Q J Med 2000; 93: 805-811
© 2000 Association of Physicians
Can clinical assessment of chest pain be made more therapeutic?
From the University Dept of Psychiatry, Warneford Hospital, Oxford 1 Department of Psychological Medicine, John Radcliffe Hospital, Oxford, and 2 Department of Medicine, Liverpool University, Liverpool, UK
Received 20 June 2000 and in revised form 13 September 2000
 |
Summary |
|---|
 Top Summary IntroductionMethodsResultsDiscussionReferences |
|---|
We describe the referral and management of consecutive patients attending a cardiac service with the presenting complaint of chest pain. Of 610 consecutive new referrals to five Oxford cardiac clinics over 12 weeks, 202 had chest pain as the presenting complaint: 91 (45%) angina, 101 (50%) non-cardiac chest pain, 8 (4%) both and 2 (1%) uncertain diagnosis. Information in clinic letters was sometimes ambiguous and contradictory and suggested a lack of precise information to patients. Patients with non-cardiac chest pain often had long histories, including considerable previous use of services and specialist investigations. There were delays in referral and assessment of patients. There are opportunities for simple changes in assessment procedures which might have substantial advantages for outcome and resource: (i) more detailed referral information from general practitioners, with an explicit statement of the reasons for referral; (ii) minor modifications to augment the assessment by provision of unambiguous information to patients and primary care at discharge.
|
Introduction |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Guidelines1–4 for referral, diagnosis and early treatment, and descriptions of rapid access clinics5,6 and special assessment units are enabling improved early diagnosis and immediate treatment of patients presenting with chest pain in out-patient and emergency department care. However, they do not appear to be effective in initiating continuing comprehensive care for those who do not have myocardial infarction. Few of those with angina and other heart disease are offered planned aftercare or rehabilitation. Even less is offered to those who are thought not to have heart disease; despite being reassured, many of these patients report continuing symptoms and disability.7,8 Delivery of care might be improved in two ways: (i) changes in the clinic assessment—the organization, procedures, advice and treatment; (ii) extra long-term care following the assessment. We need to achieve this in a manner that is effective, acceptable and feasible within the routine care of very large numbers of people.
Before suggesting extra long-term treatment services in hospital or primary care, it would be sensible to consider whether there are simple ways in which current clinical assessment (which includes initial treatment) could be ‘augmented’ to make it more effective. Our experience suggests that this may well be a crucial time for influencing patients’ subsequent concern and behaviour. We investigated the scope for changes in organization and content of assessment and for better communication with primary care in a sample of consecutive referrals of Oxfordshire patients with the presenting problem of chest pain to all cardiac clinics.
|
Methods |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
All new referrals to all five Oxford consultant cardiologists were identified over a 12-week period from clinic lists and the systematic collection of discharge letters. Patients whose main presenting complaint was chest pain were identified. Over a 12-week period, we included the 202 patients (from the total of 610) who were new referrals to the cardiac clinics for assessment of chest pain. We excluded those in whom chest pain was mentioned incidentally as a symptom, but at assessment and diagnosis was not the major problem. We also excluded a number or people for whom a diagnosis of chest pain/angina of other cause of angina had been established in the past, and whose current referral was for another symptom.
Information from letters and notes was coded on standard data sheets. Whilst we obtained almost complete information on many variables, there were a number for which little information was available, for example: date of first presentation to the general practitioner, duration of the present episode, details of distinction between current and previous episodes, number of consultations with the general practitioner. Other items whose presence or absence were not consistently recorded included, e.g. precipitants, associated symptoms, risk factors and general practitioners' provisional diagnosis. Complete data was available for all the variables analysed.
The 
2 test was used to compare category differences between the two main groups, independent group t test to compare age differences and Mann–Whitney U test to compare referral time differences. All analyses were performed using the SPSS.
|
Results |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
Over the survey period of 12 weeks, there were 610 new referrals to the five cardiac out-patient clinics. The reasons for referral were very varied, and at least a third were for non-specific complaints apparently not due to heart disease. These included 57 with palpitations but no direct evidence for arrhythmias, and many with breathlessness (45), dizziness (13) and fainting (30). There were 47 who were referred for a review of cardiac murmurs or other routine findings (Table 1
).
|
Thirty-one (15%) of 202 subjects with chest pain were referred from other hospital-based specialties within Oxford, and 34 (17%) were from outside the health district. Following their clinic visit and further investigation, 101 patients were diagnosed as non-cardiac and 91 as angina. In presenting the results we concentrate on these two large groups, but there were two small additional subgroups (5%). These included two patients in whom the diagnosis was undecided: one had no symptoms at the time of consultation and the other was thought not to require intensive investigations and was given a follow-up appointment for review. Further, dual diagnoses of both angina and non-cardiac pain were made in eight patients. Some of these were patients with proven cardiac problems (and, usually previous coronary artery surgery) who presented with pains which were thought to be partly, or possibly entirely, non-cardiac in origin. The remainder were patients with shorter histories which suggested non-cardiac causes but in whom investigations were equivocal. Three of these eight were offered further clinic follow-up.
