SEARCH

SEARCH BY CITATION

Keywords:

  • FHCHD;
  • ApoE;
  • smoking;
  • prospective study;
  • CHD risk

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

We have estimated the risk of coronary heart disease (CHD) from family history of CHD (FHCHD) in 2827 healthy European middle-aged men, and explored the extent to which this can be explained by classical and genetic risk factors. Men with FHCHD (obtained by questionnaire) had a hazard ratio of CHD of 1.73 (95% confidence interval: 1.30, 2.31) compared to those without FHCHD; after adjusting for classical risk factors this did not change substantially. Those with FHCHD had 2.3% lower Factor VIIc (p = 0.03) and 1.14% higher systolic and 1.21% higher diastolic blood pressure (p = 0.04 and p = 0.02), with evidence of interaction between blood pressure and FHCHD status on risk (p = 0.01). The risk for those with a positive family history who were also current smokers was 3.01 compared to non-smokers without FHCHD, which is greater than the risk posed by smoking or FHCHD alone (1.96 and 2.05 respectively compared to non-smokers without FHCHD), but not significantly different from a multiplicative model (p-value for interaction 0.33). Allele frequencies for 13 candidate gene variants were not significantly different between those with and without FHCHD. In those with FHCHD, current smokers who carried the APOE4 allele (e4+) had a hazard ratio of 5.66 compared to non-smokers who had no FHCHD and were not APOE4+, with a significant interaction between smoking and APOE4 in those with FHCHD p = 0.001. These data demonstrate the complex interaction between genetic and environmental factors in determining CHD risk, and suggest that the causes of the familial clustering of CHD remain largely unexplained.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Family history of coronary heart disease (FHCHD) has been found to be an independent risk factor for coronary heart disease (CHD) in a number of different studies (Boer et al. 1999; Li et al. 2000; Myers et al. 1990; Nora et al. 1980; Pohjola-Sintonen et al. 1998). Estimates of the size of the effect have ranged from 1.58 (1.17, 2.13) (Boer et al. 1999), 1.63 (1.06–2.52) (Friedlander et al. 1985), to 2.2 for MI (Colditz et al. 1991) in men, with higher estimates in women, with reported risks ranging from 2.12 (1.11–4.05) (Boer et al. 1999) to 5.0 (2.7 –9.2) (Colditz et al. 1986). Some studies have found no evidence to suggest that the relationship between FHCHD and risk of CHD is modified by classical risk factors, whereas others have found that the effect of FHCHD on risk was modified by smoking (Boer et al. 1999; Khaw et al. 1986). The relationship between FHCHD status and coronary events will have both genetic and environmental components owing to the possibility of familial aggregation of lifestyle. Some studies have reported that recognized risk factors do not account for the risk presented by FHCHD (Friedlander et al. 1998) and that FHCHD does not consistently alter risk associated with other classical risk factors (Hippe et al. 1999).

NPHSII is a prospective study of over 3000 healthy middle-aged UK men, who have to date been followed for CHD events for more than 10 years. As expected, risk of CHD events has been found to be associated with classical lifestyle, lipid and coagulation risk factors (Colditz et al. 1986; Seed et al. 2001), as well as with elevated levels of novel factors such as FXIIa (Zito et al. 2000). With regard to genetic causes of CHD, risk has been found to be significantly associated with elevated levels of Lp(a) (Seed et al. 2001) and with genotypes at the loci coding for lipoprotein lipase (LPL) (Talmud et al. 2000), apolipoprotein E (APOE) (Humphries et al. 2001b), interleukin-6 (IL6) (Humphries et al. 2001a), and PPAR-α (PPARG) (Flavell et al. 2002), with the LPL, APOE and IL6 genotypes showing interactions with smoking on CHD risk. In this study we have used questionnaire data obtained at baseline to classify FHCHD, and have determined the relative risk in men with such a history and explored the extent to which classical risk factors and genetic risk factors explain the risk effect. In particular, the potential interaction between APOE4+ and smoking with FHCHD has been examined, but interactions between rare (LPL) or more modest (IL6) genotypes with smoking and FHCHD have not been considered, due to small numbers within subgroups.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Subjects

