Treating Women With Calcified Coronary Arteries Using Intravascular Lithotripsy: Primary Results of the EMPOWER CAD Study

Abstract

Background

Percutaneous coronary intervention of severely calcified coronary lesions is associated with a higher risk of adverse events in women compared with men.

Methods

EMPOWER CAD (NCT05755711) is a prospective, multicenter, single-arm “real-world” study of women receiving intravascular lithotripsy (IVL) to treat calcific coronary artery lesions. The 30-day primary safety end point was target lesion failure (TLF), a composite of cardiac death, myocardial infarction (MI), or ischemia-driven target lesion revascularization. Procedural success was defined as successful stent delivery with residual stenosis ≤30% without in-hospital TLF.

Results

A total of 399 women were enrolled at 45 centers in Europe and the United States. Mean age was 73 ± 10 years; diabetes (46.1%), prior MI (32.3%), chronic kidney disease (27.8%), and prior percutaneous coronary intervention (38.8%) were prevalent. Of the patients, 23.3% presented with NSTE-ACS, 16.0% with LVEF ≤40%, 7.3% of target lesions were in the left main artery, and severe calcification was present in 99.3%. IVL was the first calcium-modifying treatment for 90.6% of lesions. Procedural success (87.4%) and TLF at 30 days (11.6%) were both driven by the 9.6% periprocedural MI rate per the Society for Cardiovascular Angiography & Interventions (SCAI) definition. Diameter stenosis was reduced from 78% ± 12% at baseline to 46% ± 12% after IVL, and to 12% ± 7% in-stent at final angiography. Seattle Angina Questionnaire (SAQ)-7, European Quality of Life 5 Dimension 5 Level (EQ-5D-5L), and GAD-7 quality-of-life measures all showed significant improvements at 30 days (P < .0001).

Conclusions

In the real-world EMPOWER CAD study, an IVL-first strategy demonstrated a favorable safety profile, high procedural success, and significant stenosis reduction and quality of life improvement at 30 days, consistent with or better than prior IVL studies performed in predominantly male populations.

Introduction

Despite cardiovascular disease being the leading cause of death in women, early diagnosis and understanding of optimal percutaneous coronary intervention (PCI) strategies in women with calcified coronary artery disease (CAD) are lacking.^1,2 Women are consistently underrepresented in PCI clinical trials,3, 4, 5, 6 despite having unique risk factors for cardiovascular disease,⁷ smaller vessels, and more vascular calcification.^4,8 Insufficient understanding of potential differences in care and intervention may account for the increased mortality, major adverse cardiovascular events (MACE), myocardial infarction (MI), and target lesion revascularization after PCI in women.^2,6,9

Among women, treatment of calcified lesions themselves is a predictor of poor outcomes after PCI.¹⁰ Modification of calcification before stent implantation is common^11,12; however, women have increased rates of dissection, perforation, and bleeding after rotational or orbital atherectomy compared with men.13, 14, 15 In contrast, the Shockwave intravascular lithotripsy (IVL) system’s (Shockwave Medical Inc) mechanism of action involves delivery of ultrasonic acoustic pressure waves,¹⁶ and the rates of serious angiographic complications and MACE are low and similar for men and women.^17,18

With the historical underrepresentation of women in PCI studies, an expert consensus document has identified the need for sex-specific trials on calcified plaque modification strategies.¹ Although the EMPOWER CAD study was not designed for sex-based comparisons, it directly addresses the underrepresentation of women in PCI trials by characterizing procedural safety, lesion complexity, and quality-of-life (QOL) outcomes in an exclusively female population.

Methods

Study design

EMPOWER CAD (NCT05755711) is a prospective, multicenter, single-arm, real-world observational study of women with calcified CAD treated with IVL. Details of the study design, IVL-first algorithm, and the Shockwave Medical Coronary IVL System have been published previously.^16,19 Briefly, the study design allowed enrollment of all women with limited exclusion criteria to reflect real-world practice. To meet enrollment criteria, calcification of the target lesion had to be assessed with angiography, with fluoroscopic radio-opacities noted without cardiac motion involving both sides of the arterial wall or with intravascular ultrasound/optical coherence tomography (OCT), noting the presence of ≥270° of calcium on at least one cross-section. Only patients presenting in cardiogenic shock, for treatment of in-stent restenosis, or with a life expectancy <1 year were excluded. All patients provided written informed consent, sites followed local legal and regulatory requirements for ethics committee and institutional review board approvals, and the study was conducted in accordance with the Declaration of Helsinki guidelines and good clinical practices.

