Author + information
- Received July 24, 2014
- Revision received December 15, 2014
- Accepted December 17, 2014
- Published online May 1, 2015.
- Ntobeko A.B. Ntusi, FCP (SA), DPhil∗,†,
- Stefan K. Piechnik, PhD, MScEE∗,
- Jane M. Francis, DCR (R), DNM∗,
- Vanessa M. Ferreira, MD, DPhil∗,
- Paul M. Matthews, MD, DPhil OBE‡,§,
- Matthew D. Robson, PhD∗,
- Paul B. Wordsworth, MA‖,
- Stefan Neubauer, MD∗ and
- Theodoros D. Karamitsos, MD, PhD∗,¶∗ ()
- ∗University of Oxford Centre for Clinical Magnetic Resonance Research, Division of Cardiovascular Medicine, Radcliffe Department of Medicine, John Radcliffe Hospital, Oxford, United Kingdom
- †Division of Cardiology, Department of Medicine, Groote Schuur Hospital, Cape Town, South Africa
- ‡GlaxoSmithKline Clinical Imaging Centre, London, United Kingdom
- §Division of Brain Sciences, Department of Medicine, Imperial College, London, United Kingdom
- ‖Botnar Institute, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Nuffield Orthopaedic Centre and John Radcliffe Hospital, Oxford, United Kingdom
- ¶1st Department of Cardiology, AHEPA Hospital, Aristotle University, Thessaloniki, Greece
- ↵∗Reprint requests and correspondence:
Dr. Theodoros D. Karamitsos, University of Oxford Centre for Clinical Magnetic Resonance Research, Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom.
Objectives The goal of this study was to assess the diffuse myocardial fibrosis and edema in rheumatoid arthritis (RA) using multiparametric cardiac magnetic resonance (CMR) and the association of myocardial T1 and extracellular volume (ECV) with disease activity, duration, and cardiac function.
Background RA is a connective tissue disorder, with frequent cardiovascular disease. Myocardial inflammation and diffuse fibrosis can be detected noninvasively by using native T1 mapping and ECV quantification on CMR.
Methods Thirty-nine RA patients (28 women; mean age 50 ± 12 years) and 39 matched control subjects (28 women; mean age 49 ± 12 years) underwent CMR at 1.5-T, including cine, tagging, T2-weighted, native T1 mapping (shortened modified Look-Locker inversion recovery), late gadolinium enhancement (LGE), and ECV imaging.
Results Focal fibrosis on LGE was found in 46% of RA patients compared with none of the control subjects. Patients with RA had larger areas of focal myocardial edema (10% vs. 0%), higher native T1 values (973 ± 27 ms vs. 961 ± 18 ms; p = 0.03), larger areas of involvement as detected by native T1 >990 ms (35% vs. 2%; p < 0.001), and expansion of ECV (30.3 ± 3.4% vs. 27.9 ± 2.0%; p < 0.001) compared with control subjects. Left ventricular volumes, mass, and ejection fraction were similar between RA patients and control subjects. Peak systolic circumferential strain (–16.9 ± 1.3 vs. –18.7 ± 1.2; p < 0.001) and peak diastolic circumferential strain rate (83 ± 21 s–1 vs. 112 ± 20 s–1; p < 0.001) were impaired in RA patients. Myocardial T1 and ECV were correlated with myocardial strain and RA disease activity.
Conclusions Subclinical cardiovascular disease is frequent in RA, including focal and diffuse myocardial fibrosis and inflammation, which are associated with impaired strain and RA disease activity. CMR T1 mapping provides potential added value as a biomarker for disease monitoring and study of therapies aimed at reducing diffuse myocardial fibrosis in RA.
- diffuse myocardial fibrosis
- extracellular volume estimation
- inflammatory arthropathies
- T1 time
Rheumatoid arthritis (RA) is a chronic autoimmune disease of the joints, associated with pain, swelling, limited mobility, disability, and premature death (1). RA is commonly associated with extra-articular features, including cardiovascular complications (2). Cardiovascular involvement in RA predominantly affects young female subjects and accounts for 40% to 80% of the premature mortality and a consequent 10-year reduction in life span (3–5). Myocardial disease in RA results from a combination of processes that includes chronic inflammation, myocardial fibrosis, focal nonspecific diffuse necrotizing or granulomatous myocarditis, microvascular dysfunction, unrecognized myocardial infarction (MI), and the drugs used in the treatment of RA (6).
