Which came first: the AFib or the MI?

Which came first: the atrial fibrillation or the MI?

Anna Meyendorff, MD PGY-3

(Image from: http://www.thestar.com/content/dam/thestar/life/2010/07/15/science_answers_the_question_which_came_first_the_chicken_or_the_egg/lichickenoreggjpg.jpeg.size.xxlarge.letterbox.jpeg)


Initial Presentation:

77yo AA F hx HTN, DM, CVA, former smoker presenting with chest pain. Symptoms started 24 hours prior to ED presentation while watching TV. Pain was epigastric/lower sternal, 10/10, nonradiating. Associated with SOB, but not nausea, vomiting, or diaphoresis.

Initial vitals: BP 125/61   P 114   R 21   T 98.4   95% on RA   FS 377


Initial EKG showing afib with RVR with 2mm ST elevations in III and aVF and possible small ST depressions in I, avL. Given diltiazem 20mg IVP with conversion to sinus rhythm, but not resolution of pain. Repeat EKG showing continued ST elevation in the inferior leads, so MEHEART (STEMI code) was activated by ED. Pt given aspirin, plavix, heparin IV bolus. Chest pain resolved by time of cardiology evaluation, so decision made not to proceed with transfer for immediate cardiac cath because symptoms resolved and presentation “unconvincing for STEMI” (per cardiology). In addition, lab results revealed that patient was in renal failure, further increasing the risk of dye infusion. Initial troponin was 5 (normal range <0.09). Pt was admitted to the CCU with dx: NSTEMI, plan for medical management with heparin drip.


CCU #1 course:

Overnight, patient developed worsening chest pain. Evidence of CHF/fluid overload on exam (crackles at lung bases, peripheral edema, JVD). Was given lasix 80mg IVP and started on nitro drip, without resolution of CP. She developed worsening respiratory distress requiring initiation of BIPAP. Serial troponins uptrending from initial (5) to 37 to 96. Creatinine worsened from 2.3 to 2.8. Repeat EKG revealed dynamic EKG changes with more pronounced inferior ST elevations and ST depressions in V3-V6. Pt transferred to interventional cardiology center for immediate catheterization for inferior wall STEMI.


CCU #2 course:

Cath revealed 100% circumflex occlusion and successful PCI with placement of DES. Elevated filling pressures also noted (LVEDP 35mmHg, PCWP 18mmHg), consistent with cardiogenic shock. Post-cath, pt improved and was weaned off nitro and lasix drips. Echo revealed inferolateral and inferior wall akinesis with LVEF 45%. Overnight after cath, patient noted on telemetry to be in afib with RVR (110s-140s), but was asymptomatic. She was loaded with amiodarone for rate and rhythm control. Creatinine continued to uptrend (6.7 on HD#5), urine output diminished; renal consult suspected combination of toxic AVN (contrast-induced) plus hemodynamic insults from afib, decompensated CHF, and acute MI. Discussion ongoing regarding initiation of hemodialysis at the time of this post.



So, this case got me thinking. Was the atrial fibrillation noted initially somehow related to the MI? If so, did one cause the other? Which came first?

Numerous studies have demonstrated that myocardial infarction increases risk of afib. Atrial fibrillation is common in the periinfarction period; it occurs in 6-21% of patients, usually in the first 72 hours post-MI. Development of afib in this window carries increased 30-day mortality as well as long-term mortality. Afib during acute MI can be caused by multiple factors, but is most commonly caused by atrial stretching from elevated pressure secondary to heart failure. It can also rarely be secondary to atrial ischemia or infarction, pericarditis, or atherosclerosis of the vessels supplying the SA and AV nodes. (1, 2)


New studies have been published, including several in 2015, documenting the opposite relationship: that atrial fibrillation increases risk of MI, though the mechanism is poorly understood. Soliman et al published a prospective cohort study in JAMA in January 2014. Analysis of 23,928 adult subjects without heart disease (the REGARDS cohort) over median 4.5 years demonstrated that afib conferred a nearly 2-fold increased risk of MI. Even after adjusting for smoking status, hypertension, diabetes, hyperlipidemia, BMI, aspirin use, and more risk factors, these results remained significant (Hazard Ratio 1.70 [95% CI, 1.26–2.30]). Even more interesting, the subgroup analysis also demonstrated a higher risk in women vs men (HR 2.16 vs 1.39) and in blacks vs whites (HR 2.53 vs 1.26). Recall that the patient in the case example above was both black and female. Unlike sex and race, age was not found to change the risk. Limitations of the study included lack of tracking how this risk develops over time after developing afib. It also did not evaluate asians, hispanics, or other racial groups, potentially limiting result generalizability. (3)


A second study was published by Soliman in Circulation in May 2015 with the goal of further elucidating the cause of the previously noted findings and evaluating their findings with a time variable. The study considered 14,462 participants (the ARIC cohort) over 21.6 years (median). They found that the mean time from afib diagnosis to MI was 4.82 years (median 3.35). The incidence of MI in those with afib was nearly 3-fold higher than those without afib (Incidence rate ratio 2.93), and after adjustments for cardiovascular risk factors nearly 2-fold (1.63). They replicated the findings from the previous study for women vs men (IRR 3.75 vs 2.27, P<0.001), and suggested the same in blacks vs whites (HR 2.05 vs 1.52), without statistical significance (P=0.16). In further subgroup analysis by type of MI, they found afib to be associated with increased risk of NSTEMI (HR 1.80), but not STEMI. NSTEMI is usually caused by a nonocclusive thrombus, while STEMI is usually caused by a complete arterial occlusion. According to the authors, the association with only NSTEMI suggests that direct coronary thromboembolization (the mechanism by which afib leads to stroke) is not the primary mechanism by which afib leads to MI. One proposed mechanism suggests the afib-induced prothrombotic risk (from platelet activation, inflammation, and endothelial dysfunction). An alternative mechanism proposes a rapid ventricular rate (afib with RVR) leading to demand ischemia (NSTEMI). (4)


