Impact of connective tissue disease on the surgical outcomes of aortic dissection in patients with cystic medial necrosis

We reviewed our institutional database to identify patients who underwent initial surgery for aortic dissection between April 2003 and March 2013 at the National Cerebral and Cardiovascular Center, Osaka, Japan. A total of 321 patients underwent initial surgery for aortic dissection and 298 patients (64.0 ± 15.7 years old, 56% male) underwent postoperative pathological examinations of the surgical specimen during this period, in which CMN was present in 141 patients (47.3%). Of them, 43 patients (37.0 ± 12.8 years old, 53% male) subsequently underwent genetic examinations. Our criteria for genetic examination were as follows: (1) patients with bodily features consistent with the current Ghent criteria, (2) young patients under 50 years of age, and (3) patients with family history of aortic diseases. This study was approved by the National Cerebral and Cardiovascular Center of Japan Institutional Review Board waived the need for individual patient consent under the provisions for deidentified human subject and quality improvement research.

The patient characteristics are shown in Table 1. Thirty patients (69.8%, CTD group) were diagnosed genetically as having CTD, whereas the remaining 13 patients (30.2%, non-CTD group) did not have any genetic disorders. The details of genetic disorders are shown in Table 2. Preoperative variables were compared between the CTD and non-CTD groups (Table 3). The median age at the onset of aortic dissection was significantly lower in the CTD group, namely 33.5 years in the CTD group vs. 42.0 years in the non-CTD (p = 0.030). In addition, there were differences in the Stanford classification of aortic dissection (p = 0.015) and the stages of aortic dissection (p = 0.009) between the two patient groups. In the CTD group, there were type A aortic dissections in 11 patients (37%, 7 acute and 4 chronic) and type B aortic dissections in 19 patients (63%, 1 acute and 18 chronic). In the non-CTD group, there were type A aortic dissections in 10 patients (77%, 9 acute and 1 chronic) and type B in 3 patients (23%, 3 chronic). The incidence of type B aortic dissection was higher in the CTD group. Two patients in the CTD group developed three-channel aortic dissection, although it was not found in the non-CTD group. The incidence of family history of aortic dissection was significantly higher in the CTD group (57% vs. 23%, p = 0.043). The rate of preoperative renal dysfunction (serum creatinine ≥2.0) was significantly higher in the non-CTD group (p = 0.028).

In these series, the surgical treatments were performed according to the same indication criteria at the single center with the same surgical team staff. In the acute phase, emergent or urgent surgeries were carried out, in cases with acute type A aortic dissection with patent false channels excluding intramural hematoma, and with complicated acute type B aortic dissection. In patients with chronic aortic dissection or with residual aortic dissection in the long term after the initial surgical replacement, surgical interventions were indicated when the aortic maximum diameter exceeded 50 to 55 mm. Root repairs such as aortic valve-sparing surgery or composite valve-graft root replacement were carried out with the site of the root over 40 to 45 mm, predominantly for the CTD patients [6]. In 2 patients described above, surgical treatments were indicated for 3-channel dissection, even though the diameter was less than 50 mm.

The samples were surgically resected from the dissecting aortic wall. Histopathological examination consisted of light microscopy on the surgical specimen stained with hematoxylin-eosin stain, Masson’s trichrome stain, elastica van Gieson stain, and toluidine blue stain. CMN was diagnosed if the aorta displayed fragmentation of elastic fibers associated with cystic accumulation of the extracellular matrix. The degree of loss of elastic fibers was defined as elastin fragmentation in the media stained with elastica van Gieson using a diagram of the grading system. Three grades were recognized as follows: grade 0 or 1+ (no loss or minimal loss or mild fragmentation of elastic fibers), grade 2+ (intermediate change), and grade 3+ (complete loss of elastic fibers in some full-thickness portions of the media) (Fig. 1) [5]. Afterwards, loss of elastic fibers was combined into two clinically relevant categories, namely, (1) no or minimal loss of elastic fibers (grade 0 or 1+ representing no loss or minimal loss or mild fragmentation of elastic fibers), and (2) loss of elastic fibers (grade 2+ or 3+ representing a significant loss of elastic fibers in some portion of the media) [5, 7].