Cardiac magnetic resonance (CMR) has been used in the comprehensive noninvasive assessment of myocardial abnormalities in RA (7,8). Late gadolinium enhancement (LGE), representing focal myocardial fibrosis/scarring, reportedly occurs in 39% of RA patients, likely from earlier myocarditis (7). However, LGE relies on the comparison of focal myocardial damage against unaffected normal myocardium and is thus limited in detecting diffuse myocardial fibrosis, which often occurs in RA (6). Results of endomyocardial biopsy may confirm the histologic detection of myocardial fibrosis (9). Extracellular volume (ECV) mapping based on measurement of CMR T1 relaxation times has proved to be a robust tool for the noninvasive assessment of diffuse myocardial fibrosis (10,11). T1 mapping and ECV estimation correlate well with histological evidence of myocardial fibrosis in various clinical contexts (12,13). We therefore hypothesized that CMR T1 mapping would reveal subtle forms of myocardial inflammation and diffuse myocardial fibrosis in asymptomatic RA patients with no known cardiovascular involvement compared with control subjects of similar age and sex. Furthermore, we sought to investigate the relation between CMR parameters of diffuse fibrosis and myocardial inflammation and indexes of RA disease activity and duration. Finally, we examined the relation between diffuse myocardial fibrosis and left ventricular (LV) regional systolic and diastolic function.
This was a prospective study enrolling nonselected patients with RA (N = 39) with no known cardiovascular disease. RA patients were recruited from 4 hospitals in the Thames Valley, United Kingdom. They underwent clinical assessment and CMR scanning between November 2010 and December 2012. Patients with RA included in the study were between 18 and 65 years of age and had a confirmed diagnosis of RA based on the 1987 American College of Rheumatology criteria (modified in 2010) (14), as assessed by clinical consultant rheumatologists. Exclusion criteria included inability to tolerate CMR, contraindications to CMR, nonsinus rhythm, known heart disease (previous MI, previous myocarditis on history, heart failure, arrhythmia on 12-lead electrocardiography, or other chronic cardiac condition), renal impairment (estimated glomerular filtration rate <30 ml/min), impaired liver function (alanine aminotransferase level more than twice the upper limit of normal), and known hypersensitivity to gadolinium (Figure 1). Female subjects who were pregnant, lactating, or planning a pregnancy were also excluded. Healthy volunteers were matched for age, sex, and ethnicity (N = 39); they had no cardiac history, were not on cardiovascular medications, and had a normal electrocardiography and underwent CMR as control subjects.
All subjects provided written informed consent to participate in the study. Ethical approval was granted for all study procedures by the Oxford Research Ethics Committee.
CMR image acquisition
CMR studies were performed by using a single 1.5-T magnetic resonance system (Avanto, Siemens Healthcare, Forchheim, Germany) according to standard and previously published methods. A complete stack of short-axis (SA) images were obtained during breath-hold and cardiac gating for cine, pre-contrast (native) T1 mapping, T2-weighted, and LGE imaging. T1 mapping was performed by using the shortened modified Look-Locker inversion recovery sequence, and T2-weighted CMR was performed with the black-blood short-tau inversion recovery (STIR) sequence. Three SA (basal, mid-ventricular, and apical) scans and a single long-axis (horizontal) scan were obtained for tagging. Native T1 maps and T2-weighted and cine-tagged images were acquired before administration of gadolinium contrast. LGE imaging was performed by using a T1-weighted phase-sensitive inversion recovery sequence 8 to 12 min after intravenous administration of a contrast agent (gadoterate meglumine [Dotarem, Guerbet LLC, Paris, France]; 0.15 mmol/kg of body weight). A single mid-ventricular SA slice was acquired for post-contrast T1 maps at 1, 2, 3, 4, 8, 15, and 20 min after the administration of contrast (gadoterate meglumine). Typical imaging parameters for the sequences used were as previously published (15) and are detailed in the Online Appendix.
