Risk Prediction of Coronary Artery Disease among Naval Forces

Background: Coronary artery disease (CAD) risk factors seem to cluster in some occupational groups. Objective: The present study was designed to investigate CAD risk factors among naval forces as an example of a high risk sector and to construct a risk prediction model for the disease. Methods: A case control study was carried out at the general naval hospital (GNH) in Alexandria. The study included 250 male consecutive naval CAD cases with a control group of 250 males matches for age, occupational level, sociodemographic characteristics and, free from CAD. All participants were subjected to a questionnaire about personal data, occupational history and exposures, occupational and leisure physical activity, dietary habits, smoking, and medical history. Anthropometric measurements, sitting blood pressure, and lipid profile were determined by the standard methods. Results: revealed that occupational sedentary activity and perceived occupational noise were the significantly reported special occupational characteristics together with other conventional risk factors among CAD naval cases verses controls (p<0.0001 & <0.009 respectively). Logistic regression analysis with the dependent variable as being a CAD case showed independently significant effects for family history of premature CAD, history of hypertension, smoking, history of diabetes mellitus, body mass index (BMI), leisure physical activity, fish consumption, and HDL-cholesterol. A risk prediction model utilizing these variables was constructed with an overall correct percent of 74.6%. Conclusions: Application of the model expresses the risk of having CAD in an individual eligible with criteria of the study population. These results are of special importance for design of preventive programs for CAD in similar high risk occupational groups. INTRODUCTION Coronary artery disease (CAD) is the principal cause of premature disability and mortality in most economically advanced countries. Also, developing countries are now subjected to the same risk factors showing increase in their incidence.(1,2) In Egypt, cardiovascular diseases mortality showed a sharp rise since 1998 when it 638 Bull High Inst Public Health Vol.38 No.3 [2008] represented 47% of total deaths; meanwhile, it represented only 12.4% in 1970. Urbanization may be a contributing mechanism for the increased prevalence of coronary risk factors namely; hypertension, hypercholesterolemia, low HDLcholesterol, elevated LDL-cholesterol, hypertriglyceridemia, obesity, increased fasting and post-prandial blood glucose and smoking as reported in the Egyptian National Hypertension Project (NHP 1991 – 1994).(3) On the other hand, physical activity and dietary factors greatly influence the risk of CAD. Also, occupational environment may pose an increased risk for CAD through the different physical, chemical, and psychosocial exposures. Shift work, psychosocial stress, noise, carbon monoxide, organic solvents and heavy metals are important exposures influencing coronary risk either independently or through other conventional risk factors.(4-7) Such exposures seem to cluster in some special occupational groups such as military and naval forces. The present study was designed to investigate CAD risk factors among naval forces as an example of a high risk occupational group and to construct a model for risk prediction of the disease. MATERIAL AND METHODS Study design and setting: A case control study was carried out at the general naval hospital "GNH" in Alexandria. Study population, sample size: The sample size required to accomplish the aim of the present work was determined using EPI-INFO version 6 and considering a level of confidence at 95%, an expected frequency in not ill group of 5%, a power of 90% and an odds ratio (OR) of 3.09.(8) The minimum sample size was estimated as 243 CAD cases. Accordingly, the study comprised 250 male consecutive naval cases admitted to Abd El-Nabi et al. 639 the GNH and finally diagnosed as CAD. For each case a control matched for age and occupational level was chosen from males attending or admitted to the GNH for any reason other than chest pain and free from CAD as evidenced by history and electorcardiographic examination. Methods: All participants were subjected to a questionnaire about personal and occupational data including occupational level, and history of exposure to carbon monoxide, perceived noise or high atmospheric pressure according to their specific job criteria. Occupational and leisure physical activity were inquired as the long-term habits according to the recommendations of the American Heart Association.(9) Smoking, long-term dietary habits, medical history of hypertension and diabetes mellitus, and family history of premature CAD before the age of 55 years according to the WHO guidelines.