The Hemodynamic Effects of Different Pacing Modalities After Cardiopulmonary Bypass in Patients With Reduced Left Ventricular Function

Author’s Accepted Manuscript The Hemodynamic Effects of Different Pacing Modalities after Cardiopulmonary Bypass in Patients with Reduced left Ventricular Function R.C.W. Gielgens, I.H.F. Herold, A.H. van Straten, B.M. van Gelder, F.A. Bracke, H.H.M. Korsten, M.A. Soliman Hamad, R.A. Bouwman www.elsevier.com/locate/buildenv

PII: DOI: Reference:

S1053-0770(17)30648-1 http://dx.doi.org/10.1053/j.jvca.2017.07.003 YJCAN4234

To appear in: Journal of Cardiothoracic and Vascular Anesthesia Cite this article as: R.C.W. Gielgens, I.H.F. Herold, A.H. van Straten, B.M. van Gelder, F.A. Bracke, H.H.M. Korsten, M.A. Soliman Hamad and R.A. Bouwman, The Hemodynamic Effects of Different Pacing Modalities after Cardiopulmonary Bypass in Patients with Reduced left Ventricular Function, Journal of Cardiothoracic and Vascular Anesthesia, http://dx.doi.org/10.1053/j.jvca.2017.07.003 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The Hemodynamic effects of different pacing modalities after cardiopulmonary bypass in patients with reduced left ventricular function

R.C.W. Gielgensa,b, I.H.F. Herolda, A.H. van Stratenc, B.M. van Gelderd, F.A. Bracked, H.H.M. Korstena, M.A. Soliman Hamadc, R.A. Bouwmana a

Department of Anesthesiology and Intensive-Care, Catharina Hospital, P.O. Box 1350, 5602 ZA, Eindhoven,

the Netherlands b

Faculty of Health, Medicine and Life Sciences, Maastricht University, P.O. Box 616, 6200 MD, Maastricht

c

Department of Cardiothoracic Surgery, Catharina Hospital, P.O. Box 1350, 5602 ZA, Eindhoven, The

Netherlands d

Department of Cardiology, Catharina Hospital, P.O. Box 1350, 5602 ZA, Eindhoven, the Netherlands

Corresponding author: R.C.W. Gielgens, Department of Anaesthesiology and Intensive-Care, Catharina Hospital, Eindhoven, Michelangelolaan 2, 5623 EJ Eindhoven, The Netherlands.

Corresponding e-mail: [email protected] [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]

Declaration of interest This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. B.G. provides training and education for St Jude Medical, the Netherlands and is clinical advisor for LivaNova, France. R.B and H.K. are clinical consultants for Philips Research, Eindhoven, the

1

Netherlands since 2016. R.B. received travel funding from CSL Behring to visit NATA 2015 (approx. 500 euro) and fees from ABBVie bv for preparation of educational material and lectures in 2013 (1500 euro).

Acknowledgements I.H., M.H. and, R.B. conceived the study. I.H., M.H. and, R.B. participated in the design of the study. R.G. performed the literature search. R.G. and I.H. retrieved relevant information from the articles. R.G. performed the statistical analysis. R.G., I.H., M.H. and R.B. participated in the data interpretation and drafted the manuscript. A.S., B.G., H.K. and, F.B. reviewed and the final draft. All authors read and approved the manuscript.

Abstract Objective(s): Patients with a decreased left ventricular function undergoing cardiac surgery have a higher chance of difficult weaning from cardiopulmonary bypass and a poorer clinical outcome. Directly after weaning, interventricular dyssynchrony, paradoxical septal motion, and even temporary bundle branch block might be observed. In this study, we measured arterial dP/dtmax, mean arterial pressure (MAP), and cardiac index (CI) with transpulmonary thermodilution (TPTD), pulse contour analysis and femoral artery catheter and compared the effects between right ventricular (A-RV) and biventricular (A-BiV) pacing on these parameters. Design: Prospective mono-centre pilot study. Setting: Single centre study Participants: 17 patients with normal and prolonged QRS duration and a left ventricular ejection fraction ≤ 35% undergoing coronary artery bypass grafting (CABG) with or without valve replacement.

2

Interventions: Temporary pacing wires were placed on the right atrium and both ventricles. Different pacing modalities were used in a standardised order. Measurement and main results: A-BiV pacing compared to A-RV pacing showed higher arterial dP/dtmax values (846 ± 646 mmHg/s vs. 800 ± 587 mmHg/s, p=0.023) and higher MAP values (77 ± 19 mmHg vs. 71 ± 18 mmHg, p=0.036). Conclusion: In patients with a preoperatively decreased left ventricular function undergoing CABG, biventricular pacing improves the arterial dP/dtmax and MAP in both patients with a normal and prolonged QRSduration when compared to standard RV pacing. In addition, we showed that arterial dP/dtmax and MAP can be used to evaluate the effect of intraoperative pacing. In contrast to previous studies using more invasive techniques, TPTD is easy to apply in the perioperative clinical setting. Keywords: Artificial cardiac pacing, Left Ventricular Dysfunction, Coronary Artery Bypass Graft, Resynchronization therapy, Hemodynamic effect

