Diastolic heart function in RA patients
- First Online:
- Cite this article as:
- Wislowska, M., Jaszczyk, B., Kochmański, M. et al. Rheumatol Int (2008) 28: 513. doi:10.1007/s00296-007-0473-8
- 75 Views
The results of some epidemiological studies point to the presence of an increased risk of cardiovascular disease (CVD), particularly atherosclerosis and congestive heart failure (CHF) in rheumatoid arthritis (RA). At least 50% of abnormalities remained asymptomatic. Pathological conditions contributing to myocardial dysfunction such as high serum levels of IL-6, C-reactive protein (CRP) and TNF alpha are present both in RA and CHF patients. The most common pathological mechanism leading to the development of heart failure is left ventricular (LV) diastolic dysfunction, which remains clinically asymptomatic for a long time. The aim of this study was to assess the systolic and diastolic functions of the LV in RA patients without clinically evident cardiovascular disease, using pulsed Doppler echocardiography. Our purpose was also to estimate whether there is a correlation between the duration and severity of RA and the degree of LV diastolic dysfunction. A comparison of the average values of echocardiographic measurements was made between the RA group and control group, which constituted healthy volunteers. Left ventricular mass index in RA group was significantly greater than in the control group (105.2 ± 32.6 vs. 87.9 ± 16.8; p < 0.05) so were the interventricular septum end-diastolic thickness (1.01 ± 0.33 vs. 0.86 ± 0.12; p < 0.05), the LV posterior wall end-diastolic thickness (0.94 ± 0.08 vs. 0.83 ± 0.11; p < 0.0001) and the aortic root diameter (3.18 ± 0.31 vs. 3.10 ± 0.63, p < 0.001). The ejection fraction in RA group was significantly lower than in the control group (64.4 ± 1.3 vs. 66.3 ± 1.3; p < 0.0001). The assessment of diastolic function parameters revealed significantly longer isovolumetrc relaxation time (IVRT) and shorter deceleration time (DT) in RA patients compared to the control group. Patients in stage II or III revealed significantly lower LV mass index (99 ± 17 vs. 131 ± 42; p < 0.05) and the interventricular septum end-diastolic thickness (0.94 ± 0.10 vs. 1.28 ± 0.5; p < 0.05) than those in stage IV. Mean aortic diameter was significantly greater in individuals in stages III and IV (3.73 ± 0.28) than in the stage II of the disease (2.77 ± 0.21), p < 0.05. No differences in echocardiographic parameters’ values were observed between seropositive, seronegative, nodule-present and nodule-absent persons. Echocardiographic examination revealed valvular heart disease in 24 (80%) RA and 6 (20%) control patients (p < 0.0001).
KeywordsRheumatoid arthritisDiastolic function of heartDoppler echocardiography
The results of some epidemiological studies point to the presence of an increased risk of cardiovascular disease (CVD), particularly atherosclerosis and CHF in certain autoimmune inflammatory disorders such as RA [1–3]. Necropsy examinations of RA hearts revealed a higher prevalence of clinically silent myocardial nodules, restrictive pericarditis and coronary vasculitis in RA compared to non-RA patients. At least 50% of abnormalities remained asymptomatic [4–6].
Established risk factors for the development of heart failure can be divided into four categories: clinical ones such as hypertension, ischemic and valvular heart diseases, diabetes, chronic pulmonary disease, smoking and obesity; biochemical factors, e.g. natiuretic peptides, hyperhomocysteinemia and inflammatory cytokines; pharmacological factors—especially non-steroidal anti-inflammatory drugs (NSAIDs) and other infrequent ones such as restrictive pericarditis and coronary arteritis .
Conventional CHF risk factors such as subclinical atherosclerosis, coronary artery disease, myocardial infarction , valvular heart disease  and chronic pulmonary pathology  are more prevalent in RA than in non-RA patients. The others, such as obesity and smoking are equally or even less frequent in the RA group .
The most important CHF risk factor appears to be hypertension. In RA subjects normal [8, 12, 13] as well as decreased  and elevated [1, 14] mean systolic and/or diastolic blood pressures were reported. Hypertension risk factors, such as the chronic NSAIDs and corticosteroids use however, are overrepresented in this disease .