(10) Weight (Kg), height (cm), body mass index (BMI = kg/m2), waist circumference (cm), and sitting blood pressure (mmHg) were measured by the standard methods. Fasting venous blood samples were collected by veni-puncture to determine total cholesterol, HDL-cholesterol, and triglycerides by the standard enzymatic methods. The LDL-cholesterol was computed using the Fried Weldi equation. (11) Statistical analysis: Data were analyzed using SPSS (Statistical Package for Social Sciences) version 11.(12) Continuous variables were expressed as means and standard deviations, meanwhile, qualitative data were defined as number and percentage. Comparison between cases and controls regarding the different risk factors was done using the appropriate test of significance (t-test, Pearson's 2, and multilinear 2). Logistic regression analysis, forward stepwise method, was used to obtain the best predictive model for CAD cases 640 Bull High Inst Public Health Vol.38 No.3 [2008] utilizing the independently significant factors. The level of significance for p value was considered at < 0.05. RESULTS The occupational characteristics of CAD cases and controls showed no significant differences between both groups regarding occupational level, history of exposure to carbon monoxide, and high atmospheric pressure. However, perceived occupational noise was more significantly reported among CAD cases in comparison to controls (19.2% versus 6%, respectively, p< 0.05) (Table 1). There were no significant differences for age and height between CAD cases and controls. Comparison of risk factors for the study population demonstrated significant differences with higher mean values among CAD cases versus controls for weight (93.2 vs 91.1 Kg), BMI (28.8 vs 28.1 kg/m2), waist circumference (105.3 vs 100.6 cm), SBP (133.4 vs 123.7 mmHg), DBP (87.6 vs 79.4 mmHg), total cholesterol (232.3 vs 200.4 mg/dl), LDL-cholesterol (150.2 vs 108.9 mg/dl), triglycerides (146.6 vs 132.4 mg/dl), and smoking index (319.7 vs 207.4 daily cigarettes years) (p< 0.001 for all). On the other hand, the mean value of HDL-cholesterol was significantly lower for CAD cases than controls (52.8 vs 65.0 mg/dl) (p< 0.001) (table 2). The rates for diabetes mellitus, hypertension, family history of pre-mature CAD, smoking and occupational sedentary activity were significantly higher among CAD cases in comparison to controls (p< 0.01 for all). Meanwhile, regular leisure physical activity was less significantly reported by CAD cases than controls (p= 0.014) (table 3). Dietary habits were investigated in the study population to reveal higher significant rates of excessive frank fat intake, excessive salt intake and regular full cream milk consumption, and higher mean weekly egg intake for CAD cases than controls (p< 0.005 for all). On the contrary, the rate of Abd El-Nabi et al. 641 regular raw stuff consumption and mean monthly fish intake were significantly lower for CAD cases than controls (p< 0.005 for all). There were no significant differences for daily intake of both tea and coffee by CAD cases and controls (table 4). Logistic regression analysis with the dependent variable as being CAD case or control was performed by the forward stepwise method. The final best model included family history of premature CAD, history of hypertension, smoking, history of diabetes mellitus, BMI, leisure physical activity, fish consumption and HDLcholesterol as significant independent variables (P < 0.05 for all) (table 5). The overall correct predictive percentage of the model was 74.6% (table 6). Table (1): Occupational characteristics of CAD cases and controls. Cases Controls X p no. % no. % Retired officer 223 89.2 221 88.4 Working officer 27 10.8 29 11.6 0.008 0.928 History of exposure to CO 14 5.6 12 4.8 <5 >0.05 History of exposure to high atmosphere pressure 5 2.0 9 3.6 <5 >0.05 Perceived occupational noise 48 19.2 15 6.0 19.78 0.000 *Significant at level 0.05 642 Bull High Inst Public Health Vol.38 No.3 [2008] Table (2): Continuous risk factors among CAD cases and controls. Cases Controls t-test p Mean S.D. Mean S.D. Age (years) 58.33 7.64 58.32 7.41 0.006 0.995 Height (cm) 180.02 3.51 180.12 3.29 0.30 0.763 Weight (kg) 93.24 6.31 91.09 6.05 3.89 0.000 BMI (kg/m) 28.81 2.28 28.10 2.06 3.62 0.000 Waist circumference (cm) 105.2 6.15 100.56 5.35 9.20 0.000 SBP (mmHg) 133.38 16.96 123.72 15.6 6.63 0.000 DBP (mmHg) 87.56 10.27 79.4 8.83 9.53 0.000 TC (mg/dl) 232.26 37.55 200.37 30.81 10.38 0.000 HDL-C (mg/dl) 52.75 24.35 65.01 27.15 5.31 0.000 LDL-C (mg/dl) 150.21 29.86 108.88 36.34 13.89 0.000 Tg (mg/dl) 146.55 30.87 132.42 35.27 4.77 0.000 Smoking index (daily cigarettes x years) 319.65 181.35 207.36 115.80 4.92 0.000 *Significant at level 0.05 Table (3): Categoric risk factors among CAD cases and controls. Cases Controls X p no. % no. % History of D.M 79 31.6 38 15.2 18.76 0.000 History of hypertension 103 41.2 37 14.8 43.21 0.000 Family history of premature CAD 55 13.0 21 8.4 17.94 0.000 Smoking 108 43.2 63 25.2 17.99 0.000 Occupational sedentary activity 214 85.6 187 74.8 9.49 0.009 Regular leisure physical activity 30 12.0 60 24.0 6.09 0.014 * Sig


