This chapter is not relevant to any specific Section of the 2017 CICM Primary Syllabus, and certainly not to Section G7(iv) which expects the exam candidate to "describe the methods of measurement of cardiac output including calibration, sources of errors and limitations". This chapter is all about the applications of the PA catheter to the haemodynamic management of critically ill patients. If that sounds like something that should be in ICU exams to you, that would be an appropriate reaction - but unfortunately, the CICM examiners did not think so.
The problem is that the CICM First Part exam is very basic science-oriented, which makes it a poor destination for this sort of highly practical clinical material. On the other hand, the CICM Second Part exam is too practical and clinical, i.e. the trainees at that level of seniority are expected to recognise the treacherous nature of the PA catheter, and to reject its empty promises. "The evidence does not support the routine use of PAC in the haemodynamic management of ICU patients", they are expected to say. This is not completely unexpected, as the abandonment this "cumbersome and sometimes dangerous procedure resulting in little material gain" in the early 2000s was so rapid and widespread that it is used today as an example of the successful rational deadoption of a low-value practice (Fung & Hyzy, 2019). This, of course, does not mean that this device is without its uses. Nor does it mean that one can safely go through their training without understanding something about it. It is not inconceivable that one might end up working in an ICU which is staffed by a derelict civilisation of prehistoric intensivists, a time capsule where PA catheters are still commonplace. One would not want to look foolish at their meetings.
PA catheter in the diagnosis of shock
The broader indications for the use of PA catheters are discussed elsewhere. This summary will focus on the use of the PAC for the diagnosis of haemodynamic instability in the confusing mixed-shock-state ICU patient, and how one might rely on it to guide therapy. In short, forget the biopsychosocial approach to medicine. Your patient is a system of tubes and pumps, and the behaviour of this siphonophore can be described by three main variables, of which each can only have three possible states. Therefore, there can only be 27 (33) different things wrong with the patient.
Those parameters are said to be:
- Central venous pressure (CVP)
- Cardiac Index (CI)
- Systemic Vascular Resistance (SVRI)
And so, applying the matrix,
This is not a fanciful oversimplification for the sake of comedy, this is an honest piece of advice from Paul Marino's The ICU Book (3rd ed., 2007), which unselfconsciously recommends, "These three hemodynamic parameters can be used to identify the hemodynamic problem in any patient with hypotension". There are actually quite a few such matrices, using different PAC-generated variables for different purposes. For an example of a more complicated one, here's a set of classifications for acute MI patients from Forrester et al (1976), which use cardiac index PAWP and a couple of clinical features, namely peripheral hypoperfusion (HYP) and pulmonary congestion (PC):
|Cardiac Index (L/min/m 2)||PAWP (mm Hg)|
Applications of the PA catheter in the modern era
The use of the PA catheter for the original abovementioned purpose, in the diagnosis of the haemodynamically unstable patient, still has a role to play, albeit in a restricted range of circumstances. This list of circumstances is not derived from any specific literature source or society guideline, but rather confabulated from the authors' own exiguous opinions and experiences, which makes this Grade F evidence (a level below Twitter), not to be repeated in any exam answers.
In short, you'd resort to a PA catheter...
- Wherever you can't use TTE
(and TOE is for whatever reason not appropriate)
- Open chest
- Prone patient ventilation
- Views obscured by gas:
- Immediately following open cardiac surgery
- Pneumothorax or pneumomediastinum
- Extremely high PEEP, obscuring all views
- Poorly tolerated (eg. rib fractures of the left anterior chest)
- Wherever you can't or won't use TOE
(and TTE is for some reason uninformative)
- Oesophageal perforation
- Oesophageal stricture
- Upper airway injuries prohibiting probe placement
- Where clinical features normally used to determine response to fluids are not available
(eg. right heart failure)
- Where rhythm or pulse abnormalities render pulse contour analysis useless
- Atrial fibrillation
- IABP or VAD
Additionally, there are distinct indications for right heart catheterisation or pulmonary arterial pressure monitoring which can occasionally be met in the ICU. This is nothing to do with the diagnosis of shock per se (though the patient would typically be shocked anyway); rather these are diagnostic or monitoring applications which somehow enhance or direct management.
- Diagnosis of pulmonary hypertension: PA catheterisation is still the gold standard for the differential diagnosis of pulmonary hypertension. Comparison of TTE and PAC have revealed a discordance of the two measurements of PA pressures, though the authors hasten to point out that this is not due to the inaccuracy of one modality, but rather due to the constant variation of PASP.
- Pulmonary vasodilator therapy: most people would agree that routinely used pulmonary vasodilators (especially of the inhaled or IV variety) are sufficiently finnicky and toxic that the use of a PA catheter to titrate their dose is appropriate.
- Right heart failure: the diagnosis of the causes of right heart failure, as well as the monitoring of response to therapy, is a solid indication for PA catheter use (even though you might never be able to float it).
- The PA catheter may reveal raised LA pressures and/or a waveform characteristic of mitral regurgitation, which would lay the blame for the RV decompensation upon the failure of the left heart.
- Or, the pulmonary vascular resistance (PVR) might turn out to be absurdly high, which would also explain RV failure (this time due to overwhelming afterload).
- Or one might discover an RV-PA gradient in excess of 25mmHg, which would suggest RVOT obstruction.
- Or, the PAWP and PVR may turn out to be completely normal, and one might find oneself disappointed by the right ventricular contractility.