Does diabetes accelerate progression of calcific aortic stenosis?

European Journal of Echocardiography (2009) 10, 723–725 doi:10.1093/ejechocard/jep048

CLINICAL/ORIGINAL PAPERS

Does diabetes accelerate progression of calcific aortic stenosis? Masoor Kamalesh*, Charlotte Ng, Hicham El Masry, George Eckert, and Stephen Sawada Krannert Institute of Cardiology and Roudebush VA Medical Center, Indiana University, 1481 W 10th Street, Indianapolis, IN 46202, USA Received 7 February 2009; accepted after revision 10 April 2009; online publish-ahead-of-print 30 April 2009

KEYWORDS Diabetes; Atherosclerosis; Aortic stenosis

Diabetes and progression of calcific aortic stenosis Calcific aortic valve stenosis (CAS) is now considered to be an active disease like atherosclerosis.1,2 Chronic inflammation and lipoprotein deposition has been observed on histopathological examination of these valves.3 Traditional risk factors for atherosclerosis such as male gender, age, hypertension, smoking, and elevated cholesterol are also closely associated with CAS.2,4 Despite the strong association of diabetes (D) with coronary atherosclerosis, its association with CAS has been variable2,5–9 Further, there is very limited longitudinal data on the effect of D on CAS progression over time. The primary purpose of this study was to assess the effect of D on progression of CAS. A secondary analysis was done to assess the effect of statin therapy on progression of CAS.

Methods We performed a retrospective analysis of consecutive patients with clinically suspected CAS referred to our tertiary care teaching

* Corresponding author. Tel: þ1 317 312 1472; fax: þ1 317 554 0230.

hospital between January 1997 to March 2005 for routine echocardiography. The study cohort included 166 (99%) males with age 70 + 9 years. The mean follow up duration was 2.5 years. All patients who had serial echocardiograms at least 3 months apart were included in the study. All studies were done using Acuson Sequoia echocardiography equipment. Patients were studied with the use of a 3 MHz transducer for M-mode (single-dimensional) and pulsed- and continuous-wave Doppler scanning. All measurements were determined based on previously recorded studies, averaged from three measurements, and recorded digitally according to a standard protocol. Peak and mean aortic-valve pressure gradients were calculated with the Bernoulli equation, and aortic-valve area was calculated using the continuity equation.10 Aortic valve area was measured using the continuity equation. Severity of CAS was classified based on valve area. CAS severity was categorized using valve area as mild ( 1.5 cm2), moderate (1–1.5 cm2), or severe (,1 cm2), as per 2006 American College of Cardiology/American Heart Association guidelines (http://www.acc.org/clinical/guidelines/valvular/index.pdf://www.acc.org/clinical/guidelines/valvular/index.pdf). D and non-D patients were compared for differences in gender, prevalence of hypertension, smoking, statin use, and angiotensin converting enzyme (ACE)-inhibitor use using x2 tests; for differences in baseline aortic stenosis severity using Mantel-Haenszel test for ordered categorical data; and for differences in age, baseline LDL,

E-mail address: [email protected]

Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2009. For permissions please email: [email protected].

Downloaded from by guest on March 23, 2016

Aims Calcific aortic valve stenosis (CAS) is an active disease like atherosclerosis. Effect of diabetes (D) on severity of CAS is not well documented. Methods and results We retrospectively analysed 166 consecutive patients with CAS and multiple echocardiograms from January 1997 to March 2005. Aortic valve area (AVA) was measured using the continuity equation. CAS severity was categorized using AVA. D and non-D patients were compared for differences in sex, hypertension, smoking, statin use using x2 tests. Comparisons between D and nonD for changes in AVA per year were performed using ANOVA. Study cohort included 166 males with age 70 + 9 years, of which 72 (43%) had D. Baseline CAS was mild in 66 subjects, moderate in 75, and severe in 25. D subjects smoked less (P ¼ 0.02), but all other variables were similar (P . 0.05). The interaction between D and baseline CAS severity was significant (P ¼ 0.0191), indicating comparisons should be viewed by baseline CAS severity. D had significantly larger change in AVA than non-D (P ¼ 0.0016) for those with moderate CAS at baseline only. Adjusting for statin use did not alter the results. Conclusion CAS severity progresses faster in D than in non-D in subjects with moderate CAS at baseline. Statins do not affect progression of CAS.