An increased risk of CVD in individuals with RA may result from traditional risk factors, which are more prevalent in this group of patients, but apparently the disease itself can increase the risk independently. Pathological conditions contributing to myocardial dysfunction such as high serum levels of IL-6, CRP and TNF alpha are present both in RA and CHF patients [16, 17].
The most common pathological mechanism leading to the development of heart failure is LV diastolic dysfunction, which remains clinically asymptomatic for a long time. Multiple investigations confirm that in 40–60% of patients with clinically evident heart failure LV systolic function remains normal (left ventricular ejection fraction/LVEF/more than 45–50%) [18, 19]. Many of patients’ complains in these cases can be ascribed to disturbed LV relaxation, which usually develops before its systolic dysfunction.
Isolated diastolic heart failure is a prevalent form of heart failure in the elderly, in women and in patients suffering from arterial hypertension [20, 21]. In most cases it remains clinically silent, beginning with relaxation disturbances accompanying normal systolic function . Diastolic disorders display as relaxation time retardation, increase in myocardial stiffness and decreased ventricular influx .
Echocardiography with Doppler mitral blood flow assessment is the most valuable examination of LV filling. According to its results four classes of physical effort tolerance and heart failure symptoms can be distinguished: (1) relaxation disturbances or physiological or age-related changes, (2) pseudonormalisation, (3) restricted LV influx, (4) atrial failure, irreversible damage [21, 23].
The aim of this study was to assess the systolic and diastolic functions of the LV in RA patients without clinically evident cardiovascular disease, using pulsed Doppler echocardiography. Our purpose was also to estimate whether there is a correlation between the duration and severity of RA and the degree of LV diastolic dysfunction.
Materials and methods
Thirty consecutive patients 33–63 years old (average age 51.8 ± 7.6 years) diagnosed with definite or classic RA (according to ARA criteria) , attending the Rheumatologic Outpatient Department of the Central Clinical Hospital in Warsaw, were included in the study. The group comprised 25 women (average age 52.7 ± 7.0 years) and 5 men (average age 47.2 ± 9.7 years). The disease duration ranged from 2 to 35 years (mean 12.5 ± 9.3 years, median 12). Four (13%) patients were in stage II, 20 (67%) in stage III and 6 (20%) in stage IV of the disease according to Steinbrocker’s criteria  The radiograms shown geodes and erosions of periarticular bone in all cases. Subcutaneous nodules were present in nine individuals. Patients’ body-surface area (BSA, the Mosteller formula) amounted to 1.74 ± 0.21 m2.
Waaler-Rose titre below 20 was confirmed in 14 individuals (seronegative patients). In the remaining 16 (seropositive ones), Waaler-Rose titre: 1:160 was noted in 12.1:320 in 2 and 1:640 also in 2 persons.
Virtually all the RA patients were taking NSAIDs; 14 were treated with methotrexate, 8 with leflunomide, 5 with sulphasalazine and 3 with antimalarials. Twenty-two patients received low-dose prednisolone.
Each RA patient was matched with a sex, age (±1 years) and BSA (±0.17m2) healthy volunteer control. The control group of 30 individuals 33–62 (average 51.7 ± 7.6 years old) comprised 25 women (average age 52.6 ± 6.8 years) and 5 men (average age 47.0 ± 10.1 years).
Patients with a history of myocardial infarction, arterial hypertension, rheumatic fever, type I or II diabetes or general amyloidosis were excluded from the study. All patients and volunteers agreed to participate in research and the deontologic commission’s agreement was obtained. All patients were classified I degree heart failure according to NYHA.
On the basis of clinical examinations and control chest radiographs (postero-anterior and lateral), reported by a cardiologist and a radiologist, pulmonary diseases were excluded.
All patients underwent a complete transthoracic echocardiography examination: two-dimensional, color and Doppler (continuous and pulsed wave) and the tissue Doppler, as well. Images were obtained on a Philips Hewlett-Packard Sonos 5,500 with a 2 MHz transducer. Recordings were performed in the left lateral position. All examinations were recorded and stored on videotape. All the measurements were performed according to the recommendation of the American Society of Echocardiography.
The following structures and factors were assessed during the echocardiographic examination: left atrial end-systolic diameter (LA), left ventricular end-diastolic diameter (LVEDD), right ventricular end-diastolic diameter (RVEDD), interventricular septum end-diastolic thickness (IVSEDT), left ventricular posterior wall end-diastolic thickness (LVPWEDT), aortic root diameter (AO), ejection fraction (EF) and fractional shortening (FS).
