Qualitative and quantitative comparison of systolic velocity profiles between CO, HOCM, and aortic stenosis (Figures 6 and 7)
Qualitatively , the LVCO ICG spectral Doppler profile has a similar profile to the HOCM Doppler profile (Figure 6) and hence may be confused. The HOCM LVOT initial acceleration is slow followed by a second phase of acceleration which is faster as also seen with the spectral profile of the ICG associated with LVCO. In the LVCO patients with higher gradients, however (Figure 2), the second phase of acceleration appears steeper and faster than seen with the profile of the HOCM associated LVOT gradient. Indeed, the second phase of acceleration of the ICG spectral profile appears almost exponential and can be compared to one side of an inverted half pipe skateboard ramp (Figure 6).
To quantify the difference between the profiles, we assessed the ratio between peak and mean gradients. Consistent with the exponential, scooped-out appearance of the second acceleration portion of the profile, the ratio of peak to mean gradient was significantly higher in the LVCO than HOCM patients: 3.5 (range 2.0 to 6.1) vs 2.4 (range 1.8 to 3.25) respectively (p <0.0001). The difference between LVCO and HOCM peak/mean gradient ratios was even more marked when comparing the 25 patients with LVCO ICGs greater than 35 mmHg (which may be more clinically relevant, as at that magnitude of gradient there may be more confusion with LVOT gradients associated with HOCM) as in this subgroup the mean ratio of peak/mean gradient was 3.85 (range 2.68 to 6.1). In 23/25 patients with HOCM the ratio was between 2 and <3; in 23/25 patients with LVCO and ICGs greater than 35 mmHg, the ratio was 3 or higher.
For comparison, we assessed this ratio in patients with severe AS, and the ratio was significantly lower in this group at 1.68 (range 1.46 to 2.05) reflecting that the AS spectral profile is the most symmetrically parabolic contour between LVCO, HOCM and AS. In 24/25 patients with AS the peak/mean gradient ratio was < 2 (Figure 7).
Discussion
This is the first study, to our knowledge, to assess the quantitative pathophysiologic mechanism of the ICG. This is also the first study to provide a quantitative method to distinguish the LVOT spectral Doppler profile associated with HOCM from the intracavitary gradient spectral profile associated with cavity obliteration. We are not aware of any major current echocardiography textbook that details the specific nature of the intracavitary gradient associated with LVCO.
In our laboratory, approximately 1% of patients had the term “cavity obliteration” directly entered on the report (it is not currently a “check off” option). Most patients with LVCO, in the absence of other significant cardiac conditions, have intracavity gradients less than 36 mmHg. The spectral profile associated with lower gradients differs from the patients with higher gradients. The spectral profile associated with the lower gradients is more triangular with a slow acceleration and a relatively fast deceleration. The profile of the higher gradients has an elongated fast acceleration tacked onto the initial slower acceleration, followed by a similar fast deceleration to the baseline (Figure 2).
Higher ICGs are associated with greater extent of cavity obliteration, as defined by the percentage of the end diastolic LV cavity length obliterated in systole and more prolonged LVCO, defined as the duration that the LV walls are apposed. It seems logical that other factors such as the longitudinal, radial and circumferential strain rate velocity and degree of apical twist contribute, in addition to the extent and duration of apposition, toward determining the magnitude of the gradient. As an analogy, consider the noise made when you clap your hands; it is the speed of apposing the hands as well as the surface area of hands that are apposed that correlate with the noise produced. Thirdly, higher gradients are seen in those with smaller end diastolic and end systolic cavity dimensions.
These correlations of extent and duration of LVCO with the ICG are analogous to the previously demonstrated quantitative andtemporal relationships between SAM and the LVOT gradient that showed a significant correlation between the duration and timing of SAM septal contact and the LVOT gradient4,5. Unlike SAM associated gradients, which represent LVOT obstruction between the body of the LV cavity and the LVOT, the ICG gradients are presumed to arise from the gradient between the LV apex and the body of the LV beyond the virtually closed off apical portion of the LV. That the ICG occurs when the LV is virtually closed suggests that in these patients, the overall hemodynamic significance to the LV cavity is minimal, especially compared to LVOT gradients that occur in HOCM while the LV is still emptying 4,5. They do, however, imply high pressures at the LV apex.
