Association between atherosclerotic index of plasma and long-term aortic-related adverse events in type B aortic dissection patients undergoing thoracic endovascular aortic repair

Abstract

Background

Previous research identifies the atherosclerotic index of plasma (AIP) as a key marker for cardiovascular risk, but its role in predicting outcomes in type B aortic dissection (TBAD) patients after thoracic endovascular aortic repair (TEVAR) is uncertain. This study aimed to investigate the association between AIP and long-term outcomes in TBAD patients after TEVAR.

Methods

This retrospective cohort study included 1335 patients with TBAD who underwent TEVAR. Patients were stratified into tertiles based on AIP levels. The primary endpoints were aortic-related adverse events (ARAEs) at 1 and 5 years after TEVAR. Cox regression analyses were used to evaluate the independent effect of AIP on outcomes. Kaplan–Meier (KM) analysis was conducted to compare the incidence of ARAEs among different groups. Restricted cubic spline (RCS) models were utilized to investigate the nonlinear relationship between AIP and ARAEs, and subgroup analyses assessed the stability of this association. Time-dependent receiver operating characteristic (ROC) curves were applied to assess the predictive accuracy of AIP for ARAEs over a 5-year period.

Results

The KM analysis revealed a significantly higher incidence of ARAEs in the high AIP group compared to the low AIP group (P < 0.001). However, no statistically significant differences were found in all-cause mortality and major adverse cardiovascular and cerebrovascular events (MACCEs) (all P > 0.05). Cox regression analysis demonstrated that a high level of AIP was associated with an increased risk of ARAEs (all P < 0.001). Additionally, RCS analysis indicated a linear relationship between AIP and the risk of ARAEs. In subgroup analyses, the timing of operation showed a significant interaction with 1-year ARAEs (P for interaction = 0.008). Time-dependent ROC analysis demonstrated an area under the curve approaching 0.8 throughout the 5-year period.

Conclusion

Our research indicates that AIP is independently associated with 1-year and 5-year ARAEs in patients with TBAD following TEVAR, providing a novel metabolic perspective for the prognostic evaluation of this population.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-026-03094-3.

Research insights

What is currently known about this topic?

The AIP has shown predictive value for cardiovascular diseases, but its association with long-term outcomes in patients with TBAD remains unclear.

What is the key research question?

Is there an association between AIP and the incidence of 1-year and 5-year outcomes in patients with TBAD following TEVAR?

What is new?

Our study demonstrated that AIP can independently predict 1-year and 5-year ARAEs in TBAD patients post-TEVAR, highlighting lipid metabolism’s role in TBAD prognosis.

How might this study influence clinical practice?

1. The pathogenesis of post-TEVAR ARAEs encompasses both anatomical complexity and the metabolic processes that regulate aortic remodeling. AIP may serve as a metabolic complement to anatomical indices.

2. AIP is a cost-effective and computationally simple measure that is readily accessible, as it utilizes standard lipid panel data obtained from routine pre-operative or post-operative evaluations.

3. Clinicians may enhance outcomes for patients with TBAD exhibiting elevated AIP by optimizing aortic imaging surveillance and initiating lipid-lowering therapies, thereby facilitating the early detection of ARAEs.

Introduction

Aortic dissection (AD) represents one of the most devastating cardiovascular diseases, with an annual incidence rate of 35 cases per 100,000 individuals [1]. Based on the dissection's location, AD is classified into Stanford type A aortic dissection (TAAD) and Stanford type B aortic dissection (TBAD). TBAD typically refers to dissection originating from the distal end of the left subclavian artery [2]. Treatment options for patients with TBAD primarily include open surgical intervention, pharmacotherapy, and thoracic endovascular aortic repair (TEVAR) [3]. TEVAR has demonstrated several advantages over conventional open surgical approaches, including diminished trauma and enhanced aortic remodeling, which is now considered the first-line treatment for complicated TBAD [4–6]. Nonetheless, aortic-related adverse events (ARAEs) such as retrograde type A dissection and rupture may arise following TEVAR and lead to a poor prognosis [7–10]. Consequently, there is an urgent need to identify prognostic markers to accurately assess the risk of ARAEs after TEVAR and facilitate timely intervention.

Dyslipidemia serves as a fundamental contributor to atherosclerosis and additionally influences the structural integrity and stability of the aortic wall through multiple mechanisms, thereby playing a role in the onset and progression of aortic dissection [11, 12]. In a study characterizing 389 patients diagnosed with TBAD, it was found that 24.4% of the cohort presented with hyperlipidemia [13], highlighting the pervasive nature of lipid dysregulation. Atherosclerotic index of plasma (AIP), reflecting both lipid homeostasis and progression of atherosclerosis, has emerged as a crucial biomarker with superior predictive accuracy for cardiovascular diseases compared to traditional lipid parameters [13–16]. Increasing evidence supports AIP's prognostic value in cardiovascular outcomes. The China Health and Retirement Longitudinal Survey (CHARLS) found that poor AIP control is linked to higher cardiovascular event risk in stage 0–3 cardiovascular-kidney-metabolic syndrome individuals [17]. The South Korean NHIS-HEALS study showed AIP effectively identifies high-risk individuals for cardiovascular events, even after accounting for traditional risk factors [18]. Additionally, Xicong Li et al. identified three AIP trajectory patterns, with high-stable AIP levels associated with a significantly increased risk of future cardiovascular disease [19]. Nevertheless, there is a lack of research evidence elucidating the specific relationship between AIP and the prognosis of TBAD patients following TEVAR.

Therefore, this study seeks to examine the correlation between AIP and the long-term adverse events in patients with TBAD following TEVAR.

