Author + information
- Received April 27, 2016
- Revision received September 26, 2016
- Accepted October 3, 2016
- Published online December 19, 2016.
- Jaya Chandrasekhar, MBBSa,
- George Dangas, MD, PhDa,
- Jennifer Yu, MDa,b,
- Sreekanth Vemulapalli, MDc,
- Sunil Suchindran, PhDc,
- Amit N. Vora, MDc,
- Usman Baber, MD, MSa,
- Roxana Mehran, MDa,∗ (, )
- STS/ACC TVT Registry
- aIcahn School of Medicine at Mount Sinai, New York, New York
- bPrince of Wales Hospital, Sydney, Australia
- cDuke Clinical Research Institute, Durham, North Carolina
- ↵∗Reprint requests and correspondence:
Dr. Roxana Mehran, The Zena and Michael A. Wiener Cardiovascular Institute, The Icahn School of Medicine at Mount Sinai, One Gustave L. Levy Place, Box 1030, New York, New York 10029-6574.
Background A differential impact of sex has been observed in transcatheter aortic valve replacement (TAVR) outcomes from small observational studies and subgroup analyses of randomized trials.
Objectives The goal of this study was to compare the in-hospital and 1-year outcomes in male and female subjects from the U.S. nationwide TAVR registry.
Methods National data from the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry were used for in-hospital outcomes, and data linked from the Centers for Medicare & Medicaid Services were used to provide 1-year events. Multivariable logistic regression adjustment was performed for in-hospital outcomes. Fine-Gray models were used for nonfatal 1-year outcomes to account for the competing risk of death.
Results From 2011 to 2014, a total of 11,808 (49.9%) women and 11,844 (51.1%) men underwent TAVR. Compared with male patients, female patients were older, with a lower prevalence of coronary artery disease, atrial fibrillation, and diabetes but a higher rate of porcelain aorta, lower glomerular filtration rate, and higher mean Society of Thoracic Surgeons score (9.0% vs. 8.0%; all p < 0.001). Women were treated more often by using nontransfemoral access than men (45.0% vs. 34.0%). Despite using smaller device sizes, women achieved valve cover index ≥8% more often than men (66% vs. 54%). In-hospital vascular complications were higher in women (8.27% vs. 4.39%; adjusted hazard ratio [HR]: 1.70; 95% CI: 1.34 to 2.14; p < 0.001) and a trend toward higher bleeding (8.01% vs 5.96%; adjusted HR: 1.19; 95% CI: 0.99 to 1.44; p = 0.06) was observed; however, 1-year mortality was lower (21.3% vs. 24.5%; adjusted HR: 0.73; 95% CI: 0.63 to 0.85; p < 0.001) in women than in men.
Conclusions Female patients undergoing TAVR had a different risk profile compared with male patients. Notwithstanding a greater adjusted risk for in-hospital vascular complications, 1-year adjusted survival was superior in female patients.
Transcatheter aortic valve replacement (TAVR) has increasingly been adopted as the definitive treatment for severe symptomatic aortic valve stenosis in patients at extreme high risk for surgery (1,2). Unlike percutaneous coronary intervention, approximately 50% of patients undergoing TAVR are female (3,4). Female subjects are at higher risk of procedural complications compared with male subjects, particularly bleeding and vascular complications (4,5). These events have been associated with increased post-TAVR rehospitalization and mortality (4,6,7).
Heart teams typically use algorithmic scores such as the European System for Cardiac Operative Risk Evaluation score and the Society of Thoracic Surgeons (STS) score to determine the 30-day mortality and morbidity risks, as an aid to decision-making for selection of optimal therapies (TAVR, surgical aortic valve replacement, or medical therapy). However, these scores were designed to predict surgical, not TAVR risk (8,9). Moreover, because these scores attribute a higher risk to female sex (10), early risks may be overestimated and potentially result in underutilization of definitive valve replacement.
