Abstract

Objective. To compare prevalent cardiovascular disease, conventional cardiovascular risk factors and thrombotic variables in a cohort with well‐controlled rheumatoid arthritis (RA) and in population controls.

Methods. Seventy‐six RA patients and 641 controls, randomly sampled from the local population in the North Glasgow MONICA study. Conventional cardiovascular risk factors (blood pressure, smoking, cholesterol) and thrombotic variables [fibrinogen, von Willebrand factor (vWF), tissue plasminogen activator antigen (t‐PA), fibrin D‐dimer, plasminogen activator inhibitor (PAI‐1), plasma viscosity] were measured by standard procedures.

Results. RA patients had a significantly higher prevalence of angina pectoris (P=0.03). Stroke also tended to be more common in the RA group, but the difference did not reach statistical significance (P=0.08). Diastolic blood pressure was significantly higher and serum cholesterol significantly lower in the RA group than in controls. Current smoking habits and exercise history were similar in the two groups, although RA patients were more likely to have previously smoked. Significant elevations in several thrombotic predictors of cardiovascular disease (fibrinogen, vWF, t‐PA antigen and fibrin D‐dimer) were found in the RA group.

Conclusions. In this RA patient population, diastolic blood pressure was higher than in controls and thrombotic variables were elevated compared with controls. These features are identified as potential additional cardiovascular risk factors in the RA patients studied. Prospective studies of risk modification may permit the identification of factors which could lead to a reduction in the known increased cardiovascular risk in RA.

Rheumatoid arthritis (RA) patients have an increased mortality (standardized mortality rate 1.4–3.0) [1, 2] and die on average 2.5 yr earlier in community‐based studies [3] and up to 18 yr earlier in hospital‐based cohorts [4] than the general population. Fifty per cent of deaths are due to cardiovascular disease [4]. Approximately the same proportion of the general population die of cardiovascular disease, but the reason RA patients are at risk of accelerated cardiovascular disease is not clear.

Traditional risk factors for cardiovascular disease (hypertension, smoking, hypercholesterolaemia) account for about 50% of all coronary heart disease events [5] in the general population. Several thrombotic variables have also been linked to the future risk of developing myocardial infarction or stroke. Fibrinogen [6] is the best established of these, but consistent associations have also been found with von Willebrand factor (vWF; a cofactor in platelet adhesion/aggregation) [7], tissue plasminogen activator antigen (t‐PA; a key enzyme in plasmin formation and fibrinolysis) [8] and fibrin D‐dimer (a marker of increased fibrin turnover) [9]. There is little published information on thrombotic variables in patients with RA, and no published studies comparing patients with controls from random samples of the local population, adjusting for the effects of conventional risk factors.

The aims of our study were: (i) to compare prevalent cardiovascular disease in a population of patients with RA and in population controls, and (ii) to compare potential risk factors for cardiovascular disease, including thrombotic variables.

Methods

Study and control populations

The study population comprised 76 consecutive patients (63 women, 13 men) aged 40–65 yr diagnosed with RA at the rheumatology clinic in a Glasgow teaching hospital. All patients lived in North Glasgow (defined by postcode area) and RA was diagnosed using the American College of Rheumatology (ACR) criteria.

The control population consisted of 641 men and women of the same age range as the cases, and were randomly sampled during 1989–92 from the same North Glasgow population and enrolled in the Scottish arm of the WHO MONICA study, a study of cardiovascular risk factors including thrombotic variables.

Patients and controls completed a questionnaire recording demographic data, medical history, drug history, amount of exercise per week and smoking history. The percentage of patients and controls exercising for 20 min at least once a week is presented. Angina was recorded if this condition had been diagnosed by a general practitioner or hospital physician. The Rose chest pain score [10] was used to differentiate cardiac from non‐cardiac chest pain. In RA patients, the disease duration, presence/absence of rheumatoid factor, erosive changes on X‐ray [11] and Health Assessment Questionnaire (HAQ) score (a measure of functional capacity) [12] were recorded.

