ORIGINAL RESEARCH Facilitated Data Relay and Effects on Treatment of Severe Aortic Stenosis in Europe Richard P. Steeds, MD;* Matthias Lutz, MD;* Jeetendra Thambyrajah, MD; Antonio Serra, MD; Eberhard Schulz, MD; Jiri Maly, MD; Marco Aiello, MD; Tanja K. Rudolph, MD; Guy Lloyd, MD; Alessandro Santo Bortone, MD; Karl Eugen Hauptmann, MD; Alberto Clerici, MD; Georg Delle-Karth, MD; Johannes Rieber, MD; Ciro Indolfi, MD; Massimo Mancone, MD; Loic Belle, MD; Alexander Lauten, MD; Martin Arnold, MD; Berto J. Bouma, MD; Cornelia Deutsch, MD; Jana Kurucova, MD; Martin Thoenes, MD; Peter Bramlage, MD; Norbert Frey, MD; David Messika-Zeitoun, MD Background-—Many patients with severe aortic stenosis are referred late with advanced symptoms or inappropriately denied intervention. The objective was to investigate whether a structured communication to referring physicians (facilitated data relay) might improve the rate and timeliness of intervention. Methods and Results-—A prospective registry of consecutive patients with severe aortic stenosis at 23 centers in 9 European countries with transcatheter as well as surgical aortic valve replacement being available was performed. The study included a 3- month documentation of the status quo (phase A), a 6-month intervention phase (implementing facilitated data relay), and a 3- month documentation of a legacy effect (phase-B). Two thousand one hundred seventy-one patients with severe aortic stenoses were enrolled (phase A: 759; intervention: 905; phase-B: 507). Mean age was 77.910.0 years, and 80% were symptomatic, including 52% with severe symptoms. During phase A, intervention was planned in 464/696 (67%), 138 (20%) were assigned to watchful waiting, 8 (1%) to balloon aortic valvuloplasty, 60 (9%) were listed as not for active treatment, and in 26 (4%), no decision was made. Three hundred sixty-three of 464 (78%) patients received the planned intervention within 3 months. Timeliness of the intervention improved as shown by the higher number of aortic valve replacements performed within 3 months (59% versus 51%, P=0.002) and a significant decrease in the time to intervention (3638 versus 3033 days, P=0.002). Conclusions-—A simple, low-cost, facilitated data relay improves timeliness of treatment for patients diagnosed with severe aortic stenosis, resulting in a shorter time to transcatheter aortic valve replacement. This effect was mainly driven by a significant improvement in timeliness of intervention in transcatheter aortic valve replacement but not surgical aortic valve replacement. Clinical Trial Registration-—URL: https://www.clinicaltrials.gov/. Unique identifier: NCT02241447. ( J Am Heart Assoc. 2019;8: e013160. DOI: 10.1161/JAHA.119.013160.) Key Words: Quality of care • aortic stenosis • transcatheter aortic valve implantation • surgical aortic valve replacement • facilitated data relay A ortic stenosis (AS) is a progressive, life-threatening valve disorder that is characterized by a rapid increase in risk of sudden death once symptoms appear.1 Despite clear is evidence that gaps between current recommendations and actual clinical practice exist.6–9 Patients with severe symp- tomatic AS may be referred late with advanced symptoms or American and European guidelines on management,2–5 there inappropriately denied intervention.7,8,10 Patients managed in an inappropriate or untimely way incur excess morbidity and mortality, not only because of the risk of “death on the waiting A list of the authors’ affiliations has been placed at the end of this article. list” but also from increased perioperative complications and Accompanying Figures S1 through S3 are available at https://www.ahajourna incomplete recovery.11,12 It is therefore important to identify ls.org/doi/suppl/10.1161/JAHA.119.013160 ways of improving the timeliness of healthcare delivery for *Dr Steeds and Dr Lutz are co-first authors. Correspondence to: Richard P. Steeds, MD, Queen Elizabeth Hospital, patients with AS. Mindelsohn Way, Birmingham B15 2TT, UK. E-mail: rick.steeds@uhb.nhs.uk Lack of awareness of the condition and of its conse- Received May 3, 2019; accepted August 21, 2019. quences may represent a potentially important reason for the ª 2019 The Authors. Published on behalf of the American Heart Association, undertreatment observed in patients with valvular heart Inc., by Wiley. This is an open access article under the terms of the Creative disease. In a recent community-based study,13 although Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, echocardiography correctly identified the disease and its the use is non-commercial and no modifications or adaptations are made. severity, a significant proportion of the patients were left DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 1 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH the independent ethics review board at each participating Clinical Perspective institution, and patient informed consent was obtained. The data that support the findings of this study are available from What Is New? the corresponding author upon reasonable request. • Many patients with severe aortic stenosis are referred late The study had 3 phases: baseline observation phase A, to with advanced symptoms or inappropriately denied inter- document the management of patients with severe AS vention. enrolled during a 3-month period; FDR intervention phase, • A simple, low-cost, facilitated data relay improves timeliness to evaluate the efficacy of the FDR intervention among of treatment for patients diagnosed with severe aortic patients enrolled during the subsequent 6-month period; and stenosis, resulting in a shorter time to transcatheter aortic valve replacement. follow-up observation phase B, to evaluate any legacy effect of the FDR intervention among patients enrolled during a final 3- What Are the Clinical Implications? month period (Figure 1). During the intervention phase, the