The longer term clinical outcomes after TAVR in female subjects compared with male subjects have not been extensively studied. Several observational studies and a recent meta-analysis have indicated improved survival in female patients undergoing TAVR than male patients (4,5,11,12). However, given the higher rate of procedural complications in female patients, the reasons for this finding are not fully understood. The 1-year post-TAVR clinical outcomes in female patients are of significant interest because they can influence expert consensus statements and recommendations for selection of patients for TAVR. We therefore sought to examine the sex-based differences in patients undergoing TAVR from the large national Transcatheter Valve Therapy (TVT) registry of the STS/American College of Cardiology (ACC). This registry is one of the largest of valve patients to date, receiving prospectively collected data on patients undergoing TAVR from 348 centers nationwide. All patients undergoing commercial TAVR in the United States with an approved valve device are routinely entered into this registry per the mandate of the U.S. Centers for Medicare & Medicaid Services (CMS) (1).
The STS/ACC TVT Registry was established in 2011 in accordance with the CMS National Coverage Determination requirements for TAVR reimbursements. The registry is periodically audited and includes comprehensive baseline information, as well as 30-day and 1-year follow-up data. The registry data are linked with Medicare administrative claims for detection of events requiring hospitalization. This methodology of data acquisition, event ascertainment, and analysis in the STS/ACC TVT Registry has been previously published in detail (3).
Endpoints and definitions
Endpoints were assessed in the hospital and at 1 year. In-hospital events included all-cause death, myocardial infarction (MI), stroke, major bleeding, and major vascular complications as per the Vascular Academic Research Consortium (VARC)-2 definition (13). Major adverse cardiac events (MACE) were defined as the composite of death, MI, or stroke. Net adverse cardiac events (NACE) were defined as the composite of in-hospital MACE, major bleeding, or major vascular complication. Other endpoints included composite death or stroke and death or MI.
One-year individual endpoints included time to event occurrence of death, MI, stroke, and clinically significant bleeding. Composite endpoints included MACE, death or stroke, death or MI, and a composite of MACE or clinically significant bleeding.
In-hospital events were validated at the site level before submission into the TVT Registry. In-hospital VARC-2 major bleeding was defined as in-hospital bleeding or a vascular event with hemoglobin drop or need for transfusion. VARC-2 major vascular complication was defined as the composite of major vascular access site complication or unplanned vascular surgery, annular rupture, aortic dissection, or perforation with or without tamponade. One-year events were derived from CMS data. One-year clinically significant bleeding was a composite of International Classification of Diseases-9th Revision-Clinical Modification, codes derived from CMS claims data (Online Appendix).
Groups were compared according to sex. Categorical data are presented as frequencies (percentages) and were compared by using the Pearson chi-squared test. Continuous variables are presented as mean ± SD or median (interquartile range) and were compared by using the Wilcoxon rank-sum test. Estimated odds ratios for in-hospital outcomes were generated by using logistic regression with generalized estimating equations to account for within-center clustering. One-year event rates were calculated by using the Kaplan-Meier method or as the cumulative incidence for nonfatal events. Because in-hospital events were derived from registry data and 1-year events were derived from CMS data, time-to-event curves represent a combination of the 2 data sources. Estimated hazard ratios (HRs) and confidence intervals (CIs) were generated by using a Cox model, with robust variance estimation to account for within-center clustering. Nonfatal events were analyzed by using the Fine-Gray method for competing risks (1,14). All analyses were performed in SAS version 9.4 (SAS Institute, Inc., Cary, North Carolina) and the R Environment for Statistical Computing (version 3.1.3 or greater, R Foundation for Statistical Computing, Vienna, Austria). Values of p < 0.05 were considered significant.
Outcomes were adjusted by using Cox modeling for all potential confounders based on previous studies from the TVT Registry (15). The following variables were adjusted for: age, body surface area, left ventricular ejection fraction, hemoglobin, platelet count, estimated glomerular filtration rate (GFR), number of days from registry commencement (November 1, 2011) to procedure date, race, current dialysis, left main stenosis ≥50%, proximal left anterior descending artery stenosis ≥70%, prior MI, endocarditis, prior stroke or transient ischemic attack, carotid stenosis, prior peripheral arterial disease, current/recent smoker, diabetes, New York Heart Association (NYHA) functional class IV, atrial fibrillation or flutter, conduction defect, severe chronic lung disease, home oxygen, hostile chest, porcelain aorta, access site (femoral vs. other), pacemaker or previous implantable converter-defibrillator, prior revascularization, prior cardiac operations (≥2 vs. 1 vs. 0), prior aortic valve procedure, prior nonaortic valve procedure, aortic etiology (degenerative vs. other), valve morphology (tricuspid vs. other), aortic insufficiency (moderate/severe vs. other), mitral insufficiency (moderate/severe vs. other), tricuspid insufficiency (moderate/severe vs. other), acuity (elective vs. urgent vs. shock or inotropes or assist device vs. emergency or salvage or cardiac arrest), device size, triple vessel disease, and body mass index.
