Blood perfusion status is important in the prognosis of ventricular aneurysm complicated by ventricular septal rupture

Abstract

Aims

Due to its low incidence, poor prognosis, and high mortality in the acute phase, the long‐term prognosis of the left ventricular aneurysm (LVA) complicated by ventricular septal rupture (VSR) has received little attention. This study focus on the long‐term prognosis of patients with LVA complicated by relatively stable VSR.

Methods and results

Over a decade of retrospection, 68 patients with both LVA and VSR were compared with 136 patients with LVA alone after propensity score matching. Patients with both LVA and VSR were further divided into two groups depending on whether pre‐operative intra‐aortic balloon pump (IABP) was used (23 pre‐operative IABP vs. 45 non‐pre‐operative IABP). The primary endpoint was defined as major adverse cardiovascular and cerebrovascular events, a composite endpoint including mortality, myocardial infarction, revascularization, stroke, and heart failure. Patients with both LVA and VSR were generally in a worse condition upon admission compared with those with LVA alone [percentage of patients in New York Heart Association IV: 42.6% (29/68) vs. 11.0% (15/136), P < 0.001]. Both pre‐operative and post‐operative IABP use rates were significantly higher in patients with both LVA and VSR than in patients with LVA alone [pre‐operative IABP use rates: 33.8% (23/68) vs. 0.74% (1/136), P < 0.001 and post‐operative IABP use rates: 33.8% (23/68) vs. 10.3% (14/136), P < 0.001]. No significant difference was observed in the primary endpoint between patients with both LVA and VSR and those with LVA alone (log‐rank test, P = 0.63, median follow‐up time 63 months). We further investigated the effect of pre‐operative IABP on the long‐term prognosis of patients with both LVA and VSR. Patients who applied pre‐operative IABP had a worse long‐term prognosis than those who did not (log‐rank test, P = 0.0011).

Conclusions

The long‐term prognosis of LVA combined with VSR was not inferior than LVA alone after surgery, but poor blood perfusion status was associated with a worse prognosis.

Introduction

Ventricular septal rupture (VSR) and left ventricular aneurysm (LVA) are both severe mechanical complication of acute myocardial infarction (AMI). ¹ The incidence of VSR after AMI ranges from 0.17% to 0.31%. ² The incidence of LVA is higher, reaching 10%–15% ³ , ⁴ after AMI. LVA combined with left ventricular thrombosis, ⁵ arrhythmia, ⁶ and mitral valve regurgitation ⁷ have been reported, while LVA combined with VSR has not. It is obvious that LVA worsens the long‐term prognosis of patients. ⁸ LVA combined with VSR is even more challenging. ¹ Previous literature showed that the early mortality of VSR surgery varied from 7.56% to 52.4%. ⁹ However, it remains unknown whether the prognosis of patients with LVA + VSR differs from those with LVA alone.

Due to its low incidence, poor prognosis, and high mortality in the acute phase, the long‐term prognosis of LVA complicated by VSR has received little attention. With the development of mechanical circulatory support such as IABP, many patients with severe hemodynamic dysfunction also have the opportunity for surgery. This allowed us to include more patients with LVA + VSR. Here, we compared the perioperative and long‐term prognosis of patients with LVA + VSR and LVA only, and then analysed the factors affecting the long‐term prognosis of patients with LVA + VSR. We hope to promote the understanding of the long‐term prognosis of the rare patients with LVA + VSR.

Methods

Ethics statement

This retrospective study was approved by the local ethics committee of Fuwai Hospital, Beijing, China. The approval number is 2021‐1644, 3 March 2022. The written informed consent was waived due to the retrospective design.

Patient population and data collection

A total of 948 consecutive patients undergoing LVA surgery were reviewed at Fuwai Hospital, Beijing, China, from 1 January 2010 to 1 April 2021. We excluded 56 patients whose LVA or VSR were not caused by acute MI, or who had other serious complications like pseudoaneurysm (Figure  1 ). We then excluded 160 patients who had the LVA surgery without cardiopulmonary bypass (CPB), for all the patients with LVA complicated by VSR were operated on with CPB. Among the patients included in the study, 68 individuals underwent VSR repair along with LVA surgery and 664 individuals underwent LVA surgery only. Demographic data, imaging data, operation records, and post‐operative results were obtained from the information centre and medical records of our institution. Pre‐operative cardiac ultrasound and biochemical results were the first post‐admission examination. Post‐operative cardiac ultrasound results were the last results before discharge. Post‐operative biochemical results were recorded on the second day after surgery.

