Anxiety-depression after intravascular ultrasound-guided drug-coated balloon angioplasty dissection: Predictors and 12-month cardiovascular outcomes

Abstract

BACKGROUND

Coronary dissection after balloon dilation is one of the key complications identified in coronary angioplasty, especially during drug-coated balloon (DCB) therapy in patients with coronary artery disease. Intimal dissection, albeit conservatively managed under intravascular ultrasound (IVUS) guidance, probably results in important psychiatric distress possibly impeding long-term cardiovascular prognosis. However, the summary relationship between anxiety-depression states and clinical outcomes in this particular disease population is largely unknown.

AIM

To characterize post-IVUS DCB-associated dissection presence and depression incidence, investigate their relationship with major adverse cardiovascular events (MACE), and define a multidisciplinary predictive model for early psychological risk factor stratification.

METHODS

A retrospective, single-center cohort study which consecutively enrolled 168 patients with angiographically confirmed coronary dissection after IVUS-guided DCB angioplasty was performed at the Department of Cardiology from April 2020 to September 2024. Psychological status was assessed using the Hospital Anxiety and Depression Scale (HADS, pre-procedure, 1 month and 6 months post-procedure), the Generalized Anxiety Disorder-7 questionnaire and the Patient Health Questionnaire-9. Patients (n = 168) were assigned to an anxiety-depression cohort (high score defined as HADS-A ≥ 8 or HADS-D ≥ 8 at the 1-month assessment) or a stable psychologically stable cohort, the latter included low-level anxiety/depression patients. Analyses included between-group comparisons regarding demographic profiles, procedural parameters, dissection characteristics by IVUS, inflammatory and cardiac biomarkers, and 12-month MACE rates. Multivariate logistic regression was employed to determine independent predictors and a composite predictive model was derived using receiver operating characteristic analysis.

RESULTS

At the 1-month follow-up, 65 of 168 (38.7%) patients demonstrated clinically significant anxiety-depression. Compared with the control group, the anxiety-depression cohort had a significantly higher 12-month MACE rate (27.7% vs 8.9%, P = 0.003), including increased rates of target lesion revascularization (15.4% vs 4.9%, P = 0.018), repeat myocardial infarction (7.7% vs 1.9%, P = 0.048) and unplanned cardiovascular rehospitalization (20.0% vs 7.8%, P = 0.016). Seventeen (44%) patients experienced anxiety-depression following the procedure; compared to patients without anxiety-depression. Multivariate logistic regression revealed seven independent predictors of post-procedural anxiety-depression: History of prior anxiety-depression [odds ratio (OR) = 3.45, 95% confidence interval (CI): 1.92-6.20, P < 0.001], National Heart, Lung, and Blood Institute type C dissection (OR = 3.12, 95%CI: 1.78-5.47 P < 0.001), elevated high-sensitivity C-reactive protein > 10 mg/L (OR = 2.68, 95%CI: 1.52-4.72, P = 0.001), female sex (OR = 2.35, 95%CI: 1.40-3.95, P = 0.001), residual stenosis > 30% post-DCB (OR = 2.21, 95%CI: 1.28-3.82, P = 0.004), low social support (OR = 2.05, 95%CI: 1.18-3.56, P = 0.011), and procedure duration > 90 minutes (OR = 1.88, 95%CI: 1.08-3.27, P = 0.026). The composite model yielded an area under the curve of 0.86 (95%CI: 0.80-0.92), with sensitivity of 78.5% and specificity of 82.1%.

CONCLUSION

The interpretable predictive model integrating procedural, inflammatory, and psychosocial predictors provides high discrimination for predicting vulnerable patients within 90 days of hospital discharge. These results provide a basis for the implementation of standardized psychological evaluation as part of post-interventional care pathways among patients with coronary dissection.

Key Words: Drug-coated balloon; Intravascular ultrasound; Coronary dissection; Anxiety; Depression; Major adverse cardiovascular events; Predictive model

Core Tip: Clinically significant anxiety-depression developed in 38.7% of patients following coronary dissection after intravascular ultrasound-guided drug-coated balloon angioplasty, and was independently associated with a nearly threefold higher 12-month major adverse cardiovascular event rate. Seven independent predictors were identified, and a composite model achieved an area under the curve of 0.86, supporting routine psychocardiological screening in this population.

INTRODUCTION

Coronary artery disease remains the single largest cause of cardiovascular morbidity and mortality worldwide, with percutaneous coronary intervention (PCI) being the most commonly used revascularization method for symptomatic obstructive lesions. Over the past decade, drug-coated balloon (DCB) technology has been widely investigated as an alternative to conventional drug-eluting stent implantation for in-stent restenosis, small-vessel disease and bifurcation lesions, in which permanent metallic scaffolding is undesirable[1-4]. DCB therapy delivers antiproliferative agents locally to the vessel wall with only brief balloon inflation and may, in theory, preserve vasomotion while avoiding the late stent thrombosis associated with metallic stent platforms.

