Statistics from Altmetric.com
Over the last few years, the technique of percutaneous transluminal coronary angioplasty (PTCA) has undergone great change. The use of atherectomy has become marginal, and stent implantation almost systematic, justified by both a better outcome and a better cost:effectiveness ratio.1 2 Even the provisional use of stents seems to be less effective and more expensive than systematic stenting.3 Continuing advances in technology have more or less made the preliminary balloon predilatation unnecessary, leading to a technique now known as “direct stenting”. In this way, coronary angioplasty can increasingly be classified as simply stent implantation, a simple, safe, and rapid procedure. Furthermore, a consensus has progressively appeared regarding the follow up strategy and the indications for target vessel revascularisation (TVR), no longer based on the angiographic six month control, but rather only on symptoms and non-invasive detection of ischaemia, constituting an “angioplastically correct” follow up strategy. The difficulty of effectively treating in-stent restenotic lesions, the gap between angiographic restenosis and clinical outcome, the “oculo-stenotic” reflex, and the practical impossibility of systematic angiographic control are some elements which form the basis of the current follow up strategy. Nonetheless some exceptions to this “angioplastically correct” attitude should be debated.
Difficulty of treating in-stent restenotic lesions
Quasi-systematic stenting has at least one negative consequence: in-stent restenosis is going to become the main, if not the only, form of restenosis, and management of such lesions is reputed to be more difficult than post-PTCA restenosis. Percutaneous revascularisation is feasible, but after PTCA, the risk of a second restenosis is quite high, between 25–50%, mainly depending on the length of the lesion and the elapsed time before its occurrence.4 5 Despite this limited efficacy, balloon angioplasty remains the most frequently used strategy, treating focal (< 10 mm long) lesions satisfactorily, but proving disappointing for the treatment of diffuse or proliferative lesions.6 7
For these latter lesions, apart from intracoronary β or γ radiation treatment, the availability of which remains very limited, balloon angioplasty and alternative strategies such as second stent placement, or atherectomy, do not make it possible to lower the risk of a further restenosis below 50%.
The gap between angiographic restenosis and clinical outcome
The debate about the best strategy for revascularisation of these lesions is probably no more important than the question of the indications for revascularisation and thereby the modality of detection of the restenosis. Currently, in routine practice as well as in clinical trials, there is more interest in the clinical outcome and the event free survival of the patient, (that is, without death, Q wave or non-Q wave infarction, or the need for TVR), than in the angiographic evolution. Therefore event free survival has served as a surrogate to angiographic restenosis8 and the generally accepted strategy is to revascularise (and thus to perform angiographic control) only in case of symptoms or demonstrable ischaemia. This strategy, used after PTCA as well as after stenting, reflects the practical impossibility of systematic angiographic control for all patients, but other arguments also plead against systematic control for all patients.
Firstly, although angiographic restenosis is correlated with myocardial ischaemia, it is a poor predictor of mortality.9Intermediate lesions, assessed between 50–70%, and recognised by angiography as “restenosis” have a favourable prognosis under medical treatment and without revascularisation.10 11
Furthermore, angiographic restenosis covers a sizeable proportion of lesions which are not actually tight enough to cause coronary flow disturbance and ischaemia. Intravascular ultrasound measurements12 13 or direct evaluation of coronary flow reserve by Doppler14 or pressure wire, would allow more accurate evaluation of the functional repercussions of a lesion, thereby facilitating the decision on revascularisation.
