Higher clinical class Clinical, Etiological, Anatomical, and Pathophysiological score is associated with lower odds of improvement in patient-reported outcomes after endovenous thermal ablation of truncal veins

Abstract

Objective

Endovenous thermal ablation (EVTA) of superficial lower extremity veins performed with laser or radiofrequency ablation has high rates of technical success. However, it is understudied how improvement in patient-reported outcomes after EVTA is related to the preoperative clinical class in the Clinical, Etiological, Anatomical, and Pathophysiological (CEAP) classification. This study investigated whether patients with a higher CEAP clinical class have lower odds of perceived symptomatic improvement than patients with lower preprocedural clinical class.

Methods

A retrospective cohort analysis of Vascular Quality Initiative (VQI) Varicose Vein Registry data from 2014 to 2021 was conducted. Patients >18 years of age who underwent EVTA alone and followed up within 3 months were included, and those with a clinical CEAP class of C0 or C1, nontruncal ablation, missing baseline CEAP score, or weight and height outside VQI data standards were excluded. Patient-reported limb appearance, work impact, and a composite score of heaviness, achiness, swelling, throbbing and itching (HASTI) were analyzed on a per-limb basis. Factors associated with improvement in patient-reported outcomes were tested with multivariable mixed effects logistic regression. The models adjusted for CEAP class, demographics, and anatomical variables as fixed effects and were clustered by VQI center. Demographic and comorbidity data were compared across CEAP classes with Chi-squared, Kruskal-Wallis, or analysis of variance testing as appropriate.

Results

There were 6364 patients who met inclusion the criteria, and we analyzed 7607 limbs within the study period. Patient demographics and comorbid venous pathologies were not evenly distributed among the CEAP classes. Patients with C6 disease (odds ratio [OR], 0.63; 95% confidence interval [CI], 0.45-0.88; P = .01) and patients with a history of phlebitis (OR, 0.76; 95% CI, 0.59-0.99; P = .04) had lower odds of improvement in HASTI score. For patient-reported appearance, C3 (OR, 0.79; 95% CI, 0.66-0.94; P = .01) and C6 (OR, 0.50; 95% CI, 0.38-0.67; P < .01) disease were associated with lower odds of improvement. Obesity (OR, 0.87; 95% CI, 0.76-0.99; P = .04), preoperative anticoagulation (OR, 0.77; 95% CI, 0.62-0.97; P = .03), and prior vein treatment (OR, 0.81; 95% CI, 0.71-0.93; P < .01) were also negatively associated with appearance, whereas age ≥65 years was positively associated (OR, 1.17; 95% CI, 1.01-1.35; P = .04). Clinical class showed no association with an improvement in impact on work. Female sex was positively associated with an improved impact on work (OR, 1.17; 95% CI, 1.02-1.36; P = .03).

Conclusions

Higher CEAP clinical class was associated with odds of improvement in HASTI and appearance but not associated with improvement in patient perception of impact on work. These findings are important for patient counseling regarding what outcomes they can expect with EVTA based on their preoperative clinical presentation.

Article Highlights.

• •Type of Research: Retrospective cohort study of national Vascular Quality Initiative Varicose Vein Registry data
• •Key Findings: In 6364 patients undergoing endovenous thermal ablation of truncal veins, there were lower odds of improvement in composite heaviness, achiness, swelling, throbbing, and itching for C6 disease (odds ratio [OR], 0.63; P = .01). Odds of improved appearance were lower for patients with C3 (OR, 0.79; P = .01) and C6 (OR, 0.50; P < .01) disease.
• •Take Home Message: Endovenous thermal ablation is effective in improving patient-reported outcomes across clinical class Clinical, Etiological, Anatomical, and Pathophysiological scores. Lower odds of improvement in certain classes are important for patient counseling prior to intervention.

