Article, Cardiology

The association of chest pain duration and other historical features with major adverse cardiac events

pain duration a”>Original Contribution

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage:

The association of chest pain duration and other historical features with Major adverse cardiac events

Tony Zitek, MD a,b,c,?, Elizabeth Chen, MD d,e, Armando Gonzalez-Ibarra c, Jessica Wire, MD a

a Kendall Regional Medical Center, Miami, FL, United States

b Nova Southeastern University, Kiran C. Patel College of Allopathic Medicine, United States

c University Medical Center of Southern Nevada, Las Vegas, Nevada, United States

d University of Nevada, School of Medicine, Las Vegas, United States

e Mike O’Callaghan Federal Medical Center at Nellis Air Force Base, United States

a r t i c l e i n f o

Article history:

Received 7 October 2019

Accepted 12 November 2019

Keywords:

Chest pain history Duration of pain Myocardial infarction

a b s t r a c t

Objective: We sought to determine if the duration of pain or other features of the history predict Major adverse cardiac events \(MACE\) in patients with chest pain in the emergency department (ED).

Methods: This was a prospective cohort study of patients presenting to a single ED with chest pain. Consenting patients filled out a survey about their symptoms. After 6 weeks, we assessed patients for MACE via chart review and direct contact. We used this data to calculate the Likelihood ratios of a number of historical features for acute myocardial infarction (MI) (primary endpoint) and MACE within 6 weeks (secondary endpoint). We planned a priori to analyze patients who reported chest pain for <=1 min or continuously for >=24 h.

Results: We enrolled 1002 patients, and 83.6% had successful 6-week follow up. Regarding chest pain lasting for

<=1 min, the positive LR was 0.95 (95% CI 0.24 to 3.80) for acute MI and 0.67 (95% CI 0.17 to 2.72) for MACE within 6 weeks. The positive LRs of continuous pain lasting >=24 h for acute MI and MACE within 6 weeks were 0.15 (95% CI 0.04 to 0.58) and 0.36 (95% CI 0.18 to 0.74), respectively. Amongst other historical features assessed, radiation

to the right arm was the strongest positive predictor of acute MI.

Conclusion: Patients with continuous chest pain for >=24 h are unlikely to have an acute MI. Chest pain lasting

<=1 min does not exclude acute MI.

(C) 2019

  1. Introduction

Although the history is the most basic part of the assessment for pa- tients presenting to the emergency department (ED) with chest pain, the data regarding which clinical historical features are most useful to rule in or rule out myocardial infarction (MI) are sparse and old ? gen-

erally, at least 15 years old [1-6]. Those studies were done in an era

when the diagnosis of MI was often made by creatine kinase elevation, and cardiac catheterizations were not so frequently done [7,8]. It is not certain if the conclusions from those studies still apply today. Moreover, some components of the chest pain history have not been studied at all. For example, while it has been suggested that chest pain episodes that last only seconds or continuously for hours or days are not from cardiac ischemia [6], this is not based on prospective data.

* Corresponding author at: 305-480-6602, 11750 SW 40th St, Miami, FL 33175, United States.

E-mail address: [email protected] (T. Zitek).

There have been some recent technological advancements including high sensitivity troponins [9,10] coronary computed tomographic angi- ography [11], and computed tomography myocardial perfusion [12] that have the potential to improve our diagnostic evaluation of patients presenting to the ED with chest pain. Additionally, clinical tools such as the HEART score [13] are improving our ability to risk stratify these pa- tients and help us identify which ones might benefit from additional di- agnostic testing. However, in order to properly use the new advanced diagnostic technologies or even just to calculate a HEART score, an ED provider must assess a patient’s risk based upon the history. As discussed above, the data to do so are limited. More data are needed re- garding the features of a chest pain history that substantially increase or decrease the chance that a patient is having an MI.

Thus, we performed a prospective cohort study primarily to deter- mine if the duration of a patient’s chest pain has a positive or negative association with the diagnosis of acute MI. In particular, we hypothe- sized that chest pain episodes lasting <=1 min or continuously for >=24 h would be strong negative predictors of acute MI. Secondarily, we sought

to determine if a number of other clinical historical factors can help pre- dict whether or not a patient was having an MI or will have major ad- verse cardiac events (MACE) within 6 weeks.

https://doi.org/10.1016/j.ajem.2019.11.020

0735-6757/(C) 2019

  1. Methods
    1. Study design and setting

This was a single-center, prospective, observational cohort study performed on adult patients who presented to the ED of a county, aca- demic, tertiary care facility in Las Vegas, Nevada with a chief complaint of chest pain. At this facility the ED is split into medical and trauma sec- tions. Patients who present to the ED as a result of a traumatic injury are funneled to the trauma ED. Only patients in the medical ED were enrolled. The medical ED has about 80,000 visits per year. This study was approved by the hospital’s institutional review board.