Measurements

Blood pressure, weight and height were measured as outlined in the MONICA study protocol [13]. Measurements were made by the same individual. Training and testing of the technique for blood pressure measurement was undertaken using a video used in the MONICA study to standardize the technique as much as possible. Weight and height were used to derive body mass index (BMI) (weight/height2). Non‐fasting serum cholesterol and triglyceride levels were assayed by routine methods (enzyme assay kit, Roche Pharmaceuticals). Plasma fibrinogen (Clauss assay), vWF antigen, t‐PA antigen, fibrin D‐dimer, plasminogen activator inhibitor (PAI‐1) activity and plasma viscosity (37°C) were assayed as reported previously [14]. Erythrocyte sedimentation rate (ESR) and C‐reactive protein (CRP) were measured in RA patients only.

Data analysis

Generalized linear models were used to compare patients and controls. Logistic regression was used to analyse the binary variables, adjusting for age, sex, smoking status (never smoked, ex‐smoker, current smoker) and number of cigarettes currently smoked per day. Some of the continuous variables were highly skewed, and consequently quartiles were computed as summary statistics. Significance tests were performed using general linear models, adjusting as above and for the highest level of education achieved (degree, professional, secondary, primary). BMI, fibrinogen, vWF, t‐PA and D‐dimer were analysed on the scale: PAI‐1 was analysed after raising to the power 0.1, this being the transformation that produced the best approximation to a normal shape.

Results

RA patient data (Table 1)

The median age of the RA cohort was 57 yr and median disease duration was 12.5 yr. Ninety‐seven per cent of patients had erosive X‐ray changes and 96% were rheumatoid factor‐positive. Sixty‐two of 76 patients were on disease‐modifying anti‐rheumatic drug (DMARD) therapy: 24 were on sulphasalazine, 14 on intramuscular gold, eight on hydroxychloroquine, nine on methotrexate, and seven on other drugs. Only one patient was on oral corticosteroids. At the time of this study, disease control was good, as shown by a median ESR of 22 mm/h (interquartile range 12–40), and a median CRP of 8 mg/l (interquartile range 6–22).

Table 1.

RA patient data (median, interquartile range)

Number of patients 76 (63 female, 13 male) 
Disease duration (yr) 12.5 (6–21) 
Rheumatoid factor‐positive 96% 
Erosive 97% 
HAQ score 2.12 (1.6−2.5) 
ESR (mm/h) 22 (12–40) 
CRP (mg/l) 8 (6–22) 
Number of patients 76 (63 female, 13 male) 
Disease duration (yr) 12.5 (6–21) 
Rheumatoid factor‐positive 96% 
Erosive 97% 
HAQ score 2.12 (1.6−2.5) 
ESR (mm/h) 22 (12–40) 
CRP (mg/l) 8 (6–22) 

Prevalent cardiovascular disease (Table 2)

RA patients were more likely to have a diagnosis of angina compared with controls (P=0.03). Stroke also tended to be more common in the RA group, but this did not reach statistical significance. The Rose chest pain questionnaire to distinguish cardiac chest pain was positive in 30% of RA patients compared with 15% of controls (P=0.0007).

Table 2.

Prevalent cardiovascular disease in RA patients and controls


 
RA patients
 
Controls
 
Pa
 
Number of patients 76 641  
Angina 18% 8% 0.03 
Myocardial infarction 7% 5% 0.65 
Stroke 5% 1% 0.08 
Diabetes 1% 3% 0.70 
Rose chest pain 30% 15% 0.0007 

 
RA patients
 
Controls
 
Pa
 
Number of patients 76 641  
Angina 18% 8% 0.03 
Myocardial infarction 7% 5% 0.65 
Stroke 5% 1% 0.08 
Diabetes 1% 3% 0.70 
Rose chest pain 30% 15% 0.0007 

aDifference between patients and controls after adjustment for age, sex, smoking status (current smoker, ex‐smoker, never smoked) and cigarettes per day (currently).

Major cardiovascular risk factors (Table 3)

RA patients had significantly higher diastolic blood pressure (DBP) and significantly lower serum total cholesterol than controls. Twenty per cent of the RA group were on antihypertensive therapy compared with 13% of controls (P=0.91). One per cent of both groups were on lipid‐lowering therapy. No statistically significant differences in systolic blood pressure (SBP), current smoking, exercise level, triglycerides or BMI were observed. RA patients were, however, more likely to have ever smoked (80% of RA group, 75% of control population, P=0.003).

