Chest pain in primary care: Epidemiology and pre-work-up probabilities

Abstract

Background/objective: Chest pain is a common complaint and reason for consultation. We aimed to study the epidemiology of chest pain with respect to underlying aetiologies and to establish pre-work-up probabilities for the primary care setting. Methods: We included 1212 consecutive patients with chest pain, aged 35 years and older, attending 74 general practitioners (GPs). GPs recorded symptoms and findings of each patient and provided follow-up information. An independent interdisciplinary reference panel reviewed clinical data of every patient and decided on the aetiology of chest pain at the time of patient recruitment. Results: The prevalence of chest pain among all attending patients was 0.7%. The majority (55.9%) of patients were women. Mean age was 59 (35–93) years. Of these patients, 53.2% had chest pains at the time of consultation and 29.6% presented with acute (<48 hours’ duration) chest pain. Pain originating from the chest wall was diagnosed in 46.6% of all patients, stable ischaemic heart disease (IHD) in 11.1%, and psychogenic disorders in 9.5%; 3.6% had acute coronary syndrome (ACS).

Conclusion: The study adds important information about the epidemiology of chest pain as a frequent reason for consulting primary care practitioners. We provide updated pre-work-up probabilities for IHD for each age and sex category.

Keywords::

• Chest pain
• prevalence
• primary healthcare
• family practice
• myocardial ischemia

Introduction

Chest pain is a common complaint and reason for consultation in primary care. It affects about 20 to 40% of the general population in their lifetime (1). However, incidence varies according to setting, country, and inclusion criteria (2–4).

Chest pain can be caused by a wide range of different illnesses, with life-threatening cardiac disease being of the greatest immediate concern (5,6). While serious cardiovascular disease (CVD) is not a common reason for chest pain in primary care, musculo-skeletal pain is the most frequent aetiology for chest pain in several studies (3–5,7).

For most patients suffering from chest pain, the general practitioner (GP) will be the main point of entry into the healthcare system. The primary care physician faces the challenge of having to identify patients at risk and to initiate further diagnostic procedures in a timely manner. On the other hand, GPs are expected to fulfil their gatekeeper role by protecting patients from over-diagnosis and inappropriate treatment. In this process, pre-test and work-up probabilities can be a crucial aid for the GP. However, these probabilities are subject to changes in epidemiological (incidence and prevalence of ischaemic heart disease [IHD], risk factors) and health system utilization patterns, and therefore require regular updates.

There is still a need for data on prevalence, presentation, and aetiology of chest pain in primary care derived from a larger sample in conjunction with an independent reference standard. To our knowledge, this is so far the largest study investigating the epidemiology of chest pain in a primary care setting, providing in addition updated data on pre-test probabilities for IHD according to age and sex (8).

Material and methods

We conducted a cross-sectional diagnostic study with a delayed-type reference standard in a primary care setting (9). The final diagnosis was established by an expert panel after 6 months of follow-up. In this article, we report epidemiologic results, while the accuracy of clinical criteria will be published separately.

Participating GPs and patients

We approached 209 GPs in the state of Hesse, of whom 35.4% agreed to participate in the study. Only GPs being prepared to undergo random recruitment audits could take part. Participating practices had to recruit consecutively every attending patient who had chest pain either as a presenting complaint or on questioning. The recruitment period lasted 12 weeks for each practice. For logistical reasons, recruitment was staggered in four waves between October 2005 and July 2006.

Every patient above 35 years with a complaint of pain, tightness, or oppression localized in the area between the clavicles and lower costal margins and anterior to the posterior axillary lines was to be included. Doctors were also asked to recruit at home visits and emergency calls. Patients were eligible irrespective of the acute or chronic nature of their complaint, or of previously known conditions including IHD or related risk factors. Patients whose chest pain had subsided for more than 1 month, whose chest pain had been investigated already, and/or who came for follow-up for chest pain were excluded. In emergency situations without sufficient time for patient information and written consent, relevant clinical items were documented on a case report form (CRF) kept by the GPs. Later, e.g., after discharge from hospital, the patient was asked to participate in the study. Only if he/she gave informed consent was the CRF handed over to study personnel. Like the whole study protocol, this procedure was approved by the ethics committee of the Faculty of Medicine, University of Marburg. The study complies with the Declaration of Helsinki.