Chronic venous disease (CVD) is common globally, with estimates of 21.1% to 61.3% of the population affected in various regions.1, 2, 3, 4, 5 The prevalence of CVD is similarly high in the United States with the rise in obesity and aging population contributing to the disease burden.6, 7, 8, 9, 10, 11, 12 CVD can have a significant impact on quality of life and daily function.^13,14 Treatment with endovenous thermal ablation (EVTA) is appropriate across the spectrum of CVD per the most recent practice guidelines on varicose veins and venous ulcers,15, 16, 17 and EVTA performed using laser or radiofrequency has high rates of technical success with reliable symptom improvement.18, 19, 20, 21

The Clinical, Etiological, Anatomical, and Pathophysiological (CEAP) classification was introduced to standardize the description of CVD and reporting on treatment outcomes.²² The extent to which patient-reported outcomes (PROs) after EVTA are related to the preoperative CEAP clinical class remains understudied. The aim of this analysis was to characterize improvement in PROs based on CEAP classification and to identify other factors associated with these outcomes in the Vascular Quality Initiative (VQI) Varicose Vein Registry. The hypothesis was patients with higher CEAP clinical class will have lower odds of perceived improvement relative to C2 in self-reported heaviness, achiness, swelling, throbbing and itching (HASTI), appearance, and impact on work.

Methods

This was a retrospective cohort study of VQI Varicose Vein Registry data from 2014 to 2021. Patients >18 years of age who underwent EVTA of truncal veins with 3-month follow-up data were included. Patients were excluded if they did not conform to VQI data standards (eg, vein diameter 1-30 mm and length 1-70 cm, weight 18.1-227.0 kg, and height 137-203 cm), underwent nonablation procedures, were missing a baseline CEAP classification, had C0 or C1 disease, were reported as having zero veins treated, or had an inpatient status at the time of the procedure (Fig 1). Patients were stratified by CEAP clinical class. Of note, the CEAP classification was updated in 2020, such that separate groups were added for corona phlebectatica (C4c), recurrent varicose veins (C2r), and recurrent active venous ulcer (C6r). Given that most interventions in this study occurred before this change, patients classified as C4a-c were combined into one grouping, C4, and patient classified as C2r and C6r were grouped with C2 and C6, respectively.

Fig 1.

Flow chart of patient selection. CEAP, Clinical, Etiological, Anatomical, and Pathophysiological.

Preoperative variables of interest included age, sex, race, Hispanic ethnicity, body mass index (BMI), obesity (BMI ≥30), periprocedural prophylactic anticoagulation, prior history of lower extremity deep venous thrombosis (DVT) or superficial phlebitis, varicose vein treatment and compression treatment, and the number, length, and diameter of veins treated per limb. Additional variables of interest included number of pregnancies, history of vein recanalization, and preoperative PROs (HASTI, appearance, and work impact) and Venous Clinical Severity Score (VCSS).

The primary outcome was improvement in postoperative PROs (HASTI, appearance, and work impact). Improvement was defined as any superiority in the scoring of symptoms at follow-up. Secondary outcomes included the degree of PRO and VCSS improvement, as well as thrombotic and bleeding complications. A lower score on postprocedural PRO assessments represents an improvement in symptoms. For ease of analysis, we chose to represent improvement as positive values. Patient characteristics, changes in PROs, and postoperative complications were compared across CEAP class with analysis of variance or Kruskal-Wallis tests for continuous variables and Pearson χ² testing for binary variables. When a significant Kruskal-Wallis test was obtained for change in PROs (indicating that there was a difference among CEAP clinical classes), the Conover-Iman test was performed to compare each CEAP class against the others. A P value of ≤.05 was considered significant. The following variables were tested for association with the primary outcomes on univariate logistic regression: clinical class, female sex, age ≥65 years (binary), White race, Black race, Hispanic ethnicity, obesity, maximum vein length treated, maximum vein diameter treated, total veins treated, use of perioperative anticoagulation, history of phlebitis, history of DVT, prior varicose vein treatment, number of pregnancies, prior use of compression therapy, and presence of deep venous reflux. Variables with a P value of ≤.1 on univariate analysis were included in the multiple variable regression and backward elimination using a cutoff of a P value of ≤.05 was performed to arrive at the final models. The models were clustered by VQI center, which accounts for practice variations among the different participating centers by adding random effects for each cluster. The fixed effect variables were tested for multicollinearity with a cutoff coefficient of >0.3 leading to exclusion. All analyses were performed on a per-limb basis using R Studio Version 2024.12.0+467 (RStudio, PBC). The University of Rochester Medical Center research subjects review board approved this study and waived the requirement for consent (approval number: STUDY00007618).

Results

There were 6364 patients meeting inclusion criteria for a total of 7607 limbs treated. The mean age was 56.4 years, 67.1% of patients were female, and 72.4% were White (Table I). There were greater proportions of male and Black patients in higher CEAP classes. BMI was higher in the more advanced CEAP clinical classes. A prior history of DVT and phlebitis was more common in higher CEAP classes (P < .001). Overall, 97.6% of patients underwent compression treatment, although fewer C6 patients had compression therapy relative to the other groups (P < .001). A small percentage of veins treated were recanalized or remnant veins (1.8%). In general, C5 and C6 had the worst baseline symptoms.

