Effects of mid-line thoracotomy on the interaction between mechanical ventilation and cardiac filling during cardiac surgery

D. A. Reuter1, T. Goresch1, M. S. G. Goepfert1, S. M. Wildhirt2, E. Kilger1 and A. E. Goetz*,1

1 Department of Anaesthesiology and 2 Department of Cardiac Surgery, University of Munich, Germany

*Corresponding author. E-mail: alwin.goetz{at}med.uni-muenchen.de

Accepted for publication: January 8, 2004


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Background. Mid-line thoracotomy is a standard approach for cardiac surgery. However, little is known how this surgical approach affects the interaction between the circulation and mechanical ventilation. We studied how mid-line thoracotomy affects cardiac filling volumes and cardiovascular haemodynamics, particularly variations in stroke volume and pulse pressure caused by mechanical ventilation.

Methods. We studied 19 patients during elective coronary artery bypass surgery. Before and after mid-line thoracotomy, we measured arterial pressure, cardiac index (CI) and global end-diastolic volume index (GEDVI) by thermodilution, left ventricular end-diastolic area index (LVEDAI) by transoesophageal echocardiography and the variations in left ventricular stroke volume and pulse pressure during ventilation by arterial pulse contour analysis.

Results. After thoracotomy, CI increased from 2.3 (0.4) to 2.9 (0.6) litre min–1 m–2, GEDVI increased from 605 (110) to 640 (94) litre min–1 m–2, and LVEDAI increased from 9.2 (3.7) to 11.2 (4.1) cm2 m–2. All these changes were significant. In contrast, stroke volume variation (SVV) decreased from 10 (3) to 6 (2)% and pulse pressure variation (PPV) decreased from 11 (3) to 5 (3)%. Before thoracotomy, SVV and PPV significantly correlated with GEDVI (both P<0.01). When the chest was open, similar significant correlations of SVV (P<0.001) and PPV (P<0.01) were found with GEDVI.

Conclusion. Thoracotomy increases cardiac filling and preload. Further, thoracotomy reduces the effect of mechanical ventilation on left ventricular stroke volume. However, also under open chest conditions, SVV and PPV are preload-dependent.

Br J Anaesth 2004; 92: 808–13

Keywords: anaesthetic techniques, cardiovascular; heart, cardiac index; monitoring, cvs; surgery, cardiovascular


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Coronary artery bypass grafting is a very common surgical procedure. Patients having coronary artery bypass grafts often have a reduced cardiovascular reserve and can be unstable during surgery. Assessing the effects of mechanical ventilation on the circulation has been useful for monitoring cardiac preload and predicting the response to fluid therapy.1 2 Retrospective analysis of systolic pressure variation (SPV) or pulse pressure variation (PPV), can quantify these interactions, both experimentally and clinically.3 4 The on-line measurement of cardiac output and stroke volume by arterial pulse contour analysis allows clinical assessment of variations in left ventricular stroke volume (SVV), which cause SPV and PPV.5 6 Measuring SVV allows cardiac preload to be monitored and the response to fluid administration to be predicted.79

Mid-line thoracotomy and pericardiotomy is the most common surgical approach for coronary artery bypass grafting. So during most of the surgery, the constraint on the heart, normally provided by the thoracic wall and the pericardium, is reduced or absent. The haemodynamic effects of thoracotomy and pericardiotomy in such patients have not been studied extensively.1012 Further, the effects of mid-line thoracotomy on heart–lung interactions have not been studied. Therefore, we measured haemodynamics and specifically SVV and PPV to assess heart–lung interactions before and after mid-line thoracotomy and pericardiotomy in patients having coronary artery bypass surgery.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The study was approved by the Institutional Review Board. We studied 20 patients about to have elective coronary artery bypass grafts, who all had a normal left ventricular ejection fraction (>0.50) before surgery. All patients gave written informed consent.

