Jump to
Abstract
Background—Aortic dissection type A is a disease with high mortality. Iatrogenic aortic dissection after interventional procedures is infrequent, and prognostic data are scarce. Our objective was to analyze its incidence, patient profile, and long-term prognosis.
Methods and Results—Between 2000 and 2014, we retrospectively analyzed 74 patients with dissection of the ascending aorta. Clinical and procedural data were reviewed, and later, we performed a prospective clinical follow-up by telephone or in the office. The incidence of aortic dissection was 0.06%. Our patients, predominantly male (67.6%), had a mean age of 66.9±10.8 years. With multiple cardiovascular risk factors, the main reason for cardiac catheterization was an acute coronary syndrome (n=54). The complication was detected acutely in all, trying to engage the right coronary artery in 47 and the left main artery in 30 and after other maneuvers in 2, mostly complex therapeutic procedures (78.4%). A coronary artery was involved in 45 patients (60.8%). Thirty-five patients underwent an angioplasty and stent implantation; 3 had cardiac surgery; and 36 were managed conservatively. Two patients died of cardiogenic shock after the dissection. After a median follow-up of 51.2 months (range, 16.4–104.8 months), none of the remaining patients developed complications as a result of the dissection, progression, ischemia, pain, or dissection recurrence.
Conclusions—Iatrogenic catheter dissection of the aorta is a rare complication that carries an excellent short- and long-term prognosis with the adoption of a conservative approach. When a coronary artery is involved as an entry point, it usually can be safely sealed with a stent with good long-term outcomes.
- aortic aneurysm, familial thoracic 4
- aortic diseases
- cardiac catheterization
- iatrogenic disease
- prognosis
Introduction
Type A aortic dissection is a devastating disease that carries a high mortality and often requires urgent surgical treatment.1 It usually arises spontaneously in susceptible patients, related to genetics, hypertension, drugs, aortic abnormalities, and other factors.
Clinical Perspective on p 2119
However, in the last decades, a number of cases also have been reported after medical acts, iatrogenic forms, frequently after invasive procedures.1–3 In 2002, the International Registry of Aortic Dissection reported 26 patients with this type of dissection (69% after a major surgery and 27% catheter derived) of 723 patients included at that time in the registry.1
Overall, considering the International Registry of Aortic Dissection report, epidemiological features were different from spontaneous forms. Thus, iatrogenic dissections were seen in older patients and those with diabetes mellitus, hypertension, and a higher degree of atherosclerosis or previous history of aortocoronary bypass. Moreover, iatrogenic forms have also been reported to present a poor prognosis with high mortality (International Registry of Aortic Dissection), even worse than the spontaneous forms, despite surgery. However, some recent records such as the German Registry for Acute Aortic Dissection Type A have shown that it could be similar to spontaneous2 forms.
To sum up, specifically in terms of dissections resulting from interventional procedures, the published data are still too scarce.
Considering that more interventional procedures are performed nowadays than ever before, diagnostic or therapeutic, our goal was to analyze the incidence, characteristics, and long-term prognosis of iatrogenic forms of type A aortic dissection caused by a cardiac catheterization procedure.
Methods
Between October 2000 and August 2014, we identified 47 consecutive patients in the database of our center with iatrogenic aortic dissection caused by catheter. Ours is a multipurpose prospective database, involving 3 hospitals. Six of the dissections happened in structural interventional procedures (5 during a transcatheter aortic valve implantation and 1 in a case of patent foramen oval closure in a patient with a cava filter) and were excluded from this analysis because we considered them different pathologies. Next, we added 26 consecutive patients with the same diagnosis from several European centers (Registry on Aortic Iatrogenic Dissection [RAID]). These data were used to calculate the incidence. Later, we added 6 additional nonconsecutive patients (from 1 center from 2002–2010) who were included in the follow-up and the overall analysis. Finally, 9 hospitals participated in the registry. Therapeutic strategy was decided by the attending physician. In all cases, procedure movies were reviewed by experienced interventional cardiologists, and long-term follow-up was performed in office visits or by contacting the patient directly by telephone.
Statistical Analysis
The study was observational and descriptive with a case series design. For statistical analysis, SPSS version 20.0 software (SPSS, Chicago, IL) and the Office 2010 multimedia package (Microsoft, Redmond, WA) for data processing were used. Data are expressed as mean±SD or median and quartiles (quartiles 1–3), as required. Between-group comparisons were performed with the Pearson χ2 for qualitative variables. Long-term survival curves of the different groups were obtained with the Kaplan–Meier method, and comparisons were performed with the log-rank test. The null hypothesis was rejected—no statistically significant differences—with a 2-sided value of P<0.05 used as the cutoff.