study nurse contacted the referring physician within 7 days to • Facilitated data relay has the potential to reduce perioper- implement the FDR. During the 2 observation phases, there ative and postoperative morbidity, while decreasing the was no contact with the referring physicians. Follow-up for all number of patients dying on the waiting list for aortic valve replacement. patients was capped at 3 months, when details of any • Future studies are now needed to formally demonstrate the treatment during the period were collected. impact of facilitated data relay on outcome. Patients untreated. We then collected pilot data that replicated these Patients over 18 years of age were included in the registry findings in hospitals (secondary and tertiary care), indicating based on a new finding at echocardiography of native severe that a significant proportion of patients with symptomatic AS, whether symptomatic or asymptomatic. A diagnosis of severe AS were not being referred for consideration of aortic severe AS was defined as 1 or more of the following findings valve intervention in a timely fashion. Simple, low-cost on echocardiography: an aortic valve area of <1 cm2 (com- facilitated data relay (FDR) interventions may help. We puted using a continuity equation), an indexed aortic valve hypothesized that using a dedicated nurse to inform referring area of <0.6 cm2/m2, a maximum jet velocity of >4 m/s, or a physicians of a finding of severe AS in a prompt and mean transvalvular gradient of >40 mm Hg.15 Patients were structured manner might increase the rate of referral of excluded if they had prior aortic valve intervention. patients for intervention and timeliness thereof.14 Intervention: FDR Methods A study nurse was used at each center to monitor echocar- diography results. During the first observational phase, the Study Design and Site Selection communication between the hospitals and referring physi- IMPULSE was a prospective, multinational registry of patients cians relied on usual practice in each center (eg, via a with severe AS in Europe, the rationale and design of which standard report or electronic discharge letter provided to the has been described recently.14 In brief, IMPULSE evaluated office-based physician). This letter commonly summarizes any the effect of a nurse-led FDR on the timing and rate of valve clinical findings, may give recommendations for further procedures. A total of 23 centers from 9 countries (Austria, investigations but is not highlighted, may potentially not be Czech Republic, France, Germany, Italy, The Netherlands, reviewed in a timely fashion, and there may be no mechanism Spain, Switzerland, and the United Kingdom) were involved, in place to record appropriate action. with patients enrolled between March 2015 and April 2017. During the intervention phase, the nurse contacted the The sites were selected on the basis that each offered the full relevant referring physician in addition to the aforementioned range of treatment options for AS, including surgical and measures for each center and about 1 week after a patient was transcatheter procedures, the rationale being that these enrolled, to inform them that their patient had been diagnosed might provide the best picture of service delivery. Although with severe AS on echocardiography (FDR) and had consented the 2012 guidelines applied during the conduct of the study, to participation in a study assessing follow-up of patients with the results were considered against the background of the severe AS. The FDR outlined the class 1 indications for 2017 European Society of Cardiology15 and 2017 American intervention in severe AS and gave all possible management College of Cardiology guidelines.3 The study was carried out in options, without recommending a specific technique (tran- accordance with the Declaration of Helsinki, was approved by scatheter aortic valve replacement [TAVR] or surgical aortic DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 2 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH Figure 1. Facilitated data relay. FU (follow-up) indicates 3 months follow-up period where effects of a potential intervention are collected and performance/nonperformance of an intervention are documented; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. valve replacement [SAVR]) over one another. Physicians and/or were defined as the presence of Canadian Cardiovascular patients were contacted again 3 months later to collect details Society class III or IV angina, New York Heart Association of the actual treatment given. Contact was made via a functional class III or IV, and/or dizziness on exertion/syncope. structured email, fax, or when necessary, by telephone call. A The logistic EuroScore I and the EuroScore II were calculated as standard format was used for all communications. an indication of surgical risk. Frailty was assessed according to the ability of the patient to walk 5 meters in <6 s and to perform activities of daily living.16 Activities of daily living and Data Collection life expectancy were assessed by the dedicated nurses or Data collected at baseline for all patients included demograph- physicians, but no specific list of activities of daily living or risk ics, medical history, and symptoms (chest pain, shortness of calculator was recommended. The results of the echocardio- breath, and dizziness on exertion/syncope). Severe symptoms graphic assessment were recorded, including the presence of Figure 2. Patient flow across phases. FU (follow-up) indicates 3 months follow-up period where effects of a potential intervention are collected and performance/nonperformance of an intervention are documented. Full FU refers to data available. DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 3 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH coexisting aortic regurgitation, mitral or tricuspid valve disease; the