For each practice, all patient contacts were counted during a randomly selected week of the recruitment interval and extrapolated for the whole recruitment period to calculate the prevalence of chest pain among practice attendees.

Data collection

GPs took a standardized history and performed a physical examination according to a CRF that was piloted and modified accordingly. They also recorded their preliminary diagnoses, investigations, and management related to the patients’ chest pains. Patients were contacted by phone 6 weeks and 6 months after the index consultation. Study assistants blinded to clinical data already recorded asked about the course of patients’ chest pain, treatments including hospitalization, and drugs. Discharge letters from specialists and hospitals were requested by GPs.

Precautions against bias

Participating practices were recruited from a network of research practices associated with our department. To GPs, we emphasized the importance of recruiting every patient with chest pain irrespective of the presumed likelihood of IHD. Practices were visited at 4-week intervals to check CRFs, recruitment logs, and compliance with study procedures. Random audits were performed in order to search the routine documentation of participating practices to identify cases of chest pain not included in the study.

After 6 months, a reference panel consisting of one cardiologist, one GP, and one member of the research staff at the Department of Family Medicine reviewed the baseline and follow-up data of each patient. Analysing all the information gathered during the follow-up period (results of further investigations, letters from specialists, hospital discharge reports, etc.), they decided on the most likely medical condition having caused an individual patient's chest pain at baseline.

Statistical analysis

Analyses stratified by sex, age, and known IHD status are presented as bar graphs. We used chi-square to test the significance of sex differences and the history of IHD with regard to the aetiology of chest pain. The t test was used to calculate age differences between male and female participants. Since statistical tests are regarded as being descriptive, we did not adjust significance levels to multiple testing. Proportions were calculated using two-by-two frequency tables. Ninety-five percent confidence intervals (95% CI) were calculated for proportions. Analysis was performed using SPSS (version 14.0).

Results

The majority of participating GPs were male (67%); two-thirds of practices were located in urban areas (63.5%). Mean age of GPs was 49 years. According to our estimate, these 74 GPs encountered around 190 000 patients during the study period and approached 1355 patients with chest pain. Seven patients did not meet the inclusion criteria, and 99 refused to participate in the study. Sixty cases were lost to follow-up, and 11 died but provided enough information to be judged by the reference committee; three cases were early dropouts and were therefore not included. For 34 cases, follow-up information was lacking, incomplete, or ambiguous, so that no final diagnosis could be made. We thus analysed 1212 patients for the aetiology of their chest pain (Figure 1).

Figure 1. Patient recruitment.

The prevalence of chest pain during the study period was 0.7% (95% CI: 0.66–0.73) of all patient contacts. There were slightly more women (55.9%) than men. Mean age was 59 years, ranging from 35 to 93 years. Women were on average older than men (60.6 vs 58.0 years, p <0.01).

Of the analysed patients, 53.2% had chest pains at the time of consultation and 29.6% presented with acute, i.e., duration <48 hours, chest pain. The vast majority of patients (91.9%) were known by their GP from former consultations, and most patients (87.5%) quoted chest pain as the reason for the actual consultation; 29.6% of patients presented with acute chest pain (<48 hours).

In the majority, pain originated from the chest wall (“chest wall syndrome”), followed by stable IHD, psychogenic disorders, and upper respiratory infections; 3.6% of patients had acute coronary syndrome (ACS) (Table I). We found two cases (one definite and one possible) of pulmonary embolism. Aortic dissection did not occur in this sample.

Table I. Final diagnoses in patients presenting with chest pain to their GP (n = 1212).

