Twenty-year experience of supracoronary reconstruction with fibrin glue in type A acute aortic dissection

Abstract

Objective

In surgery for type A acute aortic dissection, we perform supracoronary reconstruction with fibrin glue when the sinus of Valsalva has no intimal tears, and the diameter is ≤ 45 mm. This study aimed to clarify the surgical outcomes with fibrin glue, the incidence of aortic root events, and the influence of the preoperative severity of aortic regurgitation (AR) on aortic root events.

Methods

Among 621 patients who underwent central repair in the past 20 years, we included 546 who underwent supracoronary reconstruction (87.9%). We closed the false lumen using fibrin glue and placed inner and outer polytetrafluoroethylene felt strips. The mild AR group included patients with no, trivial, or mild preoperative AR, whereas the severe AR group included patients with moderate or severe AR.

Results

Acute aortic root events occurred in 19 patients (3.5%), and the incidence was higher in the severe AR group than in the mild AR group (9 [5.9%] vs 8 [2.2%]; P = .029). The long-term outcomes were investigated in 486 patients who underwent supracoronary reconstruction and were subsequently discharged. Late events occurred in 13 patients (2.7%), the 5-year and 10-year event-free rates were 98.4% and 95.1%, respectively. The pseudoaneurysm-free rate was 99.8% at 10 years. The 5-year and 10-year event-free rates were lower in the severe AR group than in the mild AR group (97.1% vs 99.3% and 85.8% vs 98.5%, respectively; P = .001).

Conclusions

Supracoronary reconstruction can be performed safely with fibrin glue with excellent long-term outcomes. Preoperative AR is a risk factor for aortic root events.

Graphical Abstract

Our procedure of supracoronary reconstruction with fibrin glue.

Central Message.

Supracoronary reconstruction with fibrin glue resulted in excellent acute and long-term outcomes. Preoperative AR may be a risk factor for aortic root events.

Perspective.

Although there were a few acute events due to the absence of strong glues such as GRF glue and BioGlue (CryoLife Inc) during supracoronary reconstruction in surgery for type A acute aortic dissection, the long-term outcomes were extremely good. Preoperative AR reflects the severity of the aortic root dissection and is a risk factor for aortic root events.

In emergency surgery for type A acute aortic dissection (TAAAD), new techniques, such as the introduction of frozen elephant trunk¹ and measures against malperfusion² have been reported, and clinical practice has changed significantly with improvements in treatment outcomes. However, for the aortic root, apart from the introduction of valve-sparing techniques,³ there has been no significant progress in surgical procedures over the past 30 years. Two issues regarding the aortic root remain unresolved. The first is the choice of surgical procedure: to repair the dissected sinus of Valsalva (SOV) and perform supracoronary reconstruction (SCR) or to resect the SOV and perform aortic root replacement (ARR). The second is the advantages and disadvantages of using strong, tissue-toxic biological glue with aldehyde such as gelatin-resorcinol-formalin (GRF) glue and BioGlue (CryoLife Inc) when repairing the SOV.

If the dissected SOV is resected and the aortic root is replaced, it is possible to prevent aortic root events (ARE) due to residual dissected SOV tissue; however, the surgical procedure is complex and time consuming, and there is a risk of artificial valve-related events. The guidelines recommend ARR when dissection of the SOV is severe; however, no criteria are provided for what constitutes a severe case. In most TAAADs we had seen, the dissection extended beyond the sinotubular junction into the SOV. We believe that the extent and depth of dissection in the SOV indicates the severity of aortic root dissection; however, accurate measurements and quantification are difficult. There is a large difference in the application rates of ARR between institutions and countries. In Europe and the United States, the ARR rate is 25% to 30%, whereas in Japan, it is 5.7%,⁴ raising concerns about the occurrence of long-term ARE in Japan.

