February 23rd, 2012
Reality Check: Do 42% of Women with AMI Present Without Chest Pain?
Harlan M. Krumholz, MD, SM
I am reading the new paper from NRMI in JAMA on the association of age and sex with AMI symptom presentation and am struck by the finding that 35% of the patients did not present with chest pain. This percentage is higher than I have seen elsewhere.
In our recent studies, spanning many sites, we have found that almost 90% of the patients present with typical or atypical chest pain. In the GRACE registry, “chest pain was the most common symptom for both men (94%) and women (92%).” The NRMI study reported that 42% of women presented without chest pain. They do not report why the patients did present or what other symptoms they experienced.
I wonder how this result compares with the clinical experience of our readers. Have you noted that more than 2 in 5 women in your practice who are admitted with an AMI do not have chest pain? The number seems at odds with my experience and my reading of the recent literature. What are your thoughts?
15 Responses to “Reality Check: Do 42% of Women with AMI Present Without Chest Pain?”
NEJM Journal Watch — Recent Cardiology Articles- Strategies and Criteria for the Diagnosis and Management of Myocarditis
- Sex-Based Disparity in CABG Outcomes: Are We Moving Toward Parity?
- Can Genetic and Social Risk Scores Improve Cardiac Risk Prediction?
- Extensive Ablation vs. Pulmonary Vein Isolation in Persistent Atrial Fibrillation
- Can a Novel Stent Design Improve Long-Term Outcomes of Percutaneous Coronary Intervention?
All the women in my practice with AMI or unstable angina have chest pain. Some of the women minimize symptoms, attributing them to anxiety. Some older patients with diabetes, describe minimal chest pain. I have correctly diagnosed these patients by listening carefully to their description of symptoms and viewing the symptoms in the context of their risk factors.
There are two components to any such attribution analysis that raises reliability as to the conclusions- 1.)accurate MI classification with all the comorbid confounders, particularly in the troponin era and 2.) absence of application of rigorous definition of ischemic somatic complaints. The somatic location, as we all know in clinical practice, is patient variable and specific to that patient- ie, jaw pain or interscapular or shoulder pain or arm pain or neck pain or epigastric pain may predominate and overwhelm any chest pain description by the patient, that is, if chest patient is present at all on presentation.
The cause of this high percentage may lie in making diagnosis of MI based on troponin only in a large number of patients. Positive troponin DOES NOT equal MI. High sensitivity of troponin for MI detection is offset by its low specificity. That is all “simple”, I think. (By the way, what is the proportion of patients who had any imaging method done to prove necrosis?)
Most of my ER patients have symptoms, but few have only classic symptoms. This has been shown in many studies; classic symptoms are rare.
This would of course, make sense. If you test only those who have classic symptoms, you only find the disease in this population. It is only when you test the little old ladies with weak and dizzy and find positives do you realize there is another population out there. This is called availability bias.
Excellent wiki page on cognitive biases.
External references on request, as I don’t put links in these posts.
In my clinical experience, 1/3 of AMI patients without symptoms appears to be high. This suggests a methodology or selection bias issue.
Also, this study includes patients from 1994-2006, and the early years predate routine use for troponins as a biomarker.
In the methods section, the following statement is made:
“Chest pain/discomfort was defined as any symptom of chest discomfort, sensation or pressure, or tightness; or arm, neck, or jaw pain occurring
before hospital arrival or preceding a diagnosis of acute MI. The chest pain/discomfort variable was classified as present or absent before admission, during admission, or both and may have included (but was not limited to) patients presenting with shortness of breath, nausea/vomiting, palpitations, syncope, or cardiac arrest. However, in the absence of chest pain/discomfort,the specific symptom (other than chest pain/discomfort) was not abstracted from the medical record.”
This raises some methodology issues for patient selection:
1. Some of these patients did have symptoms such as dyspnea, nausea, etc…, but they did not have chest pain. From a clinical perspective, is the population of interest those with/without “chest pain” or those with/without “symptoms”.