During the study period from November 2011 to September 2014, a total of 23,652 patients underwent TAVR, including 11,808 (49.9%) female patients and 11,844 (51.1%) male patients. Table 1 displays the baseline characteristics for male and female patients. Female patients were somewhat older than male patients, with a lower prevalence of prior percutaneous (29.5% vs. 41.9%) or surgical (16.4% vs. 46.1%) coronary revascularization, atrial fibrillation (38.94% vs. 42.7%), diabetes (35.0% vs. 39.5%), lung disease (12.9% vs. 14.6%), and lower GFR (61.2 vs. 63.3 ml/min) but higher prevalence of porcelain aorta (7.7% vs. 6.0%) (p < 0.001 for all). Female patients had a higher mean left ventricular ejection fraction (56.7% vs. 50.6%) compared with male patients but a greater prevalence of moderate or severe mitral valve regurgitation (31.5% vs. 26.4%) and NYHA functional class III or IV heart failure (82.4% vs. 80.3%) (p < 0.001 for all). The median 5-m walk test was longer in female patients (8.7 vs. 7.3 s; p < 0.001). The mean STS mortality score was 9% for female patients and 8% for male patients (p < 0.001).
TAVR was mostly performed for degenerative aortic valve disease (94.5%); 3.2% of patients overall had severe aortic insufficiency and 1.9% had bicuspid aortic stenosis.
Reasons for TAVR
Table 2 provides the heart team–derived reasons for performing TAVR (Figure 1) and procedure-related characteristics. Although an equal number of male and female patients were considered to be at prohibitive risk for surgery, more female patients were considered to be debilitated or deconditioned by the heart team (13.7% vs. 9.1%). Both male and female patients commonly underwent TAVR for a 30-day STS mortality risk score >10% (29.9% in male patients vs. 28.3% in female patients). Approximately 10% of cases were deemed inoperable for technical reasons in both sexes.
General anesthesia was used in the majority (96.7%) of cases, and most procedures were undertaken electively (90.6%). Procedural inotropes were used in nearly one-half of all patients (45.9% in male patients and 43.6% in female patients; p < 0.0001).
Female subjects underwent TAVR more often using nontransfemoral access (∼45.0% vs. 35.0%) (Figure 2) and surgical cutdown (36.8% vs. 32.4%) compared with male patients. The median sheath size was 22.0 F (range, 22 to 25 F) in female patients compared with 24.0 F (range, 20 to 26 F) in male patients. Female patients received smaller valve sizes than male patients (65.0% vs. 11.6% received 23-mm devices; 5.4% vs. 12.7% received 29-mm devices). In the overall cohort, the majority of the devices used during the study period were older generation devices, and balloon-expandable TAVR was more often used than self-expanding TAVR (87.0% vs. 13.0%), without any differences according to sex. Female patients more often than male patients achieved valve cover index ≥8% (65.7% vs. 53.9%); the reported incidence of residual severe aortic incompetence was similar (3.4% vs. 3.1%).
Discharge antithrombotic medications
The use of aspirin (88.7% vs. 87.1%; p = 0.0028), P2Y12 receptor inhibitors (63.8% vs. 62.0%; p = 0.02), and dual antiplatelet therapy (57.4% vs. 55.1%; p = 0.0004) was marginally higher in male patients than in female patients, which was consistent with greater rate of prior revascularization in male patients (Table 1). Conversely, although more male patients had baseline atrial fibrillation than female patients, the prevalence of warfarin (25.0% vs. 25.5%) or non–vitamin K oral anticoagulant (3.4% vs. 3.2%) use was similar in both groups. The combination of a P2Y12 inhibitor and an oral anticoagulant was less often used in female patients (9.1% vs. 7.4%; p < 0.0001).