Diagnosis	Frequency (n = 1212)	% (95% CI)
Chest wall syndrome	565	46.6 (43.8–49.5)
Ischaemic heart disease (stable)	135	11.1 (9.4–13.1)
Psychogenic disorders	115	9.5 (7.9–11.3)
Upper respiratory infections	98	8.1 (6.6–9.8)
Hypertension*	48	4.0 (2.9–5.2)
Acute coronary syndrome	44	3.6 (2.7–4.8)
Gastro-oesophageal reflux disease	42	3.5 (2.5–4.7)
Trauma	39	3.2 (2.3–4.4)
Benign stomach problems	26	2.1 (1.4–3.1)
Pleuropneumonia	25	2.1 (1.3–3.0)
COPD/asthma	23	1.9 (1.2–2.8)
Other	52	4.3 (3.2–5.6)

*Patients presenting with hypertensive urgency in whom chest pain consequently subsided when blood pressure was lowered. CI: confidence interval; COPD: chronic obstructive pulmonary disease.

Analyses stratified by age, gender, and history of IHD are shown in Figure 2. Male patients showed a significantly higher proportion of trauma (p<0.01) and pneumonia (p =0.04) as underlying aetiology. In contrast, female patients showed a higher proportion of psychogenic disorders (p =0.02). Stable IHD and ACS occurred more frequently in men (IHD: M 12.4%, F 9.6%; ACS: M 4.4%, F 2.8%), although these differences were not statistically significant (p =0.11 for IHD; p =0.15 for ACS).

Figure 2. Aetiology of chest pain in relation to age, gender, and history of IHD (n = 1212).

Diagnoses stratified by age groups showed an increase of IHD and ACS towards older age groups; chest wall syndrome was relatively more frequent in younger age groups. Patients with a previous history of IHD showed significantly higher proportions (p <0.01) of IHD and ACS as causes of their chest pain.

Table II shows probabilities for any IHD, stratified by age and gender. Probabilities rise in both age groups, with men already showing higher probabilities at a younger age. They represent the pre-test probabilities at which GPs start their history taking and further work-up of patients with chest pain.

Table II. Probabilities for any IHD (acute or chronic), by age and gender (n = 1212).

IHD or ACS
Age, years	Male % (n/total)	Female % (n/total)
<40	4.1 (2/49)	0.0 (0/47)
40–49	4.1 (5/122)	1.6 (2/128)
50–59	13.3 (16/120)	5.6 (8/143)
60–69	23.7 (28/118)	11.9 (18/151)
≥70	32.8 (41/125)	28.2 (59/209)

IHD: ischaemic heart disease; ACS: acute coronary syndrome.

Table III stratifies the probabilities for any IHD in patients presenting with non-anginal and anginal chest pain, by age and gender. IHD probabilities were low for both clinical categories in the age group below 40 years. For the other age groups, there was a steady rise of IHD prevalence according to age, with male patients having higher prevalences than female patients. IHD prevalences were consistently higher in patients presenting with anginal chest pain in comparison to the corresponding age and gender categories of patients with non-anginal chest pain.

Table III. Probabilities for any IHD (acute or chronic) in patients presenting with non-anginal and anginal chest pain, by age and gender (n = 1212).

	Non-anginal chest pain		Anginal chest pain*
Age, years	Male % (n/total)	Female % (n/total)	Male % (n/total)	Female % (n/total)
<40	4.4 (2/45)	0.0 (0/42)	0.0 (0/4)	0.0 (0/5)
40–49	3.6 (4/111)	0.0 (0/110)	9.1 (1/11)	11.1 (2/18)
50–59	10.7 (11/103)	3.8 (5/131)	29.4 (5/17)	25.0 (3/12)
60–69	17.0 (16/94)	7.7 (9/117)	50.0 (12/24)	26.5 (9/34)
≥70	27.7 (31/112)	24.0 (42/175)	76.9 (10/13)	50.0 (17/34)

*Patients presenting with retrosternal and exertional chest pain. IHD: ischaemic heart disease.

Discussion

Summary of main findings

The prevalence of chest pain was 0.7% among primary care patients. Only a minority of these were caused by stable IHD (11.1%) or ACS (3.6%), which occurred more frequently in men than in women. Chest wall syndrome (46.6%) was the most common aetiology. There was a steady rise of IHD prevalence according to age, with male patients having higher prevalences than female patients.