The second Northwick Park Heart Study (NPHS II) is a large prospective study of men from nine UK medical practices: Carnoustie, St Andrews, Chesterfield (classified as northern practices) and Camberly, Aylesbury, Harefield, Halesworth, Parkstone and North Mymms (classified as southern practices). The cohort consists of 3,052 healthy middle aged men. There were 3,012 eligible white Europeans, of whom 2,976 had information relating to FHCHD. Full details of the selection criteria have been published previously (Miller et al. 1995, 1996). Eligibility was conditional on the subject not having myocardial infarction (MI), cerebrovascular disease, life threatening malignancy or regular medication with aspirin or anticoagulants. Baseline characteristics, demographic information, and information on FHCHD status were obtained by a questionnaire completed at entry to the study. Family history of CHD status was determined by the answer to the following question: “Has any person in your family ever had a heart attack?” A smoker was any man who had smoked at least one cigarette/day on average for a year or more. Those smokers who had not smoked to this extent in the previous year were categorised as ex smokers. All other men were classified as “never smokers.” Height (m) was measured on a stadiometer and weight (kg) on a balance scale to calculate body mass index (BMI, kg/m2). Survivors have been recalled annually for interview and repeat measurements. A routine ECG was performed at baseline and repeated at the sixth examination. CHD events taken as end-points were fatal (sudden or not) and non-fatal MI, based on World Health Organisation (WHO) criteria (World Health Organization Regional Office for Europe et al. 1976), plus coronary artery surgery and silent MI on the follow-up ECG (in which case the time to event was assumed to have been midway between the baseline and follow-up records). Clinical information for each event was assembled by enquiries through the participating practices, hospitals attended and, for fatal events, coroners' offices. This information was collated and submitted to an independent assessor who assigned qualifying events to the appropriate category. Ethical approval was obtained for the study. The loss to follow-up in the study was 1.2%.

Plasma Measures

A 5ml sample of venous blood was taken by Vacutainer technique (Becton Dickenson, Cowley, Oxford) into a glass tube. Serum was transferred to plastic screw-cap vials (Nunc) and stored at –40°C pending analysis. Cholesterol and triglyceride concentrations were determined by automated enzyme procedures with reagents from Sigma (Poole, Dorset UK) and Wako Chemicals (Alpha Laboratories, Eastleigh UK) respectively. Serum apolipoprotein AI (apoAI) and apolipoprotein B (apoB) concentrations were measured by immunoturbidometry with reagents from Incstar (Wokingham, UK). Lp(a) was determined by an enzyme-linked immunoabsorbant assay (Biopool A.B., Umea, Sweden) (Seed et al. 2001). Clotting factors were measured by standard methods (Zito et al. 2000), and folate and homocysteine levels as described (Dekou et al. 2001).

DNA Extraction and Genotyping

DNA was extracted by the salting-out method (Miller et al. 1988) from 2,743 samples. Genotyping was carried out by PCR and restriction enzyme digestion as described (see references in Table 3 for PCR primer, condition and enzymes used).

Table 3.  Frequency of less common allele in candidate gene variants in NPHSII by FHCHD status
  FHCHD 
GenotypeReferenceNoYesP-value*
  1. *p-value for difference in less common allele frequency (and difference in genotype distribution), via chi-squared tests.

  2. APOE (apolipoprotein E), LPL (lipoprotein lipase), CETP (cholesteryl ester transfer protein), LIPC (hepatic lipase), ACE (angiotensin converting enzyme), FGA/B/G (fibrinogen), F7 (factor VII), MTHFR (methyl tetrahydrofolate reductase), CBS (cystathionine β synthase), MTR (methionine synthase), IL6 (interleukin-6), PPARG (peroxisome proliferator-activated receptor), F12 (factor XII).

APOE
 E2(Humphries et al. 2001b)0.08 (0.07, 0.09) n = 13900.07 (0.06, 0.08) n = 7780.43 (0.08)
 E4 0.16 (0.14, 0.17) n = 13900.15 (0.13, 0.16) n = 7780.71 (0.48)
LPL D9N(Talmud et al. 2000)0.015 (0.01,0.02) n = 15090.014 (0.01, 0.02) n = 8270.96 (0.83)
LPL S447XAs above0.10 (0.09, 0.11) n = 15630.10 (0.09, 0.12) n = 8580.85 (0.60)
CETP Taq1BAs above0.45 (0.43, 0.47) n = 15470.43 (0.40, 0.45) n = 8600.34 (0.49)
LIPC–480C > TAs above0.20 (0.18, 0.21) n = 14840.21 (0.19, 0.23) n = 8220.65 (0.54)
ACE I/D(O'Dell et al. 1995)0.52 (0.50, 0.53) n = 15050.52 (0.50, 0.55) n = 8260.87 (0.56)
FGA/B/G-455G > A(Thomas et al. 1996)0.19 (0.18, 0.21) n = 15550.20 (0.18, 0.22) n = 8420.93 (0.69)
F7 R353Q(Humphries et al. 1996)0.17 (0.16, 0.19) n = 15770.18 (0.16, 0.20) n = 8740.93 (0.35)
MTHFR(Dekou et al. 2001)0.31 (0.30, 0.33) n = 15710.30 (0.28, 0.32) n = 8510.66 (0.66)
CBSAs above0.09 (0.08, 0.10) n = 12000.08 (0.06, 0.09) n = 6450.37 (0.12)
MTRAs above0.18 (0.16, 0.19) n = 7750.20 (0.17, 0.22) n = 4450.48 (0.80)
IL6–174G > C(Humphries et al. 2001a)0.43 (0.41, 0.45) n = 16480.43 (0.41, 0.45) n = 9110.99 (0.69)
PPARG Intron 7(Flavell et al. 2002)0.17 (0.16, 0.19) n = 16280.18 (0.16, 0.20) n = 8970.83 (0.23)
PPARG L162VAs above0.07 (0.06, 0.08) n = 15910.06 (0.05, 0.07) n = 8770.53 (0.53)
F12(Zito et al. 2000)0.25 (0.23, 0.26) n = 15980.26 (0.24, 0.28) n = 8720.73 (0.73)

