Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
MAKE A BETTER USE OF NURSES
- IÑIGO ROMON-ALONSO (31 May 2002)
-
Hawthorne Effect Not Accounted For
- Thomas F Heston (1 June 2002)
-
Diabetes and Myocardial Infarction
- Mark W Savage (4 June 2002)
-
The main problem for thrombolysis efficacy is still prehospital patient delay.
- Jane C. Walsh (6 June 2002)
-
Time to Care, not Time to Door
- Peter A West (6 June 2002)
-
Re: Hawthorne Effect Not Accounted For
- Kathleen M. Langan (7 June 2002)
-
Use of door to needle times
- Steve Goodacre (11 June 2002)
-
Reducing delays to thrombolysis in myocardial infarction
- Russell C Davis, Ann M. Eason, and Rajai AS Ahmad (2 July 2002)
-
Focus on accident and emergency departments to reduce delays
- Chris J L Hetherington, Peter Doyle, Javid A Kayani, David F Gorman (30 July 2002)
-
Research: Findings to Implementation
- Maree F O'Sullivan (9 August 2002)
-
Caveats of fast- track thrombolysis
- Aristotle D Protopapas (17 August 2002)
-
Effective Audit to Reduce Thrombolysis Times
- Asif Qasim, Mary E. Heber (27 January 2003)
|
|
|||
|
IÑIGO ROMON-ALONSO, QUALITY MGR. BANCO DE SANGRE DE CANTABRIA. AVENIDA VALDECILLA, SANTANDER 39008 SPAIN
Send response to journal:
|
Nurses are capable of many and more complicated tasks than those assigned to them. Nurses in Spain perform a great part of the work which in Britain is still made by doctors (taking blood samples, transplant coordination, vaccination, management of chronic diseases such as diabetes and hypertension...). As a matter of fact, Spanish nurses feel underused when they work for the NHS. Training and better salaries will be a cost effective measure. I think that making a better use of our nurses is the first tool to use in order to make a better use of doctors. |
|||
|
|
|||
|
Thomas F Heston, President, Internet Medical Association Kellogg, Idaho USA 83837
Send response to journal:
|
This study found that during the time period from 1989 to 2001 thrombolytics were given more promptly to patients presenting to the hospital with a heart attack. This effect is most likely due to a greater awareness by the medical community at large that prompt initiation of thrombolytics saves lives. This is similar to the Hawthorne Effect, since with each passing year the importance of the door to needle time became more widely known. There undoubtedly was a greater emphasis upon prompt initiation of thrombolytics during phase 3 of this study. Thus, the conclusion cannot simply be that their nurse initiated protocol (phase 3) worked any better than their nurse evaluation / physician treatment protocol (phase 2). To fairly compare these two different protocols, they should be conducted at the same time. A more scientific method would be to randomly assign patients to either the phase 2 or phase 3 protocol. |
|||
|
|
|||
|
Mark W Savage, Consultant Physician Royal Hallamshire Hospital, Sheffield S10 2JF
Send response to journal:
|
Editor- Quasim et al. are to be congratulated for their achievement in reducing the door to needle time in Acute Myocardial Infarction. However, I was most disheartened to see that they have retained Diabetic Retinopathy as a relative contraindication to thrombolysis. It has been known for some time that the risk-benefit analysis is in favour of thrombolysis (1). We really must bury this erroneous belief once and for all. Whilst appreciating that the purpose of their audit was to improve door to needle time, I was nevertheless also disappointed to note that despite there being clear benefit for treating diabetic patients, and particularly newly diagnosed diabetes, with insulin and glucose (2), there seems to be no place in their protocol for the institution of this simple and highly effective therapy. I am sure that their protocol is not the only one to omit this treatment. This is despite it probably being more beneficial than thrombolysis in this group. It is most important that, with the numbers of diabetic patients increasing at an alarming rate world-wide, every opportunity be taken to realise that they represent a significant minority of patients with AMI, that they do worse than their non-diabetic fellow patients and that they benefit from aggressive therapy, which is not confined to thrombolysis. The NSF that prompted this publication, The NSF in Coronary Heart Disease, also includes advice to treat diabetic patients with insulin and glucose. 1 Mahaffey KW, Granger CB, Toth CA, White HD, Stebbins AL, Barbash GI, Vahanian A, Topol EJ, Calif RM. Diabetic Retinopathy should not be a contraindication to thrombolytic therapy for acute myocardial infarction: review of ocular hemorrhage incidence and location in the GUSTO-I trial. Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries. J Am Coll Cardiol 1997;30:1606-10. 2 Malmberg K, Norhammar A, Wedel H, Ryden L. Glycometabolic state at admission: important risk marker of mortality in conventionally treated patients with diabetes mellitus and acute myocardial infarction: long-term results from the Diabetes and Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) study. Circulation 1999;99:2626-32. |
|||
|
|
|||
|
Jane C. Walsh, Lecturer in Psychology National University of Ireland, Galway.