Table 3.

Major cardiovascular risk factors in RA patients and controls (median, interquartile range)


 
RA patients
 
Controls
 
Pa
 
Number of patients 76 641  
SBP (mmHg) 138 (123−1.5) 132 (120−147) 0.19 
DBP (mmHg) 83 (73–93) 80 (70–88) 0.02 
Exercise (once/week) 26% 34% 0.40 
Current smokers 45% 45% 0.86 
Ever smokers 80% 75% 0.003 
BMI (kg/m225.7 (22–29) 25.7 (23–29) 0.91 
Cholestrol (mmol/l) 5.3 (4.6−6.3) 6.2 (5.4−7.0) <0.0001 
Triglycerides (mmol/l) 1.5 (1.0−2.3) 1.7 (1.2−2.4) 0.57 

 
RA patients
 
Controls
 
Pa
 
Number of patients 76 641  
SBP (mmHg) 138 (123−1.5) 132 (120−147) 0.19 
DBP (mmHg) 83 (73–93) 80 (70–88) 0.02 
Exercise (once/week) 26% 34% 0.40 
Current smokers 45% 45% 0.86 
Ever smokers 80% 75% 0.003 
BMI (kg/m225.7 (22–29) 25.7 (23–29) 0.91 
Cholestrol (mmol/l) 5.3 (4.6−6.3) 6.2 (5.4−7.0) <0.0001 
Triglycerides (mmol/l) 1.5 (1.0−2.3) 1.7 (1.2−2.4) 0.57 

aDifference between patients and controls after adjustment for age, sex, smoking status (current smoker, ex‐smoker, never smoked), cigarettes per day (currently) and education level.

Thrombotic variables (Table 4)

Significantly higher plasma levels of fibrinogen, vWF, t‐PA and D‐dimers were identified in the RA group (P<0.0001). Plasma PAI‐1 levels were not significantly different between the RA and control groups, while plasma viscosity was slightly lower in RA patients (P=0.05).

Table 4.

Thrombotic variables in RA patients and controls (median, interquartile range)


 
RA patients
 
Controls
 
Pa
 
Fibrinogen (g/l) 4.3 (3.4–5.2) 2.8 (2.5–3.4) <0.0001 
vWF (IU/dl) 159 (124–2.2) 93 (72–115) <0.0001 
t‐PA (ng/ml) 7.3 (5.6–9.9) 5.0 (3.5–7.5) <0.0001 
D‐dimer (ng/ml) 148 (90–2.4) 56 (42–78) <0.0001 
PAI (% pool) 77 (61–91) 81 (65–1.0) 0.43 
Plasma viscosity (mPa s) 1.28 (1.23–1.36) 1.31 (1.26–1.36) 0.05 

 
RA patients
 
Controls
 
Pa
 
Fibrinogen (g/l) 4.3 (3.4–5.2) 2.8 (2.5–3.4) <0.0001 
vWF (IU/dl) 159 (124–2.2) 93 (72–115) <0.0001 
t‐PA (ng/ml) 7.3 (5.6–9.9) 5.0 (3.5–7.5) <0.0001 
D‐dimer (ng/ml) 148 (90–2.4) 56 (42–78) <0.0001 
PAI (% pool) 77 (61–91) 81 (65–1.0) 0.43 
Plasma viscosity (mPa s) 1.28 (1.23–1.36) 1.31 (1.26–1.36) 0.05 

aDifference between patients and controls after adjustment for age, sex, smoking status (current smoker, ex‐smoker, never smoked), cigarettes per day (currently) and education level.

Discussion

We found a higher prevalence of angina in RA patients compared with controls. In addition, significantly more RA patients (30%) were Rose chest pain positive than controls (15%) (P=0.0007 after adjustment for age and sex). RA patients were more likely to have suffered a stroke than the control population, but this did not reach statistical significance. To our knowledge, the prevalence of cardiovascular and cerebrovascular disease in RA has not been reported previously. However, these results are consistent with previous reports of increased cardiovascular risk in RA [14].