Materials and methods

Research cohort and design

This study adopted a retrospective cohort design, including 1650 patients diagnosed with TBAD who received TEVAR at Shanghai Changhai Hospital from August 2011 to June 2024. The exclusion criteria were as follows: (1) participants with traumatic aortic injury and iatrogenic aortic dissection (n = 11); (2) participants with Turner syndrome, Marfan syndrome, Ehlers-Danlos syndrome, bicuspid aortic valve, giant cell arteritis, ankylosing spondylitis, Behçet's disease, or Takayasu arteritis (n = 46); (3) participants with a history of aortic surgery (n = 24); (4) participants with a history of malignant tumors (n = 48); (5) participants lacking perioperative serum data (n = 116); and (6) individuals missing high-density lipoprotein cholesterol and triglyceride data (n = 70). Ultimately, a total of 1335 patients were included in this study (Fig. 1). The research protocol received approval from the Ethics Committee of Shanghai Changhai Hospital (CHEC-Y2020-042). Given the retrospective nature of this study, the requirement for informed consent was waived.

Fig. 1.

Flowchart of the patient selection process

Data collection and definition

Data for all research participants were obtained via the electronic medical record system, encompassing demographic characteristics, comorbidities, laboratory test results, anatomical features, intraoperative details, and discharge medication profiles. Variables exhibiting a missing data rate greater than 20% were excluded during the initial data collection and screening phase to mitigate potential bias arising from excessive missing values.

According to the reporting standards of Society for Vascular Surgery/Society of Thoracic Surgeons [5], TBAD was classified as acute (≤ 14 days, including hyperacute cases with < 24 h of symptom onset), subacute (15–90 days), and chronic (≥ 90 days) in this study. AIP was calculated as log (triglycerides/high-density lipoprotein cholesterol [TG/HDL-C]). Participants were divided into three different groups based on the tertiles of AIP: Low (AIP < 0.27, n = 688), Middle (0.27 ≤ AIP < 0.52, n = 319), and High (AIP ≥ 0.52, n = 328). Laboratory tests were performed by the Laboratory Department of Shanghai Changhai Hospital. Fasting blood samples were obtained from patients scheduled for elective surgery at 6:00 a.m. on the day of their operation. For cases of an emergent nature, blood samples were collected either in the emergency department or during the surgical procedures. An automated blood cell counter was used to count lymphocyte, neutrophil and other blood cells (XN-9000, Sysmex). Biochemical and coagulation function parameters, such as total cholesterol (TC, mg/dL), triglycerides (TG, mg/dL), high-density lipoprotein cholesterol (HDL-C, mg/dL), low-density lipoprotein cholesterol (LDL-C, mg/dL), D-dimer (mg/L), among others, were evaluated utilizing an automated biochemical analyzer (HITACHI 7600–120) and a coagulation analyzer (STAGO EVOLUTION). Previous studies have demonstrated that derivative inflammatory and metabolic indices, such as the neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), systemic immune-inflammation index (SII), aggregate index of systemic inflammation (AISI), platelet-to-lymphocyte ratio (PLR), systemic inflammation response index (SIRI), and uric acid-to-high-density lipoprotein cholesterol ratio (UHR), are predictive of the occurrence and prognosis of aortic diseases [20–22]. In this study, we compared these indices with AIP to evaluate the prediction efficacy for the outcomes of TBAD.

Endpoints and follow-up

The primary endpoints were defined as ARAEs occurring within 1-year and 5-years following TEVAR. In cases where patients experienced multiple adverse events, only the initial occurrence was included in the analysis. The evaluation of all ARAEs was performed during routine follow-up visits involving computed tomography angiography (CTA), scheduled at one month, six months, and annually thereafter, or during unscheduled CTA examinations prompted by rehospitalization due to emergent symptoms. ARAEs were defined as composite endpoints as follows [23, 24]: (1) Retrograde aortic dissection, characterized by a new intimal tear extending proximally to the aortic arch or ascending aorta, with no evidence of an iatrogenic cause during TEVAR; (2) Aortic rupture, confirmed by CTA or surgical findings, excluding traumatic or iatrogenic rupture; (3) Aortic dilation, defined by either of two criteria: an absolute maximum diameter of the dissected segment ≥ 55 mm or a relative increase in aortic diameter ≥ 5 mm compared to postoperative CTA or ≥ 5 mm per six months during follow-up; (4) Malperfusion is characterized by clinical manifestations of end-organ ischemia, such as renal insufficiency, lower limb ischemia, and severe intestinal malperfusion, in conjunction with computed CTA evidence of true lumen compression or branch artery occlusion; (5) Type I or III endoleak is indicated by persistent contrast accumulation outside the stent-graft but within the aortic wall. Secondary endpoints include all-cause death and major adverse cardiovascular and cerebrovascular events (MACCEs) within 1 year and 5 years. MACCEs are defined as a composite of nonfatal myocardial infarction, nonfatal stroke and cardiovascular death [25, 26].

All patients were followed up through medical record reviews, telephone interviews, and outpatient visits. Adverse events were assessed through a thorough review of clinical records that necessitated readjustment or evaluation during outpatient visits.

Statistical analysis

Continuous variables were presented as mean ± standard deviation (SD) or median with interquartile range (Q1–Q3), and were subsequently compared using one-way analysis of variance (ANOVA) or the Kruskal–Wallis H test, depending on the distribution of the data. Categorical variables were expressed as percentages and analyzed utilizing the chi-squared (χ²) test or Fisher's exact test. The cumulative survival curve was constructed using the Kaplan–Meier (KM) method, with intergroup differences evaluated using the log-rank test. To investigate the association between AIP and outcomes at 1 year and 5 years, Cox proportional hazards model was applied. Initially, a univariate analysis was performed, and variables exhibiting P-values less than 0.1 were subsequently included in the multivariate Cox model. Restricted cubic spline (RCS) analysis was used to evaluate the potential nonlinear associations between AIP and outcomes. Subgroup analysis was conducted to investigate the robustness of the association between AIP and ARAEs. The time-dependent receiver operating characteristic (ROC) curve was employed to assess the predictive efficacy of AIP for ARAEs within a 5-year period. Statistical analyses were carried out using R (version 4.4.0; survival package; rms package; timeROC package) and SPSS software version 27.0. In this study, P value < 0.05 was deemed statistically significant.

