European Heart Journal Advance Access published online on May 5, 2007
European Heart Journal, doi:10.1093/eurheartj/ehm153
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cardiac manifestations of AndersonFabry disease: results from the international Fabry outcome survey
Linhart1,*
1 Second Department of Internal Medicine, First Faculty of Medicine, Charles University, Prague, Czech Republic
2 Department of Pediatrics, University of Mainz, Mainz, Germany
3 Department of Cardiology, Hospital Clínico San Carlos, Madrid, Spain
4 Division of Nephrology and Dialysis, Department of Medicine III, University of Vienna, Vienna, Austria
5 Department of Haematology, Royal Free Hospital, London, UK
6 The Heart Hospital, University College London, London, UK
Received 8 November 2006; revised 7 March 2007; accepted 6 April 2007.
* Corresponding author. Tel: +420 2 2496 2605; fax: +420 2 24912154. E-mail address: alinh{at}lf1.cuni.cz
| Abstract |
|---|
|
|
|---|
Aims: AndersonFabry disease (AFD) is an uncommon X-linked disorder caused by deficient activity of the lysosomal enzyme
-galactosidase A. The Fabry Outcome Survey is a European database designed to monitor the long-term efficacy and safety of enzyme replacement therapy (ERT) with agalsidase alfa. The aim of this study was to determine the prevalence and characteristics of cardiac disease in AFD patients. Methods and results: Clinical and laboratory data were available in 714 patients from 11 countries (mean age 35 ± 17 years, 369 women, 336 treated). The prevalence of angina was 23 vs. 22%; palpitations and arrhythmias 27 vs. 26%; exertional dyspnoea 23 vs. 23%; and syncope 2 vs. 4%, in women and men, respectively (all P = NS). The frequency of all cardiac symptoms was significantly higher in treated than in untreated patients. Gender, age, and glomerular filtration rate were independent determinants of echocardiographically assessed left ventricular hypertrophy (LVH).
Conclusion: This study confirms the high prevalence of cardiac morbidity associated with AFD. The disease burden in treated women exceeds that of untreated men, suggesting that most women selected for ERT have advanced disease. The presence of LVH is associated with higher frequency of cardiac signs and symptoms and relates independently to gender, age, and renal function.
Key Words: AndersonFabry disease Cardiomyopathy Left ventricular hypertrophy Lysosomal storage disease Agalsidase alfa
| Background |
|---|
|
|
|---|
AndersonFabry disease (AFD) is an uncommon X-linked lysosomal storage disorder caused by
-galactosidase A deficiency. It is characterized by progressive accumulation of globotriaosylceramide (Gb3) in various organ systems and a wide variety of progressive signs and symptoms, including acroparaesthesiae, pain crises, gastrointestinal symptoms, angiokeratomata, and corneal dystrophy.1 While many patients die prematurely as a consequence of renal impairment, cerebrovascular disease, and cardiomyopathy, the relative contribution of disease-related abnormalities in different organ systems to long-term morbidity and mortality is unknown because most published data are derived from small and highly selected patient cohorts.2,3 The Fabry outcomes survey (FOS) is a large European registry of male and female patients with AFD that provides a unique opportunity to determine the natural history of AFD in untreated patients and in patients receiving enzyme replacement therapy (ERT) with agalsidase alfa. The aim of this study was to determine the prevalence of cardiovascular disease in men and women with AFD in the FOS database. | Methods |
|---|
|
|
|---|
The FOS database has been approved by the Ethics Institution review board of all participating centres. The diagnosis in all patients was confirmed by enzyme assay and DNA analysis. All patients gave written informed consent prior to data entry. Data collection and analysis in FOS are supported by Shire Human Genetic Therapies. The sponsor had no role in the interpretation of data or writing of the report.
Data collection
Data collection methods in FOS have been described previously.4,5 For the purpose of this study, we analysed baseline data obtained from patients included in the survey from its beginning in 2001 till October 2005. For inclusion, a confirmed diagnosis of AFD was required. Only untreated patients or patients treated with agalsidase alfa were included. Anonymised data are submitted electronically by participating physicians to the central FOS database. The following anamnestic data were recorded: dyspnoea (NYHA functional class), chest pain, a history of syncope, palpitations, documented arrhythmias, atrioventricular conduction abnormalities, the presence of a permanent pacemaker, previous myocardial infarction, coronary revascularization procedures, and known valve disease. Glomerular filtration rate (eGFR) was estimated from baseline serum creatinine values using the four-variable MDRD formula.6 A sub analysis of the cause of death in relatives with known or suspected AFD was also performed.
Echocardiography
Echocardiographic data were collected in accordance with pre-specified guidelines contained within FOS. Measurements were performed according to the American Society of Echocardiography recommendations. The following parameters were derived from M-mode recordings: left ventricular end-diastolic diameter (LVEDd), left ventricular end-systolic diameter (LVESd), interventricular septum thickness (IVSd), and posterior wall thickness (PWd) in end-diastole. LV mass was calculated using the Devereux-modified cube formula.7 Left ventricular hypertrophy (LVH) was defined as an LVM indexed to height2.7 (g/m2.7) greater than 50 g/m2.7.4,8,9
Statistical analysis
Data were analysed using SAS statistical software, version 9.1 (SAS, Carry, NC). Students two-sample two-sided tests were used for univariate comparisons of continuous variables and Cochran MantelHaenszel
2 test for categorical variables. The confidence intervals were two-sided with an alpha level of 5%. Univariate correlations were analysed using Spearman rank correlations. The relationship between LVH and clinically relevant variables including age, gender, renal function, and blood pressure was analysed using logistic regression models. All values are presented as mean ± SD. A P-value < 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
Patient's characteristics
By October 2005, the FOS database contained data on 752 patients from 11 European countries [Germany10 centres (n = 162), 21.5%; UKthree centres (n = 121), 16.1%; Czech Republicone centre (n = 85), 11.3%; France28 centres (n = 75), 10.0%; Italyeight centres (n = 74), 9.8%; Switzerlandthree centres (n = 62), 8.2%; Spain17 centres (n = 57), 7.6%; Norwayone centre (n = 36), 4.8%; Belgiumfour centres (n = 32), 4.3%; Swedentwo centres (n = 24), 3.2%; Austriafour centres (n = 24), 3.2%].
Complete clinical data were available in 714 (95%) (369 women and 345 men). The characteristics of this population are shown in Table 1.
|
Cardiac symptoms
The prevalence and age of onset of individual signs and symptoms in untreated patients and in patients receiving ERT at the time of inclusion into the FOS registry are shown in Table 2. There were no patients with exclusively cardiac involvement
|
In treated patients, the frequency of all cardiac symptoms was similar in both genders, but the age of onset was delayed in women (except for syncope, myocardial infarction, and pacemaker implantation) (Tables 2 and 3, Figure 1).
|
|
Untreated women had a similar frequency of all individual cardiac signs and symptoms as untreated men, and were of similar age at onset of most symptoms (except for LVH which was more prevalent in younger men and palpitations and arrhythmias that occurred in men at earlier age).
Table 3 shows the frequency of major clinical cardiac events and findings including LVH and valvular heart disease, as assessed by investigators. Of note, the history of myocardial infarction was reported in nine women (mean age 59 ± 8 years) and four men (mean age 48 ± 3 years). Similarly, revascularization history was noted in five patients only (three women and two men). Although valve disease was reported in 12% of women and 17% of men, only three (0.4%) had undergone valve surgery.
Left ventricular hypertrophy
Baseline echocardiographic data were available in 480 patients (304 untreated and 176 treated). The prevalence of LVH increased with age in treated and untreated patients (Figure 2). The overall frequency of LVH in untreated patients was 33% in women and 53% in men. The mean age of patients with LVH was 54 ± 13 years in untreated women and 45 ± 9 in untreated men (P < 0.001) and 43 ± 17 years in treated women and 35 ± 14 years in treated men (P < 0.001). The presence of LVH was associated with a significantly higher frequency of cardiac symptoms, arrhythmias, and valvular disease (Figure 3).
|
|
Left ventricular mass correlated with age both in men and women (Figure 4). There was a negative relation between left ventricular mass and eGFR (Figure 5). There was a significant correlation between LV mass and systolic and diastolic blood pressure in women (r = 0.49, P < 0.001 and r = 0.27, P < 0.001, respectively) but not in men (r = 0.06, P = 0.61 and r = 0.06, P = 0.61, respectively). After correction for age, in women this relation remained for systolic blood pressure (partial r = 0.24, P = 0.004) but not for diastolic blood pressure (partial r = 0.01, P = 0.86). In treated women, only non-significant correlations were observed (r = 0.21, P = 0.09 and r = 0.09, P = 0.46, for systolic and diastolic pressure, respectively). Similar observation was found in treated men (r = 0.08, P = 0.53 and r = 0.16, P = 0.19, for systolic and diastolic blood pressure, respectively).
|
|
Multivariate logistic regression analysis demonstrated that in untreated patients, gender, age, renal function, but not blood pressure were independently related to LVH. In treated patients, the odds ratio of estimated renal function did not reach the statistical significance (Table 4).
|
Cause of death in relatives
The cause of death was recorded in 113 affected relatives (41 women and 72 men). The analysis was based solely on patients description. For the purpose of the study, cardiac death was defined as sudden death or death related to heart failure, myocardial infarction, or infective endocarditis. Cardiac death was the most frequent reported cause in females (n = 12; 29.3% of all deaths in females). In males, 21 cardiac deaths were recorded (29.2% of all deaths in males), six of which were in patients with end-stage renal disease on haemodialysis or after kidney transplantation (five cases of heart failure or arrhythmia on haemodialysis and one case of endocarditis). The most frequent cause of death in males was end-stage renal failure and related problems, noted in 32 patients (44.4% of all deaths in males). The mean age of death from cardiac causes was 61 ± 8 years in women and 47 ± 9 years in men (P = 0.001), slightly higher than overall age of death (58 ± 14 years in women and 45 ± 10 years in men).
| Discussion |
|---|
|
|
|---|
This study demonstrates that cardiac disease causes substantial morbidity in men and women with AFD; the high prevalence of cardiac deaths in affected relatives suggests that it also contributes to the reduced life expectancy of patients with AFD. While there are inevitable referral biases in the study population, the fact that participating investigators were encouraged to enter all patients (including asymptomatic heterozygous females) with AFD in their care, regardless of the disease severity and treatment status, suggests that the patients recorded in FOS represent the closest approximation to date of the true spectrum of this disease. One consequence of this is that the frequency of cardiac symptoms in this study is lower than that reported previously.2,3
Coronary artery disease
AFD has been suggested as a risk factor for premature atherosclerotic coronary artery disease,10 but in spite of a high frequency of anginal chest pain in both men and women, there were relatively few records of coronary revascularization procedures or myocardial infarction. The high incidence of coronary disease in earlier reports is probably explained by patient selection and misattribution of ECG changes in patients with a history of chest pain. This is not to say, however, that patients with AFD do not experience myocardial ischaemia. Patients with LVH were more likely to complain of chest pain, suggesting that factors such as increased myocardial oxygen demand, small vessel disease, and endothelial dysfunction that cause microvascular angina in patients with other causes of LVH are also important in AFD.11,12 Similarly, we cannot exclude the possibility that AFD exacerbates cardiovascular risk in patients with other predisposing factors such as hyperlipidaemia, hypertension, and cigarette smoking.
Arrhythmia
The high frequency of arrhythmia and conduction abnormalities confirms the recent observation of atrial and ventricular arrhythmia in older patients with AFD.13 Atrial arrhythmias may be caused by substrate deposition within the atria and by dilatation of the atria due to elevated ventricular filling pressures. Myocardial fibrosis, seen in many patients on cardiac magnetic resonance imaging14 predisposes to ventricular arrhythmias. In this study, a history of arrhythmia was more frequent in patients with LVH.13 While the clinical importance of arrhythmia in AFD is yet to be determined, it is likely that they cause syncope (present in 3.2% of patients in this study) and may be one mechanism underlying stroke and transient ischaemic attacks in AFD.
Valve disease
Previous studies have suggested that the cardiac valves are frequently involved in AFD with infiltration by sphingolipids (Gb3) and secondary fibrosis and calcification.15 In this study, valve disease was reported in 14.6% of patients, but the low prevalence of valve surgery at all ages suggests that the lesions produced by Gb3 storage are rarely of haemodynamic significance.8,16
Deaths in affected relatives
As FOS was only established in 2001, there are limited follow-up data available on which to base a survival analysis (only eight deaths recorded). Therefore, in order to explore the possible effect of cardiac disease on survival, we performed an analysis of the cause of death in affected relatives. While there are obvious limitations of this methodology, the data suggest that cardiac disease is the leading cause of mortality in women and the second most frequent in men. Of note, the mean age of death among the relatives was even lower than reported in a smaller study that examined survival in relatives.2,3 It is also likely that cardiac disease contributes to the dismal prognosis of patients with AFD and end-stage renal failure.
Predictors of cardiac involvement
As AFD is an X-linked disease, gender has a major influence on cardiac involvement. As expected, the prevalence of LV hypertrophy was higher in men, but the frequency of LVH increased exponentially with age in males and females. In addition, the prevalence of LVH and cardiac symptoms in treated women was similar to that seen in male patients. These observations illustrate the inability of cells with an activated wild type X chromosome to cross-correct the metabolic defect present in cells with a mutated allele.16 Novel observations in this study include the correlation between LVH and blood pressure in untreated women, which was independent of age, and the demonstration that renal function was an independent predictor of the presence of LV hypertrophy. Although hypertension or renal impairment could have contributed to an increase in LV mass, it is likely that all three of these clinical parameters simply reflect the overall severity of the disease. Further studies are required to determine whether stabilization of renal function and stringent control of blood pressure alone delay the onset of LV hypertrophy in AFD.
Study limitations
As in any multicentre registry, a degree of ascertainment bias is very likely in FOS. For example, mildly affected patients are less likely to be diagnosed and therefore included in the database. Conversely, symptomatic patients are more likely to have complete cardiac evaluations.
Another limitation is the potential variability of echocardiographic measurements made in different centres which, together with indirect measurement of eGFR, could have influenced the correlation between LV mass and eGFR. Similarly, the use of casual blood pressure measurements may have underestimated the influence of hypertension on LV mass.
Conclusions
Accepting some inevitable limitations, the data presented in this paper represent the largest and most detailed cross-sectional description of cardiac involvement in AFD. Our data suggest that cardiovascular involvement contributes substantially to disease-related morbidity and mortality in men and that it may be the major cause of premature death in heterozygous females. Assessment of the severity of heart involvement is therefore mandatory in all patients with AFD and, if present, it should be aggressively treated using conventional therapies. The impact of ERT on cardiovascular manifestations of AFD is still under investigation.
| Acknowledgements |
|---|
|
|
|---|
FOS European Investigators Group
The data in this study were provided by the European FOS Investigators Group:
Austria: Binder C, Kotanko P, Kroepfl T, and Plecko B (Graz); Bodamer O, Hauser A-C, Kleinert J, and Sunder-Plassmann G (Vienna). Belgium: Clerbaux G, Georges B, Nassogne MC, and Pirson Y (Brussels); Dehout F, Roland D, Van Maldergem L, and Vauthier L (Charleroi); Goyens P (Liège); Eyskens F (Middelheim). Czech Republic: Bultas J, Karetová D, Linhart A, Lubanda J-C, and Magage S (Prague). France: Choukroun G (Amiens); Berthelot J (Angers); Cairey-Reomonnet A-S (Besançon); Lacombe D (Bordeaux); Benziane S (Cambrais); Khau Van Kien A (Montpellier); Mittelberger JM (Freyming Merlebach); Dobbelaere D (Lille, Jeanne de Flandres); Hachulla E (Lille, C Huriez); Dussol B (Marseille); Reade R (Maubeuge); Kaminsky P (Nancy); Guyot C (Nantes); Jaeger P (Nice); Germain D (Paris, HEGP); Lidove O (Paris, Bichat); Monlun E (Pau); Jaussaud R (Reims, R Debré); Richalet B (Saint Lô); Andres E (Strasbourg); and Caraman D (Thionville); Bazex J (Toulouse); Ouali N (Paris, Tenon); Touati G (Necker); Bataille P (Boulogne sur mer); Hardy P (Arras); Perrichot R (Vannes). Germany: von Arnim-Baas A and Hennermann J (Berlin); Cybulla M, Walter K, and Neumann HPH (Freiburg); Gal A and Schäfer E (Hamburg); Das A and Illsinger S (Hannover); Baron K, Bähner F, Beck M, Bruns K, Delgado-Sànchez S, Hartung R, Kampmann C, Keilmann, A, Lackner K, Pitz S, Schwarting A, and Whybra C (Mainz); Hoffmann B and Koletzko B (Munich); Böttcher T and Rolfs A (Rostock). Italy: Gabrielli O and Salvatori IF (Ancona); Strisciuglio P and Concolino D (Catanzaro); Borsini W and Buchner S (Florence); Menni F and Ravaglia R (Milan); Parini R and Santus F (Monza); Di Vito R (Ortona); Burlina Al, Burlina Alb, and Tognana G (Padova); Antuzzi D, Castorina M, Di Lillo M, Feriozzi S, and Ricci R (Rome). Norway: Bindoff LA, Bostad LH, Haugen OH, Hirth A, Houge G, Lægreid LM, Neckelmann G, Strømsvik N, Svarstad E, Thune TJ, and Tøndel C (Bergen); and Skarbøvik A and Kaarbøe Ø (Ålesund). Spain: Barba MA (Albacete); Botella R (Alicante); Gómez Huertas E (Central Asturias); Herrera J (General Asturias); Ara J, Bonal J, and Pintos G (Badalona); Ballarin J and Torra R (F. Puigvert-Barcelona); Torras J (Bellvitge-Barcelona); Torregrosa V (Clinic-Barcelona); González J (Cadiz); García M, Herrera C, Martin I, and Rodriguez J (Huelva); Cano Ruíz A (Madrid); Barbado FJ, García-Consuegra J, García de Lorenzo A, and López M (La Paz, Madrid); Paniagua J and Rodriguez F (Ponferrada); Hernández S (Linares); Fernández V (Santiago); Andreu J (Seville); Febrer I and Perez García A (Valencia); and Rivera A (Vigo). Sweden: Öqvist B and Green C (Lund); Dahlman, Ingrid (Stockholm). Switzerland: Ferrari P and Vogt B (Bern); Barbey F and Theytaz J (Lausanne); Schulthess G, Walter K, and Widmer U (Zurich). United Kingdom: Cox TM, Deegan P, Ramaswami U, Wright N, Allen L, Bol A, Dutka D, and Galt O (Cambridge); Baker R, Blincoe M, Bruce R, Burns A, Cadge B, Davey C, Elliott J, Elliott PM, Evans S, Ginsberg L, Hajioff D, Hughes D, Ioannidis A, Keshav S, Mehta A, Milligan A, Orteu C, Richfield L, and Shah J (London).
Conflict of interest: Data collection of FOS is supported by Shire Human Genetic Therapies. The sponsor had no role in the interpretation of data or writing of the report. All authors received honoraria and consultancy fees from Genzyme Inc. and Shire Human Genetic Therapies Inc., P.M.E. received an unrestricted educational grant from Shire Human Genetic Therapies, all authors are members of the Cardiac working group of the FOS, A.L., G.S.P., M.B. and A.M., are members of FOS Executive committee. Authors never received honoraria for their respective roles in FOS management and analysis, Shire Human Genetic Therapies covered all travel expenses associated with their work on the database.
| References |
|---|
|
|
|---|
- Desnick RJ, Ioannou YA, Eng CM.
-galactosidase A deficiency: Fabry disease. In: The Metabolic and Molecular Bases of Inherited DiseaseScriver CR, Beaudet AL, Sly WS, Valle D, eds. (2001) 8th ed. New York: McGraw-Hill. p37333774. - MacDermot KD, Holmes A, Miners AH. AndersonFabry disease. Clinical manifestations and impact of disease in a cohort of 60 obligate carrier females. J Med Genet (2001) 38:769775.
[Free Full Text] - MacDermot KD, Holmes A, Miners AH. AndersonFabry disease. Clinical manifestations and impact of disease in a cohort of 98 hemizygous males. J Med Genet (2001) 38:750760.