Our study identified differences in both traditional cardiovascular risk factors and thrombotic variables associated with cardiovascular disease in the RA population. RA patients had a trend towards higher SBP, and significantly higher DBP than the control population. Garnero et al. [15] recently found an increased prevalence of hypertension in 88 RA sufferers compared with 72 age‐ and sex‐matched controls. One possible reason may be the widespread use of non‐steroidal anti‐inflammatory drugs (NSAIDs) in this population. NSAIDs, especially indomethacin and piroxicam, cause a rise in mean blood pressure of approximately 5 mmHg [16]. Seventy‐six per cent of our RA patients were on NSAIDs (11% on indomethacin, none on piroxicam). Renal vasculitis and amyloidosis are rare causes of hypertension in RA. No such case was identified in our study.

It is now widely recognized that even modest improvement in blood pressure leads to substantial reductions in cardiovascular mortality. Reductions of 12–13 mmHg in SBP [17] and 5–6 mmHg in DBP [18] are associated with a 25% reduction in coronary heart disease and a proportional reduction in stroke in the general population. Our data and those of others [15] suggest that more attention could be paid to blood pressure control in RA patients.

Evaluation of lipids in this study indicated a significantly lower serum cholesterol level in the RA patient group (median 5.3 vs 6.2 mmol in controls) but no difference in triglyceride level. This pattern has been reported in several RA studies, both in patients on and off treatment with DMARDs [1922]. When more detailed lipid analysis and lipoprotein patterns are studied, however, inconsistencies arise between studies. Our study was not designed to measure lipoprotein levels. A reduction in cardioprotective high‐density lipoprotein (HDL) cholesterol level in RA is reported by all authors, but some report a reduction in low‐density lipoprotein (LDL) cholesterol, while others find higher levels of LDL cholesterol. Park et al. [19] recently reported a lower total cholesterol level in the RA group in their study of lipid profiles in 42 untreated RA patients vs controls. However, more detailed lipid analysis showed significantly less HDL cholesterol and higher LDL cholesterol and lipoprotein(a) levels in the RA group. This atherogenic profile may expose RA patients to higher risk of cardiovascular disease. Interestingly, they also found a significant negative correlation between CRP and HDL cholesterol and apolipoprotein A‐I levels, supporting the theory that lipoprotein levels in RA are influenced by the inflammatory process. The DMARD hydroxychloroquine has been noted to have beneficial effects on the lipid profile by increasing the level of HDL cholesterol [23]. In our study, RA patients on hydroxychloroquine had a lower median cholesterol (4.4 mmol/l) than those on other DMARD therapy.

Fewer (26%) of the RA group exercised once per week for at least 20 min compared with controls (34%). This difference was not statistically significant despite the limited functional capacity of the RA patients, as delineated in their HAQ score (median 2.12). This may reflect either the sedentary lifestyle of the control population or the difficulty in quantifying exercise capacity accurately. Forty‐five per cent of both groups were current smokers. However, the RA patients were significantly more likely to have ever smoked.

We found significant elevations in plasma fibrinogen, vWF, t‐PA antigen and fibrin D‐dimer in RA patients compared with controls. Prospective epidemiological studies consistently show an association between fibrinogen and cardiovascular disease [6]. The median fibrinogen level in our RA group was 4.3 g/l, which from prospective studies would translate into an increased risk of cardiovascular disease if the relationship was causal [6]. However, other markers of inflammation, such as CRP, white cell count and low serum albumin, are also associated with increased cardiovascular risk [6]. Hence, a causal role for increased plasma fibrinogen in cardiovascular disease remains to be proven by prospective randomized trials of its reduction. Preliminary data using bezafibrate to lower plasma fibrinogen levels show a reduction in cardiac death and stroke in patients with cardiovascular disease [24], and other trials of fibrinogen reduction in the prevention of cardiovascular disease are in progress [25].