rate of planned and performed SAVR or TAVR within 3 transvalvular gradient; left ventricle dimensions; and left months and the time to intervention. ventricular ejection fraction. At 3 months after enrollment, information on vital status Statistical Analysis (alive/dead), treatment decisions (SAVR, TAVR, balloon Data are presented descriptively, using means with SD, aortic valvuloplasty [BAV], active decision not to treat, or medians with interquartile ranges, or absolute values with watchful waiting/further diagnostics) and the number of percentages. Comparisons between patients enrolled during interventions performed were documented for all patients the FDR phase or observation phase B and observation phase enrolled into the study. Watchful waiting was defined as the A were made using a Pearson v2 or Fisher exact test for scheduling of further patient follow-up. Data were entered categorical variables, and a t test, Mann–Whitney–Wilcoxon into a standardized electronic case report form by a rank sum test, or ANOVA for continuous variables. A P value dedicated study nurse. The primary outcome measure was of <0.05 was considered statistically significant. Statistical Table 1. Patient Characteristics Total (n=2171) Observation A (n=759) Facilitated Data Relay (n=905) Observation B (n=507) P Value for Comparison MeanSD or n/N (%) MeanSD or n/N (%) MeanSD or n/N (%) MeanSD or n/N (%) Across Phases Age, y 77.910.0 77.610.2 78.09.5 78.110.5 0.520 Female sex 1041/2171 (48) 360/759 (47) 436/905 (48) 245/507 (48) 0.938 2 BMI, kg/m 27.45.3 27.45.8 27.24.8 27.75.5 0.258 Symptoms Chest pain 487/2076 (24) 166/726 (23) 201/873 (23) 120/477 (25) 0.606 Shortness of breath 1576/2123 (74) 548/744 (74) 645/888 (73) 383/491 (78) 0.084 Dizziness on exertion/syncope 471/1941 (24) 157/679 (23) 210/835 (25) 104/427 (24) 0.657 NYHA class III or IV 859/2142 (40) 315/755 (42) 338/900 (38) 206/487 (42) 0.120 Angina CCS class III or IV 91/1901 (5) 45/676 (7) 29/788 (4) 17/437 (4) 0.018 Symptomatic* 1743/2171 (80) 605/759 (80) 717/905 (79) 421/507 (83) 0.199 Severe symptoms 1122/2171 (52) 402/759 (53) 455/905 (50) 265/507 (52) 0.526 Echocardiographic valve-related parameters Aortic valve area, cm2 0.730.2 0.740.2 0.730.2 0.720.2 0.323 Indexed aortic 0.400.1 0.400.1 0.400.1 0.390.1 0.312 valve area, cm/m2 Maximum jet velocity, m/s 4.30.7 4.30.7 4.30.6 4.30.7 0.724 Mean transvalvular 47.114.7 47.215.0 47.614 46.214.7 0.237 gradient, mm Hg LVH (>12 mm thick) 1324/2132 (62) 490/758 (65) 560/902 (62) 274/472 (58) 0.068 EF, % 55.812 56.012 56.512 54.411 0.008 >50% 1492/2054 (73) 520/720 (72) 634/854 (74) 338/480 (70) 0.245 30–50% 496/2054 (24) 174/720 (24) 191/854 (22) 131/480 (27) <30% 66/2054 (3) 26/720 (4) 29/854 (3) 11/480 (2) PAP, mm Hg 39.413 39.413 40.113 38.213 0.098 Frailty (severe)† 110/2141 (5) 55/755 (7) 33/899 (4) 22/487 (5) 0.003 Surgical risk Logistic EuroScore I, % 15.613.9 15.913.8 16.314.5 14.112.9 0.157 Logistic EuroScore II, % 4.05.0 3.95.0 4.14.7 4.15.4 0.710 Comparisons were analyzed using Pearson v2 or Fisher exact test for categorical variables, and ANOVA for continuous variables. BMI indicates body mass index; CCS, Canadian Cardiovascular Society; EF, ejection fraction; LVH, left ventricular hypertrophy; NYHA, New York Heart Association; PAP, pulmonary artery pressure. *Defined as 1 or more cardiac symptoms presumably related to severe aortic stenosis (chest pain, shortness of breath, dizziness on exertion/syncope). † Defined as inability to perform 2 or more activities of daily life. DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 4 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH analysis was performed using SPSS Version 23.0 (IBM Corp., 127 patients died (death rate 5.9%) including postoperative Armonk, NY). death. The mean age of the overall (n=2171) study population was 77.910.0 years, and 80% had symptomatic AS including 52% with severely limiting symptoms. Echocardiography Results revealed a mean aortic valve area of 0.730.2 cm2, mean indexed aortic valve area of 0.400.1 cm/m2, mean maxi- Patient Characteristics mum jet velocity of 4.30.7 m/s, and mean transvalvular A total of 2171 patients with severe AS were enrolled (759 in gradient of 47.114.7 mm Hg. The mean EuroScore II was observation phase A, 905 in the FDR intervention phase, and 4.05.0%. 507 in observation phase B; Figure 2). Of those, 1562 patients Patients enrolled in the different study phases had similar were referred by cardiologists, 273 by general physicians, 111 demographic and disease characteristics, apart from a slightly by cardiac surgeons, 32 by geriatricians, and 192 by other higher proportion of patients with angina pectoris Canadian physicians (1 unknown). During the study, 44 patients were Cardiovascular Society class III/IV and severe frailty in lost to follow-up (data completeness at 3 months 98.0%) and observation phase A, and a slightly lower mean ejection Figure 3. Proportion of AVR (total), TAVR or SAVR planned (A) and performed (B) in all patients AND planned (C) and performed (D) in symptomatic patients within 3 months during the 3 study phases. AVR indicates aortic valve replacement; FDR, facilitated data relay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 5 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH fraction among patients enrolled in observation phase B (Table 1). Specifically, the rate of symptomatic patients and patients presenting with severe symptoms was not different across the different phases. Rate and Timeliness of Intervention for AS During observation phase A, which was designed to demon- strate the status quo at each participating hospital before the introduction of FDR (baseline), an AVR was planned in 464/ 696 (67%) patients (Figure 3A planned; Figure 3B performed). Of these 464, the plan was for the majority to undergo TAVR (n=306) and fewer to undergo SAVR (n=158). Of the remaining patients identified