The purpose of the present investigation was to evaluate a shortened CMR protocol in clinical routine practice. Therefore, a very heterogenous patient collective was investigated. The main findings of this study are i) T1 mapping is superior to T2 mapping in detection of unspecific myocardial pathology in a heterogenous study population of patients without LV-dysfunction and ii) a shortened protocol comprising only T1 mapping and LV-function analysis discriminates patients who will benefit from a full contrast enhanced CMR protocol from those who do not with sufficient confidence.
Need for shorter examinations
In clinical practice CMR, amongst a few other applications, is considered to be the most time consuming and therefore economically challenging examination. Due to lengthy protocols, emergency examinations cannot easily be integrated into a busy schedule even if there are drop outs. Further, the examination is technically challenging and labor intensive. In an optimal setting a CMR examination occupies one trained physician and up to two technicians [
4,
24]. This further stresses an effective and economic use of MRI scanners. Long examination times also hamper patient comfort. As a consequence, important CMR examinations can be delayed in the clinical routine while CMR reports are often urgently needed for clinical decision making [
5]. Therefore, shorter CMR protocols are desirable allowing more patients to benefit from the high diagnostic value that makes CMR the diagnostic gold standard in detection of several cardiovascular diseases. Yet, data on rigorously shortened CMR protocols is very rare.
Evaluation of the study results
The suggested protocol in this study comprises a quantitative approach for myocardial assessment and a combination of quantitative and qualitative evaluation of the left ventricle. In our study population T1 mapping was superior to T2 mapping and was therefore further investigated in the shortened CMR protocol. Prolongation of T1 relaxation time can be found in several entities of myocardial pathology such as edema, fibrosis, protein deposition and others [
7]. In contrast, T2 relaxation times are mainly increased in presence of edema and slightly in case of fibrosis. Additionally, a broad range of myocardial T2 values can be found in healthy volunteers complicating to set a strong threshold value [
25,
26]. It is to assume that those are the two main reasons why T1 mapping was superior to T2 mapping in detection of patients with a pathological finding in the final CMR report in our heterogenous study population. Further, we also found good diagnostic performance of ECV evaluation. However, determination of ECV requires T1 mapping measurement before i.v. gadolinium application and approximately 10 min after. For that reason, this technique does not necessarily lead to a shortened examination time. Yet, native T1 mapping was not inferior to ECV measurement in our population; this is of note since the latter is a contrast enhanced technique with partially inherent information about EGE/LGE [
27]. Additionally, non-enhanced protocols are desirable due to the growing knowledge on deposition of MR contrast agents in the human body [
28,
29] and increasing concerns against Gd application on the patients’ side.
One approach to shorten CMR examination time is to predict at an early state which patient will benefit from the full contrast enhanced examination and which will not. In our collective every third patient did not benefit from a full contrast enhanced CMR. 57% of those subjects were correctly identified by native T1 mapping and cine imaging alone. Regarding the entire study population, the shortened protocol correctly ruled out pathology in every fifth patient.
Yet, three cases of pathologic findings were overlooked using the shortened sequence protocol. Those three cases were consistent with chronic myocarditis. It is of note that this diagnosis does not represent acute myocardial damage with inherent treatment indication. Therefore, clinical relevance is questionable if there is no other pathologic myocardial alteration present like LV dysfunction or others. Additional T2 mapping would not have detected these 3 cases of chronic myocarditis. In summary, no case of acute myocardial damage with inherent obligation to treat has been missed.
Future applications of the shortened protocol – can it be used as a screening method?
Even though the proposed shortened CMR protocol yielded satisfying results, a standardized use in clinical routine is somewhat challenging. Application as a screening protocol with image analysis after the scan would neither be cost nor time effective since approximately 70% of the patients would have to be rescanned with the need for full contrast CMR. Although a partial scan has already been acquired and those sequences could be omitted in the full contrast scan, positioning of the patient in the scanner and geometrical planning of the sequences is a very time-consuming part of the examination. As a result, net scanner occupancy would not be reduced.
Another approach would be to immediately evaluate the image data including determination of myocardial T1 values as well as LV-parameters while the patient is still inside the scanner. If both T1 relaxation time and LV functional parameters are within normal range, further sequences and contrast application can be omitted with fair confidence. According to our mixed cohort, this will assumably be the case in approximately every fifth patient. The saved scanning time might be used to increase the number of CMR examinations offered to referring physicians.
The disadvantage is a reduced predictability of the scan length. Further, this also requires a physician permanently present at the scanner to be effective.
Assuming that a full protocol takes 60 min and the proposed shortened protocol 30 min, the second scenario would allow to stop after the shortened protocol in 31 cases from our study population. This would safe 15.5 h (10%) of scanner time at the cost of 2% incorrect diagnoses only comprising chronic myocarditis of which clinical relevance and evidence-based treatment consequence, respectively, remain unclear.
Additionally, the presented findings might be of importance when shortened protocols are a necessity instead of a choice, for example when the patients’ clinical condition does not allow for a time consuming full CMR protocol or if image quality of contrast enhanced sequences is hampered due to technical issues. Our study results indicate that in these cases the radiologist can rely on cine and mapping imaging for making a diagnosis.
Besides the economic relevance of the findings of this study, the diagnostic potential of the proposed shortened protocol has to be further evaluated prospectively in a larger multicentric study population.
It is of note that in this study sBTFE cine imaging was used for LV evaluation. Real-time imaging could further strongly shorten examination time. Recent studies have shown that measured values showed high accuracy and reliability [
30]. Therefore, a protocol comprising only T1 mapping and real-time cine imaging seems very promising since examination time can probably be reduced to less than 15 min.
Limitations
This is a retrospective single centre study. The study results are limited due to the relatively small size of the study population. Even though, the purpose of the study was to evaluate the shortened protocol for the left ventricle in daily practice conditions and therefore requiring a heterogenous cohort the composition of the study population might be dependent of statistic variations as might the subsequent results. Further, no validation against solid markers was available (e.g. laboratory values such as troponin or tissue from myocardial biopsy). In the study population right ventricular pathology was excluded. The shortened protocol would assess RV-wall movement disorders but the spatial resolution of the proposed T1 mapping sequence is probably too low to allow full and correct diagnosis on the RV limiting the method to LV evaluation.