CORRESPONDENCE: LETTERS TO THE EDITOR
Definition of failed lysis may have influenced outcome in the MERLIN trial: Reply
Andrew G.C. Sutton, MA, MB, MRCP* and
Mark A. de Belder, MA, MD, FRCP
* Cardiothoracic Division, The James Cook University Hospital, Marton Rd., Middlesbrough, TS4 3BW, Teesside, United Kingdom (Email: Andrew.Sutton{at}stees.nhs.uk).
We thank Dr. Osman and colleagues for their interest in the MERLIN trial report (1). We would like to address their comments on potential confounding factors.
First, they appear to have misread the protocol definition for failed fibrinolysis, which was "a second ECG [electrocardiogram] 60 min after the onset of fibrinolytic therapy showing failure of the ST-segment elevation in the worst lead to have resolved by 50% by comparison to the pre-treatment ECG as well as the absence of an accelerated idioventricular rhythm at the time of the 60-min ECG."
Second, we debated the most appropriate method of diagnosing failed fibrinolysis, specifically the timing of the second ECG. In a previous study we assessed the ability of a 2-h ECG, performed immediately before coronary angiography, to predict flow in the infarct-related vessel (2). In the MERLIN trial we elected to diagnose failed fibrinolysis at 60 min for the following reasons: a) time is muscle, particularly when the initial reperfusion strategy is fibrinolytic therapy, so there is a fine balance between allowing sufficient time for fibrinolytic therapy to work and not delaying to the extent that there is little myocardium to salvage. b) Although it seems logical that a 60-min ECG protocol would result in inclusion of lower-risk patients than would a 90-min (or 120-min) ECG protocol, there is evidence that a 60-min ECG identifies high-risk patients just as well as an ECG performed at 180 min (3). The MERLIN trial results support this. The 60-min ECG identified a population at significantly higher risk of adverse events than did previous studies (4). In addition, over 40% of patents in the rescue arm had Thrombolysis In Myocardial Infarction (TIMI) flow grade 3 in the infarct-related vessel, and therefore may not have been taken to the catheter laboratory on the basis of a 90-min ECG. One might expect these patients to be a low-risk subgroup, but we observed a 36% incidence of the composite secondary end point at 30 days. This high event rate cannot be attributed to the procedure itself, because less than one-third of patients with TIMI flow grade 3 had an immediate angioplasty, and all of these were technically successful. c) Relying on a later ECG for the diagnosis of failed fibrinolysis would result in further delays in the rescue angioplasty process. In particular, we were concerned that transferred patients could be disadvantaged by a longer period between the onset of fibrinolytic therapy and the diagnosis of failed fibrinolysis, given the inevitable further delay associated with transfer. In the end, 40% of patients in the rescue arm were transferred, and the time delay was almost 2 h (105 ± 32 min). d) Neither a 60-min nor 90-min ECG addresses the problem of using paired static ECGs to make a diagnosis of failed fibrinolysis. This method provides no information on whether persistent ST-segment elevation on the second ECG is the result of failure to reperfuse or reperfusion injury, nor does it provide any information on subsequent stability of the ST segments. Nonetheless, it remains the most simple and practical noninvasive test currently available.
Third, the investigators have missed our remarks on the presence or absence of ongoing chest pain. We agree that chest pain should be evaluated following administration of fibrinolytic therapy, and it is possible that this could reduce the number of patients who are exposed unnecessarily to a rescue procedure. However, our large experience of infarct angioplasty, before and including the MERLIN trial, is that the presence of chest pain is not a reliable predictor of persistent arterial occlusion and that the absence of chest pain is a very poor indicator of arterial patency. This is consistent with the observations of other investigators (5). It is unclear whether the assessment of symptoms is more useful for predicting perfusion at the cellular level. We recorded the presence or absence of chest pain in all MERLIN trial patients at the time of randomization. In the article discussion, we report that, in the rescue arm, 53% of those who were pain free at randomization had inadequate antegrade flow in the infarct-related vessel and 38% of those in pain at randomization had normal antegrade flow in the infarct-related vessel.
Furthermore, 43% of rescue-arm patients who were pain free immediately before the coronary angiogram had inadequate antegrade flow in the infarct-related vessel, and 26% of those in pain immediately before the coronary angiogram had normal antegrade flow in the infarct-related vessel. Therefore, reliance on the presence or absence of ongoing symptoms would have resulted in a large number of incorrect diagnoses and, in particular, a large number of incorrect diagnoses of successful fibrinolysis.
 |
References
|
|---|
1. Sutton AGC, Campbell PG, Graham R, et al. A randomized trial of rescue angioplasty versus a conservative approach for failed fibrinolysis in ST-segment elevation myocardial infarctionThe Middlesbrough Early Revascularization to Limit INfarction (MERLIN) trial. J Am Coll Cardiol 2004;44:287-296.[Abstract/Free Full Text]
2. Sutton AGC, Campbell PG, Price DJA, et al. Failure of thrombolysis by streptokinase: detection with a simple electrocardiographic method Heart 2000;84:149-156.[Abstract/Free Full Text]
3. Purcell IF, Newall N, Farrer M. Change in ST segment elevation 60 minutes after thrombolytic initiation predicts clinical outcome as accurately as later electrocardiographic changes Heart 1997;78:465-471.[Abstract/Free Full Text]
4. Ellis SG, da Silva ER, Heyndrickx G, et al. Randomized comparison of rescue angioplasty with conservative management of patients with early failure of thrombolysis for acute anterior myocardial infarction Circulation 1994;90:2280-2284.[Abstract/Free Full Text]
5. Nicolau JC, Lorga AM, Garzon SA, et al. Clinical and laboratory signs of reperfusion: are they reliable? Int J Cardiol 1989;25:313-320.[CrossRef][Web of Science][Medline]
Related Article
-
Definition of failed lysis may have influenced outcome in the MERLIN trial
- Faizel Osman, Amer Chit, and Peter E. Glennon
J. Am. Coll. Cardiol. 2005 45: 808.
[Full Text]
[PDF]
|
Top
References