Approach to the haemodynamically unstable cardiac surgical patient

There is an important subset of cardiothoracic ICU-related  CICM SAQs which interrogate the candidate's understanding of life-threatenening post-CABG complications. They are frequently repeated.
Such questions have included the following:

This was supposed to be a condensed revision of Ruesch and Levy's chapter on the practical aspects of post-op care for an unstable cardiothoracic surgical patient. That chapter, and the entire book, can be found here - thanks to the good people of Unfortunately, this summary is in advanced stages of apocryphal bloat. 

There are predictable reasons for why a post-bypass patient might become unstable, and I have attempted to group them all together. Specifically, haemorrhage is a major cause, and needs to be looked at in detail.

Causes of cardiovascular instability after bypass

Immediately recognisable on direct inspection

  • Artifactual - check your lines, zero to recalibrate
  • Anaphylaxis
  • Arrhythmia, eg. AF
  • Valve failure - eg. mitral regurgitation of sudden onset

Immediately recognisable on routine investigations

  • Myocardial ischaemia should not be left unrecognized
  • Pneumothorax or tension pneumothorax
  • Cardiac Tamponade
  • Haemorrhage
  • LV outflow tract obstruction

Excluded by inspection and investigations

  • Rewarming-related vasodilation
  • Excessive sedation with propofol
  • Post-bypass vasoplegia
  • Post-bypass myocardial depression

The time-poor exam candidate may only require this table of differentials.

An exploration of these abovementioned categories occurs below, but is non-essential.

Causes of instability - Immediately recognizable on direct inspection

  • Artifactual - check your lines, zero to recalibrate
  • Anaphylaxis - look for urticaria and angioedema
  • Arrhythmia, eg. AF (or pacing wire failure) - should be immediately obvious from looking at the monitor
  • Valve failure - mitral regurgitation of sudden onset may be obvious on the PA catheter waveform (it will look rather like the CVP waveform in tricuspid regurgitation). Alternatively, one should actually listen to the chest to look for murmurs.

All of these causes can be managed reasonably rapidly; apart from valve failure (which is a strong argument for an immediate return to theatre)

Causes of instability -Immediately recognizable on routine investigations

  • Get an ECG:
    • Myocardial ischaemia should not be left unrecognized
  • Get a chest Xray (or the echo probe)
    • Pneumothorax or tension pneumothorax may be the cause of obstructive shock
    • Cardiac Tamponade may be recognizable as a change in the mediastinal contour, or by finding on echo that the chambers are small and collapsed, surrounded by debris.
  • Get an ABG:
    • Haemorrhage sufficient to cause hemodynamic instability will manifest as a drop in Hb. Haemorrhage in the post-operative period is common enough and the complications of bleeding into the mediastinum or pericardium are unique enough to merit their own area of discussion.
  • Get a TTE:
    • LV outflow tract obstruction should be something you can anticipate if the patient is known to have LV hypertrophy or HOCM
    • Valve dysfunction which could not be auscultated may be discovered in this way
    • Cardiac tamponade may be discovered by skilled TTE technicians which was not picked up on crude bedside echosonography (eg. a retrocardiac clot from mitral valve surgery- hard to see on either transthoracic echo or Xray)

Causes of instability - Excluded by inspection and investigations

  • Rewarming-related vasodilation
  • Excessive sedation with propofol
  • Post-bypass vasoplegia
  • Post-bypass myocardial depression

 About these, not much can be done except to funnel more vasopressors and inotropes into your patient.

Immediate examination and investigations:

  • Auscultate chest and check trachea (midline?)
  • ABG (for haemoglobin and lactate)
  • Recalibrate arterial line
  • 12-lead ECG
  • Urgent CXR
  • Urgent TTE

Haemodynamic variables one may manipulate

  • Preload
  • Rate
  • Rhythm
  • Contractility
  • Afterload

Immediate management

Let us organise it into a system of sorts.

My simple mind favours an alphabetic system, resembling a nursery rhyme.  A is for airway management, B is for mechanical ventilation, C is for cookie. Et cetera.