with severe AS, 138 (20%) were considered suitable for watchful waiting, 8 (1%) were planned for balloon aortic valvuloplasty, 60 (9%) were listed as not for active treatment, and in 26 (4%), no decision was made at all. Data for the subgroup of symptomatic patients are displayed in Figure 3C and 3D. In total, 363/464 of patients (78% of the overall population) received the planned intervention within 3 months at a mean of 3638 days in total (Figure 4A) and 3637 days in symptomatic patients (Figure 4C). Effect of FDR During the intervention (FDR) phase B, there was a trend toward an increase in the number of AVRs planned compared with the observation phase, although this did not reach statistical significance (583/825, 71% versus 464/696, 67%, P=0.093; Figure 3A). There was no correlation between the timeliness of the FDR and the actual date of the intervention (Pearson r=0.090). There were also no differences between contact methods (phone, email/fax, or letter) for the time to intervention except for those who were referred from a physician within the hospital (P<0.001). FDR did not alter the ratio of patients listed for TAVR as opposed to SAVR. However, timeliness of the intervention improved as shown by the higher number of AVR performed within 3 months (59% versus 51%, P=0.002; Figure 3B) and a significant decrease in the time to intervention by 6 days (3638 versus 3033 days, P=0.002) (Figure 4A). Interestingly, this effect was mainly driven by a significant improvement in timeliness of intervention in TAVR but not in SAVR (Figures 3 and 4). Of interest, the effect of FDR was seen whether patients were symptomatic or not (Figure 3B and 3D). Figure 4. Time to intervention or planned date according to We performed a time-to-event analysis for the overall AVR treatment option and study phase in patients with or without symptoms (A and B) and in symptomatic patients (C). FDR group as well as for those undergoing TAVI and SAVR, indicates facilitated data relay; SAVR, surgical aortic valve respectively. The time-to-intervention (if we consider it an replacement; TAVR, transcatheter aortic valve replacement. event) was significantly reduced for the overall AVR analysis when the intervention phase was compared with the obser- 0.257, respectively) (Figure 4B). If the time-to-event analysis vational phase A with a P=0.002 using a log-rank and <0.001 is repeated for the TAVI and SAVR groups, the results using the Breslow test. This was seen when comparing reproduce the prior findings using the t test. While there is a observational phase B to observational phase A (P=0.188 and significant effect of FDR on the rate of TAVI (P=0.011 log-rank DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 6 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH P=0.001 Breslow), there is no such effect for SAVR alone differences. First, the percentage of cases planned (Figure S1) (P=0.066 log-rank, P=0.134 Breslow). Comparisons of obser- and then actually performed (Figure S2) within 3 months vational phase B to A are again not statistically significant. differs between countries, with a much higher percentage of definitive treatment delivered in a timely fashion (Figure S3) FDR: A Legacy Effect? within 3 months in Germany compared with the United Kingdom. Second, sequentially excluding 1 country and Once FDR was discontinued, there was little “legacy effect.” restricting analysis to the 4 remaining countries, much of the Comparing observation phase B with observation phase A, effect of FDR appears to be delivered by analyses that include there was neither a difference in the number of AVRs planned the United Kingdom and the impact loses significance when (332/479, 69% versus 464/696, 67%) nor in the number of the United Kingdom was excluded. Third, there was a strong AVR performed within 3 months (54.9% versus 51.3%, increase in the rate of AVR performance in Italy, which further P=0.223) (Table 2 and Figure 3B). Overall, time to interven- increased in the second observational phase for AVR and SAVR tion was not different between phase B and phase A (3433 in particular. versus 3638 days, P=0.816) (Figure 4A). Facilitated Data Relay: An International Discussion Comparison In patients diagnosed with severe AS on echocardiography, 1 Results were then investigated to determine whether there in 5 patients wait >3 months from time of diagnosis to was a difference in effect of FDR between countries. Within intervention. This delay to intervention occurred despite >80% each of the countries studied, there was an increase in the of patients being symptomatic, with the majority of these number of AVR procedures performed (Table 3), and, as a having severe limitations, classified as New York Heart result, a consistent reduction in the time to intervention Association or Canadian Cardiovascular Society class III or (Table 4). Within each single country, there was no significant IV. Moreover, the delay to intervention occurred in hospitals in difference in the effect of FDR, presumably because of lower which both SAVR and TAVR were available on-site, where one power. The data from individual countries, however, together might expect intervention to occur in a timelier fashion. A with sensitivity analysis revealed 3 important further simple, low cost, and easily applicable FDR intervention Table 2. Treatment Actually Performed After 3 Months Observation A (n=696) FDR (n=825) Observation B (n=479) n (%) n (%) P vs Phase A n (%) P vs Phase A Total AVR performed within 3 mo 357 (51.3) 487 (59.0) 0.002 263 (54.9) 0.223 If AVR not performed Date of AVR set 41 (5.9) 32 (3.9) 0.067 23 (4.8) 0.419 Date of AVR not set 66 (10) 64 (8) 0.231 46 (10) 0.945 TAVR TAVR performed within 3 mo 239 (34.3) 334 (40.5) 0.014 173 (36.1) 0.530 If TAVR not performed Date of TAVR set 29 (4.2) 17 (2.1) 0.017 14 (2.9) 0.264 No date of TAVR set 38 (5.5) 31 (3.8) 0.112 24 (5.0) 0.735 