|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Cardiovascular Diseases and Internal Medicine (Drs. Milavetz, Hayes, Seward, and Roger), the Section of Biostatistics (Ms. Weston), and the Division of Thoracic and Cardiovascular Surgery (Dr. Mullaney), Mayo Clinic and Mayo Foundation, Rochester, MN.
Correspondence to: Veronique L. Roger, MD, MPH, Division of Cardiovascular Diseases and Internal Medicine, Mayo Clinic Rochester, 200 First Street SW, Rochester, MN 55905
| Abstract |
|---|
|
|
|---|
Materials and methods: We examined 92 women and 82 men who underwent echocardiography before valve replacement for aortic stenosis.
Results: Women had a smaller cavity size (LV end-diastolic diameter 48.2 ± 7 mm in women vs 53.6 ± 7.6 mm in men; p = 0.0001) and higher ejection fraction (59% in women vs 54% in men; p = 0.02). LV mass was greater in men than women (300.4 ± 88 g in men vs 250.6 ± 85.8g in women; p = 0.0055) but when corrected for body surface area, the difference was not significant. The prevalence of LV hypertrophy was similar in both sexes (51% in women vs 49% in men; p = 0.62). The 5-year survival was 82% in women and 79% in men (p = 0.9).
Conclusion: Several descriptors of LV geometry differed between men and women. These differences were largely eliminated after normalizing for body surface area. No differences in surgical mortality or long-term outcome were noted.
Key Words: aortic valve disease outcome left ventricular geometry sex difference
| Introduction |
|---|
|
|
|---|
The present study was designed to examine clinically used indicators of LV geometry in a large series of men and women with surgical AS and to test the hypothesis that there was no sex association in LV geometry and outcome after valve replacement in AS.
| Materials and Methods |
|---|
|
|
|---|
The hemodynamic severity of AS and LV function were assessed by
echocardiography with commercially available ultrasound systems. By
using the American Society of Echocardiography measurement techniques,
short-axis measurements of the LV end-diastolic (LVEDD) and
end-systolic diameter (LVESD) were obtained, and the ejection fraction
(EF) was calculated using the modified Quinones method whenever
possible.10
The EF was visually estimated when no M-mode
measurement could be obtained.10
11
Diastolic relative
wall thickness was calculated as the ratio of twice the posterior wall
thickness to LVEDD.12
13
On the basis of published
data,6
we stratified patients by sex according to a
relative wall thickness > 0.66 vs
0.66 to examine outcomes. LV
mass was calculated from the M-mode measurements by the method of
Devereux and Reichek.14
LV hypertrophy was defined using
two different thresholds for the LV mass index: published values from
the Mayo Clinic,15
> 132 g/m2 for
women and > 144 g/m2 for men, and the criteria
of Devereux et al,16
> 110 g/m2
for women and > 134 g/m2 for men.
Aortic valve area and mean gradient were assessed by two-dimensional and Doppler echocardiography with established techniques.17 18 The recording of the optimal transvalvular Doppler signal was ascertained by systematic recordings from all ultrasound windows. The aortic valve area was calculated using the continuity equation using the maximal transstenotic velocity recorded by continuous-wave Doppler; the velocity in the LV outflow tract recorded by pulsed-wave Doppler; and the LV outflow tract diameter measured from the parasternal long-axis view.17 18 Severe AS was defined as an aortic valve area < 0.7 cm2.
Overweight status was defined as a body mass index (BMI)
27.3
kg/m2 in women and
27.8
kg/m2 in men.19
The Charlson
index,20
21
a validated summary method of classifying
comorbidity to predict short- and long-term mortality from medical
records, was used to measure the severity of comorbid illnesses.
Coronary angiography was performed in all patients as part of the preoperative evaluation.
Follow-Up
All participants in the study had provided prior approval for
use of their medical records for research protocols. The protocol and
follow-up survey were approved by our Institutional Review Board.
