Sepsis-Induced Cardiomyopathy

As far as meaningful differences go for the intensivist, it’s not clear that SATT differs clinically in a meaningful way from SICM to a bedside practitioner. They are similar in that they tend to present in patients with previous heart disease but not as a direct result of acute coronary syndrome, and they tend to present in patients in the extremes of shock. More importantly, this cardiomyopathy is also transient, and while it tends to be associated with higher levels of support and longer ICU stays, it does not outright predict higher mortality or lower survival. A nice adjunct to this consideration is a previous post on considerations for mitral valve systolic anterior motion in SATT.

Right Ventricular Dysfunction

As might be inferred from the clip in the case vignette, SICM can definitely involve the right ventricle, which is independently tied to worse mortality outcomes. While RV failure tends to co-occur with LV failure, a prospective cohort study performed at the Mayo Clinic found a large number of patients admitted to ICUs with new, isolated RV dysfunction thought to be entirely secondary to sepsis. Though more rare, isolated RV dysfunction was associated both with higher needs for invasive mechanical ventilation and higher one-year mortality. 

Diagnosis

Suspect SICM whenever a patient is requiring escalation of vasoactives beyond what is expected. Echocardiography is really the only way to diagnose this particular illness, without much in the way of sensitive/ specific biomarkers or ECG tracings. Especially once a patient requires a second vasoactive medication, at least a point-of-care echocardiogram is indicated to see if there is an inappropriate lack of contractility/ cardiac function. All abnormal point-of-care results should be followed up with formal echocardiograms and cardiology consultation if possible. 

Management

Everyone needs (some) Preload

A common fear from both ICU and ED providers is the idea that patients will become “flooded” with fluids. This, combined with recent provider distrust in both CMS and the surviving sepsis campaign guidelines’ process for weighing evidence, has led to decreased prescribing of crystalloid volumes in sepsis patients, especially in those with end stage renal disease and heart failure. Twitter-battles and published spats aside, there remains compelling evidence that patients with sepsis may benefit (or at least not be harmed) from administration of some crystalloid in the beginning of their resuscitation. Two large analyses (one prospective from New York State, and one retrospective from the Kaiser Permanente System) have shown convincingly that a 30 ml/kg bolus of IV fluids does not harm people with heart failure and may even improve outcomes. Moreover the definitive works and subsequent meta-analysis of their data that compared goal directed therapy to “usual care” showed essentially no harm with the routine administration of about 2L of routine crystalloid administration. That’s probably because prior to presentation patients were suffering from the insensible losses of poor appetite, fever, etc. Moreover, a recent study showed that vasopressor dose in septic shock only predicts mortality if patients have not received a “priming” dose of IV fluids. 

Whether patients with SICM require a different approach remains to be seen, especially given the obstacles involved in early diagnosis and recognition. For now, standard resuscitation to correct hypovolemia with repeat exams makes the most sense here, though ideally coupled with closer examination intervals. 

Beware of Escalating Afterload

The double-threat of SICM puts a patient’s perfusion in a rather precarious position. Her resuscitation will need to acknowledge the simultaneous need to improve vascular tone and peripheral vascular resistance, but not so much as to overwhelm a sickened heart with more afterload than it can handle. Indeed, some authors have theorized that patients with SICM may at first appear to have “normal” cardiac function in the setting of under resuscitated septic shock, only to have SICM’s poor contractility “unmasked” by the addition of preload (crystalloid) and the addition of vascular tone from an alpha agonist. 

Patients may at first appear to stabilize with standard management only later to fall off the right-ward part of the Frank-Starling Curve once afterload is restored. When this occurs, repeat an exam/ point-of-care echo to make sure no changes have occurred, or that LVEF has not deteriorated. Remember, a struggling heart will do poorly against afterload augmentation without improving contractility. 

Inotropy

It’s highly possible that a patient may have a mixed shock physiology. If there is any doubt regarding which shock phenotype is driving malperfusion, this may be a good time to consider PA catheter placement to get more specific numbers and resuscitative targets. That said, a PA catheter may not be necessary with some careful trial and error. If faced with escalating vasopressor doses and SICM has been recognized, it may be worth trialing a low-moderate dose of dobutamine infusion or inotrope dosing of an epinephrine infusion and trending common endpoints of resuscitation (4). Examples include improving mixed venous O2, urine output, or lactic acid levels. It may also be worth seeing if the addition of dobutamine allows for vasopressor weaning while maintaining an adequate MAP. Given that SICM is usually a transient entity that abates after temporary support, an inotrope might only be needed for a few days. 

Mechanical Circulatory Support

In select cases, there is definitely a role for extracorporeal life support (ECLS). A large retrospective cohort study specifically addressed new cardiogenic shock in sepsis victims and showed improved survival for patients placed on veno-arterial extracorporeal membrane oxygenation (12). However, as with all VA-ECMO and ECLS patients, patient selection is everything when it comes to survival. Advanced chronic disease, malignancy, baseline neurologic status, and age all need to be weighed to gauge appropriateness of ECLS salvage. However, if vasoactive/ inotropic therapy does not improve tissue oxygenation rapidly, early evaluation at an advanced center for ECLS is appropriate. 

Case Conclusion

A formal echocardiogram confirmed 15% LVEF with severe dilation of the LV and RV and moderate RV dysfunction. The patient was started on angiotensin II infusion which rapidly diminished her pressor requirements, and 6 hours later she remained on only a moderate dose of angiotensin II, vasopressin, and 0.2 mcg/kg/min of norepinephrine. At that time, dobutamine was added at 5 mcg/min which allowed rapid titration and discontinuation of all other vasoactives. 24 hours later she remained on low dose-dobutamine infusion and had normalized her lactic acid levels and hemodynamics. For her worsening kidney failure and anuria, she was started on CRRT which she tolerated well. After 6 days her cardiac function had almost completely normalized, and dobutamine was discontinued. She was extubated shortly thereafter.