Intravascular ultrasound (IVUS) has become an indispensable adjunctive imaging modality that substantially improves procedural accuracy during DCB angioplasty. Cross-sectional tomographic visualization by IVUS permits detailed assessment of plaque composition, vessel dimensions and post-angioplasty dissection characteristics that are not discernible on conventional angiography[5-7]. Coronary dissection remains an almost unavoidable sequela of balloon angioplasty, occurring in 20%-45% of patients treated with DCB[8,9]. Although most dissections are hemodynamically inconsequential and suitable for conservative, non-interventional management once confirmed by IVUS, the psychological burden may be considerable for patients who have been informed that a vessel injury requires systematic monitoring.

A growing body of evidence in the wider cardiovascular literature indicates that 25%-50% of patients develop anxiety or depressive disorders after acute coronary events and interventional procedures, with adverse effects on medication adherence, participation in cardiac rehabilitation, endothelial function and autonomic regulation[10-13]. The pathophysiological link between psychological distress and poor cardiovascular outcomes involves hypothalamic-pituitary-adrenal axis overactivation, sympathoadrenal hyperactivity, platelet activation and hyperreactivity, systemic inflammation and endothelial dysfunction[14-16]. These mechanisms may be of particular importance in patients with post-procedural dissection, in whom vascular healing depends on a balanced hemodynamic and inflammatory milieu.

Despite the recognized cardiovascular impact of emotional stress, the prevalence, predictors and prognostic significance of anxiety-depression in patients with dissection after IVUS-guided DCB angioplasty have not been systematically investigated. Previous studies have focused mainly on psychological outcomes after PCI or stent implantation, leaving a considerable knowledge gap regarding the DCB-dissection population[17-19].

Accordingly, the present study aimed to: (1) Evaluate the prevalence and longitudinal course of anxiety-depression in patients with coronary dissection after IVUS-guided DCB therapy; (2) Examine the association between anxiety-depression status and 12-month major adverse cardiovascular events (MACE); (3) Identify independent predictors using multivariate modeling; and (4) Develop and evaluate a composite predictive model for early psychological risk stratification. The study is intended to provide an evidence base for integrating psychocardiological surveillance into the post-interventional management of DCB-treated patients with dissection.

MATERIALS AND METHODS

Study design and patient selection

This retrospective cohort study was conducted in accordance with the Declaration of Helsinki and was reviewed and approved by the Ethics Committee of The Third Affiliated Hospital of Wenzhou Medical University (approval No. YJ2025145; approved on November 6, 2025). All data analyzed in the present study were generated during routine clinical care from April 2020 to September 2024 and were retrospectively extracted from the hospital medical records system only after ethics approval had been obtained. As the analysis was restricted to anonymized data from existing records and involved no additional intervention, no additional risk and no contact with patients, the Ethics Committee waived the requirement for written informed consent.

Eligibility criteria were: (1) Age 18-80 years; (2) Elective or urgent DCB angioplasty with pre- and post-procedural IVUS assessment; (3) Angiographically and/or IVUS-confirmed coronary dissection of any National Heart, Lung, and Blood Institute (NHLBI) classification after DCB inflation; (4) Availability of a neuropsychological assessment recorded at 1-month post-procedure; and (5) Complete clinical, procedural, laboratory and follow-up documentation. Exclusion criteria were: (1) Emergent bail-out stenting for flow-limiting dissection; (2) Cardiogenic shock or hemodynamic instability during the index procedure; (3) A documented diagnosis of major depressive disorder, generalized anxiety disorder, bipolar affective disorder or psychotic illness requiring ongoing pharmacotherapy at the time of the index procedure; (4) Significant cognitive impairment (Mini-Mental State Examination < 20); (5) Concomitant terminal malignancy with an anticipated survival < 12 months; (6) Administration of psychotropic agents within the 4 weeks preceding the procedure; and (7) Active substance use disorder.

Criterion (3) excluded only patients with an active psychiatric diagnosis requiring ongoing pharmacological treatment; patients with a remote or untreated history of anxiety or depressive symptoms remained eligible, and this variable was retained for analysis as “psychiatric history”.

From an initial cohort of 218 screened patients, 50 were excluded (15 for bail-out stenting, 11 for pre-existing psychiatric conditions, 8 for incomplete follow-up data, 6 for cognitive impairment, 4 for concurrent terminal illness, 3 for recent psychotropic use, 2 for substance abuse and 1 for hemodynamic instability), yielding 168 patients for the definitive analysis.