Introduction Cardiac resynchronization therapy (CRT) is frequently used for patients with heart failure associated with abnormal ventricular conduction, indicated by a wide QRS-complex, resulting in cardiac dyssynchrony. An increase of left ventricular contractility is observed in the majority of these patients. 1,2

Patients with a decreased left ventricular function undergoing cardiac surgery are at an increased risk of difficult separation from cardiopulmonary bypass (CPB) and a poor clinical outcome despite revascularisation.3,4 Directly after weaning interventricular dyssynchrony, paradoxical septal motion,

3

and even temporary bundle branch block are frequently observed. which could compromise optimal contractility and output of the left ventricle.5,6 Furthermore, atrioventricular blockade after cardiopulmonary bypass is observed in coronary artery bypass patients, which can also be effectively treated by a temporary atrioventricular (A-RV) pacemaker. 7 In patients with interventricular dyssychrony caused by RV-pacing this could diminish contractility and effectivity of the left ventricle. The presence of severe LV dyssynchrony in patients undergoing coronary artery bypass grafting was associated with high in-hospital and long-term mortality in patients with ischemic heart failure undergoing myocardial revascularization.4 It was shown that the use of CRT in these patients had a positive effect on hemodynamics, both post-operatively and directly after weaning from CPB. 8-12 Left ventricular (LV) or biventricular (BiV) pacing in patients with a prolonged QRS-duration has shown improvement of the systolic function while lowering the energy cost and the use of inotropic support. 13-15

These previous observations were derived from left ventricular (LV) dP/dtmax measurements via invasive transmurally RADI pressure tip guide wires placed directly into the left ventricle or via a pressure-volume (PV) catheter. 8,16 The less invasive transpulmonary thermodilution (TPTD) and pulse contour analysis have been shown to measure femoral dP/dtmax accurately and correlates with changes in LV dP/dtmax and volumetric preload parameters, including intrathoracic blood volume (ITBV) and global end-diastolic volume (GEDV). 17-21 The dP/dtmax measurement of the arterial pressure waveform was obtained by arterial waveform analysis. The arterial pressure waveform describes the change in pressure over time. The upslope of the arterial pressure curve represents the rise in LV pressure, therefore the dP/dtmax gives a representation of the contractility of the left ventricle.22 4

Patients with a low left ventricular ejection fraction (LVEF) and intraventricular conduction delay could benefit from biventricular pacing during cardiac surgery by a decreased need for inotropic medication, an easier separation of CPB, and a shorter ICU stay.3-15 The objective of this prospective mono-centre pilot study is to compare the effects of atrial- right or left ventricular pacing (A-RV or A-LV) with biventricular simultaneous pacing (A-BiV) on the arterial dP/dtmax, mean arterial pressure (MAP), and cardiac index (CI) via minimal invasive TPTD measurements. The secondary objective is to evaluate if there exists a difference in effect of BiV pacing on patients with normal QRS-duration (<120ms) and prolonged QRS-duration (>120ms).

Methods Ethics This prospective mono-centre pilot study was approved by the local medical ethical review board. All patients provided written informed consent. This study is registered in the ISRCTN registry under trial number 90330260.

Patients As this study was a sub-analysis of our previous work, data of a part of the patients was also used in our previous study on pulmonary blood volume measurement by contrast enhanced ultrasound. 23 All eligible patients were 18 years of age or older, had a LVEF of less than 35% and signs of dyssynchrony with and without prolonged QRS duration, were scheduled for coronary artery bypass grafting (CABG) surgery and/or valve surgery. Patients who suffered from a myocardial infarction within the past 3 months were excluded. Due to the fact that this study was a sub-analysis of our previous study 5

on pulmonary blood volume measurement by contrast enhanced ultrasound patients with a contraindication for transesophageal echocardiography were also excluded.

Preoperative procedures All patients received standard pre-operative care which included a physical examination, 12-lead electrocardiogram, and preoperative laboratory data. All patients were subjected to a preoperative echocardiogram to assess the LVEF, left ventricular end-systolic diameter (LVESD), left ventricular dyssynchrony, and the presence of concomitant valvular disease.

Operative procedures Operative procedures were equal to those followed in our previous study. 23 Following local protocol every patient received the following medication at the induction of anaesthesia; cefazolin 2 grams, tranexamic acid 1 gram, etomidate 0.2mg/kg, fentanyl 5-7µg/kg and, rocuronium 0.6mg/kg. If indicated and deemed necessary by the attending anaesthesiologist, patients received midazolam 5mg and/or dexamethasone 1mg/kg. Dosage and the decision to use inotropic/vasoactive drugs was made by the attending anesthesiologist. After induction patients were intubated and received a 8.5 F central venous catheter (DM- 12853, Arrow, Reading, PA, USA) in the right jugular vein. The medial lumen was connected to the injectate temperature sensor (PV4046, Pulsion Medical Systems, Munich, Germany). The Pulsiocath 5 F thermistor tipped catheter, PV2015L20A (Pulsion Medical Systems, Munich, Germany), was placed in the femoral artery. Both the injectate sensor and Pulsiocath were connected to the PiCCO2 monitor (Pulsion Medical Systems, Munich, Germany). General anaesthesia was maintained with propofol 5mg/kg/hr and alfentanyl 0,5mg/kg/hr via perfusor.