Postoperative morbidity (defined as sepsis, low cardiac output state, renal insufficiency requiring dialysis, permanent pacemaker implantation, inotropic support > 3 days postoperatively, ventricular arrhythmia requiring therapy, encephalopathy or neurologic disorder, and bleeding requiring reoperation) and mortality (within 30 days or same hospital stay) were examined through detailed review of the medical records. Long-term outcome was evaluated by a follow-up questionnaire designed by our Survey Research Center, focusing on the functional status of the patient. Nonresponders received a follow-up phone interview. Death certificates were obtained for all patients who died after hospital discharge. Late-outcome deaths occurring after hospital discharge were categorized as cardiac if attributable to myocardial infarction or congestive heart failure or if death occurred suddenly and unexpectedly. All other deaths during the follow-up period were defined as noncardiac.
Statistical Analysis
The data are presented in the text and tables as mean ± SD for
normally distributed variables or medians and interquartile ranges in
the case of nonnormal distributions (age, body surface area, and BMI).
LV geometry was compared across the sexes using crude values, indexed
values (divided by the body surface area), and adjusted values using
multiple linear regression analysis to adjust for sex and body surface
area. Comparisons of the baseline characteristics between men and women
were performed using the two-sample t test or the Wilcoxon
rank sum test for continuous variables and by the
2 or Fishers Exact Test for categorical
variables.
In comparing the effect of perioperative complications on perioperative
survival, the Pearson
2 test was used instead
of survival analysis owing to the small number of deaths and the short
follow-up time.
Kaplan-Meier curves were constructed to estimate survival after aortic valve replacement. Cox proportional hazards models were constructed to examine the association between several preoperative clinical and echocardiographic variables and death. Significance was judged at the level of p = 0.05.
| Results |
|---|
|
|
|---|
|
LV Function and Geometry
There were significant differences in LV function between men and
women despite a similar degree of AS (Table 2
). In particular, EF and cardiac index
were all higher in women.
|
No sex differences were observed in the septal or posterior wall thickness between the two groups. LV mass was lower in women, but after indexing, the LV mass index was similar. Fifty-one percent of the women and 49% of men (p = 0.76) had LV hypertrophy using laboratory-based criteria. Using the criteria of Devereux et al,16 53% of the women and 49% of men had LV hypertrophy (not significant).
Twelve percent of women vs 8.5% of men had a relative wall thickness > 0.66 (p = 0.46).
Five patients had missing LVESD and 59 patients had missing LVESD and LVEDD. The analyses were repeated excluding the patients with missing diameters, and the results were unchanged.
Outcome
Mean follow-up was 1,269 ± 320 days. Seven patients died within
30 days or during the same hospital stay after aortic valve
replacement. There was no sex difference in perioperative mortality
(Table 3
). A total of 19 late deaths were
noted: 10 in women and 9 in men (p = 0.98). The causes of late death
are listed in Table 3
.
|
There were no sex differences in Kaplan-Meier survival curve estimates (p = 0.9) (Fig 1 ). The 5-year Kaplan-Meier survival estimates were 82% for women and 79% for men.
|
In the subgroup of patients with a relative wall thickness > 0.66, there was no sex-related difference in long-term mortality. Two of 11 women (18%) and one of seven men (14%) died (p = 0.8).
| Discussion |
|---|
|
|
|---|
LV Geometry
Several studies have indicated that women with severe AS are more
likely than men to respond to the pressure load with concentric
hypertrophy1
2
6
22
and hyperdynamic (supernormal) LV
systolic function. Table 4
summarizes
selected findings of some of these studies. The series differ in terms
of design, number of patients, and prevalence of coronary artery
disease. Not unexpectedly, they also differ with regard to the
prevalence of sex differences in LV geometry.
|
Systolic and diastolic cavity dimensions were also uniformly smaller in women. Although there was no sex difference in wall thickness, LV mass was greater in men than in women. This is not unexpected because the calculation for LV mass incorporates the cavity dimensions, which are higher in men. The studies differ with regard to the existence of sex differences in indexed values. This may reflect differences in body habitus among the three studies. The use of indexed values to adjust echocardiographic dimensions remains controversial.23 Although we elected to present index values in the present study, when linear regression analysis was used, no sex difference in LVEDD was found in our series. This further underscores the limitation of controlling for body size by adjusting for body surface area.