Psychological evaluation and stratification

Standardized psychological assessment was performed as part of routine clinical care in our department at three predefined time points: Within 48 hours before the index procedure, at 1 month (± 5 days) and at 6 months (± 10 days) after the procedure. The present study retrospectively extracted these existing records.

The Hospital Anxiety and Depression Scale (HADS)[20], a 14-item self-administered instrument comprising discrete anxiety (HADS-A, 0-21) and depression (HADS-D, 0-21) subscales, served as the primary psychological measure. Supplementary assessments included the Generalized Anxiety Disorder-7[21] for anxiety severity and the Patient Health Questionnaire-9[22] for depressive symptom quantification. These supplementary instruments were used for confirmatory and descriptive purposes only and were not incorporated into the stratification algorithm or the predictive model.

On the basis of the 1-month HADS assessment, patients were dichotomized into an anxiety-depression cohort (HADS-A ≥ 8 or HADS-D ≥ 8; n = 65) and a psychologically stable cohort (HADS-A < 8 and HADS-D < 8; n = 103). Severity within the anxiety-depression cohort was graded as mild (HADS subscale 8-10; n = 28, 43.1%), moderate (11-14; n = 24, 36.9%) or severe (15-21; n = 13, 20.0%). Patients were defined as having de novo post-procedural anxiety-depression if they met the 1-month HADS threshold without having met it at the pre-procedural assessment.

Procedural and IVUS assessment

All DCB procedures were performed via radial or femoral access using standard interventional techniques. Pre-dilatation with a semi-compliant balloon was performed at the operator’s discretion. DCB inflation was performed at nominal pressure for 60 seconds. IVUS investigation (40-60 MHz mechanical or phased-array transducer) was mandatory before and after DCB inflation. Dissection was classified according to NHLBI criteria (types A-F) on the basis of angiographic appearance and further characterized by IVUS with respect to depth (intimal, medial, adventitial), longitudinal extent, circumferential arc and the presence of intramural hematoma. Residual stenosis was quantified by quantitative coronary angiography.

Conservative management of dissection (without bail-out stenting) was elected when IVUS confirmed adequate distal flow (Thrombolysis in Myocardial Infarction grade 3), dissection limited to the intimal-medial plane, absence of a flow-limiting flap and a residual lumen area > 3.5 mm².

Laboratory biomarkers

Peripheral venous specimens were obtained before the procedure and at 24 hours and 14 days after the procedure (07:00-08:00, fasting). Assayed parameters included inflammatory markers [high-sensitivity C-reactive protein (hs-CRP), interleukin-6, tumor necrosis factor-α and fibrinogen]; cardiac biomarkers (high-sensitivity troponin I, N-terminal pro-B-type natriuretic peptide and creatine kinase-MB); neuroendocrine indices (morning serum cortisol and adrenocorticotropic hormone); a metabolic panel (fasting glucose, glycated hemoglobin, lipid profile and homocysteine); and routine hematology including a complete blood count with calculation of the neutrophil-to-lymphocyte ratio.

Clinical endpoints and follow-up

The primary clinical endpoint was 12-month MACE, defined as the composite of cardiac death, non-fatal myocardial infarction and target lesion revascularization. Secondary endpoints included the individual MACE components, unplanned cardiovascular rehospitalization and all-cause mortality.

Follow-up was conducted through outpatient visits at 1, 3, 6 and 12 months, supplemented by telephone contact, all of which formed part of the department’s standard post-interventional follow-up program. Coronary angiographic follow-up was performed at 9-12 months or earlier if clinically indicated.

Statistical analysis

Analyses were performed using SPSS 26.0 and R 4.2.1. Continuous data with a normal distribution are summarized as the mean ± SD and were compared using the independent-samples t-test; skewed distributions are reported as the median (interquartile range) and were evaluated using the Mann-Whitney U test. Categorical variables are presented as n (%) and were compared using the χ² test or Fisher’s exact test. Variables with P < 0.10 on univariate screening were entered into multivariate logistic regression using the enter method (all variables entering simultaneously). Kaplan-Meier survival analysis with log-rank comparison was used for time-to-MACE evaluation. Spearman correlation assessed the relationship between psychological scores and MACE incidence. Model discrimination was appraised by receiver operating characteristic analysis with area under the curve (AUC) computation, and calibration was evaluated using the Hosmer-Lemeshow statistic. A two-sided P < 0.05 was considered statistically significant.