Send response to journal:
|
Despite the marvellous improvements effected by Qasim et al(1) in reducing door-to-needle times, it is important to acknowledge that the main factor thwarting the effectiveness of thrombolysis is prehospital patient delay (2-4). Studies of the time from symptom onset to hospital arrival (e.g.ISIS-2 and ISIS-3)(5-6) have indicated median presentation times of 5 hours and 4 hours, respectively. Traditionally, research has looked at the importance of demographic factors as predictors of patient delay, but more recently, health psychology has had alot to offer in terms of understanding the complex processes involved in making a decision to call for help (7-8). This is still the area that has the most potential for improvement in terms of reducing the time from onset of symptoms to thrombolysis. [1]Qasim A, Malpass K, O'Gorman DJ, Heber ME. Quality improvement report: Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction. BMJ 2002; 324: 1328-1331. [2] Walsh J, Lynch M, Murphy AW, Daly K. Effects of coping styles on patient delay in presenting to hospital with an acute myocardial infarction. Intl J of Psych 2000: 35, (3/4), 15-16. [3]Dracup K, Moser DK, Eisenberg M, Meischke H, Alonzo AA, Braslow A. Causes of delay in seeking treatment for heart attack symptoms. Soc. Sci. Med. 1995; 40(3):379-392. [4] Dracup K, Moser DK. Beyond sociodemographics: Factors influencing the decision to seek treatment for symptoms of acute myocardial infarction. Heart Lung 1997; 26(4):253-62. [5] ISIS-2 (Second International Study of Infarct Survival) Collaboration Group. Randomised trial of intravenous streptokinase, oral aspirin, both or neither among 17,187 cases suspected acute myocardial infarction". Lancet 1988; 8607(2):349-60. [6] International Study of Infarct Survival ISIS-3: A randomised comparison of streptokinase and of tissue plasminogen activator vs anistreplase and of aspirin plus heparin vs aspirin along among 41,299 cases of suspected acute myocardial infarction. Lancet 1993; 339: 753-770. [7] Dempsey, S.J., Dracup, K., Moser, D.K. Women's decision to seek care for symptoms of acute myocardial infarction. Heart & Lung Journal of Critical Care 1995; 24(6): 444-456. [8] O'Carroll, R.E., Smith, K.B., Grubb, N.R., Fox, K.A. & Masterston, G. Psychological factors associated with patient delay in attending hospital following a myocardial infarction. Journal of Psychosomatic Research 2001 51: 611- 614. |
|||
|
|
|||
|
Peter A West, Director, York Health Economics Consortium, University of York University of York
Send response to journal:
|
The paper on nurse delivery of thrombolysis highlights a bigger issue in health care. The public typically see the journey to hospital as the crucial one and assume that any hospital rationalisation will be bad for their health. This study highlights the potential for delay, even when the hospital is reached. By changing who does what, the delay before the right help was given was reduced. There may be many other areas of health care where changing who does what could improve outcomes. These might include more changes about where it is done, so that reaching the hospital (and potentially facing a significant delay nonetheless) is no longer seen as the most crucial factor in health care for every acute case. |
|||
|
|
|||
|
Kathleen M. Langan, Staff Nurse Charity Hospital Emergency Department, New Orleans, La 70112
Send response to journal:
|
Tsk, tsk! Yet another example of discounting and dismissing the contribution nurses make towards positive patient outcomes. I agree, there was probably an increased awareness of the efficacy of thrombolytics by the third phase. I am also in favor of evidence-based medicine. However, could it be that the nurses were actually doing an effective job? Just recently the New England Journal of Medicine published a study revealing a link between increased nursing care and better patient outcomes in hospitals. This comes as no surprise to most nurses. I am a British nurse with over 20 years experience in emergency and critical care nursing. I see on a daily basis the contribution nurses make towards improving patient care. "With each passing year the importance of" nurses and what they do "became more widely known" would be a favorable outcome of the study in the New England Journal of Medicine. So, come on! For once, lets hear it for the nurses! |
|||
|
|
|||
|
Steve Goodacre, Health Services Research Fellow Medical Care Research Unit, Regent Court, 30 Regent Street, Sheffield, S1 4DA
Send response to journal:
|