INTRODUCTION
Coronary artery disease (CAD) is the principal cause of premature disability and mortality in most economically advanced countries.Also, developing countries are now subjected to the same risk factors showing increase in their incidence. (1,2)In Egypt, cardiovascular diseases mortality showed a sharp rise since 1998 when it Bull High Inst Public Health Vol.38 No. 3 [2008]   represented 47% of total deaths; meanwhile, it represented only 12.4% in 1970.Urbanization may be a contributing mechanism for the increased prevalence of coronary risk factors namely; hypertension, hypercholesterolemia, low HDLcholesterol, elevated LDL-cholesterol, hypertriglyceridemia, obesity, increased fasting and post-prandial blood glucose and smoking as reported in the Egyptian National Hypertension Project (NHP 1991   -1994). (3) the other hand, physical activity

Study design and setting:
A case control study was carried out at the general naval hospital "GNH" in Alexandria.

Study population, sample size:
The sample size required to accomplish the aim of the present work was determined using EPI-INFO version 6 and considering a level of confidence at 95%, an expected frequency in not ill group of 5%, a power of 90% and an odds ratio (OR) of 3.09. (8)The minimum sample size was estimated as 243 CAD cases.
Accordingly, the study comprised 250 male consecutive naval cases admitted to the GNH and finally diagnosed as CAD.
For each case a control matched for age and occupational level was chosen from males attending or admitted to the GNH for any reason other than chest pain and free from CAD as evidenced by history and electorcardiographic examination.

