DELAYED ONSET POSTOPERATIVE PARAPLEGIA IN ACUTE TYPE A AORTIC DISSECTION.

In patients with operated type A aortic dissections, irreversible spinal cord injury (SCI) may result from several factors: prolonged circulatory arrest, extension of replacement, and hypoperfusion of segmental arteries secondary to aortic false lumen thrombosis. Careful neuroprotective strategies and shorter operative times are crucial to reduce SCI incidence. Despite optimal perioperative management, delayed-onset SCI occurs in rare cases in response to subacute aortic remodeling. This report describes the case of a 77-year-old woman who underwent ascending aorta and hemiarch replacement for type A aortic dissection and had delayed paraplegia that developed on postoperative day 12. In patients with operated type A aortic dissections, irreversible spinal cord injury (SCI) may result from several factors: prolonged circulatory arrest, extension of replacement, and hypoperfusion of segmental arteries secondary to aortic false lumen thrombosis. Careful neuroprotective strategies and shorter operative times are crucial to reduce SCI incidence. Despite optimal perioperative management, delayed-onset SCI occurs in rare cases in response to subacute aortic remodeling. This report describes the case of a 77-year-old woman who underwent ascending aorta and hemiarch replacement for type A aortic dissection and had delayed paraplegia that developed on postoperative day 12. In the frozen elephant trunk era, spinal cord injury (SCI) is a well-known complication of surgery for acute type A aortic dissection (TAAD). Occasionally, it can also complicate the postoperative course of patients who undergo less invasive aortic arch surgery, and this has been reported in only a few retrospective studies and case reports. Paraparesis or paraplegia generally occurs in the early postoperative hours, when clinical surveillance is higher. Here we report a case of acute delayed-onset postoperative paraplegia after hemiarch replacement for TAAD that was caused by false lumen (FL) thrombosis. A 77-year-old woman, a former smoker with a history of hypertension, was admitted to the emergency department of the Sant’Orsola Hospital, Bologna, Italy, with epigastric pain and diaphoresis. She was hemodynamically stable without signs of malperfusion. The computed tomographic (CT) scan showed a TAAD with severe dilatation of the ascending aorta (maximum diameter 58 mm) that extended from the aortic valve to the iliac arteries. The intimal tear was located in arch zone 0. The epiaortic vessels were not dissected, and the right intercostal arteries arose from the true lumen, whereas the left intercostal arteries arose from the FL (Figure 1), as well as the left renal artery. A few small reentry tears were identified along the thoracoabdominal aorta (Figure 1). The transthoracic echocardiogram showed a preserved ejection fraction and severe aortic regurgitation. The patient underwent an emergency operation through a full sternotomy. Cardiopulmonary bypass was established through the right femoral artery and right atrial cannulation, and antegrade cerebral perfusion with moderate hypothermic circulatory arrest at 25°C was used. Cerebral perfusion was monitored by near-infrared spectroscopy. On inspection of the aortic valve, no regurgitation was detected during the water test. Ascending aorta and hemiarch replacement was therefore performed using a 26-mm Gelweave graft (Vascutek, Inchinnan, United Kingdom). The patient was discharged from the intensive care unit after 2 days. After 12 days of an unremarkable postoperative course, with the patient in a stable condition, lower limb hypoesthesia and hyposthenia suddenly occurred, associated with lower back pain. Immediate action was taken, through intravascular volume optimization, aiming for a mean arterial pressure greater than 90 mm Hg, and cerebrospinal fluid (CSF) drainage, using the LiquoGuard system (Möller Medical GmbH, Fulda, Germany), with a target pressure of 10 mm Hg and a maximum drainage velocity of 15 mL/h. The CT scan showed FL thrombosis in the thoracic aorta, with reperfusion of the abdominal aorta. The intercostal arteries were clearly identifiable on the right side, but they were hypoperfused on the left side (Figure 2). Magnetic resonance imaging showed a T2-hyperintense central intramedullary lesion, confirming SCI at the 11th and 12th thoracic vertebral level (Figure 3).Figure 3Postoperative spinal cord magnetic resonance.View Large Image Figure ViewerDownload Hi-res image Download (PPT) After 4 days, incomplete paraplegia persisted despite low intracranial pressures, and the CSF drain was removed. The patient was eventually transferred to an SCI rehabilitation unit. SCI is a well-known complication of extensive open or endovascular repair of the thoracic aorta. Fortunately, the incidence of this complication declined, thanks to improved surgical strategies and perioperative management. However, current postoperative SCI rates seem to be persistent.1Etz C.D. Weigang E. Hartert M. et al.Contemporary spinal cord protection during thoracic and thoracoabdominal aortic surgery and endovascular aortic repair: a position paper of the Vascular Domain of the European Association for Cardio-thoracic Surgery.Eur J Cardiothoracic Surg. 2014; 47: 943-957Crossref Scopus (152) Google Scholar The theory of the Adamkiewicz artery as the pillar of spinal perfusion has been accepted for decades, but it appears to be too simplistic. The spinal cord blood supply is provided by a complex anastomotic network of anterior and posterior spinal arteries. Cranially, spinal arteries arise from both vertebral arteries, and caudally, they arise from the internal iliac, sacral, and mesenteric