Despite its infrequent incidence, aorto-coronary dissection involving the sinus of Valsalva is potentially catastrophic, leading to devastating consequences [3,4,5,6, 11, 13, 20, 21, 27]. The first case of aortic dissection as a complication of PCI was described by Moles et al. . As in our case, the majority (i.e. > 80%) of such cases have arisen secondary to RCA dissection or perforation [1,2,3,4,5,6,7, 9,10,11,12,13,14,15,16,17,18,19,20,21,22, 27], usually the proximal or ostial RCA . This could be due to differences in the histologic structure of the ostia of RCA and left main coronary artery (LMCA) [1, 8, 12, 18]. The periostial and sino tubular junctions of LMCA are formed by more concentrically arranged circular and spiral smooth muscle cells in its intima, with abundant elastic fibres than the RCA, which explains the LMCA’s greater resistance to retrograde dissection) [1, 8, 12, 18].
In the pre-stent era, dissection of sinus of Valsalva was reported secondary to prolonged balloon inflation, associated with rupture of the balloon [3, 22]. With the advent of PCI and stenting, the mechanisms for iatrogenic perforation are being attributed to: aggressive manipulation of rigid wires into subendothelial spaces, when attempting to re-canalise tight lesions [3, 11, 12, 14, 16, 19, 22]; forceful manipulation of guiding catheters that are either wedged  or in a non-coaxial position relative to the proximal segment of the coronary artery [2,3,4, 6, 8, 10,11,12,13,14,15, 17, 18, 20, 22, 27, 28]; and prolonged balloon inflations [3, 12]. A vigorous manual injections of contrast medium into the subintimal space plays a major role in extending the dissection further, by generating reverse flow proximal to the catheter tip and causing retrograde extension [3, 4, 8, 11, 14, 15, 17,18,19, 21, 28].
Perhaps the greatest risk for PCI-related dissections is the presence of heavily calcified vessels or chronic total occlusions (CTO) [13, 22], which require more aggressive catheter manipulation for better support, and the usage of stiff wires in vessels whose walls are less resistant to trauma [1, 2, 8, 11, 14,15,16]. Hydrophilic coated guide wires are associated with increased risk of coronary perforation [7, 21, 26, 29], due to low coefficient of friction and ease of distal migration . Amplatz catheters are more frequently associated with ostial coronary artery dissections [2, 7, 15, 21]. Some reports have also demonstrated the involvement of JR catheters in dissection [1, 2, 7].
A recent MI, as in our patient, poses additional risk, as the inflammatory process of infarcted vessels render them tender, with tendency to dissection [1, 2]. Iatrogenic dissections are more common in the setting of acute MI (0.19%), than for non-acute MI interventions (0.01%) . Cystic medial necrosis [3, 11, 17], smaller RCA size , arterial hypertension and age > 60 years are additional risk factors .
In our case, mechanical trauma caused by guide catheter manipulation during its withdrawal seems the most likely culprit for causing the port for further dissection: shearing forces of blood flow during systole or diastole may have assisted the propagation of the ostial perforation to involve the sinus of Valsalva [1.4.21]. Even vigorous inspiration by the patient during contrast injection could pose a risk  as in our patient. Moreover, he was restless throughout the procedure, repeatedly attempting to observe it, in spite of strong counselling against doing so.
Once the dissection was noted, vigorous contrast injections were avoided and further stenting was done with gentle manipulations of guide catheter and wires. Rapid recognition of the patient’s haemodynamic conditions  and urgent assessment by TTE or trans-oesophageal echocardiography (TOE) in the cath lab  is paramount to deciding treatment modality, which also depends on the extent of aortic involvement and underlying coronary anatomy [2, 5, 16].
In our case, there was an Ellis type II perforation in the proximal RCA which extended into the ostium and further into the sinus of Valsalva. Ellis et al. defined the angiographic classification of coronary perforations, describing type II perforations as pericardial or myocardial blush without a ≥ 1 mm exit hole . Type I and type II perforations are usually confirmed by retrospective review of the angiogram, and can be easily missed if not looked for .
Localised dissections are usually contained below the sino-tubular junctions, by the well-developed supra-valvular ridge, and believed to resolve spontaneously in the first month [1, 3, 8]. However, despite some successful outcomes [8, 13, 20, 32], localised dissections treated conservatively have sometimes rapidly progressed to extensive aortic dissection [3, 18, 22], with some even subsequently requiring surgical intervention [1, 3, 4].
Thus, immediate percutaneous stenting of the perforated ostium and sealing the presumed site of entry door for aorto-coronary dissection, is the preferred modality in haemodynamically stable and localised dissections (with suitable anatomy for stenting) [1, 5, 6, 12, 19, 22]. Bae et al. reported the first case of aorto-coronary dissection successfully treated with PTCA and stenting, without an operation .
Previously, bare metal stents were preferred, owing to their delivery ability in the ostium [11, 13, 16]. Some operators have deployed drug-eluting stents (DES) [11, 21]. However, the presence of cells in a standard stent may not fully cover the site of entry of the dissection . Thus, in free perforations causing a large entry port at the coronary ostium, stent grafts or covered stents are more effective [8, 15], as they prevent contact between vessel wall and components of blood, acting as a mechanical barrier [6, 33, 34]. Successful coronary stenting seals off the entry port within minutes of the complication developing, thus preventing further propagation of aortic dissection [1, 3, 6, 8, 21].
Abu-Ful et al. reported the first case in which acute aortic dissection complicating PCI was treated with a covered stent to seal the entry site , a strategy that was later adopted by others [7, 10]. The Graftmaster covered stent comprises of an expandable PTFE graft material sandwiched between two coaxially aligned stainless steel stents [6, 33]. They are ideally suited to fully cover and seal entry sites formed by the coronary ostium itself, where standard stents maybe insufficient [6, 33]. However, as they are bulky with low flexibility and high profile, delivery may be difficult, and high-pressure deployment and/or IVUS-guided optimisation has been recommended [13, 24, 33].
PTFE stents are associated with increased incidence of subacute stent thrombosis higher than in standard stents, estimated at 5.7 to 8.6% [23, 33, 34] and can occur later than with conventional stents. Increased incidence of restenosis and target lesion intervention is also reported [7, 33, 34]. These have been attributed to increased thrombogenicity and delayed endothelialisation of these stents [23, 26], thus warranting an extended duration of DAPT of at least 6 months to 1 year [24, 26, 34, 35]. Prasugrel or ticagrelor are suitable antiplatelet agents of choice, due mainly to their lack of intrinsic resistance [23, 35].
Once the dissection is stabilized, careful monitoring with optimum BP control and non-invasive imaging techniques, preferably in a coronary care unit setting is required, to detect progression of the dissection and ensuing complications. This can be accomplished by serial TTE/TOE, CT or MRI scan [3, 11, 12, 16, 18]. TOE/TTE appears to be superior to CT in the follow up of patients . TOE has also been recommended instead of angiography . If extension of the dissection is demonstrated, surgical repair should be considered .
This case demonstrates the greater efficacy of covered stents over bare metal stents, in immediate sealing of coronary ostium, thus limiting aorto-coronary dissections localised to the sinus of Valsalva. In addition to minimal adverse effects, covered stents also circumvents the additional risks otherwise posed by the alternative option, which is surgical intervention and are associated with better prognosis . However, appropriate treatment modalities should be reviewed on a case-by-case basis, particularly taking into account the patient’s haemodynamic conditions and coronary lesion morphology.