Selection of participants

The study enrolled a convenience sample of patients during the hours that trained research assistants were available (10 am to 10 pm). Re- search assistants identified patients in the adult medical ED who pre- sented with a chief complaint of chest pain/discomfort by looking at the ED tracking board. Written informed consent was obtained from en- rolled patients. Patients could be included if they were at least 18 years old and presented to the ED with chest pain or chest discomfort of any sort. Patients were excluded if they had an ST-elevation MI, did not speak English, or if the research assistant judged that the patient’s mental status was such that they could not answer the questions on the survey.

Measurements

Enrolled patients were instructed to complete a nine question sur- vey about their chest pain. The survey asked about the duration, quality, severity (scale 0-10), radiation, and associated symptoms of the chest pain. It also specifically asked if the patient felt they were having a heart attack and if the chest pain worsened with strenuous activities. The survey is included as an Appendix.

A research assistant stayed with the patient during survey comple- tion to make sure they understood and answered all questions. The research assistant was blinded from the results of the Diagnostic tests. The patient may not have been blinded from the results of their diag- nostic tests as their physician may or may not have told them the results at the time the survey was completed.

Six weeks after enrollment, a research assistant attempted to di- rectly contact by phone each patient to determine if they had a major adverse cardiac event after leaving the hospital. Additionally, a chart review was done by a single research assistant who was trained by the principal investigator of the study. The patient’s age, gender, race, ethnicity, highest troponin level, disposition, cardiac procedures, and di- agnoses were abstracted.

Outcomes

The primary endpoint was the diagnosis of acute MI as determined by the ED or final hospital discharge diagnosis (whichever came last) from the hospital visit during which they were enrolled in the study. Since it is possible that some patients having an MI may have been incorrectly diagnosed and discharged from the ED, we also assessed a secondary composite endpoint of acute MI, percutaneous coronary in- tervention (PCI), coronary artery bypass graft (CABG) surgery and death, within 6 weeks after presentation, together called MACE. This was determined by a review of hospital records (when available) or through direct contact with patients.

We hypothesized that a duration of chest pain for <=1 min or a contin- uous episode of chest pain for >=24 h would be strong negative predictors

of acute MI. Therefore, we planned a priori to analyze patients with chest pain duration in those specific time intervals. For our primary out- come, we calculated the positive likelihood ratio of chest pain duration for <=1 min or >=24 h continuously with regards to its ability to

predict an acute MI. We also sought to determine if chest pain duration can help predict MACE within 6 weeks. Secondarily, we sought to deter- mine if other historical factors (including severity of pain, radiation of pain, quality of pain, and associated symptoms) can help predict acute MI or MACE within 6 weeks. Finally, we performed a multivariate analysis using the historical features that were found to be significant predictors of MI or MACE on the above analyses to obtain adjusted odds ratios.

Analysis

Data were tabulated in Microsoft Excel (version 16, Microsoft Corpo- ration, Redmond, Washington), and analyzed in a statistical program called “R”. For all binary variables, LRs for acute MI and MACE within 6 weeks were calculated with 95% confidence intervals using standard technique. While LRs provide the most Clinically useful information, we further assessed the robustness of our results by obtaining adjusted odds ratios for those historical features that appeared to be significant predictors based on their LRs (95% confidence interval didn’t cross 1.0). We did so by constructing a multivariate logistic regression model using all the significant predictors of acute MI along with the available known risk factors for fatal coronary artery disease (age, gen- der, and race) [14].

For Pain severity, univariate and multivariate logistic regression analyses were done to assess for a relationship between pain severity and the diagnosis of acute MI or MACE within 6 weeks. As prior data on pain scores have disclosed that women [15,16], African Americans [15], and those of younger age [15,16] tend to report higher pain scores, gender, race, and age were used as control variables.

For all multivariate analyses, we performed some model diagnostics. In particular, we checked the assumption of linearity in the logit for age and pain score using the Box-Tidwell test. We tested the regression models for interactions, and we used the Hosmer-Lemeshow test to assess goodness of fit.

