The aim of this guideline is to provide recommendations to aid health professionals in primary care in the management of patients with chronic stable angina. When possible recommendations are based on and explicitly linked to the evidence that supports them. The guideline does not deal with unstable angina or myocardial infarction. It is an updated version of the North of England Stable Angina Guideline produced in 1996.

The development group assumes that healthcare professionals will use general medical knowledge and clinical judgment in applying the general principles and specific recommendations of this document to the management of individual patients. Recommendations may not be appropriate for use in all circumstances. Decisions to adopt any particular recommendation must be made by the practitioner in the light of available resources.

Guideline development group

The guideline development group was composed of three broad groups: relevant healthcare professionals and patients; a specialist resource and a specialist small group leader; and members of the research team. It was important that the group contained individuals from appropriate sectors; the sectors approached were general practitioners, practice nurses, and secondary care physicians. These group members were there to ensure adequate relevant discussion of the evidence, topics for which there was no evidence, and the subsequent recommendations in the guideline. The specialist resource was a consultant cardiologist. He worked closely with the research team, discussing the literature search strategy, reviewing and summarising the literature, and feeding this information back to the group. The small group leader had the role of ensuring that the group worked effectively. Together the research team was responsible for drafting the guideline and resourcing the guideline development group

Identification, review, and synthesis of evidence

The guideline was based on an original systematic review but also made use of existing systematic reviews when appropriate. When existing systematic reviews were used the authors’ conclusions were reproduced; we did not attempt to replicate their meta-analyses. However, this approach created several problems. We took on trust the quality of the reviews, and as they are all from either the Cochrane Collaboration or the NHS Centre for Reviews and Dissemination we felt justified in doing this. There were occasions when the focus of the review did not exactly match the clinical question or, if appropriately focused, did not fully answer it. When this occurred it is flagged within the guideline.

Search strategy

The aim of the evidence review was to identify and synthesise relevant evidence within the published literature to allow recommendations to be based on evidence whenever possible.(1) The search strategy was carried out using the electronic databases Medline, Embase, and the Cochrane library and covered the period January 1994 to December 1997. This built on the search for the previous version of the guideline that was based on January 1984 to September 1994. Our search strategies attempted to locate studies using a combination of MeSH and free text searches. The search strategy was backed up by the expert knowledge and experience of the group members. We did not attempt to access the grey literature (unpublished reports and abstracts) nor did we identify letters in response to original articles.

Sifting the literature

The set of references generated by the search was sifted for relevance to the clinical topic of the guideline. The initial sifting was done by a clinically qualified health services researcher on the basis of the article titles. Articles concerned with non-relevant clinical areas were removed, as were editorials and letters. If there was any doubt about an article’s relevance it was retained for the next round of sifting. This was performed by the specialist resource (the cardiologist) and was done on the basis of title and abstract (when this was available). This identified references that needed to be collected for more detailed assessment. The number of references rejected at each stage of this sifting process are listed in table 1.

Table 1 Number of references remaining at various stages of sifting process

    Stage of sifting process
    Total No of 

    papers remaining

     
    Medline and Embase searches
    8586
    After duplicate check
    5941
    After clinical health service researcher (title) sift
    324
    After consultant (abstract) sift
    141
    Plus 31 references from other sources (references cited by papers etc)
    172
    Papers within scope of guideline
    145
    Papers with acceptable method
    59
    Plus 104 papers from previous guideline
    163
 
 

Reviewing the literature

Table 2 shows the number of papers rejected at this stage and the reason for rejection. Certain criteria in the methods were not regarded as absolute flaws but were of sufficient importance to merit mention in the text. The commonest of these was an experimental study at risk of a type II error. This means that the study had a negative result (it concluded there was no difference between groups) but it did not provide any evidence that the study had sufficient power to detect a difference were one there. This is referred to in the guideline as either "at risk of a type II error" or "a negative study without a power calculation."