Before induction of anaesthesia, a 5 Fr thermistor-tipped catheter (PV2025L20, Pulsion Medical Systems AG, Munich, Germany) was inserted into a femoral artery and advanced into the distal abdominal aorta. The catheter was then connected to a haemodynamics monitor (PiCCO, V 5.1, Pulsion Medical Systems AG, Munich, Germany). Anaesthesia was induced with midazolam 0.1–0.15 mg kg–1 and sufentanil 0.6–1.0 µg kg–1. Orotracheal intubation was facilitated with pancuronium 0.1–0.15 mg kg–1. Anaesthesia was then maintained with isoflurane and continuous administration of sufentanil 1.0–1.8 µg kg–1 h–1. After induction of anaesthesia, an 8 Fr central venous catheter (Arrow, Reading, PA, USA) was inserted into an internal jugular vein. Both pressure transducers were positioned at the level of mid-axillary line and zeroed to atmospheric pressure.

Volume controlled mechanical ventilation was adjusted to obtain tidal volumes of 10 ml kg–1 of ideal body weight, with the ratio of inspiration:expiration fixed to 1:2 and a positive end-expiratory airway pressure of 5 cm H2O.

Haemodynamic monitoring
Arterial thermodilution. We used a thermistor-tipped arterial catheter to allow measurement of cardiac output by arterial thermodilution as described previously.6 13 Fifteen millilitres of iced saline 0.9% were injected into the central venous catheter at random times in the respiratory cycle. The thermal signal was detected at the tip of the arterial catheter in the distal abdominal aorta. The global end-diastolic volume, which is the volume of the four chambers of the heart, can be calculated from the mean transit time, the exponential decay time of the thermal indicator, and thermodilution cardiac output, as described previously.14 All measurements of cardiac output and global end-diastolic volume by thermodilution were performed in triplicate and the thermodilution curves were inspected to detect artefacts.

Arterial pulse contour analysis. The arterial catheter also allows continuous monitoring of arterial pressure, left ventricular stroke volume (SV), and cardiac output by pulse contour analysis, based on an algorithm developed by Wesseling and colleagues.15 With this original algorithm, SV is calculated by measuring the area under the systolic portion of the aortic pressure waveform and dividing this area by the aortic impedance, which is determined by arterial thermodilution. The algorithm used in this investigation (PiCCO, V 5.1, Pulsion Medical Systems AG, Munich, Germany) also analyses the actual shape of the pressure waveform and takes into account the patient’s aortic compliance and systemic vascular resistance as described previously.6 16 We first calibrated pulse contour cardiac output by making three consecutive measurements of thermodilution cardiac output as mentioned above. After this, heart rate, arterial pressure, pulse contour cardiac output, and stroke volume were recorded continuously.

SVV is the variation of stroke volume expressed as percentage of the mean value during the respiratory cycle and is calculated as:79

SVV=(SVmax – SVmin)/SVmeanx100

SVmean was determined using a continuously sliding time window of 30 s. This time window is further divided in four 7.5 s periods; within each period the greatest and the smallest value of SV (SVmax and SVmin, respectively) were determined and the average of the four 7.5 s intervals were used to calculate SVV.

Similarly, PPV describes the ventilation-induced variation of beat-to-beat pulse pressure (i.e. the difference between systolic and diastolic arterial pressure) as a percentage of mean pulse pressure:4

PPV=(PPmax – PPmin)/PPmeanx100

The same time window was used for assessment of SVV and PPV.

For each measurement, both SVV and PPV were recorded continuously over a period of 3 min, after which the mean was calculated.

Transoesophageal echocardiography
Transoesophageal echocardiographic (TOE) measurements were all done at end-expiration using the transgastric mid-papillary short-axis view of the left ventricle to image both the posteromedial and anterolateral papillary muscle. We used a Hewlett-Packard omni plane probe (HP 21364A) and a HP SONOS Phased Array Imaging System (Hewlett-Packard, Andover, MA, USA). TOE images and the ECG signal were recorded simultaneously on videotape for later off-line analysis. End-diastole was defined as the largest left ventricular cross sectional area immediately after R-wave peak in the ECG. Left ventricular end-diastolic area was traced edge-to-edge, including the papillary muscles. For each measurement, three consecutive cardiac cycles were analysed and the average was determined.

Study plan
During the short study period, no fluids were infused. Cardiovascular measurements were made at two times: T1, after skin incision and before thoracotomy; T2, after opening the thorax and pericardium, with the sternal retractor in place.