Result

Baseline characteristics of the participants

Table 1 presents the baseline characteristics of the patients. The average age was 59.0 ± 13.22 years, with significant differences among the groups (Low: 59.23 ± 13.37, Middle: 60.40 ± 13.47, High: 57.17 ± 12.48, P = 0.004). The high AIP group had a significantly higher BMI (Low: 24.03 ± 3.62, Middle: 24.49 ± 3.63, High: 25.01 ± 3.19, P < 0.001), more men (81.10% vs. 80.25% vs. 86.89%, P = 0.042), and higher smoking (40.26% vs. 53.29% vs. 57.62%, P < 0.001) and drinking rates (15.12% vs. 17.55% vs. 22.26%, P = 0.020). No significant differences were found in systolic or diastolic blood pressure (all P > 0.05). Diabetes mellitus was more common in the high AIP group (6.40% vs. 9.72% vs. 11.89%, P = 0.009), while chronic obstructive pulmonary disease (COPD) was more prevalent in the middle group (9.74% vs. 14.73% vs. 9.76%, P = 0.045). No significant differences were observed in other comorbidities such as pericardial and pleural effusion, major artery ischemia, hypertension, stroke, and chronic kidney disease (all P > 0.05).

Table 1.

Baseline characteristics of the study participants

Variable	AIP				P-value
	Total	Low	Middle	High
N	1335	688	319	328
Age (years)	59.0 ± 13.22	59.23 ± 13.37	60.4 ± 13.47	57.17 ± 12.48	0.004
Male	1099 (82.32)	558 (81.10)	256 (80.25)	285 (86.89)	0.042
BMI, kg/m2	24.38 ± 3.54	24.03 ± 3.62	24.49 ± 3.63	25.01 ± 3.19	< 0.001
SBP at admission, mmHg	137.85 ± 21.69	139.02 ± 22.3	136.39 ± 21.29	136.81 ± 20.68	0.155
DBP at admission, mmHg	82.13 ± 11.30	82.60 ± 11.42	81.34 ± 11.06	81.90 ± 11.25	0.409
Smoking	636 (47.64)	277 (40.26)	170 (53.29)	189 (57.62)	< 0.001
Drinking	233 (17.45)	104 (15.12)	56 (17.55)	73 (22.26)	0.020
Comorbidities
Pericardial effusion	88 (6.59)	44 (6.40)	27 (8.46)	17 (5.18)	0.233
Pleural effusion	437 (32.73)	225 (32.70)	116 (36.36)	96 (29.27)	0.157
Ischemia of major arteries	84 (6.29)	36 (5.23)	22 (6.90)	26 (7.93)	0.224
Hypertension	1002 (75.06)	509 (73.98)	234 (73.35)	259 (78.96)	0.166
Diabetes mellitus	114 (8.54)	44 (6.40)	31 (9.72)	39 (11.89)	0.009
Stroke	74 (5.54)	29 (4.22)	25 (7.84)	20 (6.10)	0.057
COPD	146 (10.94)	67 (9.74)	47 (14.73)	32 (9.76)	0.045
CKD	74 (5.54)	35 (5.09)	20 (6.27)	19 (5.79)	0.728
Laboratory tests
WBC, × 109/L	8.63 ± 3.39	8.82 ± 3.52	8.49 ± 3.25	8.37 ± 3.25	0.020
Lymphocytes, × 109/L	1.37 ± 0.59	1.30 ± 0.53	1.40 ± 0.60	1.49 ± 0.67	< 0.001
Monocytes, × 109/L	0.61 ± 0.29	0.61 ± 0.30	0.63 ± 0.29	0.59 ± 0.29	0.102
Neutrophils, × 109/L	6.48 ± 3.27	6.72 ± 3.47	6.33 ± 3.29	6.12 ± 3.19	0.001
Hemoglobin, g/L	127.88 ± 18.11	127.23 ± 17.29	128.24 ± 19.16	128.90 ± 18.73	0.205
D-dimer, mg/L	3.51 ± 4.31	3.55 ± 4.34	3.33 ± 4.04	3.61 ± 4.50	0.177
Creatinine, μmol/L	101.33 ± 103.76	102.82 ± 107.17	103.94 ± 115.6	95.66 ± 82.19	0.952
Blood glucose, mg/dL	6.53 ± 2.35	6.36 ± 1.51	6.71 ± 3.80	6.7 ± 1.92	0.023
FDP, mg/L	12.96 ± 17.57	13.08 ± 17.09	13.26 ± 19.75	12.42 ± 16.32	0.128
Platelet, × 109/L	203.28 ± 74.35	195.91 ± 68.03	209.89 ± 80.78	212.32 ± 79.03	< 0.001
TC, mg/dL	170.19 ± 37.31	166.34 ± 24.42	169.08 ± 36.16	179.33 ± 55.11	0.001
TG, mg/dL	120.41 ± 102.42	88.84 ± 23.32	108.15 ± 24.41	198.56 ± 180.83	< 0.001
LDL-C, mg/dL	41.71 ± 136.46	27.24 ± 31.06	55.32 ± 195.14	58.81 ± 189.72	< 0.001
HDL-C, mg/dL	44.15 ± 11.54	48.52 ± 11.49	43.11 ± 9.07	36.02 ± 8.85	< 0.001
Uric acid, μmol/L	99.38 ± 1091.88	92.85 ± 915.95	55.6 ± 20.41	155.69 ± 1759.38	< 0.001
NLR	6.22 ± 5.97	6.77 ± 6.43	5.72 ± 5.32	5.55 ± 5.55	< 0.001
LMR	2.68 ± 1.86	2.63 ± 2.14	2.55 ± 1.35	2.90 ± 1.60	< 0.001
SII	1195.64 ± 1237.62	1231.68 ± 1187.96	1176.69 ± 1310.95	1138.49 ± 1267.31	0.013
AISI	770.52 ± 1031.74	765.71 ± 907.58	796.28 ± 1158.18	755.57 ± 1143.1	0.026
PLR	174.21 ± 106.05	174.97 ± 98.66	173.77 ± 107.71	173.05 ± 118.94	0.456
SIRI	3.90 ± 4.91	4.16 ± 5.08	3.81 ± 4.47	3.44 ± 4.96	< 0.001
UHR	3.57 ± 49.49	1.89 ± 0.52	2.22 ± 0.89	8.42 ± 99.80	< 0.001
Anatomical characteristics
Maximal aortic diameter	39.41 ± 12.48	39.82 ± 11.89	37.99 ± 13.78	39.93 ± 12.16	0.490
False lumen diameter	27.70 ± 14.16	27.47 ± 14.14	26.29 ± 13.38	28.96 ± 14.75	0.327
Dissection length	312.58 ± 191.42	328.9 ± 195.0	290.4 ± 187.40	308.4 ± 189.70	0.339
Intraoperative details
Timing of operation					0.247
Acute	975 (73.03)	505 (73.40)	240 (75.24)	230 (70.12)
Subacute	226 (16.93)	107 (15.55)	52 (16.30)	67 (20.43)
Chronic	134 (10.04)	76 (11.05)	27 (8.46)	31 (9.45)
Branch	200 (14.98)	99 (14.39)	44 (13.79)	57 (17.38)	0.364
Adjunct	213 (15.96)	118 (17.15)	49 (15.36)	46 (14.02)	0.421
Hybrid	17 (12.73)	8 (1.16)	4 (1.25)	5 (1.52)	0.890
Discharge medications
Antiplatelet drugs	422 (31.61)	193 (28.05)	109 (34.17)	120 (36.59)	0.013
ACEI/ARB	579 (43.37)	316 (45.93)	124 (38.87)	139 (42.38)	0.100
β-blocker	627 (46.97)	337 (48.98)	137 (42.95)	153 (46.65)	0.201
CCB	728 (54.53)	389 (56.54)	171 (53.61)	168 (51.22)	0.262
Statins	85 (6.37)	37 (5.38)	19 (5.96)	29 (8.84)	0.101
Anticoagulants	17 (1.27)	9 (1.31)	4 (1.25)	4 (1.22)	0.992

