OUP user menu

Renal function and cardiovascular mortality in elderly men: the role of inflammatory, procoagulant, and endothelial biomarkers

Sasiwarang Goya Wannamethee, A. Gerald Shaper, Gordon D.O. Lowe, Lucy Lennon, Ann Rumley, Peter H. Whincup
DOI: http://dx.doi.org/10.1093/eurheartj/ehl402 2975-2981 First published online: 28 November 2006


Aims To assess the extent to which inflammatory, procoagulant, and endothelial biomarkers modify the relationship between diminished renal function and cardiovascular mortality.

Methods and results Prospective study of 4029 men aged 60–79 years followed up for a mean period of 6 years, during which 304 cardiovascular deaths occurred. Predicted estimated glomerular filtration rate (eGFR) was used as a measure of renal function. Reduced eGFR was associated with increased prevalence of established cardiovascular risk factors [cardiovascular disease, diabetes, hypertension, left ventricular (LV) hypertrophy, and dyslipidaemia] and higher levels of inflammatory markers [interleukin 6 (IL-6), C-reactive protein], endothelial markers [von Willebrand factor (vWF) and tissue plasminogen activator], activated coagulation markers (fibrin D-dimer), and blood viscosity. Cardiovascular mortality risk increased with decreasing levels of eGFR, particularly among men with eGFR <60 mL/min per 1.73 m2 even after adjustment for established risk factors (adjusted RR 1.49, 95% CI 1.10, 2.03; <60 vs. ≥70 mL/min per 1.73 m2). The association was attenuated after further adjustment for vWF, D-dimer, and IL-6 (adjusted RR 1.34, 95% CI 0.98–1.82).

Conclusion Mild-to-moderate renal insufficiency is associated with significantly increased cardiovascular mortality in elderly men, which is partly explained by the increased prevalence of established risk factors, markers of coagulation, endothelium, and inflammation.

  • Renal function
  • Cardiovascular mortality
  • Inflammation
  • Coagulation
  • Endothelial dysfunction


The prevalence of reduced renal function rises with age, particularly after 70 years.1 Chronic kidney disease (CKD) is a prevalent health problem in the elderly and is associated with increased cardiovascular mortality.2 In recent years, many prospective studies have demonstrated that even mild or moderate deterioration of renal function is an independent risk factor for cardiovascular mortality,212 particularly in high-risk populations such as the elderly28 and in subjects with some form of cardiovascular disease (CVD).2,1114 However, the mechanisms underlying the elevated risk of CVD are not clear. The increased risk is only partially explained by the increased prevalence of established risk factors such as diabetes, hypertension, and dyslipidaemia in subjects with reduced renal function. It is hypothesized that the association between renal dysfunction and cardiovascular risk may be mediated through pathways relating to inflammation, coagulation, and endothelial dysfunction.1517 There is evidence that mild or moderate renal insufficiency is associated with endothelial dysfunction, increased levels of inflammatory biomarkers such as C-reactive protein and interleukin-6 (IL-6), and increased levels of procoagulant markers including factor VIII and fibrin D-dimer,1619 all of which have been shown to be associated with CVD risk in the general population.2024 Whether these factors contribute to the increased risk of CVD in those with renal insufficiency is not well studied. We have examined the association between renal function and cardiovascular mortality in men aged 60–79 years using estimated glomerular filtration rate (eGFR) and have assessed the extent to which inflammatory markers (C-reactive protein, IL-6, and fibrinogen), markers of increased coagulability (factor VIII and fibrin D-dimer), markers of endothelial dysfunction [von Willebrand factor (vWF) and tissue plasminogen activator antigen (t-PA)], and blood viscosity influence the associations observed.


The British Regional Heart Study is a prospective study of CVD involving 7735 men aged 40–59 years selected from the age–sex registers of one general practice in each of 24 British towns, who were screened between 1978 and 1980.25 In 1998–2000, all surviving men, now aged 60–79 years, were invited for the 20th year follow-up measurements, which form the basis of this report. Ethics approval was provided by all relevant local research Ethics Committees. All men provided informed written consent to the investigation, which was carried out in accordance with the Declaration of Helsinki. The men completed a questionnaire (Q20), which included questions on their medical history and lifestyle behaviour. The men were asked to fast for a minimum of 6 h, during which they were instructed to drink only water and to attend for measurement at a specified time between 08:00 and 18:00 h. They then provided a blood sample, collected using the Sarstedt Monovette system. About 4252 men (77% of survivors) were examined. Fasting blood measurements were available in 4045 men at Q20.