Statistical Analysis

Statistical analysis was conducted using intercooled stata version 7.0. Of the 3,012 individuals, 149 answered uncertain to the FHCHD question and were not included, and hence the analysis was conducted only with the 2,827 individuals answering either yes or no. The data was reanalysed including those who answered “uncertain” in the group reporting a positive FHCHD and in the group not reporting FHCHD, and findings were similar (results not presented). Comparison of baseline characteristics was made for those with and without a FHCHD of heart disease, using one way ANOVA for continuous variables and the χ2 test for categorical variables. As a result of the non-normality of the distributions of fibrinogen, triglyceride, systolic blood pressure, diastolic blood pressure and body mass index, log transformations were conducted; in order to present the results in more familiar form, geometric means and approximate standard deviations are presented. Throughout the analysis, ex and never smokers are combined into the same group and compared against current smokers. The risk of a coronary event by FHCHD status was considered using Cox proportional hazards modelling. Adjustments were made for additional risk factors by forcing them into the model. Interaction terms between the classical variables and FHCHD were considered by including terms in the survival model, and hence represent deviations from a multiplicative model. Interactions between APOE, smoking and FHCHD have also been conducted using this method, in addition, analysis including interactions between APOE and smoking, stratified according to FHCHD status, were considered. Kaplan Meier plots were used in order to represent the difference in risk over time graphically by FHCHD status. Continuous variables used in the proportional hazards models were standardised to give results in a more comparable form. P-values of less than or equal to 0.05 were considered as statistically significant. Adjustment for multiple comparisons as required were carried out using bonferroni.

RESULTS

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Relationship Between Classical Risk Factors and FHCHD Status

Of the 3,012 eligible individuals, 1,000 answered “yes” and 1,827 answered “no” to the question of “Has any person in your family ever had a heart attack?” No further information such as whether the family member was a first degree relative or at what age they suffered from a heart attack was taken in the baseline questionnaire. Table 1 shows that the majority of baseline characteristics did not differ according to FHCHD status. There was no significant evidence to suggest a difference in the frequency of FHCHD between the nine practices overall (p = 0.06). However, there was a higher frequency of FHCHD in the Northern practices combined compared, to the Southern practices combined (38% vs. 34% p = 0.002).

Table 1.  Mean (SD) for baseline characteristics and classical risk factors of individuals in NPHSII by FHCHD status
 FHCHD 
TraitNo [n = 1827]Yes [n = 1000]P*
  1. *p-values are from ANOVA unless otherwise stated

  2. means given are the anti log of the log transformed mean and standard deviations are approximated

  3. §p-value from chi-squared test

  4. median and interquartile range given, smirnov kolmogorov test applied

  5. #means are the square of the mean of square root transformed values and standard deviations are approximate.

Age (years)56.07 (3.46)55.99 (3.54)0.52
Body mass index (kg/m2)26.18 (3.49)26.24 (3.29)0.68
Smoking Never, ex, current31.6%, 38.9%, 29.45%31.1%, 42.4%, 26.50%0.10 §
Alcohol No units 0, ≤14 > 0, >1422.3%, 66.0%, 11.8%17.4%, 73.3%, 9.3%<0.0005§
Systolic blood pressure (mmHg)136.3 (18.62)137.9 (19.13)0.04
Diastolic blood pressure (mmHG)83.37 (11.43)84.43 (11.36)0.02
Cholesterol (mmol/l)5.73 (1.01)5.76 (1.01)0.38
Triglyceride (mmol/l)1.79 (0.94)1.80 (0.95)0.79
ApoB (mg/dl)0.85 (0.24) n = 15220.87 (0.24) n = 8370.07
HDL (mmol/l)0.80 (0.25) n = 11790.80 (0.23) n = 6520.87
ApoAI (mg/dl)1.64 (0.32) n = 15221.63 (0.31) n = 8370.65
Lp(a) (mg/dL)8.3 (2.69, 24.85)9.2 (2.8, 28.2)0.24
Fibrinogen (g/l)2.72 (0.53)2. 71 (0.48)0.57
Factor VII c (% standard)108.29 (29.23) n = 1774105.8 (27.5) n = 9690.03
Factor VII ag (% standard)128.48 (35.5) n = 1678126.39 (34.7) n = 9170.15
Homocysteine (mmol/l)11.9 (10.2, 14.2) N = 88111.9 (9.8, 14.2) N = 4990.54
Folate (nmol/l)#227.00 (64.72) N = 880234.00 (63.52) N = 4970.053