Statistical analysis
All raw data were indexed by body surface area to calculate thermodilution cardiac index (CI), global end-diastolic volume index (GEDVI) and left ventricular end-diastolic area index (LVEDAI). All data are expressed as mean (SD). Liliefort’s modification of the Kolmogorow–Smirnov test, the paired t-test and the Pearson product moment correlation were used for statistical analysis. A value of P<0.05 was considered significant.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
We studied four female and 16 male patients. One patient with an unstable circulation, which required treatment before thoracotomy, was excluded. The remaining 19 patients (mean age 58 (12) [42–76] yr, mean body mass index 27.2 (4.6) kg m–2) were haemodynamically stable and received no cardiovascular agents during the observation.

Measurements before mid-line thoracotomy (T1) and after pericardiotomy (T2) are listed in Table 1. The individual values of GEDVI within each triplicate measurement were within 10%. The individual changes of the ventilation-induced SVV and PPV are shown in Figure 1A and B. Minute ventilation (7.4 (1) litre min–1) and tidal volumes (695 (100) ml) did not differ before and after thoracotomy. Peak airway pressure decreased slightly immediately after thoracotomy (from 25 (3) to 23 (3) cm H2O) as did mean airway pressure (from 10 (1) to 9 (2) cm H2O). However, neither pressure decrease was significant.


View this table:
[in this window]
[in a new window]
 
Table 1 All values are mean (SD). SAP, systolic arterial pressure; MAP, mean arterial pressure; DAP, diastolic arterial pressure; HR, heart rate; CVP, central venous pressure; GEDVI, global end-diastolic volume index; LVEDAI, left ventricular end-diastolic area index; SVV, left ventricular stroke volume variation, PPV, aortic pulse pressure variation
 


View larger version (14K):
[in this window]
[in a new window]
 
Fig 1 (A) SVV of each patient before (T1) and immediately after (T2) mid-line thoracotomy and pericardiotomy. (B) Aortic PPV of each patient before (T1) and immediately after (T2) mid-line thoracotomy and pericardiotomy.

 
Overall, measurements of SVV and PPV were closely correlated (R=0.89; P<0.001).

Before thoracotomy, SVV (R=–0.57; P<0.01) and PPV (R=–0.58; P<0.01) were inversely correlated with GEDVI, as shown in Figure 2A and B. These correlations were also noted after thoracotomy and pericardiotomy (SVV R=–0.70; P<0.001 and PPV R=–0.62; P<0.01) (Fig. 2C and D).



View larger version (20K):
[in this window]
[in a new window]
 
Fig 2 (A) Relation between GEDVI and ventilation-induced SVV before thoracotomy. The 95% confidence interval of the values is shown (R=–0.57; P<0.01). (B) Relation between GEDVI and ventilation-induced aortic PPV before thoracotomy. The 95% confidence interval of the values is shown (R=–0.58; P<0.01). (C) Relation between GEDVI and ventilation-induced SVV immediately after thoracotomy and pericardiotomy. The 95% confidence interval of the values is shown (R=–0.70; P<0.001). (D) Relation between GEDVI and ventilation-induced aortic PPV immediately after thoracotomy and pericardiotomy. The 95% confidence interval of the values is shown (R=–0.62; P<0.001).

 
Measurements of LVEDAI before and after thoracotomy were significantly correlated with GEDVI (R=0.49; P<0.05).


    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Dynamic measures of the cyclic effects of mechanical ventilation on cardiovascular function have been evaluated by different groups for assessing cardiac preload, and to optimize fluid therapy in mechanically ventilated patients.1 2 The results were promising.4 79 The rationale is that changes in intrathoracic pressure during each mechanical breath cause transient changes in cardiac preload, which, according to the Frank–Starling mechanism, change left ventricular SV.17 Thus, each mechanical breath imposes a small volume loading and un-loading manoeuvre. The change in stroke volume during the respiratory cycle depends on the responsiveness of the patients’ left ventricle that is the position on the Starling curve. The Starling (or ventricular function) curve describes the relation between preload and stroke volume.18 A steep slope of the Starling curve is associated with large SVV, whereas a shallow slope is associated with small SVV. Thus, high SVV suggests that stroke volume and cardiac output can be improved by fluid loading. Conversely, if SVV is low in a hypotensive patient, catecholamine therapy is more appropriate.

This means of functional preload monitoring has not been studied with the chest open in patients undergoing cardiac surgery, despite that during these procedures, optimization of preload may be needed to obtain haemodynamic stability.