Renal function

Serum creatinine was measured in mmol/L and converted to mg/dL by dividing the value by 88.4. Data on serum creatinine were not available in 11 men. The glomerular filtration rate (GFR) is usually regarded as a better overall measure of renal function than serum creatinine, because the latter is determined by a number of factors other than GFR, such as age, muscle mass, and ethnicity.26,27 The GFR was, therefore, estimated from serum creatinine using the Modification of Diet in Renal Disease equation developed by Levy et al.28,29 Almost all the men (99.6%) were white Caucasian. Thus, GFR=186*[(creatinine)**−1.154]*[(age)**−0.203]. eGFR is expressed as mL/min per 1.73 m2. Men with kidney failure according to the National Kidney Foundation definition of an estimated GFR of <15 mL/min per 1.73 m2 (n=5) were excluded.30 The analysis is thus based on 4029 men. Subjects with an estimated GFR between 15 and 59 mL/min per 1.73 m2 are regarded as having CKD.30 Men with no CKD (≥60 mL/min per 1.73 m2) were divided into two further groups on the basis of the eGFR level—60–69 and ≥70 mL/min per 1.73 m2. A further division of subjects with eGFR ≥70 mL/min per 1.73 m2 was not carried out as there was no further decline in mortality at higher eGFR levels; men with eGFR levels of 70–79 and ≥80 mL/min per 1.73 m2 showed similar CVD mortality rates (8.7 vs. 8.3/1000 person years). Thus, three eGFR groups were used in the analyses: 15–59, 60–69, and ≥70 mL/min per 1.73 m2. The association between eGFR and CVD mortality was also assessed using eGFR as a continuous variable.

Cardiovascular risk factors

Details of measurement and classification methods for smoking status, physical activity, body mass index, social class, alcohol intake, blood pressure, blood lipids, blood glucose, and measures of lung function [forced expiratory volume in one second (FEV1) and forced vital capacity] in this cohort have been described.3134 Men were asked to recall a doctor diagnosis of coronary heart disease (myocardial infarction or angina), heart failure, stroke, and diabetes. Pre-existing CVD included men with recall of CHD, stroke, or heart failure (n=903). Prevalent diabetes included men with a diagnosis of diabetes or men with fasting blood glucose ≥7 mmol/L (n=462).

Haemostatic and inflammatory variables

Blood was anticoagulated with K2 EDTA (1.5 mg/mL) for measurement of haematocrit, white cell count, and platelet count in an automated cell counter; plasma viscosity at 37°C in a semi-automated capillary viscometer (Coulter Electronics, Luton, UK). Blood viscosity was calculated from the haematocrit and plasma viscosity as previously described.32 Blood was also anticoagulated with 0.109 M trisodium citrate (9:1 v:v) for measurement of clottable fibrinogen (Clauss method), as well as coagulation factors VII, VIII, and IX in an MDA-180 coagulometer (Organon Teknika, Cambridge, UK). Plasma levels of t-PA antigen and D-dimer were measured with enzyme-linked immunosorbent assay (ELISA) (Biopool AB, Umea, Sweden), as was vWF antigen (DAKO, High Wycombe, UK). C-reactive protein was assayed by ultra sensitive nephelometry (Dade Behring, Milton Keynes, UK). IL-6 was assayed using a high-sensitivity ELISA (R&D Systems, Oxford, UK).


All men have been followed up for all-cause mortality from initial examination (1978–1980) to October 2005 and follow-up has been achieved for 99% of the cohort.35 This analysis is based on follow-up from re-screening in 1998–2000, a mean follow-up period of 6 years (5–7 years). Information on death was collected through the established ‘tagging’ procedures provided by the National Health Service Central Registers. Cardiovascular deaths include all those with ICD codes 400–459.