In the 1990s, Bachet and colleagues⁵ introduced GRF glue for TAAAD surgery. The use of this strong glue hardened fragile tissue and facilitated surgical manipulation and hemostasis, which in turn greatly improved surgical outcomes. However, the occurrence of pseudoaneurysms was subsequently reported,⁶ and BioGlue was used instead of GRF glue; however, the occurrence of pseudoaneurysms has also been reported with BioGlue.⁷ Regarding the advantages and disadvantages of using strong glue, prospective randomized trials are difficult because of the nature of TAAAD surgery as an emergency and lifesaving surgery. In addition, to improve the outcomes of emergency surgery, each institution often has a specific policy regarding the use of glue, and there have been no retrospective comparative studies from the same institution.

Since 2004, we have consistently performed SCR using fibrin glue without strong glue. In addition, patients with an SOV diameter of 45 mm or more or patients with an entry in the SOV have been considered indications for ARR. When the dissection extends to the commissure, aortic regurgitation (AR) occurs because of valve prolapse. Even if the AR is severe, we believe that repair is possible by fixing the commissure; therefore, AR severity is not included in the indications for ARR in our institution. By analyzing the short- and long-term results at our institution and comparing them with reports from other institutions, we present our views on surgical procedures for root surgery and the use of strong glue. In addition, assuming that the severity of AR before surgery reflects the severity of aortic root dissection, we compared the incidence of ARE at our institution according to AR severity.

Patients and Methods

At our institution, 621 emergency aortic replacements for TAAAD were performed over a 20-year period from 2004 to 2023. Almost all the patients were diagnosed using preoperative contrast-enhanced computed tomography (CT). Patients with an SOV diameter ≥45 mm on preoperative CT were candidates for ARR. The ascending aorta was incised during surgery to confirm the location of the entry tear, and ARR was considered for cases in which the entry tear was present in the SOV. Patients with hereditary aortic disease were also considered for ARR. For SCR, we closed the false lumen using fibrin glue, placed inner and outer polytetrafluoroethylene felt strips at the sinotubular junction, and secured them with 10 to 12 interrupted horizontal mattress sutures (Figure 1). Other detailed surgical procedures have been reported previously.⁸

Figure 1.

Our method for repair of the sinus of Valsalva using fibrin glue and polytetrafluoroethylene felt strips.

The patient flow diagram is shown in Figure 2. Of the 621 patients, 546 who underwent SCR were included and the occurrence of acute ARE was examined. The occurrence of long-term ARE was examined in 486 patients who survived and were discharged without acute ARE.

Figure 2.

Patient flow diagram. TAAAD, Type A acute aortic dissection; SOV, Sinus of Valsalva; SCR, supracoronary replacement; ARE, aortic root event; AR, aortic regurgitation.

Acute ARE in SCR was defined as aortic root bleeding requiring aortic reclamping, coronary artery occlusion occurring after unclamping, residual SOV dissection, pseudoaneurysm on CT scans performed during hospitalization after surgery, aortic lumen stenosis due to inner polytetrafluoroethylene felt, or residual severe AR. Long-term ARE was defined as the appearance ≥50 mm diameter of the SOV, severe AR, or the appearance of anastomotic pseudoaneurysm.

The patients were divided into 2 groups based on their preoperative echocardiographic evaluations. The mild AR group included patients with no, trivial, or mild preoperative AR, whereas the severe AR group included patients with moderate or severe AR. The occurrences of acute and long-term ARE were examined in both groups.

This retrospective observational study was approved by the Yokohama City University Institutional Review Board (F241000012; approved October 21, 2024), and the requirement for written consent was waived.

Statistical Analysis

Categorical variables are summarized as counts (percentages) and compared using χ² tests. Continuous variables are summarized as median and interquartile range and compared using the Wilcoxon rank-sum test. The follow-up time was calculated from the day of surgery to the date of the event or the last clinical follow-up visit. The Kaplan-Meier method was used to estimate survival from all-cause mortality and freedom from ARE, and differences between both groups were assessed using log-rank tests. All tests were 2-sided. Statistical analyses were performed using JMP Pro version 17.2.0 (SAS Institute Inc).