In those who reported a positive FHCHD compared to those who reported no FHCHD, fewer did not drink at all (17.4% vs. 22.5% p = 0.002). There was significant evidence to suggest a difference in mean Factor VIIc levels by FHCHD status, with those not reporting a FHCHD having 2.3% higher Factor VIIc levels than those reporting a positive FHCHD (p = 0.03). For those reporting a FHCHD mean systolic and diastolic pressure was significantly higher than for those who did not report a positive FHCHD (p = 0.04 and 0.02). In those with a FHCHD compared to those without a FHCHD, there were slightly fewer current-smokers (26.5% and 29.5% respectively) and slightly more ex-smokers (42.4% and 38.9% respectively), though the difference was not significant (p-value 0.13). Folate, but not homocysteine, levels were 3% higher for those with FHCHD compared to those without (p = 0.05). Notably, Lp(a) levels were not significantly higher in the FHCHD men (p = 0.24). After adjustment for multiple comparisons, the only significant difference seen was between alcohol group and FHCHD.

Risk and FHCHD

Of the 2,827 individuals with FHCHD status ascertained, 187 (6.6%) had a coronary event during the 26079 person years of follow up. The frequency of coronary events for those with a FHCHD was 9.0% in comparison to 5.3% for those without. Figure 1 shows risk of CHD over time as a Kaplan Meier plot, for those with and without a positive FHCHD, it is clear that the curves start to diverge early and continue to do so through the follow-up period. As shown in Table 2, using Cox proportional hazards modelling, those with a FHCHD had a 1.71 (95% confidence interval 1.28, 2.28) times higher risk of a CHD event than those with no FHCHD. Adjusting for age, practice, systolic blood pressure, alcohol status, factor VIIc, ApoB, and folate, did not substantially alter the risk estimate for FHCHD (1.90, 95% CI 1.12, 3.21). As shown in Figure 2, the risk for those with a positive family story who were also current smokers was 3.01 compared to non-smokers without FHCHD. This is greater than the risk posed by smoking or FHCHD alone, 2.05 and 1.96 respectively, compared to non smokers without FHCHD, but not significantly different from a multiplicative model (p-value for interaction 0.33).

image

Figure 1. Kaplan Meier survival curve for CHD in NPHSII, split by those with and without a family history.

Download figure to PowerPoint

Table 2.  Hazard Ratios for CHD by FHCHD for NPHSII, unadjusted and after various adjustments
 FHCHD Hazard Ratio (95% confidence interval)
AdjustmentAny CHD eventAcute MI
None1.71 (1.28, 2.28)1.73 (1.23, 2.44)
Age + practice1.73 (1.30, 2.31)1.73 (1.23, 2.44)
Age + practice, BMI, smoking, alcohol1.83 (1.37, 2.44)1.87 (1.32, 2.64)
Age + practice, BMI, smoking, alcohol,1.85 (1.38, 2.48)1.86 (1.31, 2.64)
 cholesterol, triglyceride, fibrinogen,  
Age + practice, BMI, smoking, alcohol,1.86 (1.37, 2.52)1.89 (1.31, 2.72)
 cholesterol, triglyceride, fibrinogen, Lp(a)  
image

Figure 2. Smoking risk in men with and without a FHCHD. Grey symbols are non-smokers, black are smokers. The point estimates and 95% confidence intervals for CHD risk are shown, adjustment was made for both age and practice.