arteries. Additional support is provided by the intercostal and lumbar arteries.2Etz C.D. Kari F.A. Mueller C.S. et al.The collateral network concept: a reassessment of the anatomy of spinal cord perfusion.J Thorac Cardiovasc Surg. 2011; 141: 1020-1028Abstract Full Text Full Text PDF PubMed Scopus (202) Google Scholar Within this framework, TAAD involving all the main aortic branches, along with the unpredictable anatomy and evolution of the FL, may cause reduced blood flow to the intercostal arteries. If enough arteries are affected, especially the lower thoracic segmental arteries, the ischemia may lead to irreversible SCI. Multiple thoracoabdominal reentry tears may mitigate this effect,3Kawanishi Y. Okada K. Nakagiri K. et al.Three cases of newly developed paraplegia after repairing type A acute aortic dissection.Ann Thorac Surg. 2007; 84: 1738-1740Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar, 4Girdauskas E. Kuntze T. Walther T. Mohr F.W. Delayed paraplegia associated with vertebral necrosis after type A dissection surgery.Eur J Cardiothoracic Surg. 2008; 33: 121-123Crossref PubMed Scopus (5) Google Scholar, 5Medalion B. Bder O. Cohen A.J. Hauptman E. Schachner A. Delayed postoperative paraplegia complicating repair of type A dissection.Ann Thorac Surg. 2001; 72: 1389-1391Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar but unfortunately, in our case, this did not happen. The unexpected event was the FL thrombosis, extending from the distal anastomosis to the abdominal aorta, which in our experience is an extremely rare event, especially after hemiarch replacement. It is known that aortic arch replacement, regardless of the extension, requires a period of circulatory arrest, which increases the risk of SCI. However, in our case the arrest time was short, and the patient fully recovered the day after the procedure. The debate is currently ongoing about the maximum number of intercostal arteries that can be sacrificed or covered by the stent graft with the lowest risk of paraplegia, but it is generally agreed that the lower thoracic segments (from T8 to T12) represent a critical zone.6Kozlov B.N. Panfilov D.S. Ponomarenko I.V. et al.The risk of spinal cord injury during the frozen elephant trunk procedure in acute aortic dissection.Interact Cardiovasc Thorac Surg. 2018; 26: 972-976Crossref PubMed Scopus (14) Google Scholar,7Zoli S. Roder F. Etz C.D. et al.Predicting the risk of paraplegia after thoracic and thoracoabdominal aneurysm repair.Ann Thorac Surg. 2010; 90: 1237-1244Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar In our patient, a large number of left intercostal arteries were hypoperfused, including critical segments. It is compulsory to evaluate imaging in aortic dissection carefully while considering the whole anatomy (true and FL morphology, entry and reentry tear sites) and the possible impact of a surgical procedure on remodeling. The benefits of an extensive aortic repair have previously been demonstrated, as have its consequences. Conversely, less complex lifesaving procedures, such as ascending aorta and hemiarch replacement, carry a lower risk of SCI. Prophylactic CSF drainage, currently used in our center during frozen elephant trunk procedures, is not always feasible in an emergency setting, especially in patients with unstable hemodynamics or ongoing antiplatelet or anticoagulation therapy. Investigators have reported that CSF drainage can be positioned on the first postoperative day or at the onset of symptoms.1Etz C.D. Weigang E. Hartert M. et al.Contemporary spinal cord protection during thoracic and thoracoabdominal aortic surgery and endovascular aortic repair: a position paper of the Vascular Domain of the European Association for Cardio-thoracic Surgery.Eur J Cardiothoracic Surg. 2014; 47: 943-957Crossref Scopus (152) Google Scholar In our case, SCI was mainly caused by postoperative aortic remodeling and FL thrombosis, but if we look retrospectively at the preoperative CT scan, it seems that there was already significant thrombosis of the FL, considering that most of the left intercostal arteries originated from the FL itself (Figure 2). Moreover, we considered blood pressure aggressive management to be unnecessary because the aortic repair was not so extensive and also given that we usually maintain a normal blood pressure in cases of TAAD. CSF drainage was immediately positioned at the onset of symptoms, but only with partial benefit. How to predict and prevent this complication remains an issue, especially if surgical planning does not involve an extensive aortic arch operation. Undoubtedly, during preoperative imaging evaluation, we assessed the risk of SCI, but we were not able to predict complete FL thrombosis. Moreover, we must consider that the patient was not treated with warfarin, and we can only speculate, retrospectively, that this may have played a role in the FL thrombosis. An alternative treatment could have been emergency surgical fenestration of the FL,8Miele L. Eschelman D.J. McNulty S. et al.Successful aortic fenestration to treat prolonged motor paralysis of the lower extremities after repair of type A aortic dissection.J Thorac Cardiovasc Surg. 2005; 130: 599-601Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar but we had no experience with this type of complication, despite the large volume of hemiarch replacements for TAAD performed in our center (295 cases in the last 20 years). Delayed paraplegia in partial arch replacement is almost unpredictable. For this reason, all efforts should be directed to reduce the risk: careful surgical planning according to aortic anatomy and patient characteristics, postoperative hemodynamic optimization, and aggressive management of symptoms should be adopted.