Results

The mean age of the 621 patients was 67 years (range, 57-75 years), and 338 (54.4%) were men. All patients underwent surgery within 14 days of onset, and 609 (98.0%) surgeries were initiated within 24 hours. A total of 169 patients (27.2%) had intramural hematoma. Of the 621 patients, 52 (8.4%) died in the hospital. Based on preoperative CT and examination of the intraoperative findings of the aortic root, the decision to perform ARR was made in 62 (10%) patients. In 12 (1.9%) patients, the dissection did not extend to the aortic root, and in 1 patient, BioGlue was used to close the false lumen. After excluding these cases, SCR was attempted in 546 patients. The characteristics of 546 patients according to AR severity are listed in Table 1. SCR was successfully completed in 527 out of 546 (96.5%) patients and acute ARE occurred in 19 out of 546 (3.5%) patients (Table 2). Root bleeding requiring aortic reclamping occurred in 9 out of 546 (1.6%) patients. Bleeding could be stopped with re-anastomosis alone in 3 patients, 5 underwent ARR (1 died due to cerebral infarction 7 days after surgery), and 1 patient experienced bleeding due to root rupture; given her preoperative condition (an 87-year-old woman with cardiopulmonary arrest who had been resuscitated), surgery was not continued. Coronary occlusion occurred in 3 out of 546 (0.5%) patients just after aortic unclamping determined by left ventricular akinesis. This was believed to be due to redissection of the SOV causing compression of the coronary artery, and 2 patients who underwent a saphenous vein graft bypass survived. Another 82-year-old resuscitated patient developed severe coagulopathy and left main trunk occlusion, and surgery was discontinued. Four residual dissections of the SOV were detected on the postoperative CT (4 out of 546 [0.7%]). These 4 patients were followed, and none required reoperation. Two patients had inner polytetrafluoroethylene felt stenosis, and 1 underwent reoperation (2 out of 546 [0.4%]). This complication occurred because the felt on the intimal side was short, and the proximal half (5 mm) of the 10-mm wide felt was inverted into the lumen by blood flow. After this experience, we decided to make the felt on the intimal side sufficiently long and to fix it at the proximal third. One patient had severe residual AR and required aortic valve replacement (1 out of 546 [0.2%]). The mortality among the 19 patients with acute ARE was 3 (15.8%), which was worse than the mortality of 41 (7.8%) among the 527 patients without ARE (P = .19); however, this was not statistically significant.

Table 1.

Patient characteristics according to severity of preoperative aortic regurgitation (AR)^∗

Variable	All patients	Mild AR group	Severe AR group	Missing data	P value
No. of patients	546	367 (67)	152 (28)	27 (5)
Age (y)	68 (58-75)	68 (58-75)	66 (58-76)	69 (61-74)	.8741
Male sex	282 (52)	187 (51)	81 (53)	14 (52)	.8890
Intramural hematoma	161 (29)	128 (35)	21 (14)	12 (44)	<.0001
CPA	28 (5)	15 (4)	10 (7)	3 (11)	.1815
Cardiac tamponade	169 (31)	111 (30)	47 (31)	11 (41)	.5230
Malperfusion	138 (25)	75 (20)	57 (38)	6 (22)	.0003
Total arch replacement	195 (36)	136 (37)	54 (36)	5 (19)	.1520

Values are presented as median (range) or n (%). CPA, Cardiopulmonary arrest.

^∗Twenty-seven patients had no data and could not be classified. Statistical analysis was performed only on the differences between the mild AR group and severe AR group.

Table 2.

Occurrence of aortic root events according to the severity of preoperative aortic regurgitation (AR)

Variable	All patients	Mild AR group	Severe AR group	Missing data
No. of patients	546	367	152	27
Acute ARE
Root bleeding	9 (1.6)	5 (1.4)	4 (2.6)	0
Coronary occlusion	3 (0.5)	1 (0.3)	0	2 (7.4)
Residual dissection of SOV	4 (0.7)	1 (0.3)	3 (2.0)	0
Inner felt stenosis	2 (0.4)	1 (0.3)	1 (0.7)	0
Residual severe AR	1 (0.2)	0	1 (0.7)	0
Total	19 (3.5)	8 (2.2)	9 (5.9)	2 (7.4)
Long-term ARE
Pseudoaneurysm	1 (0.2)	1 (0.3)	0	0
SOV >50 mm and/or severe AR	12 (2.2)	3 (0.8)	9 (5.9)	0
Total	13 (2.4)	4 (1.1)	9 (5.9)	0

Values are presented as or n (%). ARE, Aortic root event; SOV, sinus of Valsalva.

