580910 research-article2015
CNU0010.1177/1474515115580910European Journal of Cardiovascular NursingO’Keefe-McCarthy et al.
EUROPEAN SOCIETY OF CARDIOLOGY ®
Original Article
Prodromal symptoms associated with acute coronary syndrome acute symptom presentation
European Journal of Cardiovascular Nursing 1–8 © The European Society of Cardiology 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1474515115580910 cnu.sagepub.com
Sheila O’Keefe-McCarthy1, Michael H McGillion2, J Charles Victor3, Jeremy Jones1 and Judith McFetridge-Durdle4
Abstract Objective:Prodromal symptoms (PS), indicative of myocardial ischemia, are frequently unrecognized by individuals prior to an acute coronary syndrome (ACS). ACSs are the leading cause of death worldwide. This study describes (1) the prevalence and association of PS with patients’ baseline ACS-related acute symptoms of pain intensity and state anxiety and (2) the relationship of PS to co-morbidity. Methods:An exploratory sub-analysis was performed. Cross sectional data identified prodromal predictors of ACS pain intensity (numeric rating scale 0–10 (NRS)) and state anxiety (Speilberger state-trait anxiety personality inventory (STAI)). ACS patients (n=121) admitted to a community rural emergency department completed the prodromal symptom screening scale (PS-SS) and reported baseline cardiac pain intensity, state, and trait anxiety. Results:Increased ACS pain intensity was associated with PS. Median pain scores were higher by two points for those with prodromal headache, p=0.006, and anxiety, p=0.017, and one point higher for those with sleep disturbances, p=0.012. PS were not associated with state or trait anxiety. Hypertensive individuals were 7.5 times more likely to experience prodromal fatigue prior to their ACS event. Conclusion:Results extend current knowledge of the predictive value that prodromal headache, sleep disturbance and anxiety may have on individuals’ acute symptom presentation. A prospective, prognostic study is required in order to determine whether PS are predictive of adverse cardiac events and if PS are a stronger predictor of ACS acute symptom presentation, compared with typical ACS-related co-morbidities. Keywords Prodromal symptoms, acute coronary syndrome, pain, anxiety, sleep disturbance, headache, sex differences Date received 31 July 2014; revised 17 March 2015; accepted 18 March 2015
Introduction The value of immediate recognition of myocardial ischemia-related symptoms has been well documented. Multiple studies provide undeniable evidence that early symptom recognition of coronary heart disease (CHD) reduces morbidity and mortality.1–5 Despite advances in treatment for acute coronary syndromes (ACS) and increased public awareness of its benefits, hundreds of thousands of people with unstable angina (UA) or acute myocardial infarction (AMI) fail to recognize the warning signs—prodromal symptoms (PS)—prior to their ischemic cardiac-related event.6–8 PS are specific and non-specific, transitory sensations, indicative of myocardial ischemia, that should serve as a warning sign of an impending ACS.6
Incidence of PS ranges from 49%–92%.6 PS may be experienced days,6,9 weeks,8,10 or months11 prior to an index cardiac event. Early recognition of an ACS onset is essential for rapid amelioration of chest pain with institution of immediate lifesaving reperfusion therapies.12 1Ross
Memorial Hospital, Canada University, Canada 3University of Toronto, Canada 4Florida State University, USA 2McMaster
Corresponding author: Sheila O’Keefe-McCarthy, Ross Memorial Hospital, 10 Angeline St North, Lindsay, Ontario, K9V 4E8, Canada. Email:
[email protected] Downloaded from cnu.sagepub.com at UNIV OF WISCONSIN OSHKOSH on November 13, 2015
2
European Journal of Cardiovascular Nursing
International data on delay times indicate the time from symptom recognition to the decision to seek medical care as the most problematic phase.13 In the United States, median delay time from experiencing symptoms to hospital arrival ranges from 1.5 to 6 hours,13–15 Australian data reports 6.4 hours,16 with some European countries reporting longer delay time compared to United States data.17 Poor recognition of PS one may experience transiently before an AMI may explain the reason for misinterpretation and resultant treatment delay. Canto et al.12 identified three potential reasons for symptom confusion and differences in prodromal and/or acute ACS symptom presentation.12 First, sex-related biological anatomical and physiological factors such as sex hormones, autonomic nervous system, or micro vascular disease.18 Second, the socio-demographic differences with varying degree of risk factor profile (age, race, education) or co-morbid conditions19,20 and last, sex-related cultural factors may also influence prodromal symptom presentation and when a person may decide to seek help for cardiac-related symptoms.21–23
