Author + information
- Received December 18, 2003
- Revision received April 16, 2004
- Accepted April 27, 2004
- Published online November 2, 2004.
- William C. Roberts, MD, MACC* ( and )
- Jong M. Ko, BA
- ↵*Reprint requests and correspondence:
Dr. William C. Roberts, Baylor Heart & Vascular Institute, Baylor University Medical Center, 3500 Gaston Avenue, Dallas, Texas 75246
Objectives The purpose of this study was to correlate the weights of operatively excised stenotic aortic valves to preoperative transvalvular peak systolic gradients and to calculated aortic valve areas.
Background No previous publication has correlated the weights of stenotic aortic valves to the transvalvular gradients or to the calculated aortic valve areas.
Methods We weighed operatively excised stenotic aortic valves in 324 adults who had undergone preoperative left-sided cardiac catheterization.
Results As the weights of the operatively excised stenotic aortic valves increased (from <1 g to >6 g), the average transvalvular peak systolic pressure gradients progressively increased. For any valve weight, in general, the women had higher average transvalvular gradients (p ≤ 0.005) and lower average valve areas (p ≤ 0.008) than did the men. Correlation between aortic valve weight and transvalvular gradient improved further when gender was taken into account.
Conclusions Preoperative transvalvular peak systolic pressure gradients across stenotic aortic valves correlate better with the weights of the operatively excised valves than do the calculated valve areas.
Recently, we reported weights of operatively excised stenotic unicuspid, bicuspid, and tricuspid aortic valves in 499 adults and their relation to age, gender, body mass index (BMI), and presence or absence of concomitant coronary artery bypass grafting (CABG) (1). No previous studies have compared weights of stenotic aortic valves to transvalvular pressure gradients across them or to calculated aortic valve areas. Such is the purpose of this study.
From January 1998 through June 2003 (65 months), a total of 1,272 patients had one or more cardiac valves or portions of cardiac valves submitted to the surgical pathology unit of the Department of Pathology of Baylor University Medical Center (Fig. 1).After elimination of the 33 patients with active infective endocarditis, the 6 who had isolated tricuspid valve replacement, the 444 who had mitral valve replacement or repair, and the 58 who had combined mitral and aortic valve replacements, 731 patients who had isolated aortic valve replacement remained. Of them, 151 had the valve replacement for pure aortic regurgitation (no element of valve stenosis) and 580 for aortic valve stenosis (with or without some associated aortic regurgitation). Of the 580 patients with stenotic aortic valves, the operatively excised valve was weighed by one of us (W.C.R.) in 575 patients; of them, 251 were eliminated from this study because cardiac catheterization was not performed before aortic valve replacement, the data obtained at cardiac catheterization was not available to us, or the left-sided cardiac hemodynamic data was missing or incomplete. The present study is limited to the 324 patients who had isolated aortic valve replacement to excise a stenotic aortic valve, had peak left ventricular (LV) to systemic arterial peak systolic pressure gradients ≥10 mm Hg, had calculated aortic valve areas obtained at cardiac catheterization before the aortic valve replacement, and had the operatively excised stenotic aortic valve weighed.
Each operatively excised cardiac valve arrived at the surgical pathology unit in a container filled with formaldehyde. The valve was removed from the container, placed on absorbing paper to absorb excess formaldehyde, and then weighed on an Ohaus scale (Ohaus Corp., Florham Park, New Jersey), which has an accuracy to 0.01 g. The age and gender of each patient were obtained from the patient identification label pasted on the container submitted from the operating room. The type of dysfunction of the operatively excised aortic valve was determined initially by gross inspection of the valve, and confirmation that the valve was stenotic was obtained from the cardiac catheterization report. The interval from the cardiac catheterization to aortic valve replacement was <3 months (usually <1 week) in 312 (96%) of the 324 patients, from >3 to 6 months in 5 patients, from >6 to 12 months in 6 patients, and 14 months in 1 patient. Most of the pertinent clinical information, including age, gender, BMI, whether or not concomitant CABG was performed, was obtained from the Baylor University Medical Center Apollo Advance Cardiovascular Data Integration system. In the 26 patients in whom cardiac catheterization was not performed at Baylor University Medical Center, the record room chart was reviewed to obtain the pertinent hemodynamic data.