In this regard, and to show the difference between LVOT gradients associated with HOCM and ICG gradients associated with LVCO, we studied another 25 patients with HOCM and severe SAM. We chose to compare the LVCO patients with gradients of 36 mmHg or greater, as that level the size of the gradient lies within the realm of the gradients seen with severe SAM where the confusion may arise.
The distinction in peak/mean gradient ratios between LVCO and HOCM may be helpful to determine quantitatively the origin of a high systolic velocity obtained from the apex when the origin is uncertain or the shape of the spectral profile is ambiguous or unclear qualitatively as the ICG and LVOT spectral profiles are somewhat similar with an initial slow acceleration followed by a second faster rate of acceleration. Meticulous placement of the continuous wave Doppler cursor through the body of the left ventricle to separate the LVOT profile from the LCO profile is not always possible, especially when LV cavity size is small or the cavity is not perfectly vertically aligned to the apical acoustic window. These different origin gradients may also be confused because of depth ambiguity, a known phenomenon of continuous wave Doppler, where spectral profiles from MR, LVOT and ICG may overlap (Figure 8). Confusion may also arise as Valsalva maneuver increases both the ICG and the LVOT gradient (Figure 9). We have noticed that some of our less experienced sonographers may confuse LVOT spectral profiles with LVCO ICG profiles, perhaps because, in addition to the foregoing, patients with HCM and LVCO have similar hyperdynamic left ventricular contraction. The obstructed LVOT gradient spectral profile in HOCM has been likened to a dagger, presumably due to its pointed appearance. Daggers, however, have many different shapes (Figure 6), and this is an imprecise way of identifying a profile, especially as the ICG spectral profile, also has a “point” and resembles a dagger. The ICG gradient, particularly in those with a peak gradient of more than 35 mmHg, however, has a distinctive appearance and resembles an inverted skateboard half-pipe slope (Figure 6). The quantitative index of separating the contours as described by the peak/mean ratio may be especially helpful clinically when the origin of the high velocity is in doubt. The distinction has clinical relevance, because treating the ICG gradient as if it were an LVOT gradient associated with HOCM11, with measures such as disopyramide, septal ablation or surgical myectomy, would be inappropriate, and potentially harmful although there is one case report 12 of using cibenzoline to reduce the intracavitary gradient from 65 to 35 mmHg with improvement in dyspnea.
The lower peak/mean gradient ratio for HOCM patients (2.4) than for the LVCO patients (especially those with peak gradients equal to or more than 36 mmHg) (3.8) lends weight to the known hemodynamic and clinical significance of LVOT gradients. For example, a peak HOCM LVOT gradient of 64 mmHg translates to a mean gradient of approximately 27mmHg, whereas a peak ICG gradient of 64 mmHg is equivalent to a mean gradient of 17 mmHg.
As a further comparison, as a contrast, and because depth ambiguity may overlay systolic velocity profiles obtained from the apex simultaneously, we also looked at 25 patients with severe AS. The peak/mean gradient ratio was lowest in this group, at 1.7, consistent with the more symmetrically shaped parabolic contour (Figure 6) associated with the AS spectral profile. For an equivalent peak gradient of 64 mmHg, there would be a mean gradient of 38 mmHg.
The clinical relevance of these intracavitary gradients is uncertain. It seems plausible that the higher intracavitary gradients may have significance, as the resulting high apical pressures, and the potential accompanying apical ischemia 9, provide a possible reason for the association between LVCO and adverse outcomes of the combined end-point of sudden death and potentially lethal arrhythmic events, in patients who also have HOCM 13. These high apical pressures are also the presumed etiology for the development of apical aneurysms in patients with HOCM and coexisting LVCO14. Furthermore, intense catecholamine excess, which may produce severely high apical pressures secondary to LVCO, may be the cause of the apical wall motion abnormality seen in Takotsubo cardiomyopathy 15 and the LV wall motion abnormalities associated with subarachnoid hemorrhage 16.