2. AMI patients who present with syncope, cardiac arrest, and possibly Killip 4 or shock are extremly high risk and included among those without chest pain. This subgroup is not be comparable to those AMI patients presenting with nausea, vomiting, palpitations.
I would have liked to see what % of women vs. men present with the highest risk AMI (cardiac arrest, shock), as that variable is likely driving the mortality risk.
Dr. Ting, as an astute clinical trialist/methodologist, makes manifest how diffused and inclusive the definition of “chest pain” is in the NRMI registry. Perhaps, it is semantics but chest is a fairly well defined anatomic location and it is confounding to headline and use “chest pain” as an all encompassing definition. A more binary and accurate semantic for prospective classification would be ischemic(which is what the NRMI definition includes) vs. nonischemic symptoms.
Great comments. Of note, the data collection for this study was conducted from 1994-2006. With NRMI winding down after about 2000, I expect that most of the data were from the 1990s. The authors do not report the percentage of patients who were diagnosed with troponins, but it may be a minority. They also do not report the percentages before and after the change in the MI definition. They do say that “the introduction of troponin assays in the registry by restricting the analysis to the STEMI cohort who would not be directly affected by this change in MI definition, and possible bias caused by earlier deaths (such as cardiac arrest) and potential for incomplete symptom characterization in this group by excluding individuals who died within 24 hours and 48 hours.” They report that the restriction did not affect the adjusted sex-specific differences, but we do not know the actual rates. I think it is unlikely that their finding of 42% of women did not have chest pain is a result of the biomarker-only MI.
What did you find in VIRGO?
I agree with you, dr Krumholz it is very surprising that nearly 1 woman over 2 suffers from a MI with totally atypical symptoms.
a recent study* performed “experimental ischemia ” in women by occlusion (during less than 2 minutes-thanks-) of a coronary artery and symptoms related too ischemia in women and men.
the main investigator (a woman) reports more frequent pains in the jaw or throat in women but she also insisted on the fact that “no sex differences were found in rates of chest or typical ischemic discomfort regardless of ischemic status”
may be at the beginning of the registry, used in JAMA paper, they were not aware about Yentl’ syndrome as we are now…
but the take home message may be :” the sensitive tracks in women are unfathomable…”
And thank you for your always remarkable comments.
*Mackay MH, Ratner PA, Johnson JL et coll. Gender differences in symptoms of myocardial ischemia. Eur. Heart J. 2011; 32: p 3107-14.
We can speculate about the numbers indefinitely,it clearly doesn’t make sense(mostly numbers but even discussion about it). The only thing one can conclude that the quality of NRMI data for whatever reason is unfortunately poor which may affect other studies coming out of it.
i do not see those high percents without any pain.
definition of chest pain is of importance. patients answers what you ask. if you ask chest pain in the correct way then answer will be correct. that is simple. if you do not ask epigastric pain as chest pain, patients will not so…
There are valid concerns about this specific variable “absence of chest pain” in the NRMI registry as well as suggesting a causal association with mortality.
However, I would not “throw out the baby with the bathwater”, because there are numerous high quality studies that have come from NRMI — including several by our esteemed colleague Dr. Harlan Krumholz.
My experiency coincides with the study published in JAMA. For over twenty years, practicing in Brazil and in the US, I’d say, 2 out of five women with AMI present mostly with dizziness, lightheadness, palpitation, shortness of breath and some chest discomfort, but not pain.
Just as Luciene says: “some chest discomfort, but not pain”. Patients do not always use the word of PAIN, rather frequently other words and their words designate unpleasant feeling equivalent to pain. This is partially the background of big number of “painless” MIs.
Absence of angina depends on what type of patient and emergency service we are talking. An open door, with spontaneous arrive, or a central service with 24h primary angioplasty facility? I agree that walking patients came to hospital more frequently by angina presentation; but we can receive patients from others clinics (surgery, for example), with acute pulmonary edema and elevated troponin (type 2 MI). Women who presents MI are older and have more comorbities (diabetes, hypertension). There are several confounders factors and a particular interpretation of information of database could be a bias of study. But, the best, most of us agree that there is a real difference in women’s MI presentation (maybe not so great) in relation to men.