CMR image analysis
All CMR images and maps were analyzed offline and in a blinded fashion.
For cine images, analysis of LV ejection fraction was performed by using Argus software version VB17 (Siemens Healthcare). LV SA epicardial and endocardial borders were manually contoured at end-diastole and end-systole. LV end-systolic and end-diastolic volumes were used to calculate stroke volume and LV ejection fraction as stroke volume/end-diastolic volume. LV myocardial mass was calculated by subtracting the endocardial volume from the epicardial volume, based on earlier knowledge of myocardial specific gravity (1.05 g/cm3). Left atrial (LA) diameter was measured in the LV outflow tract (3-chamber) view.
For tagged cine images, post-processing and semiautomated analysis was performed by using Cardiac Image Modeler software (CIMTag2D, University of Auckland, Auckland, New Zealand) by aligning a grid to the myocardial tagging planes in end-diastole. End-systole was determined visually, and tags were adjusted at each frame through the cardiac cycle. From the mid-SA slice, peak circumferential systolic strain and peak diastolic strain rate were derived.
For STIR images, semiquantitative analysis was performed by comparing the LV myocardium in the SA against adjacent skeletal muscle in the same slice, verified on a corresponding steady-state free precession image. The T2 signal intensity (SI) ratio was calculated as T2 SImyocardium:skeletal = SImyocardium/SIskeletal muscle. Myocardial edema was diagnosed when the myocardial T2 SI ratio was >1.9. Care was taken to exclude nonsuppressed blood pool signal due to slow-flow adjacent to the subendocardium and to avoid using areas with abnormally low signal for normalization.
LGE images were evaluated qualitatively for the presence or absence, pattern (e.g., subendocardial, mid-wall, subepicardial, transmural), and regional distribution of LGE areas by 3 observers, each with at least 4 years of CMR experience. The detection of LGE was made by consensus of all 3 observers. In addition, focal areas of LGE were defined semiquantitatively as those with SI ≥2.0 SDs above the mean SI of normal myocardium according to MC-ROI software (programmed by S.K.P. in Interactive Data Language version 6.1, Exelis Visual Information Solutions, Boulder, Colorado).
For T1 maps, they were manually contoured by using in-house software MC-ROI to outline the endocardium and epicardium. They were then divided into 6 segments per slice by using the anterior right ventricular–LV insertion point as reference for comparing segments among sequences.
Consistent with earlier reports on ECV estimation (10,12), we measured pre-contrast and post-contrast myocardial and blood T1 values, and the estimation of ECV and lambda was based on multipoint regression, incorporating all available pre-contrast and post-contrast points, to increase the robustness of the estimates by increasing the number of underlying data points. ECV was calculated as: (1 – hematocrit). For calculation of post-contrast T1 values, the post-contrast T1 map acquired at 20 min was utilized for ECV calculation.
For areas of myocardial involvement by STIR and native T1 mapping, the following was used: on dark-blood T2-weighted images, edema was diagnosed when myocardial T2 SI was >1.9 times that of remote skeletal muscle. On T1 maps, acute myocardial injury was diagnosed when T1 was >990 ms. For all quantitative analyses of T2-weighted and T1-mapping images, only regions of myocardium with a contiguous area of ≥40 mm2 above the specified thresholds were considered relevant (to reduce the detection of noise as positive findings). To calculate the extent of a subject’s myocardial involvement detected by using tissue characterization techniques, the percentage of abnormal myocardium (as defined earlier) was determined for each segment and then averaged for that subject (16).
RA disease activity and duration
Disease activity was assessed by using a disease activity score that incorporates a 28 tender joint and swollen joint count in addition to a measure of general health, together with the serum C-reactive protein level (DAS28-CRP). An absolute level of disease activity can be selected as a clinically meaningful goal for therapeutic intervention; a value of ≤3.2 was defined as the threshold for a low disease activity state and <2.6 as the threshold for remission. CMR biomarkers were also correlated with disease duration. In keeping with clinical practice, a cutoff of 2 years was used to distinguish early RA disease duration from established duration.