Conclusion
Our study provides insight into the mechanism of the ICG in patients with LVCO. Just as SAM is not an all-or-none phenomenon, ranging from late and minimal septal contact which produces a small LVOT gradient, to early and prolonged septal contact that produces a large LVOT gradient4,5, so too LVCO is not an all-or-none phenomenon. Greater extent, and longer duration of obliteration are associated with higher intracavitary gradients.
Our study also highlights the different qualitative differences between the Doppler spectral profiles of HOCM and LVCO. For the first time, this study reports a quantitative method of differentiation by using the ratio of peak/mean gradient. The difference between the profiles has clinical implications as the ICG associated with LVCO may be confused with the LVOT gradient of HOCM, and treatment for the former as if it were the latter would be inappropriate and potentially harmful.
Limitations of the study
This is a retrospective study. A prospective study performing simultaneous echocardiographic and Doppler studies would provide more insight into the pathophysiology between LVCO and the resulting ICG. Three dimensional echocardiographic assessment of apical obliteration might provide more accurate spatial quantification of the degree of obliteration. In addition, strain measurements of LV shortening would likely be illuminating.
References
1. Criley JM, Lewis KB, White RI, Ross RS. Pressure gradients without obstruction. A new concept of ”hypertrophic subaortic stenosis”. Circulation 1965;32:881-887.
2. Raizner AE, Chahine RA, Ishimori T and Awdeh M. Clinical Correlates of Left Ventricular Cavity Obliteration. Am J Cardiol 1977;40:303-309
3. Wigle ED, Auger P, Marquis Y. Muscular subaortic stenosis. The initial left ventricular inflow tract pressure as evidence of outflow tract obstruction. Circulation 1967;35:1100-1117
4. Pollick C, Rakowski H, Wigle ED. Muscular subaortic stenosis: the quantitative relationship between systolic anterior motion and the pressure gradient. Circulation 1984;69:43-9.
5. Pollick C, Morgan CD, Gilbert BW, Rakowski H, Wigle ED.Muscular subaortic stenosis: the temporal relationship between systolic anterior motion of the anterior mitral leaflet and the pressure gradient. Circulation 1982;66:1087-94.
6. Maron BJ. Clinical course and management of hypertrophic cardiomyopathy. N Engl J Med 2018; 379:655-668
7. Khamal S, Daggubati R, Gaalla A, Shah PM, Pai RG. Left ventricular cavity obliteration during dobutamine stress echocardiography. J Am Soc Echocardiogr 1998;11:957-60
8. Secknus M, Niedermaier O, Lauer M, Marwick T. Diagnostic and prognostic implications of left ventricular cavity obliteration response to dobutamine echocardiography. Am J Cardiol 1998; 81:1318-1322
9. Matsubara K, Nakamura T, Kuribayashi T, Azuma A, Nakagawa M. Sustained cavity obliteration and apical aneurysm formation in apical hypertrophic cardiomyopathy. J Am Coll Cardiol 2003;42:288-95.
10. Mitchell C, Rahko PS, Blauwet LA, Canaday B, Finstuen JA, Foster MC et al. Guidelines for performing a comprehensive transthoracic echocardiographic examination in adults: Recommendations from the American Society of Echocardiography. J Am Soc Echocardiogr 2019;32:1-64
11. Jaber WA, Nishimura RA, Ommen SR. Not All Systolic Velocities Indicate Obstruction in Hypertrophic Cardiomyopathy: A Simultaneous Doppler Catheterization Study. J Am Soc Echocardiogr 2007; 20:1009.e5-e7.
12. Isobe S, Ieda M, Kohno T, Hishiyama T, Maekawa Y, Tsuruta H, Murata M, Yashima F, Yanagisawa R, Tanaka M, Inami S, Nasuno T, Hruyama A, Sano M, Hayashida K, Fukuda K. Effective cinezoline treatment in a patient with midventricular obstruction after transcatheter aortic valve implantation.
