoperatively among the three echocardiographic variables reflecting systolic function were in the range of 1–2 and far below delta 2 levels required for a statistical intergroup difference with 80% power in a sample size of 75 patients per group (Table 6). Therefore, we conclude that there are no differences between the two groups concerning left ventricular dimensions or contractility following surgery. In some patients, pre-operative echocardiographic examinations were performed several days before surgery itself. One could argue that data obtained then are not representative of the status immediately before surgery. However, all patients were in a stable condition, without evidence of acute coronary syndrome, decompensated heart failure or arrhythmia in the period between the pre-operative echocardiogram and surgery. In addition, the majority of patients were studied within nine days of surgery, which is acceptable in our opinion. © 2014 The Association of Anaesthetists of Great Britain and Ireland

Lindholm et al. | Echocardiographic variables in major vascular surgery

We have experienced disappointing results in obtaining complete right ventricular volumes from the apical window with two- and three-dimensional echocardiography [38] (Appendix S1). To include the right ventricular outflow tract simultaneously, the epigastric view is needed, which was not possible postoperatively in our patients, as they had recently undergone major abdominal surgery. Therefore, assessments of right ventricular dimensions and function were not included. The significant increase in heart rate seen on the first postoperative day is a confounding factor when assessing echocardiographic changes. As it is not routine in our department to adjust echocardiographic measurements, none of the reported values were normalised for heart rate. As there were no clinically important or statistical differences between the two anaesthetic groups with respect to heart rate, we believe that this confounder was of minor importance for the echocardiographic comparisons. Other important confounding factors like fluid volume shifts and use of vasopressors and inotropic agents could also potentially affect haemodynamic indices and cardiac systolic and diastolic function. However, we did not find any differences between the two groups in any of these variables. The number of patients with valvular disease was small and similar between the two groups and therefore probably of limited importance for interpretation of the prevalence of left ventricular dysfunction. Selection bias cannot be ruled out, as only survivors were included on the first or second postoperative day. Patients who died before the postoperative echocardiographic examination was performed were excluded from the study. As reflected in the low and almost identical early mortality in patients selected for echocardiographic examination and those who were not, such a bias seems highly unlikely. Mortality in the study was low and similar to other published data [39–41]. In the ABSENT study, we have previously reported no differences in troponin-T release between volatile and TIVA in open abdominal aortic surgery [7], in accordance with data from Lurati Buse et al. [42] and Zangrillo et al. [43]. Data in this study showing no protective effects of volatile agents in echocar© 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 558–572

diographic variables following major vascular surgery further underline the notion that cardioprotective effects of volatile agents are not as evident in non-cardiac surgery as in cardiac surgery. The American College of Cardiology/American Heart Association guidelines recommend volatile anaesthesia for non-cardiac major vascular surgery in haemodynamically stable patients at risk for peri-operative myocardial ischaemia [6]. The clinical implications of the present and other recent studies in non-cardiac surgery [7, 42, 43] indicate that there are no substantial clinical benefits of volatile agents compared with total intravenous anaesthesia. Thus, the question arises whether the American College of Cardiology/American Heart Association guidelines need to be revised. We conclude that there are no significant differences in early postoperative echocardiographic indices of cardiac function between patients anaesthetised with volatile anaesthesia or TIVA during abdominal aortic surgery. Thus, choice of anaesthetic regimen does not influence cardiac function as evaluated by echocardiography after major vascular surgery. The postoperative changes that we have reported are most likely to be related to loading conditions due to a substantial iatrogenic fluid surplus in our patients in the immediate postoperative period. A component of peri-operative myocardial ischaemia, at least in some patients, cannot be excluded, as N-terminal pro-BNP levels were increased 30 days after surgery and some patients developed acute myocardial infarction. The prognostic significance of these observations should be addressed in larger scaled studies with longer follow up.

Acknowledgements Funding was provided solely from institutional and departmental sources. We thank Camilla Noren, Marit Eide and Tone Næss, who are study nurses in the department of Anaesthesiology, Vestfold Hospital Trust, Tønsberg, Norway. We also thank Else-Marie Ringvold, head of department and Odd Helmers, former head of department of Anaesthesiology at Vestfold Hospital Trust. We are also grateful to Ingar Holme, professor in statistics, University Hospital in Oslo, Norway for performing the post hoc power analysis. Finally, we also thank Gisle Kjøsen for proofreading 569

Anaesthesia 2014, 69, 558–572

Lindholm et al. | Echocardiographic variables in major vascular surgery

the manuscript and Erna Engelstad, who is a laboratory assistant at the Vestfold Hospital Trust. The study was funded by the Vestfold Hospital Trust, Tønsberg, Norway. EL has received fees for presentations by Baxter AS Norway, Oslo, Norway.