Qasim et al raise an interesting issue regarding the use of door to needle times to measure performance in managing acute myocardial infarction.[1] The proportion of patients receiving thrombolysis within the target time may be maximised by either increasing the numerator, i.e. the number of patients receiving thrombolysis within the target time, or reducing the denominator, i.e. the total number of patients receiving thrombolysis. The former may be achieved by using strategies to accelerate door to needle times. The latter may be achieved by either withholding thrombolysis for, or excluding from analysis, those patients with a complicated presentation who are likely to undergo delays prior to thrombolysis. The table presented by Qasim et al shows that the denominator appeared to be relatively constant over the first two phases of the study (160 and 157 patients respectively), but then appeared to fall by approximately 40% (to 93 patients) during phase three. The numerator, on the other hand, remained relatively constant between phases two and three. This suggests that the increase in the proportion achieving thrombolysis within the target time was due to fewer patients receiving thrombolysis rather than more patients achieving the target time. I cannot find any obvious explanation for this finding in the description of the study methods. Each phase seems to last for the same duration. Can the authors offer any explanation as to why the rate of thrombolysis has fallen in phase three? Was there a change in service delivery that may have resulted in a smaller number of patients receiving thrombolysis in stage three? Did patients differ in age, gender, co- morbidity or ECG findings between the phases of the study? This observation has important implications for routine audit of thrombolysis times.2 If improvements in the proportion thrombolysed within the target time are achieved by increasing selection for thrombolysis or inclusion in audit, rather than by accelerating time to thrombolysis, then changes recorded by audit may not relate to improvements in patient care. 1 Qasim A, Malpass K, O'Gorman DJ, Heber ME. Quality improvement report: Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction. BMJ 2002; 324: 1328-1331. 2 The National Service for Coronary Heart Disease. Department of Health, 1999. London. |
|||
|
|
|||
|
Russell C Davis, Consultant Cardiologist Sandwell General Hospital, Lyndon, West Bromwich, UK B71 4HJ, Ann M. Eason, and Rajai AS Ahmad
Send response to journal:
|
EDITOR-Qasim et al.1 are to be congratulated on having reduced delays to thrombolysis for patients with acute myocardial infarctions, and having done so of their own initiative some years before the publication of the National Service framework for Coronary Heart Disease,2 which lays down standards which they were already exceeding. Their system of assessment of patients on the coronary care unit, rather than in the Accident & Emergency department, with all thrombolytic administration being on coronary care, is contrary to the recommendation of the NSF, but the unit’s success shows that different local policies can lead to similar success. In Sandwell, we have achieved greatly improved door-to-needle times via a refinement of the existing arrangements, rather than radical changes. Despite administration of thrombolysis in the accident and emergency department, our previous delays were unacceptably high. From January to March 2001, 16 of 42 patients (38%) with ‘barn door’ infarcts received thrombolysis within 30 minutes, median time being 44 minutes. A systematic process-mapping exercise revealed that delays occurred at a number of points. Ambulance staff now alert the hospital that they are bringing in a patient with suspected cardiac pain; all chest pain patients are triaged as highest priority; electrocardiograms are recorded by nursing staff within 5 minutes of arrival, and examined by a member of the Medical team; ‘barn door’ infarcts are seen immediately, by a casualty officer or member of the medical team, and thrombolysis prescribed after a brief check to ensure it is warranted and that there are no contra- indications; and the rapid administration of thrombolytics, once prescribed, has been encouraged. From January to March 2002, 33 of 39 patients (85%) with ‘barn door’ infarcts received thrombolysis within 30 minutes, median time being 19 minutes. We feel more radical changes are likely to be needed to achieve further reductions in delays. Electrocardiogram recording in the ambulance, ideally with transmission to the hospital, could save approximately 5 minutes , and the use of Tenecteplase rather than Alteplase or Streptokinase would likely reduce the median door-to-needle times by a further 5 minutes. We feel that this would be better use of resources than the employment of dedicated ‘thrombolysis’ nurses, who are unlikely to be able to work 24 hours a day. Russell C Davis