Primary and secondary end points

The primary safety end point of target lesion failure (TLF) at 30 days was defined as a composite of cardiac death, MI attributable to the target vessel, or ischemia-driven target lesion revascularization (ID-TLR). The protocol prespecified that periprocedural MI (PPMI) within 48 hours of procedure would be defined using the Society for Cardiovascular Angiography & Interventions (SCAI) definition²⁰ and the Fourth Universal Definition²¹ would be used for spontaneous MI (>48 hours after procedure). The primary efficacy end point of procedural success was defined as successful stent delivery with a residual in-stent stenosis ≤30% in all target lesions (core laboratory) and without adjudicated in-hospital TLF.¹⁹

Other end points included serious procedural angiographic complications, MACE, Seattle Angina Questionnaire-7 (SAQ-7), European Quality of Life 5 Dimension 5 Level (EQ-5D-5L), and Generalized Anxiety Disorder-7 (GAD-7) questionnaires. For comparative purposes, PPMI was prospectively adjudicated using SCAI²⁰ (primary), Fourth Universal Definition,²¹ and the Modified Academic Research Consortium²² definitions (see Supplemental Table S2). Women-specific history included pregnancy history, menopausal status, hormone replacement therapy, past and current use of hormonal contraceptives, hysterectomy, and oophorectomy status. The EMPOWER CAD study included a subset of patients receiving OCT-guided IVL, although that prespecified subanalysis is not included in this primary analysis. We report the primary study end point and secondary outcomes at 30 days; study follow-up will continue through 3 years.

Statistics

Results are presented with descriptive statistics. Categorical variables were compared with a χ² test, Fisher exact test, or McNemar’s χ², and exact confidence intervals were generated for estimates of proportions. Within-patient changes in continuous variables were assessed parametrically using the paired t test or nonparametrically using the sign-rank test. Multivariate logistic regression analyses with stepwise selection were performed to assess predictors of the primary safety and efficacy end points. Baseline demographics, lesion characteristics, and procedural data were initially considered in the model; a P value of .05 was used as the significance threshold, and final variables in the model are reported in forest plots. Statistical analyses were performed using SAS version 9.4 or higher (SAS Institute).

Results

From May 2023 to October 2024, a total of 399 patients were enrolled at 45 centers. This was more rapid than originally anticipated, and therefore, enrollment was completed 4 months ahead of schedule. The study leadership included 32 women (71%) as site principal investigators at the 45 enrolling sites in the United States and Europe. Ten patients were not eligible for their 30-day follow-up, including 2 patients who withdrew consent and 8 deaths prior to 30 days, of which 4 were cardiac related and 4 were noncardiac related. The 4 cardiac deaths were included in the calculation of TLF and MACE at 30 days. Of the 389 patients who were eligible for their 30-day visit, follow-up compliance was 92.5% (360/389).

Baseline characteristics (Table 1) included a mean age of 73.1 ± 9.9 years, 89.5% (357/399) with hypertension, 46.1% (184/399) with diabetes, 32.3% (129/399) with prior MI, 27.8% (111/399) with chronic kidney disease, and 38.8% (155/399) with prior PCI. The indication for PCI included 2.5% (10/399) of patients with ST-segment elevation myocardial infarction (STEMI) and 23.3% (93/399) with NSTEMI-ACS. At baseline, 99.3% (418/421) of target lesions were severely calcified, with a mean lesion length of 33.5 ± 17.3 mm, a diameter stenosis of 77.7% ± 12.3%, and 36.7% (158/431) were bifurcation/trifurcation lesions according to the core lab (Table 2).

Table 1.

Baseline characteristics.