Figure 1.

Study flowchart. AOCA, anomalous origin of coronary artery; CPB, cardiopulmonary bypass; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; LVA, left ventricular aneurysm; VSD, ventricular septal defect; VSR, ventricular septal rupture.

Diagnosis of left ventricular aneurysm and ventricular septal rupture

Transthoracic echocardiography (TTE) was used for assessing LVA and VSR. Quantitative measurement of LVA was performed at the end of diastole in the apical four‐chamber view of TTE. Two parameters were measured including LVA basal diameter and height. VSR was diagnosed via Doppler echocardiography characterizing interventricular septal echo interrupted and left‐to‐right shunt.

Surgical strategy and post‐operative management

The operation indication of true ventricular aneurysm: (1) angina; (2) congestive heart failure; (3) recurrent ventricular arrhythmias; (4) ventricular mural thrombus. LVA complicated by VSR is deemed to have surgery without debate. All patients in this study were operated on CPB and hypothermia. The left ventricle was incised at the site of LVA 1–2 cm parallel to the left anterior descending artery or posterior descending artery. For LVA, the linear suture was mostly often performed, and purse string or patch plasty would be used when needed. For VSR, a Dacron patch or autologous pericardial patch was sutured to the ruptured ventricular septum. When the perforation area of the septum is relatively small, the direct suture can also be used. All patients were admitted to the intensive care unit (ICU) after surgery.

Follow‐up

All patients were suggested to visit our outpatient clinic 3, 6 months, and 1 year after discharge, and once every year afterward. If the patients returned to another clinic, we would make a phone call to obtain the follow‐up data. In short, follow‐up data were obtained via a combination of outpatient records and telephone interviews.

Study endpoint

The primary endpoint was a major adverse cardiovascular and cerebrovascular event, a composite of mortality, myocardial infarction, revascularization, stroke, and rehospitalization for heart failure.

Statistical analysis

Analysis was performed with SPSS 26.0 (IBM, Armonk, NY, USA) and R statistical software 4.2.1. (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables with a normal distribution were expressed as the mean and standard deviation (SD) and continuous variables with a non‐normal distribution were expressed as the median and interquartile range (IQR). Student's t‐test for independent samples was used to compare variables with a normal distribution, and the Mann–Whitney U test was used for variables that were not normally distributed. Paired t‐test was used to compare the changes of the same index before and after an operation. Categorical variables were expressed as frequencies and percentages and compared via the χ ² test or Fisher's exact test. The Kaplan–Meier method was applied to compare the cumulative risks of endpoints and the log‐rank test was used to compare the distributions of the time from surgery to the first event during the follow‐up. Cox regressions of the primary endpoint were used to select different variates that showed significance (P < 0.10) in univariable models. After adjusting for significant baseline covariates, the different variates in univariable models were then assessed using multivariable Cox regression with forward: LR variable selection. Receiver operator characteristic (ROC) and linear correlation analysis were used. Univariate binary logistic regression analysis was used to detect the efficiency of single‐factor regression. P < 0.05 was considered to be statistically significant. All tests were two‐tailed in this article.

We employed a 1:2 propensity score matching (PSM) analysis based on 16 characters: age, sex, BMI, smoking history, family history of coronary heart disease, hypertension, diabetes mellitus, hyperlipidaemia, ischaemic stroke history, previous PCI, left main coronary lesion, the number of lesioned vessels, LVA basal diameter, LVA height, LVA site, and left ventricular thrombus with a nearest neighbour‐matching algorithm. The standardized mean differences (SMD) < 0.1 indicated a balance between the two groups. Notably, those potential intermediate variables for the effects of VSR on surgical outcomes were not included in the matching, such as New York Heart Association (NYHA) functional classification, the timing of surgery, pre‐operative left ventricular end‐diastolic diameter (LVEDD), pre‐operative left ventricular ejection fraction (LVEF), and biochemical test results.

Results

Baseline characteristics

Baseline characteristics of patients in the two groups before and after propensity score matching were shown in Table 1 . After PSM, 16 baseline characteristics, like sex, age, previous history, ventricular aneurysm characters, and so on, were balanced between the two groups. Finally, 68 patients in the LVA + VSR group and 136 in the LVA group were included in the further analysis.