Statistical analyses

Kaplan–Meier curves were used to construct cumulative mortality (%) curves across the three eGFR groups. Cox's proportional hazards model was used to assess the multivariable-adjusted relative risk for the eGFR. In the adjustment, we included traditional established CV risk factors as well as haemostatic and inflammatory markers that have shown to be associated with CVD mortality. In the multivariable analysis, smoking [never, long-term ex-smokers (>15 years), recent ex-smokers (<15 years), and current smokers], social class (manual vs. non-manual), physical activity (four groups), alcohol intake (five groups), diabetes (yes/no), and pre-existing CVD (yes/no) were fitted as categorical variables. Age, BMI, FEV1, albumin, and the haemostatic and inflammatory markers were fitted as continuous variables; eGFR was fitted as three groups (<60, 60–69, and ≥70 mL/min per 1.73 m2). Tests for trend were carried out fitting eGFR as a continuous variable. The proportionality assumption for the Cox regression model was tested by including an interaction term for eGFR and follow-up time in a time-dependent analysis. There was no evidence that the proportional hazards model had been violated.


The mean estimated GFR at re-screening was 71.95 (std 12.57) mL/min per 1.73 m2. During the mean follow-up period of 6 years, there were 304 deaths from cardiovascular causes (189 CHD deaths, 58 stroke deaths, and 57 deaths from other CVD causes).

Table 1 shows the characteristics in 4029 men by levels of eGFR at re-screening. eGFR of <60 mL/min per 1.73 m2 was present in 630 men (15.7%). Prevalence of CHD, stroke, heart failure, diabetes, hypertension, and left ventricular hypertrophy (LVH) increased significantly with decreasing eGFR. Men with lower eGFR were significantly older, had significantly higher systolic blood pressure, lower HDL-C, lower albumin, and lower lung function, but no association was seen with total serum cholesterol. The inflammatory and procoagulant markers increased with decreasing eGFR, and with the exception of fibrinogen, these associations persisted even after adjustment for lifestyle characteristics and other CV risk factors (Table 2).

View this table:
Table 1

Characteristics of the study population according to estimated GFR levels at baseline re-screening

eGFR (mL/min per 1.73 m2)P-value trend across groups
No. of men63010742325
Age (years)71.3 (5.6)69.3 (5.4)67.7 (5.3)<0.0001
Pre-existing disease
 % CHD (n)27.1 (173)19.3 (208)15.7 (370)<0.0001
 % Stroke (n)9.6 (60)5.5 (63)4.1 (98)<0.0001
 % Heart failure (n)4.2 (23)1.6 (16)1.1 (24)<0.001
 % Diabetes (n)15.5 (97)10.2 (112)10.9 (253)0.01
 % Hypertension (n)38.8 (243)30.4 (329)26.0 (606)<0.0001
 % LVH (n)4.1 (26)1.5 (16)1.1 (26)<0.001
Lifestyle characteristics (Q20)
 % Manual (n)53.6 (338)55.3 (592)53.3 (1235)0.61
 % Current smokers (n)10.0 (63)12.4 (132)13.4 (313)0.03
 % Inactive (n)42.4 (257)35.0 (362)31.8 (706)<0.0001
 % Heavy drinkers (n)3.3 (21)2.7 (29)4.5 (104)0.04
Physical characteristics
 BMI (kg/m2)27.3 (3.5)27.1 (3.6)26.7 (3.7)0.005
 Height (m)171.9 (6.4)172.3 (6.3)172.5 (6.6)0.03
Biological and traditional risk factors (Q20)
 FEV1 (L/min)2.50 (0.65)2.57 (0.62)2.64 (0.67)<0.0001
 Albumin (g/L)43.67 (2.62)43.99 (2.66)44.38 (2.77)<0.0001
 Haemoglobin (g/dL)14.42 (1.31)14.69 (1.17)14.57 (1.14)0.17
 Cholesterol (mmol/L)5.94 (1.09)60.3 (1.05)6.01 (1.09)0.34
 SBP (mmHg)152.9 (26.7)149.5 (24.3)147.8 (23.2)<0.0001
 HDL-C (mmol/L)1.24 (0.33)1.30 (0.32)1.36 (0.35)<0.0001
Inflammation and haemostasis
 WCC (109/L)a7.10 (6.0–8.3)6.75 (5.7–8.0)6.82 (5.7–8.1)<0.0001
 C-reactive protein (mg/L)a2.36 (1.10–4.70)1.77 (0.91–3.34)1.57 (0.74–3.10)<0.0001
 IL-6 (pg/mL)a2.92 (1.79–4.33)2.48 (1.64–3.42)2.32 (1.49–3.29)<0.0001
 Fibrinogen (g/L)3.43 (0.78)3.26 (0.69)3.24 (0.72)<0.0001
 t-PA (ng/mL)11.77 (4.79)11.26 (4.27)10.79 (4.42)<0.0001
 VWF (IU/dL)154.2 (48.9)141.2 (47.9)135.3 (43.7)<0.0001
 Factor VIII (IU/dL)139.3 (32.7)132.4 (32.1)130.2 (30.7)<0.0001
 Blood viscosity (mPa s)3.41 (0.32)3.43 (0.29)3.38 (0.29)0.0002
 D-dimer (ng/mL)a108.9 (58–180)87.4 (51–131)76.71 (46–117)<0.0001
  • All numbers are expressed as mean values (standard deviation) unless stated. n, number of men in parenthesis.