In a M-mode technology, the end-systolic antero-posterior diameter in the parasternal projection of the left side of a patient is accepted as the size of the left atrium (LA).
The end-diastolic antero-posterior diameter in the parasternal projection of the left side of a patient is accepted as the size of the left ventricle (LV), right ventricle (RV), thickness of an interventricular septum (IVS), left ventricle posterior wall (LVPW) and aortic root diameter (AO).
The ejection fraction was calculated as follows:
LV diastolic volume − LV systolic volume/LV diastolic volume × 100 .
The shortening fraction (left ventricular end-diastolic − end-systolic distance/end-diastolic × 100) was calculated .
Pericardial effusion was measured as the largest distance between the parietal and visceral pericardium at peak systole behind the left ventricular posterior wall.
In order to assess mitral valve insufficiency a semi-quantitative method that is based on the assessment of a revolving wave extension, which penetrates inside an atrium or a ventricle  was used. A four-grade scale was used. When a revolving wave, whose width was greater than 2 mm, registered during the whole period of systole reached up to one-quarter of the atrium, then mitral insufficiency of degree I was noted. When the revolving wave, whose width was approximately 5 mm, registered in left atrium reached up to one-third of the atrium, then a mitral insufficiency of degree II was noted. When a revolving wave, wider than 5 mm, reached up to one-half of the atrium, a mitral insufficiency of degree III was noted. Finally, a mitral insufficiency of IV degree was noted when a revolving wave took up more than 50% of the atrium.
The insufficiency of aortic valve cusps was also assessed using a four-degree scale: degree I, revolving wave under the valve; degree II, revolving wave registered during the LV reflux at the level of an anterior mitral cusp; degree III, revolving wave reaching the level of the papillary muscles; degree IV, revolving wave taken in a horizontal cross-section of at least 50% of a ventricle surface at the level of the papillary muscles.
A trace of mitral, aortal and trigeminal insufficiency was recognised when a thin revolving wave appeared just under a valve during a short period of time, i.e. for less than 100 ms. As these types of changes are often seen in healthy people, patients with such a complication were excluded from the statistical analysis.
The parameters of diastolic function measured included: (1) peak early velocity (E wave; m/s) and (2) peak velocity at the time of atrial contraction (A wave; m/s), (3) E to A ratio ( E/A), (4) E wave deceleration time (DT; ms), (5) isovolumetric relaxation time (IVRT; ms), (6) Tissue Doppler imaging of the mitral annulus velocities (TDI): early diastolic (E′) and late diastolic (A′) velocity and E′/A′ ratios were measured.
LV diastolic function was assessed by measuring the mitral flow velocity recorded in the apical four-chamber view by pulsed-wave Doppler sampling during diastole. Peak early (E) and late (A) mitral inflow velocities were measured. E/A ratio and deceleration time of E wave were obtained. IVRT of left ventricle was obtained as the time interval from the cessation of LV outflow to the onset of mitral valve inflow. TDI was applied in the pulsed Doppler mode of the mitral flow annulus velocity with the same echocardiographic unit.
Finally, a comparative analysis of recognised changes in the heart between both groups was undertaken.
After a 4 h fasting blood was collected for erythrocyte sedimentation rate (ESR) [according to Westergren method (mm/h)], morphology, C-reactive protein (CRP) (mg/l), cholesterol (mmol/l), HDL cholesterol (mmol/l), LDL cholesterol (mmol/l), triglycerides (mmol/l), glucose (mmol/l) and a Waaler-Rose test. All examinations were performed at the Central Diagnostic Laboratory of the Central Clinical Hospital.
A value for an average quantitative variable was assessed on the basis of a Student’s t-test (differences between seronegative and seropositive patients), paired t-test (differences between RA patients and Control group patients) and ANOVA followed by Scheffe’s test (differences between three groups: RA stages according to Steinbrocker’s stages) .
Correlations were sought using Pearson’s correlation coefficient .
Discrete variabIes were described in terms of frequencies and percentages. Fisher exact test or chi-square test judged differences in proportions.