Table 1.

Baseline characteristics before and after propensity score matching

	Before matching			After matching
	LVA + VSR (N = 68)	LVA (N = 664)	P value	LVA + VSR (N = 68)	LVA (N = 136)	P value	SMD
Age, years	63.01 ± 7.95	57.73 ± 9.95	<0.001	63.01 ± 7.95	62.74 ± 8.57	0.822	0.034
Female	25 (36.8)	88 (13.3)	<0.001	25 (36.8)	48 (35.3)	0.959	0.031
BMI, kg/m2	23.85 ± 3.15	25.43 ± 5.75	0.026	23.85 ± 3.15	23.90 ± 3.68	0.928	0.014
Smoking history	39 (57.4)	437 (65.8)	0.208	39 (57.4)	76 (55.9)	0.960	0.030
CHD family history	10 (14.7)	79 (11.9)	0.631	10 (14.7)	19 (14.0)	1.000	0.021
Hypertension	36 (52.9)	327 (49.2)	0.651	36 (52.9)	72 (52.9)	1.000	<0.001
Diabetes mellitus	21 (30.9)	175 (26.4)	0.510	21 (30.9)	44 (32.4)	0.958	0.032
Hyperlipidaemia	34 (50.0)	414 (62.3)	0.063	34 (50.0)	70 (51.5)	0.961	0.029
Ischaemic stroke history	8 (11.8)	107 (16.1)	0.445	8 (11.8)	15 (11.0)	1.000	0.023
Previous PCI	15 (22.1)	160 (24.1)	0.821	15 (22.1)	34 (25.0)	0.772	0.069
LMC lesion	4 (5.9)	96 (14.5)	0.076	4 (5.9)	10 (7.4)	0.922	0.059
Number of lesioned vessels	2.04 ± 0.80	2.39 ± 0.82	0.001	2.04 ± 0.80	2.08 ± 0.94	0.783	0.042
LVA basal diameter, mm	42.59 ± 10.18	41.07 ± 10.65	0.260	42.59 ± 10.18	42.60 ± 11.76	0.996	0.001
LVA height, mm	29.41 ± 7.69	26.06 ± 8.52	0.002	29.41 ± 7.69	29.43 ± 9.95	0.987	0.002
LVA site			0.133			1.000	0.030
Apex and anterior wall	64 (94.1)	650 (97.9)		64 (94.1)	127 (93.4)
Inferior and posterior wall	4 (5.9)	14 (2.1)		4 (5.9)	9 (6.6)
LVT	9 (13.2)	191 (28.8)	0.009	9 (13.2)	15 (11.0)	0.818	0.068

Data are presented as mean ± SD or in parameter counts n (%).

BMI, body mass index; CHD, coronary heart disease; LMC, left main coronary; LVA, left ventricular aneurysm; LVT, left ventricular thrombus; PCI, percutaneous coronary intervention; SD, standard deviation.

Pre‐operative data not included in PSM were compared in Table 2 . Before surgery, LVEF (left ventricular ejection fraction) of the LVA + VSR group was higher than that of the LVA group [45% (40, 55) vs. 40% (35, 45), P < 0.001], and LVEDD smaller [55 mm (50, 59) vs. 58 mm (53, 63), P = 0.013]. Patients in the LVA + VSR group had worse cardiac function than the LVA group (NYHA IV 42.6% vs. 11.0%, P < 0.001) on admission, accompanied by worse renal function [creatinine 100.33 μmol/L (82.68, 124.49) vs. 83.61 μmol/L (72.05, 100.47), P < 0.001], decreased ALB [36.00 g/L (33.38, 40.00) vs. 40.75 g/L (38.40, 43.80), P < 0.001], and increased inflammation level [HSCRP 10.66 mg/L (3.81, 12.43) vs. 2.66 mg/L (1.45, 5.76), P < 0.001]. In particular, patients in the LVA + VSR group have higher rates of pulmonary hypertension (20.6% vs. 2.9%, P < 0.001) and lower HDL‐C [0.76 mmol/L (0.66, 0.92) vs. 1.06 mmol/L (0.88, 1.28), P < 0.001], without differences in cholesterol, triglyceride or LDL‐C.

Table 2.