The prevalence of LV hypertrophy using predefined as well as laboratory-specific criteria did not differ according to sex in the present study, which is at odds with the findings reported by Douglas et al.22 In that series, however, the population was older and had a greater prevalence of heart failure, which is likely to affect LV geometry and could conceivably confound some of the sex-specific findings. The lack of control for hypertension may also compromise the interpretation of these findings. In an exploratory analysis of the present series, excluding persons with a history of hypertension resulted in nonclinically significant changes of the sex differences in LVEDD, EF, and cardiac index. With regards to LV mass index, excluding patients with hypertension indicated a significantly lower LV mass index in women (130.7 ± 35.6 in women vs 151.7 ± 30.2 g/m2 in men; p = 0.043), adding more support toward the lack of an excessive degree of hypertrophy in women as compared with men. Although these results have to be interpreted with caution inasmuch as they were obtained from secondary analyses, survival analysis stratified by hypertension status did not unmask any sex difference in outcome. Thus, although there is some consistent evidence across published studies that LV function indexes in severe AS differ according to sex, some uncertainties remain with regard to sex-specific patterns of LV geometry (mass index, relative wall thickness, prevalence of LV hypertrophy) associated with severe AS in clinical practice.
Outcome
Regardless of their interpretation, the sex differences discussed
above pose an important clinical question because some series have
reported worse operative mortality for valve replacement in patients
with concentric LV hypertrophy, small cavity size, and hyperdynamic LV
systolic function.2
6
Inasmuch as these findings are more
prevalent in women, they raise the concern of possible excess operative
mortality among women with surgical AS. Yet there is little data on the
outcome associated with the reported sex-differences in LV
geometry.2
3
5
22
The outcome of the present population does not support this concern of an excess operative mortality among women inasmuch as no sex difference in postoperative or long-term outcome was noted. In contrast to other reports,8 9 our study represents a group of patients with isolated AS. Excluding coexisting coronary artery disease is important to assist in the interpretation of the outcomes, because its presence may adversely and independently affect outcome.2 6 Indeed, revascularization adds a small increase in the operative risk of aortic valve replacement, which could conceivably apply more to women.9 It has been suggested that because women are older at the time of revascularization, have smaller target vessels, and have more comorbid illnesses, this increases mortality rates from revascularization procedures compared with men.24 Finally, in some of these series, the operative mortality for valve replacement was 22%,6 which is higher than the reported average mortality for aortic valve replacement in other studies of 4.2 to 6.6%,9 25 26 27 28 raising the concern about selection bias and limited relevance to the general population.
The present series differs from previous reports in that it included a large number of consecutive patients who underwent aortic valve replacement for AS during the study period without associated coronary disease. The evaluation of patients with AS is routinely conducted using Doppler echocardiography,29 which protects from the selection biases outlined above and positively impacts the generalizability of our data. The operative mortality in the present series is similar to previously reported figures.
Limitations
This study was retrospective, and it is conceivable that some
determinants of outcome were not captured at aortic valve replacement.
In this surgically defined population, the decision made by the
physician to refer the patient for aortic valve replacement may have
been influenced by factors possibly confounding the association between
sex or other variables and outcome.
Some determinants of LV geometry were not captured in our series. In particular, no measure of wall stress was obtained. However, all clinically used indexes of LV geometry were included. Although our study showed that women had a higher EF and a larger LVEDD index, and that > 50% of the women had LV hypertrophy, intracavitary flow acceleration was not routinely assessed. For consistency, we elected to use measurements similar to those used in previous studies; however, LV volume measurements are important in the analysis of LV geometry and should be the subject of future studies. Volumetric measures of the LV would have been desirable but were not performed routinely in our practice during the study period.
It has been suggested that a relative wall thickness
0.66 is
associated with a trend toward increased mortality.6
No
increased mortality was associated with a relative wall thickness
0.66 in our series, but the series may lack power for this
analysis.
| Summary |
|---|
|
|
|---|
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication April 15, 1999. Accepted for publication October 5, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Doenst, J. Ivanov, M. A. Borger, T. E. David, and S. J. Brister Sex-specific long-term outcomes after combined valve and coronary artery surgery. Ann. Thorac. Surg., May 1, 2006; 81(5): 1632 - 1636. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. C. Sharma, P. Barenbrug, S. Pokharel, W. R. M. Dassen, Y. M. Pinto, and J. G. Maessen Systematic review of the outcome of aortic valve replacement in patients with aortic stenosis Ann. Thorac. Surg., July 1, 2004; 78(1): 90 - 95. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |