German Origin Clusters for High Cardiovascular Risk in an Italian Enclave

Edoardo CASIGLIA,1 MD, Giancarlo BASSO,2 MD, Francesco GUGLIELMI,1,2 MD, Bortolo MARTINI,2 MD, Alberto MAZZA,1 MD, Valérie TIKHONOFF,1 MD, Roberta SCARPA,1 MD, Mario SAUGO,3 MD, Sandro CAFFI,3 MD, and Achille C. PESSINA,1 MD

SUMMARY Mortality and morbidity appear to be higher in a Cimbrian population representing an enclave of people who migrated from medieval Germany to the secluded Leogra valley in Italy. A population-based study was organized, recruiting 881 elderly subjects of Cimbrian origin and comparing them with a standard control population (SCP, n = 3282) having comparable general characteristics and lifestyle. Serum lipids and glucose, blood pressure, heart rate, respiratory function, ECG abnormalities, and historical events were used as risk indicators. Age-adjusted systolic and pulse pressure were higher in the Cimbrians than in the SCP, while diastolic blood pressure was comparable. The prevalences of arterial hypertension, isolated systolic hypertension, and pulse hypertension were significantly more represented among Cimbrians than SCP. The prevalences of diabetes, hypercholesterolemia, and hypertriglyceridemia were higher among the former than the latter. The ratio between apolipoproteins B and A1 was also higher, while the HDL fraction was significantly lower in Cimbrians than in the SCP. In Cimbrians, the relative risk (RR) for ischemic heart disease was 1.92 (1.57-2.34) in women, 2.30 (1.54-3.43) in men and 1.03 (1.00-1.06) in women for stroke, 2.43 (1.54-3.83) in men and 1.45 (1.01-1.12) in women for atrial fibrillation, 3.85 (2.83-5.24) in men and 1.39 (1.20-1.60) in women for respiratory disease, 1.97 (1.32-2.94) in men and 6.81 (4.38-10.60) in women for intermittent claudication, and 3.31 (2.44-4.50) in men and 2.30 (1.76-3.01) in women for left ventricular hypertrophy. The subjects living in the secluded Leogra valley are at higher cardiovascular risk than the standard controls. Whether this depends on genetic factors, lifestyle, or both will need to be clarified by further analysis. (Int Heart J 2005; 46: 489-500)

Key words: Cardiovascular risk, Special populations, Genetics, Epidemiology

From the 1Department of Clinical and Experimental Medicine, University of Padova, Padova, 2Cardiac Unit, General Hos- pital of Schio-Thiene, Schio, and 3Direction and Epidemiological Department of the Local Health Unit No. 4, Thiene, Italy. Address for correspondence: Edoardo Casiglia, MD, Department of Clinical and Experimental Medicine, University of Padova, Via Giustiniani, 2 35128 Padova, Italy. Received for publication August 12, 2004. Revised and accepted December 17, 2004. 489 Int Heart J 490 CASIGLIA, ET AL May 2005 NO experimental model of ageing is available in humans and epidemiological observation of the general population is the best approximation. Migration and segregation are quasi-experimental models at a population level.1-3) Our group previously had the opportunity to verify this model at a population level by studying African and South American people living in different contexts.1,2,4,5) In 1999, a special Italian population came to our attention. This population - called Cimbrian - has 4 interesting characteristics: it is of middle-European origin although the people have been living in Italy for many centuries, it is located in a secluded valley, it underwent emigration but low or absent immigration, and it shows higher mortality and morbidity due to complications of arteriosclerosis in comparison to neighbouring areas.6) In a preliminary analysis of mortality based on records from the government, hospitals, and physicians, the standard- ised mortality ratio for coronary heart disease (CAD) was 1.36 (95% confidence 6) intervals [95%CI] 1.13-1.62) in males, and 1.29 (95%CI 1.06-1.57) in females. From 983 to 1287 AD, people from Germany crossed the Alps and settled in different uninhabited areas of the North Italian mountains between Vicenza, Verona, and Trento. They were called Cimbrian (from tzimbarn, woodcutter),7) maintained German traditions and a medieval German-like language (the Toish or Tausch), and preferred not to integrate or assimilate with the natives. Linguis- tic analysis demonstrates that the first came from Bavaria and particularly from an area under the jurisdiction of the Benediktbeuern abbey. Groups of German tzimberer then migrated to Asiago, Posina, Folgaria, Lavarone, Luserna, Valli del Leno, and Vallarsa, and in 1287 to the Lessinia area. Those now living in the valley or by the Leogra River, coming in 1200 AD from Posina, remained secluded, intermarried, and are still recognizable for their German or German-like family names and traditions. In Europe, they are the most South-settled Germanic- speaking people. While important emigration interested North Italy in past centuries, practically no immigration was recorded in the Leogra valley, so the present inhabitants must be considered to be of German strain.7) This study was performed in order to evaluate whether the prevalence of cardiovascular risk factors actually is higher in elderly subjects from the above- mentioned Cimbrian population than in elderly subjects living in the neighbouring areas, which may explain their less favourable outcome, by comparing these Cimbrians with a control standard population. The decision to focus on the elderly was based on the consideration that the epidemiology of cardiovascular risk in the elderly is far from being defined. Although a limited number of elderly subjects have been included as fringe groups in general epidemiological investigations, only a few studies have had an elderly population as the main target group. The only Italian study with these characteristics is the CArdiovascular STudy in the ELderly (CASTEL),8-10) per- Vo l 4 6 No 3 CIMBRIANS OF GERMAN ORIGIN IN ITALY 491 formed by our group on a representative cohort of 3,282 subjects of the Italian general population.

METHODS Cimbrian study population: The LEOGRA study8) was conceived as a longitudinal study aimed at evaluating in the long term the incidence and relative weight of the cardiovascular risk factors in the elderly, as well as the feasibility of an interventional program of cardiovascular prevention. The study cohort consisted of 881 subjects (350 men and 531 women), aged 65 years or over, representative of all the elderly living in the Leogra valley (namely in Valli del Pasubio and Tor- rebelvicino). The data obtained from the initial cross-sectional survey are described herein. Data from longitudinal analysis, including the incidence of new cases of hypertension, morbidity, and mortality, will be available in the next few years. For the purpose of the present analysis, 881 elderly subjects aged 65 years or over living in the Leogra valley, identified as “Cimbrian” according to criteria based on history and on family names,7) were invited by letter to visit to a local general hospital. Although the area selected for the study offered a great number of pre-existing medical facilities (some even being considered the most advanced in Italy in this respect), a special epidemiological unit was organized within a hospital, with a staff from the University of Padua dedicated to the study. Before the clinical examination, all subjects underwent blood tests and a very detailed questionnaire concerning personal data, lifestyle, smoking and drinking habits, nutrition, physical activity, quality of life, and personal and familial medical history.11-13) Anthropometrics were taken in light underwear without shoes. Body mass index (BMI) was calculated from the Quatelet algorithm. Blood pressure (diastolic Korotkoff phase 5) was taken in triplicate both in lying and standing posi- tions by trained doctors using a mercury sphygmomanometer at 10-min intervals, taking special care to avoid any terminal digit preference. The average of the last two clinostatic measurements was taken as BP, to minimise white-coat effects, if any. Heart rate was also taken at the same time. Standard 12-lead electrocardiograms were performed in the morning and blindly codified according to the Min- nesota code14) by an expert who did not know the aim and design of the study. All subjects underwent spyrometry by means of a Pony Spyrometer class I type B (Cosmed, Rome, Italy). Vital capacity (VC in l) and forced expiratory volume in

1 second (FEV1 in l) were measured. Theoretical VC and FEV1 were then calculated for each subject, and individual pulmonary function was defined for each individual as a percent of the theoretical values. Int Heart J 492 CASIGLIA, ET AL May 2005

Standard control population: Elderly subjects included in the CASTEL, who did not differ from the study population with respect to lifestyle, diet, and habits (Table I), were used as the standard control population (SCP). The protocol of the CASTEL, described elsewhere,9,10,15-18) was identical to that used for studying the Cimbrian elderly population. All elderly subjects aged 65 years or over in the Ital- ian towns of Castelfranco, Veneto, and Chioggia were invited by letter to visit to a clinic organised at the local general hospital; 3,282 (73%) accepted the study protocol and were enrolled. After signing an informed consent form, each subject underwent the same procedures as in the LEOGRA protocol. Statistical analysis: SAS software (version 8.1, SAS Institute, Cary, NC) was used for statistical analysis. According to power analysis based on the results of the CASTEL study, the number of Cimbrian subjects was adequate for statistical analysis. Values of continuous variables are expressed as the mean and standard devi-

ation and the 95%CI is indicated when appropriate. Analysis of variance using the

Table I. General Characteristics of the Study Cohort and of the Standard Control Population

Standard control population Study cohort (n = 881) The CASTEL study (n = 3282)

Men (n = 350) Women (n = 531) Men (n = 1282) Women (n = 2001)

Age (years) 71.2 ± 5.5 73.1 ± 6.2† 73.1 ± 4.9* 74.2 ± 5.5*† Married subjects (%) 90.8 93.0† 90.2 92.8† Body mass index (kg/m2) 27.5 ± 4.5 27.5 ± 3.7 25.9 ± 3.8* 26.7 ± 4.6* Overweight (%) 54.0 48.0 36.2 44.1 Smoking habits: Smokers (%) 14.0 12.1† 24.9* 10.1† Cigarettes/day♦ 1.7 ± 4.9 1.0 ± 3.2† 2.5 ± 5.3* 0.4 ± 2.0*† Former smokers (%) 62.3 21.1† 43.4* 3.8†* with morning cough of catharr (%) 28.6 27.7† 32.1 16.1†* ± † † Ethanol from wine (mL/week) 413 302 189 ± 58 542 ± 398* 284 ± 119 * From beer (mL/week) 10 ± 46 1 ± 9† 10 ± 54 3 ± 25† ± † † From drinks (mL/week) 20 44 5 ± 27 7.5 ± 18* 2 ± 9 * Subjects adding salt to food (%) 21.1 19.8 16.9 13.7† Coffee drinkers (%) 90.3 91.3 93.0 89.7 ± Caffeine intake (mg/week) 678 347 691 ± 355 1077 ± 641* 965 ± 615* At least one parent with: † Hypertension (%) 22.6 33.0 15.8* 26.7*† Diabetes (%) 9.1 12.6 9.0 12.5 Sudden death (%) 0.3 0.2 0.25 0.2 Coronary artery disease (%) 23.1 28.8 18.0* 18.9*

Overweight: body mass index > 27 kg/m2; ♦among smokers only; †P < 0.01 versus men in same cohort, *P < 0.01 versus same gender in study cohort. Vo l 4 6 No 3 CIMBRIANS OF GERMAN ORIGIN IN ITALY 493 Bonferroni correction was then used to compare grouped continuous variables, and Pearson's χ2 test was used to compare the prevalence of categorical variables in the two cohorts. Differences were statistically significant when the P value was < 0.05 and not significant (NS) when it was ≥ 0.05. As age and body mass index were slightly but significantly different in the two populations, the analysis was adjusted, when necessary, for these items using the direct method. The determinants of dichotomic categorical variables and prevalence of disease were ascertained through stepwise logistic regression. Relative risk (RR) adjusted for the age and BMI of men versus women and of Cimbrian versus SCP was determined, for each dichotomic item, from the logistic equations and expressed together with its 95%CI.

RESULTS The mean age of the Cimbrians was 1.5% lower than the SCP subjects. BMI was significantly higher and parental history of arterial hypertension and CAD were significantly more prevalent in the Cimbrians than in the SCP for both men and women (Table I). Cigarette smoking was less common among the Cimbrian men and more common among the Cimbrian women than in the SCP. The prevalence of respiratory symptoms (morning cough or catarrh) was less among Cim- brian than control women (Table I). Ethanol consumption (mainly from wine) was significantly lower in the Cimbrian (means of 443 mL/week in men and 195 mL/week in women) than in the SCP population (542 and 284 mL, respectively). Weekly caffeine intake from Italian coffee (1 cup = 80 mg) was also significantly lower in the Cimbrian (men -37%, women -28%) than in the control subjects. The amount of salt added to food after cooking was comparable between the two populations (Table I). Age-adjusted systolic and pulse BP values were higher in the Leogra valley subjects than in the SCP, while diastolic BP was comparable in the two cohorts (Table II). The prevalence of arterial hypertension (systolic BP ≥ 140 or diastolic BP ≥ 90 mmHg or current antihypertensive treatment) was significantly higher (P < 0.0001) among the Cimbrians than in the SCP (91.5 versus 86%, ie. 92 versus 83.5% in men, and 91 versus 87.8% in women; nonsignificant difference between genders); the RR of arterial hypertension compared to the SCP was 1.88 (95%CI 1.24-2.87) for Cimbrian men and 1.45 (95%CI 1.01-1.12) for Cimbrian women. Isolated systolic hypertension (systolic BP ≥ 140 and diastolic BP < 90 mmHg) was more prevalent in the Cimbrians than in the SCP, a difference that was statistically significant in women (40% versus 33%, P < 0.001) but not in men (39% versus 34%). The prevalence of pulse hypertension (pulse pressure > Int Heart J 494 CASIGLIA, ET AL May 2005

Table II. Blood Pressure Values

Standard control population Study cohort (n = 881) The CASTEL study (n = 3282)

Men (n = 350) Women (n = 531) Men (n = 1282) Women (n = 2001)

Systolic blood pressure 161.8 ± 21.2† 162.8 ± 22.6 156.8 ± 24.0†* 161.8 ± 24.9* (mmHg) (159.6 - 164.0) (160.8 - 164.7) (155.5 - 158.1) (160.7 - 162.9) Diastolic blood pressure 91.4 ± 9.8† 89.0 ± 10.5 87.6 ± 11.3†* 89.4 ± 11.5* (mmHg) (89.3 - 91.4) (88.1 - 89.9) (87.0 - 88.3) (88.9 - 89.9) Pulse pressure 71.4 ± 18.1† 73.8 ± 18.6 69.2 ± 18.9†* 72.4 ± 19.3* (mmHg) (69.5 - 73.3) (72.5 - 75.4) (68.1 - 70.2) (71.5 - 73.2)

†P < 0.01 versus women in same cohort, *P < 0.0001 versus same gender in study cohort.

Table III. Blood Lipids in the Two Cohorts of Elderly Subjects From General Population

Standard control population Study cohort (n = 881) The CASTEL study (n = 3282)

Men (n = 350) Women (n = 531) Men (n = 1282) Women (n = 2001)

Total cholesterol (mmol/L) 5.62 ± 1.13 6.06 ± 1.16† 5.41 ± 1.08 5.85 ± 1.26†* (5.51 - 5.74) (5.95 - 6.16) (5.33 - 5.46)* (5.80 - 3.05) LDL-cholesterol (mmol/L) 3.83 ± 0.95 4.19 ± 1.01† 3.34 ± 0.95 3.60 ± 1.03†* (3.75 - 3.93) (4.09 - 4.27) (3.28 - 3.39)* (3.55 - 3.65) HDL-cholesterol (mmol/L) 1.11 ± 0.31 1.22 ± 0.28† 1.42 ± 0.41 1.55 ± 0.41†* (1.09 - 1.14) (1.19 - 1.24) (1.39 - 1.45)* (1.54 - 1.58) ± ± Triglycerides (mmol/L) 1.47 0.88 1.41 ± 0.73 1.37 0.81 1.46 ± 0.82† (1.38 - 1.56) (1.36 - 1.48) (1.32 - 1.42)* (1.42 - 1.50) ± ± Apolipoprotein A1 (g/L) 1.47 0.31 1.58 ± 0.29† 1.54 0.32 1.60 ± 0.30† (1.44 - 0.15) (1.56 - 1.61) (1.52 - 1.57)* (1.59 - 1.62) 1.09 ± 0.29 1.28 ± 0.30 Apolipoprotein B (g/L) 1.11 ± 0.29 1.35 ± 0.31†* (1.06 - 1.12) (1.09 - 1.14) (1.26 - 1.31)* (1.33 - 1.37) 0.77 ± 0.25 0.88 ± 0.27* ApoB/apoA1 ratio 0.74 ± 0.34 0.87 ± 0.26* (0.74 - 0.79) (0.85 - 0.88) (0.71 - 0.77) (0.86 - 0.89)

Mean ± standard deviation; 95% confidence intervals in brackets. †P < 0.01 versus women in same cohort, *P < 0.0001 versus same gender in study cohort. LDL and HDL = low- and high-density lipoprotein fractions.

60 mmHg)17) was also significantly higher among the Cimbrian subjects than in the SCP (72% versus 66%, P < 0.001); in men it was 68% versus 62% (P < 0.05), and in women 75% versus 69% (P < 0.001). The lipid pattern was more unfavourable in the Cimbrians than in the SCP. Serum lipid values were significantly higher in the former than in the latter (Table III). The prevalence of hypercholesterolemia (serum total cholesterol ≥ 5.2 Vo l 4 6 No 3 CIMBRIANS OF GERMAN ORIGIN IN ITALY 495 mmol/L) was also greater in the Cimbrians than in the SCP both in men (63% versus 55%, P < 0.0001) and women (77% versus 71%, P < 0.0001). The RR of hypercholesterolemia was 1.24 (95%CI 1.01-1.60) in men and 1.34 (95%CI 1.06- 1.68) in women of Cimbrian origin. The ratio between apolipoproteins B and A1 was also higher in the Leogra valley subjects than in the SCP, while the HDL fraction was significantly lower (Table III). Hypertriglyceridemia (serum triglycerides ≥ 1.7 mmol/L) was more prevalent in Cimbrian men than in male SCP (28% versus 20%, P < 0.001), while no difference was observed in women (24.5% versus 26%). No between-cohort difference in the prevalence of diabetes mellitus was detected after adjusting for age and body mass index. CAD (Minnesota code 1.1 or 1.2 or 1.3 if absent 6.4.1, or 4.1 or 4.4 if absent 6.4.1, 7.1.1 and 7.2.1, or 5.1 or 5.2 or 5.3 or 5.4 if absent 6.4.1, 7.1.1, 7.2.1 and 7.4, or positive myocardial scintigraphy, or positive stress test, or history of myocardial infarction confirmed by hospital records, or history of angina pectoris confirmed by hospital or physician records, or appropriate antianginal chronic treatment) was more prevalent in the Leogra valley subjects than in the SCP

(Table IV). The RR of CAD was 3.40 (95%CI 2.65-4.37) in men and 1.92 (95%CI 1.57-2.34) in women who were Cimbrian in origin, as compared to controls. Historical stroke (neurological signs or positive history confirmed by hospital or physician records) was more prevalent in the study cohort than in the control cohort for both genders (12% versus 6% in men, P < 0.0001; 10 versus 1.7% in women, P < 0.0001). Compared to the SCP, the RR of having a stroke in the clinical history was 2.30 (95%CI 1.54-3.43) for Cimbrian men and 1.03 (95%CI 1.00-1.06) for Cimbrian women. Atrial fibrillation (Minnesota code 8.3) was more prevalent in the Leogra valley subjects than in the SCP, with a RR of 2.43 (95%CI 1.54-3.83) in men and 1.45 (95%CI 1.01-1.12) in women. In the Cimbrian subjects but not in the SCP, this arrhythmia was more prevalent in men than in women (RR of male gender 1.96,

95%CI 1.16-3.30). Intraventricular conduction defects (Minnesota code 7 > 0), cardiac pre-excitation (Minnesota 6.4), and second-degree atrioventricular block (Minnesota 6.2) were also more prevalent in the former than in the latter (Table IV). Respiratory parameters are also presented in Table IV. Respiratory disease was significantly more prevalent in Cimbrian than in SCP subjects, with a RR of

3.85 (95%CI 2.83-5.24) in men and 1.39 (95%CI 1.20-1.60) in women. This was due to the higher prevalence of bronchospasm and to lower vital capacity and forced expiratory volume values in the former than in the latter, while the incidences of morning cough and catarrh were comparable in the two cohorts. In both cohorts, respiratory disease was significantly more prevalent in men than women. Int Heart J 496 CASIGLIA, ET AL May 2005

Table IV. Clinical and Historical Cardiovascular and Respiratory Parameters in the Two Cohorts of Elderly Subjects From General Population

Standard control population Study cohort (n = 881) The CASTEL study (n = 3282)

Men (n = 350) Women (n = 531) Men (n = 1281) Women (n = 2001)

Left ventricular hypertrophy 27.1† 18.6* 11.0* 9.6* Atrial fibrillation 9.4 6.2* 4.1* 4.6* Cardiac pre-excitation 2.0 3.0* 0.23* 0.25* 2nd-degree A-V block 0.9 0.2 0.01* 0.0 IV conduction defects 19.4† 11.7 16.5*† 10.7 History of CAD 46.9 41.4 19.9* 22.3* Myocardial infarction 30.3† 9.6 13.0*† 6.5* Angina pectoris 12.9 7.9 4.2* 5.5* Silent ECG ischemia 4.2† 13.9 2.6*† 11.3 Intermittent claudication 11.4† 10.4 6.2*† 1.7* Respiratory disease 76.8† 48.2 51.4*† 43.3* † † Low FEV1 46.5 25.9 28.1* 30.0 Low vital capacity 57.2† 14.5 12.8*† 17.7 Morning cough 30.6† 15.2 29.8† 14.3 Morning catarrh 31.4† 13.9 29.9† 12.8 Bronchospasm 18.3 14.9 11.3*† 5.9*

Numbers and percent rates are shown. A-V = atrioventricular; IV = intraventricular; CAD = coronary

artery disease; FEV1 = forced expiratory volume in 1 second; low FEV1 and vital capacity = < 70% of the individual theoretical value. †P < 0.01 versus women in same cohort, *P < 0.001 versus same gender in study cohort. Intraventricular (IV) defects include bundle branch blocks and hemiblocks.

The RR of suffering from leg intermittent claudication (calf pain arising dur- ing walking and disappearing at rest) was 1.97 (95%CI 1.32-2.94) in Cimbrian men and 6.81 (95%CI 4.38-10.60) in Cimbrian women compared to the SCP subjects (Table IV). Left ventricular hypertrophy (Minnesota code 3.1 or 3.3) was twice as prevalent in the Leogra valley cohort than in the CASTEL cohort (Table IV), with a

RR for the Cimbrian elderly subjects of 3.31 (95%CI 2.44-4.50) in men and 2.30 (95%CI 1.76-3.01) in women. In the Leogra valley but not in the CASTEL cohort, left ventricular hypertrophy was more prevalent in men than in women (Table

IV), with a RR for male gender of 1.74 (95%CI 1.25-2.42).

DISCUSSION The reasons why some elderly subjects retain a young build and healthy body, while others are afflicted by a number of diseases or die prematurely, are still unknown. Arteriosclerosis and its complications, chronic respiratory dis- Vo l 4 6 No 3 CIMBRIANS OF GERMAN ORIGIN IN ITALY 497 eases, and the occurrence cardiovascular events are the basis of unhealthy ageing. Both genetic and lifestyle-dependent environmental factors are most likely involved. The problem is very significant from a social point of view. If environment and lifestyle are the main causes of early ageing, evidence-based public health strategies must be implemented in order to promote healthy ageing of the population.19,20) If, on the contrary, genetic factors are implicated, public health strategies must address the early identification of high-risk subjects, and resources must be allocated for following-up such persons. The need for experimental studies in this field is therefore very strong, but to date, the only way to discriminate between genetic and environmental risk factors is to analyse with epidemiological instruments, at the population level, different human groups that have different origins and live in the same ecogenetic context. Europe is a good model for this purpose, as it is composed of groups with ancient origins, who tended to remain separated for many centuries for political and social reasons, as well as linguistic barriers. In selected cases, migration and seclusion produced in Europe enclaves of subjects with special characteristics. This is the case of the Leogra valley in Italy, an area whose inhabitants are Ger- man in origin, and whose survival and quality of ageing are different from those of the Italian people living in the neighbouring areas, although the habits and life- styles are similar. Register's Office reports indicate that they have greater cardiovascular mortality,6) and the results reported here confirmed that they actually do have an inferior risk pattern compared to SCP of the same age class. In fact, practically all the most important risk factors (arterial hypertension, and namely systolic and pulse hypertension,17,21) dislipidemia,22) left ventricular hypertrophy,23) atrial fibrillation,24,25) and cardiac conduction defects24)) were more prevalent in the Leogra valley than in the SCP, and the prevalences of chronic diseases such as CAD, respiratory disease, intermittent claudication, and stroke were greater as well. High blood pressure, and particularly high pulse pressure, is one of the most important risk factors in the elderly. Its effect on risk is independent and graded, with an excess mortality of 1.01 for every mmHg.17,21,27,28) In Italian subjects having pulse hypertension, the RR of a CAD event was three-fold greater in longitudinal studies in comparison to the normotensives.17,27) In the Cimbrians of the Leogra valley, BP was higher than in the SCP for all three components (systolic, diastolic and pulse). Dyslipidemia is another crucial risk factor. All of the main phenotypic expressions of dyslipidemia (hypercholesterolemia,22,23) hypertriglyc- 22) 29) 30) eridemia, low HDL-fraction level, and high apoliproprotein B1/A ratio ) were more prevalent in the Cimbrians than in the SCP (Table III), particularly in women. The hypothesis that the direct association between hypercholesterolemia and the risk of CAD becomes less robust with increasing age31) is under debate, Int Heart J 498 CASIGLIA, ET AL May 2005 but in the Italian general population - and particularly in the geographic area where the Leogra and the CASTEL studies were performed - cholesterol contin- ues to play a role in predicting coronary mortality even after the age of 65, at least in men and specially in those who smoke.22) Left ventricular hypertrophy, highly prevalent in the elderly,32) is another independent risk factor, whose impact is pos- sibly reduced24) but not abolished by age.28,29) When CAD is taken into consideration, the RR of hypertrophy is > 1.70 in elderly Italian people.33) Atrial fibrillation, a common disease in the elderly,27) predisposes a patient to stroke via formation of atrial thrombi. In elderly subjects with atrial fibrillation, the RR of stroke is increased.32) Finally, we have previously demonstrated24) that in the Ital- ian general population conduction defects independently increase mortality by 40 to 49%. The high prevalence of historical chronic diseases among the Cimbrians also deserves comment. Although they act as mere risk indicators rather than as risk factors, historical items (CAD, stroke, intermittent claudication, and respiratory disease) constantly predicted, in our previous experience, overall and cardiac mortality.16,22,28) The convergence of all the above-mentioned risk factors and indicators in the Cimbrians suggests they are at high risk for cardiovascular complications, and accounts for the unusually high mortality and morbidity recorded in these subjects of German origin. It is unlikely that these differences were attributable to differences in environmental factors, anagraphics and/or lifestyle since marital status, school-attendance rate, salt consumption, physical activity, and type of work were comparable in the two populations, coffee consumption and alcohol intake were less prevalent, and mean age was lower in the Leogra valley than in the SCP. On the contrary, the prevalence of familial disease such as arterial hypertension and CAD was significantly higher among the Cimbrians. Possible expla- nations therefore are that of an unfavourable genetic pattern producing undesirable phenotype variations, or an unfavourable genotype/environment interaction. Genetic analyses will be performed and reported upon in the near future.

ACKNOWLEDGEMENTS We thank Dr. Giuseppe Cavedon for allowing the use of the chemical laboratory and Professor Vinicio Filippi for his advice in selecting the Cimbrian people and checking the historical aspects of the manuscript. Vo l 4 6 No 3 CIMBRIANS OF GERMAN ORIGIN IN ITALY 499

REFERENCES

1. Pavan L, Casiglia E, Pauletto P, et al. Blood pressure, serum cholesterol and nutritional state in Tanzania and in the Amazon: comparison with an Italian population. J Hypertens 1997; 15: 1083-90. 2. Pavan L, Casiglia E, Braga LM, et al. Effects of a traditional lifestyle on the cardiovascular risk profile: the Amondava population of the Brazilian Amazon. Comparison with matched African, Italian and Polish populations. J Hypertens 1999; 17: 749-56. 3. Wilson TW, Grim CE. Biohistory of slavery and blood pressure differences in blacks today. A hypothesis. Hypertension 1991; 17 (Suppl 1): 1122-8. 4. Pauletto P, Puato M, Caroli MG, et al. Blood pressure and atherogenic lipoprotein profiles of fish-diet and veg- etarian villagers in Tanzania: the Lugalawa study. Lancet 1996; 348: 784-8. 5. Marcovina SM, Kennedy H, Bittolo Bon G, et al. Fish intake, independent of apo(a) size, accounts for lower plasma lipoprotein(a) levels in Bantu fishermen of Tanzania: The Lugalawa Study. Arterioscler Thromb Vasc Biol 1999: 19: 1250-6. 6. Veneto Region Council Bullettin, Venezia, 1995-2003. 7. Filippi V. Breve Storia della Letteratura Cimbra dei Sette Comuni Vicentini, Edizioni Taucias Gareida , Giazza, Verona, 1982. 8. Casiglia E, Mazza A, Tikhonoff V, et al. Protocol and experimental design of the Leogra (Last Evidences Of Genetic Risk factors in the Aged) study, a population-based genetic approach to cardiovascular risk. J Hyper- tens 2000; 18 (Suppl. 2): 174. 9. Casiglia E, Palatini P. Cardiovascular risk factors in the elderly. J Hum Hypertens 1998; 12: 575-81. (Review) 10. Casiglia E, Mazza A, Tikhonoff V, et al. Weak effect of hypertension and other classic risk factors in the elderly who have already paid their toll. J Hum Hypertens 2002; 16: 21-31. 11. Perneger TV, Kossovsky MP, Cathieni F, di Florio V, Burnand B. A randomized trial of four patient satisfaction questionnaires. Med Care 2003; 41: 1343-52. 12. Rose GA, Blackburn H, Gillum RF, Prineas RJ. Cardiovascular Survey Methods. Geneva: World Health Orga- nization; 1982: 64-77. 13. Eaker ED. Use of questionnaires, interviews, and psychological tests in epidemiological studies of coronary heart disease. Eur Heart J 1988; 9: 698-704. 14. Annex 1. Classification of the electrocardiograms for population studies. In: Rose GA, Blackburn H, Gillum RF, Prineas RJ, eds. Cardiovascular Survey Methods. Geneva: World Health Organization; 1982: 123-45. 15. Casiglia E, Spolaore P, Mormino P, et al. The CASTEL project (Cardiovascular Study in the Elderly): protocol, study design, and preliminary results of the initial survey. Cardiologia 1991; 36: 569-76. 16. Palatini P, Casiglia E, Julius S, Pessina AC. High heart rate: a risk factor for cardiovascular death in elderly men. Arch Intern Med 1999; 159: 585-92. 17. Casiglia E, Tikhonoff V, Mazza A, Piccoli A, Pessina AC. Pulse pressure and coronary mortality in elderly men and women from general population. J Hum Hypertens 2002; 16: 611-20. 18. Casiglia E, Mazza A, Tikhonoff V, Pessina AC. Population-based studies improve outcome in hypertensive patients. Am J Hypertens 2002; 15: 605-8. 19. Kostis JB, Rosen RC, Brondolo E, Taska L, Smith DE, Wilson AC. Superiority of nonpharmacologic therapy compared to propranolol and placebo in men with mild hypertension: a randomized, prospective trial. Am Heart J 1992; 123: 466-74. 20. Dal Palu C, Magnani B, Zanchetti A, et al. Structured approach to the treatment of hypertension: Italian project. High Blood Pressure 1995; 4: 141-53. 21. Staessen JA, Gasowski J, Wang JG, et al. Risks of untreated and treated isolated systolic hypertension in the elderly: meta-analysis of outcome trials. Lancet 2000; 355: 865-72. 22. Casiglia E, Mazza A, Tikhonoff V, Scarpa R, Schiavon L, Pessina AC. Total cholesterol and mortality in the elderly. J Intern Med 2003; 254: 345-62. 23. Casiglia E, Daskalakis C, Maniati G, et al. Left ventricular hypertrophy is not a predictor of mortality in the elderly. High BP 1997; 6: 12-21. 24. Casiglia E, Spolaore P, Ginocchio G, et al. Mortality in relation to Minnesota code items in elderly subjects. Sex-related differences in a cardiovascular study in the elderly. Jpn Heart J 1993; 34: 567-77. Int Heart J 500 CASIGLIA, ET AL May 2005

25. Casiglia E, Maschio O, Spolaore P, et al. [Atrial fibrillation in a cohort of the elderly: etiopathogenic role of occult hyperthyroidism and diagnostic and therapeutic considerations. Results of the CASTEL (Cardiovascular Study in the Elderly)] Cardiologia 1991; 36: 685-91. 26. Lee ML, Rosner BA, Weiss ST. Relationship of blood pressure to cardiovascular death: the effects of pulse pressure in the elderly. Ann Epidemiol 1999; 9: 101-7. 27. Casiglia E, Mazza A, Tikhonoff V, Scarpa R, Guglielmi F, Pessina AC. Arterial hypertension and mortality in the elderly. Am J Hypertens 2002; 15: 958-66. 28. Mazza A, Pessina AC, Privato G, Tikhonoff V, Pavei A, Casiglia E. Pulse pressure: an independent predictor of coronary and stroke mortality in elderly females from the general population. Blood Pressure 2001; 10: 205-11. 29. Houterman S, Boshuizen HC, Verschuren WM, et al. Predicting cardiovascular risk in the elderly in different European countries. Eur Heart J 2002; 23: 294-300. 30. Kuyl JM, Mandelsohn D. Observed relationship between ratios HDL-cholesterol/total cholesterol and apolipoprotein A1/apolipoprotein B. Clin Biochem 1992; 25: 313-6. 31. Delahaye F, Brukert E, Thomas D, Emmerich J, Richard JL. Serum cholesterol and cancer. Is there a causal relationship? Arch Mal Coeur Vaiss 1992; 85 (Spec 3): 37-45. 32. Casiglia E, Maniati G, Daskalakis C, et al. Left-ventricular hypertrophy in the elderly: unreliablity of ECG criteria in 477 subjects aged 65 years or more. The cardiovascular study in the elderly (CASTEL). Cardiology 1996; 87: 429-35. 33. Hart RG, Halperin JL, Pearce LA, et al. Lessons from the Stroke Prevention in Atrial Fibrillation trials. Ann Intern Med 2003; 138: 831-8. (Review)