6

The effect of the different pacing modalities was monitored continuously with a sample every 12 seconds after cardiopulmonary bypass during hemostasis of the thorax in a period with minor hemodynamic fluctuations with no change in inotropes and vasopressor doses. The pacemaker technician and anesthesiologist (BG/IH) performed the different pacing modalities once and the values of arterial dP/dtmax, MAP and, CI were registered by PiCCO monitor automatically. Times were registered by the attending anaesthetic nurse. The dP/dtmax, MAP and CI were collected by pulse contour analysis of the arterial pressure wave. Therefore, a calibration measurement was performed by injection of 20 ml cold saline as indicator in the central venous line.23 At the tip of the femoral artery PiCCO catheter a thermistor registered the temperature change in time. The cardiac index was estimated from the thermodilution curve by using the Stewart-Hamilton equation. The injection of this indicator automatically started the measurement on the PiCCO2-monitor. All TPTD data were stored via PiCCO-Win software on a computer that was linked to the PiCCO2-monitor. Volume replacement was standardized. Before surgery no volume was added, volume status was evaluated pre- and post-bypass via TEE. The cardiopulmonary bypass was primed with 1 liter saline and 0,5 liter colloid Voluven® (Fresenius Kabi Norge, Halden, Norway). After weaning from cardiopulmonary bypass, the aim was to empty the reservoir completely, otherwise it was given as rest volume or cellsaved and returned to the patient. Other blood loss was cellsaved and returned to the patient. The preload and contractility of the left and right ventricle was visually estimated by echocardiography pre and post bypass.24

For pacing, temporary wires were sutured to the right atrium and right ventricular free wall, according to standard procedures. Similar to the RV, additional temporary LV leads were sutured to the basal 7

obtuse marginal part of the LV. Pacing leads were connected to a standard external dual chamber pacemaker, Medtronic Temporary Pacemakers, model 5388 (Medtronic, Minneapolis, Mn, USA). To connect the RV and LV lead to this device a splitter was placed in the ventricular channel thus enabling simultaneous BiV pacing, figure 1.

Pacing protocol After confirmation of an adequate pacing and sensing threshold, the different pacing modalities were used in the standardised order: Atrial pacing 5-10 beats above the intrinsic atrial rhythm for patients with normal AV conduction. By using this minimal overdrive pacing the heart rate is kept constant during the hemodynamic measurements to avoid changes in hemodynamics due to unpredictable changes in heart rate. For patients with prolonged AV conduction RV pacing was used as a backup with an AV interval of 120ms. These settings are considered as baseline conditions. In patients with intact AV conduction A-RV pacing is performed with an AV interval equal to 75% of the intrinsic PQ time. Pacing protocol for A-LV and simultaneous A-biventricular pacing was identical to A-RV pacing, as described above. These settings are considered as baseline conditions. Next, RV pacing for patients with intact AV conduction atrium as well as RV will be paced at the heart rate 5-10 beats above the intrinsic rate, with the AV interval set at 75% of the intrinsic PQ time. For patients with prolonged atrial conduction the hemodynamic effect is similar to atrial pacing as described under atrial pacing. Pacing protocols for LV and simultaneous biventricular pacing (BiV VV-0ms) pacing was identical to RV pacing, as described above. Simultaneous BiV pacing was obtained by the use of a splitter in the ventricular channel for simultaneous RV and LV pacing (fig 1).

8

Determination of the pacemaker stimulation threshold was performed for each pacemaker lead by slowly turning the output (mA) to a higher range until capture was obtained on the electrocardiogram, which is the stimulation threshold. The output of the pacemaker was subsequently set at two times the stimulation threshold. Baseline was defined as the average A-RV pacing preceding and following each pacing mode, depending on the status of AV conduction. To minimise the effect of drift over time, all data were obtained for a period of minimally 45 seconds per pacing modality to reach a steady state and a baseline pacing mode was first performed between different pacing modalities, both as recommended by Thibault et al.25 We have seen that after changing pacing conditions positive and negative LVdP-dt are stabilized within these 45 seconds.8 The effect of the different pacing modalities was monitored continuously after cardiopulmonary bypass during drying the thoracotomy wound and closure of the thorax.

