Care for the patient following a TAVR

These people are by definition unfit for any sort of major surgery, which makes them fragile and susceptible to complications. And they bring these complications to the ICU, because that is where the worst of them often end up. Fast tracking of patients through high-volume TAVR services only deflects the straightforward femoral punctures through a well-oiled CCU pathway, leaving a volume of crumbly high-risk patients in the hands of the intensivist. It is therefore not surprising that TAVR complications should have come up in Question 23 from the first paper of 2023, nor that the exam candidates appeared to be well-prepared to handle them. 

The best peer-reviewed resource for this actually happens to be a book chapter; specifically Chapter 60 from Interventional Cardiology: Principles and Practice (2023, 3rd ed). It is written by Matthew I. Tomey, who seems to have been writing about this for about ten years (see Tomey et al, 2013), which means the chapter probably represents the distillation of some considerable personal experience as well as a summary of the evidence.

But before we continue, is it TAVI or TAVR? These days, TAVR is the preferred nomenclature. The change appears to have occurred somewhere between 2010 and 2011, around the time the procedure was being considered for FDA approval in the US. Transcatheter aortic valve implantation is the term this procedure was referred to throughout the process of its development, and some argue that it should remain TAVI because the term replacement  implies that the original valvular structure is removed to make way for a new one, whereas this procedure merely moves it to the side. 

The real reason for the rebranding, according to the cynical vampires of Forbes, is that the "reimbursement for the procedure, ...would be much higher if it were based on a comparison with surgical replacement rather than surgical repair." So in this fashion, we in the rest of the world have sacrificed the precision of our communication to support the billing practices of American cardiologists.

Airway management in post-TAVR care

Airway issues are rarely a factor, as these patients are almost all by definition poor candidates for general anaesthesia, and so wherever possible the procedure will be carried out with a combination of sedation and local. Only occasionally (eg. where the valve is being inserted by a transcarotid, transapical or transaortic approach, or where a TOE is an essential part of the procedure) does the patient need intubation and a general anaesthetic. Interestingly, the SOLVE-TAVI trial demonstrated no real difference in outcomes between patients who underwent a GA, versus those who had local and sedation. If the patient returns from this procedure intubated, their extubation will usually follow a pathway similar to any post op cardiac surgical patient.

Ventilation management in post-TAVR care

Ventilation is usually spontaneous and generally should not pose a management problem. Whatever heart failure-related shortness of breath they had before should be much better now that they have a competent valve. On the other hand, there are a few matters to consider:

• Pulmonary oedema can definitely occur due to (for example) paravalvular leak or mitral apparatus damage.
• Atelectasis can also develop, whether with GA or conscious sedation
• They may have brought their own pleural effusion from home

The most important thing is not to let them fail this stress test:

• These are often elderly smokers with LV hypertrophy; in other words their respiratory effort should not be allowed to become excessive, as the increased myocardial workload will predispose them to myocardial ischaemia.
• They are not allowed to sit upright for some time following the procedure, and they are still under some residual anaesthetic and opioid effect, making NIV use somewhat dangerous (as it poses a risk of aerophagy and aspiration)
• High flow nasal oxygen seems like a reasonable compromise while they are in this vulnerable state, transitioning to NIV(CPAP) if they remain in respiratory failure after the period of supine bed rest has elapsed.

Haemodynamic management in post-TAVR care

The circulation should be better now that the aortic valve is competent. However, it is not the same as the valve you were born with, and the heart has remodelled after years of pressure overload, producing predictable haemodynamic challenges:

• The LV contracts forcefully, and there may be hypertension in the post-op period (in fact this is healthy and appears to be associated with a better prognosis). 
• The LV has poor compliance and will have poor diastolic function, which means these patients will generally require a reasonable amount of volume. With the AS resolved they should be more tolerant of fluid resuscitation.
• The LV chamber may be small, and contractility will be good (with afterload decreased and coronary flow now optimised), which means the LV outflow tract may become dynamically obstructed - again a reason to keep a high-normal intravascular volume.

In short: keep them well filled, and keep their blood pressure high. Specific blood pressure targets have not been established, but it appears that low-normal BP in the medium term is associated with increased mortality:  Lindman et al (2019) observed that SBP under 120 and DBP under 60 at 30-day follow-up was associated with an increased risk of death. What this means in the first hour after the valve is deployed, is up to speculation, but generally most people would agree that some high-normal value should be the target (eg. a MAP of 75-85 mmHg), and that fluid boluses should be the first option used to achieve this target. For the patient who achieves and over-achieves this goal, aggressive antihypertensive management is probably unnecessary in the acute period; the intensivist can turn a blind eye and the cardiologist can intensify their antihypertensives in the ward. The only real argument for any serious blood pressure control here would be in the context of bleeding from the arterial puncture site.

Causes of haemodynamic deterioration following TAVR:

This is definitely not a comprehensive list, as the total number of things that can go wrong following a TAVR defies the boundaries of human imagination. What follows is merely a short concentrated inventory of the most likely TAVR-specific problems that are organised into vague "cardiac" and "non-cardiac" categories to help answer Question 23 from the first paper of 2023.