  1. Results

Between April 25, 2017 and January 9, 2019, 1002 unique patients

were enrolled. Of the 1002, 464 (46.3%) were female, and 538 (53.7%) were male. The median age was 54 years (IQR: 44-62) with a range of 18 to 95 years. Regarding race, 525 patients were white (52.4%), 345 were black (34.4%), 48 were Asian (4.8%), and 84 did not select any of those three. Of the 1002 patients, 22.9% reported their ethnicity as His- panic or Latino.

All patients had an electrocardiogram done, and almost all patients (97.8%) had a troponin ordered. In total, 685 patients (68.4%) were ad- mitted to the hospital while 272 (27.1%) were discharged from the ED, 42 (4.2%) left against medical advice from the ED, and three (0.3%) were transferred from the ED to another facility.

As shown in Fig. 1., 6-week follow up was successful in 83.6% of pa- tients. In total, 64 patients had MACE within 6 weeks (7.6% of those with successful follow up), including 44 who were diagnosed with an acute MI during their initial hospital stay and four who had an MI after leaving the hospital. We found 37 patients who had PCI, 12 patients who had CABG surgery, and 6 who died.

In addition to MI, the discharge diagnoses of enrolled patients included a number of dangerous diseases. In particular, 24 were diag- nosed with pneumonia (2.4%), 15 with pulmonary embolism (1.5%), 1

with aortic dissection (0.1%), 1 with a pericardial effusion (0.1%), 1 with pericarditis (0.1%), and 3 with pneumothorax (0.3%). That being said, by far the most common diagnosis for the patients enrolled in this trial was nonspecific chest pain with 598 of 1002 enrolled patients (59.7%) being given this diagnosis, while the second most common diagnosis was “non-cardiac chest pain” with 61 patients (6.1%). Table 1 provides more details with regards to the discharge diagnosis of the enrolled patients.

1069 patients approached who were English speaking adults with chief complaint listed as chest pain/discomfort.

1002 patients enrolled and completed the survey.

67 patients excluded: 2 for STEMI,

22 refused, 19 deemed unable to complete survey, 24 denied chest pain as chief complaint.

After six weeks, 838 follow up successful.

After six weeks, 164 lost to follow up.

64 patients had MACE: 44 acute MIs, 4 MIs after discharge, 37 PCI, 12 CABG, and 6 deaths.

Fig. 1. Flow diagram of the study cohort.

Regarding the primary outcome, 47 patients presented to the ED for episodes of chest pain that lasted <=1 min, and two of them (4.3%) were diagnosed with an acute MI. One of these patients was a 63 year-old female who reported chest pain for only ten seconds. Although she had chronic troponemia from end-stage renal disease, she had an

acute increase in her troponin and was diagnosed with a type 1 non- ST elevation MI. The other patient in this group who had an acute MI was an 82 year-old male who reported chest pain for 1 min associated with an episode of supraventricular tachycardia. He had an elevated tro- ponin, and was taken for a cardiac catheterization that showed multivessel coronary artery disease. He then had a CABG surgery.

Compared to the 4.3% MI rate for patients who reported <=1 min of

chest pain, 42/941 (4.5%) patients who had chest pain for N1 min had an acute MI; 14 patients did not specify how long their chest pain epi- sode lasted. Therefore, the positive LR of chest pain lasting <=1 min for

Table 1 The frequency of various, notable discharge diagnosis among the 1002 enrolled patients who presented to the ED with chest pain.

Discharge Diagnosis Number (%)

acute MI was 0.95 (95% CI 0.24 to 3.80). Among the patients who had successful follow up at 6 weeks, 38 presented for chest pain episodes lasting <=1 min, and 2 of those (5.3%) had MACE within 6 weeks. Com- paratively, the rate of MACE at 6 weeks for those who had chest pain

episodes lasting N1 min was 7.8% (61/786). Thus, the positive LR of chest pain lasting <=1 min for MACE within 6 weeks was 0.67 (95% CI 0.17 to 2.72).

In total, 292 patients presented to the ED for a chest pain episode that had lasted for >=24 h continuously. Of those, 2 (0.7%) were diag- nosed with an acute MI during their hospital stay. Comparatively,

6.03% (42/696) who had chest pain episodes less than 24 h were diag- nosed with an acute MI. Again, 14 patients didn’t specify how long their chest pain lasted. Therefore, the positive LR of chest pain lasting

>=24 h continuously for acute MI was 0.15 (95% CI 0.04 to 0.58). Among the patients who had successful follow up at 6 weeks, 239 patients had reported their chest pain had lasted for >=24 h continuously, and 7 of those had MACE (2.93%). Comparatively, the rate of MACE at