SAVR SAVR performed within 3 mo 118 (17.0) 153 (19) 0.419 90 (18.8) 0.418 If SAVR not performed Date of SAVR set 12 (1.7) 15 (2) 0.890 9 (1.9) 0.844 No date of SAVR set 28 (4) 33 (4) 0.982 22 (4.6) 0.634 Comparisons were analyzed using Pearson v2 or Fisher exact test. AVR indicates aortic valve replacement; BAV, balloon aortic valvuloplasty; FDR, facilitated data relay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. Number of patients in whom BAV was performed within 3 months: 5/696 (0.7%) during observation phase A; 6/825 (0.7%) during intervention (facilitated data relay) phase; and 7/479 (1.5%) during observation phase B. DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 7 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH Table 3. Treatment Planned, and Treatment Actually Performed After 3 Months: Country-Specific Analysis Observation A Facilitated Data Relay Observation B Total n (%) n (%) P vs Phase A n (%) P vs Phase A Germany n=520 n=191 n=209 n=120 AVR Planned 472 (90.8) 169 (88.5) 191 (91.4) 0.333 112 (93.3) 0.158 Performed 449 (86.3) 164 (85.9) 181 (86.6) 0.830 104 (86.7) 0.842 TAVR Planned 390 (75.0) 137 (71.7) 153 (73.2) 0.741 100 (83.3) 0.019 Performed 374 (71.9) 134 (70.2) 147 (70.3) 0.969 93 (77.5) 0.156 SAVR Planned 82 (15.8) 32 (16.8) 38 (18.2) 0.707 12 (10.0) 0.096 Performed 75 (14.4) 30 (15.7) 34 (16.3) 0.878 11 (9.2) 0.097 UK n=478 n=178 n=114 n=186 AVR Planned 247 (51.7) 92 (51.7) 62 (54.4) 0.652 93 (50.0) 0.748 Performed 135 (28.2) 48 (27.0) 42 (36.8) 0.075 45 (24.2) 0.544 TAVR Planned 143 (29.9) 51 (28.7) 41 (36.0) 0.189 51 (27.4) 0.794 Performed 83 (17.4) 27 (15.2) 29 (25.4) 0.030 27 (14.5) 0.861 SAVR Planned 104 (21.8) 41 (23.0) 21 (18.4) 0.347 42 (22.6) 0.918 Performed 52 (10.9) 21 (11.8) 13 (11.4) 0.918 18 (9.7) 0.513 France n=356 n=107 n=202 n=47 AVR Planned 271 (76.1) 83 (77.6) 150 (74.3) 0.520 38 (80.9) 0.648 Performed 241 (67.7) 71 (66.4) 133 (65.8) 0.928 37 (78.7) 0.123 TAVR Planned 150 (42.1) 42 (39.3) 83 (41.1) 0.754 25 (53.2) 0.108 Performed 133 (37.4) 33 (30.8) 76 (37.6) 0.235 24 (51.1) 0.017 SAVR Planned 121 (34.0) 41 (38.3) 67 (33.2) 0.366 13 (27.7) 0.202 Performed 108 (30.3) 38 (35.5) 57 (28.2) 0.186 13 (27.7) 0.340 Italy n=327 n=108 n=145 n=74 AVR Planned 242 (74.0) 77 (71.3) 105 (72.4) 0.845 60 (81.1) 0.133 Performed 198 (60.6) 51 (47.2) 91 (62.8) 0.014 56 (75.7) <0.001 TAVR Planned 169 (51.7) 60 (55.6) 78 (53.8) 0.781 31 (41.9) 0.070 Performed 133 (40.7) 38 (35.2) 68 (46.9) 0.062 27 (36.5) 0.857 SAVR Planned 73 (22.3) 17 (15.7) 27 (18.6) 0.550 29 (39.2) <0.001 Performed 65 (19.9) 13 (12.0) 23 (15.9) 0.389 29 (39.2) <0.001 Continued DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 8 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH Table 3. Continued Observation A Facilitated Data Relay Observation B Total n (%) n (%) P vs Phase A n (%) P vs Phase A Others n=319 n=112 n=155 n=52 AVR Planned 147 (46.1) 43 (38.4) 75 (48.4) 0.105 29 (55.8) 0.037 Performed 84 (26.3) 23 (20.5) 40 (25.8) 0.317 21 (40.4) 0.008 TAVR Planned 47 (14.7) 16 (14.3) 27 (17.4) 0.492 4 (7.7) 0.230 Performed 23 (7.2) 7 (6.3) 14 (9.0) 0.405 2 (3.8) 0.720 SAVR Planned 100 (31.3) 27 (24.1) 48 (31.0) 0.218 25 (48.1) 0.002 Performed 61 (19.1) 16 (14.3) 26 (16.8) 0.582 19 (36.5) 0.001 Comparisons were analyzed using Pearson v2 or Fisher exact test. No adjustment for multiple testing was made, because the analyses are meant to explore the country-specific effect, but with lower statistical power compared with the main analysis. AVR indicates aortic valve replacement; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. resulted in a greater proportion undergoing treatment within service.21,22 One study showed that symptoms were detected 3 months, mainly because of faster delivery of TAVR. FDR did earlier among patients with severe AS followed up in a not affect the rate of AVR. Delay to delivery of treatment structured valve clinic compared with those referred from differs between the countries included, with fewer receiving other services, which facilitated optimized timing of surgery.23 treatment within 3 months of a decision in the United Such services do not have to be physician-led. For example, Kingdom and a particular improvement with FDR in Italy. the introduction of a sonographer-led heart valve clinic Delay between the diagnosis of severe AS and an increased the proportion of patients with valvular disease appropriate intervention is responsible for patients dying managed according to best-practice guidelines from 41% to while waiting for an intervention and can increase perioper- 92%.24 In our study, a simple communication by a nurse ative and late complications.11,12 Given the impact of unduly resulted in faster delivery of treatment in centers that had delayed intervention, the introduction of methods such as FDR fully capable facilities available and which one would expect that do reduce delays in management should be mandated. should already deliver the highest quality of care. A nurse was This is particularly important in patients with severe AS who used in the study; however, the FDR communication itself have advanced disease at a stage when full recovery can no highlighted the diagnosis, the indications for intervention, and longer be guaranteed.6–8,17 In the current study, 52% of asked for options regarding management, and so could patients had severely limiting symptoms and 27% had an potentially be carried out automatically or by nonmedical ejection fraction of ≤50% at the time of enrollment. Since the staff. All healthcare systems are currently under financial prognosis is poor for such patients with severe AS who do not constraint and FDR is a low-cost intervention that was shown receive appropriate treatment,18–20 a failure to deliver out- in this study to improve timeliness of care in AS. come-modifying treatment in 1 in 5 patients within 3 months There is evidence that this