Results
Patients, Epidemiology, and Incidence Profile
Overall, RAID analyzed the period from October 1, 2000, to August 2014. In that period of time, in our centers, we performed 108 083 consecutive cardiac catheterizations, 66 354 diagnostic procedures, and 41 729 therapeutic procedures. Among these patients, we found 68 with an iatrogenic aortic dissection. Thus, the incidence during the selected time was 0.062%. The incidence of only the forms involving the coronary tree was 0.039% (0.006% for diagnostic and 0.098% for therapeutic procedures). For the remainder of the analysis, we involved all patients included in RAID (n=74).
Patients included in the study had a mean age of 66.9±10.8 years and were predominantly male (67.6%). Table 1 details the overall epidemiological profile of the patients.
Patient Features of the Overall Cohort (n=74)
Details of the Procedure
The main reason for catheterization was chest pain, with a diagnosis of acute coronary syndrome in 54 patients (73.0%). In 27 patients (36.5%), the study was urgent/emergent. Nineteen patients entered the laboratory with the diagnosis of an acute coronary syndrome with ST-segment elevation. Vascular access was performed by right femoral artery in 49 patients (66.2%), by radial approach in 25 (14 right radial and 11 left radial artery), and by both in 7 patients. All radial accesses were performed since 2010. We found no relevant differences in the clinical profile, diagnosis, and procedure timing in terms of the type of vascular access.
In 16 patients, the procedure had a diagnostic intention only, whereas in 58 patients (78.4%), a coronary intervention was successfully performed at that time. In 10 patients, despite the use of guiding catheters, the procedure had to be stopped without intervention because of the dissection. The amount of contrast media used was 237.53±91.9 cm3, and the procedural median time was 68.5 minutes (quartiles 1–3, 53.5–100.5 minutes). The problem was identified during the procedure in all patients, coinciding with a maneuver engaging a vessel in 72 patients (97.2%), trying to address the right coronary artery in 42 patients (56.8%) or the left coronary in 30 (40.5%), and with other maneuvers in 2. In 5 patients, there was an anomalous origin of the coronary ostium (4 presented an aberrant circumflex and 1 patient had a right coronary artery arising from the left sinus), whereas in the rest of patients, the coronary anatomy was normal. We found that a catheter caused the dissection in 68 patients (91.8%); in 4 patients, the problem was the 0.035-in wire; and in 2 patients, we could not identify the cause (catheter or wire). Usually, the dissection occurred with guiding catheters (70.3%) and 6F catheters (90.5%). The most frequent curve responsible in the femoral access cohort was a Judkins (42.9%; followed by the Amplatz, 25.0%) and for the radial group was the Amplatz (38.9%; Judkins, 27.8%; P=0.55). Table 2 provides further details on the curve distribution.
Procedural Data (n=74)
Initially, 5 patients had left main disease (stenosis of at least 50%), 37 patients presented at least a 50% stenosis of the left anterior descending artery level, 39 displayed a similar lesion of the circumflex artery, and the right coronary was involved in 53 patients. Dominance was right in 90.5% of the patients (n=67).
Most patients, 72, entered the room under antithrombotic treatment, at least aspirin. One had received fibrinolysis with tenecteplase; 5 patients received abciximab; and all were anticoagulated during the procedure.
Events and Follow-Up
After a median follow-up of 51.2 months, only 2 deaths were recorded. Six patients had a follow up <1 month, and 2 of them died during this period. Both patients with a previous diagnosis of ischemic heart disease died of cardiogenic shock during hospitalization.
One patient, a 67-year-old man, underwent a diagnostic catheterization for non–ST-segment–elevation myocardial infarction. He left the catheterization laboratory hemodynamically stable but with an extensive dissection of the ascending and descending aorta (type I) without involvement of major branches, but he experienced moderate back pain. Back in intensive care unit, he suddenly developed intense pain, signs of deep anterior myocardial ischemia, and refractory ventricular fibrillation with quick cardiogenic shock and death, before surgical treatment could be performed. The other patient, a 74-year-old man with severe 3-vessel disease and a left ventricular ejection fraction <30%, also was admitted for non–ST-segment–elevation myocardial infarction. During a therapeutic procedure, the patient experienced a small aortic dissection (Dunning type 1) with involvement of the left coronary during treatment of the left anterior descending artery, all sealed successfully with 3 drug-eluting stents and double-checked by transesophageal echocardiography. The patient was discharged 5 days later. Unfortunately, despite treatment of the coronary artery disease, the patient was readmitted and died 1 month later of cardiogenic shock.
Of the surviving patients, most were managed conservatively (35 patients), 34 patients underwent a percutaneous coronary intervention, and only 3 patients were referred for surgery (2 for aortic surgery and 1 for coronary artery bypass grafting).
