Get PulmCCM’s Weekly Email Update
Stay up-to-date in pulmonary and critical care. No spam.
“I'm burnin' through the sky, yeah
Two hundred degrees,
That's why they call me Mister Fahrenheit”
Part 1 gave a brief overview of the mechanisms of atrial fibrillation [AF] as well as pertinent features of rate controlling agents and anticoagulation. In this brief second part, nuanced features of ventricular filling are considered as groundwork for reasoning around the hemodynamic consequences of AF.
Atrial Fibrillation and Hemodynamics
Approach to managing atrial fibrillation in the intensive care unit follows a few – sometimes challenging – questions:
- What is the hemodynamic effect of AF and how much of this is due to the AF itself [see below]?
- Are there reversible factors that can be addressed [see part 1]?
- Is AF still present despite the above and causing adverse effects [see below]?
If there is hemodynamic instability and/or on-going myocardial ischemia as a consequence of AF, direct current cardioversion [DCCV] is indicated based on low-quality evidence. The likelihood of success, however, is small in critical illness; accordingly, concurrent administration of rate or rhythm controlling agents should be strongly considered. The decision to introduce anticoagulation in the setting of DCCV should be made on a case-by-case basis given that the benefit in the critically-ill has never been quantified and this must be weighed against the patient’s bleeding risk. In the non-critically-ill, embolic stroke following cardioversion occurs almost entirely within the first 10 days. The quoted risk of embolic stroke following cardioversion is between 0 and 7% with an average risk of 1.5%.
The most commonly-encountered patient, however, is one in whom it is difficult to ascertain the hemodynamic consequences of the atrial fibrillation – is this rapid rate harming the patient’s stroke volume or is the atrial fibrillation with rapid ventricular response merely a bystander signaling an underlying evil? Certainly, hard-and-fast rules are challenging because of each patient’s individual cardiac function – including diastolic stiffness, contractility, afterload, valve function, etc.
Filling Only During Acceleration of the E-wave
One theoretical approach to this problem is to consider how AF affects diastolic filling. In AF, the a-wave in diastole is lost and filling is dominated by the early, E-wave - see figure 1A. The E wave is triangular in shape with a typical acceleration time [AT] of 80-100 milliseconds regardless of degree of LV stiffness. Thus, if we consider an extreme situation where the totality of left ventricular filling is limited to only the normal acceleration time – a filling time assumed to directly impair left ventricular end-diastolic volume – can we know what heart rate this is and consider this an upper limit of safety [see figure 1B]?In a fascinating, and perhaps forgotten, study of AF, an equation was derived relating R-R interval to total diastolic left ventricular filling time [LVFT].
LVFT = [0.834 x (R-R interval) – 219] +/- 48 [in milliseconds]
If we use this equation to limit LVFT to allow for only the acceleration time of the E-wave [see figure 1B], we find the RR interval to be 360 milliseconds [range of 310 to 405]. This RR interval represents a heart rate of 166 bpm [range of 150 bpm to 190 bpm]. In other words, in atrial fibrillation with rapid ventricular rates between 150 bpm and 190 bpm, there is only enough time for left ventricular filling to accept the first upstroke of the E wave! Interestingly, in patients of varying degrees of diastolic dysfunction, the AT is relatively preserved [ranging between 80 & 100 milliseconds]. Accordingly, ventricular rates above 150 bpm in atrial fibrillation will only afford enough diastolic time to accept the initial rise of the E-wave and – as Josh Farkas points out – ventricular rates above 150 are likely directly contributing to impaired hemodynamics.
Filling During Acceleration & Deceleration of the E-wave
Obviously, the deceleration time [DT] of the E-wave is an important contributor to diastolic ventricular filling. The DT varies the most between patients with different grades of diastolic dysfunction. For example, in grade I diastolic dysfunction, the DT can be ~2.5 times the length of the AT while in grade IV diastolic dysfunction the AT to DT ratio is nearly 1. If we assume that grade I diastolic dysfunction displays, on average, the longest E-wave from start-to-finish [~ 330 milliseconds], these patients will require a comparatively low heart rate to ensure that the long E-wave completely contributes to diastolic filling [figure 1C]. From the equation listed above, we can solve for the RR interval and find that a ventricular rate of 90 bpm [range 85 to 100 bpm] allows for the full E-wave to contribute to left ventricular filling in typical grade I diastolic dysfunction. Accordingly, a low SV at a ventricular rate approaching 100 bpm almost certainly has an etiology beyond that of the shortened LVFT in AF. If a shorter E-wave duration is assumed, this heart rate cutoff rises to roughly 110 to 120 bpm in AF.
Importantly, the discussion of E-wave filling time above is derived from echocardiograms of stable – and typically ambulatory – patients. The duration of the E-wave in a patient with grade IV diastolic dysfunction, but with a low left atrial pressure secondary to hypovolemia or sepsis or other insult may be quite prolonged and diminished.
Finally, the above analysis only speaks for left ventricular filling [i.e. end-diastolic volume - EDV]; as we know, increasing EDV does not necessarily translate into increased stroke volume [SV]. The SV depends not only upon EDV, but also afterload, valve function and contractility. For example, lowering the heart rate and adding EDV in a patient with a dilated left and/or right ventricle can be detrimental.
The Atrial Kick
Notably, the loss of the a-wave [figure 1A] may still play an important role in maintaining EDV and SV – even if LVFT is optimized to allow for the entirety of the E-wave. It has been observed in AF that a systolic contraction following an RR interval of at least 800 ms [i.e. a HR < 75 bpm] produces a nearly equivalent hemodynamic response as compared to the same patient converted to normal sinus rhythm. Accordingly, and with respect to question 3 above, if hemodynamics normalize or improve tremendously following beats of particularly long R-R intervals [e.g. above 800 milliseconds or HR < 75] then the loss of atrial kick may be especially important and rhythm control to regain the a-wave may be helpful.
Return to Case
The fellow is extremely concerned about the patient’s low pulse pressure, rapid ventricular rate and severe mitral stenosis. In addition to treatment for community-acquired pneumonia and anticoagulation for valvular atrial fibrillation [and acute pulmonary embolus], the fellow initiates the patient on intravenous esmolol in the MICU with serial Doppler assessments of the patient’s narrowed mitral orifice and left ventricular outflow tract. With each drop in ventricular rate by 10 beats per minute, she assess the mitral and aortic valves. To her chagrin, the fellow sees both prolongation of the E-wave and increase in the LVOT Doppler velocity time integral [VTI] with each 10 beat decrement in ventricular rate. At a ventricular rate of 80-90 bpm, there is no further increase LVOT VTI and the esmolol dose is maintained. The patient’s blood pressure is 122/70 and her oxygenation is improved to 95% without supplemental oxygen.
Dr. Kenny is the cofounder and Chief Medical Officer of Flosonics Medical; he also the creator and author of a free hemodynamic curriculum at heart-lung.org