type of quality improvement suggests that there is an urgent need to improve quality of strategy, which involves the transmission of clinical data using care. It is also noteworthy that consistently across all phases a means other than an existing automated system, is effective of the study, no decision was made regarding management of in other medical conditions.25 A systematic review of quality- these high-risk patients in 3% to 4% of cases. Given the risk of improvement strategies for the management of hypertension sudden cardiac death, irreversible myocardial damage, and found that FDR of clinical data was associated with median development of congestive heart failure, which directly impact reductions in systolic and diastolic blood pressure of 8.0 and outcome, any intervention that can speed up management 1.8 mm Hg, and a 25% increase in the proportion of patients pathways should lead to improved outcomes, and in large achieving target systolic blood pressure.26 The results of the populations, may reduce mortality. current study indicate that an FDR strategy can increase the There is a consensus that the best care for patients with rate of delivery of TAVR to patients with severe AS. A trend valvular heart disease is provided by a specialist heart valve towards an increase in the number of surgical AS DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 9 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH Table 4. Time to Intervention: Country-Specific Analysis Observation A FDR Observation B Total MeanSD MeanSD P vs Phase A MeanSD P vs Phase A Total AVR 33.034.7 36.4537.6 29.933.0 0.002 34.033.0 0.816 TAVR 29.234.2 35.539.1 24.831.2 <0.001 28.430.5 0.182 SAVR 40.734.5 38.334.2 40.434.4 0.798 44.535.1 0.091 Germany AVR 24.629.2 23.625.7 23.130.2 0.119 28.532.1 0.347 TAVR 21.428.2 22.225.6 17.428.4 0.004 26.330.6 0.373 SAVR 40.329.1 29.525.7 47.125.9 0.003 47.739.0 0.072 UK AVR 50.941.5 54.245.5 47.035.4 0.615 50.441.8 0.606 TAVR 53.242.9 60.550.9 50.337.8 0.605 47.937.3 0.410 SAVR 47.239.2 45.035.0 39.328.6 0.667 53.847.8 0.968 France AVR 22.026.2 22.222.1 22.929.5 0.242 18.320.0 0.209 TAVR 20.725.8 23.420.9 21.529.8 0.078 14.415.5 0.053 SAVR 23.626.7 21.223.3 24.729.3 0.965 25.525.4 0.672 Italy AVR 35.934.7 54.844.7 29.130.2 <0.001 26.115.6 0.005 TAVR 35.135.1 53.544.0 26.727.3 <0.001 23.820.1 <0.007 SAVR 37.733.9 59.148.4 35.837.3 0.071 28.39.3 0.115 Others AVR 62.834.8 65.341.3 60.433.7 0.886 64.928.1 0.820 TAVR 60.641.0 72.258.8 53.925.4 0.786 50.520.5 0.693 SAVR 63.732.4 61.528.9 63.436.8 0.668 66.328.8 0.676 Comparisons were analyzed using Mann–Whitney U test. No adjustment for multiple testing was made, because the analyses are meant to explore the country-specific effect, but with lower statistical power compared with the main analysis. AVR indicates aortic valve replacement; FDR, facilitated data relay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. interventions being pursued was also seen during the FDR TAVR may be more flexible because it requires fewer phase, although this did not achieve statistical significance. resources and allows for a higher rate of AVR performance This study provides proof of concept and offers the possibility and a substantial shortening of the time to intervention. While that similar benefit could be gained by automated communi- an increase in TAVR performance capacity may contribute to cations containing the same information and sent to the this, it is unlikely to have a played a role in our setting referrer (for example, by e-mail generated by the echocardio- because enrollment in each institution was limited to grapher or the machine itself), to minimize delays. 3 months for observation A, 6 months for FDR, and 3 months The reasons for the difference in the effect of FDR on the for observation B. The fact that there are differences between rate of delivery of TAVR and SAVR are unknown but could countries included in this study suggest that other factors possibly relate to the increased awareness of the role of TAVR may also have an impact, which could include issues of in patients who are at high surgical risk or for whom surgery is logistics and infrastructure that appear to be much less of a contraindicated (such as very elderly patients with comor- problem in Germany and Italy than in the United Kingdom. All bidities). A potential explanation is that there is a fixed the centers involved in this study had the capability of capacity for SAVR in a given hospital and merely adding delivering SAVR and TAVR, and it is not known whether the patients to the wait list does not result in a higher throughput. effect of FDR might be different in centers without direct On the contrary, the potential turnover of patients undergoing access to these interventions. Given the adverse effect that DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 10 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH delays can have on outcomes for patients with severe AS (as content. All authors gave final approval of the version to be illustrated by the mortality rate of 6% within 3 months noted published. across each phase of the study), these results suggest that additional measures may be necessary to further reduce delays in the delivery of intervention.27 Sources of Funding There were limitations to the IMPULSE registry. As an This work was supported by a research grant provided by observational, cross-sectional study, the outcomes after valve Edwards Lifesciences (Nyon, Switzerland) to the Sponsor, the intervention beyond 3 months were not recorded. Further- Institute for Pharmacology and Preventive Medicine (IPPMed, more, although treatment decisions were documented, the Cloppenburg, Germany). actual treatment that each patient underwent may have differed. Finally, despite a total number of patients of n=2171, Disclosures the study was not powered to detect a difference in outcomes