Normality of data was tested by using the Kolmogorov-Smirnov test. Normally distributed data are presented as mean ± SD and nonparametric data as median (interquartile range); categorical data are presented as number (percentage). The chi-square test or the Fisher exact test was used to compare dichotomous data. The unpaired Student t test (when normally distributed) or Mann-Whitney U test (for nonparametric data) was used to compare continuous variables between RA patients and control subjects, as appropriate. Segmental data were averaged on a per-subject basis before group comparisons versus control for clustering of segments within each subject. Bivariate correlations were assessed by using the Pearson’s (R) or the Spearman’s (RS) coefficient, as appropriate. All statistical tests were 2-tailed, and a p value <0.05 was considered statistically significant. All analyses were performed using SPSS version 20 (IBM SPSS Statistics, IBM Corporation, Armonk, New York).
The patients with RA were well matched with the control subjects for age, sex, and comorbidities (Table 1). A majority of the RA patients were on methotrexate and chloroquine. Most patients had been diagnosed with RA for a median of 7 years (interquartile range: 4 to 11 years). The median duration of anti-inflammatory treatment was 4 years (interquartile range: 3 to 6). The mean DAS28-CRP score was 3.3 ± 1.3, which signifies ongoing disease activity in most patients.
LV and right ventricular structure and function
There was no significant difference in LV size, mass, or ejection fraction between RA patients and control subjects (Table 2). However, peak systolic circumferential strain (–16.9 ± 1.3 vs. –18.7 ± 1.2; p < 0.001) and diastolic strain rate (83 ± 21 s–1 vs. 112 ± 20 s–1; p < 0.001) at mid-SA were impaired in RA patients compared with control subjects. In keeping with diastolic dysfunction, the LA diameter was larger in RA patients compared with control subjects (32 ± 5 mm vs. 27 ± 5 mm; p < 0.001).
Focal myocardial fibrosis (LGE imaging)
As previously reported by others (7), we found an increased incidence of LGE in RA patients compared with control subjects (46% vs. 0%), as shown in Table 2. The majority of LGE was nonischemic in nature: 28% of RA patients had patchy mid-wall enhancement in the basal inferior and lateral walls (Figure 2); 13% had patchy or linear enhancement involving the septum; and 5% had evidence of previously unknown MI. In these 2 patients with previous MI, subsequent coronary angiography confirmed epicardial coronary artery disease. Quantitative analysis of LGE revealed that, overall, RA patients had a small volume of scarring (3.7 ± 0.4% of total LV mass).
Myocardial edema (T2-weighted CMR)
On conventional dark-blood T2-weighted imaging, there was no significant difference in the overall global myocardial T2 SI ratio between RA patients and control subjects (1.6 ± 0.2 vs. 1.5 ± 0.1; p = 0.7). However, patients with RA had significantly more areas of focal myocardial edema within the left ventricle (median 10% vs. 0% in control subjects; p < 0.001).
Diffuse myocardial fibrosis (T1 mapping)
Native T1 mapping and post-contrast T1 mapping were performed in all patients with RA. The 2 RA patients with MI were excluded from this analysis. RA patients had significantly higher native shortened modified Look-Locker inversion recovery T1 values (973 ± 27 ms vs. 961 ± 18 ms; p = 0.03) and expanded ECV (30.3 ± 3.4 vs. 27.9 ± 2.0; p < 0.001) (Figure 3). Patients with RA had significantly larger areas of myocardial involvement detected by using native T1 mapping (median 35% vs. 2% in control subjects; p < 0.001). When RA patients were stratified according to the presence of LGE, there was no significant difference in native T1 and ECV values (Table 3). However, patients with LGE had greater areas of myocardial involvement on volume fraction of T1 >990 ms (median 33% vs. 13%; p < 0.001). When grouped according to duration of RA diagnosis, there was no difference in native T1 values (966 ± 27 vs. 963 ± 36; p = 0.79) and ECV (29.8 ± 3.8 vs. 30.4 ± 4.0; p = 0.59) between RA patients with disease duration <2 years and those with disease duration >2 years, respectively. Finally, there were also no differences in native T1 and ECV when RA patients were separated according to rheumatoid factor positivity or the presence of anti–cyclic citrullinated peptide antibodies.