Table I.

Patient characteristics and baseline symptoms

No. of limbs	C2	C3	C4	C5	C6	P value
	1700	3528	1582	214	583
Age, years	52.12 ± 13.55	55.53 ± 13.47	60.36 ± 12.96	62.14 ± 13.30	61.81 ± 13.47	<.001
Female sex	336 (19.8)	966 (27.4)	624 (39.4)	100 (46.7)	254 (43.6)	<.001
Whitea	1186 (69.8)	2650 (75.3)	1095 (69.4)	143 (66.8)	425 (73.1)	<.001
Black or African American	56 (3.3)	241 (6.8)	126 (8.0)	28 (13.1)	81 (13.9)	<.001
Asian	42 (2.5)	58 (1.6)	26 (1.6)	3 (1.4)	6 (1.0)	.123
Native American	1 (0.1)	2 (0.1)	1 (0.1)	0 (0.0)	1 (0.2)	.879
Other or unknown	372 (21.9)	493 (14.0)	299 (18.9)	33 (15.4)	55 (9.4)	<.001
Hispanic ethnicity	106 (6.3)	211 (6.0)	100 (6.4)	16 (7.5)	31 (5.3)	.800
BMI, kg/m2	27.41 ± 5.82	30.32 ± 6.89	32.25 ± 7.61	33.81 ± 8.01	33.83 ± 8.94	<.001
Periprocedural anticoagulation	189 (11.1)	548 (15.5)	260 (16.4)	26 (12.1)	111 (19.0)	<.001
History of phlebitis	137 (8.1)	332 (9.4)	263 (16.7)	39 (18.3)	77 (13.3)	<.001
Prior DVT	67 (3.9)	188 (5.3)	144 (9.1)	30 (14.0)	87 (15.0)	<.001
Prior varicose vein treatment	634 (37.4)	1243 (35.3)	551 (34.9)	68 (31.8)	157 (27.0)	<.001
Prior compression treatment	1672 (98.4)	3459 (98.1)	1542 (97.5)	209 (97.7)	541 (92.8)	<.001
Veins treated per limb	1.16 ± 0.62	1.09 ± 0.39	1.17 ± 0.61	1.11 ± 0.34	1.12 ± 0.36	<.001
Truncal vein ablated	1683 (99.0)	3468 (98.3)	1555 (98.3)	206 (96.3)	572 (98.1)	.031
Truncal recanalized/remnant vein ablated	20 (1.2)	67 (1.9)	29 (1.8)	9 (4.2)	11 (1.9)	.026
Presence of deep venous reflux	816 (48.5)	1256 (35.9)	694 (44.4)	88 (41.7)	292 (50.9)	<.001
Maximum vein length treated, cm	35.88 ± 15.12	37.18 ± 14.63	37.88 ± 15.45	36.95 ± 14.63	39.90 ± 15.62	<.001
Maximum vein diameter treated, mm	7.11 ± 2.79	7.93 ± 3.58	8.29 ± 3.53	8.06 ± 2.96	8.23 ± 2.98	<.001
No. of pregnancies	2.64 ± 2.72	2.90 ± 2.85	2.10 ± 2.77	1.87 ± 2.78	2.09 ± 2.95	<.001
HASTI	8.13 ± 5.05	9.73 ± 5.09	9.89 ± 5.33	10.87 ± 5.92	10.29 ± 5.61	<.001
Heaviness	1.81 ± 1.49	2.16 ± 1.48	2.09 ± 1.56	2.26 ± 1.63	1.83 ± 1.59	<.001
Achiness	2.31 ± 1.36	2.45 ± 1.38	2.37 ± 1.42	2.40 ± 1.56	2.35 ± 1.57	.023
Swelling	1.53 ± 1.52	2.39 ± 1.52	2.59 ± 1.58	2.96 ± 1.50	2.73 ± 1.59	<.001
Throbbing	1.48 ± 1.43	1.65 ± 1.45	1.54 ± 1.45	1.61 ± 1.52	1.84 ± 1.59	<.001
Itching	1.01 ± 1.31	1.08 ± 1.32	1.32 ± 1.41	1.61 ± 1.54	1.53 ± 1.46	<.001
Appearance	2.44 ± 1.17	2.18 ± 1.26	2.44 ± 1.24	2.41 ± 1.29	2.22 ± 1.36	<.001
Work impact	1.51 ± 1.16	1.79 ± 1.26	1.78 ± 1.30	2.18 ± 1.46	2.13 ± 1.59	<.001
VCSS	5.30 ± 2.01	6.94 ± 2.09	9.38 ± 2.98	10.55 ± 3.35	15.90 ± 4.24	<.001

BMI, Body mass index; C, clinical class; DVT, deep venous thrombosis; HASTI, heaviness, achiness, swelling, throbbing, itching; VCSS, Venous Clinical Severity Score.