724

M. Kamalesh et al.

Table 1 Patient characteristics

Age, mean (SD) Male, n (%) HTN, n (%) ACE-I, n (%) Statin, n (%) LDL, mean (SD) AS severity, n (%) Mild Moderate Severe AS area, mean (SD) Smoking, n (%) Follow-up years, mean (SD)

All

Non-D

D

P-value

69.7 (8.9) 164 (99) 141 (85) 130 (78) 103 (62) 112 (36)

69.0 (9.9) 93 (99) 79 (84) 69 (73) 58 (62) 114 (36)

70.6 (7.4) 71 (99) 62 (86) 61 (85) 45 (63) 109 (35)

0.27 0.85 0.71 0.08 0.92 0.29

66 (40) 75 (45) 25 (15) 1.45 (0.50) 51 (31) 2.51 (1.57)

34 (36) 44 (47) 16 (17) 1.41 (0.49) 36 (38) 2.63 (1.60)

32 (44) 31 (43) 9 (13) 1.50 (0.50) 15 (21) 2.34 (1.53)

0.24

Table 2 Changes in aortic valve area (cm2/year): mean (SD) All

All Mild Moderate Severe

20.22 20.28 20.19 20.17

(0.21) (0.24) (0.17) (0.18)

Non-D

D

20.20 (0.19) 20.28 (0.24) 20.14 (0.13) 20.17 (0.12)

20.26 20.29 20.25 20.17

(0.23) (0.25) (0.20) (0.27)

baseline aortic valve area, and follow up time using two-sample t-tests. Because the distribution of the baseline aortic valve area was not normally distributed, a log transformation of the baseline aortic valve area was used when comparing D and non-D patients. The change in aortic valve gradient (peak and mean) and area per year was computed to account for the variability in time between echocardiograms for each patient. Differences in the change in aortic valve area were compared between D and non-D using ANOVA that included terms for statin use, diabetes status, baseline aortic stenosis severity, and the interaction between diabetes and severity. Analyses were also performed by statin use, due to assess the potential effect of statin therapy on CAS progression. Because the distribution of the change in aortic valve area was not normally distributed, a log transformation of the change was used for the analyses. For the D group, glycosylated haemoglobin levels were obtained at baseline and follow up echo.

Results Of the 166 subjects, 72 subjects (43%) had D. Baseline CAS was mild in 66 subjects, moderate in 75, and severe in 25. Clinical characteristics were comparable between D and non-D patients at baseline (Table 1) except a lower percentage of D were smokers (P ¼ 0.01). Table 2 shows the mean changes in aortic valve area for all patients and for the D and non-D cohorts based on stenosis severity at baseline. The interaction between D and baseline aortic stenosis severity was significant (P ¼ 0.02), indicating comparisons between D and non-D should be viewed by baseline aortic stenosis severity. For the cohorts with mild stenosis and severe stenosis at baseline, there was no significant difference in the progression of stenosis between D and non-D (P ¼ 0.98 for mild and 0.34 for severe). For those with moderate stenosis at baseline, D had faster progression of stenosis than non-D (P ¼ 0.0016) (Table 2). The annual rate