  • aGeometric mean (interquartile range).

View this table:
Table 2

Renal function and mean levels (95% CI) of inflammatory and haemostatic markers adjusted for age, smoking status, physical activity, alcohol intake, BMI, FEV1, albumin, pre-existing CVD, pre-existing diabetes, history of hypertension, LVH, systolic blood pressure, and HDL-cholesterol

eGFR (mL/min per 1.73 m2)P-value trend across groups
No. of men63010742325
Inflammatory and haemostatic variables
 WCC (109/L)a6.95 (6.82, 7.10)6.75 (6.62, 6.89)6.82 (6.75, 6.92)0.25
 C-reactive protein (mg/L)a1.99 (1.82, 2.14)1.69 (1.60, 1.80)1.67 (1.60, 1.73)<0.0001
 IL-6 (pg/mL)a2.61 (2.46, 2.71)2.44 (2.35, 2.52)2.41 (2.35, 2.47)0.001
 Fibrinogen (g/L)3.31 (3.26, 3.37)3.24 (3.20, 3.28)3.27 (3.24, 3.30)0.26
 t-PA (ng/mL)11.22 (10.89, 11.55)11.18 (10.96, 11.44)10.98 (10.79, 11.17)0.02
 VWF (IU/dL)147.3 (143.6, 150.7)139.7 (137.0, 142.0)137.6 (135.8, 139.5)<0.0001
 Factor VIII (IU/dL)135.2 (132.7, 137.7)131.6 (129.8, 133.4)131.4 (130.2, 132.6)0.02
 Blood viscosity (mPa s)3.41 (3.39, 3.43)3.43 (3.41, 3.45)3.38 (3.36, 3.40)<0.0001
 D-dimer (ng/mL)a92.8 (86.5, 98.5)83.9 (80.6, 88.2)82.3 (79.0, 85.6)0.002
  • aGeometric mean.

Figure 1 shows the cumulative mortality incidence in all men and separately in men with and without pre-existing CVD. Men with prevalent CVD had significantly higher mortality rates than men without prevalent CVD overall. Mortality increased with decreasing eGFR in both men with (903 men; 126 cases; 23.7/1000 person years) and without CVD (3126 men; 178 cases; 8.6/1000 person years) (P=0.002 and 0.03 for trend, respectively), with mortality substantially increased at levels <60 mL/min per 1.73 m2. However, absolute CVD mortality rates in men with CKD without prevalent CVD was comparable with mortality rates in men with prevalent CVD and high eGFR (≥70 mL/min per 1.73 m2) (16.2/1000 vs. 16.3/1000 per-years). The age-adjusted RR (95% CI) in men with <60 when compared with those with ≥70 mL/min per 1.73 m2 was 1.78 (1.15, 2.75) and 1.59 (1.08, 2.32) in men with and without CVD, respectively. No significant interaction was seen between the eGFR mortality relationship by cardiovascular status (P=0.91). The analyses were, therefore, carried out in all men with adjustment for prevalent CVD.

Figure 1

Kaplan–Meier survival curve for CVD mortality stratified by the levels of eGFR at baseline re-screening in all men and separately in men with and without prevalent CVD.

