The objective of this study was to examine a novel hybrid SPECT/CT molecular imaging approach for detecting myocardial sympathetic dysfunction in patients with suspected or diagnosed ARVC/D by analyzing the left and right ventricular uptake separately. To this end, we performed early and delayed planar and delayed SPECT/CT 123I-MIBG imaging in a group of 27 patients, 17 of whom proved to have ARVC/D. Using the SPECT/CT fusion images, we were able to obtain separate assessment of the left and right ventricular sympathetic innervation. Whereas planar scintigraphy provided only limited information, the delayed RV/M ratio in SPECT/CT was significantly lower in patients with diagnosed ARVC/D and could distinguish with high sensitivity and specificity between ARVC/D patients and patients with other underlying cardiac heart diseases in our study cohort. To our knowledge, this is the first SPECT/CT study to demonstrate the feasibility of separate left and right ventricular in vivo evaluation of the sympathetic myocardial innervation.
Planar H/M and Washout Ratio
A number of different ROI-based methods to quantify the myocardial sympathetic innervation in planar images have been published,13,14,15,16,17 but all different methods showed similar results and correctly predicted a worse prognosis for patients with a low H/M ratio. The patients in our ARVC/D group had a mean H/M ratio of 1.5, slightly below the published cut-off value of 1.6 in the ADMIRE-HF study.15 Thus, our slightly lower H/M ratio is in line with earlier findings, since ARVC/D is apparently resulting in partial myocardial sympathectomy, resulting in lower 123I-MIBG uptake. The mean H/M ratio in our reference group was 1.7, indicating generally less severe heart disease than in the ARVC/D group. The small numerical difference did not differ significantly between groups, likely because the planar H/M ratio is weighted towards tracer uptake in the left ventricle, which is relatively spared in earlier stages of the disease.
In a previous 123I-MIBG study, Ogita et al.18 showed that a healthy cohort had a WOR of 9.6% ± 8.5%, whereas those patients with a WOR above 27% had significantly higher mortality than did patients with WOR < 27%. Both of our cohorts showed a WOR much greater than 27%, probably due to the nature of their underlying arrhythmogenic diseases. Nevertheless, the WOR was distinctly higher in our ARVC/D group, pointing towards a more pronounced impairment of the myocardial innervation, as compared to that in the reference group.
SPECT/CT LV/M and RV/M Ratio
Planar scintigraphy is the established method in the clinical setting to quantify the sympathetic innervation with 123I-MIBG. As a 3D modality, SPECT/CT can bring additional valuable information about the distribution of tracer uptake in the heart. In our approach, we used the additional morphological information of the low-dose CT scan to define separate VOIs of the left and right ventricle.
Veen et al. demonstrated a high correlation between planar and left ventricular H/M ratios,12 which is in good accordance with our findings. We could demonstrate a systematic underestimation of the cardiac 123I-MIBG uptake by planar imaging as compared to SPECT, most likely arising from the somewhat higher mediastinal tracer uptake in planar imaging, also reported by Chen et al.19 In an early SPECT study of 48 patients with ARVC/D Wichter et al. showed abnormally reduced 123I-MIBG uptake in the LV myocardium, most notably in the posteroseptal basal segments.9 We, too, found a significantly lower LV/M ratio in our ARVC/D group, as compared to the reference cohort (3.2 vs 3.9; P = 0.014).
Previous studies demonstrated that a present sympathetic dysfunction, as visualized by 123I-MIBG imaging, predicts poor prognosis in heart failure patients.6,15 Paul et al. confirmed the generalizability of this relationship to ARVC/D without present heart failure.10 These patients had a significantly higher risk to develop life-threatening ventricular tachyarrhythmia to follow-up. Thus, there is an emerging role of 123I-MIBG SPECT/CT for individual risk stratification and the possibility to detect the disease in the early phase. However, we contend that previous methods have been inadequate since they were restricted to consideration of LV, and provided no information regarding the RV sympathetic innervation.10 We now used the superior spatial and anatomic assignment of SPECT/CT to assess separately the RV, which showed a significantly reduced RV/M ratio in the ARVC/D group as compared to the reference group (1.6 ± 0.3 vs. 2.0 ± 0.2; P = 0.001), and indeed proved to be the best method to distinguish between the two patient groups. Nonetheless, to establish the absolute magnitude of reduced RV innervation a comparison to individuals without heart disease will be mandatory.
It is well known that ARVC/D is characterized by a fibrofatty infiltration of the heart muscle, which primarily affects the right ventricle1,20 and in the further course of the disease can lead to the involvement of the left ventricle.2 During disease progression 123I-MIBG uptake is hampered due to the damaging of sympathetic fibers.10 This might also lead to a change in the patient’s RV/M, LV/M, and H/M ratios over the course of time, which could be a potential tool to non-invasively monitor disease progression.
ROC-analysis revealed a threshold for the H/M ratio of (1.66) comparable to that previously published for patients with heart failure (1.60) by Jacobson et al.,15 but with a rather low sensitivity (75%) and specificity (60%). The AUC of only 0.73 also indicates that planar H/M ratio is not adequate to distinguish the ARVC/D and IVF groups. In contrast, the SPECT/CT findings for the LV/M ratio showed clearly superior results for the cut-off threshold of 3.41, which gave higher sensitivity (77%), specificity (80%), and AUC (0.78). The best parameter of all was the RV/M ratio with a cut-off value of 1.88, which gave 94% sensitivity, 80% specificity, and an AUC of 0.93.
In this context, it is of great interest to compare the sensitivities and specificities of our SPECT/CT results with those of the modified task force major criteria; parameters determined by echocardiography required 95% specificity, with sensitivity of the individual parameters ranging from 55% to 75%. Based on our findings yielding a high specificity and sensitivity for the RV/M ratio the inclusion of the 123I-MIBG imaging in the revised McKenna criteria could be a future consideration. However, in order to establish this parameter for the diagnosis of ARVC/D, it would be essential to validate our findings in a larger cohort.
In addition, the progressive fibro-fatty replacement of the RV with a subsequent enlargement of the ventricle leads to abnormalities in the regional wall motion. These changes can be assessed by cardiac magnetic resonance imaging.21 Complementary nuclear medicine techniques like gated SPECT equilibrium radionuclide angiocardiography can be used to calculate ventricular functions, volumes, synchrony, and entropy or regional wall motion abnormalities. This techniques could be helpful in identifying patients with suspected ARVC/D22,23,24 and should be further evaluated in the future to enable a faster and more reliable diagnosis in this rare but endangered cohort.
A major limitation of our study is the low number of patients and the heterogeneous clinical stage of the disease, both of which are due to rarity of ARVC/D. Furthermore, the lack of a healthy control group—due to the ethical requirements of radiation exposure to healthy subjects—is limiting the transfer of this results into the clinical routine, since the published thresholds were established to distinguish the cardiac 123I-MIBG uptake between individuals with heart disease.