Characteristic	N = 399
Age, y	73.1 ± 9.9
Body mass index, kg/m2	27.9 ± 5.8 (398)
Diabetes mellitus	46.1% (184/399)
Hypertension	89.5% (357/399)
Hyperlipidemia	87.0% (347/399)
Prior stroke or TIA	13.3% (53/399)
Myocardial infarction	32.3% (129/399)
Prior PCI	38.8% (155/399)
Prior CABG	8.3% (33/399)
Peripheral vascular disease	14.8% (59/399)
Congestive heart failure	24.1% (96/399)
Arrhythmia	26.1% (104/399)
Chronic obstructive pulmonary disease	18.0% (72/399)
Smoking/tobacco use	48.9% (195/399)
Chronic kidney disease	27.8% (111/399)
Left ventricular ejection fraction ≤40%	16.0% (64/399)
Pregnancy related
Gravida (no. of pregnancies)
0-1	24.3% (96/395)
2	25.6% (101/395)
≥3	49.4% (195/395)
Parida (no. of births)
0-1	28.9% (114/395)
2	33.4% (132/395)
≥3	37.2% (147/395)
History of miscarriage	26.1% (103/395)
History of gestational diabetes	9.6% (38/395)
History of gestational hypertensive disorders	9.9% (39/395)
Other reproductive history
Past use of hormonal contraceptives	40.0% (158/395)
Menopause	7.1% (28/395)
Postmenopause	89.4% (353/395)
History of hormone replacement therapy	17.2% (68/395)
History of oophorectomy	22.0% (87/395)
History of hysterectomy	33.9% (134/395)

Values are mean ± SD or % (n/N). CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention; TIA, transient ischemic attack.

Table 2.

Procedural characteristics.

Characteristics	399 patients with 449 lesions
Lesions treated with IVL	99.8% (448/449)
Target lesion vessel
Left anterior descending coronary artery	48.3% (217/449)
Right coronary artery	30.1% (135/449)
Left circumflex	14.3% (64/449)
Left main	7.3% (33/449)
PCI indication
STEMI	2.5% (10/399)
NSTE-ACS	23.3% (93/399)
Stable angina	45.6% (182/399)
Unstable angina	10.8% (43/399)
Other	17.8% (71/399)
Subjects treated with IVL	99.7% (398/399)
Target lesions treated per subject
1	87.4% (348/398)
2	12.6% (50/398)
Total IVL catheters per subject	1.2 ± 0.5 (398)
Access site
Femoral	45.2% (180/398)
Radial	54.0% (215/398)
Other	0.8% (3/398)
Guide catheter used
5F	2.5% (10/398)
6F	65.8% (262/398)
7F	29.6% (118/398)
8F	2.0% (8/398)
Core lab-assessed lesion characteristics
Lesion length, mm	33.5 ± 17.3 (396)
RVD, mm	3.0 ± 0.5 (416)
RVD <2.5 mm	9.8% (42/430)
MLD, mm	0.7 ± 0.4 (416)
% DS	77.7 ± 12.3 (416)
Eccentrica	8.4% (36/431)
Calcification
Moderate	0.7% (3/422)
Severe	99.3% (419/422)
Calcification length, mm	45.2 ± 22.0 (420)
Bifurcation/trifurcation	36.7% (158/431)

Values are % (n/N) or mean ± SD (N). DS, diameter stenosis; IVL, intravascular lithotripsy; MLD, minimal lumen diameter; NSTE-ACS, non-ST-elevation acute coronary syndromes; PCI, percutaneous coronary intervention; RVD, reference vessel diameter; STEMI, ST-segment elevation myocardial infarction.

^aEccentric is defined as a vessel that has only 1 of its luminal edges compromised by >25%.