Pre‐operative data

	LVA + VSR (N = 68)	LVA (N = 136)	P value
LVEF, %	45 [40, 55]	40 [35, 45]	<0.001
LVEDD, mm	55 [50, 59]	58 [53, 63]	0.013
NYHA			<0.001
I	0 (0.0)	5 (3.7)
II	18 (26.5)	63 (46.3)
III	21 (30.9)	53 (39.0)
IV	29 (42.6)	15 (11.0)
Arrhythmia	10 (14.7)	25 (18.4)	0.560
Atrial fibrillation	5 (7.4)	6 (4.4)	0.512
Ventricular arrhythmia	7 (10.3)	21 (15.4)	0.391
Pulmonary hypertension	14 (20.6)	4 (2.9)	<0.001
Creatinine, μmol/L	100.33 [82.68, 124.49]	83.61 [72.05, 100.47]	<0.001
Cholesterol, mmol/L	3.70 [3.08, 4.17]	3.79 [3.20, 4.47]	0.239
Triglyceride, mmol/L	1.47 [1.08, 1.77]	1.50 [1.06, 1.92]	0.471
LDL‐C, mmol/L	2.20 [1.72, 2.87]	2.24 [1.89, 2.96]	0.543
HDL‐C, mmol/L	0.76 [0.66, 0.92]	1.06 [0.88, 1.28]	<0.001
Uric acid, μmol/L	438.25 [359.66, 577.44]	365.25 [291.35, 436.78]	<0.001
ALB, g/L	36.00 [33.38, 40.00]	40.75 [38.40, 43.80]	<0.001
ALT, IU/L	28.50 [15.75, 52.00]	22.00 [15.75, 36.25]	0.215
HSCRP, mg/L	10.66 [3.81, 12.43]	2.66 [1.45, 5.76]	<0.001
MR			0.061
0	32 (47.1)	43 (31.6)
1	9 (13.2)	23 (16.9)
2	23 (33.8)	46 (33.8)
3	4 (5.9)	16 (11.8)
4	0 (0.0)	8 (5.9)
IABP	23 (33.8)	1 (0.74)	<0.001

Data are presented as median (IQR) or in parameter counts n (%).

ALB, albumin; ALT, alanine aminotransferase; HDL‐C, high‐density lipoprotein cholesterol; HSCRP, high sensitivity C reactive protein; IABP, intra‐aortic balloon pump; IQR, interquartile range; LDL‐C, low‐density lipoprotein cholesterol; LVEDD, left ventricular end diastolic dimension; LVEF, left ventricular ejection fraction; MR, mitral regurgitation, 0 means no regurgitation, 1 minimal, 2 small amount, 3 moderate, 4 extensive; NYHA, New York Heart Association.

To further investigate the role of HDL‐C in distinguishing patients with both LVA and VSR from patients with LVA alone, ROC curve was plotted (Figure 2 A ). The AUC (area under the curve) was 0.79033, indicating that low HDL‐C was associated with more VSR. Univariate binary logistic regression analysis yielded consistent results [odds ratio (OR), 0.007; 95% confidence interval (CI), 0.001–0.038; P < 0.001]. The HDL was in a negative correlation with HSCRP (Figure 2 B ).

Figure 2.

ROC analysis of HDL and linear regression of HDL and HSCRP. a, ROC curve of HDL in distinguishing LVA + VSR from LVA; b, linear regression of HDL and HSCRP in all patients.

There was no significant difference in CBP time [115 min (81, 135) vs. 103 (79, 135), P = 0.552] and cross‐clamping time [69.5 (57, 89) vs. 69 (54, 97), P = 0.741] (Table  3 ). The intraoperative bleeding volumes of the LVA + VSR group were more than those of the LVA group [600 (600, 632) vs. 600 (400, 600), P = 0.011], so was the blood transfusion rate (19.1% vs. 5.9%, P = 0.006).

Table 3.