Long-term outcomes were investigated in 486 patients who underwent SCR without acute ARE and were discharged. The mean follow-up period was 6.5 years, with 5-year and 10-year survival rates of 90% and 78%, respectively. Long-term ARE occurred in 13 out of 486 (2.7%) cases, of which 12 had severe AR and/or an SOV diameter ≥50 mm, and 1 developed a pseudoaneurysm 6 months postoperatively. The pseudoaneurysm did not expand thereafter; no reoperation was required. Dilatation of the SOV to 50 mm or more was observed in 6 patients, 4 of whom also had severe aortic regurtitation. Most of the AR was caused by poor valve coaptation due to SOV dilatation, and no recurrence of prolapse was observed. The 5-year and 10-year ARE-free rates were 98.4% and 95.1%, respectively. The pseudoaneurysm-free rate was 99.8% at 10 years. Aortic root reoperation was performed in 8 out of 486 (1.6%) patients. Six patients developed long-term ARE and underwent ARR, and 2 patients underwent AVR for moderate AR simultaneously with total arch replacement for distal aortic dilatation. The 5-year and 10-year rates of freedom from aortic root reoperation were 99.4% and 97.7%, respectively.

Among the 546 patients, 367 belonged to the mild AR group, 152 belonged to the severe AR group, and 27 had missing data. The severe AR group had a significantly lower rate of intramural hematomas and a higher rate of malperfusion than the mild AR group. Echocardiography was performed before discharge in 495 patients, 480 (97.3%) patients had mild or less AR and 15 (2.7%) had moderate aortic regurgitation. The incidence of acute ARE was higher in the severe AR group than in the mild AR group (9 [5.9%] vs 8 [2.2%]; P = .029) (Table 2). The 5-year and 10-year ARE-free rates were higher in the mild AR group than in the severe AR group (99.3% vs 97.1% and 95.8% vs 85.8%, respectively; P = .001) (Figure 3).

Figure 3.

Long-term aortic root event-free rate. Confidence limits are 95%. AR, Aortic regurgitation; ARE, aortic root event.

Discussion

There are many reports on the long-term outcomes of aortic root surgery for TAAAD, but most compare the SCR and ARR at each institution. Because the indication for each surgery is decided according to the location of the tear and the SOV diameter, it is difficult to interpret the results of different surgeries performed on patients with different conditions. In addition, the rate of strong glue use varies depending on the institution's policy and the historical period of surgery, and it is believed that the use of strong glue also affects the comparison of outcomes between SCR and ARR. Therefore, we extracted only the SCR group from previous reports and compared the results (Table 3).^5,9, 10, 11, 12, 13, 14, 15, 16, 17 The rate of ARR varies greatly depending on the institution and is low (8.0%, 12.7%, and 10%) in reports from Japan (Nakajima and colleagues,¹⁰ Nishida and colleagues,¹⁴ and our current study). In addition, the aortic root reoperation-free rate in reports that clearly stated not using strong glue appears to be lower than that in reports with strong glue. Rylski and colleagues¹¹ reported the results of a series in which BioGlue was used routinely and stated that new aortic root lesions occurred in 26 (27%) of 119 patients who underwent SCR after an average of 4.4 years. This is an important study in that it pointed out that dissection of the three sinuses of Valsalva and dissection reaching the pelvis are risk factors for the occurrence of new aortic root lesions, but it seems that the use of glue had a large influence. Nishida and colleagues¹⁴ also used BioGlue in all 276 cases of SCR, and although the in-hospital mortality rate was 4.7%, the 5-year ARE-free rate was low at 68.7%. Concistre and colleagues⁶ reported that GRF glue is a risk factor for ARE, and even BioGlue is likely to be a risk factor for ARE.