Mutation analyses were performed by bidirectional Sanger sequencing of exons and exon-intron boundaries of FBN1, TGFBR1, and TGFBR2 genes first [8‐11]. PCR products were purified and sequenced using BigDye Terminator chemistry v.1.1 on an ABI Prism3130xl or 3730xl (Applied Biosystems). In cases in which mutations of FBN1 gene were not found, mutations were further examined with the multiple ligation probe amplification method on an ABI Prism3130xl (Applied Biosystems). In cases in which these two methods failed to find mutations in patients, SMAD3, ACTA2, and TGFB2 genes were additionally analyzed by bidirectional Sanger sequencing of exons and exon-intron boundaries. For patients without mutations in FBN1, TGFBR1, TGFBR2, SMAD3, ACTA2, or TGFB2 genes, exome sequencing was performed after TruSeq Exome enrichment on HiSeq1000 (Illumina) for searching mutations in MYH11, COL3A1, COLIAI (COL1A1), and COL1A2 genes [12‐16]. Nonsense, missense, or splicing variations in these genes were further analyzed by Sanger sequencing if they were not present in SNP databases, predicted to be damaging by PolyPhen-2, or the SIFT program, or previously reported to be a pathogenic mutation.

Data were collected from hospital admission and outpatient medical records. All patients were regularly followed, either at our center or at other local hospitals. All of them underwent strict blood pressure control before leaving the hospital, with β-blockers in all cases and angiotensin II receptor blockers in some cases. The follow-up rate was 100% and the duration was 57.1 ± 43.0 (range, 6–132) months.

Two-tailed Student’s t tests for continuous variables and chi-square tests for categorical variables were used to make univariate comparisons between the two groups. The long-term outcomes including re-dissection and re-operation were analyzed using Kaplan-Meier methods and compared with log-rank tests. The risk factors (see Appendix) for mortality and late aortic re-operations were estimated using a multivariate proportional hazard regression analysis (Cox model). A p value of <0.05 was considered statistically significant. SPSS software, version 22.0 for Windows (IBM SPSS Inc., Chicago, IL) was used for all calculations.

In this study, between the CTD and the non-CTD groups, the early and late outcomes of the initial aortic surgery for aortic dissection were compared to demonstrate the impact of CTD on the surgical outcomes in the patients with aortic dissection associated with CMN.

As shown in Table 4, in the CTD group, there were initial surgical repairs of the ascending aorta in 5 patients and the aortic arch in 9 patients. Simultaneous aortic root repairs were performed in 11 of these 14 patients. The others underwent repairs of the descending aorta (in 10 patients), the thoracoabdominal aorta (in 2 patients), and the abdominal aorta (in 4 patients). In the non-CTD group, the surgical repairs were for the ascending aorta in 3 patients and the aortic arch in 7 patients. Simultaneous aortic root repair was carried out in 1 patient. The other 3 patients underwent repairs of the descending aorta (in 1 patient) and the thoracoabdominal aorta (in 2 patients). There were concomitant mitral valve surgeries in 2 patients in the CTD group.

There were no early deaths. In the long term, despite of there being no late deaths, new aortic dissection occurred at the same sites (three-channel aortic dissection) in 3 patients or at different sites in 3 patients (a total of 6 patients [20.0%] of the CTD group), whereas only 1 patient (7.7%) in the non-CTD group developed it at a different site (p = 0.310). The absence of re-dissection was not significantly different between the two groups (log-rank, p = 0.380) (Fig. 2), that is, 83.2% in the CTD group and 85.7% in the non-CTD at 5 years.

There were no significant differences in the re-operation rates between the Stanford aortic dissection classification groups (p = 0.332), that is, 13 of 21 with type A (62.0%) and 17 of 22 with type B (77.3%). There were also no significant differences in the absence of aortic re-operation (log-rank, p = 0.404), that is, 13.8% with type A and 14.5% with type B at 5 years. In contrast, re-operations were required more frequently (p = 0.037); there were 24 (80.0%) patients in the CTD group versus 6 patients (46.0%) in the non-CTD group. The absence of aortic reoperation tended to be worse (log-rank, p = 0.052), that is, 7.9% in the CTD group and 31.0% in the non-CTD group at 5 years (Fig. 3). Twelve of 24 CTD patients (50.0%) and 3 of 6 non-CTD patients (50.0%) required thirdly surgical interventions. Despite no substantial differences in the initial and following surgical procedures between the two patient groups, the rate of surgical interventions to the descending aorta was significantly higher in the CTD group (p = 0.021), that is, 16 patients (53.3%) versus 2 patients (15.4%). The rate of surgical interventions to the aortic root was also significantly higher in the CTD group (p = 0.037), that is, 15 patients (50.0%) versus 2 patients (15.4%). Regarding the time frame for aortic re-interventions, 15 of 24 patients (62.5%) in the CTD group underwent re-operations during the two years of follow-up after the initial surgery, compared with 3 of 6 patients (50.0%) in the non-CTD group. Univariate analysis revealed that patent false channel (p = 0.019), and simultaneous aortic root repair (p = 0.024) were significant risk factors for re-operation. In the multivariate analysis of risk factors for re-operation, patent false channel, simultaneous aortic root repair, and preoperative systemic hypertension were independent risk factors (Table 5).