Procedural and in-hospital outcomes
There was no difference in the rate of device success, post-implant aortic valve gradient, or post-implant aortic valve area in either group. Complications related to device implantation were rare (Table 3) but more frequent in female patients. Female patients were more likely to have the procedure aborted due to access related or navigation issues (1.4% vs. 1.0%). Similarly, conversion to open surgery (1.7% vs. 1.0%) and use of emergent cardiopulmonary bypass (2.9% vs. 2.2%) was more frequent in female patients. The incidence of aortic dissection (5 in male patients vs. 22 in female patients), annulus rupture (12 vs. 28), ventricular rupture (15 vs. 45), and coronary obstruction (2 vs. 19) requiring conversion to open surgery was extremely low but occurred more commonly in female patients than in male patients (Figure 3). Device embolization needing surgical conversion was rare, and male patients more often than female patients had device embolization in the left ventricle (0.4% vs. 0.2%). Urgent implantation of a second TAVR valve was more often required in male patients (3.0% vs. 2.1%).
Female patients had a higher incidence of in-hospital vascular complications (8.3% vs. 4.4%; adjusted HR: 1.70; 95% CI: 1.34 to 2.14; p < 0.001) and a trend for more bleeding (8.0% vs. 6.0%; adjusted HR: 1.19; 95% CI: 0.99 to 1.44; p = 0.06) (Table 4, Central Illustration). There was no difference in the incidence of death, MI, stroke, or MACE. The incidence of in-hospital NACE was numerically higher in female patients compared with male patients (19.0% vs. 13.8%; adjusted HR: 1.14; 95% CI: 0.99 to 1.30; p = 0.06).
Periprocedural event rates were also examined according to year, and they generally decreased with time in both sexes (data not shown).
1-Year clinical outcomes
Table 5, Central Illustration, and Figure 4 show the 1-year outcomes in male patients and female patients. At 1 year, female patients demonstrated higher survival (78.7% vs. 75.5%; adjusted HR: 0.73; 95% CI: 0.63 to 0.85; p < 0.001) and lower MACE (25.3% vs. 28.1%; adjusted HR: 0.80; 95% CI: 0.70 to 0.92; p = 0.0012), composite death or MI (22.7% vs. 26.2%; adjusted HR: 0.74; 95% CI: 0.64 to 0.85; p < 0.001), and death or stroke (24.2% vs. 26.6%; adjusted HR: 0.80; 95% CI: 0.70 to 0.92; p < 0.001) compared with male patients. No differences were observed in the incidence of 1-year clinically significant bleeding.
The current analysis presents the largest observational report on comparative outcomes in male and female patients undergoing TAVR. The main findings are as follows: 1) although male and female patients have comparable prevalence of TAVR for significant aortic valve disease, female patients present with fewer comorbidities but a different risk profile compared with male patients, including older age, frailty, higher prevalence of porcelain aorta, moderate to severe mitral regurgitation, lower GFR, and higher STS score; 2) female patients undergo nontransfemoral TAVR more often and have a higher incidence of device-related coronary obstruction and conversion to open surgery than male patients (despite smaller TAVR device sizes, female patients achieved a more optimal valve cover index than male patients); 3) the adjusted risk of in-hospital vascular complications was higher in female patients, and there was a trend for greater in-hospital bleeding and NACE without significant differences in death or stroke; and 4) at 1 year, female patients demonstrated higher survival compared with male patients with no differences in stroke or clinically significant bleeding.
Sex-specific baseline risk profiles
Although findings on the baseline vascular comorbidities in female patients are in line with results from previous studies (4,5,11,16), they highlight the fact that female patients have a distinctive risk profile compared with male patients, with more frequent co-existent moderate or severe mitral valve disease, NYHA functional class III/IV symptoms, porcelain aorta, lower body surface area, and GFR. Indeed, moderate or severe mitral valve regurgitation and NYHA functional class IV have been shown to be independently associated with 1-year survival (17). We also observed that heart teams deemed female patients more often than male patients to be frail. However, a universal and simple frailty score continues to evade clinicians, and studies conducting formal frailty assessments have not shown sex differences in frailty status of TAVR subjects (18–21). These findings underscore the need for rigorous TAVR-specific and female-specific risk assessments in decision-making for the most optimal treatment option in patients with significant aortic valve disease. Although early TAVR was offered exclusively to prohibitive and extreme-risk patients, use in the future is expected to expand to intermediate-risk and lower risk patients (22,23). Given the high proportion of female patients requiring aortic valve intervention and the relatively high perception of frailty, it is paramount to appropriately identify those female patients who may benefit from TAVR.