Bramlage has received research funding for the present such as mortality. Future study at the population level will project and honoraria for consultancy from Edwards determine whether FDR altered appropriateness of interven- Lifesciences. Frey, Steeds, and Messika-Zeitoun have tion. The strengths of the IMPULSE registry include its received honoraria for advisory board meetings and Rudolph prospective design and that it is the largest prospective speakers’ honoraria from Edwards Lifesciences. The insti- registry to date that documents clinical characteristics and tutions of these 3 and those of the remaining authors management of contemporary patients. representing study centers have received funding for employing a study nurse. The remaining authors have no disclosures to report. Conclusions In conclusion, a significant proportion of patients diagnosed with severe AS on echocardiography wait >3 months from Authors’ Affiliations time of diagnosis to intervention. This delay to intervention From the Queen Elizabeth Hospital & Institute of Cardiovas- occurred despite the majority being symptomatic and mostly cular Sciences, University of Birmingham, UK (R.P.S.); with severe limitation. A simple, low-cost FDR process may Department of Cardiology and Angiology, University of Kiel, improve treatment pathways for patients diagnosed with Germany (M.L., N.F.); James Cook Hospital, Middlesbrough, severe AS, by increasing the rate of intervention performed UK (J.T.); Interventional Cardiology Unit, Hospital de la Santa within 3 months and decreasing the time to TAVR. Although Creu i Sant Pau, Barcelona, Spain (A.S.); Cardiology Depart- the effect of FDR was seen, the data from this study also ment I, University Clinic Mainz, Germany (E.S.); Department of highlighted differences in the delivery of aortic valve inter- Cardiovascular Surgery, Institute for Clinical and Experimental vention to patients, with many more experiencing delay in the Medicine, Prague, Czech Republic (J.M.); Department of United Kingdom compared with Germany. Although FDR did Cardiovascular Surgery, Second Faculty of Medicine, Charles not alter the overall rate of intervention, this process could be University, Prague, Czech Republic (J.M.); Department of improved and has the potential to reduce perioperative and Cardiothoracic Surgery, Foundation IRCCS Policlinico S. postoperative morbidity, while decreasing the number of Matteo, Pavia, Italy (M.A.); Department of Cardiology, Univer- patients dying on the waiting list for AVR. Future studies are sity of Cologne Heart Center, Cologne, Germany (T.K.R.); St now needed to formally demonstrate the impact of FDR on Bartholomew’s Hospital, London, UK (G.L.); University of Bari, outcome. Italy (A.S.B.); Krankenhaus der Barmherzigen Br€uder Trier, Germany (K.E.H.); University of Turin, Italy (A.C.); 4th Medical Acknowledgments Department, Hietzing Hospital, Vienna, Austria (G.D.-K.); Herzkatheterlabor Nymphenburg and Department of Cardiol- Data were captured using the s4trials Software provided by Software for Trials Europe GmbH, Berlin, Germany. ogy, University of Munich, Germany (J.R.); Division of Cardi- ology and URT CNR of IFC, University Magna Graecia, Catanzaro, Italy (C.I.); Sapienza University of Rome, Italy (M.M.); Centre Hospital d’Annecy, France (L.B.); German Author Contributions Centre for Cardiovascular Research (DZHK), University Heart Frey, Steeds, Messika-Zeitoun, Kurucova, Thoenes, and Bram- Center & Charite, Berlin, Germany (A.L.); Department of lage were involved in the conception and design of the study. Cardiology, University Hospital Erlangen, Germany (M.A.); Steeds, Lutz, and Bramlage drafted the manuscript and all University of Amsterdam, Netherlands (B.J.B.); Institute for other authors revised the article for important intellectual Pharmacology and Preventive Medicine, Cloppenburg, DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 11 Timeliness of Treatment for Severe AS Steeds et al ORIGINAL RESEARCH Germany (C.D., P.B.); Edwards Lifesciences, Nyon, Switzerland 12. Lund O, Nielsen TT, Emmertsen K, Flo C, Rasmussen B, Jensen FT, Pilegaard HK, Kristensen LH, Hansen OK. Mortality and worsening of prognostic profile (J.K., M.T.); University of Ottawa Heart Institute, Ottawa, during waiting time for valve replacement in aortic stenosis. Thorac Cardiovasc Canada (D.M.-Z.). Surg. 1996;44:289–295. 13. Dziadzko V, Clavel MA, Dziadzko M, Medina-Inojosa JR, Michelena H, Maalouf J, Nkomo V, Thapa P, Enriquez-Sarano M. Outcome and undertreatment of mitral regurgitation: a community cohort study. Lancet. 2018;391:960–969. References 14. Frey N, Steeds RP, Serra A, Schulz E, Baldus S, Lutz M, Pohlmann C, Kurucova 1. Eveborn GW, Schirmer H, Heggelund G, Lunde P, Rasmussen K. The evolving J, Bramlage P, Messika-Zeitoun D. Quality of care assessment and improve- epidemiology of valvular aortic stenosis. The Tromso study. Heart. ment in aortic stenosis—rationale and design of a multicentre registry 2013;99:396–400. (IMPULSE). BMC Cardiovasc Disord. 2017;17:5. 2. Baumgartner H, Falk V, Bax JJ, De Bonis M, Hamm C, Holm PJ, Iung B, 15. Baumgartner H, Hung J, Bermejo J, Chambers JB, Edvardsen T, Goldstein S, Lancellotti P, Lansac E, Rodriguez Munoz D, Rosenhek R, Sjogren J, Tornos Lancellotti P, LeFevre M, Miller F Jr, Otto CM. Recommendations on the Mas P, Vahanian A, Walther T, Wendler O, Windecker S, Zamorano JL. 2017 echocardiographic assessment of aortic valve stenosis: a focused update from ESC/EACTS Guidelines for the management of valvular heart disease. Eur the European Association of Cardiovascular Imaging and the American Society Heart J. 2017;38:2739–2791. of Echocardiography. J Am Soc Echocardiogr. 2017;30:372–392. 3. Otto CM, Kumbhani DJ, Alexander KP, Calhoon JH, Desai MY, Kaul S, Lee JC, 16. Katz S, Downs TD, Cash HR, Grotz RC. Progress in development of the index of Ruiz CE, Vassileva CM. 2017 ACC expert consensus decision pathway for ADL. Gerontologist. 