During the long-term follow up, after discharge, none of these patients developed complications caused by the iatrogenic dissection such as dissection progression, myocardial ischemia, chest or back pain, or dissection recurrence.
Figure 1 shows the long-term probability for complications and major adverse cardiac events after a catheter-derived iatrogenic dissection by dissection type (with or without coronary artery involvement) or vascular access. There were no differences in the combined major adverse cardiac events variable after stratification for coronary involvement (Figure 1A) or access point (femoral/radial; Figure 1B) after a log-rank analysis.
Kaplan–Meier curve of long-term major adverse cardiac event (MACE)–free survival. MACEs included all cause deaths, cerebrovascular complications, heart failures, reinfarctions, or unstable angina admissions. A significant number happened during the acute phase. One patient was excluded as a result of being lost to follow-up immediately after discharge. A, There were no short- and medium-term prognostic differences in the presence of coronary artery involvement in the dissection, although in the long term the curves begin to separate (not significant). B, Vascular access also displayed no differences.
Interestingly, in 5 patients in whom imaging tests were repeated (3 computed tomography scans and 2 angiographic magnetic resonance images), dissection/hematoma was completely solved ad integrum. Figure 2 shows a representative case.
Multimodality assessment of a type A aortic dissection (Dunning type 3) involving the left main circumflex artery with flow compromise. The patient, a 77-year-old woman, was admitted for a non–ST-segment–elevation myocardial infarction, and during a left anterior descending artery intervention, she developed an extensive dissection. Finally, the dissection was sealed successfully with 3 drug-eluting stents. She was discharged uneventfully 15 days later on dual antiplatelet therapy. The white arrow denotes the dissection plane. A, Angiographic right oblique view after the detection of the complication. B, Computed tomography (CT) reconstruction 5 hours later depicting the same projection as in A. C, Transthoracic echocardiographic long-axis view displaying the dissection (arrow) minutes later. D, Intravascular ultrasound imaging at the left main level showing the dissection plane. This approach has therapeutic usefulness for guiding the angioplasty. E, The CT scan, short-axis view, 5 hours later. F, The same CT scan view 5 days later and (G) 25 days later (after discharge).
Discussion
When the ascending portion of the vessel is involved, aortic dissection is a disease that, despite early surgical treatment, carries with high morbidity and mortality.1–4 Although dissection is usually spontaneous, with the introduction of new surgical and interventional procedures, we are witnessing the development of occasional iatrogenic cases, with a profile very dissimilar to that of the classic patient. Because of its low overall frequency, estimated at <0.1% of procedures globally, we have few data available on its natural history and long-term outcomes.1
The purpose of our study was to provide an update of and a glance at the long-term outcome of patients with dissection of the ascending aorta or arch produced during a cardiac catheterization procedure. With respect to this complication, few data are available, and the data that are available are limited to small case series or isolated single-center case reports. Most of the case series, many with a limited number of patients and published several years ago, are difficult to extrapolate to our practice because different accesses are used now, often requiring different materials and different techniques, frequently with the patient on intense antithrombotic treatment.4–7 More specifically, in 2000, Dunning et al4 published a series of 9 patients with extensive coronary dissection to the aorta (0.02% incidence) and proposed a simple classification of 3 grades. Type 1 was the aortic dissection limited to the sinus of Valsalva; a type 2 dissection reached the ascending aorta but was <4 cm; and the type 3 dissection, the worst one, exceeded 4 cm in length.4 The authors suggested that stenting was usually sufficient to solve the problem when a coronary artery was involved, but in some cases, surgery was warranted (type 3).4,8 On the other hand, such a complex surgery, coronary or aortic (supracoronary conduit or other techniques), in a patient who is, in many cases, under the influence of strong antithrombotic drugs and has recently experienced a myocardial infarction could have catastrophic results.1–3
However, after following up our patients, the largest series described, for a median of 5 years, we perceived that, after the acute phase, the spontaneous evolution of this complication is excellent. Interestingly, this happened even though the majority of patients presented an acute coronary syndrome, and many continued receiving antithrombotic treatment as indicated in this context, something that would be unthinkable in a spontaneous form or even a postsurgical form. In addition, iatrogenic postsurgical forms seem to carry a worse prognosis, as previously shown.1,2
Thus, the profile of the patient who develops this rare complication is a man beyond the sixth decade with frequent cardiovascular risk factors, mostly atherosclerotic and ischemic, who underwent a complex procedure, sometimes with difficult coronary catheterization, generally therapeutic but with a more or less (atherosclerotic) healthy aortic wall. As previously published,1 the clinical presentation was also different. There was less back pain, but when it existed, the pain presented different characteristics and showed a higher frequency of hypotension and shock, and patients were much more likely to experience myocardial ischemia and infarction (36% and 15%, respectively, as reported by the International Registry of Aortic Dissection investigators).1
Nowadays, the use of improved instrumental probably can be offset by the realization of more complex procedures, explaining a relatively similar incidence in older reports.1 However, operator experience and careful, gentle handling of wires and catheters probably would prevent some cases of this dreadful complication.