Previous history
There were considerable differences in the amount of information provided by the referring general practitioners about history, medication, possible psychological and social factors and previous investigations and current treatment. There was little evidence in the letters about how long the chest pain had been treated in general practice. Only a minority asked very precise questions about diagnosis and management.
Those eventually diagnosed as having non-cardiac pain were much less likely than those diagnosed with angina to have been given a definite cardiac diagnosis in primary care (35% compared to 78%, p<0.001). Only one fifth of the letters about those eventually diagnosed as having non-cardiac pain mentioned non-cardiac causes as likely.
One hundred and eleven patients were diagnosed as suffering from angina or ischaemic heart disease by the general practitioner, and this diagnosis was confirmed by the cardiologist in 69 (62%). Twenty-seven patients were diagnosed in the referral letter, and this was confirmed in all except five.
Subjects diagnosed as having non-cardiac chest pain were slightly more likely to have consulted general practitioners or psychiatrists about nervous problems than those with angina (22% vs. 15%; NS). Twenty-four percent of the non-cardiac group were noted to have histories of consultations for previous unexplained physical symptoms compared with 11% of those with angina (p<0.05). Twelve percent had previously been referred to a cardiac clinic, two in the course of their current episode of symptoms and 11% to another specialist clinic. Nineteen (19%) of the non-cardiac and 18 (20%) of the angina group had previously attended the emergency department for chest pain. Clinical characteristics are shown in Tables 2 and 3. The characteristics of the 41 subjects who had previously attended cardiac and other specialist clinics for assessment of their chest pain are listed in Table 4.
|
|
|
Assessment
The median time from referral letter to out patient appointment was 57 days (range 5–161) for angina patients and 66 days (range 5–192) for non-cardiac subjects (p=0.07). The median time from first appointment to conclusion of assessment was 4 days (range 0–366) for non-cardiac patients and 42 days (range 0–402) for the angina group (p<0.001). There were differences between the five clinics in waiting times and to a lesser extent in duration of assessment.
A major limiting factor in completing the assessment was the problem of long delays caused by patients being placed on waiting lists for specialist investigation, especially angiography. This was particularly evident for patients with a final diagnosis of angina, but it also delayed conclusions for a number of non-cardiac patients. For example, although the median time before establishing a diagnosis of non-cardiac chest pain was 4 weeks, 25 (25%) patients waited between 10 and 52 weeks. (See Figure 1.)
|
Angina
Fifty-two patients diagnosed with angina had angiography (one at another hospital before referral), which in 33 cases resulted in referral for coronary artery surgery. For all, except four of the remainder, changes in drug treatment were recommended. Only two patients attended for regular follow-up after the diagnosis had been made.
There were relatively few comments in letters to general practitioners about other aspects of medical management, such as changes in life-style or effects on everyday behaviour. In a minority there were comments that associated social or psychological problems may have been contributing to disability and problems in care. It was usually unclear from the letter what information had been given to patients.
Non-cardiac chest pain
Of the 101 patients with non-cardiac chest pain, 45 were given a specific anxiety or minor physical explanation: of these, three were given more than one specific diagnosis (hyperventilation and stress, musculo-skeletal and stress, musculo-skeletal or gastrointestinal). Two non-cardiac patients were offered further follow-up appointments following completion of the assessment. Non-cardiac patients were younger than cardiac patients (mean age of 52 compared with 64, p<0.001) and more likely to be women (51% compared with 22%, p<0.001). They were more likely to mention palpitations and other non-specific physical symptoms (apart from breathlessness).
Cardiologists referred to the difficulties of reassuring some anxious patients. One letter stated ‘I am convinced her symptoms are nothing to do with her heart but she is resistant to this explanation’. Another said of a 65-year-old woman: ‘I spent a long time with her... unfortunately she had been told that she had an abnormal heart in the past—this is incorrect’. For another it was said: ‘the symptoms are difficult to assess in the setting of what seems to be a chronic anxiety state and multiple somatic symptoms... on examination... she seemed anxious and rather depressed.’