An editorial in the same issue of Circulation includes the following figure as a graphic for understanding the interaction between these two entities. (5)

pasted image 0

Patients found with new onset atrial fibrillation in the emergency department are often admitted for initiation of anticoagulation therapy with the primary goal of preventing stroke. Atrial fibrillation increases stroke risk by 4-5 fold. Based on the newly emerging evidence, it bears noting that these patients also carry a ~2 fold increased risk for MI. More studies must be performed to evaluate whether this risk should alter the standard management of newly diagnosed atrial fibrillation. What medication regimen is optimal for both MI and CVA prevention?



(1)Angeli FF. Atrial fibrillation and mortality in patients with acute myocardial infarction: a systematic overview and meta-analysis. Current cardiology reports. 2012-10;14:601-610.


(2)Schmitt JJ. Atrial fibrillation in acute myocardial infarction: a systematic review of the incidence, clinical features and prognostic implications.. European Heart Journal. 2009-05;30:1038-1045.


(3)Soliman EE. Atrial fibrillation and the risk of myocardial infarction. JAMA internal medicine. 2014-01;174:107.


(4)Soliman EE. Atrial Fibrillation and Risk of ST-Segment-Elevation Versus Non-ST-Segment-Elevation Myocardial Infarction: The Atherosclerosis Risk in Communities (ARIC) Study. Circulation (New York, N.Y.). 2015-05;131:1843-1850.


(5)Vermond RR. Does myocardial infarction beget atrial fibrillation and atrial fibrillation beget myocardial infarction?. Circulation (New York, N.Y.). 2015-05;131:1824-1826.

1 Comment on "Which came first: the AFib or the MI?"

  • Get the posts but didn’t have time for this one. Pity. I would have chimed in. This is a sad case in the end.

    Sometimes you do studies and find something unexpected, like the potential harmfulness of Supplemental O2.

    Unfortunately most studies are done find something obvious, and they are only revelations because we have forgotten basic “commandments”.

    Commandment number 1. the heart perfuses in diastole. Therefore necessarily any increase in heart rate requires the trade off of time in diastole. UCAF can cause MI in a well known bunch of ways; by inducing demand ischemia (which is what we usually see ) or by robbing the patient of their “atrial kick”, which can contribute up to 20% of the EF in valvular diseases like aortic stenosis. If this occurs in the setting of someone with subclinical stenotic lesions viola; MI. But the hallmark is if you treat the Arrythmia you treat the chest pain and they get better be the EKG resolves. This patient’s didn’t.

    2. Ischemia is a setup for reentry arrythmias. As a matter of fact we are not two far from the day when therapy for an mi include aspirin, morphine, and a laxative to prevent straining/valsalva, and all the antiarrythmics to prevent/treat what killed patients in the first 24 hrs; arrythmias. It was the major killer of MI patients in the first 48hours. Cardizem built up its “street credibility” in the treatment of these arrythmias.

    3. CHF induce by an MI is bad. Syncope induced by an MI is bad. We are so used to admitting CHF and syncope patients that we forget why we we were taught to do it in all patients by our forefathers; if a healthy person suddenly loses >40% of their cardiac output they are going into fulminant cardiogenic shock. More than that and you syncopize. Or sudden cardiac death. MI’s don’t have the monopoly on this. A massive PE or sudden valvular insufficiency that ribs you of 40% of your EF gets you just as dead.

    4. Modern cardiology has changed all that. We decided to go “old school” on this one and the patient paid.

    5. New commandment on the block Calcium channel blockers are “OUT” in the management of ischemic heart disease. They do nothing to affect morbidity/mortality and there is evidence to suggest that they increase it. We cling to it out of some misguided love of the fact that it controls heart rate better; they might but that is all they do. Measure that against the mountains of evidence supporting BBlocker use.

    Now to the case.
    This was an MI. Person in usual state of health suddenly gets struck down with chest pain. Presents with afib. Patients arrythmia is treated successfully and STILL has pain and still has the EKG changes. The fact that the chest pain is gone by the time cardiology gets down is only a testimony to the power of calcium channel blockers as antiarrythmics. They still have work to do even if the pain is gone. Relief of the heart rate should resolve the pain and ANY rate related EKG changes. Cath lab. Period. Sorry they needed their beauty sleep but that is irrelevant.

    Sounds like they didn’t get any sleep anyway because even though they didn’t see it she tried to prove to them that she lost 40% of her EF by going into APE (no ACEI? No hydralazine/NTG if ACEI intolerant? This case gets better and better- her renal insufficiency was a sign of “forward failure; especially if prerenal) and making troponins.
    The next day suddenly she is an emergency she gets transferred and gets a dye load in her prerenal state made worse by high dose lasix, and every is shocked she goes into renal failure. BTW this could have been avoided or the risk doing the stents only and by foregoing the venticulogram EDP calculations, these could have been safely calculated by echo after the Cath

    Afib did little in this case but kick up a smokesceen


Leave a Reply