Values are mean ± standard deviation or number (%).

Boldface entries indicate statistical significance.

^aRacial demographics may not add up to 100% given overlap with Hispanic ethnicity.

In every CEAP class, there was postprocedural improvement in all PROs and VCSS. Thrombotic (2.5%) and bleeding (0.2%) complications were rare. On Kruskal-Wallis testing there was no difference among CEAP clinical classes on the work impact PRO, although differences were detected for HASTI and appearance where C6 saw the smallest improvements (see Supplementary Tables I and II, online only for individual class comparisons). The greatest improvement in VCSS score was seen in the C6 subgroup (Table II).

Table II.

Degree of improvement in patient-reported outcomes (PROs) and complications

No. of limbs	C2	C3	C4	C5	C6	P value
	1700	3528	1582	214	583
ΔHASTI	5.10 ± 5.00	6.04 ± 5.44	6.07 ± 5.84	6.07 ± 6.15	4.65 ± 5.43	<.001
Δ Appearance	1.18 ± 1.33	0.96 ± 1.34	1.13 ± 1.39	0.99 ± 1.31	0.69 ± 1.51	<.001
Δ Impact on work	0.83 ± 1.32	0.92 ± 1.46	0.90 ± 1.45	0.91 ± 1.68	0.80 ± 1.45	.373
Δ VCSS	1.84 ± 3.08	2.45 ± 3.07	3.50 ± 4.00	3.07 ± 4.88	4.41 ± 5.34	<.001
Thrombotic complication	34 (2.1)	65 (2.0)	57 (3.9)	5 (2.5)	18 (3.2)	.001
Bleeding complication	2 (0.1)	4 (0.1)	4 (0.3)	0 (0.0)	1 (0.2)	.733

C, Clinical class; HASTI, heaviness, achiness, swelling, throbbing, itching; VCSS, Venous Clinical Severity Score.

Values are mean ± standard deviation or number (%).

Boldface entries indicate statistical significance.

Positive delta value denotes improvement.

On multiple variable mixed-effects logistic regression, there were lower odds of improvement in HASTI for patients with C6 disease (odds ratio [OR], 0.63; 95% confidence interval [CI], 0.45-0.88; P = .01) relative to C2 and for patients with a history of phlebitis (OR, 0.76; 95% CI, 0.59-0.99; P = .04). CEAP classes C3 (OR, 1.12; 95% CI, 0.91-1.39; P = .29), C4 (OR, 1.08; 95% CI, 0.84-1.40; P = .53), and C5 (OR, 0.66; 95% CI, 0.39-1.14; P = .14) all had similar odds of improvement in HASTI as C2 (Fig 2). For limb appearance, both C3 (OR, 0.79; 95% CI, 0.66-0.94; P = .01) and C6 (OR, 0.50; 95% CI, 0.38-0.57; P < .01) disease were negatively associated with odds of improvement. Obesity (OR, 0.87; 95% CI, 0.76-0.99; P = .04), perioperative anticoagulation (OR, 0.77; 95% CI, 0.62-0.97; P = .03), and prior vein treatment (OR, 0.81; 95% CI, 0.71-0.93; P < .01) were also negatively associated, whereas age ≥65 years was positively associated (OR, 1.17; 95% CI, 1.01-1.35; P = .04). CEAP classes C4 (OR, 1.00; 95% CI, 0.81-1.24; P = .98) and C5 (OR, 0.70; 95% CI, 0.44-1.11; P = .13) had similar odds of improved appearance as C2 (Fig 3). Clinical class was not associated with the patient-reported impact on work. Rather, female sex was positively associated with this outcome (OR, 1.17; 95% CI, 1.02-1.36; P = .03) (Fig 4).

Fig 2.