14.

15.

Competing interests No other competing interests declared.

16.

References 1. Gualandro DM, Campos CA, Calderaro D, et al. Coronary plaque rupture in patients with myocardial infarction after noncardiac surgery: frequent and dangerous. Atherosclerosis 2012; 222: 191–5. 2. Hur DJ, Kizilgul M, Aung WW, Roussillon KC, Keeley EC. Frequency of coronary artery disease in patients undergoing peripheral artery disease surgery. American Journal of Cardiology 2012; 110: 736–40. 3. McFalls EO, Ward HB, Moritz TE, et al. Coronary-artery revascularization before elective major vascular surgery. New England Journal of Medicine 2004; 351: 2795–804. 4. Landoni G, Biondi-Zoccai GG, Zangrillo A, et al. Desflurane and sevoflurane in cardiac surgery: a meta-analysis of randomized clinical trials. Journal of Cardiothoracic and Vascular Anesthesia 2007; 21: 502–11. 5. Symons JA, Myles PS. Myocardial protection with volatile anaesthetic agents during coronary artery bypass surgery: a meta-analysis. British Journal of Anaesthesia 2006; 97: 127– 36. 6. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery. Circulation 2007; 116: 1971–96. 7. Lindholm EE, Aune E, Noren CB, et al. The Anesthesia in Abdominal Aortic Surgery (ABSENT) Study: a prospective, randomized, controlled trial comparing troponin T release with fentanyl-sevoflurane and propofol-remifentanil anesthesia in major vascular surgery. Anesthesiology 2013; 119: 802–12. 8. Dickstein K, Cohen-Solal A, Filippatos G, et al. Guidelines for the diagnosis and treatment of acute and chronic heart failure. European Heart Journal 2008; 29: 2388–4428. 9. Biccard BM, Lurati Buse GA, Burkhart C, et al. The influence of clinical risk factors on pre-operative B-type natriuretic peptide risk stratification of vascular surgical patients. Anaesthesia 2012; 67: 55–9. 10. Kragelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-terminal pro-B-type natriuretic peptide and long-term mortality in stable coronary heart disease. New England Journal of Medicine 2005; 352: 666–75. 11. Otterstad JE, Froeland G, St John SM, Holme I. Accuracy and reproducibility of biplane two-dimensional echocardiographic measurements of left ventricular dimensions and function. European Heart Journal 1997; 18: 507–13. 12. Aune E, Baekkevar M, Rodevand O, Otterstad JE. Reference values for left ventricular volumes with real-time 3-dimensional echocardiography. Scandinavian Cardiovascular Journal 2010; 44: 24–30. 13. Aune E, Baekkevar M, Roislien J, Rodevand O, Otterstad JE. Normal reference ranges for left and right atrial volume indexes and ejection fractions obtained with real-time three570

17.

18.

19. 20.

21.

22.

23.

24.

25.

26.

27.

28. 29.

30.