Characteristic of RA group
Duration of disease
12.5 ± 9.3 (2–35) median 12
RA stage according to Steinbrocker’s
Number of painful joints
6.2 ± 2.3 (3–14) median 6
Number of swollen joints
3.2 ± 1.8 (0–8) median 3
Morning stiffness (h)
1.6 ± 0.8 (0.5–3) median 2
29.8 ± 20 (5–80)
24.6 ± 26.7 (0.71–0.90) median 13
133 ± 42 (89–204) median 124
36.2 ± 14.7 (10–60) median 40
3.8 ± 1.4 (0.7–7.7) median 3.9
Echocardiographic measurements in 30 RA patients and 30 controls
Control group patients
4.62 ± 0.47
4.73 ± 0.28
100.0 ± 24.3
104.7 ± 14.4
183.6 ± 62.4
153.8 ± 38.3
Lvmass index (g/m2)
105.2 ± 32.6
87.9 ± 16.8
1.01 ± 0.33
0.86 ± 0.12
0.94 ± 0.08
0.83 ± 0.11
2.34 ± 0.23
2.28 ± 0.19
3.56 ± 0.38
3.54 ± 0.35
3.18 ± 0.31
3.10 ± 0.63
64.4 ± 1.3
66.3 ± 1.3
37.8 ± 0.8
36.9 ± 0.8
63.0 ± 9.9
67.9 ± 8.4
0.83 ± 0.19
0.74 ± 0.11
0.76 ± 0.14
0.70 ± 0.17
1.16 ± 0.33
1.12 ± 0.29
240 ± 23
219 ± 48
111 ± 22
89 ± 18
0.12 ± 0.03
0.09 ± 0.03
0.12 ± 0.08
0.11 ± 0.02
1.18 ± 0.56
0.85 ± 0.40
LV mass index in RA group was significantly greater than in the control group (105.2 ± 32.6 vs. 87.9 ± 16.8; p < 0.05) so were the interventricular septum end-diastolic thickness (1.01 ± 0.33 vs. 0.86 ± 0.12; p < 0.05), the left ventricular posterior wall end-diastolic thickness (0.94 ± 0.08 vs. 0.83 ± 0.11; p < 0.0001) and the aortic root diameter (3.18 ± 0.31 vs. 3.10 ± 0.63, p < 0.001).
The ejection fraction in RA group was significantly lower than in the control group (64.4 ± 1.3 vs. 66.3 vs. 1.3; p < 0.0001).
The assessment of diastolic function parameters revealed significantly longer isovolumic relaxation time (IVRT) and shorter deceleration time (DT) in RA patients compared to the control group.
The values of echocardiographic parameters of RA patients in different disease stages (according to Steinbrocker), seropositive, seronegative, with rheumatoid nodule and without were compared. Patients in stage II or III revealed significantly lower LV mass index (99 ± 17 vs. 131 ± 42; p < 0.05) and the interventricular septum end-diastolic thickness (0.94 ± 0.10 vs. 1.28 ± 0.5; p < 0.05) than those in stage IV. Mean aortic diameter was significantly greater in individuals in stages III and IV (3.73 ± 0.28) than in the stage II of the disease (2.77 ± 0.21), p < 0.05. No differences in echocardiographic parameter values were observed between seropositive, seronegative, nodule present and nodule absent persons.
Echo-Doppler echocardiographic findings in 30 RA patients and 30 controls group
Control group patients
Normal mitral valve
Mitral insufficiency grade I
Mitral insufficiency grade I
Normal tricuspid valve
Tricuspid insufficiency grade I
Normal aortic valves
Aortic insufficiency grade I
Aortic insufficiency grade II
The correlation coefficient between echocardiographic measurements’ values and values of ESR, DAS28, duration of disease and age RA patients
Duration of disease
0.40 < 0.05
−0.42 < 0.05
−0.39 < 0.05
−0.41 < 0.05
Laboratory parameters in RA patients and the control group
Control group patients
198 ± 48
215 ± 27
122 ± 34
129 ± 25
67 ± 16
66 ± 19
109 ± 51
115 ± 40
82 ± 11
79 ± 8
Cardiac involvement in rheumatoid arthritis has been reported previously. Relaxation abnormalities are well known in patients with this disease, also in asymptomatic. Munstonen et al.  showed isovolumic relaxation time longer and peak filling rate lower in 12 RA patients than in control subjects. Corrao et al.  reported diastolic abnormalities affecting echo-Doppler parameters of LV filling in RA individuals.