Comments in the notes and letters frequently suggested some degree of doubt about the diagnosis. In some of these cases, cardiologists felt that further investigation would not be justified, for example: ‘I explained that his history was not typical of angina. However it is impossible to exclude it entirely...’. ‘He may have mild angina, but in view of the excellent results on the exercise test, he would be inappropriate to be investigated at the moment’. Some letters gave ambiguous or inconsistent advice (for example, reassurance about diagnosis of non-cardiac pain but continuing anti-anginal medication) and provided little information about what was said to the patient or advice on future management. Explicit recommendations about stopping medication were made for only 19 of the 60 patients who were taking cardiac drugs (including aspirin) before the consultation. Furthermore, it was often not clear from correspondence if beta blockers were continued for the treatment of chest pain or for hypertension. The evidence from the notes and the letters suggests that few had specific reassurance, and that in many cases advice given to general practitioners was to some degree inconsistent and ambiguous. There were a number of comments in the letters which might have given mixed messages to general practitioners and certainly to patients: ‘I reassured her very strongly that the pain is not angina... GTN if she wishes’ (Table 5
).
|
|
Discussion |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
We have suggested that, as in the standard treatment of hypertension, care for patients with chest pain should be ‘stepped’. This would result in all patients receiving good routine care, with those who need extra help obtaining the treatment specifically tailored to their needs.9 The present review of a cohort of attenders at a District hospital clinic, together with previous evidence from prospective studies of patients with chest pain, suggests ways in which changes might be made. The advantages of the present study are twofold. First, that it is one of the few studies of new out-patient referrals to a cardiac clinic, and second, it pays as much attention to those whose chest pain was not thought to be due to heart disease as to those who received a final diagnosis of angina. Inevitably, there are limitations to any study based solely on examination of hospital notes and correspondence. However, the characteristics of patients described from records in the present study are consistent with those of an interview study of chest pain and palpitation attenders previously reported.10 We should also be cautious about generalizing from a single health district to other areas and other centres. Even so, published evidence and clinical experience suggest that despite limitations of data based on letters and notes, the study provides a clear account of the general clinical problems facing all district cardiac services.
Can we improve the referral process?
Referral letters varied considerably in the extent of information provided and in the extent to which there was an explicit reason for referral, such as diagnosis and advice about treatment. It seems likely that if general practitioners and other referring doctors had guidelines about referral then they could, in many cases, have provided more explicit and relevant information about previous histories and care, as well as formulating specific questions for the cardiologists. Furthermore, probable non-cardiac chest pain was often apparent at the time of referral, from both the letter from the general practitioner and previous hospital notes. Questions might include asking the cardiologist to provide reassurance and a satisfactory explanation for the pain if all investigations are normal.
Can we improve the cardiac clinic assessment?
Administration
The study shows that there were delays in referral, between referral and appointment (particularly evident at the time in which the survey was conducted) and in access to investigations. The practical problems of very busy clinics with delays in appointments, difficulty and long delays in arranging both simple and more specialist investigations, and shortage of time to talk to patients were all conspicuous. These delays and the associated uncertainty and concern about heart disease for patients are likely to reinforce their anxiety and lead to disability which may then become difficult to reverse by simple reassurance.
Angina
The diagnosis or confirmation of diagnosis of angina frequently resulted in significant changes in treatment: either recommendation for surgery (in a third) or major changes in medication. It is notable however that there was little advice about secondary prevention e.g. lifestyle change or advice about physical activity or other behaviour. Again, patients with angina might have benefited from further intervention to treat associated anxiety and behavioural problems. Patients with dual diagnoses of chest pain and ischaemic heart disease (4% of the total) appeared to have caused particular diagnostic difficulty.
Non-cardiac pain
The non-cardiac patients were overall consumers of substantial amounts of health resources, and many had histories of previous consultation and investigation. Drug prescribing did not closely follow diagnosis, and in many instances there were no recommendations about discontinuing beta blocker, aspirin or nitrates. In some cases there was a suggestion that these should be continued. Such a policy may have adverse consequences: prescription of anti-anginal medication for patients with non-cardiac chest pain may have iatrogenic potential, and lead to the development of chronic pain and disability.11
Undecided
Further investigation is sometimes clinically not justified to establish a definite diagnosis. Patients differ considerably in their ability to cope with uncertainty, and follow-up review in the clinic or primary care is sensible to ensure that patients have returned to full everyday life.
Dual diagnosis
The combination of ischaemic heart disease and non-cardiac chest-pain is common and can be associated with considerable, unnecessary disability and with extra use of resources. Specific extra intervention may be indicated and greater attention to specialist after-care is often warranted.
Is augmenting the cardiological assessment feasible?
It is unrealistic to expect cardiologists to devote substantial extra time to patients with non-cardiac pain. But recent surveys of out-patient and rapid assessment chest pain services have confirmed that they comprise between 50 and 60% of their work load.12 It is important therefore to optimize the assessment so that subsequent management of these patients might be improved. Price13 recently reviewed the possible therapeutic benefits of adopting such an approach. Simple measures, such as enquiring after the patient's major concerns, and providing a satisfactory alternative explanation for the pain, for example panic disorder or oesophagitis may be helpful. The benefits of this approach, accompanied by unambiguous reassurance, are often enhanced when delivered in the presence of a spouse or close friend. Provision of a simple handout explaining the ways in which non-cardiac chest pain can occur might assist this process. Withdrawing or tapering anginal medication would give a powerful message to the patient and, in the absence of conspicuous ischaemic heart disease, would prevent the development of potential iatrogenic complications. Finally, sending a letter to the patient's GP with advice about medication, exercise, etc. would augment the power of the explanation in the clinic.