6 weeks for those who had chest pain for less than 24 h continuously was 56/585 (9.57%). Thus, the positive LR of chest pain lasting >=24 h continuously for MACE within 6 weeks was 0.36 (95% CI 0.18 to 0.74). Likelihood ratios for a number of other features related to the

non-specific chest pain Non-cardiac chest pain

598 (59.7%)

61 (6.1%)

patients? chest pain were assessed as detailed in Table 2. Regarding

the diagnosis of acute MI, notable findings included a positive LR of

Musculoskeletal chest pain

47 (4.7%)

2.10 (95% CI 1.22 to 3.61) for radiation to the right shoulder/arm, a

Non-ST elevation MI

44 (4.4%)

positive LR of 2.72 (95% CI 1.32 to 5.62) for radiation to both shoul-

Intraabdominal Cause

non-specific abdominal pain Cholecystitis

Gastritis

36 (3.6%)

22 (2.2%)

3 (0.3%)

2 (0.2%)

ders/arms, a positive LR of 0.32 (95% CI 0.11 to 0.98) for vomiting, a pos- itive LR of 1.39 (95% CI 1.08 to 1.80) for chest pain described as

“pressure”, and a positive LR of 0.44 (0.21 to 0.93) for chest pain

Pancreatitis

3 (0.3%)

described as “sharp” or “stabbing”. Additionally, zero of the 47 patients

Bowel obstruction

2 (0.2%)

who described their chest pain as a “tingling” or “pins and needles” sen-

Other intraabdominal cause

4 (0.6%)

sation were diagnosed with an acute MI, resulting in a positive likeli-

Pneumonia

Arrhythmia

24 (2.4%)

27 (2.2%)

hood ratio of zero with an incalculable confidence interval. To address

Atrial fibrillation

16 (1.6%)

this statistical issue, a univariate logistic regression was performed

Supraventricular tachycardia

5 (0.5%)

showed pins and needles sensation not to be a useful predictor of

atrial flutter

Other arrhythmia Pulmonary embolism Pneumothorax

2 (0.2%)

4 (0.4%)

15 (1.5)

3 (0.3%)

acute MI (p = 0.99). Of all the historical factors assessed, only continu- ous pain for >=24 h and vomiting were significant predictors for MACE within 6 weeks.

Aortic dissection

1 (0.1%)

As demonstrated in Table 2, when patients were specifically asked if

Pericarditis Pericardial effusion

1 (0.1%)

1 (0.1%)

they thought they were having a heart attack, 80 of 1002 patients stated

“yes,” and 8 of those 80 (10%) were, in fact, having an acute MI. A total of

Table 2

The percentage of patients reporting each historical feature who had an acute MI and MACE within 6 weeks and the associated positive likelihood ratio (LR) for MI and MACE.

Historical feature

% with Feature who had a MI

+LR for MI (95% CI)

% with feature who had MACE

+LR for MACE (95% CI)

Duration <=1 min

2/47 (4.3%)

0.95

2/38 (5.3%)

0.67

Duration >=24 h continuously

2/292 (0.7%)

(0.24-3.80)

0.15*

7/239 (2.93%)

(0.17-2.72)

0.36*

(0.04-0.58)

(0.18-0.74)

Radiation to back

12/260 (4.6%)

1.05

17/214 (7.94%)

1.04

(0.64-1.73)

(0.68-1.60)

Radiation to left Shoulder/arm

15/313 (4.79%)

1.10

21/272 (7.72%)

1.01

(0.72-1.67)

(0.70-1.46)

Radiation to right Shoulder/arm

11/125 (8.8%)

2.10*

11/108 (10.2%)

1.37

(1.22-3.61)

(0.78-2.42)

Radiation to both Shoulders/arms

7/63 (11.1%)

2.72*

6/54 (11.1%)

1.51

(1.32-5.62)

(0.67-3.40)

Radiation to Neck/throat

7/168 (4.17%)

0.95

10/145 (6.9%)

0.90

(0.47-1.89)

(0.50-1.62)

Radiation to Abdomen

5/116 (4.31%)

0.98

5/103 (4.85%)

0.62

(0.42-2.28)

(0.26-1.46)

Described as Pressure

26/432 (6.02%)

1.39*

34/359 (9.47%)

1.27

(1.08-1.80)

(0.99-1.62)

Described as Tightness

8/212 (3.77%)

0.85

11/179 (6.15%)

0.79

(0.45-1.62)

(0.45-1.38)

Described as Sharp/stabbing

6/303 (1.98%)

0.44*

12/253 (4.74%)