Statistical analyses All data were analysed using SPSS statistics (IBM, version 21, Armonk, NY, USA). Because of the total amount of included patients being below 20 non-parametrical testing was used. Therefore, continuous variables were described in terms of medians and interquartile range (IQR). Differences in the arterial dP/dtmax values between settings of the pacing modalities were tested using Wilcoxon’s test. Differences in arterial dP/dtmax values between normal and prolonged QRS duration were tested with the Mann-Whitney test. QRS duration < 120ms was defined as normal, a QRS duration ≥ 120ms was defined as prolonged. A p-value of p < 0.05 was considered significant.

9

Results A total of 17 patients undergoing cardiac surgery between January 2010 and July 2013 met the inclusion criteria of this study. Two patients were excluded; in one patient pacing was not possible due to the perioperative use of an intra-aortic balloon pump (IABP) and in one patient there was malfunction of the PiCCO catheter. Pre-operative patient characteristics, subdivided in the groups QRS<120ms and QRS>120ms, are shown in table 1. There were no complications reported due to the introduction of the PiCCO catheter. No intra- or postoperative complications were observed with the placement of temporary pacing wires or the completion of the pacing protocol.

Arterial dP/dtmax No difference was found regarding dP/dtmax when comparing RV pacing to LV pacing (800 ± 587 mmHg/s vs. 810 ± 597 mmHg/s, p=0.382). However, simultaneous BiV pacing showed significantly higher dP/dtmax values compared to RV pacing (846 ± 646 mmHg/s vs. 800 ± 587 mmHg/s, p=0.023) (figure 2a). When comparing the dP/dtmax between patients with a normal and a prolonged QRS duration, we found significant differences in RV pacing (629 mmHg/s vs 956 mmHg/s, p=0.028), LV pacing (562 mmHg/s vs 994 mmHg/s, p=0.04) and simultaneous BiV pacing (663 mmHg/s vs 966 mmHg/s, p=0.049), figure 3. Within the intrinsic and baseline group the dP/dtmax values between the normal and prolonged QRS patients were not significantly different.

10

Mean arterial pressure MAP was not different when comparing RV pacing to LV pacing (70.8 ± 18 mmHg vs. 71.0 ± 18 mmHg, p=0.650). However, simultaneous BiV pacing showed higher MAP values compared to RV pacing (70.8 ± 18 mmHg vs. 77 ± 19 mmHg, p=0.036), figure 2b. No significant difference was found when comparing the MAP between patients with a normal and a prolonged QRS in the different pacing modalities.

Pulse contour cardiac index We found no significant difference regarding Pulse contour cardiac index (PCCI) when comparing RV pacing to LV pacing (3.34 ± 0.88 l/min/m2 vs. 3.42 ± 0.83 l/min/m2, p=0.254). simultaneous BiV pacing showed no significantly higher PCCI values (3.23 ± 0.86 l/min/m2, p=0.875) than RV pacing. When comparing the PCCI between patients with a normal and a prolonged QRS we found no significant differences in the different pacing modalities.

Discussion In this study, we showed that simultaneous BiV pacing improved hemodynamics (arterial dP/dtmax and MAP) in patients with a reduced left ventricular function undergoing cardiac surgery when compared to standard RV pacing. Furthermore, patients with prolonged QRS duration showed a significant higher arterial dP/dtmax compared to patients with a normal QRS duration for both RV, LV, and simultaneous BiV pacing. Moreover, our results indicate that arterial dP/dtmax and pressure measurements are feasible to evaluate the effect of intraoperative pacing.

The effect of simultaneous BiV pacing on cardiac function has been well established and simultaneous BiV pacing has become the gold standard for treating heart failure patients with a LVEF of less than 11

35%, a QRS-complex of more than 130ms and sinus rhythm.26 Multiple studies showed the acute beneficial effect of CRT on pulse pressure, LV dP/dtmax and cardiac output2,13,27 Only recently, BiV pacing after coronary surgery with cardiopulmonary bypass gained research interest. One of the first studies that focused on the hemodynamic effects after elective coronary artery bypass grafting was done by Foster et al. 9 They investigated 18 patients without a atrioventricular and intraventricular conduction delay, including 14 patients with a LVEF > 40%. It was found that 12-36 hours after coronary artery revascularization atrio-biventricular pacing increased the cardiac index significantly compared to AAI pacing. In 54 patients with a mean QRS duration of 124ms and a mean LVEF of 25%, Dzemali et al. found that simultaneous BiV pacing improved cardiac output and that this improvement lasted 6-24h postoperatively. 10 Flynn et al. found similar results for the cardiac index one hour postoperatively, but mainly included patients without conduction delay.11 These studies are in agreement with our findings where the difference in response between normal and prolonged QRS is not significant different. However, the absolute values were higher in the prolonged QRS groups for all pacing modes. The latter was not an expected finding, which should be evaluated, an explanation could be found in the peripheral position of the PiCCO catheter or the low number of patients (7 with a QRS duration below 120ms). CRT is indicated in patients with decreased ejection fraction (<35%) with signs of dyssychrony between the left and right ventricle. As earlier reported, the response to CRT is limited to 60-70%.28 A positive response to CRT is defined by different criteria, like decrease in N-terminal pro brain natriuretic peptide, improvement in ejection fraction or an increase in functionality. The most frequently used criteria is an decrease in left ventricular end-systolic volume by more than 15%.29-31 In contrary to these criteria, we evaluated response by a changes in dP/dt, MAP and CI peri-operatively. 12