Table 3.

Surgical outcomes of supracoronary replacement for type A acute aortic dissection in previous reports

Author	Journal	No. of patients	Rate of root surgery (%)∗	Hospital mortality (%)	Survival rate	Freedom from root reoperation	GRF or BioGlue† use
Bachet and colleagues, 19995	ATS	204	21.1	21.0	66%/5 y	87.6%/5 y	GRF 98%
Lai and colleagues, 20039	JTCVS	82	13.3	15	69%/6 y	95%/6 y	BioGlue sometimes
Nakajima and colleagues, 200710	ATS	100	8.0	9	75%/5 y	98%/5 y	No
Concistre and colleagues, 20126	ATS	173	25.2	15.6	73%/5 y	82%/5 y‡	GRF 22%
Rylski and colleagues, 201311	JTCVS	119	21.7	8	89.2%/5 y	91.6%/5.4 y	BioGlue 100%
Dell’Aquiia and colleagues, 201312	JTCVS	319	14.5	34	82%/5 y§	97%/5 y	Glue sometimes
Rylski and colleagues, 201413	ATS	489	22	11	69%/5 y	96%/5 y	Small amount of BioGlue
Nisida and colleagues, 201614	ATS	276	12.7	4.7	91.9%/5 y	68.7%/5 y	BioGlue 100%
Chiu and colleagues, et al, 201815	JTCVS	212	27.6	9	80%/5 y	88.2%/2.3 y	Glue sometimes
Sievers and colleagues, 201816	JTCVS	96		12.5	69.4%/5 y	92.2%/5 y	Glue sometimes
Brown and colleagues, 202417	JTCVS	370	38.4	11.6	77.1%/5 y	99.1%/5 y	No
Uchida and colleagues, 2025	Current study	527	10.0	8.4	90%/5 y§	99.4%/5 y	No

ATS, Annals of Thoracic Surgery; JTCVS, Journal of Thoracic and Cardiovascular Surgery.

^∗Rate of root surgery shows the percentage of patients who underwent root replacement or aortic valve and ascending replacement among all surgeries for type A acute aortic dissection reported in each article.

^†BioGlue (Cryolife Inc).

^‡Including reoperation for arch.

^§Survival among discharged patients.

Before the introduction of GRF glue 30 years ago, the procedure was performed without glue and only with felt reinforcement. Although it is impossible to examine the data from that time in detail, the surgical outcomes for type A dissection at that time were extremely poor. Redissection of the SOV, coronary artery occlusion, and aortic root rupture were frequently observed. Our purpose in using glue to close the false lumen was to restore the morphology of the dissected and deformed SOV and to prevent leakage from the needle hole of the proximal anastomosis into the false lumen. For this purpose, fibrin glue, which is not strong, was sufficient. Felt reinforcement must be used to ensure anastomotic strength. Neomedia, as reported by Rylski and colleagues,¹³ is an excellent method that can simultaneously prevent needle hole leakage and ensure anastomotic strength; however, it is easier to restore the morphology of the SOV with glue.

Nakajima and colleagues¹⁰ in 2007 and Brown and colleagues¹⁷ in 2024 reported a series that did not use a strong glue. In both studies, the hospital mortality rate was sufficiently low (9%, 11.6%), and the long-term survival (75%/5 year, 77.1%/5 year) and freedom from reoperation rates (98%/5 year, 99.1%/5 year) were excellent. However, the patients included in their study ultimately underwent SCR, and did not include cases in which the patient was converted to ARR for reasons such as aortic root rupture or re-dissection during surgery. Therefore, it is possible that the actual surgical results without using strong glue were overestimated. Therefore, we conducted this study on patients in whom SCR was attempted rather than on those in whom only SCR was performed.