We found that SVV and PPV were less after thoracotomy and GEDVI was greater.

The immediate effects of thoracotomy on biventricular preload are not agreed.1012 19 However, assessment of central blood volumes by GEDVI and intrathoracic blood volume by transcardiopulmonary thermodilution is clinically useful for estimation of cardiac preload volume.13 14 20 21 The greater GEDVI in our patients after thoracotomy suggests that cardiac preload volume is increased by opening the chest. This is supported by the increase in LVEDAI and the significant correlation between GEDVI and LVEDAI measurements that we found. Adrenergic responses to sternotomy may also have contributed to the changes in haemodynamics that we found.10

We described recently changes of SVV and PPV after cardiac surgery when tidal volume was changed.22 SVV increased with larger tidal volumes and decreased when tidal volume was reduced. At each tidal volume, SVV decreased after fluid was given to volume-responsive patients. We argued that if the force affecting these heart–lung interactions (i.e. the size of the mechanical breath) changed within the closed thorax, SVV had to change as well. With the same reasoning for thoracotomy, SVV should decrease when the constraint of the thoracic wall and the pleura is released. These interpretations of our previous findings were discussed in a commentary by Michard and colleagues,23 who suggested that altering tidal volume might itself affect fluid responsiveness. If the tidal volume (i.e. the force responsible for SVV and PPV) increases, the mean cardiac preload is consequently reduced, leading to a leftward shift on the Starling curve. Thus, adequately resuscitated patients, who are preload-independent, may become preload-dependent if tidal volume is increased with a subsequent increase in intrathoracic pressure.

Thus, there may be two potential components, which could contribute to the decrease in SVV and PPV after thoracotomy. First, SVV and PPV decrease because intrathoracic pressure varies less during the ventilatory cycle after opening the chest. This is consistent with the findings of Pizov, who studied changes in SPV during ventilation-synchronized chest compression.24 The second component is that preload volume also increases after thoracotomy, as shown by the increases of GEDVI and LVEDAI in our patients. Thoracotomy thus changes SVV and PPV as the patients’ left ventricular function shifts to the right on the Starling curve. The increase in preload volume makes the patients less preload-dependent. We tested this hypothesis by analysing the relation between preload volume and both SVV and PPV. Before thoracotomy, we found significant correlations of both SVV and PPV with GEDVI, showing the preload dependency of the heart–lung interactions under closed thorax conditions that are the conditions of previous studies.4 79 However, SVV and PPV were significantly correlated with GEDVI after thoracotomy as well. Thus, SVV and PPV are also dependent on preload under open thorax conditions. This suggests that functional preload monitoring may be suitable for detecting volume responsiveness under open thorax conditions and that monitoring changes in SVV and PPV can help to guide fluid therapy during cardiac surgery. However, these findings need to be confirmed and extended by investigating changes in SVV and PPV by fluid loading with the chest open. We should also define threshold values indicating volume responsiveness in this clinical situation.

In summary, we found that thoracotomy increases preload volume and reduces SVV and PPV in cardiac surgery patients. SVV and PPV are preload-dependent, both under closed chest conditions and with the thorax opened. We conclude that measuring SVV or PPV can allow assessment of responsiveness to fluid therapy during cardiothoracic surgical procedures requiring thoracotomy.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest 2002; 121: 2000–8[Abstract/Free Full Text]

2 Pinsky MR. Functional hemodynamic monitoring. Intensive Care Med 2002; 28: 386–8[CrossRef][ISI][Medline]

3 Perel A, Pizov R, Cotev S. Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded hemorrhage. Anesthesiology 1987; 67: 498–502[ISI][Medline]

4 Michard F, Boussat S, Chemla D, et al. Relation between respiratory changes in arterial pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care 2000; 162: 134–8[Abstract/Free Full Text]

5 Frank O. Schätzung des Schlagvolumens des menschlichen Herzens aufgrund der Wellen- und Windkesseltheorie. Z Biol 1930; 90: 405

6 Felbinger TW, Reuter DA, Eltzschig HK, Moerstedt K, Goedje O, Goetz AE. Comparison of pulmonary arterial thermodilution and arterial pulse contour analysis: evaluation of a new algorithm. J Clin Anesth 2002; 14: 296–301[CrossRef][ISI][Medline]