RESULTS

Baseline demographic and clinical profiles

Of the 168 patients analyzed, 65 (38.7%) met the criteria for clinically significant anxiety-depression at the 1-month assessment. Table 1 describes the baseline characteristics. Compared with the psychologically stable cohort, patients in the anxiety-depression cohort were more likely to be female (52.3% vs 33.0%, P = 0.013), to have a psychiatric history (18.5% vs 3.9%, P = 0.002), to have lower social support as measured by the Social Support Rating Scale (30.5 ± 8.2 vs 41.2 ± 7.8, P < 0.001) and to have higher pre-procedural anxiety and depression scores. The cohorts did not differ significantly in age, body mass index, or the prevalence of hypertension, diabetes or hyperlipidemia (Table 1).

Table 1 Baseline demographic and clinical characteristics, n (%)/mean ± SD.

Characteristic	Anxiety-depression cohort (n = 65)	Stable cohort (n = 103)	P value
Age (years)	62.8 ± 10.5	64.5 ± 11.2	0.308
Female	34 (52.3)	34 (33.0)	0.013
BMI (kg/m2)	25.2 ± 3.1	24.8 ± 3.4	0.425
Current smoker	22 (33.8)	38 (36.9)	0.683
Hypertension	38 (58.5)	55 (53.4)	0.520
Diabetes mellitus	18 (27.7)	22 (21.4)	0.343
Hyperlipidemia	28 (43.1)	40 (38.8)	0.581
Prior MI	10 (15.4)	12 (11.7)	0.482
Prior PCI	15 (23.1)	20 (19.4)	0.566
Psychiatric history	12 (18.5)	4 (3.9)	0.002
Family psychiatric history	6 (9.2)	3 (2.9)	0.081
SSRS score	30.5 ± 8.2	41.2 ± 7.8	< 0.001
Pre-procedural HADS-A	8.2 ± 3.5	4.8 ± 2.2	< 0.001
Pre-procedural HADS-D	7.5 ± 3.2	4.2 ± 1.9	< 0.001
Pre-procedural GAD-7	9.0 ± 3.8	5.2 ± 2.5	< 0.001
Pre-procedural PHQ-9	8.5 ± 3.5	4.8 ± 2.2	< 0.001
LVEF (%)	55.2 ± 8.5	57.8 ± 7.2	0.105

BMI: Body mass index; MI: Myocardial infarction; PCI: Percutaneous coronary intervention; SSRS: Social Support Rating Scale; HADS: Hospital Anxiety and Depression Scale; GAD-7: Generalized Anxiety Disorder-7; PHQ-9: Patient Health Questionnaire-9; LVEF: Left ventricular ejection fraction.

Procedural and dissection characteristics

Patients in the anxiety-depression cohort had more severe dissection (NHLBI type C or greater: 35.4% vs 15.5%, P = 0.003), greater dissection length on IVUS (12.8 ± 4.5 mm vs 8.2 ± 3.2 mm, P < 0.001), a larger circumferential arc (125° ± 38° vs 85° ± 28°, P < 0.001) and higher residual stenosis (32.5% ± 10.8% vs 22.8% ± 8.5%, P < 0.001). Procedure duration was also substantially longer (95.2 ± 28.5 minutes vs 72.5 ± 22.0 minutes, P < 0.001). The distribution of target vessels, the number of treated lesions and DCB specifications did not differ significantly between groups (Table 2).

Table 2 Procedural and dissection characteristics, n (%)/mean ± SD.

Parameter	Anxiety-depression cohort (n = 65)	Stable cohort (n = 103)	P value
Target vessel			0.425
LAD	28 (43.1)	42 (40.8)
LCx	15 (23.1)	25 (24.3)
RCA	22 (33.8)	36 (35.0)
Multi-vessel disease	25 (38.5)	32 (31.1)	0.318
Lesion length (mm)	22.5 ± 8.2	20.8 ± 7.5	0.172
Reference diameter (mm)	2.85 ± 0.52	2.92 ± 0.48	0.362
DCB diameter (mm)	2.88 ± 0.50	2.95 ± 0.45	0.345
Pre-dilatation	52 (80.0)	78 (75.7)	0.514
Procedure time (minutes)	95.2 ± 28.5	72.5 ± 22.0	< 0.001
Contrast volume (mL)	185 ± 52	158 ± 42	0.012
NHLBI dissection type			0.003
Type A	18 (27.7)	45 (43.7)
Type B	24 (36.9)	42 (40.8)
Type C	16 (24.6)	12 (11.7)
Type D	5 (7.7)	3 (2.9)
Type E	2 (3.1)	1 (1.0)
Type F	0	0
Dissection length on IVUS (mm)	12.8 ± 4.5	8.2 ± 3.2	< 0.001
Circumferential arc (°)	125 ± 38	85 ± 28	< 0.001
Medial involvement	18 (27.7)	10 (9.7)	0.002
Intramural hematoma	12 (18.5)	8 (7.8)	0.035
Residual stenosis (%)	32.5 ± 10.8	22.8 ± 8.5	< 0.001
Residual MLA (mm2)	4.8 ± 1.2	5.5 ± 1.0	0.002

LAD: Left anterior descending artery; LCx: Left circumflex artery; RCA: Right coronary artery; DCB: Drug-coated balloon; NHLBI: National Heart, Lung, and Blood Institute; IVUS: Intravascular ultrasound; MLA: Minimum lumen area.