There was a total of 448 lesions treated with IVL (Table 2). Target lesions were in the left anterior descending in 48.3% (217/449), right coronary artery in 30.1% (135/449), left circumflex in 14.3% (64/449), and the left main stem in 7.3% (33/449); 9.7% (42/431) of lesions had a reference vessel diameter <2.5 mm. Radial access (54.0%, 215/398) was more common than femoral (45.2%, 180/398), with most IVL delivered through a 6F guide catheter (65.8%, 262/398). A total of 103 patients had OCT-guided IVL, intravascular ultrasound–guided IVL was used in 252 patients, and 42 patients had angiography guidance only. Most target lesions (90.6%, 407/449) were treated with IVL without the prior use of atherectomy or scoring balloons (Figure 1). Atherectomy was used in 7.8% (35/449) of lesions, with all but 2 of the atherectomy uses occurring pre-IVL. Cutting/scoring balloons were used in 4.9% (22/449) of lesions, with similar use pre- and post-IVL. Most lesions were stented after IVL (98.4%, 442/449); additional procedural characteristics are available in Supplemental Material 1.

Figure 1.

Procedural characteristics flow chart (site-reported). IVL, intravascular lithotripsy.

Safety and effectiveness end points

Procedural success was achieved in 87.4% (341/390), with 98.5% (384/390) of patients achieving ≤30% residual stenosis and 89.2% (348/390) free from in-hospital TLF events. The in-hospital TLF events consisted of 0.8% (3/399) cardiac death, 9.8% (39/399) target vessel MI, and 1.3% (5/399) ID-TLR, with 30-day TLF rates in Table 3. SCAI-defined PPMI was 9.5% (38/397),²⁰ and spontaneous MI after discharge was 1.5% (6/397) based on the Fourth Universal Definition.²¹ Of the 38 PPMI, 52.6% (20/38) were due to elevated biomarkers with no clinical symptoms. The rate of PPMI was 4.5% (18/397) using the Fourth Universal Definition²¹ and 1.5% (6/397) based on modified Academic Research Consortium-2 definitions (Table 4).20, 21, 22 Additional details of troponin and CK-MB measures and thresholds are available in Supplemental Material 1. Predictors of TLF at 30 days and procedural success are shown in Figure 2, and only procedural duration was an independent predictor of the safety and efficacy end points (P < .001). At 30 days, 1.0% (4/397) of patients had primary stent thrombosis, of which 2 were definite, and 2 were probable.

Table 3.

Outcomes at 30 days.

End points	N = 397
Target lesion failure	11.6% (46/397)
Cardiac death	1.0% (4/397)
MI	10.6% (42/397)
Periprocedural MI (<48 h)a	9.6% (38/397)
Spontaneous MI (>48 h)b	1.5% (6/397)
ID-TLR	1.3% (5/397)

Values are % (n/N). ID-TLR, ischemia-driven target lesion revascularization; MI, myocardial infarction.

^aSCAI definition used for periprocedural MI.

^bThe Fourth Universal Definition (type 4a) used for spontaneous MI.

Table 4.

Periprocedural myocardial infarction by alternative definitions.

Name	Abbreviated definition	PPMI in EMPOWER CAD
SCAI20	•CK-MB >10× or >5× when new Q waves or LBBB occur, or •Tn ≥70× measured within 48 hours, or •Tn ≥35× with new pathologic Q waves in ≥2 contiguous leads, or •New persistent LBBB	9.6% (38/397)
Fourth Universal Definition21	•Tn >5× PLUS chest pain, or •Ischemic ECG changes, or •Loss of myocardium on serial imaging, or •Periprocedural complications	4.5% (18/397)
Modified Academic Research Consortium-222	•Tn >35× OR CK-MB >5× PLUS Q waves, or •Unresolved PCI complications at procedure end, or •Loss of myocardium on serial imaging	1.5% (6/397)

ECG, electrocardiogram; LBBB, left bundle branch block; PCI, percutaneous coronary intervention; PPMI, periprocedural myocardial infarction.

Figure 2.

Multivariable logistic regression analysis of predictors of end points. (Top) Target lesion failure (TLF) at 30 days and (bottom) procedural success. CABG, coronary artery bypass graft; CKD, chronic kidney disease; DS, diameter stenosis; LVEF, left ventricular ejection fraction; MLD, minimal lumen diameter; NSTE-ACS, non ST-elevation acute coronary syndrome; RVD, reference vessel diameter; STEMI, ST-segment elevation myocardial infarction.