Forest plot of factors associated with improved composite of heaviness, achiness, swelling, throbbing, and itching. C class odds ratios referenced to C2. CI, confidence interval; OR, odds ratio.

Fig 3.

Forest plot of factors associated with improved appearance. C class odds ratios referenced to C2. CI, confidence interval; OR, odds ratio.

Fig 4.

Forest plot of factors associated with improved impact on work. C class odds ratios referenced to C2. CI, confidence interval; OR, odds ratio.

Discussion

In a study of 7607 limbs treated with EVTA in truncal veins, PROs improved for HASTI, appearance, and impact on work across all CEAP classes demonstrating excellent patient-centered results of this treatment at 3 months. However, results were not equal for all preoperative CEAP classes. For the composite HASTI score, C6 had lower odds of improvement relative to the reference C2. A history of phlebitis was also negatively associated with this outcome, which appears consistent with prior literature. Vuoncino et al²³ found that patients with a history of phlebitis were more likely to undergo recanalization of ablated veins. Perhaps this factor plays a role in the persistence of these patients' postoperative symptomatology. The slightly lower rates of compression therapy noted among C6 patients may also play a role in this finding. The nature of retrospective database analysis makes it difficult to speculate why lower rates of adherence to compression were found in this group. However, a recent meta-analysis by Stevenson et al²⁴ found that noncompliance with compression therapy in chronic venous insufficiency was due to “physical limitations, health literacy, discomfort, financial issues, and psychosocial issues.” Perhaps of equal import is the success of C3 to C5 disease in improving the HASTI symptoms relative to C2, demonstrating the effectiveness of this intervention. Importantly, this study did not consider a minimal clinically important difference (MCID) as a threshold for the primary outcome of improvement in symptoms, although MCID of 20% improvement correlates with a 1-point improvement on the Likert scales used in this study.²⁵ Paty et al²⁶ attempted to correlate HASTI scores to the Patient Global Impression of Change questionnaire, which typically defines MCID as “moderately improved.” In that study, patients who were “a little improved” had greater improvement in HASTI scores than those who were “moderately improved” on the Patient Global Impression of Change demonstrating the inherent difficulties in interpreting PROs.²⁶ Future studies should attempt to better define MCID in this population using the Varicose Vein Symptom Questionnaire.

Patient-reported appearance was less likely to improve in C3 and C6 disease, as well as among obese patients. Zottola et al²⁷ previously studied the impact of obesity on PROs after ablation in C3 disease and showed worse appearance outcomes in these patients. Interestingly, HASTI and work impact actually improved more among patients with obesity in that analysis.²⁷ Perioperative anticoagulation and prior vein treatment were also negatively associated with appearance, perhaps indicating the lower likelihood of improvement in patients with recalcitrant disease. These data counterintuitively demonstrated higher odds of improvement in appearance for patients aged ≥65 years. However, this finding is not without basis; prior studies have shown excellent outcomes for older patients undergoing ablation. Sutzko et al²⁸ showed that patients >65 years experienced significant improvement in their PROs after ablation, although not quite to the same degree as their younger counterparts. However, that study did not comment on the likelihood of the binary outcome of improvement vs no improvement. Perhaps patient expectations of improved appearance change with age, although this speculation hints at the inherent uncertainty in analyzing PROs. The lower odds of improved appearance in C6 disease may be partly due to lower rates of compliance with compression therapy. In an attempt to explain this finding, a subgroup analysis of the 583 C6 patients showed that 114 (28.3%) were classified as C5 (healed ulcer) at follow-up, and 249 (61.8%) remained at C6 (180 did not have a follow-up clinical classification recorded). In this study, only one follow-up timepoint was considered, although perhaps at further follow-up appointments improvement was noted. Additionally, there are challenges in interpreting what improved appearance might mean for a patient with C6 disease at baseline vs C2. Encouragingly, C4 and C5 disease were equally likely to improve as C2. These difficulties in PRO interpretation suggest the need for further qualitative studies to better understand patient expectations and responses.