Categorical variables were presented as n(%), while continuous variables were expressed as mean ± standard deviation (SD). BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease. WBC, white blood cell; FDP, fibrinogen degradation products; TC, total cholesterol; TG, triglycerides; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; NLR, neutrophil-to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio; SII, systemic immune-inflammation index; AISI, advanced inflammation-based score index; PLR, platelet-to-lymphocyte ratio; SIRI, systemic inflammatory response index; UHR, uric acid to high-density lipoprotein cholesterol ratio

In the high AIP group, there was a significant elevation in the concentrations of lymphocytes, platelets, TC, TG, LDL-C, LMR, and UHR compared to the other two groups (all P < 0.05). Conversely, the levels of white blood cells (WBC), neutrophils, HDL-C, NLR, SII, and SIRI were significantly reduced in the high AIP group (P < 0.05). The middle AIP group exhibited higher values of blood glucose and AISI (P = 0.023 and P = 0.026, respectively). No significant differences were observed in other laboratory parameters, including monocytes, hemoglobin, D-dimer, creatinine, and fibrinogen degradation products (FDP) (all P > 0.05). Additionally, the utilization rate of antiplatelet drugs was higher in the high AIP group (28.05% vs. 34.17% vs. 36.59%, P = 0.013). Conversely, no significant differences were observed in anatomical characteristics, intraoperative details, or other discharge medications. (all P > 0.05).

1-year and 5-year outcomes

The cumulative incidences of ARAEs, MACCEs, and all-cause mortality at both 1-year and 5-year intervals are detailed in Table 2. Within the 1-year follow-up period, 131 cases of ARAEs, 43 cases of all-cause mortality, and 25 cases of MACCEs were identified. Over the 5-year follow-up period, the incidences increased to 199 cases of ARAEs, 114 cases of all-cause mortality, and 145 cases of MACCEs. The cumulative incidence of ARAEs was most pronounced in the group with elevated AIP levels, exhibiting rates of 19.51% at the 1-year mark and 24.39% at the 5-year mark. Both of these rates were statistically significant (P < 0.001), as illustrated in Fig. 2. The analysis indicated no statistically significant differences in all-cause mortality and MACCEs among the three groups at both the 1-year (P = 0.420 and P = 0.407, respectively) and 5-year (P = 0.336 and P = 0.135, respectively) follow-up periods (refer to Fig. S1).

Table 2.

Outcomes of TBAD patients receiving TEVAR grouped according to AIP

Variable	Total	Low	Middle	High	P-value
1-year outcomes
Cumulative incidence of 1-year ARAEs	131 (9.81)	40 (5.81)	27 (8.46)	64 (19.51)	< 0.001
Cumulative incidence of 1-year all-cause death	43 (3.22)	21 (3.05)	8 (2.51)	14 (4.27)	0.420
Cumulative incidence of 1-year MACCEs	25 (1.87)	11 (1.60)	5 (1.57)	9 (2.74)	0.407
5-year outcomes
Cumulative incidence of 5-year ARAEs	199 (14.91)	71 (10.32)	48 (15.05)	80 (24.39)	< 0.001
Cumulative incidence of 5-year all-cause death	114 (8.54)	57 (8.28)	23 (7.21)	34 (10.37)	0.336
Cumulative incidence of 5-year MACCEs	145 (10.86)	71 (10.32)	29 (9.09)	45 (13.72)	0.135

Values are presented as n (%). TBAD, type B aortic dissection; TEVAR, thoracic endovascular aortic repair; AIP, atherogenic index of plasma; ARAEs, aortic-related adverse events; MACCEs, major adverse cardiovascular and cerebrovascular events

Fig. 2.