Angiographic results

Significant diameter stenosis improvements are shown in Figure 3A. Baseline mean minimal lumen diameter (0.7 ± 0.4 mm) improved to 1.6 ± 0.4 mm after IVL to 2.6 ± 0.4 mm in-stent after the final procedure, with a final in-stent acute gain of 2.0 ± 0.5 mm (Figure 3B). The rate of serious angiographic complications at final angiography was 0.2% (1/420) (Table 5). A single perforation was sealed with a stent, and the patient was discharged the next day and had no further complications through their 30-day follow-up. There were 3 perforations poststenting that were treated with covered stents (n = 2) or a coil (n = 1).

Figure 3.

Angiographic results. (A) Diameter stenosis and (B) cumulative distribution function for the mean lumen diameter (MLD) showing significant improvements post-IVL and at final angiography as compared to the baseline. IVL, intravascular lithotripsy.

Table 5.

Core lab-adjudicated serious angiographic complications.

Complications	Post-IVLa	Post-stent	Final
Serious angiographic complications	2.6% (6/233)	1.2% (5/410)	0.2% (1/420)
Dissections
Type D	2.6% (6/233)	0.2% (1/410)	0% (0/420)
Type E	0% (0/233)	0% (0/410)	0% (0/420)
Type F	0% (0/233)	0.2% (1/410)	0% (0/420)
Perforation	0% (0/233)	0.7% (3/410)	0.2% (1/420)
Abrupt closure	0% (0/233)	0.5% (2/410)	0% (0/420)
Persistent slow flow or no reflow	0% (0/233)	0% (0/410)	0% (0/420)

IVL, intravascular lithotripsy.

^a233/399 patients had angiography after IVL.

Secondary end points

At 30 days, the overall rate of MACE was 12.1% (48/397), including 1.0% (4/397) cardiac deaths, 11.1% (44/397) MI, and 1.3% (5/397) of patients requiring target vessel revascularization. Of the 44 protocol-defined MI, 2 were not attributed to the target vessel; 5 were STEMI, 38 were NSTEMI, and 1 type was undetermined. All QOL measures showed significant improvements compared with baseline at 30 days: SAQ-7 summary scores increased by a mean of 17.9 ± 22.7 from baseline (P < .0001), with significant improvements in all domain scores (Figure 4). Compared with baseline values, EQ-5D-5L increased by a mean of 7.8 ± 20.9 (P < .0001). GAD-7 anxiety questionnaires also significantly improved with a mean decrease in anxiety score of 2.2 ± 4.8 (P < .0001) and 72.5% (258/356) of patients reporting minimal anxiety at 30 days.

Figure 4.

Seattle Angina Questionnaire-7 (SAQ-7) scores for the EMPOWER CAD study. All comparisons between baseline (dark blue) and 30 days (light blue) were significant (P < .0001; a higher score indicates improvement).

Discussion

EMPOWER CAD is the first real-world study to exclusively evaluate an all-women population with calcified coronary artery lesions treated with IVL. The study addresses 2 important gaps: (1) the underrepresentation of women in PCI trials by expanding the body of available evidence and benchmarking their outcomes following plaque modification of severely calcified lesions, and (2) establishment of the safety and feasibility of an IVL-first approach in a population whose acute complications and long-term outcomes with non-IVL therapies have been worse than that reported in men.^2,4 The primary findings of the EMPOWER CAD study (Central Illustration) demonstrated that in a highly complex population of women with severely calcified lesions: (1) an IVL-first strategy was feasible in over 90% and was successful in modifying severely calcified coronary lesions; (2) adverse event rates were consistent with prior calcified CAD trials of lower risk patients; and (3) women had quantifiable improvements in all QOL measures as early as 30 days after the procedure.

Central Illustration.

In the real-world EMPOWER CAD study, an intravascular lithotripsy (IVL)-first strategy demonstrated a favorable safety profile, high procedural success, and significant stenosis reduction and quality of life improvement at 30 days. %DS, % diameter stenosis; EQ-5D-5L, European Quality of Life 5 Dimension 5 Level; NSTE-ACS, non ST-elevation acute coronary syndrome; STEMI, ST-segment elevation myocardial infarction.