Detailed review of surgical records revealed that 19 out of 546 patients (3.5%) had acute ARE. Among them, 6 out of 546 cases (1.1%) of ARE (residual severe AR, inner felt stenosis, and root bleeding in which reanastomosis alone was successfully completed) were believed to be due to the learning curve of young surgeons. On the other hand, 13 out of 546 cases (2.4%) of ARE (severe root bleeding, coronary artery occlusion, and residual SOV dissection) were believed to be due to insufficient strength of the fibrin glue. These events may have been prevented if a strong glue had been used. The operative mortality rate in patients with acute ARE was 15.8%, which was higher than the mortality of 7.8% among the patients without acute ARE, although not statistically significant. Given that saving lives during emergency surgery for TAAAD is of the utmost importance, the significance of using strong glue cannot be completely denied. Although the in-hospital mortality rate for TAAAD during the entire study period at our institution was a favorable 8.5%, it is possible that the use of strong glue could further improve acute outcomes.

However, our long-term results with fibrin glue were satisfactory. The 5-year and 10-year ARE-free rates were 98.4% and 95.1%, respectively. The pseudoaneurysm-free rate was 99.8% at 10 years. The aortic root reoperation free rate was 99.4% at 5 years and 97.7% at 10 years. These data were very good compared with previous reports using strong glue.

Patients with TAAAD are young and often require long-term reintervention of the distal aorta, and it is not uncommon for them to require life-threatening surgery 3 or more times in their lifetime. Considering the patient's quality of life, it is of great significance to be able to avoid reoperation on the aortic root. Considering the overall situation, we believe that the long-term benefits of using fibrin glue are greater than those of using strong glue in the acute phase.

Our ARR rate was 10%, which is lower than that in the West. The results of this study suggest that in 90% of patients with TAAAD in whom we performed SCR, moderate or severe AR may be a risk factor for the development of ARE, both during the acute phase and in the long-term. Most long-term ARE were worsening AR, accompanied by SOV expansion. The severity of preoperative AR may reflect the severity of SOV dissection. Should the indications for ARR be expanded to include severe AR? In our cohort, the rate of aortic root surgery was 17% if severe preoperative AR was added as an indication for ARR and 34% if moderate AR was also added, which is similar to the rate in Western countries. However, increasing the ARR may result in longer operative times and worse hospital mortality. Brown and colleagues¹⁷ performed SCR in 61.6% of patients with TAAAD and ARR in 38.4% and reported that the long-term outcomes of SCR were as good as those of root replacement. The long-term survival rate and freedom from root reoperation were similar to our data despite our lower ARR rate. Bojko and colleagues¹⁸ performed ARR in 18.3% of patients with TAAAD and reported that the cumulative incidence of root reoperation was unexpectedly higher than that in patients who underwent SCR, most often because of infection. Even in the severe AR group, postoperative AR was well controlled by SCR, and there were great benefits in reducing the operation time and avoiding the use of prosthetic valves. When preoperative AR is moderate or severe, it may be possible to actively consider ARR when the SOV diameter is close to 45 mm. Careful follow-up is required for patients undergoing SCR to check for ARE in the long term.

Study Limitations

This was a retrospective study conducted at a single institution, and the results were compared with those of previous reports from other institutions; however, the patient backgrounds and surgical methods were different. Because this was an emergency operation, there was a small amount of missing data regarding preoperative AR findings. Our hospital is an educational institution, and many surgeons have performed the surgery; therefore, it is likely that there was a learning curve for each surgical technique.

Conclusions

Acute ARE occurred in 3.5% of patients in whom SCR with use of fibrin glue was attempted, and in 2.4% of cases, the cause was believed to be insufficient strength of the glue. In contrast, long-term outcomes of SCR using fibrin glue were excellent compared with previous reports from other institutions that used strong glue. We will continue to perform SCR using fibrin glue instead of strong glue. Preoperative AR was a risk factor for the occurrence of ARE. In cases with an SOV diameter close to 45 mm and a preoperative AR of moderate or higher, it may be possible to actively consider the indications for ARR. If SCR is performed in such a situation, careful follow-up is required.

Conflict of Interest Statement

The authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.