0.60

(0.21-0.93)

(0.36-1.01)

Described as Tingling or pins and needles

0/47 (0%)

0

2/40 (5.0%)

0.64

(N/A)

(0.16-2.58)

Worse with activity

20/492 (4.07%)

0.92

34/419 (8.11%)

1.07

(0.66-1.28)

(0.84-1.36)

Presence of nausea

12/260 (4.6%)

0.79

20/367 (5.45%)

0.70

(0.52-1.20)

(0.48-1.01)

Presence of vomiting

3/204 (1.47%)

0.32*

5/170 (2.94%)

0.37*

(0.11-0.98)

(0.16-0.86)

Presence of diaphoresis

20/399 (5.01%)

1.15

26/336 (7.74%)

1.01

(0.82-1.60)

(0.75-1.38)

Presence of lightheadedness

19/500 (3.8%)

0.86

28/436 (6.4%)

0.83

(0.61-1.21)

(0.62-1.10)

Presence of cough

11/302 (3.64%)

0.82

19/266 (7.14%)

0.93

(0.49-1.39)

(0.63-1.38)

Feels like a heart attack

8/80 (10.0%)

2.01*

8/72 (11.1%)

1.31

(1.04-3.88)

(0.66-2.60)

* Signifies statistically significant predictor.

701 patients said “no”, they did not think they were having a heart attack, and yet 33 (4.7%) of those were ultimately diagnosed with an acute MI. Meanwhile, 221 patients stated they were unsure or did not specify if they thought they were having a heart attack or not. The pos- itive LR associated with the above values of the patient’s perception of whether or not they were having a heart attack for diagnosed acute MI was 2.01 (95% CI 1.04 to 3.88).

To determine if certain clinical features were independent predictors of acute MI or MACE within 6 weeks, a multivariate analysis was done using age, gender, race, and all historical features that were significant positive or negative predictors of acute MI. As radiation to the right shoulder/arm and radiation to both shoulders/arms are similar vari- ables, they are not likely independent, and therefore, only radiation to the right shoulder/arm was used in the model. The results of the multi- variate analysis are shown in Table 3. As demonstrated, only radiation to

note, one 77 year-old male patient who had an MI reported a chief com- plaint of chest pain, but ranked his maximum pain score as zero, de- scribing an unusual sensation in his chest, “not a pain.”

  1. Discussion

This is the first study in over 15 years [6] to prospectively assess the association of features of the history in Undifferentiated chest pain patients with acute MI and other MACE, and it is the only study to

Table 3

The adjusted odds ratios and associated confidence intervals of chest pain historical fea- tures for acute MI and MACE within 6 weeks as determined by a multivariate analysis of features that were significant predictors on univariate analysis.

the right shoulder/arm was found to be an independent positive predic- tor of acute MI, and only continuous pain for >=24 h was found to be an independent negative predictor of acute MI and MACE within 6 weeks. Finally, regarding pain severity, the mean maximum pain score (0-10) for the 44 patients who were diagnosed with an MI was 7.4,

compared to 7.5 for the 958 who were not (a difference of 0.1 [95% CI

-0.8 to 0.9]). Univariate logistic regression analyses showed no statistically significant relationship between pain score and acute MI (p = 0.79) or MACE within 6 weeks (p = 0.12). Moreover, multivariate

logistic regression analyses controlling for age, race, and gender found no statistically significant relationship between pain score and the diag- nosis of acute MI (p = 0.24) or MACE within 6 weeks (p = 0.76). Of

Historical feature Odds ratio for MI (95% CI)

Duration >=24 h continuously 0.11* (0.02-0.36)

Radiation to right shoulder/arm 3.56*

(1.55-7.80)

Described as pressure 1.58

(0.77-3.40)

Described as Sharp/stabbing 0.53

(0.18-1.40)

Presence of vomiting 0.35

(0.08-1.02)

* Signifies statistically significant predictor.

Odds ratio for MACE (95% CI)

0.30* (0.12-0.64)

2.05

(0.93-4.25)

1.34

(0.72-2.54)

0.71

(0.32-1.50)

0.42

(0.14-1.00)

date to specifically assess whether particularly short (<=1 min) or long (>=24 h) episodes of chest pain can help rule out acute MI. Previous data about how long the pain from an MI lasts are scant. Duration

N60 min has previously been studied, and not shown to be very helpful in distinguishing MI from non-cardiac pain [4]. A review article states that pain that lasts only seconds is rarely indicative of Ischemic pain, but the authors admit that this has not been demonstrated in formal studies [21].