1Qasim A, Malpass K, O’Gorman DJ, Heber ME. Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction. BMJ 2002;324:1328-31. 2Department of Health. National service framework for coronary heart disease. London: HMSO, 2000. |
|||
|
|
|||
|
Chris J L Hetherington, SpR Emergency Medicine Selly Oak Hospital, Birmingham, B29 6JD, Peter Doyle, Javid A Kayani, David F Gorman
Send response to journal:
|
“SAFETY AND EFFICACY OF NURSE INITIATED THROMBOLYSIS IN PATIENTS WITH ACUTE MYOCARDIAL INFARCTION” Focus on accident and emergency departments to reduce delays EDITOR- We read with interest the paper entitled above under the Education and Debate section of the BMJ dated 1st June 20021. The stepwise improvement in thrombolysis times over the three phases is very impressive and a credit to the medical and nursing staff of the coronary care unit. One hopes that things will continue to improve into phase 4 enabling the 2003 national service framework targets to be met. However we feel that the point has been somewhat missed. The accident and emergency department gets merely a cursory mention, in the context of reasons/causes of delays in treatment. It would appear from the paper that this was not deemed an important area to focus attention with regard to streamlining the management of patients with acute myocardial infarction. Yet during Phase 1 it was one of the “two main reasons” for delays identified. Phase 2 does not address this problem whatsoever. Patients with suspected acute myocardial infarction in the accident and emergency department were still transferred to CCU for further assessment by a nurse. Presumably the patient had already seen a doctor in the A&E department? Why not initiate treatment there? The coronary care nurse still then had to call the on call medical team to see the patient; this, rather unsurprisingly, was identified as a source of delay. Phase 3 improved upon this with the nurse-initiated thrombolysis, yet the patients coming from the A&E department still had already been assessed prior to this. The diagnosis of a barn door MI is based on the first ECG; if this is done in A&E, then why not treat it there? The CCU nurse apparently repeats the ECG on these patients, which again represents an unnecessary delay. In our unit we have set up a system of providing prompt assessment and treatment of patients who attend the A&E department with acute myocardial infarction. All chest pains are triaged as red, undergo an ECG and are considered immediately for thrombolysis (or other intervention) by the A&E medical staff in the A&E department. Since all of the nursing and medical staff are fully aware of this system, and the urgency with which acute MI needs to be treated, this runs very smoothly. Our results are carefully and frequently audited to identify and minimise delays and inappropriate thrombolysis. In the six month period prior to April 2002 we thrombolysed 92 barn-door MIs, 92% of which were within 30 minutes, and 85% of which were within 20 minutes. We feel strongly that the accident and emergency department is the most appropriate place for these patients to be assessed and treated, and our methods have proven that with training of staff and audit of data the NSF targets are readily achievable, with the patients being the ones who benefit most. C J L Hetherington SpR in Emergency Medicine, hethers@doctors.org.uk P Doyle Consultant in Emergency Medicine J A Kayani Consultant in Emergency Medicine D F Gorman Consultant in Emergency Medicine Accident and Emergency Department Selly Oak Hospital University Hospital Birmingham NHS Trust Raddlebarn Road Birmingham B29 6JD References 1. A Qasim, K Malpass, D J O’Gorman, M E Heber. Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction. BMJ 2002;324:1328-31. “SAFETY AND EFFICACY OF NURSE INITIATED THROMBOLYSIS IN PATIENTS WITH ACUTE MYOCARDIAL INFARCTION” < |
|||
|
|
|||
|
Maree F O'Sullivan, Senior Medical Officer Gold Coast Sexual Health Clinic, 2019 Gold Coast Highway, Miami, QLD, 4220, Australia
Send response to journal:
|