Statistical analysis was performed by using SigmaStat Version 2.0 (SPSS Inc., Chicago, Illinois). Parametric tests were used for the analysis of the data that passed the assumption tests of normality and equal variance: for continuous variables, the unpaired ttest was performed for two variable comparisons and one-way analysis of variance for more than two variable comparisons. When the data did not pass the assumption tests, nonparametric tests (i.e., Mann-Whitney rank-sum test for two variable comparisons and Kruskal-Wallis one-way analysis of variance on ranks test for more than two continuous variable comparisons) were used. The correlation coefficient (r) was obtained by using either the parametric Pearson product moment test or the nonparametric Spearman rank order test. A value p < 0.05 was considered to be statistically significant. The study protocol was approved by the Institutional Review Board of Baylor University Medical Center.
The weights of the operatively excised stenotic aortic valves ranged from 0.69 to 11.30 g (mean 2.72 g), a 16-fold difference between the smallest and the largest (Table 1).The ages, BMIs, LV to aortic peak systolic pressure gradients, calculated aortic valve areas, whether or not concomitant CABG was performed, and ejection fractions are summarized for both men and women according to the aortic valve weights in Table 1. The transvalvular peak systolic pressure gradients in both men and women increased as the valve weights increased (p < 0.001, Kruskal-Wallis one-way analysis of variance on ranks test) (Table 1), the calculated aortic valve areas also decreased as the valve weights increased (p ≤ 0.002, Kruskal-Wallis one-way analysis of variance on ranks test), but the changes were not as impressive (Table 1). The men in general had higher valve weights than did the women (p < 0.001, Mann-Whitney rank-sum test), lower transvalvular gradients (p ≤ 0.005, unpaired ttest), and higher valve areas (p ≤ 0.008, unpaired ttest) (Table 1).
The valve weights, gradients, and valve areas in both men and women according to aortic valve structure (unicuspid, bicuspid, or tricuspid) are summarized in Table 2.The fewer the number of aortic valve cusps, the heavier the valves were, the greater the peak transvalvular gradients, and the smaller the valve areas.
Valve weights, gradients, and valve areas were not significantly different (by Kruskal-Wallis one-way analysis of variance on ranks test) in patients ≤40 years, 41 to 70 years, and 71 to 90 years of age (Table 3).Valve weights, gradients, and mean areas varied relatively little in patients with BMIs <25 kg/m2, 25 to 30 kg/m2, and >30 kg/m2(Table 4).
Compared with the 166 patients having coronary bypass, the 158 patients not having concomitant bypass grafting had heavier valves (p < 0.001, Mann-Whitney rank-sum test), higher transvalvular peak pressure gradients (p = 0.045, Mann-Whitney rank-sum test), but similar valve areas (Table 5).
Valve weights, gradients, and areas were insignificantly different (by either Kruskal-Wallis one-way analysis of variance on ranks test or one-way analysis of variance) when women were compared with women and men with men with LV ejection fractions ≤35%, 36% to 45%, and >45% (Table 6).
The relation of aortic valve weight to transvalvular peak systolic gradient for each of the 201 men is illustrated in Figure 2and in the 123 women, in Figure 3.The relation of aortic valve weight to aortic valve area in each of the 201 men is shown in Figure 4and in each of the 123 women in Figure 5.
Of the 324 patients, only 13 had moderate (3+ of 4+) or severe (4+ of 4+) aortic regurgitation, and the data in these 13 patients are tabulated in Table 7.Although these patients were considered to have moderate or severe aortic regurgitation after cardiac catheterization, only three had dilated LV cavities, and these were the three with low LV ejection fractions.
The present study shows that the peak systolic pressure gradients across stenotic aortic valves correspond reasonably well with the weights of the operatively excised stenotic aortic valves in adults when gender is considered (Figs. 2and 3). Although the average transvalvular peak systolic pressure gradients were similar in both men and women (approximately 52 mm Hg), the mean weights of the stenotic aortic valves in the men were nearly twice (1.8 ×) those is the women. Thus, for any given weight, the peak systolic pressure gradients across stenotic aortic valves were greater in the women than in the men. The smallest valves belonged to the women, and the largest ones to the men (Table 1).