Table 3 shows the association between eGFR and cardiovascular mortality with adjustment for confounders and potential mediators. Adjustment for prevalent CVD, diabetes, FEV1, albumin, and lifestyle factors (model 1) attenuated the increased risk, but it remained markedly elevated. Further adjustment for history of hypertension, LVH, systolic blood pressure, and HDL-cholesterol (model 2) reduced the risk further, but there was still almost a 50% increase in risk associated with low eGFR. To assess the contribution of the inflammatory and procoagulant biomarkers, we in turn adjusted for C-reactive protein, IL-6, t-PA, vWF, blood viscosity, and fibrin D-dimer. Fibrin D-dimer and vWF showed the largest effects, followed by inflammatory markers (IL-6 and C-reactive protein) and t-PA. Blood viscosity made minimal contribution to the increased risk. Adjustments for IL-6, vWF, and D-dimer simultaneously attenuated the risk further and the residual risk was of marginal significance [adjusted RR: 1.34 (0.98, 1.02); P=0.07 for <60 vs. ≥70 mL/min per 1.73 m2]. The inverse trend with eGFR was now of marginal significance (P=0.08).

View this table:
Table 3

Renal function and adjusted relative risk of cardiovascular mortality

eGFR mL/min per 1.73 m2Test for trendP-value
All men
No. of cases8894122
Rate/1000 person years23.414.08.5
Age-adjusted RR1.83 (1.38, 2.43)1.37 (1.05, 1.81)1.00<0.0001
Model 11.61 (1.19, 2.17)1.33 (1.01, 1.76)1.000.001
Model 21.49 (1.10, 2.03)1.27 (0.95, 1.68)1.000.008
Model 2+C-reactive protein1.44 (1.06, 1.96)1.24 (1.11, 1.38)1.000.003
Model 2+IL-61.41 (1.04, 1.92)1.25 (0.94, 1.66)1.000.02
Model 2+t-PA1.45 (1.06, 1.97)1.26 (0.96, 1.67)1.000.01
Model 2+vWF1.40 (1.03, 1.98)1.24 (0.93, 1.65)1.000.03
Model 2+D-dimer1.38 (1.02, 1.88)1.23 (0.93, 1.64)1.000.04
Model 2+viscosity1.47 (1.08, 1.99)1.26 (0.94, 1.67)1.000.01
Model 2+D-dimer+vWF1.35 (0.99, 1.86)1.22 (0.91, 1.62)1.000.08
Model 2+D-dimer+vWF+IL-61.34 (0.98, 1.02)1.21 (0.91, 1.61)1.000.08
  • Model 1: adjusted for age, smoking status, physical activity, alcohol intake, BMI, pre-existing CVD, pre-existing diabetes, FEV1, and albumin; model 2: adjusted for model 1 and systolic blood pressure, LVH, history of hypertension, and HDL-cholesterol. Test for trend carried out fitting eGFR in its original continuous form.


It has been suggested that renal insufficiency is associated with increased risk of CVD only in the presence of anaemia.36 Anaemia (haemoglobin <13 g/dL) was present in only 311 men. Exclusion of these men made little difference to the associations seen.


In this study of men aged 60–79 years, the risk of cardiovascular mortality increased with decreasing levels of renal function based on estimated GFR, with risk significantly raised at levels <60 mL/min per 1.73 m2, levels currently defined as CKD.30 This association was seen in both men with and without prevalent CVD. In men without prevalent CVD, absolute CVD mortality rates in those with reduced eGFR (CKD) were similar to mortality rates seen in men with CVD with high eGFR levels (≥70), emphasizing the high risk associated with CKD in elderly men. This supports the contention that patients with CKD should be considered to be in the highest risk group when considering prevention, detection, and treatment of CVD risk factors.2


It has been observed in many studies that reduced eGFR is associated with a high prevalence of CVD, electrocardiographic LVH, diabetes, hypertension, and dyslipidaemia and that these factors explain to some degree the association between renal function and CVD mortality risk.2 In the present study, eGFR is strongly associated with these risk factors, but there still remained a significant 50% increase in risk after adjustment for these established risk factors. Our study confirms previous findings of an association between mild impairment of renal function and cardiovascular mortality independent of traditional risk factors39 and extends previous findings by examining the contribution of inflammation, procoagulant markers, and markers of endothelial dysfunction to this relationship.