There will be some patients in whom there is continued diagnostic doubt, as well as the small group with both ischaemic heart disease and non-cardiac pain. These patients may require follow-up and extra help, perhaps provided by specialist nurses or in the community; a minority may even require specialist referral for psychological treatment.9 These changes could be expected to improve patient outcome and ultimately to lead to savings in the cost of continuing to provide primary and secondary care.
Can we improve communication with primary care?
There would be advantages in general practitioners providing more clinical detail and explicit questions for cardiologists at referral. Following the assessment by the cardiologist there are opportunities for conveying clear advice to general practitioners about medication, secondary prevention and planning of measures to minimise distress and disability. It would be useful if letters had a systematic format and included a section summarizing information given to the patient.
Can we do more to improve psychiatric and psychological expertise in cardiology services?
Whilst we believe there should be easier access to clinical psychologists and psychiatrists for the assessment and treatment of the most severe persistent psychological problems, improving routine care for the increasing numbers of patients who are being referred for the assessment of chest pain will depend upon cardiologists being more aware of psychological issues and the need to address them. It is also likely to require the appointment of specialist chest pain nurses whose responsibilities include explanation, discussion of worries and the provision of self-help information together with follow-up for a proportion of subjects.
Where can we go from here?
The findings of this audit reinforce conclusions from interview studies that ‘stepped care’ offers considerable opportunities for delivering proven and recommended treatment for patients with both angina and non-cardiac chest pain. The first step, which we believe involves changes in the initial assessment procedure, requires further specification and evaluation. At the same time we should consider the ways in which medical secondary prevention and other care can be provided in an individualized manner to those who require extra help in further therapeutic steps.
|
Notes |
|---|
Address correspondence to Professor R.A. Mayou, University Department of Psychiatry, Warneford Hospital, Oxford OX7 3JX. e-mail: richard.mayou{at}psych.ox.ac.uk
|
References |
|---|
|
TopSummaryIntroductionMethodsResultsDiscussionReferences |
|---|
1. North of England Stable Angina Guideline Development Group. North of England evidence based guidelines development project: summary version of evidence based guideline for the primary care management of stable angina. Br Med J1996; 312:827–32.
2. NHS Centre for Research & Dissemination. Management of stable angina. Effective Health Care1997; 3(5):1–7.
3. Working party of the joint audit committee of the British Cardiac Society and the Royal College of Physicians of London. The investigation and management of stable angina. J Roy Coll Physicians1993; 27:267–71.
4.
Hlatky MA. Evaluation of chest pain in the emergency department. N Engl J Med1997; 337:1687–8.
5.
Davie AP, Caesar D, Caruana L, Clegg G, et al. Outcome from a rapid-assessment chest pain clinic. Q J Med1998; 91:339–43.
6.
Newby DE, Fox KAA, Flint LL, Boon NA. A ‘same day’ direct-access chest pain clinic: improved management and reduced hospitalization. Q J Med1998; 91:333–7.
7. Potts SG, Bass CM. Psychological morbidity in patients with chest pain and normal or near-normal coronary arteries: a long-term follow-up study. Psychol Med1995; 25:339–47.[Web of Science][Medline]
8. Gill D, Mayou R, Dawes M, Mant D. Presentation, management and course of angina and suspected angina in primary care. J Psychom Res1999; 46:349–58.
9. Mayou R, Bass C, Bryant B. Management of non-cardiac chest pain: from research to clinical practice. Heart1999; 91:387–92.
10.
Mayou RA, Bryant B, Forfar C, Clark DM. Non-cardiac chest pain and benign palpitations in the cardiac clinic. Br Heart J1994; 72:548–53.
11. Kouyyanou K, Pither C, Wessely S. Iatrogenic factors in chronic pain. Psychosom Med1997; 59:597–604.[Abstract]
12.
Capewell S, McMurray J. ‘Chest pain—please admit’: is there an alternative? Br Med J2000; 320:951–2.
13. Price JR. Managing physical symptoms: the clinical assessment as treatment. J Psychosom Res2000; 48:1–10.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Tyrer From the Editor's desk The British Journal of Psychiatry, June 1, 2009; 194(6): 578 - 578. [Full Text] [PDF]
|
|
|
|
 |
|
 G. Nijher, J. Weinman, C. Bass, and J. Chambers Chest pain in people with normal coronary anatomy BMJ, December 8, 2001; 323(7325): 1319 - 1320. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||