Are dizziness/lightheadedness helpful symptoms for telephone triage of females and males calling the out-of-hours service for acute chest discomfort? A cross-sectional study

Authors: Anna S M Dobbe, Dorien L Zwart, Laurens van der Hem, D Carmen Erkelens, Loes T C M Wouters, Sander van Doorn, Hester M den Ruijter, Frans H Rutten

Abstract

Objectives

Symptoms of dizziness/lightheadedness are more often reported by females than males with an acute coronary syndrome (ACS). Therefore, we investigated if these symptoms are helpful for triage and diagnosis of ACS in females and males with acute chest discomfort.

Design

Cross-sectional study.

Setting

Nine out-of-hours service in primary care (OHS-PC) in the Netherlands participated, covering both rural and urban areas, serving a total population of 1.5 million people.

Participants

2195 patients were selected calling the OHS-PC with acute chest discomfort, 55.4% were females. Calls were selected based on International Classification of Primary Care codes and keywords, that is, chest pain, heart complaints, heart, myocardial infarction, heart attack, heart infarction or common abbreviations of these keywords. Recordings were excluded if (1) they were of poor quality, (2) the conversation was not a triage conversation, (3) patients were <18 years, (4) patients did not live in the vicinity of the OHS-PC, and (4) general practitioners did not want to provide follow-up information including the final diagnosis.

Primary and secondary outcome measures

The prevalence of dizziness/lightheadedness among patients calling the OHS-PC with acute chest discomfort and the relation between dizziness/lightheadedness and (1) urgency allocation and (2) a final diagnosis of ACS.

Results

Among 2195 patients who called the OHS-PC with acute chest discomfort, 251 (11.4%) had an ACS (8.3% females, 15.3% males). Females more often reported concurrent dizziness/lightheadedness than males (14.7% vs 10.9%, p=0.008). However, this was not observed in those with confirmed ACS (9.9% of the females vs 9.3% of the males, p=0.881). Reporting of dizziness/lightheadedness was not positively related to an ACS diagnosis in females (OR 0.61 (95% CI 0.31 to 1.21)) or in males (OR 0.82 (95% CI 0.45 to 1.47)). In contrast, both females and males with concurrent dizziness/lightheadedness more often received a high urgency allocation than those without these symptoms.

Conclusions

Dizziness/lightheadedness are common symptoms but are not positively related to an ACS diagnosis in both females and males with acute chest discomfort. It seems not worthwhile to give females and males with acute chest discomfort and concurrent dizziness/lightheadedness a higher urgency allocation than those without these symptoms.

Trial registration number

NTR7331.

Introduction

For decades, medical professionals aim to improve symptom recognition of acute coronary syndrome (ACS) in patients with acute chest discomfort.^1^ Despite advancements in the recognition of ACS, particularly in females, there remains a need for further improvement, as delayed or misdiagnosis of ACS is still prevalent.^2^ In countries where the general practitioner (GP) or the out-of-hours service in primary care (OHS-PC) is the first point of contact for healthcare, adequate symptom recognition is especially important, since there is limited availability of electrocardiography or blood measurements of troponin in these settings. In the Netherlands, around 80% of those seen at the emergency department (ED) for suspected ACS are referred by the GP or the OHS-PC, while the remaining 20% directly called the ambulance service (112) or were self-referrals.^3^

At the OHS-PC, triage nurses are supervised by GPs and make crucial decisions on the following trajectory based on telephone triage. It is challenging to optimally triage those with a potential ACS based on history taking only, while adequately balancing between safety and efficiency.^4 5^ On the one hand, timely recognition of ACS is crucial for early interventions to prevent life-threatening ventricular arrhythmias, reduce myocardial necrosis (‘time is muscle’), and consequently improve patient outcome and prognosis.^6^ On the other hand, referring everyone with acute chest discomfort, including those with a very low suspicion of ACS, would reduce efficiency and cause overflow at the ED. This leads to reduced patient safety as the most urgent cases could end up being managed with delay.^7^ Currently, around 6.5%–11% of those referred by the GP or OHS-PC for suspected ACS indeed have an ACS in the Netherlands.^8 9^