Predictors for response to CRT are defined within different categories. Ischemic cardiomyopathy is known to respond less than non-ischemic origin of left ventricular dysfunction, as are patients with a severely decreased functional capacity as described by New York Heart Association (NYHA) classification and low quality of life scores (Minnesota living with heart failure score. 29,32,33 Furthermore, extremely dilated left ventricular volume, body mass index < 30 kg/m2, smaller left atrial size, and the male sex are also prone for non-response. Concluding there seems to be a limit to which response seems possible to expect. Our patient population was mainly ischemic and therefore less prone to respond, we had however, a large group of LBBB (n=7) which are known to have a higher response rate when the QRS duration is above 150 ms.34 However the mean QRS duration in our prolonged QRS duration group was just below this margin with 148 seconds.

Multiple studies focused on the effect of CRT directly after weaning from cardiopulmonary bypass (CPB). A recent study performed by Russel et al.12 included 38 patients with a LVEF <35% undergoing on-pump surgical revascularization with or without associated valve surgery. They reported an increased CO of 7% and 9% for BiV pacing compared to AAI and RV pacing respectively. The study of Hanke et al.16 found that both CI and LV dP/dtmax improved when comparing BiV pacing to RV pacing. Although both studies did not exclude patients with a QRS-duration above 130ms the average QRSduration in the BiV group was lower than the average QRS-duration in our study, 113.3ms and 98ms, respectively versus 126.8ms in our study. Soliman Hamad et al.8 also studied the effects of BiV CRT in patients with poor left ventricular function and focused on patients who had a QRS-duration above 130ms with a left bundle branch pattern. Mean LV dP/dtmax was measured transmurally using a RADI sensor-tipped pressure 13

guidewire. Although only 11 patients were included, they also found a significant increase in mean LV dP/dtmax when comparing simultaneous BiV pacing to RV pacing.

Whereas this study focused on measuring the effect of CRT directly after weaning from CPB using a minimal invasive approach, previous studies on the same subject where preformed with more invasive measurement methods. Hanke et al.16, Soliman Hamad et al.8 and Russel et al.12 used a PVcatheter, a transmurally-placed RADI sensor-tipped pressure guided wire and a pulmonary artery catheter, respectively. The articles of De Hert et al.17 and Morimont et al.21 both concluded that femoral dP/dtmax underestimates LV dP/dtmax but accurately reflects changes in LV dP/dtmax during various interventions. Due to the fact that femoral artery dP/dtmax measurement is less invasive than the methods used in previous research on this subject it is also easier to incorporate in to clinical routine and is less time consuming. An additional advantage of the femoral artery catheter over the more invasive measurement methods that it can be used during surgery as well as postoperatively. Adequate vascular filling is the only essential condition for accurate assessment of hemodynamics via PiCCO. 21

Limitations The relatively small number of included patients may limit interpretation of the present results, as the number of patients was not large enough to stratify for QRS-duration and compare the effect of the different pacing modalities. Our study focused on the effect of simultaneous BiV resynchronisation and LV pacing versus standard RV pacing. We did not incorporate sequential BiV resynchronisation, which is pacing both ventricles with different intraventricular conduction delay, in our research. However, earlier research on 14

patients with severe heart failure showed that when simultaneous BiV resynchronisation is compared to sequential BiV resynchronisation, the latter improves the LV systolic and diastolic performance even more.2,27 Recent studies on the effect of sequential BiV resynchronisation in patients with LV dysfunction undergoing CABG seems to confirm that sequential biventricular resynchronisation also improves cardiac function peri-operatively.8,12,35 As stated above, previous studies found an improved CO or CI up to 32 hours postoperatively with BiV pacing compared to AAI pacing.10-12 However, the presented results show differently, and may relate to the variability in the cardiac output measurements and the subtility of the change in the hemodynamic parameters. Therefore, subtle changes in cardiac output/cardiac index may have remained undetected in this relatively small sample sized study. In our study we found no significant difference in CI, which could be due to the relatively small number of included patients. The study of Hanke and Garcia-Bengochea both found relatively small differences in CI between pacing modalities (respectively 0,53L/m2, p=<0.01, n=21 and 0,27L/m2, p=<0.05, n=50). 16,36 These expected subtle changes in hemodynamic parameters combined with the variability of cardiac output measurements in itself may explain the lack of increase of cardiac index. Moreover, different studies showed a high agreement between the measured CO by pulse contour analysis but a less reliable trending in comparison to TPTD CO.37,38 On the other hand a recent study in dogs showed a high accuracy of the pulse contour analysis in measuring CO during large volume shifts. Due to these hemodynamic changes, we did not perform different sequential BiV pacing modalities. A further limitation to this study was that no stratification was performed for the use of inotropic drugs. Inotropic drugs could influence the effect on arterial dP/dtmax and MAP. However, as shown in 15