Download figure to PowerPoint

The majority of the variables presented in Table 1 are related to CHD risk (Seed et al. 2001), and since those which differ by FHCHD (i.e. systolic blood pressure, diastolic blood pressure, alcohol and Factor VIIc) may confound or mediate the relationship between FHCHD and risk of heart disease, the relationship between these variables, FHCHD, and CHD, risk was investigated further. As shown in Figure 3, blood pressure at baseline was higher for those who subsequently developed CHD, regardless of FHCHD status; but the difference between baseline blood pressure for those developing CHD, or not, was greater for those without a FHCHD compared to those with a FHCHD (interaction between FHCHD and systolic blood pressure on risk, p = 0.01). Similar results were found for diastolic blood pressure (interaction p = 0.03). After adjustment for systolic blood pressure, the risk posed by a positive FHCHD was 1.67 (95% CI 1.25, 2.23) compared with 1.71 (95% CI 1.28, 2.28) without adjustment. However, there was significant evidence of an interaction between FHCHD and systolic blood pressure on risk. The risk posed by a positive FHCHD and systolic blood pressure was slightly less than would be expected if the two had independent multiplicative effects on risk. For those with average systolic blood pressure the risk associated with FHCHD was 1.87, whereas in those with a family history and one standard deviation above the average systolic blood pressure, the risk was 2.03. Interactions between factor VIIc and FHCHD on risk, or between alcohol status and FHCHD on risk, were not seen (p = 0.58, 0.48).

image

Figure 3. Mean and standard error for systolic blood pressure, by CHD status and family history of CHD status.

Download figure to PowerPoint

Relationship Between Genotypes and FHCHD Status

Differences in the frequency distribution of common polymorphisms in 13 candidate genes so far studied in NPHSII were examined by FHCHD status. The less common allele frequency did not differ between those who reported and did not report a FHCHD for any of the polymorphisms considered (see Table 3). As a result of the previously reported interactions between smoking and APOE, the genotype distribution by FHCHD was considered separately in non-smokers and current smokers, but no differences in the distributions by FHCHD were seen in either group.

For APOE we have previously shown that carriers of the APOE4 allele (APOE4+) had a higher risk of CHD only if they smoke (genotype × smoking interaction p = 0.006 using updated survival data) (Humphries et al. 2001b), and we examined whether the effect of smoking on risk and the interaction with APOE was similar in those with, and without, a FHCHD. There was evidence of significant interaction between APOE and smoking on risk in those with a FHCHD, but not in those without a FHCHD (p = 0.001 and 0.46 respectively). In the group without a FHCHD 9.1% of individuals who were APOE4+ and current smokers had an event compared to 21.4% for APOE4+ smokers with FHCHD. The risk in both of these groups was significantly higher than those who neither smoked nor were APOE4+. As shown in Figure 4, the odds ratio for those men who were APOE4+, a current smoker and reported a positive FHCHD was 5.66 (95% CI 2.91, 10.98), compared to those who did not carry the E4 allele, were non-smokers and reported a negative FHCHD.

image

Figure 4. APOE genotype and smoking risk in men with and without FHCHD, adjusted for age and practice. P-values for interaction between genotype and smoking, were calculated from two separate models, stratified by FHCHD.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

This prospective study of UK middle aged men shows that the risk of CHD carried a hazard ratio of 1.7 for those individuals with a FHCHD, and that this risk was independent of genotypes at 13 candidate gene regions and also a number of classical risk factors. This confirms results from a number of other studies (Boer et al. 1999; Friedlander et al. 1985; Li et al. 2000). The size of the effect in these men is similar to the 1.58 fold higher risk reported in Dutch men (Boer et al. 1999), to the 1.63 fold higher risk in Israel (Friedlander et al. 1985), and to the 1.41 increase reported by Li et al. for white men.

The question used to obtain information on FHCHD was similar to that used in other studies, but relatively unspecific, and did not distinguish between blood and non-blood relatives or between early and late onset CHD. Since the response was based entirely on subject recall it may also not be completely accurate; however, in a population based study of familial cardiovascular disease in middle aged probands in four US communities, the answer given to questions relating to FHCHD by the proband regarding their spouse, parent and sibling were found to be in 87%, 85% and 81% concordance with the actual FHCHD status (Bensen et al. 1999). In addition, as this is a prospective rather than a case-control study, there should be no bias in the recall between the CHD group and those subsequently not having an event. It is however possible that if more precise information was available on the FHCHD (e.g. specified relative, age at time of event), an even higher risk associated with FHCHD might have been observed. It was not possible to ascertain if the group answering “uncertain” were predominantly those with or without FHCHD; however, inclusion of these subjects with either the “yes” or the “no” subjects did not materially effect any of the risk estimates (not shown).

Of the classical risk factors considered here, plasma levels of FVIIc were lower and blood pressure was moderately higher in those with a positive family history status. It has previously been reported that hypertension, BMI, fibrinogen, total cholesterol and C-reactive Protein (CRP) levels are associated with a positive family history of MI (Margaglione et al. 2000), although De Bacquer et al. reported that lifestyle related risk factors did not differ by family history of premature fatal CHD (De Bacquer et al. 1999). Previous studies have found high levels of Lp(a) to be associated with FHCHD (Durrington et al. 1988), but levels were not significantly higher in the NPHSII men. It is possible that the lack of differences seen in the NPHSII men with and without a FHCHD may have resulted from the reduced information obtained by an unverified and relatively unspecific question. If more precise information on the FHCHD of an individual were available, larger differences in baseline risk factors may have been detected.