Kaplan–Meier survival analysis of 1-year and 5-year ARAEs for TBAD patients undergoing TEVAR. A The 1-year freedom from ARAEs, B The 5-year freedom from ARAEs. ARAEs, aortic-related adverse events; TEVAR, thoracic endovascular aortic repair; TBAD, type B aortic dissection

The association between AIP and outcomes of TBAD patients receiving TEVAR

In the univariate model, a significant association was found between AIP (categorical) and the risk of 1-year ARAEs (HR = 3.40, 95% CI: 2.29–5.05; P < 0.001). After adjusting for potential confounders, the high AIP group presented an independent risk of 1-year ARAEs compared to the low AIP group (HR = 3.39, 95% CI: 2.27–5.08; P < 0.001) (Table 3). The link between the AIP levels and 5-year ARAE was shown in Table 4. Cox analysis indicated that, after adjusting for variables, compared with patients having the lowest tertile of AIP, patients with the highest AIP tertile presented a significantly elevated risk of ARAEs (HR = 2.33, 95% CI: 1.63–3.33; P < 0.001). When assessed as a continuous variable, AIP was identified as a significant risk factor in both the unadjusted model (HR = 2.32, 95% CI: 1.19–4.55; P = 0.014) and the fully adjusted model (HR = 2.28, 95% CI: 1.15–4.55; P < 0.019) for 1-year ARAEs, demonstrating predictive efficacy that surpasses all other composite indicators. In the analysis of 5-year ARAEs, AIP showed significance in the univariate analysis (HR = 1.77, 95% CI: 1.03–3.10; P < 0.047) but did not achieve statistical significance in the multivariable analysis (HR = 1.78, 95% CI: 0.95–3.32; P < 0.072) (Table 5). Additionally, univariate Cox regression analysis showed that AIP was not significantly associated with either all-cause mortality or MACCEs over the 5‑year follow‑up (Tables S1–S2).

Table 3.

Univariate and multivariate Cox proportional hazards model for 1-year ARAEs

Variable	Univariate analysis			Multivariate analysis
	HR	95% CI	P-value	HR	95% CI	P-value
Age	1.00	(0.99, 1.02)	0.473
Male	1.31	(0.81, 2.14)	0.271
BMI	1.00	(0.96, 1.05)	0.921
Smoking	1.12	(0.79, 1.57)	0.530
Drinking	1.00	(0.66, 1.53)	0.987
SBP at admission	1.00	(0.99, 1.01)	0.580
DBP at admission	1.00	(0.98, 1.02)	0.953
Comorbidities
Pericardial effusion	2.14	(1.28, 3.56)	0.004	2.02	(1.20, 3.38)	0.008
Pleural effusion	1.06	(0.73, 1.52)	0.765
Ischemia of major arteries	2.25	(1.27, 3.99)	0.006	1.67	(0.52, 5.33)	0.385
Hypertension	1.04	(0.69, 1.55)	0.856
Diabetes mellitus	1.36	(0.79, 2.33)	0.265
Stroke	2.11	(1.21, 3.68)	0.008	1.92	(1.08, 3.41)	0.027
COPD	0.66	(0.35, 1.26)	0.207
CKD	1.65	(0.91, 2.99)	0.097	1.40	(0.77, 2.56)	0.272
Laboratory tests
WBC	1.05	(1.01, 1.10)	0.014	1.06	(1.02, 1.11)	0.007
Hemoglobin	0.99	(0.98, 1.00)	0.110
Creatinine	1.00	(1.00, 1.00)	0.212
FDP	1.00	(0.99, 1.01)	0.985
Blood glucose	1.03	(0.98, 1.07)	0.228
AIP groups
Low	1.00			1.00
Middle	1.41	(0.87, 2.30)	0.167	1.36	(0.83, 2.24)	0.218
High	3.40	(2.29, 5.05)	< 0.001	3.39	(2.27, 5.08)	< 0.001
Anatomical characteristics
Maximal aortic diameter	1.05	(1.00, 1.09)	0.052	1.16	(0.94, 1.44)	0.169
False lumen diameter	1.10	(1.01, 1.21)	0.036	1.00	(0.87, 1.16)	0.956
Dissection length	1.00	(1.00, 1.01)	0.081	1.00	(1.00, 1.01)	0.378
Intraoperative details
Timing of operation
Acute	1.00
Subacute	0.98	(0.63, 1.52)	0.932
Chronic	1.25	(0.73, 2.16)	0.416
Branch	1.28	(0.83, 1.97)	0.273
Adjunct	1.04	(0.66, 1.63)	0.861
Hybrid	2.03	(0.65, 6.37)	0.226
Discharge medications
Antiplatelet drugs	0.88	(0.60, 1.27)	0.488
ACEI/ARB	1.40	(1.00, 1.98)	0.052	1.41	(1.00, 1.99)	0.051
β-blocker	0.78	(0.55, 1.10)	0.153
CCB	0.76	(0.54, 1.07)	0.112
Statins	1.16	(0.61, 2.22)	0.646
Anticoagulants	1.77	(0.56, 5.56)	0.328

Variables with a P value < 0.1 in univariable analysis were entered in the multivariable models. ARAEs, aortic-related adverse events; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; WBC, white blood cell; FDP, fibrinogen degradation products; AIP, atherogenic index of plasma; ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor blocker; CCB, calcium channel blockers; HR, hazard ratio; CI, confidence interval

Table 4.

Univariate and multivariate Cox proportional hazards model for 5-year ARAEs

Variable	Univariate analysis			Multivariate analysis
	HR	95% CI	P-value	HR	95% CI	P-value
Age	1.01	(1.00, 1.02)	0.143
Male	1.25	(0.85, 1.84)	0.261
BMI	0.98	(0.94, 1.02)	0.369
Smoking	1.08	(0.82, 1.43)	0.577
Drinking	1.08	(0.78, 1.50)	0.636
SBP at admission	1.00	(0.99, 1.01)	0.827
DBP at admission	1.00	(0.98, 1.01)	0.681
Comorbidities
Pericardial effusion	2.01	(1.30, 3.10)	0.002	1.89	(1.15, 3.10)	0.012
Pleural effusion	1.08	(0.80, 1.45)	0.609
Ischemia of major arteries	1.68	(0.99, 2.84)	0.054	1.62	(0.58, 4.52)	0.352
Hypertension	1.15	(0.82, 1.61)	0.416
Diabetes mellitus	1.37	(0.89, 2.12)	0.151
Stroke	1.98	(1.22, 3.21)	0.006	2.04	(1.23, 3.40)	0.006
COPD	1.00	(0.63, 1.57)	0.990
CKD	1.60	(0.98, 2.64)	0.062	1.19	(0.65, 2.16)	0.577
Laboratory tests
WBC	1.03	(1.00, 1.07)	0.077	1.04	(1.00, 1.09)	0.046
Hemoglobin	0.99	(0.99, 1.00)	0.049	1.00	(0.99, 1.01)	0.459
Creatinine	1.00	(1.00, 1.00)	0.184
FDP	1.00	(0.99, 1.01)	0.626
Blood glucose	1.01	(0.96, 1.06)	0.629
AIP groups
Low	1.00			1.00
Middle	1.47	(1.02, 2.12)	0.040	1.45	(0.97, 2.16)	0.069
High	2.47	(1.80, 3.41)	< 0.001	2.33	(1.63, 3.33)	< 0.001
Anatomical characteristics
Maximal aortic diameter	1.04	(1.00, 1.08)	0.035	1.14	(0.94, 1.37)	0.178
False lumen diameter	1.11	(1.03, 1.20)	0.010	1.05	(0.92, 1.19)	0.484
Dissection length	1.00	(1.00, 1.01)	0.053	1.00	(1.00, 1.01)	0.240
Intraoperative details
Timing of operation
Acute	1.00
Subacute	1.19	(0.84, 1.68)	0.324	1.15	(0.80, 1.64)	0.448
Chronic	1.73	(1.15, 2.61)	0.009	1.93	(1.26, 2.94)	0.002
Branch	1.14	(0.79, 1.64)	0.495
Adjunct	0.98	(0.68, 1.42)	0.932
Hybrid	1.17	(0.37, 3.65)	0.793
Discharge medications
Antiplatelet drugs	0.86	(0.63, 1.17)	0.329
ACEI/ARB	1.08	(0.82, 1.44)	0.574
β-blocker	0.75	(0.57, 1.00)	0.049	0.86	(0.60, 1.23)	0.412
CCB	0.76	(0.58, 1.01)	0.055	0.98	(0.69, 1.41)	0.928
Statins	1.39	(0.85, 2.29)	0.191
Anticoagulants	1.16	(0.37, 3.62)	0.803

Variables with a P value < 0.1 in univariable analysis were entered in the multivariable models. ARAEs, aortic-related adverse events; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; WBC, white blood cell; FDP, fibrinogen degradation products; AIP, atherogenic index of plasma; ACEI, angiotensin converting enzyme inhibitors; ARB, angiotensin receptor blocker; CCB, calcium channel blockers; HR, hazard ratio; CI, confidence interval

Table 5.

Univariate and multivariate Cox proportional hazards model for 1-year and 5-year ARAEs

Variable	1-year ARAEs				5-year ARAEs
	Univariate analysis		Multivariate analysis		Univariate analysis		Multivariate analysis
	HR and 95% CI	P-value	HR and 95% CI	P-value	HR and 95% CI	P-value	HR and 95% CI	P-value
AIP	2.32 (1.19, 4.55)	0.014	2.28 (1.15, 4.55)	0.019	1.77 (1.01, 3.10)	0.047	1.78 (0.95, 3.32)	0.072
NLR	1.03 (1.01,1.05)	0.001	1.02 (1.00, 1.05)	0.065	1.03 (1.01,1.04)	0.004	0.93 (0.83, 1.04)	0.187
LMR	0.85 (0.74,0.99)	0.039	0.97 (0.86, 1.08)	0.572	0.95 (0.86,1.04)	0.270
SII	1.00 (1.00,1.00)	0.767			1.00 (1.00,1.00)	0.945
AISI	1.00 (1.00,1.00)	0.782			1.00 (1.00,1.00)	0.694
PLR	1.00 (1.00,1.00)	0.440			1.00 (1.00,1.00)	0.964
SIRI	1.03 (1.01,1.06)	0.007	1.03 (0.99, 1.07)	0.186	1.02 (1.00,1.05)	0.064	0.92 (0.82, 1.04)	0.176
UHR	1.00 (0.99,1.01)	0.870			1.00 (0.99,1.01)	0.870

Variables with a P value < 0.1 in univariable analysis were entered in the multivariable models. AIP, atherogenic index of plasma; NLR, neutrophil-to-lymphocyte ratio; LMR, lymphocyte-to-monocyte ratio; SII, systemic immune-inflammation index; AISI, advanced inflammation-based score index; PLR, platelet-to-lymphocyte ratio; SIRI, systemic inflammatory response index; UHR, uric acid to high-density lipoprotein cholesterol ratio; HR, hazard ratio; CI, confidence interval; ARAEs, aortic-related adverse events

RCS analyses were conducted to examine the nonlinear relationship between AIP levels and ARAEs, as illustrated in Fig. 3. A linear association between AIP and ARAEs was identified at the 1-year mark. This linear relationship persists consistently over a 5-year period. Subgroup analyses showed that the timing of operation demonstrated a statistically significant interaction with 1-year ARAEs (P-interaction = 0.008). AIP demonstrated a positive correlation with 1-year ARAE in patients with acute and subacute TBAD, whereas it exhibited a negative correlation in those with chronic TBAD (Fig. 4). We stratified patients by TBAD phases and used Cox regression analysis, finding that high AIP levels increased the 1-year ARAE risk in acute and subacute phases (Table S3). Additionally, no interaction was observed between AIP and subgroups for the AIP-ARAE association over 5-years post-TEVAR (all P-interaction > 0.05) (Fig. 5).

Fig. 3.

Restricted cubic splines for the relationship between AIP and the incidence of ARAEs in TBAD patients after TEVAR. A The relationship between AIP and 1-year ARAEs risk, B The relationship between AIP and 5-year ARAEs risk. RCS, restricted cubic spline; AIP, atherogenic index of plasma; ARAEs, aortic-related adverse events; TBAD, type B aortic dissection; TEVAR, thoracic endovascular aortic repair; HR, hazard ratio; CI, confidence interval

Fig. 4.

Subgroup analysis for association between AIP and 1-year ARAEs in TBAD patients after TEVAR. AIP, atherogenic index of plasma; ARAEs, aortic-related adverse events; SBP, systolic blood pressure; DBP, diastolic blood pressure; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; HR, hazards ratio; Cl, confidence interval

Fig. 5.

Subgroup analysis for association between AIP and 5-year ARAEs in TBAD patients after TEVAR. AIP, atherogenic index of plasma; ARAEs, aortic-related adverse events; SBP, systolic blood pressure; DBP, diastolic blood pressure; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; HR, hazards ratio; Cl, confidence interval

A time-dependent ROC analysis was performed to evaluate the predictive efficacy of AIP for ARAEs over a 5-year follow-up period (Fig. S2). The AUC values for AIP in predicting ARAEs remained consistently high, ranging from 0.76 to 0.87 throughout the follow-up period (Table S4). The peak AUC of 0.87 was recorded at the 1-year follow-up, with a gradual decline observed over time.

Discussion

Atherosclerosis is the key underlying cause of cardiovascular diseases, primarily driven by lipid metabolism disorders [27–30]. AIP, a novel metric for assessing lipid metabolism, offers the advantage of integrating the dual atherosclerotic impacts of elevated triglycerides and reduced high-density lipoprotein cholesterol [31]. In recent years, AIP has emerged as a focal point of research in the domain of cardiovascular diseases. Prior studies have established that elevated AIP is significantly associated with the risk of adverse cardiovascular events [17–19]. However, the clinical significance of AIP in aortic dissection remains inadequately understood. To our knowledge, this is the first research to investigate the association between AIP and outcomes in TBAD patients receiving TEVAR. Our study identified AIP as an independent risk factor for adverse events following TEVAR in patients with TBAD. Specifically, patients exhibiting higher AIP levels demonstrated an elevated risk of ARAEs. In patients with TBAD and a high AIP, enhanced aortic imaging surveillance combined with lipid-lowering treatment is recommended to enable early identification of ARAEs, thus improving long-term outcomes.

The correlation between AIP and adverse prognosis in individuals with cardiovascular diseases has been extensively researched. A study utilizing data from the CHARLS [17] demonstrated that inadequate control of AIP levels correlates with an elevated risk of cardiovascular events among individuals with cardiovascular kidney-metabolic syndrome. Ramin et al. [32] conducted a systematic review and meta-analysis, confirming that elevated AIP is strongly linked to increased coronary artery disease risk, increased severity of lesions and poor prognosis. The findings of this study align with previous research on AIP in various cardiovascular diseases, while also extending the investigation to the population having aortic dissection. Regarding the association between AIP and long-term ARAEs in patients with TBAD observed in this study, we proposed several possible mechanisms to elucidate the correlation between AIP and AD: Firstly, AIP is strongly associated with metabolic risk factors like hypertension, hyperglycemia, dyslipidemia, and obesity [33, 34], which are linked to TBAD susceptibility and poor post-TEVAR outcomes [35]. Elevated AIP is independently associated with obesity [33]. This link is especially strong in patients with metabolic syndrome, where AIP reflects the combined effects of high triglycerides and low HDL-C, exacerbated by visceral fat. Moreover, AIP is strongly linked to hypertension and type 2 diabetes in large national surveys, and these conditions are known risk factors for aortic dissection [35]. Hypertension, hyperglycemia, and dyslipidemia collectively weaken the aortic wall. In TBAD patients, this cluster of conditions, worsened by high AIP, creates a persistent high-risk state even after TEVAR, leading to complications like false lumen expansion and stent-graft migration. Secondly, AIP directly reflects the severity of atherosclerosis, a key driver of aortic wall remodeling and TBAD progression. Elevated AIP leads to the buildup of atherogenic lipids, such as sd-LDL and remnant lipoprotein cholesterol, in the aortic intima-media layer [11]. These lipids oxidize easily and are absorbed by macrophages, transforming them into foam cells, which are the initial cellular components of atherosclerotic plaques. A recent study of aortic wall samples from TBAD patients found that higher AIP levels are associated with more foam cell infiltration and fewer smooth muscle cells in the aortic media [36]. Elevated AIP also induces ox-LDL and proinflammatory cytokines, which in turn trigger smooth muscle cell apoptosis, leading to elastic fiber degradation and increased collagen deposition, thereby weakening the aortic wall structure [37]. These changes disrupt aortic wall stability, making it more fragile and less compliant, which are critical factors for TBAD development and post-TEVAR ARAEs. Moreover, elevated AIP may lead to atherosclerotic changes in endoleak. Pathological studies have confirmed that atherosclerotic aortic wall elasticity is reduced, elastic fiber rupture is present and collagen deposition is increased, which will impair the seal between the stent graft and the aortic wall, especially in the proximal anchoring zone. Imaging studies have further confirmed that focal atherosclerotic plaque or calcification in the landing zone is an independent risk factor for type IA endoleak [38, 39], as irregular vessel walls can hinder complete stent graft apposition. For type IB endoleaks, AIP-associated medial weakness [18] may exacerbate distal aortic wall compliance mismatch [40], increasing the risk of stent graft-induced new entry point and persistent false lumen perfusion [5]. In addition, elevated AIP is associated with increased systemic inflammation [41, 42], which exacerbates aortic medial necrosis and elastic fiber degeneration. Retrograde dissection after TEVAR is closely associated with proximal aortic wall fragility [43–45]; Imaging studies have shown that AIP is associated with aortic calcification [46] and stent thrombosis [47]. During TEVAR, the radial force generated by the stent graft may further destroy the already fragile middle layer in patients with high AIP value, thereby triggering retrograde extension of the dissecting valve [5]. Finally, atherosclerotic plaques in the aortic arch can cause local stress concentration during stent graft implantation, increasing the risk of retrograde tear. These pathways suggest that AIP does not directly induce complications after TEVAR, but rather serves as a surrogate indicator for underlying aortic wall lesions that compromise stent-graft compatibility and vascular integrity. Future studies integrating preoperative angiography with pathological analysis of aortic wall specimens could further validate these mechanisms.

In this study, AIP was categorized into tertiles (Low: < 0.27, Middle: 0.27–0.52, High: ≥ 0.52). These cutpoints aligned with established metabolic syndrome-related thresholds, where an AIP > 0.21 is widely recognized as a marker of abnormal lipid metabolism and elevated cardiovascular risk [31]. Our Low AIP group (< 0.27) is slightly higher than this established threshold but still falls within the range generally considered low-risk for metabolic-related cardiovascular events, highlighting the similarities between our stratification and previous studies. Our stratification further refines this risk gradient: the middle tertile (0.27–0.52) captures a moderate-risk profile, while the high-risk group (≥ 0.52) identifies individuals with the most pronounced risk of ARAEs. Due to the lack of universal AIP thresholds for TBAD, using tertile grouping based on our study's distribution is a reliable alternative. Our high-risk threshold (≥ 0.52) aligns with the established high metabolic risk threshold (≥ 0.50) observed in the general population, thereby maintaining clinical relevance [48]. This threshold is corroborated by a comprehensive study that demonstrated that individuals in the highest AIP quartile (≥ 0.50) exhibited the greatest risk of cardiovascular events [18]. Our findings indicate that a threshold of 0.52 signifies a heightened risk for ARAEs, marked by advanced lipid metabolism issues such as increased small, dense LDL particles [11] and decreased HDL function [37]. Using this cutpoint enhances risk assessment, identifying patients who might benefit from targeted lifestyle changes and closer cardiometabolic monitoring.

Subgroup analysis showed that patients undergoing subacute surgery have the highest risk of AIP-related ARAEs, due to progressive aortic wall destruction, while those with chronic surgery have the lowest risk, as chronic lesions are more stable and may have partial thrombosis of the false lumen. These findings are preliminary and require further validation in larger studies to confirm their clinical relevance.

Based on the research findings, this study holds significant clinical implications in two primary areas. Firstly, our study addresses a crucial gap in TBAD prognostic systems by adding a metabolic perspective. Traditional models for post-TEVAR ARAEs focus on anatomy, disease stage, or procedures, often missing the long-term effects of lipid metabolism on aortic wall homeostasis and lesion progression. We found that AIP is a simple, reliable marker that can predict 1-year and 5-year ARAEs, underscoring the importance of metabolic factors in long-term TBAD prognosis. By incorporating AIP, existing models can improve accuracy and better identify high-risk patients who might be overlooked by conventional methods. Secondly, AIP is ideal for routine clinical use and large-scale risk assessment due to its low cost, ease of measurement, and availability. Unlike advanced imaging or complex biomarkers, AIP comes from standard lipid panels already used in preoperative and postoperative evaluations. For TBAD patients with high AIP, clinicians can improve outcomes by increasing aortic imaging frequency and initiating personalized lipid-lowering treatments to mitigate long-term complications.

Limitations

First, the single-center retrospective design may introduce selection bias due to varying patient characteristics, referral patterns, perioperative management, and operator expertise across institutions, which can influence preoperative AIP levels and postoperative outcomes. Furthermore, our study's reliance on data-driven confounding adjustments rather than a predefined predictor set for ARAEs, mortality, and MACE could result in residual confounding bias. Caution is advised in generalizing these findings, and future multicenter, large-sample prospective studies are necessary to validate our results and clarify the relationship between AIP and patient outcomes. Secondly, our study is its reliance on baseline AIP measurement, without ongoing assessments. Since AIP can change with lipid metabolism and lifestyle, one measurement doesn't reflect its dynamic nature. Future studies should incorporate time-updated medication data and time-dependent covariate analysis to better understand AIP's impact on TBAD prognosis. Thirdly, the relationship between AIP and other AD biomarkers is unclear, making it uncertain whether AIP influences prognosis through different pathways or synergistically. Our subgroup analysis of chronic TBAD may be affected by survivor bias, as patients reaching this stage without intervention might have milder conditions or better initial characteristics, with potential residual confounding.

Conclusion

Our research indicates that AIP is independently associated with 1-year and 5-year ARAEs in patients with TBAD following TEVAR, providing a novel metabolic perspective for the prognostic evaluation of this population over an extended period.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AIP

Atherogenic index of plasma

AD

Aortic dissection

TBAD

Type B aortic dissection

TEVAR

Thoracic endovascular aortic repair

ARAEs

Aortic-related adverse events

K–M

Kaplan–Meier curve

RCS

Restricted cubic spline

ROC

Receiver operating characteristic

MACCEs

Major adverse cardiovascular and cerebrovascular events

TC

Total cholesterol

TG

Triglyceride

HDL-C

High-density lipoprotein cholesterol

LDL-C

Low-density lipoprotein cholesterol

NLR

Neutrophil-to-lymphocyte ratio

LMR

Lymphocyte-to-monocyte ratio

SII

Systemic immune-inflammation index

AISI

Advanced inflammation-based score index

PLR

Platelet-to-lymphocyte ratio

SIRI

Systemic inflammatory response index

UHR

Uric acid to high-density lipoprotein cholesterol ratio

CTA

Computed tomography angiography

COPD

Chronic obstructive pulmonary disease

WBC

White blood cells

CHARLS

China Health and Retirement Longitudinal Survey

sdLDL

Small and dense low-density lipoprotein

ox-VLDL

Oxidized modified very low-density lipoprotein

Author contributions

Shuangshuang Li, Wen Li, Jiahe Zhang and Kaiwen Zhao contributed equally to this research. Shuangshuang Li was involved in research design, data collection, data analysis, and manuscript preparation. Wen Li and Jiahe Zhang contributed to data collection, data analysis, and manuscript writing. Kaiwen Zhao and Zhichen Ding assisted in revising the manuscript. Jianli Ren, Wenping Hu, Qingsheng Lu and Jian Zhou participated in the research design. All authors reviewed and approved the final manuscript.

Funding

The study was funded by the National Natural Science Foundation of China [82570560], Scientific Research Cultivation Project of the Third Affiliated Hospital of Naval Medical University (2025QN04), Clinical Research Special Scientific Research Project funded by the Science and Technology and Economy Commission of Yangpu District (YPM202545), and the Tengfei Project Talent Program of the Third Affiliated Hospital of Naval Medical University.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The requirement for consent was waived, and the Clinical Research Ethics Committee of Changhai Hospital, Naval Medical University, approved the research protocol. The protocol number is CHEC-Y2020-042, and the ratification date is August 21, 2020.

Competing interests

The authors declare no competing interests.