Table 3 Aortic valve gradient changes (in mmHg) in moderate stenoses group Peak gradient—D Baseline mean 39.7, SD 15.7, range 16.4–81.1 Follow-up mean 47.5, SD 23.6, range 13.4–125.0 Change per year mean 4.5, SD 13.1, range 225.2 to 28.5 Peak gradient—non-D Baseline mean 35.6, SD 13.5, range 17.8–85.3 Follow-up mean 47.2, SD 19.5, range 12.8–94.1 Change per year mean 3.8, SD 7.2, range 223.7 to 24.2 Mean gradient—D Baseline mean 20.5, SD 8.1, range 9.0–43.1 Follow-up mean 25.7, SD 14.8, range 4.6–81.6 Change per year mean 3.5, SD 8.6, range 213.7 to 23.1 Mean gradient—non-D Baseline mean 18.4, SD 8.6, range 6.8–57.3 Follow-up mean 26.9, SD 18.6, range 8.2–128.0 Change per year mean 2.6, SD 6.1, range 26.6 to 35.7 Comparisons Mean gradient change per year between D and non-D, P ¼ 0.62

of peak and mean gradient change per year was higher for D, although it did not reach statistical significance (Table 3). The mean follow up time was different for the three groups. The longest follow-up was in the mild severity group—median of 2.93 years, followed by 2.40 years for moderate severity group, and 1.69 years for the severe group. Mean glycoyslated haemoglobin levels at baseline were 7.4% (SD 1.8, range 5.0–14.7%) and at follow up 7.2% (SD 1.6, range 5.0–12.0). Similar results were obtained when analyses were performed separately for patients using statin and patients not using statin therapy. For those with moderate stenosis at baseline, D has faster progression of stenosis than non-D whether they were taking a statin (P ¼ 0.02) or not (P ¼ 0.02). For those with mild or severe stenosis at baseline, there was no significant difference in progression of stenosis between D and non-D for those taking a statin (P ¼ 0.75 for mild and 0.94 for severe) or not (P ¼ 0.73 for mild and 0.29 for severe).

Discussion Our data reveal that diabetes accelerates progression of CAS in subjects who have moderately severe CAS at

Downloaded from by guest on March 23, 2016

AS severity

0.18 0.0156 0.24

Diabetes and aortic stenosis

opposed to prior retrospective studies, and this may need to be factored in future larger prospective trials. Limitations of our study need to mentioned. First, it is a retrospective analysis and thus suffers from all drawbacks of such a study including bias in patient selection and follow-up. This is especially important because prior studies in CAS have show how retrospective and prospective studies can differ in their results. Thus, our study needs to be validated by prospective evaluation. Second, the patient numbers particularly in subgroups are relatively small. Future studies should be conducted in larger numbers of subjects. Our study subjects were almost exclusively males. Thus, extrapolating these conclusions to women subjects need to be done with caution.

Conclusions Our study shows that diabetes accelerates progression of CAS. This effect is a function of baseline severity of CAS. Statin therapy does not appear to have an effect of progression of CAS in D. These data are helpful in guiding frequency of follow-up to patients with aortic stenosis with serial echocardiograms and in selecting patients for potential therapies that may delay CAS progression. A recent report concludes that patients with asymptomatic haemodynamically significant CAS become symptomatic within 5 years.12 Diabetic subjects with moderate stenosis at baseline may need intensive follow-up for their CAS than non-D based on our study results.