Chest discomfort is among the top five reasons for contacting the OHS-PC, accounting for 2.1% of all telephone consultations at the OHS-PC.^10^ Triage nurses at the OHS-PC use a semiautomatic digital decision support system for triage of patients based on their ‘main complaint’. The Netherlands Triage Standard (NTS) is such a support system, and it is based on the Manchester triage system. It consists of 56 ‘main complaints’ including ‘chest discomfort’.^11^ After the basic life support ABCDE check, the triage system generates an urgency level according to a scaling system with six possible urgency levels linked to a response time within which a caller should receive medical help (see figure 1).^12^

The last decade, much attention has been paid to differences in symptoms between females and males with established ACS.^13 14^ Previous studies reported that females with ACS more often report ‘less specific’ symptoms compared with males with ACS. Shortness of breath, but also dizziness are often mentioned as ‘typical’ symptoms for females with ACS.1419 A systematic review on sex differences in symptom presentation in patients with ACS showed that females more often report dizziness or lightheadedness compared with males with ACS (OR 1.28; 95% CI 1.15 to 1.44).^14^ These findings consequently led to the general viewpoint that dizziness or lightheadedness are important symptoms when considering ACS in females.^18^ However, these studies assessed symptom presentation in males and females with established ACS.142022 To be able to conclude that dizziness or lightheadedness are indeed important symptoms for considering ACS in females, it should be assessed in females suspected of ACS. Thus, comparing females who eventually show to have an ACS with females who have another cause for their symptoms.

Therefore, in this study, we investigated the prevalence of dizziness or lightheadedness as reported symptoms in patients calling the OHS-PC with acute chest discomfort. Subsequently, we studied whether these symptoms were useful in differentiating females with ACS from females without ACS, and similarly so for males. We also assessed the relation between dizziness or lightheadedness with urgency allocation in females and males calling the OHS-PC with acute chest discomfort.

Methods

Study design

This study is part of the Safety First project that had the overarching aim to analyse and enhance the process of telephone triage, specifically in patients suspected of ACS or suspected of transient ischaemic attack (TIA)/stroke.^23^ In this study, we focus on dizziness or lightheadedness as potential predictors for ACS in either sex.

Study population

Patients calling the OHS-PC with acute chest discomfort (pain, pressure, tightness or discomfort) between 2014 and 2016. We considered everybody with chest discomfort as ‘acute’ from the perspective of the patient; he or she perceived their symptoms to be urgent and thought it could not wait until seeing their GP during daytime office hours. Nine OHS-PC in the Netherlands participated, covering both rural and urban areas, serving a total population of 1.5 million people.

Calls were selected based on International Classification of Primary Care codes and keywords, that is, chest pain, heart complaints, heart, myocardial infarction, heart attack, heart infarction or common abbreviations of these keywords. Recordings were excluded if (1) they were of poor quality, (2) the conversation was not a triage conversation, (3) patients were <18 years, (4) patients did not live in the vicinity of the OHS-PC and (5) GPs did not want to provide follow-up information including the final diagnosis.

Data collection

In total, 3207 telephone triage recordings were relistened to collect information about the characteristics of patients and the conversation; on symptoms, medical history, urgency allocation and whether a supervising GP was involved in the triage. This was done while the researchers were blinded for the outcome (the final diagnosis). Information on final diagnoses was retrieved from the GP’s electronic medical file, and cardiologists’ and ED discharge letters. ACS diagnosis was based on the cardiologist’s opinion, information retrieved from (high sensitivity) troponin levels and results from electrocardiography. In 1012 cases (31.6%), the patient’s GP refused diagnosis retrieval from their files, mainly because of fear to violate the European General Data Protection Regulation. Thus, from 2195 patients with acute chest discomfort, we obtained follow-up data on the final diagnosis. We could compare the 2192 (except for 3) patients with and the 1012 patients without information on the final diagnosis (see table 1).^24^ There were no clinically meaningful differences in patient or call characteristics between the patients with information on the final diagnosis, and the patients in whom knowledge of the final diagnosis was missing. Other life-threatening events (LTEs) were defined as pulmonary embolism, thoracic aortic dissection, acute heart failure, TIA/stroke, aortic aneurysm, sepsis and severe chronic obstructive pulmonary disease (COPD) exacerbation.

Data analyses

Analyses were restricted to patients in whom the final diagnosis was retrieved. We defined dizziness or lightheadedness as present if patients expressed to have (1) lightheadedness or feeling faint, (2) a sense of motion or spinning (vertigo) or (3) a feeling of floating, wooziness or heavy-headedness after asking for it or mentioned them spontaneously during the call. Thus, the presence of these symptoms is based on the patient’s subjective perception. We defined dizziness or lightheadedness as absent in all other callers (combining those who denied presence after asking for it with those in whom either symptom was not mentioned during the call). Because lightheadedness and dizziness were used interchangeably, we combined both symptoms during data collection. Symptoms related to balance disorders and (near) fainting were not considered as dizziness or lightheadedness. In this study, symptoms of nausea, vomiting, sweating and pallor were classified as autonomic nervous system-related symptoms, distinct from dizziness or lightheadedness. Throughout this manuscript, we will indicate symptoms of dizziness or lightheadedness as follows ‘dizziness/lightheadedness’. Urgency allocation was dichotomised into ‘high urgency’ (U1 and U2) and ‘low urgency’ (U3, U4 and U5).

Patient and call characteristics were compared between those with and without dizziness/lightheadedness. The Pearson’s χ^2^ test or Fisher’s exact test (in case of groups with less than 10 people) was used to compare categorical variables and the independent sample T-test was used to compare continuous variables. For analysis of the relation between dizziness/lightheadedness and (1) urgency allocation, (2) final diagnosis of ACS and (3) final diagnosis of ACS or other LTEs OR were calculated. Adjusted ORs (AORs) were calculated after correction for age. A p<0.05 was considered as statistically significant. Data analyses were performed with IBM SPSS V.28.0. and R Studio V.4.0.5.

Patient and public involvement

No patients were involved in setting the research question or the outcome measures, or in developing plans for design; however, they were involved in the implementation of the study. In addition, the results will be shared and discussed with the national patient community of cardiovascular diseases (‘Harteraad’).

Results

Population characteristics and the prevalence of dizziness/lightheadedness

The study population consisted of 2195 patients who called the OHS-PC with symptoms suggestive of an ACS (figure 2). The mean age was 59.1 (SD 19.5) years, and 55.4% (1215/2195) were females. In total, 251 (11.4%) of these patients had an ACS; 8.3% (101/1215) of the females and 15.3% (150/980) of the males. Concurrent symptoms of dizziness/lightheadedness were reported in 286 (13.0%) patients. Females more often reported these symptoms compared with males; 14.7% (179/1215) vs 10.9% (107/980) (p=0.008), respectively. Patients reporting dizziness/lightheadedness were younger (mean age 55.7 vs 59.6 years; p=0.002), had a longer median call duration (7:13 vs 48 min:s, p=0.013), more often had a history of cardiac arrhythmia (14.3% vs 10.0%, p=0.036) and less often used cardiovascular medication (32.9% vs 39.9%, p=0.014). Furthermore, patients with concurrent dizziness/lightheadedness more often reported nausea/vomiting (35.0% vs 20.4%, p=0.001), sweating (30.8% vs 24.7%, p=0.020), palpitations (25.9% vs 11.1%, p=0.026) and pressing/tightening chest pain (49.3% vs 37.0%, p<0.001) at presentation compared with those without these symptoms (see also table 1).