  • If the patient has already been extubated, there may not be a strict indication to reintubate them, but reintubation would make the management of the rest of their problems much easier.


  • Maintain  SpO2 ~ 100% to maintain satisfactory tissue oxygen delivery.
  • Adjust positive pressure, with two goals in mind:
    • Enhancement of preload by controlling the intrathoracic venous flow
    • Enhancement of afterload by increasing LV transmural pressure
    • Some kind of tamponade effect? The college answer to Question 13 from the first paper of 2012 recommends increasing the PEEP to 10. Pushing against the bleeding with PEEP is a standard cardiothoracic ICU thing people are taught to do, but it may not have much evidence to support it. Considering that this PEEP corresponds to about 7.4 mmHg, and that only a fraction of the PEEP is transmitted to the mediastinal contents, it seems illogical to expect it to work magic on a post-operative bleeder. Still, people do it. Ilabaca et al (1980) reported a case series where bleeding cardiac surgical patients stopped bleeding when a PEEP of 10-15 was applied. Remember that in the 1980s very low PEEP was often the default, so any PEEP seemed revolutionary.


  • Preload
    • Maintain satisfactory diastolic pressure with noradrenaline and/or vasopressin
    • Ensure there is no cardiac tamponade (it makes for suboptimal preload)
    • Give a fluid bolus; ensure there is satisfactory filling (consider performing some of the static and dynamic tests for fluid responsiveness). The topic of preload and fluid responsiveness following bypass surgery is interesting enough to have developed extensive bloat (see below).
    • Diastolic failure is a frequent cause of post-cardiotomy hemodynamic instability. Its not just that the myocardium is fat and hypertrophied, leaving a smaller chamber- though that may also be the case. The myocardium is also stiff and oedematous, which decreases its compliance. And on top of that, there is a pericardial effusion, which decreases diastolic filling.  Patients with LV hypertrophy require higher LV filling pressures. Reduced cardiac output in these patients may be due to inadequate preload. A fluid bolus would be the correct response if preload is inadequate.
    • Levosimendan, milrinone and to a lesser extent dobutamine will all improve LV diastolic relaxation. This improves the compliance of the chamber, allowing the extra fluid you have given to improve LV filling.
    • One may consider re-opening the chest. The benefit of delayed sternal closure lies in the analysis of what precisely keeps the cardiac output down post-operatively. Turns out a fair proportion of post-cardiotomy myocardial depression is due to diastolic failure. This is how delayed sternal closure produces an improved cardiac output; the myocardium, free to expand as much as it likes, tends to fill more and thus pump better when it is not restricted by the ribcage.
    • Lastly, "preload failure" might actually be the consequence of catastrophic post-operative haemorrhage. Haemorrhagic complications of coronary artery bypass grafting are discussed in greater detail in a chapter dedicated to that specific topic, and I won't go into too much detail here- suffice to say one always needs to consider massive haemorrhage as a differential.
  • Right ventricular failure
    • There are many reasons why the right ventricle might struggle post-bypass, not the least of which is bypass-associated pulmonary hypertension.
    • Additionally, if fluid is collecting in the pericardium, the right ventricle will notice this first, because it is a lower pressure system.
    • This complex topic can be summarised by saying that the management of a failing right ventricle will be some combination of volume loading and inotrope support. Milrinone is a popular favourite, as it has a tendency to decrease right ventricular afterload.
    • Again, an open chest is going to be helpful because the right ventricle may fill better without the restriction of a closed chest cavity.
  • Rate
    This is pretty straightforward. Cardiac output = heart rate x stroke volume. If you cant do much for the stroke volume, you can try to increase the heart rate. Obviously, if there is diastolic failure, this strategy will begin to fail in proportion to the rate increase. In most other circumstances this is a good way to go. With the availability of epicardial pacing leads, one can choose to set the pacemaker box to atrial pacing, which would take over at rates below 70 or 80. If you have just had nice new bypass vessels grafted to your coronary arteries, then perhaps ischaemia should not be such a major concern, particularly at sensible rates.
    • Maintain heart rate with pacing.
    • Consider increasing rate to 90 in the absence of significant valve disease.