References 1. Roberts WC. The senile cardiac calcification syndrome. Am J Cardiol 1986;58:572–4. 2. Freeman RV, Otto CM. Spectrum of calcific aortic valve disease. Pathogenesis, disease progression, and treatment strategies. Circulation 2005;111:3316–26. 3. Otto CM, Kuusisto J, Reichenbach DD et al. Characterization of the early lesion in degenerative valvular aortic stenosis: histological and immunohistochemical studies. Circulation 1994;90:844–53. 4. Peltier M, Trojette F, Sarano ME et al. Relation between cardiovascular risk factors and non-rheumatic severe calcific aortic stenosis among patients with a three cuspid aortic valve. Am J Cardiol 2003;91:97–9. 5. Katz R, Wong ND, Kronmal R et al. Features of the metabolic syndrome and diabetes mellitus as predictors of aortic valve calcification in multiethnic study of atherosclerosis. Circulation 2006;113:2113–9. 6. Lindroos M, Kupari M, Valvanne J et al. Factors associated with calcific aortic valve degeneration in the elderly. Eur Heart J 1994;15:865–70. 7. Nassimiha D, Aronow WS, Ahn C, Goldman ME. Association of coronary risk factors with progression of valvular aortic stenosis in older persons. Am J Cardiol 2001;87:1313–4. 8. Stewart BF, Siscovick D, Lind BK, Gardin JM, Gottdiener JS, Smith VE et al. Clinical factors associated with calcific aortic valve disease. Cardiovascular Health Study. J Am Coll Cardiol 1997;29:630–4. 9. Agmon Y, Khandheria BK, Meissner I, Sicks JR, O’Fallon WM, Wiebers DO et al. Aortic valve sclerosis and aortic atherosclerosis: different manifestations of the same disease? Insights from a population-based study. J Am Coll Cardiol 2001;38:827–34. 10. Oh JK, Seward JB, Tajik AJ. Valvular Heart Disease. The Echo Manual. Lippincot, Williams & Wilkins; 1994. p87–113. 11. Aronow W, Ahn C, Kronzhon I, Goldman M. Association of coronary risk factors and use of statins with progression of mild valvular aortic stenosis in older persons. Am J Cardiol 2001;88:693–5. 12. Pellikka PA, Sarano ME, Nishimura RA et al. Outcome of 622 adults with asymptomatic hemodynamically significant aortic stenosis during prolonged follow up. Circulation 2005;111:3290–5.

Downloaded from by guest on March 23, 2016

baseline. We also found no effect of statin therapy on progression of CAS. There is lack of data on the longitudinal effect of diabetes on progression of severity of CAS.2 In the only other study we could find in the English literature, Aronow et al.11 report on 180 older subjects (mean age 82 years) with mild aortic stenosis at baseline. There were 48 subjects with D. Mean follow up period was 33 months. The mean annual progression in peak systolic gradient was 5.7 mmHg in the D group compared with 4 mmHg in the non-D group. Although the difference appears to be of minor clinical significance, it was statistically significant. Valve area was not reported. In our study, valve area was used to assess stenosis progression. Continuity equation valve area may be a more reliable way to assess the changes in stenosis severity when compared with peak gradient. Changes in contractility and stroke volume may result in significant changes in peak gradient in the absence of changes in stenosis severity.10 In our study, D accelerated progression of disease only in patients with moderate stenosis at baseline. The explanation for this finding is speculative. Changes in valve area were most pronounced in mild disease, but were nevertheless there identical in patients with and without D (20.29 cm2 valve area vs. 20.28 cm2) and that the difference was only significant in moderate AS (20.25 vs. 20.14 cm2) with again identical progression in severe AS. Furthermore, differences in gradients and the annual rate of progression were not significant even in moderate AS. Patients with severe disease at baseline had the shortest mean duration between echocardiographic studies, and it is possible that, with longer follow up, a difference in the rate of progression between D and non-D may have occured.2 Alternatively, D may not affect progression of disease in those with heavily calcified, and severely stenotic valves. Once the valve is heavily calcified, atherosclerosis and inflammation may play less of a role in progression of stenosis than in a valve with less severely diseased or calcified tissue. Patients with mild disease, despite their longest period of follow up, showed no influence of diabetes on disease progression. Although differences in gradients (both peak and mean) and annual rate of progression was numerically higher among D, this was not statistically significant. It is possible that with larger number of subjects we may see a statistically significant difference. Although the negative results for statin use may appear to be in line with recent randomized studies in CAS, they are nevertheless surprising. These studies with negative results were obtained in populations without cardiovascular risk factors, completely different from the present population. Previous—although again retrospective—studies of patients with indications for statin use showed almost consistently positive results and are in disagreement with the present results. However, these results should be interpreted with caution and need confirmation in larger studies due to small number of subjects in subgroups. Other possibility that may explain the different results in our study when compared with prior retrospective studies may be due to differences in baseline characteristics of patients involved. Better and more aggressive control of cardiovascular risk factors in our study may be a confounding factor as

725