Functional and clinical correlates of diffuse myocardial fibrosis
Both native T1 (R = 0.54, p < 0.001) and ECV (R = 0.61, p < 0.001) were moderately correlated with disease activity. Native T1 and ECV also correlated with peak systolic strain (R = 0.46, p = 0.003; and R = 0.52, p = 0.001) and peak diastolic strain rate (R = –0.44, p < 0.001; and R = –0.49, p < 0.001), respectively (Figure 4). In addition, the DAS28-CRP score was associated with peak systolic strain (R = 0.41, p = 0.01) and peak diastolic strain rate (R = –0.37, p = 0.02) (Figure 5). Finally, areas of myocardial injury detected by using T1 mapping correlated with peak systolic strain (RS = 0.62, p < 0.001), peak diastolic strain rate (RS = –0.68, p < 0.001), and DAS28-CRP score (RS = 0.44, p = 0.006), respectively.
The principal finding of the current study is that CMR detects diffuse and focal myocardial fibrosis, as well as inflammation, in patients with RA. Second, in these RA patients with no known cardiovascular disease, indexes of diffuse myocardial fibrosis and inflammation were elevated regardless of the presence of LGE and correlate with the DAS28-CRP score, a measure of disease activity in RA as well as with systolic and diastolic strain. Third, diffuse myocardial fibrosis is unrelated to RA disease duration and is not associated with the presence of rheumatoid factor or anti–cyclic citrullinated peptide antibodies. Our results indicate that assessment of novel tissue characteristics with T1-mapping CMR can provide novel imaging biomarkers of disease activity, beyond those provided by standard techniques such as STIR T2-weighted and LGE imaging.
Myocardial fibrosis is due to deposition of extracellular matrix proteins, including collagen (17), and has been shown to be associated with adverse cardiovascular outcomes (18). Myocardial fibrosis in RA likely represents the final common expression of microvascular disease, macrovascular coronary artery disease, and chronic myocardial inflammation (myocarditis and cardiomyopathy) (6). The pathophysiology of LGE is premised on the combination of increased volume of distribution for gadolinium-based contrast agent and prolonged washout related to the decreased capillary density within the scarred myocardium (19). The increase in gadolinium concentration within fibrotic tissue results in T1 shortening, and the fibrotic myocardium appears as bright SI on inversion recovery images compared with the dark, nonfibrotic myocardium. We found that the prevalence of focal LGE in RA was 46%, with most patients having a nonischemic pattern of patchy mid-wall enhancement. Similarly, a previous study reported LGE in 7 (39%) of 18 RA patients with no symptomatic cardiac disease (7). We also documented 2 RA patients with previously undiagnosed MIs. Although the risk of coronary artery disease is higher in patients with RA (2,3), RA patients are less likely to report symptoms of angina and are more likely to experience silent MIs and sudden cardiac death (5).
LGE CMR is an accurate method for measuring focal myocardial fibrosis. However, its sensitivity is limited for the assessment of diffuse myocardial fibrosis, in which the difference in SI between normal and fibrotic myocardium may not be a discriminator. T1 mapping, before and after contrast, directly measures the underlying T1 relaxation times with high spatial resolution. T1 mapping also allows direct parametric signal quantification on a standardized scale for each myocardial voxel, and it provides better characterization of myocardial tissue composition on a global and regional level than conventional LGE (19). ECV is derived from the ratio of T1 change in the blood and myocardium, taking into account the hematocrit level. In the absence of MI, infiltration, or edema, increases in ECV quantitatively depict the extent of diffuse myocardial fibrosis (20). We found increased pre-contrast T1 values and expanded ECV in RA patients, indicating the presence of subclinical myocardial disease. Our group has previously published research on the use of volume fraction of T1 as a sensitive biomarker of myocardial involvement in patients with myocarditis and systemic sclerosis (11,15). Elevation in the volume fraction of native T1 above a threshold of 990 ms may indicate myocardial fibrosis or myocardial inflammation not detected by using standard techniques such as T2-weighted STIR or LGE.