table 2, only 7 patients received inotropic drugs. The dosage and the decision to use inotropic/vasotropressor drug was made by the attending anesthesiologist. All measurements per patient were performed under the same dosage of inotropic/vasopressor drugs. Pacing protocol was performed in a relative stable post-bypass phase where no extra fluids needed to be administered. If the dP/dtmax and/or MAP would have been altered by the use of inotropic/vasopressor drugs this effect would be (roughly) the same for each pacing modality, as the dosage did not alter during the different passing modalities. Therefore, we presume the effect of inotropic drugs on the arterial dP/dtmax and MAP was limited in this study.

Future studies As stated previously we found a significant higher arterial dP/dtmax in the groups with the prolonged QRS-duration for the RV, LV and simultaneous BiV pacing modalities, which can be explained by the presence of dyssynchrony in the patients with a prolonged QRS-duration. Stratification for this variable could be interesting for future studies with larger patient groups. The present study only focused on the acute improvement of cardiac function after CPB and did not investigate the possible long term benefits of the improvement of cardiac function. According to Steendijk et al.39 the long term hemodynamic effects of simultaneous biventricular pacing in patients with heart failure showed that the acute hemodynamic improvements of CRT are maintained and that ventricular-atrial coupling, mechanical efficiency and chronotropic response are improved after 6 months of CRT. The study of Cleland et al.28 which included 813 patients with heart failure who were randomly assigned to receive either medical therapy only or medical therapy with CRT found that CRT improved the quality of life, improved symptoms and reduced complications and the risk of death. In our knowledge, as of yet, there are no studies that focused on the long-term outcome of CRT in

16

patients with a reduced LV function undergoing cardiac surgery. This might be a focus for upcoming studies.

Conclusion In conclusion, in patients with a preoperatively decreased left ventricular function undergoing CABG, biventricular pacing improves the arterial dP/dtmax and MAP in both patients with a normal and a prolonged QRS-duration when compared to standard RV pacing. Patients with a prolonged QRSduration seem to benefit more from RV, LV and BiV pacing than patients with a normal QRS-duration as shown by larger effect of BiV on the arterial dP/dtmax, there was no difference in MAP and CI between the groups. As it was feasible to evaluate the effect of pacing on cardiovascular function using a clinical measurement technique, we feel that CRT strategies in the intra-operative setting after CPB warrant further exploration in different patient populations with regard to QRS duration and ventricular function.

Figure 1. Schematic presentation of external simultaneous biventricular pacing.

Figure 2. Perioperative arterial dP/dtmax (A) and mean arterial pressure (MAP) (B) measurements during different pacing modes in comparison with RV pacing as a baseline for both. *NS, **p = 0.023, and ***p = 0.036.

Figure 3. Boxplots displaying the difference in arterial dP/dtmax between patients with a normal or prolonged QRS-duration for RV (A, p=0.028), LV (B, p=0.040) and BiV (C, p=0.049) pacing modalities. 17

Reference List

[1] Auricchio A, Stellbrink C, Block M, et al.: Effect of pacing chamber and atrioventricular delay on acute systolic function of paced patients with congestive heart failure. The Pacing Therapies for Congestive Heart Failure Study Group. The Guidant Congestive Heart Failure Research Group. Circulation99(23):pp. 2993-3001,1999. [2] van Gelder BM, Bracke FA, Meijer A, et al.: Effect of optimizing the VV interval on left ventricular contractility in cardiac resynchronization therapy. Am J Cardiol93(12):pp. 1500-1503,2004. [3] Denault AY, Tardif JC, Mazer CD, et al.: Difficult and complex separation from cardiopulmonary bypass in high-risk cardiac surgical patients: a multicenter study. J Cardiothorac Vasc Anesth26(4):pp. 608-616,2012. [4] Penicka M, Bartunek J, Lang O, et al.: Severe left ventricular dyssynchrony is associated with poor prognosis in patients with moderate systolic heart failure undergoing coronary artery bypass grafting. J Am Coll Cardiol50(14):pp. 1315-1323,2007. [5] Elmi F, Tullo NG, Khalighi K: Natural history and predictors of temporary epicardial pacemaker wire function in patients after open heart surgery. Cardiology98(4):pp. 175-180,2002. [6] Reynolds HR, Tunick PA, Grossi EA, et al.: Paradoxical septal motion after cardiac surgery: a review of 3,292 cases. Clin Cardiol30(12):pp. 621-623,2007. [7] Baerman JM, Kirsh MM, de BM, et al.: Natural history and determinants of conduction defects following coronary artery bypass surgery. Ann Thorac Surg44(2):pp. 150-153,1987. 18