Procedural characteristics and outcomes
In contrast with a recent sex-based meta-analysis (4), we found that female patients underwent TAVR via nontransfemoral access more often than male patients, which is likely a reflection of available sheath sizes and center preference during early experience. Notwithstanding, access selection for TAVR can significantly affect device delivery and device-related complications (24,25). For the first time, we show that female patients had a greater incidence of coronary obstruction and a higher rate of conversion to open surgery, for reasons including annular, aortic, and ventricular rupture (4). Male patients, conversely, experienced a greater incidence of valve embolization and the need for a second TAVR implant. Female patients not only have smaller annuli and left ventricular outflow tract dimensions but also shorter distances from the coronary ostia to the annulus and greater prevalence of aortic calcification (26–28) and horizontal aorta, which may explain some of these complications. Conversely, the smaller annuli may have allowed for a more optimal cover index. Moreover, although we noted more frequent failure to advance the device in female patients, this finding may be due to smaller peripheral vessels (4) and greater likelihood of horizontal aorta compared with male patients. Such anatomical differences may account for disparities in our results compared with studies conducted primarily with smaller, more advanced devices and delivery systems.
Short- and longer-term outcomes
Aligned with previous data, we found that female patients experienced more vascular complications and a trend for greater bleeding than male patients, despite smaller sheath sizes. The plausible reasons for this outcome are smaller body mass index, frailty, sheath–to–femoral artery ratio >1 in small peripheral vessels (29), or differential rate of closure device success, which warrants detailed evaluation. Moreover, vascular complications and bleeding may not only be related to ipsilateral valve access site but may also be nonaccess site related (6) or indeed related to the contralateral/complementary arterial or venous access sites, which has not been separately examined. Notably, the incidence of major bleeding in this registry was lower than that observed elsewhere, possibly stemming from different study populations and endpoint definitions (4,5,23). Recent data from the TVT Registry have indicated decreasing trends in vascular complications and bleeding, consequent to new device iterations with smaller and more flexible delivery systems and evolving center experience (30).
Despite these differences in early outcomes between male and female patients in our study, survival was similar in both sexes at the time of discharge and higher in female patients at 1 year. As other studies have inferred, this outcome may be a consequence of several factors: fewer traditional clinical risks in female patients at baseline (e.g., lower coronary artery disease, atrial fibrillation, higher left ventricular ejection fraction [4,5] and a better valve “fit”) as we have shown, with higher cover index after valve implantation. Female patients may also have earlier remodeling and left ventricular regression after TAVR compared with male patients, as indicated in surgical aortic valve replacement analyses (31,32); these findings may have a favorable effect on rehospitalizations (33) and survival. Moreover, vascular complications have recently been shown to have a much lower impact on 30-day mortality than bleeding, suggesting that vascular complications may be amenable to prompt and successful management in the majority (4). This scenario may provide an explanation for the higher 1-year survival in female patients despite greater vascular complications. However, further study is needed on the medium- to long-term causes of death in these patients to evaluate these hypotheses. Device advancements and 3-dimensional printing technology promise to improve device-related outcomes in both male and female patients (34). Dedicated studies on vascular biology and patterns of aortic valvular and annular calcification will inform TAVR practice to further reduce the already decreasing rates of vascular complications. Concurrently, randomized studies on the optimal discharge pharmacotherapy will guide future strategies to reduce out-of-hospital and longer term bleeding events (GALILEO [Global Study Comparing a Rivaroxaban-based Antithrombotic Strategy to an Antiplatelet-based Strategy After Transcatheter Aortic Valve Replacement to Optimize Clinical Outcomes] study; NCT02556203).
Our findings are based on observational registry data, which were internally validated by sites but not centrally adjudicated. One-year data were drawn from claims records, and although this method may result in some underestimation of events, information from these data sources can enable improvement in health care and population management (35,36). Although we adjusted for potential confounders, there may be unmeasured confounders accounting for biased reporting of 1-year survival. The majority of the devices implanted during the study period were balloon-expandable in nature. We did not examine for differences in outcomes according to TAVR time period or adjust for device types. It is possible that sex-based differences in vascular complications are attenuated with contemporary technologies. Device and technical improvements should allow for reduction in the early complications and greater uptake of this procedure in female patients.