For both systolic and diastolic blood pressure, mean levels were higher in those with, compared to those without, FHCHD. A European study which considered paternal history of premature CHD found that diastolic blood pressure was higher in females with FHCHD than those without, and found a higher prevalence of personal history of high blood pressure in males (Masana et al. 1996). As well as having a genetic component (Beevers et al. 1980), it is clear that environmental factors such as dietary salt intake play an important part in determining blood pressure (Frost et al. 1991). It is likely that to some extent there may be a familial aggregation of such factors that may contribute to the elevated blood pressure seen in those with FHCHD. However, the blood pressure in those who subsequently went on to have an event was 8.9 mmHg higher (p < 0.0005) in those with no FHCHD, while in those with a history this difference was only 2.3 mmHg and non-statistically significant. One interpretation of these data is that in those with no FHCHD the elevation in blood pressure is playing a more important role in determining CHD events, while in those with a history blood pressure it is less important, suggesting that other factors may be more relevant. The identification of such factors would be of obvious interest. Another possible explanation for blood pressure being higher for those with a FHCHD is that this relationship is confounded by one or more genotypes, and that the alleles related to raising blood pressure are more common in those with a FHCHD. Although this relationship was in no way explained by considering the genotypes that were measured in NPHSII, other unexplained genotypes may be important. Alternatively, this relationship may be due to shared environment between relatives, for example the high blood pressure in those with a FHCHD may be as a result of poor diet, family use of salt and lack of exercise, all of which may tend to cluster in family units.

For none of the 13 genes examined to date in the NPHSII men were there significant frequency differences by FHCHD status. A study from Holland which also considered genotypes found that the APOE4 isoform and the LPL-N9 allele were more frequent in those with a FHCHD compared to those without, and reported that these polymorphisms accounted for a proportion of the risk posed by FHCHD (Boer et al. 2001). Neither of these findings were confirmed in this sample of men. A previous study considering the Factor VII gene HVR4 polymorphism found that the frequency of the less common allele was lower in those with a family history compared to those without a family history (Di Castelnuovo et al. 2000). This was not examined in this sample, but the F7 R353Q genotype showed no frequency difference.

It is possible to propose mechanisms for the observed interaction between FHCHD, smoking and the APOE genotype, although these must be interpreted with caution due to the complexity of the comparisons and the small numbers within certain subgroups. We have previously proposed that the increased risk of CHD in E4 carriers who smoke is due to the poor antioxidant activity of the E4 protein compared to E3 and E2. This poor antioxidant activity has been demonstrated in vitro (Smith et al. 1998) and E4 carriers have higher levels of markers of oxidative stress than others (Chait et al. 1993; Jolivalt et al. 2000). Smoking is a major oxidant stress and E4 smokers may therefore be at greater risk of oxidative damage to small dense LDL particles and to vascular cells, which would increase the progression of atherosclerosis. If this hypothesis is correct, the data suggest that those with a FHCHD might be particularly at risk of oxidative damage, which could be due to both shared environmental factors (e.g. a similar antioxidant-poor diet) and to genetic factors.

Although the genotype and smoking interaction observed here is of potential interest in understanding the pathology of CHD and how these processes are influenced by FHCHD, it must be interpreted with caution. The conclusions are based on relatively small numbers of events, and although some of the risk estimates and interaction terms are statistically significant, this data must be regarded as hypothesis generating, and the observations will need to be confirmed in subsequent studies. For this reason we have not pursued the other smoking-genotype interactions reported in NPHSII men, which involved rarer genotypes (LPL) (Talmud et al. 2000) or more modest effects (IL6) (Humphries et al. 2001a). However, if confirmed, further studies to explore the mechanism of the APOE effects are warranted, and may lead to novel therapeutic approaches such as the use of antioxidants in those with FHCHD.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