Clinical outcomes at 12 months

Clinical outcomes at 12 months differed markedly between groups (Table 3). Anxiety-depression was associated with a significantly higher rate of composite MACE (27.7% vs 8.7%, P = 0.003). Target lesion revascularization (15.4% vs 4.9%, P = 0.018), non-fatal myocardial infarction (7.7% vs 1.9%, P = 0.048) and cardiac death (4.6% vs 1.9%, P = 0.247) all occurred more frequently in the anxiety-depression cohort than in the psychologically stable cohort. The risk of unplanned cardiovascular rehospitalization was almost threefold higher in the anxiety-depression cohort (20.0% vs 7.8%, P = 0.016). Kaplan-Meier analysis showed significantly worse event-free survival in the anxiety-depression cohort (log-rank P = 0.001; Figure 1). Spearman analysis demonstrated positive correlations between the 1-month HADS-A score and 12-month MACE (r = 0.38, P < 0.001) and between the HADS-D score and MACE (r = 0.35, P < 0.001).

Figure 1 Kaplan-Meier curves for freedom from major adverse cardiovascular events over 12 months in the anxiety-depression cohort and the psychologically stable cohort (log-rank P = 0.001). MACE: Major adverse cardiovascular event; AD: Anxiety-depression.

Table 3 Twelve-month clinical outcomes, n (%).

Endpoint	Anxiety-depression cohort (n = 65)	Stable cohort (n = 103)	P value
Composite MACE	18 (27.7)	9 (8.7)	0.002
Cardiac death	3 (4.6)	2 (1.9)	0.247
Non-fatal MI	5 (7.7)	2 (1.9)	0.048
Target lesion revascularization	10 (15.4)	5 (4.9)	0.018
Target vessel revascularization	12 (18.5)	7 (6.8)	0.022
Unplanned CV rehospitalization	13 (20.0)	8 (7.8)	0.016
All-cause mortality	4 (6.2)	3 (2.9)	0.272
Stroke/TIA	2 (3.1)	1 (1.0)	0.340
New-onset heart failure	5 (7.7)	3 (2.9)	0.141
Recurrent angina (CCS ≥ II)	18 (27.7)	12 (11.7)	0.006

MACE: Major adverse cardiovascular events; MI: Myocardial infarction; CV: Cardiovascular; TIA: Transient ischemic attack; CCS: Canadian Cardiovascular Society.

Laboratory biomarker profiles

Biomarker assessment on day 14 after the procedure showed marked separation between the groups (Table 4). The anxiety-depression cohort demonstrated consistently increased systemic inflammation: Hs-CRP (12.8 ± 4.5 mg/L vs 6.2 ± 2.8 mg/L, P < 0.001) and interleukin-6 (9.5 ± 3.2 pg/mL vs 5.2 ± 1.8 pg/mL, P < 0.001).

Table 4 Laboratory biomarkers on post-procedural day 14, mean ± SD.

Biomarker	Anxiety-depression cohort (n = 65)	Stable cohort (n = 103)	P value
hs-CRP (mg/L)	12.8 ± 4.5	6.2 ± 2.8	< 0.001
IL-6 (pg/mL)	9.5 ± 3.2	5.2 ± 1.8	< 0.001
TNF-α (pg/mL)	15.2 ± 4.0	9.8 ± 2.5	< 0.001
NLR	4.8 ± 1.6	2.9 ± 1.1	< 0.001
Fibrinogen (g/L)	4.2 ± 0.8	3.5 ± 0.6	< 0.001
Peak hs-TnI (ng/mL)	0.85 ± 0.32	0.42 ± 0.18	< 0.001
NT-proBNP (pg/mL)	385 ± 125	245 ± 88	< 0.001
Peak CK-MB (U/L)	28.5 ± 10.2	18.2 ± 6.5	< 0.001
Morning cortisol (nmol/L)	418 ± 98	325 ± 72	< 0.001
ACTH (pg/mL)	42.5 ± 9.2	35.2 ± 7.0	< 0.001
Homocysteine (μmol/L)	16.8 ± 5.5	12.2 ± 3.8	< 0.001
HbA1c (%)	6.5 ± 1.2	6.2 ± 1.0	0.082
LDL-C (mmol/L)	2.85 ± 0.82	2.72 ± 0.75	0.385
HDL-C (mmol/L)	1.02 ± 0.25	1.18 ± 0.28	0.012

hs-CRP: High-sensitivity C-reactive protein; IL-6: Interleukin-6; TNF-α: Tumor necrosis factor-α; NLR: Neutrophil-to-lymphocyte ratio; hs-TnI: High-sensitivity troponin I; NT-proBNP: N-terminal pro-B-type natriuretic peptide; CK-MB: Creatine kinase-MB; ACTH: Adrenocorticotropic hormone; HbA1c: Glycated hemoglobin; HDL-C: High-density lipoprotein cholesterol; LDL-C: Low-density lipoprotein cholesterol.