CEAP class was not associated with impact on work, and only female sex was positively associated with this outcome. This finding is particularly noteworthy given the significant impact that venous ulcer disease has on work and disability.²⁹ One possible explanation for this is the greater proportion of females with C2 and C3 vs C5 and C6 disease. This finding may indicate that, in general, treatment at earlier stages of CVD in women results in better functional outcomes, even though this signal was not found within any single CEAP class. Notably, work impact is not reported on a per-limb basis, and thus a subgroup analysis was performed on patients who underwent unilateral interventions. Again, clinical class showed no association, whereas female sex was positively associated with improvement. The relatively low symptom burden at baseline may also have played a role in the overall lack of improvement in this outcome. The mean baseline score for patients with C2 to C4 disease fell within “full work/activity but have symptoms,” whereas patients with C5 and C6 disease had “mildly reduced work/daily activities.” For C2 to C4 patients to improve, they would have to reach the point of their disease having “no effect” on work or daily activities. C5 and C6 patients would have had to improve 2 points on the Likert scale to be asymptomatic. Future studies might focus on patients who report severe work impairment at baseline.

This study was limited to the treatment of CVD with ablation of truncal veins. Marston et al³⁰ suggested that phlebectomy be performed concomitantly with EVTA to decrease the risk of ulcer recurrence. Perhaps inclusion of adjunctive procedures with EVTA would show improved work outcomes in C5 and C6 relative to C2 disease. Marston et al's study³⁰ also found that deep venous reflux was associated with worse outcomes for patients with venous ulcers. However, this analysis found that the presence of deep venous reflux was not associated with any of the primary outcomes on univariate regression. This result certainly invites further study of whether and why the severity of CVD did not impact the odds of improving a patient's ability to work.

It bears mention that the maximum vein diameter and length of vein treated, as well as the number of veins treated, were not significant in the final regression models for the various PROs. This finding may indicate that EVTA is effective across the spectrum of presenting anatomy or perhaps shows that patient perception of improvement is unrelated to the severity of disease. It also aligns with previous findings by Pinto Rodríguez et al,³¹ who demonstrated no significant difference in VCSS based on vein size. In this study, VCSS did improve to a greater extent in higher clinical classes, although this may reflect the validity of VCSS in accurately measuring disease severity rather than greater treatment effectiveness in these patients.

This study is limited by the issues inherent to any retrospective database analysis, namely, selection bias and differences in practice patterns both in patient selection and follow-up among VQI participants. This study attempted to account for these differences by clustering the mixed-effects model by VQI center, which accounts for random effects specific to each cluster. Despite clustering, this strategy may not fully account for all the possible differences. Additionally, there were significant baseline demographic differences between clinical classes. Results should be interpreted in light of these group differences, which the authors attempted to control for using multiple variable logistic regression. Further, patients who would have been classified as C2r, C4c, or C6r under the new CEAP classification were grouped with C2, C4, and C6, respectively, given that most of the procedures in this study were conducted before the new classification. Although the clinical class is arranged in ascending order of disease severity, it is ultimately a classification tool rather than linearly discriminating degree of disease.

The primary outcomes are patient reported, and each patient's estimation of their own symptoms may be different. This study attempted to account for that by focusing mainly on whether patients felt their symptoms had improved at all, rather than analyzing the degree of improvement in depth. However, there are likely patient factors that affect one's self-assessed symptoms that are not captured in this study. Given the potential for patients who underwent bilateral interventions to impact the analysis, mixed effects regressions were repeated, clustering by VQI center and patient. No significant differences were found in those models. Additionally, the follow-up dataset does not include whether any adjunctive procedures were performed in a delayed fashion, which could impact the reported PROs. VCSS scores may improve for different reasons in each clinical class, but the authors were unable to report a breakdown of such improvements. Work impact is not reported on a per-limb basis, although the authors chose to analyze it in this way to detect any associations with clinical class without excluding patients with bilateral disease. The results of this study may not be generalizable, given that they only include centers that participate in the VQI, which likely represent a biased sample of vascular surgery practices.

Conclusions

In 7606 limbs undergoing EVTA on truncal veins, CEAP clinical class was associated with odds of improvement in HASTI and appearance but not associated with improvement in patients' perception of impact on work. Overall, EVTA showed excellent outcomes for improving patient-reported symptoms across the spectrum of CVD. These findings are important for patient counseling regarding what outcomes they can expect with EVTA based on their preoperative clinical presentation.

Author contributions

Conception and design: DL, JG, JE, KN

Analysis and interpretation: DL, JG, BK, GP, MS, JE, KN

Data collection: DL

Writing the article: DL, JG, BK, GP, MS, JE, KN

Critical revision of the article: DL, JG, BK, GP, MS, JE, KN

Final approval of the article: DL, JG, BK, GP, MS, JE, KN

Statistical analysis: DL, KN

Obtained funding: Not applicable

Overall responsibility: KN

Funding

None.

Disclosures

None.