Dizziness/lightheadedness and final diagnosis of ACS

The 286 patients calling with acute chest discomfort with concurrent dizziness/lightheadedness tended to have a lower risk of ACS compared with those without these symptoms; 8.4% vs 11.9%, OR 0.68 (95% CI 0.44 to 1.05, p=0.085). For females, this was 5.6% vs 8.8%, OR 0.61 (95% CI 0.31 to 1.21, p=0.156), and for males, 13.1% vs 15.6%, OR 0.82 (95% CI 0.45 to 1.47, p=0.500) (see table 2). After correction for age, the AOR for females was 0.77 (95% CI 0.39 to 1.54, p=0.461) and for males 0.83 (95% CI 0.46 to 1.51, p=0.542). There was no significant difference between females and males with established ACS, they both equally often reported dizziness/lightheadedness compared with those without these concurrent symptoms; 9.9% (10/101) vs 9.3% (14/150), p=0.881.

Dizziness/lightheadedness and final diagnosis of ACS or other LTEs

Patients with concurrent symptoms of dizziness/lightheadedness tended to have a lower risk of an ACS or other LTEs compared with those without these symptoms (11.2% vs 14.9%, OR 0.72 (95% CI 0.49 to 1.06, p=0.095). In females, this was 7.8% vs 11.1% and an OR of 0.68 (95% CI 0.38 to 1.21). In males, this was 16.8% vs 19.5% with an OR of 0.84 (95% CI 0.49 to 1.43) (see table 3). After correction for age, the AOR was 0.85 (95% CI 0.47 to 1.55, p=0.602) in females and 0.85 (95% CI 0.49 to 1.47, p=0.559) in males. There was no significant difference between females and males with established ACS or other LTE reporting dizziness/lightheadedness compared with those without these concurrent symptoms 10.9% (14/129) vs 9.6% (18/188), p=0.711.

Dizziness/lightheadedness and urgency allocation

The 286 patients with dizziness/lightheadedness significantly more often received a high urgency allocation than those without these symptoms; 75.5% vs 67.3%, OR 1.50 (95% CI 1.13 to 2.00, p=0.005). Following stratified analyses, this was not observed in females (OR 1.23: 95% CI 0.86 to 1.75, p=0.252), but only in males (OR 2.18: 95% CI 1.31 to 3.61, p=0.002) (see table 4). However, after correction for age, the AOR for a high urgency allocation was similar for females and males (female, 1.59: 95% CI 1.10 to 2.31, p=0.015 and males, 2.38: 95% CI 1.41 to 4.03, p=0.001), suggesting that age was an important confounder.

Discussion

Our findings show that dizziness and lightheadedness are not important symptoms for assessing ACS risk in females nor in males with acute chest discomfort calling the OHS-PC. Females with acute chest discomfort significantly more often reported concurrent symptoms of dizziness/lightheadedness compared with males (14.7% vs 10.9%, p=0.008). However, no difference was found between females and males with an established ACS diagnosis, as both equally often reported dizziness/lightheadedness (9.9% vs 9.3%, p=0.881). Additionally, in both females and males reporting dizziness/lightheadedness along with acute chest discomfort, no positive association was found with a final diagnosis of ACS or other LTE compared with those without these symptoms. Therefore, dizziness and lightheadedness do not seem to be discriminative symptoms for an ACS diagnosis in patients with acute chest discomfort calling the OHS-PC. Moreover, these symptoms do not appear to be ‘typical’ for females with ACS.

Comparison with literature

In our study, females with an ACS equally often reported dizziness/lightheadedness as males with an ACS (9.9% vs 9.3%; p=0.881). This is in contrast with a recent meta-analysis reporting a pooled OR of 1.28 (95% CI 1.15 to 1.44) for females compared with males with ACS reporting symptoms of dizziness/lightheadedness.^14^ Our findings also do support what the most recent guideline of the American Heart Association for the evaluation and diagnosis of chest pain reported on lightheadedness; that it is a more common associated symptom with chest pain among females compared with males. In this guideline, however, chest pain was not further subdivided into acute and non-acute or stable chest pain.^18^