We also noted similar results. Mean IVRT in RA patients was 25% longer than in the control group (111 ± 22 and 89 ± 18, respectively). No correlation between IVRT and the presence of rheumatic nodules, Waaler-Rose titre or the stage of the disease progression was observed, however. In 100 RA patients it was found that Diastolic LV diameter and aortic root diameter were larger and the ejection fraction, as well as mean velocity of circumferential fibre shortening and the fractional shortening-smaller than in the control OA group .
The differences we noted between the mean aortic diameter and the EF values in RA versus non-RA patients were in accordance with Montecucco observations . Significantly higher LV mass and LV mass index in RA individuals reflected significantly greater IVSD and PWD. Significantly greater LV mass index and IvsD were noted in patients in stage IV of RA compared to those in stage II and III, whereas the aortic diameter was significantly smaller in stage II than in stage III or IV patients.
Significant reduction in various indices of LV diastolic function was present, whereas no differences in LV end-diastolic diameter, systolic function and parietal thickness were found in RA patients compared to controls. The values of the early to late filling waves of mitral inflow Doppler (E/A) correlated with patients’ age and were independent of disease duration. The peak-lengthening rate of the LV diameter correlated closer with the disease duration than with the patients’ age .
In a study of 32 patients, suffering from RA and the equal number of matched controls, reported by Alpaslan et al.  systolic function was found normal in all subjects. Echocardiographic indices of LV diastolic function (peak E velocity, E velocity/A velocity ratio, IVRT, MPI and TFPV) in the RA group were significantly different form those of the controls. Marked difference in RV echocardiographic indices between the two groups was not observed, however . Statistically significant differences between mean E, E′ E′/A′ values and IVRT and DT were also observed in our RA patients in comparison with the control group.
Di Franco et al. concluded that RA individuals with no evidence of heart disease presented diastolic dysfunction characterised by decreased E/A and S/D ratios which findings were correlated with the disease duration . In our RA patients, negative correlation between E/A values and the disease duration and patients’ age was observed.
Levendoglu et al. observed left and right ventricular diastolic dysfunction in RA patients. They reported a direct relationship between values of some of the LV diastolic parameters and the disease duration .
Another investigation of 47 patients with RA and sex- and age-matched controls showed significantly higher frequency of pulmonary artery systolic pressure(35 mmHg in patients with RA (21%), than in controls (4%). Diastolic dysfunction caused by impaired relaxation was more common in patients with RA (66%), than in controls and correlated with extraarticular manifestations of the disease .
The reported findings indicate the presence of subclinical myocardial involvement in RA. This is most probably due to the nonspecific myocarditis observed in RA patients, previously described as asymptomatic, rarely influencing cardiac size or function, predominantly affecting LV diastole, usually diffuse pattern scarcely clinically significant.
The question arises, of course, as to what extent cardiac pathology in RA observed by us and other investigators, emerge from inflammatory myocarditis itself or is secondary to other pathology or drug use in this disease.
Our patients were free from symptoms and signs of coronary artery disease, pulmonary diseases as well as arterial hypertension, which were excluded prior to investigation. Significant valvular disturbances were not diagnosed in any case, nor was the pericardial pathology. None of the patients presented with diabetes or uncontrolled hypercholesterolaemia. The patients were non-smokers.
We did not perform histomorphological examinations yet in absence of other immediate causes of CHF and considering that our patients did not undergo any treatment with drugs of known and immediate cardiodepressive effects we assume that the CHF features present in them were RA complications. We assume that in absence of hypertension the main reason for diastolic heart failure in RA patients is probably myocardial fibrosis secondary to inflammation of systemic disease.
The lack of history of cardiotoxic antirheumatic drug use in patients under surveys confirms the hypothesis that the abnormalities can be due to RA itself.
Doppler echocardiography as a noninvasive, simple technique is widely used to detect mild asymptomatic heart abnormalities. The results of our study indicated that it could be useful in RA patients in detection of early stages of diastolic dysfunction. Determining whether the detection of such abnormalities in patients with no clinical symptoms will influence therapeutic decisions requires further investigation.