Operation data and post‐operative results

	LVA + VSR (N = 68)	LVA (N = 136)	P value
Operation data
CBP time, min	115 [81, 135]	103 [79, 135]	0.552
Cross‐clamping time, min	69.5 [57, 89]	69 [54, 97]	0.741
LVA surgical strategy			0.019
Linear suture only	57 (83.8)	93 (68.4)
Purse string/patch plasty with or without linear suture	11 (16.2)	43 (31.6)
Concomitant CABG	46 (67.6)	115 (84.6)	0.007
Number of grafted vessels	1.22 ± 1.09	1.74 ± 1.03	0.001
Concomitant mitral valve surgery	4 (5.9)	22 (16.2)	0.045
Bleeding volume, mL	600 [600, 632]	600 [400, 600]	0.011
Blood transfusion	13 (19.1)	8 (5.9)	0.006
Post‐operative results
Creatinine, μmol/L	106.46 [85.42, 138.12]	97.70 [80.20, 122.93]	0.077
Mechanical ventilation, hour	31 [20, 62]	21[17, 37]	0.003
ICU stay, h	115.5 [91, 183]	87.5 [41, 135]	<0.001
Post‐operative length of stay, day	12 [9, 14]	10 [7, 13]	0.047
Drainage volume, mL	820 [558, 1,345]	873 [573, 1,310]	0.618
LVEF, %	45 [40, 51]	45 [40, 50]	0.283
LVEDD, mm	49.5 [45, 54]	53 [48, 57]	0.006
Reexploration surgery	4 (5.9)	3 (2.2)	0.225
Ischaemic stroke	0 (0.0)	2 (1.5)	0.553
MI	0 (0.0)	1 (0.7)	1.000
Renal replacement therapy	4 (5.9)	4 (2.9)	0.445
IABP	23 (33.8)	14 (10.3)	<0.001
LVAD	1 (1.5)	2 (1.5)	1.000
ECMO	0 (0.0)	2 (1.5)	0.553
30‐day death	2 (2.9)	3 (2.2)	1.000

Data are presented as mean ± SD, median (IQR) or in parameter counts n (%).

CABG, coronary artery bypass grafting; CBP, cardiopulmonary bypass; ECMO, extracorporeal membrane oxygenation; IABP, intra‐aortic balloon pump; ICU, intensive care unit; IQR, interquartile range; LVA, left ventricular aneurysm; LVAD, ventricle assisted device; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; MI, myocardial infarction; SD, standard deviation.

After surgery, the duration of mechanical ventilation, ICU stay, post‐operative length of stay, and the ratio of post‐operative IABP usage were all significantly higher in the LVA + VSR group. All the data above indicated that patients in the LVA + VSR group were in a more severe condition, while the 30‐day death did not differ between the two groups (LVA + VSR group 2.9% vs. LVA group 2.2%, P = 1.000). After left ventricular reconstruction surgery, LVEDD in the LVA group decreased by an average of 5.55 mm (paired t‐test, P < 0.001), and the LVA + VSR group by 5.62 mm (paired t‐test, P < 0.001). LVEF in the LVA group increased by an average of 3.62% (paired t‐test, P < 0.001). However, LVEF did not change significantly after surgery in the LVA+VSR group (paired t‐test, P = 0.456). More than half of the patients with LVA+VSR had a decrease in LVEF after surgery. No difference in discharge medication was observed between the two groups (Table  S1 ).

One‐year and long‐term follow‐up outcomes

About 90% of patients were followed up for more than 1 year. Three all‐cause deaths and five MACCEs happened in the LVA + VSR group; 4 deaths and 10 MACCEs in the LVA group. (Table  4 ) There was no statistical difference between two groups in the incidence of all‐cause death (Fisher's exact test, P = 0.684) and MACCE (Fisher's exact test, P = 1.000).

Table 4.

One‐year follow‐up of death and MACCE

	LVA + VSR	LVA	P value
Readmission for heart failure	1/56 (1.8)	0/118	‐
Death	3/58 (5.2)	4/121 (3.3)	0.683
MACCE	5/60(8.3)	10/124 (8.1)	1.000

Data are presented in parameter counts n/total (%).

MACCE, major adverse cardiovascular and cerebrovascular events.

The follow‐up rate was 98.0%, with the median follow‐up time 63 months (interquartile range: 32–93 months). No VSR recurrence after surgery was recorded. During the follow‐up, 5 all‐cause deaths and 16 MACCEs were found in the LVA + VSR group. Correspondingly, 15 deaths and 40 MACCEs in the LVA group. As was demonstrated in Figure 3 A , no significant difference was observed in the primary endpoint of MACCE (log‐rank test, P = 0.63) between the two groups. To avoid bias, we compared the long‐term prognosis of all LVA patients before PSM with those with LVA + VSR, and found no statistical difference (log‐rank test, P = 0.072). (Figure 3 B ).

Figure 3.