Methods:
All participants were subjected to a questionnaire about personal and occupational data including occupational level, and history of exposure to carbon monoxide, perceived noise or high atmospheric pressure according to their specific job criteria.Occupational and leisure physical activity were inquired as the long-term habits according to the recommendations of the American Heart Association. (9)Smoking, long-term dietary habits, medical history of hypertension and diabetes mellitus, and family history of premature CAD before the age of 55 years according to the WHO guidelines. (10)ight (Kg), height (cm), body mass index (BMI = kg/m 2 ), waist circumference (cm), and sitting blood pressure (mmHg) were measured by the standard methods.
Fasting venous blood samples were collected by veni-puncture to determine total cholesterol, HDL-cholesterol, and triglycerides by the standard enzymatic methods.The LDL-cholesterol was computed using the Fried Weldi equation. (11)atistical analysis:  3).
Dietary habits were investigated in the study population to reveal higher significant rates of excessive frank fat intake, excessive salt intake and regular full cream milk consumption, and higher mean weekly egg intake for CAD cases than controls (p< 0.005 for all).On the contrary, the rate of regular raw stuff consumption and mean monthly fish intake were significantly lower for CAD cases than controls (p< 0.005 for all).There were no significant differences for daily intake of both tea and coffee by CAD cases and controls (table 4).This finding may enforce the assumption that a neurohormonal activation mediated by noise possibly imposes an indirect effect through other risk factors such as hypertension and dyslipidemia. (5)nventional CAD risk factors by quantitative or qualitative measures were more significantly observed among CAD cases versus controls including high levels of BMI, waist circumference, SBP, DBP, total cholesterol, LDL-cholesterol, triglycerides and smoking index, and low HDL level.Also, significant high rates of diabetes mellitus (31.6%), hypertension (41.2%), family history of premature CAD (13%), smoking (43.2%), occupational sedentary activity (85.6%), and absence of regular leisure physical activity (88%) were reported among CAD cases.7)(18) All coronary risk factors that were prevalent in this study among CAD naval cases are modifiable except the genetic factor that was reported in 13% of cases only.Hypertension, blood lipids, diabetes mellitus, smoking, obesity, and physical activity are the most important modifiable CAD risk factors. (14)e negative dietary habits that were significantly reported in this study among CAD cases were excessive fat intake, including frank fat, full-cream milk and/or high weekly egg intake, and excessive salt intake.On the other hand, the positive dietary habits that were less significantly observed among CAD cases were regular raw stuff consumption and monthly times of fish intake.Dietary intervention is an essential tool in cardio-prevention to Bull High Inst Public Health Vol.38 No. 3 [2008]   reduce body weight and to contribute in control of other risk factors such as dyslipidemia, hypertension and diabetes. (19)od availability and occupation may influence type of diet of the individual.Fish is a particularly common diet among naval population that should be encouraged along with other healthy food as a part of a balanced diet. (20)ta of the current occupational group Study prediction score (21) and others. (22)wever, BMI, leisure physical activity and fish consumption in our model have replaced total or LDL-Cholesterol in the traditional models.

Logistic regression analysis
This was observed to reduce CAD events. (25)reover, the favorable effect of fish consumption on CAD risk was previously reported. (26)This is possibly explained by the high concentration of omega 3 fatty acids that may have an antiprostaglandin like action aborting the inflammatory process of atherogenesis.Also, it has a favorable effect on blood lipids particularly triglycerides. (27)ese findings draw the attention for

CONCLUSION AND RECOMMENDATIONS
The suggested model may be used to risk factors.(4)(5)(6)(7)Such exposures seem to cluster in some special occupational groups such as military and naval forces.The present study was designed to investigate CAD risk factors among naval forces as an example of a high risk occupational group and to construct a model for risk prediction of the disease.
were utilized to construct a CAD risk model to be used in the future for prediction of CAD in any individual eligible with the criteria of study population.The independently significant risk factors in the best model were hypertension, smoking, diabetes mellitus and HDL-cholesterol.These are common risk predictors in this model, as well as in Framingham Heart the special needs of this occupational group to optimize CAD risk profile through controlling risk factors and promoting favorable factors, in community-based preventive programs.Application of the CAD risk prediction model is the novel aspect in this study to define high risk subjects in similar situations, in order to provide them with intensified risk control measures in individual-based preventive programs.
predict CAD risk in individuals having similar demographic and occupational characteristics without CAD in primary prevention or with CAD in secondary prevention.Also, more studies are needed for other occupational groups.Global CAD risk management is needed to get the best preventive outcome.Particular emphasis on controlling hypertension, smoking and diabetes mellitus with increasing HDL-cholesterol is needed.Also, promoting the favorable triad of weight reduction, physical activity and fish consumption is a fruitful option for CAD prevention in general and among naval forces in particular.