[8] Hamad MA, van Gelder BM, Bracke FA, et al.: Acute hemodynamic effects of cardiac resynchronization therapy in patients with poor left ventricular function during cardiac surgery. J Card Surg24(5):pp. 585-590,2009. [9] Foster AH, Gold MR, McLaughlin JS: Acute hemodynamic effects of atrio-biventricular pacing in humans. Ann Thorac Surg59(2):pp. 294-300,1995. [10] Dzemali O, Bakhtiary F, Dogan S, et al.: Perioperative biventricular pacing leads to improvement of hemodynamics in patients with reduced left-ventricular function--interim results. Pacing Clin Electrophysiol29(12):pp. 1341-1345,2006. [11] Flynn MJ, McComb JM, Dark JH: Temporary left ventricular pacing improves haemodynamic performance in patients requiring epicardial pacing post cardiac surgery. Eur J Cardiothorac Surg28(2):pp. 250-253,2005. [12] Russell SJ, Tan C, O'Keefe P, et al.: Optimized temporary bi-ventricular pacing improves haemodynamic function after on-pump cardiac surgery in patients with severe left ventricular systolic dysfunction: a two-centre randomized control trial. Eur J Cardiothorac Surg42(6):p. e146-e151,2012. [13] Nelson GS, Berger RD, Fetics BJ, et al.: Left ventricular or biventricular pacing improves cardiac function at diminished energy cost in patients with dilated cardiomyopathy and left bundlebranch block. Circulation102(25):pp. 3053-3059,2000. [14] Nguyen HV, Havalad V, ponte-Patel L, et al.: Temporary biventricular pacing decreases the vasoactive-inotropic score after cardiac surgery: a substudy of a randomized clinical trial. J Thorac Cardiovasc Surg146(2):pp. 296-301,2013. 19

[15] Bakhtiary F, Dogan S, Dzemali O, et al.: Impact of different pacing modes on left ventricular contractility following cardiopulmonary bypass. Pacing Clin Electrophysiol30(9):pp. 10831090,2007. [16] Hanke T, Misfeld M, Heringlake M, et al.: The effect of biventricular pacing on cardiac function after weaning from cardiopulmonary bypass in patients with reduced left ventricular function: a pressure-volume loop analysis. J Thorac Cardiovasc Surg138(1):pp. 148-156,2009. [17] De Hert SG, Robert D, Cromheecke S, et al.: Evaluation of left ventricular function in anesthetized patients using femoral artery dP/dt(max). J Cardiothorac Vasc Anesth20(3):pp. 325-330,2006. [18] Sakka SG, Bredle DL, Reinhart K, et al.: Comparison between intrathoracic blood volume and cardiac filling pressures in the early phase of hemodynamic instability of patients with sepsis or septic shock. J Crit Care14(2):pp. 78-83,1999. [19] Della RG, Costa GM, Coccia C, et al.: Preload index: pulmonary artery occlusion pressure versus intrathoracic blood volume monitoring during lung transplantation. Anesth Analg95(4):pp. 83543, table,2002. [20] Huber W, Umgelter A, Reindl W, et al.: Volume assessment in patients with necrotizing pancreatitis: a comparison of intrathoracic blood volume index, central venous pressure, and hematocrit, and their correlation to cardiac index and extravascular lung water index. Crit Care Med36(8):pp. 2348-2354,2008.

20

[21] Morimont P, Lambermont B, Desaive T, et al.: Arterial dP/dtmax accurately reflects left ventricular contractility during shock when adequate vascular filling is achieved. BMC Cardiovasc Disord12:p. 13,2012. [22] Esper SA, Pinsky MR: Arterial waveform analysis. Best Pract Res Clin Anaesthesiol28(4):pp. 363380,2014. [23] Herold IH, Soliman Hamad MA, van Assen HC, et al.: Pulmonary blood volume measured by contrast enhanced ultrasound: a comparison with transpulmonary thermodilution. Br J Anaesth115(1):pp. 53-60,2015. [24] Gudmundsson P, Rydberg E, Winter R, et al.: Visually estimated left ventricular ejection fraction by echocardiography is closely correlated with formal quantitative methods. Int J Cardiol101(2):pp. 209-212,2005. [25] Thibault B, Dubuc M, Karst E, et al.: Design of an acute dP/dt hemodynamic measurement protocol to isolate cardiac effect of pacing. J Card Fail20(5):pp. 365-372,2014. [26] Strickberger SA, Conti J, Daoud EG, et al.: Patient selection for cardiac resynchronization therapy: from the Council on Clinical Cardiology Subcommittee on Electrocardiography and Arrhythmias and the Quality of Care and Outcomes Research Interdisciplinary Working Group, in collaboration with the Heart Rhythm Society. Circulation111(16):pp. 2146-2150,2005. [27] Sogaard P, Egeblad H, Pedersen AK, et al.: Sequential versus simultaneous biventricular resynchronization for severe heart failure: evaluation by tissue Doppler imaging. Circulation106(16):pp. 2078-2084,2002.