13. Kim H, Park JH, Won KB, Yoon HJ, Park HS, Cho YK, Nam CW, Han S, Hur SH, Kim YN, Kim KB. Significance of apical cavity obliteration in apical hypertrophic cardiomyopathy. Heart 2016;102:1215-20
14 Minami Y, Kajimoto K, Terajima Y, Yashiro B, Okayama D, Haruki S, Nakajima T, Kawashiro N, Kawana M, Hagiwar N. Clinical Implications of Midventricular Obstruction in Patients with Hypertrophic Cardiomyopathy. J Am Coll Cardiol 2011;57:2346 –55
15. Pelliccia F, Kaski JC, Crea F, Camici PG. Pathophysiology of takotsubo syndrome. Complete review on the pathophysiological mechanisms of TTS with particular emphasis on the role of central and autonomic nervous system. Circulation. 2017;135:2426–41.
16. Pollick C, Cujec B, Parker S, Tator C. Left ventricular wall motion abnormalities in subarachnoid hemorrhage: an echocardiographic study. J Am Coll Cardiol. 1988;12:600-5.
Figure Legends
Figure 1
2D echo apical 4 chamber frames in a LVCO patient with peak ICG of 2.1 mmHg. End diastolic (ED) frame at top left. Apical length from apex to mitral annulus is 5.6 cm. Cavity obliteration first occurs in frame 7, and has already ended by frame 9 (post CO) which shows a tiny gap between the LV walls; therefore only 2 frames; 7 and 8: are obliterated. At a frame rate of 50 Hz, this equals 20 msec per frame, and therefore obliteration lasts 2 x 20 msec, or 40 msec. End obliteration apex to annulus length in frame 8 is 2.9 cm. Percent obliteration = 5.6-2.9/2.9 = 48%.
Figure 2
Left panel: upper – typical spectral profile in a patient with a small peak ICG (yellow arrow): lower – typical spectral profile in a patient with a higher peak ICG (green arrow)
Right panel: schematic of the spectral profile in a patient with a small ICG gradient (A) with an initial slow acceleration (1) that speeds up (2) and then decelerates (3). For the higher gradients (B), the acceleration continues (4) to a peak and then rapidly decelerates (5).
Figure 3
Peak ICG gradient vs peak/mean ratio for all 87 patients
Figure 4
Upper graph: Peak ICG vs % obliteration in 65 patients; Lower graph: Peak ICG gradient vs time in obliteration in 65 patients
Figure 5
Upper panel: Apical length % obliteration in patients subgrouped into peak gradients <36 mmHg and =/> 36 mmHg; Lower panel: Time in obliteration in patients subgrouped into peak gradients <36 mmHg and =/> 36 mmHg
Figure 6
Left panel: AS, HOCM, and ICG profiles. Right panel: real life images of a parabola (inverted city arch), daggers, and inverted half pipe skateboard ramp.
Figure 7
Comparison of peak/mean gradients between patients with AS, HOCM, and LVCO
Figure 8
Spectral continuous wave Doppler profile in a patient (not in the study) with overlapping (in order of peak velocity) mitral regurgitation (green arrow), SAM associated LVOT obstruction (blue arrow), and LVCO (yellow arrow), demonstrating depth ambiguity
Figure 9
Patient with LVCO and ICG at rest (yellow arrow - left panel) that doubles with Valsalva maneuver (green arrow – right panel)
Figure 1
2D echo apical 4 chamber frames in a LVCO patient with peak ICG of 2.1 mmHg. End diastolic (ED) frame at top left. Apical length from apex to mitral annulus is 5.6 cm. Cavity obliteration first occurs in frame 7, and has already ended by frame 9 (post CO) which shows a tiny gap between the LV walls; therefore only 2 frames; 7 and 8: are obliterated. At a frame rate of 50 Hz, this equals 20 msec per frame, and therefore obliteration lasts 2 x 20 msec, or 40 msec. End obliteration apex to annulus length in frame 8 is 2.9 cm. Percent obliteration = 5.6-2.9/2.9 = 48%. Figure 2
Left panel: upper – typical spectral profile in a patient with a small peak ICG (yellow arrow): lower – typical spectral profile in a patient with a higher peak ICG (green arrow)
Right panel: schematic of the spectral profile in a patient with a small ICG gradient (A) with an initial slow acceleration (1) that speeds up (2) and then decelerates (3). For the higher gradients (B), the acceleration continues (4) to a peak and then rapidly decelerates (5).
Figure 3