EMPOWER CAD recruited a “real-world” population of women and represents a challenging cohort enriched with comorbidities such as diabetes, hypertension, smoking, and renal insufficiency compared with the Disrupt CAD IVL approval studies.3, 4, 5 It included patients with left main lesions,²³ STEMI and NSTEMI-ACS,^24,25 and LVEF ≤40%, all high-risk CAD groups known to have worse outcomes after PCI. Unique to EMPOWER CAD is the reproductive history of the women enrolled: 26.1% had a history of miscarriage or recurrent miscarriage, a risk factor associated with cardiovascular death and stroke later in life.^26,27 Most enrolled women were postmenopausal, and nearly a quarter had a prior oophorectomy; whether this estrogen-poor population explains the extent of CAD and the severe calcium phenotype of the population can only be postulated.^28,29 As expected from a women-only study, vessel diameters were small, and likely a contributor to acute complications when combined with the extensive calcification observed in the study.^4,8,10

Despite the challenging population and lesion types treated in EMPOWER CAD, 90.6% were treated with the intended IVL-first approach and achieved an overall 87.4% procedure success comparable to prior, less complex IVL coronary approval clinical trials.³ The need for adjunctive therapies such as cutting/scoring balloons (4.9%) or atherectomy (7.8%) in EMPOWER CAD was low compared to other real-world IVL registries, which have reported 31% cutting/scoring balloon and 13% to 30% rotational/orbital atherectomy adjunctive use.^11,12 It is likely that the IVL-first protocol contributed to low usage of other calcium modification, in comparison to the REPLICA-EPIC18 study, where only 9.4% of patients were treated with IVL alone in a selected population.¹¹ The low rate of adjunctive therapies suggests that IVL alone was sufficient to effectively prepare lesions for stenting. The final residual in-stent stenosis (11.6%) in EMPOWER CAD reflects the complexity of lesions treated and is comparable to other “real-world” studies (15%) evaluating calcium-modifying technologies.^11,30

Several important factors account for the 11.6% TLF rate at 30 days. First is the challenging all-comer patient population with its complex lesion characteristics, representative of the intended real-world study, which are known to contribute to worse outcomes and cannot justifiably be compared to far more selective populations of the IDE approval trials. The second factor relates to PPMI and the impact of definitions and systematic biomarker ascertainment on the outcome. Although the debate over the ever-evolving PPMI definitions continues, EMPOWER CAD provides added clarity on the impact of accepted definitions, as these were all prospectively adjudicated by protocol in the study. The main contributor to 30-day TLF events was PPMI (9.6%), selected to be SCAI defined for the prespecified primary end point of the study, and is comparable to other studies showing rates from 6% to 18%.³¹ Importantly, rates of PPMI are heavily dependent on the definitions used and biomarker collection protocols that can vary from study to study.^20,21 For context, patients treated with orbital atherectomy in the recent multicenter ECLIPSE trial³² had a similar overall TLF rate (11.5%) with more cardiac death (4% vs 1.0% in EMPOWER CAD) and ID-TLR (3.4% vs 1.3% in EMPOWER CAD). The difference in procedural MI in ECLIPSE (4% vs 9.6% in EMPOWER CAD) was that enzyme collection was done only if clinically indicated in the ECLIPSE trial, rather than the systematic serial assessment in all EMPOWER CAD patients, irrespective of symptoms suggestive of ischemia. As a result of EMPOWER CAD's strict biomarker collection protocol, over half of the PPMI were exclusively asymptomatic biomarker elevations without other objective evidence of myocardial injury.