dimensional echocardiography. European Journal of Echocardiography 2009; 10: 738–44. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978; 58: 1072–83. Knutsen KM, Stugaard M, Michelsen S, Otterstad JE. M-mode echocardiographic findings in apparently healthy, non-athletic Norwegians aged 20-70 years. Influence of age, sex and body surface area. Journal of Internal Medicine 1989; 225: 111–5. Nagueh SF, Appleton CP, Gillebert TC, et al. Recommendations for the evaluation of left ventricular diastolic function by echocardiography. Journal of the American Society of Echocardiography 2009; 22: 107–33. Paulus WJ, Tschope C, Sanderson JE, et al. How to diagnose diastolic heart failure: a consensus statement on the diagnosis of heart failure with normal left ventricular ejection fraction by the Heart Failure and Echocardiography Associations of the European Society of Cardiology. European Heart Journal 2007; 28: 2539–50. Poole-Wilson PA, Lubsen J, Kirwan BA, et al. Effect of longacting nifedipine on mortality and cardiovascular morbidity in patients with stable angina requiring treatment (ACTION trial): randomised controlled trial. Lancet 2004; 364: 849–57. Thygesen K, Alpert JS, White HD, et al. Universal definition of myocardial infarction. Circulation 2007; 116: 2634–5322. Ommen SR, Nishimura RA, Appleton CP, et al. Clinical utility of Doppler echocardiography and tissue Doppler imaging in the estimation of left ventricular filling pressures: A comparative simultaneous Doppler-catheterization study. Circulation 2000; 102: 1788–94. Canty DJ, Royse CF, Kilpatrick D, Bowman L, Royse AG. The impact of focused transthoracic echocardiography in the preoperative clinic. Anaesthesia 2012; 67: 618–25. Canty DJ, Royse CF, Kilpatrick D, Williams DL, Royse AG. The impact of pre-operative focused transthoracic echocardiography in emergency non-cardiac surgery patients with known or risk of cardiac disease. Anaesthesia 2012; 67: 714–20. White HD. Pathobiology of troponin elevations. Do elevations occur with myocardial ischemia as well as necrosis? Journal of the American College of Cardiology 2011; 57: 2406–8. Panagopoulou V, Deftereos S, Kossyvakis C, et al. NTproBNP: an important biomarker in cardiac diseases. Current Topics in Medicinal Chemistry 2013; 13: 82–94. Wang TJ, Larson MG, Levy D, et al. Plasma natriuretic peptide levels and the risk of cardiovascular events and death. New England Journal of Medicine 2004; 350: 655–63. Hirsch AT, Haskal ZJ, Hertzer NR, et al. ACC/AHA 2005 Practice Guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic). Circulation 2006; 113: e463–654. Bitoh H, Nakanishi K, Takeda S, Kim C, Mori M, Sakamoto A. Repair of an infrarenal abdominal aortic aneurysm is associated with persistent left ventricular diastolic dysfunction. Journal of Nippon Medical School 2007; 74: 393–401. Gelman S. The pathophysiology of aortic cross-clamping and unclamping. Anesthesiology 1995; 82: 1026–60. Bown MJ, Nicholson ML, Bell PR, Sayers RD. Cytokines and inflammatory pathways in the pathogenesis of multiple organ failure following abdominal aortic aneurysm repair. European Journal of Vascular and Endovascular Surgery 2001; 22: 485– 95. Sarkar S, GuhaBiswas R, Rupert E. Echocardiographic evaluation and comparison of the effects of isoflurane, sevoflurane © 2014 The Association of Anaesthetists of Great Britain and Ireland

Lindholm et al. | Echocardiographic variables in major vascular surgery

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

and desflurane on left ventricular relaxation indices in patients with diastolic dysfunction. Annals of Cardiac Anaesthesia 2010; 13: 130–7. Bolliger D, Seeberger MD, Kasper J, et al. Different effects of sevoflurane, desflurane, and isoflurane on early and late left ventricular diastolic function in young healthy adults. British Journal of Anaesthesia 2010; 104: 547–54. Pagel PS, Schmeling WT, Kampine JP, Warltier DC. Alteration of canine left ventricular diastolic function by intravenous anesthetics in vivo. Ketamine and propofol. Anesthesiology 1992; 76: 419–25. Graham MR, Thiessen DB, Mutch WA. Left ventricular systolic and diastolic function is unaltered during propofol infusion in newborn swine. Anesthesia and Analgesia 1998; 86: 717– 23. Filipovic M, Michaux I, Wang J, Hunziker P, Skarvan K, Seeberger M. Effects of sevoflurane and propofol on left ventricular diastolic function in patients with pre-existing diastolic dysfunction. British Journal of Anaesthesia 2007; 98: 12–8. Wong GT, Huang Z, Ji S, Irwin MG. Remifentanil reduces the release of biochemical markers of myocardial damage after coronary artery bypass surgery: a randomized trial. Journal of Cardiothoracic and Vascular Anesthesia 2010; 24: 790–6. Mullenheim J, Molojavyi A, Preckel B, Thamer V, Schlack W. Thiopentone does not block ischemic preconditioning in the isolated rat heart. Canadian Journal of Anesthesia 2001; 48: 784–9. Zaugg M, Lucchinetti E, Spahn DR, Pasch T, Garcia C, Schaub MC. Differential effects of anesthetics on mitochondrial K(ATP) channel activity and cardiomyocyte protection. Anesthesiology 2002; 97: 15–23. Aune E, Baekkevar M, Rodevand O, Otterstad JE. The limited usefulness of real-time 3-dimensional echocardiography in obtaining normal reference ranges for right ventricular volumes. Cardiovascular Ultrasound 2009; 7: 35. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. New England Journal of Medicine 2002; 346: 1128–37. Chiesa R, Tshomba Y, Psacharopulo D, et al. Open repair for infrarenal AAA: technical aspects. Journal of Cardiovascular Surgery (Torino) 2012; 53: 119–31. Lederle FA, Freischlag JA, Kyriakides TC, et al. Outcomes following endovascular vs open repair of abdominal aortic aneurysm: a randomized trial. Journal of the American Medical Association 2009; 302: 1535–42. Zangrillo A, Testa V, Aldrovandi V, et al. Volatile agents for cardiac protection in noncardiac surgery: A randomized controlled study. Journal of Cardiothoracic Vascular Anesthesia 2011; 25: 902–7. Lurati Buse GA, Schumacher P, Seeberger E, et al. Randomized comparison of sevoflurane versus propofol to reduce perioperative myocardial ischemia in patients undergoing noncardiac surgery. Circulation 2012; 126: 2696–704.