There is increasing evidence that increased plasma levels of vWF [7, 9], t‐PA antigen [8] and fibrin D‐dimer [9, 26] are also associated with increased risk of coronary heart disease and stroke. Such studies suggest roles for these thrombotic variables in cardiovascular disease, e.g. by promotion of platelet adhesion and aggregation (vWF) and up‐regulation of fibrinolytic activity (t‐PA, D‐dimer), perhaps as a response to fibrin formation [27]. We have shown that even well‐controlled patients with RA have significant elevations of these thrombotic variables, which are predictive of cardiovascular disease. Wallberg‐Jonsson et al. [26] have also measured thrombotic variables in RA patients. They studied 74 RA patients with mean age 63.5 yr and mean disease duration 21.9 yr. Thirty‐six patients were on corticosteroid therapy. Patients were followed up for 8 yr for cardiovascular events. vWF, PAI‐1, ESR and haptoglobin were significantly elevated in those patients who sustained a cardiovascular event. Are these associations causal or non‐causal?

Ninety‐six per cent of our RA patients were rheumatoid factor‐positive, which in some individuals is known to be associated with vascular injury (vasculitis). Such injury may increase plasma vWF and t‐PA antigen levels by inducing vascular endothelial disturbance, which may be prothrombotic (e.g. by promoting platelet–vessel wall interaction, in which vWF is an important cofactor). Up‐regulation of the fibrinolytic system, as shown by increased levels of t‐PA antigen and fibrin D‐dimer, may arise as a response to fibrin formation, either intravascularly or extravascularly. Even if this is extravascular (e.g. in RA), intravascular increases in plasma fibrinogen and vWF may still be thrombogenic e.g. by increasing platelet adhesion and aggregation [27].

What are the clinical implications of our findings? We have confirmed that patients with RA have an increased prevalence of cardiovascular disease (angina and stroke), which could be partly explained by higher blood pressure, possibly induced by NSAIDs. Tighter control of blood pressure might therefore reduce the risk of cardiovascular disease in RA patients. The substantial elevations in thrombotic potential risk factors in RA suggest prospective evaluation of anti‐thrombotic therapies, e.g. aspirin or other anti‐platelet agents. Oral anti‐coagulants, e.g. warfarin, which are effective in reducing the risk of coronary heart disease, appear less practical in RA patients, because of the risk of potentiating NSAID‐induced gastrointestinal bleeding. The role of fibrinogen‐reducing agents remains to be established in ongoing trails. Finally, the role of lifestyle advice (e.g. stopping smoking, exercise) should not be ignored; there is increasing evidence that an unhealthy lifestyle is related to thrombotic variables in the general population [14].

Correspondence to: A. McEntegart.

The authors thank Miss Elaine Smith for typing this manuscript.

References

1
Mutru O, Laakso M, Isomaki H, Koota K. Ten year mortality and causes of death in patients with rheumatoid arthritis.
Br Med J
 
1985
;
290
:
1797
–9.
2
Prior P, Symmons DPM, Scott DL, Brown R, Hawkins CF. Cause of death in rheumatoid arthritis.
Br J Rheumatol
 
1984
;
23
:
92
–9.
3
Myllykangas‐Luosujärvi R, Aho K, Kautiainen H, Isomäki H. Cardiovascular mortality in women with rheumatoid arthritis.
J Rheumatol
 
1995
;
22
:
1065
–7.
4
Wolfe F, Mitchell DM, Sibley JT et al. The mortality of rheumatoid arthritis.
Arthritis Rheum
 
1994
;
37
:
481
–94.
5
Heller RF, Chinn S, Tunstall‐Pedoe HD, Roase G. How well can we predict coronary heart disease?
Br Med J
 
1984
;
288
:
410
–1.
6
Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C‐reactive protein, albumin or leukocyte count with coronary heart disease.
J Am Med Assoc
 
1998
;
279
:
1477
–82.
7
Meade TW, Cooper JA, Stirling Y, Howarth DJ, Ruddock V, Miller GJ. Factor VIII, ABO blood group and the incidence of ischaemic heart disease.
Br J Haematol
 
1994
;
88
:
601
–7.
8
Juhan‐Vague I, Pyke SD, Alessi MC, Jespersen J, Haverkate F, Thompson SG. Fibrinolytic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris.
Circulation
 