Previous studies have suggested similar outcomes with IVL independent of sex,¹⁸ and the EMPOWER CAD results are consistent with prior IVL studies that included both women and men.³ In this study, procedural outcomes were comparable to prior male-predominant studies such as REPLICA-EPIC18 (20% female)¹¹ and ECLIPSE (27% female),³² while also delivering measurable early QOL improvement. Orbital atherectomy has been associated with more severe dissections in women compared to men (6.4% vs 1.7%, P = .01).¹⁵ More dissections (women 4.6%, men 1.3%, P = .004) and a higher incidence of perforation (women 2.1%, men 0.4%, P = .046) have been reported with rotational atherectomy, although there was no difference in AE between men and women at later follow-ups.^13,15 Others have postulated disease severity, lesion complexity, or vessel diameter as potential reasons for the difference in procedural complications between men and women.^1,2 Although only hypothesis-generating, procedure duration was the only predictor of procedural success and TLF at 30 days. Other studies have found that increased lesion complexity, whether that be due to severe calcification³³ or the presence of CTO,³⁴ is associated with longer procedural times. Potentially, the severity of calcification and lesion complexity across the entire EMPOWER CAD study cohort may have led to the unique regression analysis results, and future assessments into this signal are warranted.

A pooled patient-level analysis of the Disrupt CAD studies demonstrated low procedural complications and similar outcomes in women compared to men treated with IVL,³⁵ leading an SCAI expert consensus to suggest that IVL may be a preferred first-line therapy for women with calcified lesions requiring plaque modification.¹ EMPOWER CAD, therefore, fulfills an evidence gap serving to benchmark acute and long-term outcomes for women that support an IVL-first recommendation. Although sex-based differences are helpful to highlight potential vulnerabilities to higher acute complications in female patients, it is important to emphasize that men should not be the standard by which outcomes in female patients are measured. Biological and clinical differences, including smaller vessel caliber, plaque composition, and hormonal influences, necessitate the establishment of independent benchmarks for women.^1,7 In the absence of needed randomized data for women and failed recruitment efforts in traditional trials, clinicians and patients are too often left with therapeutic uncertainty for their female patients. EMPOWER CAD sets a new standard for establishing the expanded evidence base needed to establish a predictable pathway of care that, in this context, supports SCAI statement endorsing IVL as a first-line approach in female patients with complex calcified CAD.

Another important perspective of the EMPOWER CAD study is its focus on QOL, with paired baseline and 30-day QOL assessments, distinguishing it from most interventional CAD device trials, including Disrupt CAD III/IV.¹⁸ Defining quantifiable symptom recovery trajectories among women undergoing complex PCI in this population will assist in the patient-centered dialog and personal decision-making they face when considering revascularization options. Previous studies have shown that after PCI, women often experience slower and less pronounced early improvements in QOL compared with men.^36,37 However, in EMPOWER CAD, significant and consistent improvements were observed by 30 days across all QOL instruments (SAQ-7, EQ-5D-5L, GAD-7). In addition, women have historically been at greater risk of vascular access site complications,² and have had higher rates of femoral access compared to men.^13,18 Although radial access (54.0%) was higher than femoral access (45.2%) in EMPOWER CAD, the use of radial access was lower than the 55.7% to 75% that was reported for women in other studies.^13,14,18 This most likely reflects the complexity of CAD treated, and the anticipated need for larger guide catheters and/or improved support.

Finally, the EMPOWER CAD study recognized the contributions of women physicians in the field of interventional cardiology through its nomination of its principal investigators and inclusion of over 70% of centers with women in leadership. The study had expedited enrollment, which has generally not been the case in other PCI studies.3, 4, 5, 6 The emphasis on creating more career advancement, research, and networking opportunities for women³⁸ could be a model for future clinical trials to address the significant gap in recruiting and retention of women.

Limitations

The lack of a control group in the EMPOWER CAD study limited comparisons to other treatment modalities or comparisons to men. Also, single-arm studies in which patients and providers were not blinded may have led to bias in patient-reported QOL outcomes. The high success of stand-alone IVL without the need for other adjunctive therapies resulted in a low number of other calcium modification therapies, and assessments of differences in outcomes based on differences in treatment strategy are inherently limited in this study. Other randomized controlled trials, such as Short-Cut (NCT06089135), DECALCIFY (NCT04960319), and SONAR (NCT05208749) will provide additional information on comparative outcomes with different treatment strategies.

Conclusion

The EMPOWER CAD study evaluated outcomes in a high-risk real-world population of women with complex calcified coronary artery lesions who were treated with IVL. An IVL-first strategy in this population resulted in favorable angiographic and clinical outcomes as well as significant improvements in QOL.