21

[28] Cleland JG, Daubert JC, Erdmann E, et al.: The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med352(15):pp. 1539-1549,2005. [29] Bleeker GB, Bax JJ, Fung JW, et al.: Clinical versus echocardiographic parameters to assess response to cardiac resynchronization therapy. Am J Cardiol97(2):pp. 260-263,2006. [30] Bax JJ, Bleeker GB, Marwick TH, et al.: Left ventricular dyssynchrony predicts response and prognosis after cardiac resynchronization therapy. J Am Coll Cardiol44(9):pp. 1834-1840,2004. [31] Hoogslag GE, Hoke U, Thijssen J, et al.: Clinical, echocardiographic, and neurohormonal response to cardiac resynchronization therapy: are they interchangeable? Pacing Clin Electrophysiol36(11):pp. 1391-1401,2013. [32] Auger D, van Bommel RJ, Bertini M, et al.: Prevalence and characteristics of patients with clinical improvement but not significant left ventricular reverse remodeling after cardiac resynchronization therapy. Am Heart J160(4):pp. 737-743,2010. [33] Diaz-Infante E, Mont L, Leal J, et al.: Predictors of lack of response to resynchronization therapy. Am J Cardiol95(12):pp. 1436-1440,2005. [34] Hsu JC, Solomon SD, Bourgoun M, et al.: Predictors of super-response to cardiac resynchronization therapy and associated improvement in clinical outcome: the MADIT-CRT (multicenter automatic defibrillator implantation trial with cardiac resynchronization therapy) study. J Am Coll Cardiol59(25):pp. 2366-2373,2012.

22

[35] Wang DY, Richmond ME, Quinn TA, et al.: Optimized temporary biventricular pacing acutely improves intraoperative cardiac output after weaning from cardiopulmonary bypass: a substudy of a randomized clinical trial. J Thorac Cardiovasc Surg141(4):pp. 1002-8, 1008,2011. [36] Garcia-Bengochea JB, Fernandez AL, Calvelo DS, et al.: Temporary epicardial left ventricular and biventricular pacing improves cardiac output after cardiopulmonary bypass. J Cardiothorac Surg7:p. 113,2012. [37] Godje O, Friedl R, Hannekum A: Accuracy of beat-to-beat cardiac output monitoring by pulse contour analysis in hemodynamical unstable patients. Med Sci Monit7(6):pp. 1344-1350,2001. [38] Tang R, Noh HJ, Wang D, et al.: Candidate genes and functional noncoding variants identified in a canine model of obsessive-compulsive disorder. Genome Biol15(3):p. R25,2014. [39] Steendijk P, Tulner SA, Bax JJ, et al.: Hemodynamic effects of long-term cardiac resynchronization therapy: analysis by pressure-volume loops. Circulation113(10):pp. 12951304,2006.

Preoperative characteristics of patients, QRS≥ 120ms

QRS<120ms Patients, n Age, yrs (range) Weight, kg (sd) Height, cm (sd) Male/female ratio, n ASA III, n (%) Surgical procedure, n CABG

7 62 (47-80) 83,5 (16,1) 174 (7) 7/0 3 (100)

8 68,6 (59-78) 81,3 (7.2) 171 (10) 7/1 3 (100)

6

8 23

CABG + valve repair LVEF, % (sd) Rhythm Sinus (n)

1 32,3 (7.3)

Atrium Fibrilation (n) Pacemaker Rhythm (n) QRS duration, msec, mean (sd) PR time, msec (sd) Normal QRS duration LBBB Incomplete LBBB RBBB CPB Time, min (sd) AoX time, min (sd) Graft number, mean Type of cardioplegia (n) Blood St. Thomas Cardioplegia dose, ml (sd) Blood St. Thomas

0 28,8 (4.1)

6

8

0 0

0 1

106 (5,8) 177 (30,8)

148 (28.4) 156 (25,0) 4 0 3 0

68 (24) 52 (20)

484 (520) 1095 (224)

0 7 0 1 69 (23) 44 (11)

4

4

2 5

1 7 555 (-) 818 (429)*

Abbreviations: ASA Classification = Anesthesiological risk assessment system of the American Society of Anesthesiologists; CABG = Coronary Artery Bypass Grafting; LVEF = left ventricular ejection fraction; LBBB = Left Bundle Branch Block; RBBB = Right Bundle Branch Block; CPB time = Cardio Pulmonary Bypass time; AoX time = Crossclamp time; sd = standard deviation; n = number*Two times missing dose.

24

Use of inotropic drugs after CPB Type Any inotropic drug Dobutamine (µg/kg/min) Milrinon (µg/kg/min) Norepinepherin (µg/kg/min)

n (%) 7 (47%) 4 (27%) 1 (7%) 3 (20%)

Mean dose (range) 4,4 (3,7-6,3) 0,08 0,04 (0,03-0,07)

Abbreviations: CPB = cardiopulmonary bypass

25

26