Cardiac biomarkers were also elevated: Peak high-sensitivity troponin I (0.85 ± 0.32 ng/mL vs 0.42 ± 0.18 ng/mL, P < 0.001) and N-terminal pro-B-type natriuretic peptide (385 ± 125 pg/mL vs 245 ± 88 pg/mL, P < 0.001). Diurnal dysregulation of neuroendocrine output was supported by higher morning cortisol concentrations (418 ± 98 nmol/L vs 325 ± 72 nmol/L, P < 0.001). The anxiety-depression cohort also had higher homocysteine concentrations (16.8 ± 5.5 μmol/L vs 12.2 ± 3.8 μmol/L, P < 0.001) and lower high-density lipoprotein cholesterol (1.02 ± 0.25 mmol/L vs 1.18 ± 0.28 mmol/L, P = 0.012).

Multivariate logistic regression

Seven independent predictors of post-procedural anxiety-depression were identified by multivariate logistic regression incorporating all variables reaching univariate significance (P < 0.10) (Table 5 and Figure 2). The strongest predictor was a prior history of anxiety-depression [odds ratio (OR) = 3.45], followed by NHLBI type C dissection (OR = 3.12) and elevated hs-CRP (OR = 2.68). Model calibration was adequate, as confirmed by the Hosmer-Lemeshow goodness-of-fit test (χ² = 6.85, P = 0.553).

Figure 2 Forest plot depicting odds ratios with 95% confidence intervals for all variables entered into the multivariate logistic regression model. Square markers denote statistically significant independent predictors (P < 0.05); diamond markers indicate non-significant variables. The dashed vertical line at an odds ratio of 1.0 represents the null hypothesis of no association. IVUS: Intravascular ultrasound; DCB: Drug-coated balloon; hs-CRP: High-sensitivity C-reactive protein; SSRS: Social Support Rating Scale; OR: Odds ratio; CI: Confidence interval.

Table 5 Independent predictors of post-procedural anxiety-depression on multivariate logistic regression.

Variable	β	SE	Wald	OR (95%CI)	P value
Prior anxiety-depression history	1.238	0.300	17.02	3.45 (1.92-6.20)	< 0.001
Type C dissection	1.138	0.286	15.82	3.12 (1.78-5.47)	< 0.001
hs-CRP > 10 mg/L	0.986	0.290	11.56	2.68 (1.52-4.72)	0.001
Female sex	0.855	0.265	10.40	2.35 (1.40-3.95)	0.001
Residual stenosis > 30%	0.793	0.280	8.02	2.21 (1.28-3.82)	0.004
Low social support (SSRS < 35)	0.718	0.282	6.48	2.05 (1.18-3.56)	0.011
Procedure time > 90 minutes	0.631	0.283	4.97	1.88 (1.08-3.27)	0.026
Candidate variables not retained
Diabetes mellitus	0.482	0.289	2.78	1.62 (0.92-2.85)	0.092
Multi-vessel disease	0.438	0.290	2.28	1.55 (0.88-2.73)	0.128
Age > 65 years	0.351	0.280	1.57	1.42 (0.82-2.46)	0.210
Hypertension	0.300	0.280	1.15	1.35 (0.78-2.34)	0.282
Smoking history	0.247	0.272	0.82	1.28 (0.75-2.18)	0.362

OR: Odds ratio; CI: Confidence interval; hs-CRP: High-sensitivity C-reactive protein; SSRS: Social Support Rating Scale.

Predictive model performance

The seven independent predictors combined into a composite model discriminated post-procedural anxiety-depression with an AUC of 0.86 (95% confidence interval: 0.80-0.92), a sensitivity of 78.5% and a specificity of 82.1% at the Youden-optimized threshold (Figure 3). The AUCs of individual predictors were 0.79 for the pre-procedural HADS score, 0.75 for hs-CRP, 0.71 for dissection severity and 0.65 for female sex (male sex as reference). DeLong testing confirmed that the composite model performed significantly better than each individual predictor (all P < 0.05).