Although the available information is limited, as often happens in conditions of very low incidence, what we provide here is probably enough to establish some general recommendations for action. On the basis of what was observed in the patients described, for retrograde, non–coronary-related iatrogenic aortic dissections, it seems reasonably safe to maintain a conservative approach under a strict monitoring by imaging techniques to detect potential problems early.
If the patient has no major symptoms, the coronary tree is patent, and the dissection is small and does not seem to evolve or grow in the imaging techniques, the patient will probably evolve well. If a coronary artery is involved, frequently as an entry point, that usually is enough to seal the dissection at that spot with a stent (big dissections, flow compromise).4,6,7,9,10 Otherwise, perhaps it may be advisable to consider early surgical consultation.
In addition, although a type A dissection is still an independent risk factor for mortality in patients with spontaneous dissections,11 our findings are congruent with a recent report by Kim et al,12 pointing out the fact that an acute retrograde type A aortic dissection presents a more favorable prognosis than antegrade dissections (spontaneous).
Thus, we found 2 main types of aortic dissections during cardiac catheterization, an anterograde form, usually with an entry point inside a coronary artery, which can be solved (and closed) with a stent, and a retrograde form, which usually seals alone without further treatment, with the collaboration of the antegrade aortic blood flow. Something that got our attention is that all accesses and all catheter curves can be related to this condition. In addition, the most frequent catheters associated with these iatrogenic forms are 6F, that is, relatively small.
Currently, what seems clear is that, after a period of a few days, with the process stabilized, healing occurs completely ad integrum, and the problem does not recur in further long-term follow-up.10 This fact allowed us in some patients in our series to perform the percutaneous intervention needed days after the procedure in which the dissection occurred, with the consequent intense antithrombotic treatment.
Limitations
The study presents the constraints of an observational and retrospective study, so the therapeutic recommendations should be interpreted with caution. Nonetheless, a disease with so low casuistry per se makes it difficult to use other designs, and the data presented are likely very close to current routine clinical practice.
Conclusions
Iatrogenic catheter dissection of the aorta is a rare complication that carries an excellent short- and long term prognosis with the adoption of a conservative, nonsurgical approach. When a coronary artery is involved as an entry point, it usually can be sealed safely with a stent with good results. However, the profile of patients, usually elderly men with multiple cardiovascular risk factors, ischemic risk, and on intense antithrombotic treatment, makes monitoring them closely during the acute phase advisable.
Disclosures
None.
Footnotes
* A complete list of the RAID Investigators can be found in the online-only Data Supplemental.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.115.015334/-/DC1.
- Received January 6, 2015.
- Accepted April 10, 2015.
- © 2015 American Heart Association, Inc.
References
CLINICAL PERSPECTIVE
Type A aortic dissection is a dreadful disease that carries a high mortality and often requires urgent surgical treatment. It usually arises spontaneously in susceptible patients, related to genetics, hypertension, drugs, aortic abnormalities, and other factors. However, in the last decades, a number of cases have been reported after medical acts (iatrogenic forms), frequently after invasive procedures. We review here a multinational series of 74 patients (Registry on Aortic Iatrogenic Dissection [RAID]) with an acute iatrogenic dissection caused by a catheter. In >100 000 cardiac catheterizations, the incidence was 0.06%. Our patients, predominantly male, had a mean age of 66.9 years. With multiple cardiovascular risk factors, the main cause for cardiac catheterization was an acute coronary syndrome. The complication was detected acutely in all, trying to engage the right coronary artery in 47 and the left main artery in 30 and after other maneuvers in 2, mostly complex therapeutic procedures (78.4%). A coronary artery was involved in 45 cases (60.8%); 35 patients underwent an angioplasty and stent implantation; 3 underwent cardiac surgery; and 36 were managed conservatively. Two patients died of cardiogenic shock. After a median follow-up of 51.2 months, none of the remaining patients developed complications as a result of the dissection, progression, ischemia, pain, or dissection recurrence. Thus, iatrogenic catheter dissection of the aorta is a rare complication that usually carries an excellent short- and long-term prognosis with the adoption of a conservative, nonsurgical approach. When a coronary artery is involved as an entry point, it usually can be sealed safely with a stent with good results.
This Issue
Jump to
Article Tools
Share this Article
- Incidence, Management, and Immediate- and Long-Term Outcomes After Iatrogenic Aortic Dissection During Diagnostic or Interventional Coronary ProceduresCLINICAL PERSPECTIVE