• Cardiac causes (i.e. related to TAVI device or procedure)
  • Complete heart block 
  • Unstable arrhythmia
  • Aortic annular rupture or aortic root rupture 
  • Paravalvular aortic regurgitation
  • Displacement of the TAVI device
  • Iatrogenic ventricular septal defect
  • Fistula from the left ventricle to the atria (iatrogenic Gerbode defect)
  • Perforation of the anterior mitral valve curtain
  • Acute coronary artery occlusion
  • Cardiac tamponade
  • LV outflow tract obstruction ("left ventricular suicide")
• Non-cardiac causes (related to anaesthesia or patient comorbidities):
  • Retroperitoneal bleed from the femoral access
  • Cardiotoxic effects of anaesthetic agents
  • Sepsis due to undeclared perioperative infection
  • Bowel ischaemia due to atheromatous emboli
  • Anaphylaxis
  • Post-TAVR autodiuresis

Assessment of the haemodynamically unstable TAVR patient

From the answer to Question 23 from the first paper of 2023, an approach to the assessment of a desperately hypotensive TAVR patient could look like this:

• History,  to look for
  • Anaphylaxis (allergy list and medication chart)
  • Drug error ("did you seriously give 100mg IV metoprolol")
  • Procedural complications
• Physical examination looking for:
  • Conduction disturbance
    • Pulse rate
    • Arrhythmia 
  • Aortic pathology
    • Radioradial and radiofemoral delay
    • Murmur
  • Tamponade
    • Pulse pressure variation, pulsus paradoxus
    • CVP trend
    • Neck veins (distended?)
    • Heart sounds (muffled?)
  • Bleeding
    • Pallor
    • Abdominal distension
    • Brusing or swelling of the arterial puncture sites
    • Pulses in the distal limbs (occluded by haematoma?)
• Bedside tests 
  • ​​​​​​​ECG, looking for arrhythmias and conduction blocks
  • TTE/TOE, looking for paravalvular leak and other cardiac/valvular injury, and of course for effusion and tamponade
  • TOE can also assess for aortic dissection in a limited way
• ​​​​​​​Biochemistry
  • ​​​​​​​ABG, to immediately assess lactate and haemoglobin (as well as any electrolyte abnormalities that might be contributing to a conduction disturbance or arrrhythmia)
  • FBC, in case 
  • ACT or TEG to find any coagulopathy
• Imaging
  • CXR looking for mediastinal widening and TAVR device embolisation
  • CT aortogram 

This is based on a diagram from the excellent Tomey chapter (2023).

Neurological complications following TAVR

Delirium: these patients, though they are getting younger and less frail, remain a group of individuals with rather advanced age, serious vascular disease, and multiple comorbidities. They are extremely susceptible to delirium. It appears the conscious sedation and transfemoral access are somewhat protective: Tse et al (2015) found that only 12% of these tend to become delirious, vs. 53% of the fully anaesthetised transapical approach patients. 

Stroke: chances are, that's a well calcified aorta you're digging around in, and moreover the patient is likely to have a fairly diseased cerebral circulation from all their risk factors. And many are chronically in AF. And then you expose them to random and erratic blood pressure fluctuations during the procedure. The risk of stroke is therefore fairly high - Eggebrecht et al (2015) give a figure of 3.3%, which is apparently twice as high as the risk of stroke from a surgical AVR. From the ICU point of view, there is little to do other than vigilantly observe, defend the CPP, and have a low threshold for CT.  Considering the recency of largebore arterial puncture, nobody will have any appetite for giving systemic thrombolysis, but endovascular clot retrieval can still be an option, and that's exactly the kind of superspecialised interventional service that would be available in a centre that offers TAVR.

So, in summary:

• Avoid GA for these people
• Maintain a high-normal blood pressure
• Do not reverse the heparin at the end of the procedure
• Perform careful regular neuro observations, and wake/extubate the patient as soon as possible (for those that return intubated)
• Low threshold for imaging, considering the high risk of stroke and the confounding effects of residual sedaton
• Endovascular clot retrieval is the main management option

Fluid and electrolyte management following TAVR

Electrolyte derangement is likely to play a minor role in the post-TAVR arrrhythmogenesis, but most reasonable people would agree that these fragile patients are not the group in whom you could be cavalier about the biochemistry. Surely they need every possible protective factor against AF or VT. 

Acute kidney injury is much more important, and occurs in 20-30% of patients. From the ICU point of view, the only real modifiable risk factor we can attend to is going to be fluid resuscitation and blood pressure management, as the others (atheromatous emboli, contrast exposure, pre-existing renal impairment) we can do nothing about. Yet another reason to keep these people well-filled.

Anticoagulants and antiplatelet agents following TAVR 

Periprocedural heparinisation following TAVR is may be incompletely reversed. The ACT recommended by the ACCF/AATS/SCAI/STS expert consensus is at least 300, and they suggest you reverse it with protamine at the end of the procedure, but this is not always followed.

Post-TAVR antiplatelet agents are dual for 6 months, and then single for the rest of the patient's life, but this gets complicated if they also have AF or other reasons to be anticoagulated. Valvo et al (2019) performed a thorough review and created several excellent flowcharts to help guide antithrombotic management for these patients, based on their risks of bleeding and clotting. These are nice to be able to refer to but probably do not impact on the immediate postprocedure management of these patients from the ICU point of view, and so will not be included here.

Bleeding complications following TAVR

Understandably, the need to anticoagulate and protect from stroke conflicts with the concurrent need to puncture huge holes in large arterial structures, and the result is a risk of bleeding complications. According to Mangieri et al (2019), this risk is improving with time: they offered a change from a 24.2% are of bleeding complications in the first PARTNER trial in 2010, to a risk of merely 3.6% in the third PARTNER trial in 2019. To be fair, the latter was a study of TAVR in low risk patients, but still this is a rather impressive change, attributed mainly to mechanical factors such as a decrease in the diameter of the introducer sheath and a decreased reliance on femoral access. From the ICU point of view, the most important factor now becomes the vigilant observation of any arterial puncture site, keeping in mind the possibility of a high femoral stab producing a retroperitoneal haematoma.