“Doctor knows best”… Perhaps; but which one?

Br J Cardiol 2010;17:64-6 Leave a comment
Click any image to enlarge

We continue our series in which Consultant Interventionist Dr Michael Norell takes a sideways look at life in the cath lab…and beyond. In this column, he considers the choice between surgery and percutaneous coronary intervention.

So there I was, sitting in one of our twice-weekly multi-disciplinary team (MDT) meetings. I was proffering my sixpence worth on the merits of surgery (coronary artery bypass graft [CABG]) or percutaneous coronary intervention (PCI) (occasionally neither, and – rarely – both), as a succession of clinical data, scans of various types and coronary angiograms were laid before us. And I got to thinking, “is this the way it should be?”

We have come a long way in tailoring treatment to patients. When PCI, or percutaneous transluminal coronary angioplasty (PTCA) as it was then, emerged as a young and promising technique in the late seventies, that decision process was more straightforward. Ballooning (alone) was a reasonable choice in accessible, discrete, concentric and non-calcified single vessel disease, in order to alleviate ischaemic symptoms.

You will note that I avoided the term ‘proximal’, because that hallowed segment of the left anterior descending (LAD) particularly, was then – and perhaps still is now – considered surgical territory in view of its prognostic impact. Nevertheless, in that era, the nature of disease uncovered by angiography was such that the bulk of revascularisation was still surgical and, as a result, discussion regarding other options was largely unnecessary.

New realms

But it was not long before Kansas City’s Geoff Hartzler, and many other pioneers, led us into the realms of multi-vessel PTCA. The randomised trials in the eighties comparing CABG with PTCA were consistently reassuring. They confirmed that the only price paid by embarking upon a less invasive strategy was a higher likelihood of a repeat procedure because of re-stenosis. Things are really no different 20 years later. Predictably, the magnitude of that ‘disadvantage’ became smaller with the introduction of stents, and was to reduce even further with the arrival of their drug-eluting cousins.

The turning point occurred around 10 or so years ago. We were getting better at intervention and were investigating – let alone, treating – many more patients earlier in the course of their disease, their anatomy being far more attractive to the interventionist. Along with more unstable presentations came advancing age and co-morbidity, all three of these factors causing surgeons, still reeling under the public scrutiny of the Bristol enquiry, to pause. And so at this juncture, PCI activity in the UK began to overhaul that of CABG, and – to be honest – we have not looked back since.

Harder choices

Recent ‘real-world’ studies have shown that similar outcomes can be achieved with either PCI or CABG in selected patient subsets, and so there is now an increasing need for debate about a specific individual’s treatment in even more cases. Traditionally, these ‘discussions’ have consisted of brief corridor chats or a few grabbed minutes with a CD in one hand and a tuna sandwich in the other. But perhaps the structure of that process now deserves to be better defined.

How should such a meeting be constructed and which of the multi disciplines should attend? Presumably, at least a surgeon and an interventionist would be a minimum, but what about our anaesthetic colleagues, let alone an individual with none of these skills (which might at least be refreshing)? Should all cases be presented, or just those that demand revascularisation? Or should we limit the forum even further by focusing upon those who need treatment and in whom either surgery or PCI are both deemed to be feasible? But then who, exactly, does the deeming?

Acknowledging that an increasing proportion of most units’ PCI activity is unplanned, there needs to be an acceptance that many cases with multi-vessel disease will still be managed percutaneously (and without discussion) simply because expedience is required on clinical – or sometimes, administrative – grounds.

Influencing outcomes

Once you start down such a formal process, you would be surprised where it leads. There is no doubt that the participants at a MDT meeting can influence its outcome, whether by one speciality outnumbering – and, conceivably, intimidating another (is that actually possible?) – or by the presence or absence of an

individual with specific expertise that might have relevance in certain cases. A ‘pooled’ system may also demand some fine tuning; further intra-speciality discussion might be required to avoid a particularly high-risk case being served up to an operator who might be less than keen to take it on.

Naturally, all relevant clinical data need to be available in order for a decision to be made. The proceedings should be documented (so as to form a record of the discussion), with such a ‘memorandum of understanding’ emphasising particular issues that impacted upon the result. The outcome should be signed off by a Chair, which could rotate between specialities in the hope of avoiding at least the perception of bias (as if…).


I have considered two modifications to our own meetings, which I am happy to share with the wider public. First, in order to reassure ourselves that the quality of decision making is robust, I have speculated as to the merits of re-presenting a case to my colleagues (after a suitable delay, of course) to see if we reach the same result; now that would be interesting.

Second: feedback. If a decision rested upon the anticipation of grafting a particular vessel, attending to a valve, re-canalising a chronic occlusion or evaluating a lesion with a pressure-wire, then there should be a mechanism whereby it can be determined that this actually occurred. MDT meetings commonly provoke individuals to ‘talk the talk’, but all parties must be reassured that, when push came to shove, they also ‘walked the walk’.

More often than not, in cases that could be revascularised to an equivalent extent with either modality, the consensus often rests upon just how much ‘work’ PCI would involve. Re-opening a lengthy chronic occlusion of the right, bifurcation stenting in the circumflex and dealing with long and diffuse LAD disease, might all require a total of five or six stents. Depending on the surgical risk, that numeric level might be considered a threshold at which CABG starts to become ‘acceptable’, so to speak. Such an approach (the Anticipated Stent Summation score) might be seen as a surrogate for procedural difficulty, analogous to the system utilised in the recent SYNTAX trial (I don’t recommend using the acronym, however).

Patient opinion

There is of course another contributor to the MDT process, as yet not mentioned, namely the patient. When all else fails, putting the information in front of the person to be treated and inviting him or her to choose their management, does seem somewhat intuitive. However, human nature being what it is, the patient might well accept a less invasive and low-risk procedure now, even if it means the possibility of further ‘attention’ at a later date. That seems entirely reasonable.

When I was training, many of the patients with acute coronary syndromes whom we now stent routinely, would never have got to the cath lab in the first place. Thus, I for one am not going to lose too much sleep if a patient with severe multi-vessel disease is managed surgically, even if I know I could do the same job at lower risk – all be it with a high ASS score (sorry about that).

At the end of the day it’s all about openness and transparency. If a choice of treatment is made on a patient’s behalf then the route to that decision should at least be made clear.