EDITOR – Qasim et al [1] report commendable efforts by the Princess Royal Hospital to improve their compliance with and implementation of the National Service Framework for Coronary Heart Disease for the In-Hospital Management of Acute Myocardial Infarction. The use of fast-tracking areas within the coronary care unit, the innovative use of the Coronary Care Nurse in patient assessment and management, and improved liaison between medical and nursing staff have demonstrated the feasibility of a rapid response to the management of patients with myocardial infarction. However, it is of concern that the original randomised trial results of the use of intravenous streptokinase were published as long ago as August 1988. [2] This trial clearly demonstrated survival benefits with the use of this therapeutic intervention. The importance of these findings was underlined by the 1994 Guidelines set out by the British Heart Foundation Working Group for the Early Management of Patients with Myocardial Infarction. [3] & nbsp; These principles of management outlined responsibilities for general practitioners, ambulance services, hospitals and health authorities, setting out implementation processes and timetables. Although the hospital involved in this report has made progress with respect to these management guidelines, they appear to be working in isolation from the other potential professionals who may be involved in patient management. While this implementation of the management guidelines is creditable, it has been undertaken in a less than timely manner. From a population perspective, national guidelines aim to produce best practise, providing a reliable and transferable process between hospitals and health districts both regionally and nationally. The speed of implementation of the recommendations from the 1988 research is of concern given the potential and likely adverse health effects. Whole-of-episode view of patient management, encompassing not just hospital personnel but also ambulance services, general practitioners and health authorities needs urgent attention at the national level. It is of concern that potentially life-saving research seems to have stalled at the policy level and has not translated into procedural implementation. Is this a best-practise outcome in terms of both research funding and medical research implementation? Maree F O’Sullivan senior medical officer< o:p> Gold Coast Sexual Health Centre, 2019 Gold Coast Highway, Miami, QLD, 4220, Australia [1]< /span> Qasim A, Malpass K, O’Gorman DJ, Heber ME. Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction. BMJ 2002;324:1328-31. [2]< /span> Second International Study of Infarct Survival Collaborative Group. Randomised trial of intravenous streptokinase, oral aspirin, both or neither among 18, 187 cases of suspected acute myocardial infarction. Lancet 1988;ii:349-60. [3]< /span> Weston CF, Penny WJ, Julian DG. Guidelines for the early management of patients with myocardial infarction. British Heart Foundation Working Group. BMJ 1994;308:767-71. |
|||
|
|
|||
|
Aristotle D Protopapas, Specialist Registrar in Cardiothoracic Surgery (Locum) Barts Hospital London EC1A
Send response to journal:
|
To the Editor of the BRITISH MEDICAL JOURNAL Dear Sir or Madam: I read with interest the paper of Qasim et al.(1)on expediting thrombolysis for acute myocardial infarction (AMI) by specialist nurse‘fast-tracking’. A word of surgical caution though, on a perhaps well-known caveat: AMI can be confused with aortic dissection (AD), a life-threatening condition. In AD a ‘flap’ is created on the aortic wall and blood escapes to a false lumen within the vessel; optimal treatment is by urgent major surgery(2).When inadvertently administered in patients with AD, thrombolysis propagates the dissecting process and often leads to iatrogenic acute cardiac tamponade, multiplying morbidity and mortality , as has been reported by Kahn et al.(3)and several other workers over the last decade. Clinical differentiation between AMI and AD can be challenging for the most astute diagnostician (4 ), even more perhaps with ‘just a brief history, baseline observations, and a 12 lead electrocardiogram’ as proposed by Qasim(1). The electrocardiogram in particular is of no value in AD complicated with a myocardial event (4).The most thorough implementation of guidelines(1) can be no substitute for a complete clinical assessment and relevant investigations in such a confusing situation. The only element of the pre-thrombolysis checklist (1) that may alert to the differential diagnosis of AD (and thus spare the patient an unnecessary, costly and most perilous therapy) is the differential arterial blood pressure. Unfortunately, it has been documented that this sign is found in less than 40% of the cases of AD(5). It is therefore unavoidable, in ‘this fast-track’ set-up, to ‘thrombolyse’ some unfortunate patient with chest