NPHSII was supported by the British Medical Research Council, the US National Institute of Health (grant NHLBI 33014) and Du Pont Pharma, Wilmington, USA. SEH, PJT, and EH are all supported by the British Heart Foundation. The following general practices collaborated in the study: The Surgery, Aston Clinton, Upper Gordon Road, Camberley; The Health Centre, Carnoustie; Whittington Moor Surgery, Chesterfield; The Market Place Surgery, Halesworth; The Health Centre, Harefield; Potterells Medical Centre, North Mymms; Rosemary Medical Centre, Parkstone, Poole; The Health Centre, St Andrews.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References
  • Beevers, D. G., Hawthorne, V. M. & Padfield, P. L. 1980. Salt and blood pressure in Scotland. Br Med J 281, 641642.
  • Bensen, J. T., Liese, A. D., Rushing, J. T., Province, M., Folsom, A. R., Rich, S. S. & Higgins, M. 1999 Accuracy of proband reported family history: the NHLBI Family Heart Study (FHS). Genet Epidemiol 17, 141150.
  • Boer, J. M., Feskens, E. J., Kuivenhoven, J. A., Schouten, E. G., Havekes, L. M., Kastelein, J. J., Seidell, J. C. & Kromhout, D. 2001 Parental history of myocardial infarction: lipid traits, gene polymorphisms and lifestyle. Atherosclerosis 155, 149156.
  • Boer, J. M., Feskens, E. J., Verschuren, W. M., Seidell, J. C. & Kromhout, D. 1999 The joint impact of family history of myocardial infarction and other risk factors on 12-year coronary heart disease mortality. Epidemiology 10, 767770.
  • Chait, A., Brazg, R. L., Tribble, D. L. & Krauss, R. M. 1993 Susceptibility of small, dense, low-density lipoproteins to oxidative modification in subjects with the atherogenic lipoprotein phenotype, pattern B. Am J Med 94, 350356.
  • Colditz, G. A., Rimm, E. B., Giovannucci, E., Stampfer, M. J., Rosner, B. & Willett, W. C. 1991 A prospective study of parental history of myocardial infarction and coronary artery disease in men. American Journal of Cardiology 67, 933938.
  • Colditz, G. A., Stampfer, M. J., Willett, W. C., Rosner, B., Speizer, F. E. & Hennekens, C. H. 1986 A prospective study of parental history of myocardial infarction and coronary heart disease in women. Am J Epidemiol 123, 4858.
  • De Bacquer, D., De Backer, G., Kornitzer, M. & Blackburn, H. 1999 Parental history of premature coronary heart disease mortality and signs of ischemia on the resting electrocardiogram. J Am Coll Cardiol 33, 14911498.
  • Dekou, V., Gudnason, V., Hawe, E., Miller, G. J., Stansbie, D. & Humphries, S. E. 2001 Gene-environment and gene-gene interaction in the determination of plasma homocysteine levels in healthy middle-aged men. Thromb Haemost 85, 6774.
  • Di Castelnuovo, A., Orazio, A., Amore, C., Falanga, A., Donati, M. B. & Iacoviello, L. 2000 The decanucleotide insertion/deletion polymorphism in the promoter region of the coagulation factor VII gene and the risk of familial myocardial infarction. Thromb Res 98, 917.
  • Durrington, P. N., Ishola, M., Hunt, L., Arrol, S. & Bhatnagar, D. 1988 Apolipoproteins (a), AI, and B and parental history in men with early onset ischaemic heart disease. Lancet 1, 10701073.
  • Flavell, D. M. Jamshidi, Y., Hawe, E., Pineda Torra, I., Askinen, M.-R., Frick, M. H., Nieminen, M. S., Kesäniemi, Y. A., Pasternack, A., Staels, B., Miller, G., Humphries, S. E., Talmud, P. J. & Syvänne, M. 2002 Peroxisome Proliferator-Activated Receptor Gene Variants Influence Progression of Coronary Atherosclerosis and Risk of Coronary Artery Disease. Circulation 105, 14401445.
  • Friedlander, Y., Kark, J. D. & Stein, Y. 1985 Family history of myocardial infarction as an independent risk factor for coronary heart disease. British Heart Journal 53, 382387.
  • Friedlander, Y., Siscovick, D. S., Weinmann, S., Austin, M. A., Psaty, B. M., Lemaitre, R. N., Arbogast, P., Raghunathan, T. E. & Cobb, L. A. 1998 Family history as a risk factor for primary cardiac arrest. Circulation 97, 155160.
  • Frost, C. D., Law, M. R. & Wald, N. J. 1991 By how much does dietary salt reduction lower blood pressure? II– Analysis of observational data within populations. British Medical Journal 302, 815818.
  • Hippe, M., Vestbo, J., Hein, H. O., Borch-Johnsen, K., Jensen, G. & Sorensen, T. I. 1999 Familial predisposition and susceptibility to the effect of other risk factors for myocardial infarction. J Epidemiol Community Health 53, 269276.
  • Humphries, S., Temple, A., Lane, A., Green, F., Cooper, J. & Miller, G. 1996 Low plasma levels of factor VIIc and antigen are more strongly associated with the 10 base pair promoter (−323) insertion than the glutamine 353 variant. Thromb Haemost 75, 567572.