Prodromal symptoms (PS) Decades of research has reported PS in individuals with UA and myocardial infarction.6–12,24 The most prevalent PS reported by patients in 28 studies were chest pain (49%), fatigue, (37%) and shortness of breath (SOB) (25%).24 Evidence indicates that PS may vary in frequency. For example, McSweeney and colleagues reported that women (n=515) experienced unusual prodromal fatigue, sleep disturbance, and SOB one month prior to AMI.8 Prodromal chest pain and increasing anxiety were reported at four weeks, with gastric symptoms occurring greater than six weeks prior to an adverse cardiac event.10 Others have described prodromal chest pain/discomfort from 48 hours to days prior to an ACS6,9 In studies that examined ACS prodromata in men and women, PS were reported more frequently by women as compared to men.7–12 Men presented with more prodromal chest pain, whereas women experienced greater fatigue, anxiety, and headrelated PS (headaches and dizziness) compared to their male counter parts.8–11 Few studies have examined the relationship between PS and acute symptoms experienced during an ACS episode. In a multicentre, cross-sectional evaluation of 533 Norwegian patients with AMI, half of the sample reported PS.11 These PS were associated with patients’ acute symptoms experienced during their AMI.11 A new systematic review reported across seven studies of 6716 individuals with confirmed CHD, that PS experienced prior to an ACS index hospitalization were predictive of patients’ ACS-related acute symptom presentations and associated cardiac events (UA or AMI).6 Identifying the association between co-morbidity and PS has not been well documented in the prodromal literature. One study examined the relationship of co-morbidities and PS and
found that those with hypertension were one and a half times more likely to experience prodromal SOB (OR=1.55; 95% CI=1.03–2.31) and unusual fatigue (OR=1.60; 95%CI=1.06-2.43) [11]. Those with diabetes were almost three times at risk for prodromal SOB than those without diabetes (OR=2.55; 95%CI=1.32-4.93), but PS were not associated with individuals who had documented hyperlipidemia.11 Contemporary accounts of individual’s unique, prehospital PS are limited and require further examination. Minimal evidence is available that examines the predictive value of PS on ACS acute symptom presentation or identifies the PS which may be associated with common comorbidities within the context of an ACS event.
Aims The aim of this study was to describe the prevalence, frequency, and severity of PS in a sample of ACS patients. In particular, we examined (1) the relationship of PS on baseline ACS acute symptoms of pain intensity, and state anxiety and (2) the association of PS with co-morbidity.
Methods The data for this article were collected as part of another study that examined ACS pain and anxiety and pain management practices that has been described in detail previously.25 As a result, the findings reported here contain only those data relevant to the present prodromal symptoms sub-analysis.
Design/Setting This descriptive, cross sectional study examined patients diagnosed with ACS in one emergency department (ED) in southeastern Ontario, Canada. This study was conducted in a rural, small town, farming community of predominately older, retired Caucasian adults.
Participant recruitment This study was conducted in two phases: (1) ACS patient recruitment and data collection and (2) a prospective patient chart audit. Patients were eligible if they were (1) diagnosed with ACS (either UA or non ST-elevation myocardial infarction [NSTEMI]), (2) reported prodromal symptoms, (3) reported chest pain for more than 20 minutes duration, and/or described comparable anginal equivalent symptoms suggestive of ACS-related pain (i.e. SOB, chest tightness, pressure, syncope, diaphoresis, and nausea/vomiting), (4) presented with a positive electrocardiogram (ECG) change (t-wave changes, and/or ST depression or elevation) in one or more ECG leads, and (5) were able to speak, read, and understand English. Patients excluded were those who: (1) were deemed emergent (had a Canadian Triage Acuity
Downloaded from cnu.sagepub.com at UNIV OF WISCONSIN OSHKOSH on November 13, 2015
3
O’Keefe-McCarthy et al. Scale (CTAS) score of II, indicating emergent status, and required assessment within 10–15 minutes of ED triage26), (2) had recent coronary bypass grafting or heart/valve replacement (within the last two years—which may confound the acute pain symptom presentation, should persistent post-operative pain develop post median sternotomy). and (3) were unable to give a verbal and written informed consent. This study was approved by the Research Ethics Board at the University of Toronto, (protocol reference number 25999) and the community hospital’s Ethics Review Committee. The research protocol was in line with the Declaration of Helsinki (1975) and revised Hong Kong declaration of 1989. Patients were recruited by champions (nurses in charge without a patient assignment). Recruitment champions identified potential ACS participants after initial triage and stabilization and notified the primary investigator (PI). Eligibility was confirmed by the PI and patients were approached after consenting to speak about the study, being first provided with an oral explanation and then with a written form. Baseline, demographic data, the PS screening scale, and patients’ cardiac pain intensity, state, and trait anxiety data were collected via patient interview by the PI. Following initial baseline data collection, a prospective review of patients’ healthcare records was used to complete and verify accurate documentation of patients’ socio-demographic data, previous medical history, and pertinent clinical variables.