7 Berkenstadt C, Margalit N, Hadani M, et al. Stroke volume variation as a predictor of fluid responsiveness in patients undergoing brain surgery. Anesth Analg 2001; 92: 984–9[Abstract/Free Full Text]

8 Reuter DA, Felbinger TW, Schmidt C, et al. Stroke volume variations for assessment of cardiac responsiveness to volume loading in mechanically ventilated patients after cardiac surgery. Intensive Care Med 2002; 28: 392–8[CrossRef][ISI][Medline]

9 Reuter DA, Kirchner A, Felbinger TW, Kilger E, Lamm P, Goetz AE. Usefulness of left ventricular stroke volume variation to assess fluid responsiveness in patients with reduced cardiac function. Crit Care Med 2003; 31: 1399–404[CrossRef][ISI][Medline]

10 Hoar PF, Stone JG, Faltas AN, Bendixen HH, Head RJ, Berkowitz BA. Hemodynamic and adrenergic responses to anesthesia and operation for myocardial revascularization. J Thorac Cardiovasc Surg 1980; 80: 242–8[ISI][Medline]

11 Estefanous FG, Urzua J, Yared JP, Zurick AM, Loop FD, Tarazi RC. Pattern of hemodynamic alterations during coronary artery operations. J Cardiovasc Surg 1984; 87: 175–82

12 Milocco I, Löf BA, William-Olsson G, Appelgren LK. Haemodynamic stability during anaesthesia induction and sternotomy in patients with ischaemic heart disease. A comparison of six anaesthetic techniques. Acta Anaesthesiol Scand 1985; 29: 465–73[ISI][Medline]

13 Hoeft A, Schorn B, Weyland A, et al. Bedside assessment of intravascular volume status in patients undergoing coronary bypass surgery. Anesthesiology 1994; 81: 76–86[ISI][Medline]

14 Sakka SG, Ruhl CC, Pfeiffer UJ, et al. Assessment of cardiac preload and extravascular lung water by single transpulmonary thermodilution. Intensive Care Med 2000; 26: 180–7[CrossRef][ISI][Medline]

15 Wesseling KH, de Wit B, Weber JAP, et al. A simple device for the continuous measurement of cardiac output. Adv Cardiovasc Phys 1983; 5: 16–52

16 Goedje O, Hoeke K, Goetz AE, et al. Reliability of a new algorithm for continuous cardiac output determination by pulse contour analysis during hemodynamic instability. Crit Care Med 2002; 30: 52–9[ISI][Medline]

17 Jardin F, Farcot JC, Gueret P, et al. Cyclic changes in arterial pulse during respiratory support. Circulation 1983; 83: 266–74

18 Sonnenblick EH, Strobeck JE. Current concepts in cardiology: derived indexes of ventricular and myocardial function. N Engl J Med 1977; 296: 978–82[ISI][Medline]

19 Werner O, Malmkvist A, Beckman S, Stahle S, Nordström L. Gas exchange and haemodynamics during thoracotomy. Br J Anaesth 1984; 56: 1342–9

20 Della Rocca G, Costa MG, Coccia C, Pompei L, Di Marco P, Pietropaoli P. Preload index: pulmonary artery occlusion pressure versus intrathoracic blood volume monitoring during lung transplantation. Anesth Analg 2002; 95: 835–43[Abstract/Free Full Text]

21 Reuter DA, Felbinger TW, Moerstedt K, Kilger E, Weis F, Goetz AE. Intrathoracic blood volume index by thermodilution for preload monitoring after cardiac surgery. J Cardiothorac Vasc Anesth 2002; 16: 191–5[CrossRef][ISI][Medline]

22 Reuter DA, Bayerlein J, Goepfert M, et al. Influence of tidal volumes on left ventricular stroke volume variation. Intensive Care Med 2003; 29: 476–80[ISI][Medline]

23 Michard F, Teboul JL, Richard C. Influence of tidal volume on stroke volume variation. Does it really matter? Intensive Care Med 2003; 29: 1613[CrossRef][ISI][Medline]

24 Pizov R, Ya'ari Y, Perel A. The arterial pressure waveform during acute ventricular failure and synchronized external chest compression. Anesth Analg 1989; 68: 150–6[Abstract]