    • Chronotropes which may be appropriate would probably include milrinone and dobutamine; one's choice would depend on the presence or absence of pulmonary hypertension.
    • Rate may be the only variable left to optimise. One would need to consider the fact that the stroke volume is fixed; that is to say, the hypertrophied left ventricle has a chamber with reduced compliance, with a fixed (small) volume. Thus, once filling pressure, LV relaxation and the "atrial kick" are optimal, one's means of increasing cardiac output are limited to increasing the heart rate. This can be achieved by use of pacing. Ideally, one would pace the atrium, because the synchronised atrial-ventricular system pumps more effectively than a dyssynchronous one. Obviously, there will be a point at which one loses this benefit. At very rapid heart rates, the LV does not have time to fill.
  • Rhythm
    The best rhythm to be in is of course sinus. The addition of atrial contraction to ventricular preload is such that up to 30% of cardiac output could be lost if this atrial  "kick" disappears with AF. Thus, either keep them in sinus rhythm chemically, or control their rhythm with atrial pacing.  Give up and pace the ventricle if you need to pace at all.
    • Rhythm control needs to be achieved with pacing and antiarrhythmic drugs (eg. amiodarone).
    • The addition of "atrial kick" is a welcome cardiac output enhancement. Contraction of the left atrium is crucial for the patient in diastolic failure; poor left ventricular filling can result from a lazy atrium failing to inject a sufficient amount of blood into the ventricular chamber. Thus, a patient in AF will find their hypertrophied left ventricle empty, and haemodynamic compromise will ensue. Control of the rhythm is achieved mainly by amiodarone. Digoxin is pretty well useless in this setting:  if you're full of catecholamines, it simply won't work. Similarly beta-blockade (though usually very helpful) may be counterproductive. Cardioversion should be considered only if there is severe haemodynamic compromise.
    • Of course, this wont always be possible. If the patient has been in AF for the last 20 years, you have no chance of converting this to sinus rhythm by any means.
    • Thankfully, "new AF" post bypass is usually transient. 20-30% revert within 2 hours, and up to 80% revert within 24 hours. The linked-to article suggests that sotalol and amiodarone may be equivalent in efficacy when it comes to chemical cardioversion.
  • Contractility
    • In general, the choice of agent is a topic is interesting enough to merit a chapter to compare the available inotropes for post-bypass cardiogenic shock.
    • In brief, one can increase RV + LV contractility with milrinone or levosimendan; dobutamine probably acts more as a chronotrope than an inotrope, and has little effect on pulmonary vascular resistance, which makes it a second choice.
    • Ensure that the patient has not infarcted their grafts,  as this would create an environment of unsatisfying contractility.
    • On top of the abovementioned diastolic failure, the post-bypass myocardium simply refuses to pump in systole- because its hypothermic, because it is "stunned", because there has been some reperfusion injury, or because the systemic vascular resistance is incredibly high (post hypothermia).
  • Afterload
    Well, one is slightly less likely to be absurdly hypertensive after bypass surgery, in comparison to being hypotensive. However, it i possible to have increased afterload while having low systemic arterial blood pressure. Increased afterload is not desirable. GTN, sodium nitroprusside and milrinone all contribute to systemic arterial vasodilation, which in turn decreases myocardial workload and oxygen consumption.
    • Decrease RV and LV afterload with milrinone or levosimendan (pulmonary and systemic vasodilation). Aortic pressure, if it is too high, will increase myocardial oxygen demand, and will decrease the ejection fraction, thus decreasing the cardiac output. Thus, one may consider using vasodilators to reduce afterload.
    •  IABP may be indicated to decrease myocardial workload with especially if there is evidence of ischaemia with inotrope use.
    • Increase the diastolic arterial blood pressure with vasopressors.  By increasing the  diastolic blood pressure, one tends to improve the diastolic filling of coronary arteries. Unfortunately, vasopressors tend to be counterproductive for this task, as increasing the systolic pressure also increases afterload. The main contribution of vasopressors is actually in causing venoconstriction, thus increasing the preload, and improving left ventricular filling. Apart from the IABP, there is no specific agent which acts specifically on the diastolic pressure.
  • In the opposite scenario, systemic arterial HYPOtension is treated with dear old familiar noradrenaline, and with vasopressin if it comes to that.