Measures Prodromal questionnaire A prodromal symptom screening scale (PS-SS) was developed and was based on current prodromal literature of the most typical symptoms patients may experience prior to an ACS event.8–11,27–31 Patients completed the PS-SS at baseline. The scale indentifies nine cardiac-related prodromal symptoms (unusually located aches/and/or pain, unusual fatigue, sleep disturbances, chest pain, anxiety, headaches, dizziness, and SOB). To thoroughly describe patients’ unique prodromal symptoms, the category of other symptom [item 9] was offered to patients, in order to capture all patients’ subjective prodromal symptoms. Each symptom was recorded as a binary outcome (Yes or No and scored 1 or 0, respectively), followed by questions pertaining to symptom intensity (mild, moderate, or severe; scored 1–3) and frequency (daily, few times week, once a week, two to three times per month, once a month, and less than monthly; scored 6, 5, 4, 3, 2, and 1, respectively). Each prodromal symptom item was scored and summed. The global scale score is achieved by summing all item sums. Total possible score is 90. Pilot testing of the PS-SS was evaluated for its content validity index by a panel of nine nurses certified in cardiac critical care. Total scale average item CVI was 0.85, demonstrating
strong content validity. Two internal consistency estimates were computed along with individual item analysis for the PS-SS: one for the overall nine-item scale and one for a seven-item revised scale, removing the two lowest correlated items (sleep disturbance r=206 and anxiety r=217) to determine if the internal consistency reliability would be significantly different than the overall calculated value. The values for coefficient alpha (Chronbach’s alpha ) were 0.613 and 0.611, respectively, each indicating acceptable reliability and therefore the full nine-item scale was retained.32
Pain intensity ACS pain intensity was measured once at baseline with the numeric pain rating scale (NRS). The 11 point (0–10) intensity scale allows patients to select the number that best represents how much pain intensity they feel; 0 being no pain, 1–3 mild pain, 4–6 moderate pain, and 7–10 severe pain. The NRS has well established reliability, construct validity, and concurrent validity in cardiovascular and ED populations.33–36
Anxiety State and trait anxiety were measured using the 20 item state and trait subscales of the state–trait anxiety personality inventory (STAI) developed by Spielberger et al. for adults.37 State and trait anxiety were collected at baseline. The state anxiety subscale evaluates how respondents feel in the moment. It measures the subjective feelings of anxiety that may be typically elevated in stressful situations and can fluctuate and vary in intensity over time. Alternately, the trait anxiety subscale measures more stable individual personality differences in a persons’ tendency to perceive situations as threatening and to experience ‘states’ of anxiety. The tool has been shown to have well established psychometric properties.37
Data analyses All data were analyzed using SAS 9.1 software.38 The sample size is described elsewhere.24 Briefly, a sample of 150 ACS patients was projected to detect correlations of 0.3 (medium size) with 90% power and a two sided alpha [(.05) with a 25% lost to follow up.25 Baseline socio-demographic and clinical characteristics were expressed as means and standard deviations for normally distributed continuous factors; medians and interquartile ranges were reported for non-normally distributed continuous factors. Frequencies and proportions were calculated for categorical factors. Prodromal sub-analysis. The association between baseline ACS pain intensity and presence or absence of prodromal symptoms was evaluated using Wilcoxon Rank–Sum nonparametric test because of a positively skewed distribution
Downloaded from cnu.sagepub.com at UNIV OF WISCONSIN OSHKOSH on November 13, 2015
4
European Journal of Cardiovascular Nursing
in pain intensity scores.39 State and trait anxiety were examined by prodromal symptoms using Student’s t-tests.39 Unadjusted comparisons of prodromal symptoms between men and women were evaluated with use of chi-square tests.40 The associations between co-morbidities and prodromal symptoms were conducted using chi-square tests for association and multivariable logistic regression controlling for all confounders.41 P values