Table 2 Criteria for rejecting papers

    Reason for rejecting paper
    No

    rejected 

     
    Relevance:  
    Not relevant (patient group, setting)
    27
    Superseded (eg, by good review or interim results by final results)
    4
    Methodological criteria:  
    <80% follow up
    3
    Group sizes <20
    33
    No control group
    2
    Not a helpful design (eg not RCT in clinical area where RCTs already available)
    18
    Open when could be blinded
    5
    Non-systematic review
    9
    Other design flaws
    7
    Duplicate publication (results reported in more than one journal)
    5
 
 

The studies on drug use were categorised with explicit criteria into two groups. Those studies that had high internal and external validity, reported results of exercise tests, and measured the outcomes that were likely to be available to a general practitioner (glyceryl trinitrate use, frequency of angina) were most relevant. Studies that did not fulfil all these criteria were used as long as they had high internal and external validity and reported duration of exercise. Studies that did not fulfil these criteria were cited only as "supporting papers."

Synthesising the literature

The guideline group members considered the relevant clinical areas of the guideline and defined the questions that they wanted the evidence to answer within these areas. This guided the interpretation of the evidence. The retrieved papers were assessed for their quality and their ability to provide valid answers. Assessment of study quality concentrated on questions of internal validity (the extent to which the study measured what it intended to measure), external validity (the extent to which study findings could be generalised to other treatment settings), and construct validity (the extent to which measurement corresponds to theoretical understandings of a disease).(2) After individual papers had been checked for methodology and clinical significance, the information was synthesised. Questions were answered by using the best evidence available. When we considered the effect of an intervention, if a question could be answered by category I evidence provided by a meta-analysis or randomised controlled trial then we did not use studies of weaker design (controlled studies without randomisation). The evidence was synthesised by using qualitative methods. These involved summarising the content of the papers into brief statements that the group thought accurately reflected relevant evidence. Quantitative (meta-analysis) techniques were not used de novo. Recommendations were derived with informal consensus methods.

Evidence categories were adapted from the Agency for Health Care Policy and Research Classification. (3) Four categories are available; this categorisation is most appropriate for questions of causal relations. Similar taxonomies for other types of research questions do not yet exist.
 

Categories of evidence

Strength of recommendation

In the meetings summaries of the evidence were presented to the group. These were interpreted in the light of the clinical area and the recommendations in the previous version of the guideline. Recommendations were graded A to D as shown below. The guideline distinguishes between the category of evidence and the strength of the associated recommendation. It is possible to have methodologically sound (category I) evidence about an area of practice that is clinically irrelevant or has such a small effect that it is of little practical importance and would therefore attract a lower strength of recommendation. More commonly, a statement of evidence would cover only one part of an area in which a recommendation had to be made or would cover it in a way that conflicted with other evidence. Consequently, to produce comprehensive recommendations, the group had to extrapolate from the available evidence. This leads to lower strength recommendations based on evidence category I statements.

Areas without evidence

We accepted that there would be areas without evidence where recommendations had to be made and that consensus would be required to deal with these areas. The group used informal consensus. When this process identified important unanswered research questions these were recorded at the end of the relevant section of the guideline.
 

Scheduled review of the guideline

The guideline should be reviewed no later than three years after its completion.

Background

Most patients with angina are initially seen and managed in general practice. The most recent national survey of consultations with general practitioners (GPs) suggests a crude incidence of new episodes of angina (ICD-9 code 413, international classification of diseases, 9th revision) of 0.52% and prevalence of 1.14% per 1000 population.(4) These rates are similar to those from studies on selected populations after adjustment for age.(5) (6) (7) In all age groups angina is more common in men than women, but because of their relative longevity more women consult for angina in old age (figure). On average a GP can expect to provide care for 23 patients with angina who typically make 2-3 visits each year associated with the condition.

Figure 1


 Prevalence of patients consulting for angina pectoris in English primary care (assuming a list size of 2000)

In a national survey of self reported angina in England in 1994, 4.3% of men reported experiencing angina at some time, 3.1% in the past year; in women these figures were 3.4% and 2.3%.(8) Adults aged 16 or over were surveyed, which partially explains the higher rates found than in the primary care survey. Of respondents reporting angina, 8.4% reported having had a clinically diagnosed heart attack in the previous 12 months (9.7% of men and 7.1% of women). Other studies have pointed to the raised risk of myocardial infarction in patients with angina. In a cohort of 499 patients in Nottingham, identified by prescribed use of nitrates and followed for seven years, 26% were admitted with a suspected myocardial infarction, although this was diagnosed as probable only in 14%.(9) A prospective cohort study in an open access chest pain clinic in Southampton identified 110 out of 467 referred patients as having typical angina. In these 110, after a median follow up of 16 months, 11% of patients had spontaneously remitted, 19% had undergone revascularisation, 7% had suffered a non-fatal myocardial infarction, and 4% had died.