No previous attempts have been made to estimate the valve weight from the transvalvular peak pressure gradient. The present data suggest that in women, each 1-mm Hg peak systolic pressure gradient represents about 35 mg of stenotic aortic valve, and in men each 1-mm Hg peak gradient represents about 65 mg of stenotic aortic valve. Thus, a 53-mm Hg peak systolic gradient across a stenotic aortic valve in a woman calculates into a 1.80-g valve, and a 52-mm Hg transvalvular peak gradient across a stenotic aortic valve in a man translates into a 3.27-g valve.
The peak systolic transvalvular gradients correlated better with valve weights than did the valve areas. In the men, as the valve weights progressively increased from >1 to >6 g, the average transvalvular peak systolic pressure gradients progressively increased from an average of 36 to 87 mm Hg (Fig. 2), whereas the valve areas decreased from 0.86 to 0.71 cm2(Fig. 4). In the women, as the valve weights progressively increased from <1 to 4 g, the transvalvular peak systolic pressure gradients progressively increased from 28 to 85 mm Hg (Fig. 3), whereas the valve areas decreased from 0.83 to 0.51 cm2(Fig. 5).
Although valve area is used in many medical centers as the gold standard of valve stenosis (2–4), the present study—the first to correlate either transvalvular peak systolic pressure gradient or the calculated aortic valve area with valve weight—demonstrates that the peak systolic transvalvular pressure gradient is more indicative of the degree of valvular obstruction than is the calculated valve area, particularly in the patients with low transvalvular peak systolic pressure gradients.
We are well aware that the gradient across a stenotic valve is dependent on flow across that valve. Several publications have discussed patients with “low gradient-severe aortic stenosis,” but in none of these reports has the operatively excised valve been illustrated, described, or weighed (5–11). The peak systolic gradient across a stenotic aortic valve is, of course, a direct measurement. The aortic valve area, in contrast, is indirect and calculated from the cardiac output, the heart rate, the systolic LV ejection period, a constant (44.3), and the square root of the mean transvalvular gradient. The major message of this study is that care must be taken in concluding that “severe” stenosis can or does occur in the absence of a large peak systolic pressure gradient across the valve. The data herein support the view that a low gradient—irrespective of flow—means non-severe stenosis, and that the valve area calculation is, or can be, very unreliable in patients with low peak systolic gradients across stenotic aortic valves.
The positive features of the present study are the fact that a large group of patients were analyzed, that all valves were weighed by the same person, and that all cases analyzed were seen over a relatively short period (65 months). The negative feature of this study is the fact that the cardiac catheterizations were performed by numerous cardiologists and that the standards of each probably varied, and that most (98%) of the ejection fractions were estimated rather than calculated.
- Abbreviations and acronyms
- body mass index
- coronary artery bypass grafting
- left ventricular
- Received December 18, 2003.
- Revision received April 16, 2004.
- Accepted April 27, 2004.
- American College of Cardiology Foundation
- Cannon S.R.,
- Richards K.L.,
- Crawford M.
- Burwash I.G.,
- Thomas D.D.,
- Sadahiro M,
- et al.
- Cannon J.D. Jr..,
- Zile M.R.,
- Crawford F.A. Jr..,
- Carabello B.A.
- Brogan W.C. III.,
- Grayburn P.A.,
- Lange R.A.,
- Hillis L.D.
- Connolly H.M.,
- Oh J.K.,
- Schaff H.V,
- et al.
- Monin J.-L.,
- Monchi M.,
- Gest V.,
- Duval-Moulin A.-M.,
- Dubois-Rande J.-L.,
- Gueret P.
- Nishimura R.A.,
- Grantham J.A.,
- Connolly H.M.,
- Schaff H.V.,
- Higano S.T.,
- Holmes D.R. Jr.
- Grayburn P.A.,
- Eichhorn E.J.