    It would seem, in fact, that vasopressin may be better.

  • Uncontrollable vasoplegia with adequate cardiac output
    • Noradrenaline is the first choice; vasopressin may be added as a noradrenaline-sparing agent.
    • Consider methylene blue as a third agent in cases of refractory vasoplegia.
    • Angiotensin may also be available.

Sedation, analgesia, paralysis:

  • Sedate and paralyse patient to decrease whole-body oxygen demand.

Electrolyte and environment optimisation:

  • Re-warm patient

Haematology: blood transfusion and correction of coagulopathy

  • Keep the haemoglobin around 80. It seems to be the ideal value.
  • Too little haemoglobin may result in tissue hypoxia, especially given that the cardiac output is fluctuating.
  • Too much haemoglobin adds to the myocardial workload by increasing the mass of the blood which may end up being counterproductive.


Frederick A. Hensley, Jr., M.D., Donald E. Martin, M.D.,  Glenn P. Gravlee, M.D. A Practical Approach to Cardiac Anaesthesia, 3rd ed. Sibylle A. Ruesch and Jerrold H. Levy. CHAPTER 9. The Postcardiopulmonary Bypass Period: A Systems Approach. 2003 by LIPPINCOTT WILLIAMS & WILKINS

André, Arthur C. St, and Anthony DelRossi. "Hemodynamic management of patients in the first 24 hours after cardiac surgery." Critical care medicine 33.9 (2005): 2082-2093.

Eagle, Kim A., et al. "ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery).Circulation 110.14 (2004): e340.

Goepfert, Matthias SG, et al. "Goal-directed fluid management reduces vasopressor and catecholamine use in cardiac surgery patients." Intensive care medicine 33.1 (2007): 96-103.

Marik, Paul E. "Hemodynamic parameters to guide fluid therapy." Transfusion Alternatives in Transfusion Medicine 11.3 (2010): 102-112.

Almassi, G. Hossein, et al. "Atrial fibrillation after cardiac surgery: a major morbid event?." Annals of surgery 226.4 (1997): 501.

Maisel, William H., James D. Rawn, and William G. Stevenson. "Atrial fibrillation after cardiac surgery.Annals of Internal Medicine 135.12 (2001): 1061-1073.

BUCKLEY, MORTIMER J., et al. "Intra-aortic balloon pump assist for cardiogenic shock after cardiopulmonary bypass." Circulation 48.1S3 (1973): III-90.

Licker, Marc, et al. "Clinical Review: Management of weaning from cardiopulmonary bypass after cardiac surgery.Annals of cardiac anaesthesia15.3 (2012).

Lavana JD, Fraser JF, Smith SE, Drake L, Tesar P, Mullany DV. Influence of timing of intraaortic balloon placement in cardiac surgical patients. J Thorac Cardiovasc Surg 2011;140(1):80-5.

Maas, Jacinta J., et al. "Cardiac Output Response to Norepinephrine in Postoperative Cardiac Surgery Patients: Interpretation With Venous Return and Cardiac Function Curves*." Critical care medicine 41.1 (2013): 143-150.

Hajjar, L., et al. "Vasopressin Versus Norepinephrine for the Management of Shock After Cardiac Surgery (VaNCS study): a randomized controlled trial." Critical Care17.Suppl 2 (2013): P222.

Ilabaca, Patricio A., John L. Ochsner, and Noel L. Mills. "Positive end-expiratory pressure in the management of the patient with a postoperative bleeding heart." The Annals of thoracic surgery 30.3 (1980): 281-284.