Kaplan–Meier analysis of the MACCE between LVA + VSR group and LVA group. (A) KM curve of LVA + VSR group and LVA group after PSM; (B) KM curve of LVA + VSR group and LVA group without matching.

Despite the fact that patients with LVA + VSR had worse pre‐operative outcomes, the data in our study do not support that patients with LVA + VSR have a worse long‐term prognosis than patients with LVA alone.

Blood perfusion status affects the long‐term prognosis

To further analyse what may affect the long‐term prognosis of patients with LVA + VSR, we conducted univariable cox regression analysis. Predictors of the primary endpoint were listed in Table 5 . Other insignificant variates detected in the univariable cox regression model were listed in Table S2 . We subsequently included age, pre‐operative IABP, post‐operative IABP, mechanical support, post‐operative LVEF, pre‐operative creatinine, post‐operative creatinine, pre‐operative uric acid, and female into the multivariable model after adjusting for age and sex. After adjusting, the pre‐operative IABP and post‐operative creatinine were still significant, especially the former one. (Table  6 ) We then compared the long‐term prognosis of patients with or without pre‐operative IABP. As shown in Figure 4 , patients who had used IABP before surgery had a significantly worse prognosis than those who have not (log‐rank test, P = 0.0011). In addition, we also compared the pre‐operative and post‐operative Society for Cardiovascular Angiography and Interventions (SCAI) classifications of the two groups, and the corresponding results are presented in Table S3 . Overall, it appears that patients who did not receive IABP pre‐operatively had lower pre‐operative and post‐operative SCAI grades, indicating better blood perfusion status. The pre‐operative vasoactive inotropic score (VIS) also indicated that the use of active medications in the IABP group far exceeded that in the non‐IABP group [200.00 (100.00, 400.00) vs. 0.00 (0.00, 150.00) P < 0.001] (Table  S4 ).

Table 5.

Predictors of primary endpoint on univariable Cox regression analysis

	HR (95% CI)	P value
Age	1.081(1.003–1.165)	0.042
Pre‐operative IABP	4.684(1.688–12.999)	0.003
Post‐operative IABP	3.543(1.281–9.799)	0.015
Mechanical support	2.114(0.949–4.707)	0.067
Post‐operative LVEF, %	0.936(0.875–1.001)	0.054
Pre‐operative creatinine, μmol/L	1.014(1.004–1.025)	0.008
Post‐operative creatinine, μmol/L	1.009(1.004–1.015)	0.001
Pre‐operative uric acid, μmol/L	1.004(1.002–1.006)	0.001

IABP, intra‐aortic balloon pump; LVEF, left ventricular ejection fraction.

Table 6.

Predictors of primary endpoint on multivariable cox regression analysis

	HR (95% CI)	P value
Pre‐operative IABP	3.520(1.172–10.567)	0.025
Post‐operative creatinine	1.006(1.000–1.013)	0.038

IABP, intra‐aortic balloon pump; LVEF, left ventricular ejection fraction.

Figure 4.

Kaplan–Meier analysis of the MACCE between non‐Pre‐operative IABP group and Pre‐operative IABP group.

Discussion

We reviewed 68 patients who underwent LVA + VSR surgery and 664 patients who underwent LVA surgery alone. In order to avoid the influence of factors such as age, gender, and size of ventricular aneurysm on the analysis, we performed PSM between the two groups. After PSM, 68 patients who had LVA + VSR surgery and 136 patients who had LVA surgery alone were further analysed.

On admission and perioperative results showed that patients with LVA + VSR were in an apparently worse condition than LVA alone. However, this retrospective PSM study showed that LVA complicated by VSR may not lead to a worse long‐term prognosis than LVA alone (log‐rank test, P = 0.63). Analysis without PSM also showed the same result (log‐rank test, P = 0.072). After univariable and multivariable models of Cox regression analysis, we identified that pre‐operative IABP and post‐operative creatinine were independent predictors of MACCE. Both of the two factors indicate that patients with hemodynamic disorders have a worse prognosis. The pre‐operative use of IABP and post‐operative elevation of creatinine both indicate inadequate systemic blood perfusion in patients. Attention to and active improvement of systemic perfusion in these patients are beneficial for improving long‐term outcomes.