1996
;
94
:
2057
–63.
9
Lowe GDO, Yarnell JWG, Sweetnam PM, Rumley A, Thomas HF, Elwood PC. Fibrin D‐dimer, tissue plasminogen activator, plasminogen activator inhibitor and the risk of major ischaemic heart disease in the Caerphilly study.
Thromb Haemost
 
1998
;
79
:
129
–33.
10
Rose GA. The diagnosis of ischaemic heart pain and intermittent claudication in field surveys.
Bull World Health Organ
 
1967
;
27
:
645
–58.
11
Larsen A. How to apply Larsen score in evaluating radiographs of rheumatoid arthritis in long‐term studies.
J Rheumatol
 
1995
;
22
:
1974
–5.
12
Fries JF, Spitz P, Kraines RG, Holman HR. Measurement of patient outcome in arthritis.
Arthritis Rheum
 
1980
;
23
:
137
–45.
13
WHO MONICA Project Part III. Population survey; Section 1: Population survey data component. In: MONICA manual, Geneva: World Health Organization,
1990
.
14
Yarnell JWG, Sweetnam PM, Rumley A, Lowe GDO. Lifestyle and haemostatic risk factors for IHD: the Caerphilly study.
Arterioscler Thromb Vasc Biol
 
2000
;
20
:
271
–9.
15
Garnero A, Fasciolo D, Accarrdo S, Seriolo B. Cardiac risk factors in patients with rheumatoid arthritis [abstract].
Ann Rheum Dis
 
1999
;
(Suppl.)
:
69
–70.
16
Chou CM. Evaluation and treatment of hypertension.
Rheum Dis Clin North Am
 
1999
;
25
:
521
–37.
17
Whelton PK, He J. Elevated systolic blood pressure and risk of cardiovascular and renal disease; overview of evidence from observational epidemiological studies and randomized controlled trials.
Am Heart J
 
1999
;
138
:
211
–9.
18
Collins R, Peto R, MacMahon SW et al. Blood pressure, stroke and coronary heart disease. II: short‐term reductions in blood pressure: overview of randomised drug trials in their epidemiological context.
Lancet
 
1990
;
335
:
827
–38.
19
Park YB, Lee SK, Lee WK et al. Liquid profiles in untreated patients with rheumatoid arthritis.
J Rheumatol
 
1999
;
26
:
1701
–4.
20
Rantapää‐Dahlqvist S, Wållberg‐Jonsson S, Dahlēn G. Lipoprotein (a), lipids and lipoproteins in patients with rheumatoid arthritis.
Ann Rheum Dis
 
1991
;
50
:
366
–8.
21
Lazarevic MB, Vitic J,  Mladenovic V, Myones BL, Skosey JL, Swedler WI. Dyslipoproteinaemia in the course of active rheumatoid arthritis.
Semin Arthritis Rheum
 
1992
;
22
:
172
–80.
22
Lorber M, Aviram M, Linn S, Scharf Y, Brook JG. Hypocholesterolaemia and abnormal high‐density lipoprotein in rheumatoid arthritis.
Br J Rheumatol
 
1985
;
24
:
250
–5.
23
Munro R, Morrison E, McDonald AG, Hunter JA, Madhok R, Capell HA. Effect of disease modifying agents on the lipid profiles of patients with rheumatoid arthritis.
Ann Rheum Dis
 
1997
;
56
:
374
–7.
24
Behar S. Lowering fibrinogen levels: clinical update. BIP Study Group. Bezafibrate infarction prevention.
Blood Coagul Fibrinolysis
 
1999
;
10(Suppl.)
:
S41
–3.
25
MacCallum PK, Meade TW. Haemostatic function, arterial disease and the prevention of arterial thrombosis.
Baillière's Best Pract Clin Haematol
 
1999
;
12
:
577
–99.
26
Wallberg‐Jonsson S, Cederfelt M. Rantapaa‐Dahlqvist S. Haemostatic factors and cardiovascular disease in active rheumatoid arthritis: an 8 year followup study.
J Rheumatol
 
2000
;
27
:
71
–5.
27
Lowe GDO, Rumley A. Use of fibrinogen and fibrin D‐dimer in prediction of arterial thrombotic events.
Thromb Haemost
 
1999
;
82
:
667
–72.

Comments

0 Comments