  • Humphries, S. E., Luong, L. A., Ogg, M. S., Hawe, E. & Miller, G. J. 2001a The interleukin-6 -174 G/C promoter polymorphism is associated with risk of coronary heart disease and systolic blood pressure in healthy men. European Heart Journal 22, 22432252.
  • Humphries, S. E., Talmud, P. J., Hawe, E., Bolla, M., Day, I. N. & Miller, G. J. 2001b Apolipoprotein E4 and coronary heart disease in middle-aged men who smoke: a prospective study. Lancet 358, 115119.
  • Jolivalt, C., Leininger-Muller, B., Bertrand, P., Herber, R., Christen, Y. & Siest, G. 2000 Differential oxidation of apolipoprotein E isoforms and interaction with phospholipids. Free Radic Biol Med 28, 129140.
  • Khaw, K. T. & Barrett-Connor, E. 1986 Family history of heart attack: a modifiable risk factor Circulation 74, 239244.
  • Li, R., Bensen, J. T., Hutchinson, R. G., Province, M. A., Hertz-Picciotto, I., Sprafka, J. M. & Tyroler, H. A. 2000 Family risk score of coronary heart disease (CHD) as a predictor of CHD: the Atherosclerosis Risk in Communities (ARIC) study and the NHLBI family heart study. Genet Epidemiol 18, 236250.
  • Margaglione, M., Cappucci, G., Colaizzo, D., Vecchione, G., Grandone, E. & Di Minno, G. 2000 C-reactive protein in offspring is associated with the occurrence of myocardial infarction in first-degree relatives. Arteriosclerosis, Thrombosis, and Vascular Biology 20, 198203.
  • Masana, L., Farinaro, E., De Henauw, S. & Nicaud, V. 1996 Blood pressure in young adults with and without a paternal history of premature coronary heart disease in Europe: the EARS Study. European Arteriosclerosis Study. J Hum Hypertens 10, 207213.
  • Miller, G. J., Bauer, K. A., Barzegar, S., Cooper, J. A. & Rosenberg, R. D. 1996 Increased activation of the haemostatic system in men at high risk of fatal coronary heart disease. Thromb Haemost 75, 767771.
  • Miller, G. J., Bauer, K. A., Barzegar, S., Foley, A. J., Mitchell, J. P., Cooper, J. A. & Rosenberg, R. D. 1995 The effects of quality and timing of venepuncture on markers of blood coagulation in healthy middle-aged men. Thromb Haemost 73, 8286.
  • Miller, S. A., Dykes, D. D. & Polesky, H. F. 1988 A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 16, 12151215.
  • Myers, R. H., Kiely, D. K., Cupples, L. A. & Kannel, W. B. 1990 Parental history is an independent risk factor for coronary artery disease: the Framingham Study. Am Heart J 120, 963969.
  • Nora, J. J., Lortscher, R. H., Spangler, R. D., Nora, A. H. & Kimberling, W. J. 1980 Genetic–epidemiologic study of early-onset ischemic heart disease. Circulation 61, 503508.
  • O'Dell, S. D., Humphries, S. E. & Day, I. N. 1995 Rapid methods for population-scale analysis for gene polymorphisms: the ACE gene as an example. Br Heart J 73, 368371.
  • Pohjola-Sintonen, S., Rissanen, A., Liskola, P. & Luomanmaki, K. 1998 Family history as a risk factor of coronary heart disease in patients under 60 years of age. European Heart Journal 19, 235239.
  • Seed, M., Humphries, S. E., Ayres, K. L. & Miller, G. J. 2001 Lipoprotein (a) as a predictor for myocardial infarction in the Second Northwick Park Heart Study—a prospective study in middle-aged men. Am J Med Assoc 200, 2227.
  • Smith, J. D., Miyata, M., Poulin, S. E., Neveux, L. M. & Craig, W. Y. 1998 The relationship between apolipoprotein E and serum oxidation-related variables is apolipoprotein E phenotype dependent. Int J Clin Lab Res 28, 116121.
  • Talmud, P. J., Bujac, S., Hall, S., Humphries, S. E. & Miller, G. J. 2000 Substitution of asparagine for aspartic acid at residue 9 (D9N) of lipoprotein lipase markedly augments risk of coronary heart disease in male smokers. Atherosclerosis 149, 7581.
  • Thomas, A. E., Green, F. R. & Humphries, S. E. 1996 Association of genetic variation at the beta-fibrinogen gene locus and plasma fibrinogen levels; interaction between allele frequency of the G/A-455 polymorphism, age and smoking. Clin Genet 50, 184190.
  • World Health Organization Regional Office for Europe and Myocardial infarction Community register. 1976 Public Health in Europe 5.
  • Zito, F., Drummond, F., Bujac, S. R., Esnouf, M. P., Morrissey, J. H., Humphries, S. E. & Miller, G. J. 2000 Epidemiological and genetic associations of activated factor XII concentration with factor VII activity, fibrinopeptide A concentration, and risk of coronary heart disease in men. Circulation 102, 20582062.