Lastly, the temporal sequence of the phenomenon of restenosis must be considered. The phenomenon of restenosis culminates during the 3–4 months after angioplasty, but may regress in the long term; several studies have shown a 10–15% reduction in the severity of restenosis 3–5 years after stent implantation.15 16 This may justify an attitude of “watchful waiting” in the case of asymptomatic or intermediate restenosis.17
Thus, clinical follow up presents a threefold advantage: it reflects the natural evolution of patients; it avoids the risks and costs of systematic angiographic control; and it reduces the number of patients undergoing TVR, since the frequency of TVR corresponds to less than half the rate of actual angiographic restenosis.1 8 9
The irresistible oculo-stenotic reflex
In the early days of angioplasty, follow up was focused on the angiographic detection of restenosis. It appeared logical that a coronary stenosis which justified angioplasty should necessarily be revascularised in the case of restenosis; indeed, in the past, 70% of restenotic lesions diagnosed by systematic angioplasty were treated.8 9 The Benestent II study (second Belgium-Netherlands stent study),1 18 with its double randomisation, pointed out the gap between restenosis and TVR. In the stent group, 16% of the patients had angiographic restenosis. TVR was carried out in 13.5% of the patients with systematic angiographic control, but only in 5.4% of patients with clinical follow up. Basically, the TVR level varied from one- to threefold, depending on the type of follow up—clinical or angiographic. This difference is generally classified as an excess of revascularisation caused by the oculo-stenotic reflex, which implies that there was no objective reason for treating. Therefore, systematic angiographic control seems not only to be unnecessary, but responsible for misjudgement in case of restenosis.
Nevertheless, it is remarkable that, both in the context of a clinical trial1 and in routine practice,9interventional cardiologists continue to revascularise the majority of restenotic lesions observed by angiography. This attitude may be justified by a risk:benefit ratio in favour of systematic revascularisation of in-stent restenosis lesions once detected by angiography. Indeed, balloon angioplasty in such lesions is generally safe and systematic revascularisation of all angiographically observed lesions may not be without argument. In the past, the ACIP (asymptomatic cardiac ischemia pilot) study demonstrated that a strategy of systematic revascularisation was shown to be more effective than medical treatment or ischaemia guided revascularisation.19 Similarly, in the Benestent II study, the revascularisation of in-stent restenosis was not only free of complications, but would also appear to have improved the functional status of the patients.18
So, is there still an interest in a systematic six month angiographic control?
Although quite provocative, the debate about the interest of systematic angiographic control in selected patients deserves to be resumed. Let us remember that, in the past, two studies dealt with this issue. Weintraub and colleagues9 compared the evolution of 3363 patients who were re-evaluated systematically by angiography with 3858 patients followed clinically. As expected, the rate of revascularisation was higher in the group with angiographic follow up. However, survival at six years was higher in the angiographic follow up group than in the clinical follow up group (0.94 (0.01)v 0.92 (0.01), p < 0.01). This was confirmed in another study, based on a registry of 400 patients followed over 10 years: mortality was 2.7 times higher in those patients who refused systematic angiographic control than in those who were re-evaluated systematically.20 Even considering the methodological limits inherent to registry studies, the favourable impact of the systematic angiographic control and of the oculo-stenotic reflex on a criterion as important as survival should not be underestimated. All in all, we cannot ignore the fact that systematic angiographic control and revascularisation of these lesions may lead to better long term outcome; therefore, the logical approach to revascularisation of in-stent restenosis would be to customise the strategy according to the clinical situation.
Exceptions to the “angioplastically correct” attitude
The question of follow up after stent implantation and the strategy to follow in cases of asymptomatic restenosis is a dilemma which today's interventional cardiologist often has to face. The prevailing attitude of angiographic control and revascularisation guided solely by symptoms or residual ischemia should not overshadow either the fact that non-invasive methods for detecting myocardial ischaemia have only a mediocre positive predictive value,21 22 or the impact of such follow up strategies on mortality.9 20
The safety of the angioplasty procedure and the low TVR rate assured by the use of stents should not make us forget the risk of restenotic lesions going unrecognised. Thanks to stenting, coronary angioplasty is increasingly performed in high risk patients—for example, those with low left ventricular ejection fraction, multivessel disease, or with complex and unfavourable lesions23 located on the ostium or even on the left main coronary artery.24 This being so, it may be more sensible to advocate systematic angiographic control and consider revascularisation when there is angiographic restenosis, even in the absence of symptoms and even if this departs from an “angioplastically correct” attitude.