[Abstract/Free Full Text] - Mehta A, Ricci R, Widmer U, Dehout F, Garcia de Lorenzo A, Kampmann C, Linhart A, Sunder-Plassmann G, Ries M, Beck M. Fabry disease defined: baseline clinical manifestations of 366 patients in the Fabry Outcome Survey. Eur J Clin Invest (2004) 34:236242.[CrossRef][ISI][Medline]
- Beck M, Ricci R, Widmer U, Dehout F, de Lorenzo AG, Kampmann C, Linhart A, Sunder-Plassmann G, Houge G, Ramaswami U, Gal A, Mehta A. Fabry disease: overall effects of agalsidase alfa treatment. Eur J Clin Invest (2004) 34:838844.[CrossRef][ISI][Medline]
- Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med (1999) 130:461470.
[Abstract/Free Full Text] - Devereux RB, Alonso DR, Lutas EM, Gottlieb GJ, Campo E, Sachs I, Reichek N. Echocardiographic assessment of left ventricular hypertrophy: comparison to necropsy findings. Am J Cardiol (1986) 57:450458.[CrossRef][ISI][Medline]
- Linhart A, Palecek T, Bultas J, Ferguson JJ, Hrudova J, Karetova D, Zeman J, Ledvinova J, Poupetova H, Elleder M, Aschermann M. New insights in cardiac structural changes in patients with Fabrys disease. Am Heart J (2000) 139:11011108.[CrossRef][ISI][Medline]
- de Simone G, Devereux RB, Daniels SR, Koren MJ, Meyer RA, Laragh JH. Effect of growth on variability of left ventricular mass: assessment of allometric signals in adults and children and their capacity to predict cardiovascular risk. J Am Coll Cardiol (1995) 25:10561062.[Abstract]
- Desnick RJ, Brady R, Barranger J, Collins AJ, Germain DP, Goldman M, Grabowski G, Packman S, Wilcox WR. Fabry disease, an under-recognized multisystemic disorder: expert recommendations for diagnosis, management, and enzyme replacement therapy. Ann Intern Med (2003) 138:338346.
[Abstract/Free Full Text] - Elliott PM, Kindler H, Shah JS, Sachdev B, Rimoldi O, Thaman R, Tome MT, McKenna WJ, Lee P, Camici PG. Coronary microvascular dysfunction in male patients with AndersonFabry disease and the effect of treatment with alpha galactosidase A. Heart (2006) 92:357360.
[Abstract/Free Full Text] - Kalliokoski RJ, Kalliokoski KK, Sundell J, Engblom E, Penttinen M, Kantola I, Raitakari OT, Knuuti J, Nuutila P. Impaired myocardial perfusion reserve but preserved peripheral endothelial function in patients with Fabry disease. J Inherit Metab Dis (2005) 28:563573.[CrossRef][ISI][Medline]
- Shah JS, Hughes DA, Sachdev B, Tome M, Ward D, Lee P, Mehta A, Elliott PM. Prevalence and clinical significance of cardiac arrhythmia in AndersonFabry disease. Am J Cardiol (2005) 96:842846.[CrossRef][ISI][Medline]
- Moon JC, Sachdev B, Elkington AG, McKenna WJ, Mehta A, Pennell DJ, Leed PJ, Elliott PM. Gadolinium enhanced cardiovascular magnetic resonance in AndersonFabry disease. Evidence for a disease specific abnormality of the myocardial interstitium. Eur Heart J (2003) 24:21512155.
[Abstract/Free Full Text] - Desnick RJ, Blieden LC, Sharp HL, Hofschire PJ, Moller JH. Cardiac valvular anomalies in Fabry disease. clinical, morphologic, and biochemical studies. Circulation (1976) 54:818825.
[Abstract/Free Full Text] - Kampmann C, Baehner F, Whybra C, Martin C, Wiethoff CM, Ries M, Gal A, Beck M. Cardiac manifestations of AndersonFabry disease in heterozygous females. J Am Coll Cardiol (2002) 40:16681674.
[Abstract/Free Full Text]
This article has been cited by other articles:
![]() |
T. Kovacevic-Preradovic, M. Zuber, C.H. Attenhofer Jost, U. Widmer, B. Seifert, G. Schulthess, A. Fischer, and R. Jenni Anderson-Fabry disease: long-term echocardiographic follow-up under enzyme replacement therapy Eur J Echocardiogr, April 14, 2008; (2008) jen129v1. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||