pain, ambiguous ischaemic electrocardiogram and no differential sphygmomanometry measurements, who still bleeds into an aortic false lumen (and this, even worse, after the administration of soluble aspirin(1). Would it be fair for the specialist nursing colleagues(and if the proposed phase 5(1) is implemented, for the paramedics) to assume the responsibility of a misdiagnosis and mistreatment in this oft-repeated clinical scenario? One might argue that the probability of such a blunder is negligible if compared to the potential benefit of expediting life- saving thrombolysis in bona fide AMI. Is this a satisfactory argument in our litigious times? Would the situation warrant an’ ad hoc’ drafted consent form with mention of the specific risk(amongst others)? Once again, a case for the Hippocratic dictum: The gods have put diagnosis well before treatment….. REFERENCES 1.Qasim A, Malpass K, O'Gorman DJ , Heber ME . Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction BMJ 2002; 324:1328-1331 2. Edmunds LH. Cardiac Surgery in the adult, chapter 38, McGraw-Hill 1997 ISBN: 0070189633 (on-line edition:http://www.ctsnet.org/book/edmunds/) 3. Kamp TJ, Goldschmidt-Clermont PJ, Brinker JA, Resar JR. Myocardial infarction, aortic dissection, and thrombolytic therapy. Am Heart J 1994 Dec;128(6 Pt 1):1234-7 4. Slater EE, DeSanctis RW: The clinical recognition of dissecting aortic aneurysm. Am J Med 1976; 60: 625. 5.Ergin MA, Griepp RB: Dissections of the Aorta, p. 2277 In: Glenn’s Thoracic and Cardiovascular Surgery, Sixth Edition, Appleton and Lange, Stamford 1996. |
|||
|
|
|||
|
Asif Qasim, Specialist Registrar Princess Royal Hospital, Telford, TF1 6TF, Mary E. Heber
Send response to journal:
|
The publication of a Quality Improvement Report in the BMJ detailing the experience at Princess Royal Hospital in Telford, in developing a system for Nurse Initiated Thrombolysis [1], has led to considerable comment and correspondence on the subject of improving the speed of diagnosis and treatment in acute myocardial infarction The developments in our unit were as a result of detailed, comprehensive and systematic audit of the process of care that patients received following arrival at the hospital with a suspected acute myocardial infarction. There was a notable and regular delay in the prescribing and administration of thrombolytic therapy after an ECG diagnosis of acute myocardial infarction had been made by the cardiac triage nurse. This led to the consideration of Nurse Initiated Thrombolysis as one potential solution. As in any hospital trust the changes in service were restrained and directed by a number of local factors such as the skills and experience of personnel, hospital layout, financial considerations, and the interaction between different departments within the trust. These factors are clearly in addition to the issues of safety, effectiveness and clinical governance which guide such developments. It was after assessing all such considerations that the decision to proceed to Nurse Initiated Thrombolysis was made, and subsequent audits have demonstrated a marked improvement in performance. A number of equally valid alternative approaches, such as the administration of thrombolytic therapy in the Accident and Emergency Department, have been suggested, and have demonstrated improvements within a number of trusts. The key to effective change, however, is the use of comprehensive and detailed audit. It is crucial to identify the stages at which delays are occurring to allow changes to be appropriately targeted. The use of Nurse Initiated Thrombolysis is only likely to be beneficial in units where there is a delay in prescribing therapy after an ECG diagnosis has been made. In many hospitals the main delay is in performing an ECG, and then having it reviewed by an experienced practitioner. In such situations the main changes relate to the initial response to the arrival of a patient with suspected myocardial infarction, and involve establishing an effective process of cardiac triage. Extending the role of experienced and well trained nurses to include the initiation of thrombolysis in acute myocardial infarction has proven to be safe and effective in reducing door to needle times in our unit. It may well be a useful approach in many other units but such changes should be directed by well conducted audit. 1 Quality improvement report: Safety and efficacy of nurse initiated thrombolysis in patients with acute myocardial infarction Asif Qasim, Kerry Malpass, Daniel J O'Gorman, and Mary E Heber BMJ 2002; 324: 1328-1331. Competing interests: None declared |
|||