For the NHS in England in 1994-5, there were 77 565 inpatient admissions for angina and 284 292 admissions for ischaemic heart disease as a whole.(10) Ischaemic heart disease accounted for 3.5% of all inpatient admissions in England and 2.7% of all bed days. Admission to hospital costs an estimated £86 million for angina and £311 million for ischaemic heart disease (applying specialty costs from the 1997 CiPFA/HfM Health Database to outpatient and inpatient episodes).

Deaths from ischaemic heart disease, including myocardial infarction, are common. In England and Wales in 1995 there were 133 861 recorded deaths for ischaemic heart disease, 23.5% of all recorded deaths.(11)
 

Cost effectiveness of management in primary care

Patients presenting with angina in general practice are likely in the course of time to be high users of healthcare resources. Ideally, it would be desirable to know how different strategies of diagnosis, management, and use of drugs, implemented at presentation, might effect patient outcomes and the long term use of resources. This level of information is not available. Hence the scope for formal cost effectiveness analyses of different approaches to treatment or management in primary care is limited.

Clinical assessment and evaluation, with identification and modification of risk and precipitating factors, are identified in this guideline as important elements of a baseline assessment in general practice. An exercise ECG is advocated for its prognostic information, particularly in screening for patients requiring further investigation. The value of all this information in incremental improvment in health outcomes (by subsequent intervention) has not been shown.

Patients with angina may use a sequence of drugs, beginning with a sublingual glyceryl trinitrate preparation for rapid symptomatic relief. Forms include tablet and aerosol, but as the effect may last only 30 minutes, modified release and percutaneous preparations have been developed, as well as isosorbide dinitrate and mononitrate for prophylactic use. Prophylactic treatment is recommended for patients requiring regular relief of symptoms and should first be attempted with a b blocker.

Considerable variations exist in the cost of nitrates, although treatment provided for immediate symptomatic relief cannot necessarily be compared with prophylactic forms. In 1997 the Prescription Pricing Authority reimbursed prescriptions for nitrates in English primary care to the value of £60 million (table 3). This amount is totally dwarfed by over £213 million paid out for calcium channel blockers prescribed for a range of conditions including angina. However, it is of interest to note that two thirds of the cost of nitrates went in the purchase of isosorbide mononitrate and nearly all of this accrued because of the use of expensive proprietary modified release forms, implying its use for prophylaxis of symptoms. It is unclear how many patients currently receiving isosorbide mononitrate alone could be more appropriately managed on a b blocker alone. A simple analysis of volume of use is provided, in terms of script items, although these may not strictly be comparable across classes of drugs (table 3)

Table 3 Reimbursement for selected drugs used for angina and other conditions, England 1997

     
    Reimbursed cost

    (£ millions)

     
    Script items (thousands)
    Cost/item (£)
    Nitrates:
    60.5 
    8370 
    7.23
    Glyceryl trinitrate
    17.8 
    3222 
    5.51
    Isosorbide dinitrate
    1.6 
    529 
    3.04
    Isosorbide mononitrate
    41.5 
    4616 
    9.00
    b blockers
    78.0 
    14 811 
    5.26
    Calcium channel blockers
    213.5 
    13 390 
    15.94

b blockers themselves vary greatly in price. A recent review of the use of b blockers after myocardial infarction found no evidence of improved efficacy or compliance between b blockers with different selectivities.

There are considerable long term costs associated with angina and its sequelae. Appropriate sequencing of drugs, reflecting the evidence of effectiveness presented in this guideline, and avoiding the use of expensive proprietary forms of drugs that have questionable additional benefits, may ensure the best use is made of limited resources.