Native T1 and ECV were correlated with abnormal systolic and diastolic strain parameters in these patients with RA. Myocardial fibrosis has been linked to abnormal indexes of myocardial strain (21). The pathophysiology and sequence of events occurring before the development of structural heart disease in RA remain unclear. Others have postulated that myocardial fibrosis detected by using CMR likely precedes the development of myocardial relaxation abnormalities (22). Our study adds to the growing body of evidence regarding the association between diffuse myocardial fibrosis and impaired myocardial strain (in the context of adverse structural remodeling in inflammatory heart disease).
We also found that native T1 and ECV correlated with disease activity in RA. Our data support the hypothesis that systemic inflammatory response is critical to development of cardiovascular complications. Early epidemiological observational and clinical studies are commensurate with this proposition (23). Elevated DAS28-CRP scores in RA correlate with elevations in pro-inflammatory cytokines, including interleukin-1-alpha, interleukin-1-beta, interleukin-6, tumor necrosis factor-alpha, leptin, and vascular endothelial growth factor (24); these factors play a crucial role in mediating pathophysiologically important effector functions within the cardiovascular system, including up-regulation of vascular adhesion molecules, oxidization of low-density lipoprotein cholesterol, increased oxidative stress, endothelial and microvascular dysfunction, and promotion of atherothrombosis.
In the current study, we concentrated on pre-contrast and post-contrast tissue characteristics in RA. Although native myocardial T1 values and ECV may be increased due to other causes (e.g., cardiac amyloidosis) (15,25), there was no evidence of myocardial infiltration, and none of the RA patients had an LGE pattern suggestive of MI (the 2 patients with MI were excluded from the analysis). Hence, we are confident that the increased native T1 and ECV reflect diffuse myocardial fibrosis and myocardial inflammation. Moreover, we have not validated T1 and ECV measurements against histological assessments; endomyocardial biopsy was not possible in these asymptomatic patients. Although T1 mapping can detect both myocardial edema and fibrosis, the technique is not sufficiently evolved to be able to distinguish the underlying mechanism behind increased extracellular water. Myocardial biopsy is the only certain way to distinguish these 2 pathological features of the myocardium. However, previous publications have shown good correlation between CMR indexes of diffuse myocardial fibrosis and histological fibrosis in other clinical contexts (12,26). Another limitation of this study was the lack of serum biomarkers of myocardial injury, including troponins, CRP, and B-natriuretic peptides. Finally, we report LA diameter, which is less accurate compared with LA volume as a measure of LA size. However, measuring LA volume accurately with CMR would require a complete stack of SA cines through the atria; this was not part of our scanning protocol, which was already long in duration.
We have shown that RA is associated with subclinical myocardial fibrosis. It remains unclear how much of these observations reflect RA disease activity and progression versus the effect of drugs such as methotrexate, which reportedly causes hepatic and pulmonary fibrosis (27,28). Indeed, methotrexate was associated with mild liver fibrosis in 1% to 52% of patients with RA (27). Some studies have shown that although most methotrexate-induced hepatic enzyme dysfunction is reversible, many patients who develop hepatic fibrosis also have other risk factors for fibrosis, including obesity, excessive alcohol intake, diabetes mellitus, advanced age, and drug-related factors such as high total methotrexate dose and long duration of therapy (29). In humans, methotrexate has not been shown to cause myocardial fibrosis. In an animal model of autoimmune myocarditis, methotrexate was shown to be anti-inflammatory, antifibrotic, and to be associated with significant improvement in ejection fraction (30). These data would support the observed changes being due to the disease rather than to the use of methotrexate.
Our results demonstrate the potential of CMR to depict diffuse and focal myocardial fibrosis and inflammation in RA patients with apparently normal hearts, in whom indexes of diffuse myocardial fibrosis and inflammation are elevated regardless of the presence of LGE and correlate with disease activity and impaired systolic and diastolic strain indexes. Novel CMR T1-mapping tissue characterization provides added value as promising imaging biomarkers for disease monitoring and the study of therapies aimed at reducing diffuse myocardial fibrosis in RA. Finally, in the future, CMR including T1 mapping may potentially be useful in RA patients at different stages of the disease to assess the extent of myocardial involvement and to delineate myocardial disease progression.