Theoretically, the shape of the ventricle after left ventricular reconstruction is improved, and the LVEF value is also improved. Our data of LVA group also identified that LVEF increased by an average of 3.62% (paired t‐test, P < 0.001) after left ventricular reconstruction. However, LVEF in LVA + VSR group did not change significantly (paired t‐test, P = 0.456). Further analysis found that more than half of the patients in LVA + VSR group have a decreased LVEF after surgery. This phenomenon might be attributed to the elevated LVEF value caused by the left‐to‐right shunt before surgery. After repair of ventricular septal defect, LVEF also decreased. Additionally, patients with both LVA and VSR had a higher rate of pulmonary hypertension than those with LVA alone (20.6% vs. 2.9%, P < 0.001). It means that the left‐to‐right shunt through the rupture site was sufficient to place a greater load on the right ventricle and even affect pulmonary artery pressure within several months after MI. In the LVA group, the prevalence of severe mitral regurgitation is 5.9%, while no patients with severe regurgitation were observed in the LVA + VSR group. The proportion of concomitant mitral valve surgery is also greater in the LVA group than in the LVA+VSR group (16.2% vs. 5.9%, P = 0.045). This phenomenon may also be related to the ventricular septal defect. The defect site can reduce left ventricular pressure, thereby decreasing the stress on the mitral valve.

Intraoperative data indicates that, compared with patients in the LVA group, patients in the LVA + VSR group are more likely to undergo linear closure during left ventricular reconstruction (linear suture rate, 83.8% vs. 68.4%, P = 0.019), and this surgical preference may be related to the specific procedural steps. During the repair of the ventricular septal rupture, typically, a cut is made along the course of the anterior descending artery to fully expose the site of the rupture, whereas the patch technique, involving a circular incision at the infarcted site, is not conducive to optimal site exposure. This may be one of the reasons for the higher proportion of linear closure.

Patients in the LVA + VSR group also showed a significantly lower number of bypass grafts compared with the LVA group. (Number of grafted vessels, 1.22 ± 1.09 vs. 1.74 ± 1.03, P = 0.001) Ventricular septal rupture is often related to acute myocardial infarction, and once a defect occurs, surgery is typically performed promptly. Conversely, patients with isolated ventricular aneurysms tend to have a more stable condition, allowing for surgery to be performed several months after the myocardial infarction, often accompanied by diffuse coronary atherosclerosis. This may lead to a higher number of required bypass grafts in the LVA group. Additionally, the patients in the LVA + VSR group are often in critical condition, and the main objective of the surgery is to repair the ventricular septal defect. In order to save surgical time and reduce surgical ischaemic damage to the patients, a more conservative approach may be taken when deciding on the necessity of bypass grafts.

Previous studies have shown that older age and female gender were risk factors for VSR prognosis. ¹⁰ In our study, patients in the LVA + VSR group were older, more female, and had higher pre‐operative creatinine levels than those in the LVA group.

This study also has many limitations. Even though we included patients over a decade, our hospital only had 68 patients undergoing surgery for LVA + VSR. Future multi‐center studies may be possible to provide a more comprehensive and objective description of the prognosis of this kind of rare patients. There are still defeciencies in the description of the morphology of VSR. In nearly half of patients, the size of VSR was not reported on ultrasound report, so the analysis of the influence of VSR morphology on prognosis was unavailable. The disunity of BNP and ProBNP also made the analysis of this important heart failure indicator molecule unavailable.

In 68 LVA + VSR patients, there was only one patient used LVAD after surgery. The application of LVAD in China is not as common as abroad. In the future, with the wider application of LVAD in China, the perioperative and long‐term prognosis of these patients is expected to be improved.

Conclusions

Despite worse perioperative performance in the LVA + VSR group, no significant difference was observed in the long‐term prognosis between LVA + VSR and LVA alone after surgery, but poor blood perfusion status in the LVA + VSR group was associated with a worse prognosis.

Funding

This research was supported by National High Level Hospital Clinical Research Funding (2022‐GSP‐GG‐30).

Conflict of interest

The authors declare no conflicts of interest.

Supporting information

Table S1. Discharge with medicine.

Table S2. Insignificant variates detected in the univariable cox regression model.

Table S3. SCAI classifications pre and post surgery.

Table S4. VIS before surgery.

Huang, S. , Zhang, S. , Song, Y. , and Feng, W. (2024) Blood perfusion status is important in the prognosis of ventricular aneurysm complicated by ventricular septal rupture. ESC Heart Failure, 11: 1657–1665. 10.1002/ehf2.14733.