Appendix 1 Upper and lower normal limits calculated from mean obtained with 2D and 3D transthoracic echocardiography. Vaues are +/
2SD except for LVMI where only upper limits are given. © 2014 The Association of Anaesthetists of Great Britain and Ireland

Anaesthesia 2014, 69, 558–572

3D Variables

Men
2

LVEDVI; ml.m LVESVI; ml.m
2 LVEF;% LA max volume; ml.m
2 LA min volume; ml.m
2 LAEF;% LVMI*; g.m
2

2D Women

Men

Women

46–86 17–41 49–65 15–42

42–74 13–33 49–73 15–39

58–110 24–56 46–66 12–44

53–93 21–45 46–66 13–45

6–20

5–18

2–18

2–18

46–77

44–80

48–80 125

47–83 110

LVEDV, left ventricular end diastolic volume index; LVESVI, left ventricular end systolic volume index; LVEF, left ventricular ejection fraction; LA, left atrium; LAEF, left atrial ejection fraction; LVMI, Left ventricular mass index.

From references [12, 13, 15–17] and Otterstad JE, Aune E, Baekkevar M, Frøland G, Knudsen K. Hjerteforum. 2009;4:47–51. * LVMI is derived from the formula [Devereux RB, Alonso DR. Lutas EM, Gottlieb GJ, Campo E, Sachs I, et al. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. American Journal of Cardiology 1986; 57:450–8]: Left ventricular mass = 0.8 9 (1.04 (Left ventricular internal diameter in diastole + Posterior wall thickness in diastole + Septal wall thickness in diastole)3
(Left ventricular internal diameter in diastole))3 + 0.6 g:_______g

Appendix 2 The following N-terminal pro-BNP cut-off values were used to diagnose heart failure [8]: < 47 pmol.l
1 unlikely 47–237 pmol.l
1 uncertain > 237 pmol.l
1 likely Diagnostic criteria and definitions for preoperative conditions and diseases: Preoperative coronary artery disease: Diagnosed by one or more of the following criteria: Sustained acute myocardial infarction verified from hospital records, previous percutaneous coronary intervention and/or coronary artery bypass grafting, angiographically verified coronary artery stenosis ≥50% or a positive exercise test combined with a history of typical angina pectoris [20]. 571

Anaesthesia 2014, 69, 558–572

Lindholm et al. | Echocardiographic variables in major vascular surgery

History of stroke/transient ischaemic attack: Diagnosis verified by hospital records including findings at computer tomography and/or magnetic resonance imaging. Peripheral artery disease: Diagnosis verified by one or more angiographic significant stenoses (≥50%) of aortoiliacal and infrainguinal arteries. Carotid artery stenosis: Angiographic or ultrasonographic evidence of one or more significant stenosis (≥50%). Atherosclerotic disease: Presence of one or more of the diagnoses listed above. Systolic left ventricular dysfunction at baseline: Left ventricular ejection fraction 15 abnormal 8–15 “grey zone” 15 or/and of brain natriuretic pep• N-Terminal prohormone
1  ≥8 or at least tide >26 pmol.l provided E/E-ratio one of the following:

572

° ° ° ° •

Left atrial max volume >44 ml.m
2. Left ventricular mass index >125 g.m
2 in men and >110 g.m
2 in women. E/A ratio 280 ms in patients >50 years. Atrial fibrillation.

or/and  8 ≤E/E-ratio ≤ 15 and at least one of the following:

° ° ° °

Left atrial max volume >44 ml.m
2. Left ventricular mass index >125 g.m
2 in men and >110 g.m
2 in women. E/A ratio 280 ms in patients >50 years. Atrial fibrillation.

Postoperative myocardial infarction: A rise in troponin T to >30 ng.l
1 and one or more of the following criteria [21]: symptoms of ischaemia, new electrocardiogram changes indicative of ischaemia (ST and/or T changes or new left bundle branch block) and development of pathological Q waves.

© 2014 The Association of Anaesthetists of Great Britain and Ireland