Figure 3 Receiver operating characteristic curves comparing the discriminatory capacity of the composite predictive model and individual predictors for post-procedural anxiety-depression in patients with dissection after intravascular ultrasound-guided drug-coated balloon angioplasty. The composite model attained the highest area under the curve (0.86) and significantly outperformed each individual predictor (DeLong test, all P < 0.05). ROC: Receiver operating characteristic; IVUS: Intravascular ultrasound; DCB: Drug-coated balloon; AUC: Area under the curve; HADS: Hospital Anxiety and Depression Scale; hs-CRP: High-sensitivity C-reactive protein.

Severity-stratified subgroup analysis

Analysis across strata of anxiety-depression severity demonstrated a dose-response relationship for virtually every clinical and biomarker endpoint (Table 6). Patients with severe symptoms (HADS ≥ 15) had the highest 12-month MACE rate (46.2%), the greatest hs-CRP elevation (16.5 ± 5.2 mg/L), the highest troponin concentrations (1.05 ± 0.38 ng/mL), the lowest social support (Social Support Rating Scale: 26.8 ± 6.5) and the poorest quality of life (EuroQol five-dimension visual analogue scale: 48.5 ± 12.0). These results indicate a severity gradient, with increasing psychological distress associated with poorer inflammatory, cardiac and functional parameters.

Table 6 Clinical and biomarker parameters stratified by anxiety-depression severity, mean ± SD.

Parameter	Mild (n = 28)	Moderate (n = 24)	Severe (n = 13)	P for trend
HADS-A score	9.5 ± 1.2	13.0 ± 1.5	17.8 ± 2.0	< 0.001
HADS-D score	9.0 ± 1.0	12.5 ± 1.2	17.2 ± 2.2	< 0.001
12-month MACE (%)	17.9	33.3	46.2	0.005
TLR (%)	10.7	16.7	23.1	0.028
Recurrent angina (%)	17.9	29.2	46.2	0.008
hs-CRP (mg/L)	9.8 ± 3.5	13.5 ± 4.2	16.5 ± 5.2	< 0.001
IL-6 (pg/mL)	7.8 ± 2.5	10.2 ± 3.0	12.8 ± 3.8	< 0.001
Peak hs-TnI (ng/mL)	0.68 ± 0.25	0.88 ± 0.30	1.05 ± 0.38	0.002
NT-proBNP (pg/mL)	325 ± 105	395 ± 120	475 ± 145	0.003
Morning cortisol (nmol/L)	385 ± 85	425 ± 95	465 ± 115	0.005
NLR	4.0 ± 1.2	4.8 ± 1.5	5.8 ± 1.8	0.002
SSRS score	33.2 ± 7.5	29.5 ± 8.0	26.8 ± 6.5	0.008
EQ-5D VAS	65.2 ± 10.5	55.8 ± 11.2	48.5 ± 12.0	< 0.001
Medication adherence (%)	82.5 ± 8.5	72.8 ± 10.2	62.5 ± 12.5	< 0.001
Cardiac rehabilitation participation (%)	71.4	54.2	30.8	0.002

HADS: Hospital Anxiety and Depression Scale; MACE: Major adverse cardiovascular events; TLR: Target lesion revascularization; hs-CRP: High-sensitivity C-reactive protein; IL-6: Interleukin-6; hs-TnI: High-sensitivity troponin I; NT-proBNP: N-terminal pro-B-type natriuretic peptide; NLR: Neutrophil-to-lymphocyte ratio; SSRS: Social Support Rating Scale; EQ-5D VAS: EuroQol five-dimension visual analogue scale.

Sensitivity analysis

To isolate the psychological impact of the dissection event from pre-existing psychological vulnerability, a sensitivity analysis was performed in patients without pre-procedural anxiety-depression (pre-procedural HADS-A < 8 and HADS-D < 8; n = 156). Within this subgroup, 53 patients (34.0%) developed de novo clinically significant anxiety-depression at 1-month follow-up. The association with 12-month MACE remained statistically significant (32.1% vs 10.2%, P = 0.003), and the composite predictive model retained acceptable discrimination (AUC = 0.81, 95% confidence interval: 0.72-0.90). These findings suggest that the association with MACE is not solely attributable to pre-existing psychological burden and that the dissection event itself contributes independently to post-procedural anxiety-depression risk.

DISCUSSION

To our knowledge, the present study represents one of the first systematic evaluations of the prevalence, associated factors and long-term outcomes of anxiety-depression among patients with coronary dissection after IVUS-guided DCB angioplasty. The key findings, that 38.7% of such patients develop clinically significant psychological distress and that this distress is associated with a nearly threefold increase in 12-month MACE, have important implications for contemporary interventional cardiology practice.

