In this study we demonstrate the utility of ungated RTC CMR imaging to identify significant retrosternal adhesions in patients with previous sternotomies for congenital heart disease. Choosing a midline sagittal plane orientation allowed for a relatively short scan time. Imaging during deep breathing yielded superior results when compared to regular breath-holding. Our results agreed well with intraoperative findings and CMR was able to predict the presence or absence of adhesions in ≥80%. Thus, this imaging technique is a valuable tool in pre-operative planning and can guide surgeons to take precautionary measures such as cannulation for cardiopulmonary bypass via femoral vessels before performing the redo sternotomy.
This is the first published report to use RTC CMR to evaluate retrosternal adhesions during both cardiac and breathing motions. CT and CMR have been used to evaluate retrosternal adhesions via different techniques. Pre-operative CT has been shown to reduce the rate of surgical complications [
10]. Classically, the distance between the sternum and the structure immediately posterior to it was identified as a marker of adhesions [
6]. Furthermore, presence of retrosternal fat on static CT is considered a sign of less significant adhesions, while absence of fat suggests severe adhesions [
7,
11,
12]. Malguira et al identified adhesions as linear fibrous bands on static retrospective electrocardiogram (ECG)-gated CT [
9]. Adhesions were confirmed by detecting deformation and tethering of mediastinal structures on cine CT imaging. Dynamic 4D CT imaging is a relatively new technique that assesses differential mobility to identify tethering between sternum and underlying structures during breathing [
13]. Image acquisition occurs while patient is spontaneously breathing for two respiratory cycles over few seconds. Using this modality, Narayanan et al. demonstrated excellent correlation with intraoperative findings without false negative cases [
13]. Similar to the CMR RTC technique, the 4D CT incorporates cardiac and respiratory motion to determine tethering; however, it involves a significantly higher radiation dose than coronary CT angiogram [
13]. CMR has been used to detect and grade the severity of retrosternal adhesions using tagged cine imaging with aid of a finite element model [
8]. Yoshioka et al were able to calculate the strain at the surface of the RV on the basis of tag displacement over the cardiac cycle with acceptable correlation at the time of surgery [
8]. This method utilized special software that may not be widely available.
RTC CMR has been successfully used in several clinical scenarios including cardiac arrhythmias, impaired breath holding ability or poor ECG gating [
14]. RTC allows ventricular functional assessment comparable to bSSFP [
15‐
18]. The utility of RTC lies beyond assessment of intracardiac anatomy, it has been widely used in thoracic imaging of extra-cardiac chest tumors [
19‐
21]. Lee et al have determined that the combined use of non-breath hold cine and static CMR for assessing mediastinal and chest wall tumor invasion have higher specificity and better diagnostic confidence than that of a static CMR alone [
21].
We did not attempt to grade the severity of adhesions due to limitations of spatial resolution. Another limitation of this study is the small sample size. Further improvement to this sequence that yields higher image quality, a larger sample size and a prospective design may allow for detailed grading of adhesions and more generalizability.