Indeed, pain that lasts only seconds does not seem to fit with the pathophysiology of MI as one would expect a longer duration of pain from a coronary artery occlusion. However, our data fits into the well- established concept that patients with MIs may have Atypical clinical presentations [22], especially if they are elderly [23]. That being said,

the data for chest pain lasting <=1 min is driven by just two patients who had MIs, so it may be that with a larger sample size we would

have, in fact, found the presence of this feature to have negative predic- tive value for MI.

On the other hand, the data for chest pain that lasted >=24 h continu- ously were fairly robust in providing value as a negative predictor for

both acute MI and MACE within 6 weeks, with only 2 acute MIs among the 292 patients who reported this feature. Moreover, this his- torical feature was the only significant predictor of acute MI and MACE of those evaluated my multivariate regression analysis. Previous data have suggested that no single historical variable can identify a low risk patient amongst those presenting to the ED for chest pain [5]. While this may still be true, continuous pain for >=24 h makes acute MI

unlikely. This previously unstudied historical feature should be incorpo-

rated into the risk assessment for patients who present to the ED with chest pain, perhaps via the history component of the HEART score.

Some of the secondary findings in this study confirm previous work about the most useful features of a chest pain history. For example, we found that chest pain radiation to the right shoulder/arm and radiation to both shoulders/arms were insensitive for acute MI (occurring in 25% and 16% of acute MI cases, respectively), but when present, they were the strongest positive predictors of acute MI. Prior studies support these findings [4,20,21]. Additionally, our finding that chest pain described as a pressure is a weak positive predictor of acute MI is consis- tent with prior data [18,24]. Next, our finding that chest pain severity does not help predict acute MI is also consistent with prior literature [21,25], and is consistent with an emerging theme in the literature suggesting that self-reported pain scores may not have predictive clini- cal utility in general [26]. Lastly, the description of the pain as sharp or stabbing was found to be a negative predictor of MI in our study and previously [5,17]. Although one study that looked at this component of the chest pain history did not find it to be a significant negative pre- dictor of MI, there was still a trend towards that result [6], while a prior systematic review concluded that sharp or stabbing pain decreases the likelihood of MI [21].

Some of our findings run in contrast to prior studies about the most useful features of a chest pain history. Namely, in studies from over 25 years ago, diaphoresis was found to be a positive predictor of acute MI [17,18,19]. However, a more recent study found that diaphoresis is not predictive of acute MI [6], and so did our study. Given the conflicting results, we suspect that diaphoresis is, at best, a weak predictor of acute MI. Additionally and similar to the case with diaphoresis, old data found nausea and vomiting to be positive predictors of acute MI [4,17,18,27]. On the other hand, a more recent study found nausea and vomiting not to be predictive of MI [6], while we actually found vomiting to be a negative predictor of acute MI. Notably, the sample sizes from the more recent study [6] and in our study were much larger than any of these previous studies. Therefore, while vomiting may have prognostic importance for a patient having an MI [27], we do not think that the presence of vomiting in a patient with undifferentiated chest pain should raise a physician’s suspicion for acute MI.

Finally, we were surprised that we did not find chest pain that worsens with activity to be a useful predictor of acute MI or MACE

since, intuitively, this would seem to be an important feature of the chest pain history. Moreover, prior data have found that exertional chest pain is a useful predictor of acute MI [6]. We may have failed to identify exertional chest pain as a useful predictor of acute MI because we used patient-completed surveys for our data. Determining whether or not the patient’s pain is truly worse with exertion or just worse from movement of the arm (because of Musculoskeletal pain) may require some physician interpretation.

Limitations

There are some limitations to consider when interpreting the re- sults of our study. First, this was a single-center study, and the results may not be generalizable to other facilities. Second, while our total number of patients was large, the number of patients who had MACE was relatively small, and the number of patients reporting cer- tain historical features (especially chest pain lasting <=1 min and pain

described as a tingling or pins and needles sensation) were small

enough that one or two patients could have had a large impact on the results.

Additionally, the results came directly from patient responses to surveys, while previous data about the historical features of chest pain used surveys filled out by physicians [4,17,18,19] or a specialist nurse [6]. We chose to use patient responses rather than health care provider-generated data to reduce bias. A physician caring for a pa- tient with known coronary artery disease may be more likely to re- cord a history that fits with their preconceived notions of what cardiac chest pain is. Conversely, for a very young person with no risk factors for coronary artery disease, a physician may be more in- clined to record a history that does not fit with their notion of cardiac chest pain. That being said, ultimately, the physician’s interpretation of what the patient says (rather than exactly what the patient says) is what drives their decision making.