The prevalence of anxiety and depression observed here is consistent with estimates for wider post-PCI populations (25%-50%)[10-12] but may also reflect the additional psychological burden specific to post-DCB dissection. In contrast to routine stent implantation, in which patients are reassured by the deployment of a scaffold, conservative management of dissection entails a period of clinical surveillance during which the patient is aware that a vessel injury remains unscaffolded. This experiential difference may amplify health anxiety, catastrophic thinking and hypervigilance to bodily sensations, constructs that are well described in the cardiac anxiety literature[19,23-25].

The identification of NHLBI type C dissection as the second strongest predictor (OR = 3.12) highlights the psychological importance of procedural severity. More extensive dissections require longer procedural times, heightened clinician vigilance, more intensive post-procedural surveillance and more detailed patient counselling, all of which may increase the patient’s perception of procedural severity. This strengthens the link between objective procedural severity and subjective psychological distress[8,9], and is reinforced by the finding that IVUS-derived dissection length, circumferential arc and medial involvement were all significantly greater in the anxiety-depression cohort.

Several convergent pathophysiological mechanisms may underlie the observed difference in MACE (27.7% vs 8.7%). Hypothalamic-pituitary-adrenal axis dysregulation, sympathetic hyperactivation and pro-inflammatory cytokine enhancement may act as psychoneuroimmunological pathways leading to endothelial injury at the dissection site, neointimal hyperplasia and a prothrombotic tendency[14-16]. Cardiovascular risk may be further compounded by behavioral mediators, including reduced medication adherence (a graded deficit observed across severity strata in the present cohort), lower rates of cardiac rehabilitation participation and poorer lifestyle modification[26-28].

The association between elevated hs-CRP and psychological distress (OR = 2.68) is consistent with the inflammatory hypothesis linking depression to cardiovascular sequelae. Systemic inflammation, whether triggered by the procedural insult itself or reflecting an underlying inflammatory diathesis, may concurrently promote depressive symptomatology and adverse vascular remodeling[29-31]. This finding raises the hypothesis that anti-inflammatory therapeutic approaches and inflammation-guided psychological screening might serve as complementary intervention targets.

Residual stenosis > 30% was identified as an independent predictor (OR = 2.21), emphasizing the importance of procedural optimization in reducing psychological sequelae. Patients aware of a suboptimal angiographic result may experience greater dissatisfaction with the procedure, prognostic uncertainty and anxiety about the future. This observation supports meticulous IVUS-guided procedural optimization together with open, empathetic communication of procedural outcomes[32,33].

The composite predictive model (AUC = 0.86) provides a potentially useful tool for systematic risk stratification. Pragmatic implementation would involve pre-procedural screening for known psychological vulnerability (psychiatric history, sex, social support) to guide targeted post-procedural biomarker measurement and psychological assessment in high-risk patients. Early referral for cognitive behavioral therapy, pharmacological anxiolysis or structured cardiac rehabilitation may reduce psychological morbidity and subsequent cardiovascular risk[34-36].

Limitations

Several limitations warrant acknowledgement. First, the retrospective, single-center design limits external validity and precludes causal inference regarding the direction of the anxiety-depression-MACE association. Second, the HADS, Generalized Anxiety Disorder-7 and Patient Health Questionnaire-9 are self-administered screening instruments rather than diagnostic reference standards; clinically significant anxiety-depression was therefore defined by score thresholds rather than by structured psychiatric interview. Third, the sample size was modest relative to the number of candidate variables entered into the multivariate model, and the predictive model was developed and evaluated in the same dataset without internal or external validation, so the reported AUC is likely to be optimistic. Fourth, no adjustment was made for multiple comparisons across the numerous clinical and biomarker endpoints examined. Fifth, although the 12-month horizon was adequate for MACE ascertainment, it may be insufficient to capture late psychological recovery or deterioration. Sixth, as informed consent was waived and the analysis relied on existing records, subjective variables such as illness perception and patient-reported procedural experience could not be captured. Future work should include prospective multicenter validation, randomized assessment of psychological interventions on MACE outcomes, serial IVUS evaluation to relate dissection healing kinetics to psychological status, and integration of genetic polymorphisms affecting stress reactivity and inflammatory responses.

CONCLUSION

Anxiety-depression is an under-recognized, prognostically adverse and potentially modifiable complication of coronary dissection after IVUS-guided DCB angioplasty. The present study identified several independent predictors and developed a composite model with an AUC of 0.86 for the early identification of psychologically vulnerable patients. The severity-graded association between psychological distress and 12-month MACE, together with plausible mediation through inflammatory, neuroendocrine and behavioral pathways, provides a strong rationale for integrating psychocardiological screening into post-interventional follow-up of DCB-treated patients with dissection. Targeted psychological intervention in this population may reduce psychological suffering while improving long-term cardiovascular prognosis.