CLINICAL COMPETENCY: Subclinical cardiovascular disease is frequent in RA, including focal and diffuse myocardial fibrosis and inflammation, which are associated with impaired strain and RA disease activity. Novel CMR T1-mapping tissue characterization provides added value as a promising imaging biomarker for disease monitoring and the study of therapies aimed at reducing diffuse myocardial fibrosis in RA.
TRANSLATIONAL OUTLOOK: Future studies are warranted to elucidate the relationship between myocardial fibrosis and inflammation with myocardial strain. Studies are also needed to clarify to what extent the earlier observations are due to RA disease activity and progression versus the effect of RA therapies.
The authors gratefully acknowledge the role of the rheumatologists and rheumatology specialist nurses in patient recruitment from the following hospitals: John Radcliffe Hospital, Oxford; Nuffield Orthopaedic Centre, Oxford; Wexham Park Hospital, Slough; and Royal Berkshire Hospital, Reading.
The authors acknowledge support from the National Institute for Health Research Oxford Biomedical Research Centre Programme. Drs. Piechnik and Robson have competing nonfinancial interests including the following: 1) US patent 61/387,591: Drs. Piechnik and Robson: systems and methods for shortened Look-Locker inversion recovery cardiac gated mapping of T1; September 29, 2010; all rights sold exclusively to Siemens Medical. 2) US patent pending 61/689,067: Drs. Piechnik and Robson: Color map design method for immediate assessment of the deviation from established normal population statistics and its application to cardiovascular T1-mapping images. Dr. Matthews was employed by GlaxoSmithKline during the conduct of the study; has received research funds from Biogen Idec and GlaxoSmithKline; consultancy fees or speakers honoraria from Novartis, Biogen Idec, and GlaxoSmithKline; and holds stock in GlaxoSmithKline. Dr. Neubauer has received support from the Oxford British Heart Foundation Centre for Research Excellence. Dr. Karamitsos has received investigator-led grants from GlaxoSmithKline and Guerbet. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- cardiac magnetic resonance
- C-reactive protein
- disease activity score with 28 tender and swollen joint count, incorporating C-reactive protein
- extracellular volume
- long atrial
- late gadolinium enhancement
- left ventricular
- myocardial infarction
- signal intensity
- short-tau inversion recovery
- rheumatoid arthritis
- Received July 24, 2014.
- Revision received December 15, 2014.
- Accepted December 17, 2014.
- American College of Cardiology Foundation
- Solomon D.H.,
- Goodson N.J.,
- Katz J.N.,
- et al.
- Cooper L.T.,
- Baughman K.L.,
- Feldman A.M.,
- et al.
- White S.K.,
- Sado D.M.,
- Fontana M.,
- et al.
- Flett A.S.,
- Hayward M.P.,
- Ashworth M.T.,
- et al.
- Iles L.,
- Pfluger H.,
- Phrommintikul A.,
- et al.
- Ferreira V.M.,
- Piechnick S.K.,
- Dall’Armellina E.,
- et al.
- Weber K.T.,
- Janicki J.S.,
- Shroff S.G.,
- et al.
- Ling L.H.,
- Kistler P.M.,
- Ellims A.H.,
- et al.
- Mewton N.,
- Liu C.Y.,
- Croisille P.,
- et al.
- Jellis C.,
- Martin J.,
- Narula J.,
- Marwick T.H.
- Moreo A.,
- Ambrosio G.,
- De Chiara B.,
- et al.
- Libby P.,
- Ridker P.M.,
- Maseri A.
- Karamitsos T.D.,
- Piechnik S.K.,
- Banypersad S.M.,
- et al.
- Wong T.C.,
- Piehler K.,
- Meier C.G.,
- et al.
- Barrera P.,
- Laan R.F.,
- Van Riel P.L.,
- et al.