Comparison with existing literature

The observed prevalence of chest pain in our study is similar to the results of a retrospective study in a group practice in Iceland. Svavarsdottir et al. found a prevalence of 0.68% (2). Two prospective studies both showed a slightly higher prevalence. Nilsson et al. reported a prevalence of 1.5%, although patients with known IHD were excluded (3). Verdon et al. reported a prevalence of 2.7%, including patients with chest pain as an already known or new symptom; a subanalysis of patients with chest pain as a new symptom and reason for consultation showed a prevalence of 1.0% (7).

With regard to the aetiology of chest pain, the 42.7% for chest wall syndrome quoted by Verdon et al. and 48.9% by Svavarsdottir et al. correspond largely with our findings (2,7). Other studies further differentiate between different types of musculoskeletal pain. When these aetiologies are added together, the results (29% and 33.5%) show a lower proportion of musculoskeletal pain than in our study (5,6). A possible explanation is different inclusion criteria. While our study also included patients with a history of trauma or chest pain not being the main reason for consultation, these patient groups were excluded in the study conducted by Klinkman et al., which might have led to a shift to other potentially more serious diagnoses (6). Our study found a proportion of 15% for overall IHD (11.1% stable angina pectoris; 3.6% ACS), as the diagnosis second in frequency. This is similar to the results from the Iceland study but higher than the Swiss, American, and Belgian studies (5–7). In particular, the rate for ACS differs by 3.6% from the rates of two of the above-quoted studies that both report 1.5% (6,7). Only Buntinx et al., with 4.8%, quote a similar cumulative number for serious cardiovascular disorders (5). Psychogenic disorders and gastro-oesophageal reflux disease showed similar distributions in other studies (5,6).

It is noteworthy that, even in patients with a known history of IHD, other aetiologies such as chest wall syndrome or respiratory disease still made up 50% of the acute episode of chest pain, while IHD and ACS constituted the other 50%.

Pre-work-up probabilities for IHD (see Table II) show the typical pattern according to age and gender. Further stratification for patients presenting with non-anginal versus anginal chest pain (see Table III) shows a similar pattern, with prevalences being higher in men than in women, and rising in each age group. However, in comparison to the results of a study undertaken by Diamond et al., we found lower pretest likelihoods for corresponding age/gender categories (8). This is not surprising, as our data are derived from a low-prevalence setting. This underscores the need to first look at the overall disease prevalence in a given clinical setting and then to estimate the resulting pre-work-up probabilities before applying further diagnostic tests (10). For our data, this would mean that any patient presenting with chest pain has a general probability of 14.7% for any IHD (see Table I, stable IHD and ACS added together). This probability would rise to 23.7% in a 65-year-old male patient (see Table II), and the same patient would have an IHD probability of 50.0% when presenting with anginal chest pain (see Table III).

Limitations

To our knowledge, this is the largest study investigating the epidemiology and aetiology of chest pain in primary care using an independent reference standard. Patients were consecutively recruited in a large number of urban and rural practices. The participating GPs’ demographic characteristics are similar to the population of GPs in the state of Hesse (data available upon request). Study procedures, including random audits, reduced the possibility of selection bias.

We did not interfere with the work-up provided by participating GPs. As a result of this, for some patients, only limited clinical data were available to the reference panel. This might have resulted, for example, in under-diagnosis of panic disorders or under-diagnosis of ACS in elderly patients. Some uncertainty therefore remains, although decisions on diagnoses were made according to current guidelines. However, we are confident that, during the follow-up of 6 months, relevant cardiac disease would have manifested itself (9). Given the low probability of IHD in most patients, standardized invasive investigations, e.g., coronary angiography, would not have been justifiable from an ethical point of view.

Conclusions

The 0.7% prevalence of chest pain in a primary care setting is rather low. Chest wall syndrome is the most frequent diagnosis. Compared to a hospital setting, cardiovascular disorders still make up a significant, but much lower proportion of the remaining underlying aetiology. We provide updated IHD pretest probabilities as a starting point for diagnostic processes.