This study was performed to determine whether dizziness and lightheadedness are helpful symptoms for differentiating females with ACS from females without ACS and likewise for males. Contrary to the expectations based on studies performed in patients with established ACS, our study showed a trend that reporting dizziness/lightheadedness was predictive of another diagnosis than ACS in females with acute chest discomfort (OR of ACS 0.61 (95% CI 0.31 to 1.21). A similar trend was seen in males; OR of ACS 0.82 (95% CI 0.45 to 1.47). Importantly, these findings are consistent with the results of a study performed by Devon et al in a study at the ED; females who reported lightheadedness along with chest pain were less likely to be diagnosed with an ACS compared with females without this concurrent symptom (40% vs 51%, p=0.118). The same was seen in males (34% vs 42% p=0.088).^25^

Thus, there are contrasting findings in the literature on the diagnostic value of dizziness/lightheadedness in patients suspected of or with established ACS, with most studies reporting that females with ACS more often have these symptoms compared with males with ACS.^14 21 22^ Differences in how symptoms of dizziness/lightheadedness are defined could partly explain this, such as whether (near) fainting is included in the definition. It are highly subjective symptoms, with patients using different wording, or having other cognitive connections with dizziness/lightheadedness than physicians and triage nurses. Consequently, the method of data collection, whether through questionnaires or routine care data, may also influence findings and contribute to variations in prevalence and incidence rates. In our study, the prevalence of concurrent dizziness/lightheadedness in patients with established ACS was relatively low, in 9.9% of the females and in 9.3% of the males. Two previous studies, respectively, reported lightheadedness in 40% and 36% of the females with ACS and in 34% and 32% of the males with ACS.^25 26^

Strengths and limitations

To the best of our knowledge, this is the first study to assess the diagnostic value of dizziness/lightheadedness for ACS in females and males with acute chest discomfort. An important strength is that we assessed this in the clinically relevant domain; namely, patients suspected of ACS. Additionally, the data were obtained by listening to the original telephone triage conversation with researchers blinded to the clinical outcome, thus without hindsight bias. Finally, our findings are generalisable to the whole of the Netherlands because we retrieved our results from nine OHS-PC locations covering both urban and rural areas of the Netherlands. Our findings may also be applicable to patients with acute chest discomfort who contact the GP during office hours in the Netherlands, as telephone triage is conducted there using the same decision support tool (NTS). Moreover, our results are likely generalisable to other countries with a similar primary healthcare system, for example, Scandinavian countries, Germany and the UK.^27^

A limitation of our study is that there were missing data on some patient characteristics, a common finding in studies using routine care data. A further limitation is that inter-researcher variation in coding characteristics may be present because multiple researchers assessed the telephone triage recordings. Because there are likely differences in interpretation by researchers of the wording patients use for dizziness/lightheadedness non-differential errors may occur. The direction of the error is, however, uncertain. It could either strengthen or weaken the observed association.^28^ Consequently, there is uncertainty in both our findings and the reported prevalence rate, a limitation that is also present in other studies reporting on dizziness/lightheadedness. Lastly, the ACS diagnoses and other cardiac diagnoses were based on the treating cardiologist’s opinion. Other diagnoses were also based on the treating physician’s opinion. This may have introduced some differential bias in the diagnoses, potentially reducing the certainty of our findings. The use of an expert panel would have been preferable to provide greater diagnostic consistency but was unfeasible to organise.

Implications for research and/or practice

Our findings suggest that dizziness and lightheadedness are not important symptoms when considering ACS in females, at least not in a positive sense. Females presenting with acute chest discomfort and concurrent dizziness/lightheadedness tended to have a lower risk of ACS compared with females without dizziness/lightheadedness. This was similarly so in males. Contrary to current literature suggesting these symptoms are related to an ACS diagnosis in females, the presence of dizziness/lightheadedness should not warrant a higher urgency allocation during OHS-PC triage than based on other symptoms such as transpiration and oppressive chest pain.

Conclusion

Dizziness and lightheadedness are common symptoms but are not positively related to an ACS diagnosis in both females and males with acute chest discomfort. It seems not worthwhile to give females and males with acute chest discomfort and concurrent dizziness/lightheadedness a higher urgency allocation than those without these symptoms.