Endothelial dysfunction, inflammation, and coagulation

Patients with renal dysfunction have shown to have evidence of endothelial dysfunction and activation of inflammation and blood coagulation.3739 It has been hypothesized that these factors may mediate the association between renal function and cardiovascular risk.1517 We have observed a strong association between renal function and inflammatory markers (C-reactive protein, IL-6, and viscosity), procoagulant markers (fibrin D-dimer), and vWF and t-PA (markers of endothelial dysfunction), consistent with reports from other large population studies in the USA and Europe.9,16,17

The results suggested that vWF (a marker of endothelial dysfunction), fibrin D-dimer (procoagulant marker), and IL-6 (proinflammatory cytokine) could be intermediate markers in the renal function–cardiovascular mortality relationship, which was appreciably reduced in strength after adjustment for them. In contrast, C-reactive protein and t-PA appeared less important in explaining the association between reduced GFR and CVD mortality. This is consistent with previous studies that have shown renal insufficiency to be associated with CVD, independent of C-reactive protein.9 IL-6 appeared to contribute more to the renal insufficiency–CVD mortality relationship than C-reactive protein, possibly because it may be a strong predictor of progressive atherosclerosis.40

Endothelial dysfunction may be an important mechanism, linking mildly impaired renal function to CVD. Endothelium-dependent vasodilation appears to be impaired in patients with CKD.41 Endothelium normally has anti-atherothrombotic properties such as promotion of vasodilation and inhibition of vascular smooth muscle cell proliferation, thrombosis, and inflammatory activity.9 vWF, which has prothrombogenic properties through its involvement in platelet adhesion and aggregation and in transport of blood coagulation factor VIII,42 has been suggested to be a marker of generalized endothelial dysfunction43 and has been associated with increased risk of CHD.23,24 Our findings are consistent with those of the Hoorn Study, which suggested that endothelial dysfunction contributed to the renal function–CVD mortality relationship,9 whereas C-reactive protein played a minimal role in explaining the excess risk. D-dimer has been shown to be associated with increased prevalence of CVD in patients on chronic haemodialysis,39 but its contributing role to CVD risk in subjects with mild renal insufficiency has not been studied. Fibrin D-dimer reflects activation of the coagulation system, thrombin formation, and turnover of cross-linked intravascular fibrin and is associated with increased risk of coronary heart disease.20,44 The findings of this study suggest that the coagulation system is activated in those with mild renal insufficiency and is an important contributor to CVD mortality in these men.

Strengths and limitations

GFR estimated from serum creatinine may not be as accurate a measure of renal function as direct measures of GFR. However, direct measures of renal function are not feasible in large epidemiological studies.45 We used the Modification of Diet in Renal Disease Study Group equation to estimate GFR, which takes into account age, and it has been validated in large populations.28,29 Repeating the analyses by applying the Cockcroft–Gault formula, an alternative predictor equation for GFR,46 yielded very similar results. As measures of inflammation, coagulability, and endothelial dysfunction were obtained at the same time as eGFR, we cannot assess whether renal dysfunction leads to these abnormalities or is a consequence of these abnormalities. It is also possible that these biomarkers may be confounding factors, rather than mediating factors of the eGFR–CVD association. The slight excess risk of CVD mortality still seen in those with reduced eGFR after analysis with full adjustment may reflect a true association between renal dysfunction and risk of CVD mortality independent of these factors which would be consistent with the findings of other studies including the Valsartan in Acute Myocardial Infarction Trial (VALIANT)47 or may be due to residual confounding by established and/or novel risk factors. Although vWF is a valid marker of endothelial damage,43 it is possible that adjustment for more detailed measures of endothelial dysfunction would attenuate the association further. Moreover, subjects with reduced eGFR may have had more severe hypertension as reflected by the high prevalence of LVH or dyslipidaemia and may therefore have suffered more vascular damage secondary to hypertension or dyslipidaemia.


CVD mortality increased with declining levels of estimated GFR, with a substantial increase in those with CKD (eGFR <60 mL/min per 1.73 m2). We confirmed that CKD is common in the elderly and is associated not only with high prevalence of CVD, diabetes, hypertension, and dyslipidaemia, but also with increased inflammation, endothelial dysfunction, and hypercoagulability. Established risk factors, together with markers of inflammation, activated coagulation, and endothelial dysfunction, determine to a large extent the excess risk seen in those with CKD. These findings have important implications for the prevention of CVD in those with mild renal insufficiency. Although the control of blood pressure and cholesterol levels is important in this group, the results of the present study raise the possibility that medications with effects on the endothelium and inflammatory pathways (particularly perhaps statins and antiplatelet drugs)48 could be of particular benefit in these high risk men.


The British Regional Heart Study is a British Heart Foundation Research Group and also receives support from the Department of Health (UK). The measurements and laboratory analyses reported here were supported by the British Heart Foundation Project Grants PG97012 and PG97027. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health.

Conflict of interest: none declared.


View Abstract