Female patients undergoing TAVR for significant aortic valve disease have a different risk profile compared with male patients. Notwithstanding a greater adjusted risk for in-hospital vascular complications, 1-year survival was superior in female patients compared with male patients.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: Women undergoing TAVR generally have lower prevalence of coronary artery disease than men but a higher prevalence of porcelain aorta, moderate or severe mitral regurgitation, and higher STS scores. In a cohort treated mainly with older generation devices in the United States, female patients more often underwent TAVR via nonfemoral access and experienced more severe procedure-related complications, yet had better adjusted survival at 1 year and a risk of stroke similar to that in male patients.
TRANSLATIONAL OUTLOOK: A comparative sex-based analysis in warranted in patients undergoing TAVR with newer generation devices to examine for temporal changes in procedural complications and long-term outcomes.
For a list of codes derived from U.S. Centers for Medicare & Medicaid Services claims data, please see the online version of this article.
Dr. Vemulapalli has received research grant support (significant) from Abbott Vascular; and consulting fees from Novella. Dr. Mehran has received research grant support from Eli Lilly/DSI, AstraZeneca, The Medicines Company, Bristol-Myers Squibb, OrbusNeich, Bayer, and CSL Behring; has received consulting fees from Janssen Pharmaceuticals Inc., Medscape, Osprey Medical Inc., and Watermark Research Partners (modest level [< $5,000/year]); and has served on the scientific advisory board of Abbott Laboratories. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Molly Szerlip, MD, served as Guest Editor for this paper.
- Abbreviations and Acronyms
- confidence interval
- Centers for Medicare & Medicaid Services
- glomerular filtration rate
- hazard ratio
- major adverse cardiac events
- myocardial infarction
- net adverse cardiac events
- New York Heart Association
- odds ratio
- Society of Thoracic Surgeons/American College of Cardiology
- transcatheter aortic valve replacement
- Transcatheter Valve Therapy
- Valve Academic Research Consortium
- Received April 27, 2016.
- Revision received September 26, 2016.
- Accepted October 3, 2016.
- American College of Cardiology Foundation
- O'Connor S.A.,
- Morice M.C.,
- Gilard M.,
- et al.
- Genereux P.,
- Cohen D.J.,
- Mack M.,
- et al.
- Genereux P.,
- Webb J.G.,
- Svensson L.G.,
- et al.
- Nashef S.A.,
- Roques F.,
- Hammill B.G.,
- et al.
- STS. Online STS Adult Cardiac Risk Surgery Calculator. Available at: http://riskcalc.sts.org/stswebriskcalc. Accessed April 1, 2016.
- Ad N.,
- Barnett S.D.,
- Speir A.M.
- Hayashida K.,
- Morice M.C.,
- Chevalier B.,
- et al.
- Humphries K.H.,
- Toggweiler S.,
- Rodes-Cabau J.,
- et al.
- Edwards F.H.,
- O’Brien D.J.,
- Peterson S.M.,
- et al.
- Williams M.,
- Kodali S.K.,
- Hahn R.T.,
- et al.
- Green P.,
- Woglom A.E.,
- Genereux P.,
- et al.
- Rodes-Cabau J.,
- Mok M.
- Chandrasekhar J.,
- Mehran R.
- Leon M.B.,
- Smith C.R.,
- Mack M.J.,
- et al.
- Chandrasekhar J.,
- Hibbert B.,
- Ruel M.,
- et al.
- Mack M.J.
- Buellesfeld L.,
- Stortecky S.,
- Kalesan B.,
- et al.
- Hayashida K.,
- Lefevre T.,
- Chevalier B.,
- et al.
- Holmes D.R. Jr..,
- Nishimura R.A.,
- Grover F.L.,
- et al.
- Clavel M.A.,
- Webb J.G.,
- Rodes-Cabau J.,
- et al.
- Petrov G.,
- Regitz-Zagrosek V.,
- Lehmkuhl E.,
- et al.
- Lindman B.R.,
- Stewart W.J.,
- Pibarot P.,
- et al.
- Psaty B.M.,
- Delaney J.A.,
- Arnold A.M.,
- et al.