Next, the patients in this study were not blinded from their diag- nostic test results. Doing so was not technically feasible in our de- partment without risking delays in care. We did not track what test results the patient knew about at the time they completed the sur- vey, although it is highly likely that at least the electrocardiogram was done prior to enrollment of all patients. We suspect that the pa- tient’s knowledge of their diagnosis or test results had only a small impact on the results of the study with one exception. The answer to the question “do you feel like you are having a heart attack now?” would likely be compromised by the patient’s knowing their test results. Thus, while there was a positive association of a “yes” an- swer to that question and the patient being diagnosed with an MI, that data is of questionable value.

Finally, while we looked at a number of historical features of chest pain, there are others that may be positively or negatively predictive of acute MI. For example, previous data have found that pleuritic chest pain and positional chest pain are negative predictors of MI [20]. Additionally, some patients (perhaps a quarter of them) have an MI without chest pain at all [3]. We did not include all possible historical features that could be associated with MIs as we wanted to keep the survey short to keep the patient’s answers as accurate as possible.

  1. Conclusions

For patients presenting to the ED with undifferentiated chest pain, no historical features can definitively rule in or rule out an acute MI. Even when the pain lasts for <=1 min, the patient may still have had an

MI. That being said, continuous pain lasting for >=24 h reduces the chance

that the patient is having an MI as does a description of the pain as sharp or stabbing. Amongst the assessed historical features, pain radiating to the right shoulder/arm or both shoulders/arms are the strongest posi- tive predictors that the patient is having an MI, and when the chest

pain is described as a pressure, there is a slight increase in the chance that the patient is having an MI.

  1. Presentations

Poster presentation at Florida ACEP August 2019. Abstract was pre- sented at ACEP 2019 in Denver.

  1. Financial sources

None.

Declaration of Competing Interest

The authors declared that there is no conflict of interest.

Appendix

Patient Sticker or Acct #

Survey #: (Lead RA) Phone #: & e-mail: Best time to call:

Adult Emergency Department Chest Pain patient survey

PART 1 – To be completed by patient while in the emergency department.

  1. Did you come into the emergency room because of pain or discomfort in your chest (circle one)?

Yes No

  1. How long has your current episode of chest pain/discomfort been going on? If you don’t currently have chest pain/discomfort, how long did your last episode of chest pain/discomfort last?

Weeks: Days: Hours: Minutes: Seconds:

  1. On a scale of 0-10 (with 0 being no pain, and 10 being the worst pain you could ever imagine), how bad was your chest pain/discomfort at its worst?

Pain scale score:

  1. Where does your chest pain/discomfort radiate (circle all that apply)?

Back Left shoulder/arm Right shoulder/arm Neck Abdomen Chest only Other

  1. Are any of the following symptoms present along with your chest pain/discomfort (circle all that apply)?

Nausea Vomiting Sweating Lightheadedness Cough None of these

  1. Which of the following best describes the quality of your chest discomfort (circle one)?

Pressure Tightness Ripping/tearing Sharp/stabbing Pins and needles Other

  1. Do you feel like you are having a heart attack now?

Yes No Not Sure

  1. If you have had a previous heart attack, does this episode of chest pain/discomfort compare:

No previous heart attack Similar to previous heart attack Similar, but worse Completely different

  1. Does your chest pain/discomfort get worse with strenuous activities such as walking or climbing stairs? Yes No

References

  1. Constant J. The diagnosis of nonanginal chest pain. Keio J Med 1990;39(3):187-92.
  2. Sandler G, Illahi MA, Lawson CW. Glyceryl trinitrate in angina pectoris. Lancet 1963; 1:1130-6.
  3. Uretsky B, Farquhar D, Berezin A, Hood WB. Symptomatic myocardial infarction without chest pain: prevalence and clinical course. Am J Cardiol 1977;40:498-503.
  4. Berger JP, Buclin R, Haller E, Van Melle G, Yersin B. Right arm involvement and pain extension can help to differentiate coronary diseases from chest pain of other origin: a prospective emergency ward study of 278 consecutive patients admitted for chest pain. J Intern Med 1990;227:165-72.
  5. Lee TH, Cook EF, Weisberg M, Sargent RK, Wilson C, Goldman L. Acute chest pain in the emergency room. Identification and examination of low-risk patients. Arch In- tern Med 1985;145(1):65.
  6. Goodacre S, Locker T, Morris F, Campbell S. How useful are clinical features in the di- agnosis of acute, undifferentiated chest pain? Acad Emerg Med 2002;9:203-8.
  7. Every NR, Larson EB, Litwin PE, et al. The association between on-site cardiac cathe- terization facilities and the use of coronary angiography after acute myocardial in- farction. N Engl J Med 1993;329:546-51.
  8. Reed GW, Rossi JE, Cannon CP. Acute myocardial infarction. The Lancet 2017;389 (10065):197-210.
  9. Bandstein N, Ljung R, Johansson M, Holzmann MJ. Undetectable high-sensitivity car- diac troponin T level in the emergency department and risk of myocardial infarction. J Am Coll Cardiol 2014;63(23):2569-78.
  10. Mueller C, Giannitsis E, Christ M, et al. Multicenter evaluation of a 0-Hour/1-hour al- gorithm in the diagnosis of myocardial infarction with high-sensitivity cardiac tro- ponin T. Ann Emerg Med 2016;68(1):76-87.e4.
  11. Hoffmann U, Truong Q, Schoenfeld D. Coronary CT angiography versus standard evaluation in acute chest pain. N Engl J Med 2012;367(4):299-308.
  12. Pontone G, Andreini D, Guaricci AI, et al. Incremental Diagnostic value of stress com- puted tomography myocardial perfusion with whole-heart coverage CT scanner in intermediate- to high-risk symptomatic patients suspected of coronary artery dis- ease. JACC Cardiovasc Imaging 2019;12(2):338-49.
  13. Six A, Backus B, Kelder J. Chest pain in the emergency room: value of the HEART score. Neth Heart J 2008;16(6):191-6.
  14. Safford MM, Brown TM, Muntner PM, et al. Association of race and sex with risk of incident acute coronary heart disease events. JAMA 2012;308(17):1768-74.
  15. Marco CA, Kanitz W, Jolly M. Pain scores among emergency department (ED) pa- tients: comparison by ED diagnosis. J Emerg Med 2012;44:46-52.
  16. Singh JA, Gabriel S, Lewallen D. The impact of gender, age, and preoperative pain se- verity on pain after TKA. Clin Orthop Relat Res 2008;466:2717-23.
  17. Tierney WM, Fitzgerald D, McHenry R. Physicians’ estimates of the probability of myocardial infarction in emergency room patients with chest pain. Med Decis Mak- ing 1986;6:12-7.
  18. Jonsbu J, Rollag A, Aase O. Rapid and correct diagnosis of myocardial infarction: stan- dardized case history and clinical examination provide important information for correct referral to monitored beds. J Intern Med 1991;229:143-9.
  19. Solomon CG, Lee TH, Cook EF. Comparison of clinical presentation of acute myocar- dial infarction in patients older than 65 years of age to younger patients: the multi- center chest pain study experience. Am J Cardiol 1989;63:772-6.
  20. Panju A, Hemmelgarn B, Guyatt G, Simel D. Is this patient having a myocardial in- farction? JAMA 1998;280(14):1256-63.
  21. Swap CJ, Nagurney JT. Value and limitations of chest pain history in the evaluation of patients with Suspected acute coronary syndromes. JAMA 2005;294(20):2623-9.
  22. Brieger D, Eagle KA, Goodman SG, et al. Acute coronary syndromes without chest pain, an underdiagnosed and under- treated high-risk group: insights from the global registry of acute coronary events. Chest 2004;126:461-9.
  23. Grosmaitre P, Le Vavasseur O, Yachouh E. Significance of atypical symptoms for the diagnosis and management of myocardial infarction in elderly patients admitted to emergency departments. Arch Cardiovasc Dis 2013;106(11):586-92.
  24. Chun AA, McGee SR. bedside diagnosis of coronary artery disease: a systematic re- view. Am J Med 2004;117:334-34315336583.
  25. Horner SM. Chest pain — no difference in severity between those having a myocar- dial infarction and chest pain from other causes. Int J Cardiol 1989;24(3):371-2.
  26. Zitek T, Pellman L, Uribe J, Guillen A. Pain scores are not predictive of radiographi- cally evident intraAbdominal pathology in patients with abdominal pain. Acad Emerg Med 2019. https://doi.org/10.1111/acem.13841 [Epub ahead of print].
  27. Herlihy T, McIvor ME, Cummings CC, Siu CO, Alikahn M. Nausea and vomiting during acute myocardial infarction and its relation to infarct size and location. Am J Cardiol 1987;60:20-2.

Leave a Reply

Your email address will not be published. Required fields are marked *