1970;10:20–30. transcatheter aortic valve replacement in the management of adults with 17. Iung B, Messika-Zeitoun D, Cachier A, Delahaye F, Baron G, Tornos P, Gohlke- aortic stenosis: a report of the American College of Cardiology Task Force on Barwolf C, Boersma E, Ravaud P, Vahanian A. Actual management of patients Clinical Expert Consensus Documents. J Am Coll Cardiol. 2017;69:1313–1346. with asymptomatic aortic valve disease: how practice fits with guidelines. Am 4. Vahanian A, Alfieri O, Andreotti F, Antunes MJ, Baron-Esquivias G, Baumgartner Heart J. 2007;153:696–703. H, Borger MA, Carrel TP, De Bonis M, Evangelista A, Falk V, Lung B, Lancellotti 18. Schwarz F, Baumann P, Manthey J, Hoffmann M, Schuler G, Mehmel HC, P, Pierard L, Price S, Schafers HJ, Schuler G, Stepinska J, Swedberg K, Schmitz W, Kubler W. The effect of aortic valve replacement on survival. Takkenberg J, Von Oppell UO, Windecker S, Zamorano JL, Zembala M. Circulation. 1982;66:1105–1110. Guidelines on the management of valvular heart disease (version 2012): 19. Horstkotte D, Loogen F. The natural history of aortic valve stenosis. Eur Heart the Joint Task Force on the Management of Valvular Heart Disease of J. 1988;9(Suppl E):57–64. the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS). Eur J Cardiothorac Surg. 2012;42: 20. Rosenhek R, Zilberszac R, Schemper M, Czerny M, Mundigler G, Graf S, S1–S44. Bergler-Klein J, Grimm M, Gabriel H, Maurer G. Natural history of very severe aortic stenosis. Circulation. 2010;121:151–156. 5. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP III, Guyton RA, O’Gara PT, Ruiz CE, Skubas NJ, Sorajja P, Sundt TM III, Thomas JD, Members 21. Lancellotti P, Rosenhek R, Pibarot P, Iung B, Otto CM, Tornos P, Donal E, AATF. 2014 AHA/ACC guideline for the management of patients with valvular Prendergast B, Magne J, La Canna G, Pierard LA, Maurer G. ESC Working heart disease: a report of the American College of Cardiology/American Heart Group on Valvular Heart Disease position paper–heart valve clinics: organi- Association Task Force on Practice Guidelines. Circulation. 2014;129: zation, structure, and experiences. Eur Heart J. 2013;34:1597–1606. e521–e643. 22. Chambers JB, Ray S, Prendergast B, Taggart D, Westaby S, Grothier L, Arden C, 6. Iung B, Baron G, Butchart EG, Delahaye F, Gohlke-Barwolf C, Levang OW, Wilson J, Campbell B, Sandoe J, Gohlke-Barwolf C, Mestres CA, Rosenhek R, Tornos P, Vanoverschelde JL, Vermeer F, Boersma E, Ravaud P, Vahanian A. A Otto C. Specialist valve clinics: recommendations from the British Heart Valve prospective survey of patients with valvular heart disease in Europe: the Society working group on improving quality in the delivery of care for patients Euro Heart Survey on Valvular Heart Disease. Eur Heart J. 2003;24: with heart valve disease. Heart. 2013;99:1714–1716. 1231–1243. 23. Zilberszac R, Lancellotti P, Gilon D, Gabriel H, Schemper M, Maurer G, Massetti 7. Freed BH, Sugeng L, Furlong K, Mor-Avi V, Raman J, Jeevanandam V, Lang RM. M, Rosenhek R. Role of a heart valve clinic programme in the management of Reasons for nonadherence to guidelines for aortic valve replacement in patients with aortic stenosis. Eur Heart J Cardiovasc Imaging. 2017;18:138– patients with severe aortic stenosis and potential solutions. Am J Cardiol. 144. 2010;105:1339–1342. 24. Taggu W, Topham A, Hart L, Carr-White G, Sulke N, Patel NR, Lloyd G. A 8. Heuvelman HJ, van Geldorp MW, Kappetein AP, Geleijnse ML, Galema TW, cardiac sonographer led follow up clinic for heart valve disease. Int J Cardiol. Bogers AJ, Takkenberg JJ. Clinical course of patients diagnosed with severe 2009;132:240–243. aortic stenosis in the Rotterdam area: insights from the AVARIJN study. Neth 25. Shojania KG, Ranji SR, Shaw LK, Charo LN, Lai JC, Rushakoff RJ, McDonald KM, Heart J. 2012;20:487–493. Owens DK. Closing the Quality Gap: A Critical Analysis of Quality Improvement 9. Chan RH, Shaw JL, Hauser TH, Markson LJ, Manning WJ. Guideline adherence Strategies (Vol. 1: series overview and methodology). Rockville, MD: Agency for for echocardiographic follow-up in outpatients with at least moderate valvular Healthcare Research and Quality (US); 2004. Available at: https://www.ncbi. disease. J Am Soc Echocardiogr. 2015;28:795–801. nlm.nih.gov/books/NBK43908/. 10. Ahmed A, Sorajja P, Garberich RF, Farivar RS, Harris KM, Gossl M. Association 26. Walsh JM, McDonald KM, Shojania KG, Sundaram V, Nayak S, Lewis R, Owens of guideline adherence for serial evaluations with survival and adverse clinical DK, Goldstein MK. Quality improvement strategies for hypertension manage- events in patients with asymptomatic severe aortic stenosis. JAMA Cardiol. ment: a systematic review. Med Care. 2006;44:646–657. 2017;2:1141–1146. 27. Wijeysundera HC, Wong WW, Bennell MC, Fremes SE, Radhakrishnan S, 11. Malaisrie SC, McDonald E, Kruse J, Li Z, McGee EC Jr, Abicht TO, Russell H, Peterson M, Ko DT. Impact of wait times on the effectiveness of transcatheter McCarthy PM, Andrei AC. Mortality while waiting for aortic valve replacement. aortic valve replacement in severe aortic valve disease: a discrete event Ann Thorac Surg. 2014;98:1564–1570; discussion 1570–1561. simulation model. Can J Cardiol. 2014;30:1162–1169. DOI: 10.1161/JAHA.119.013160 Journal of the American Heart Association 12 SUPPLEMENTAL MATERIAL Figure S1. Country specific analysis - Proportion of AVR (total), TAVR or SAVR planned within 3 months during the three study phases. AVR, aortic valve replacement; FDR, facilitated data relay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. Figure S2. Country specific analysis - Proportion of AVR (total), TAVR or SAVR performed within 3 months during the three study phases. AVR, aortic valve replacement; FDR, facilitated data relay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement. Figure S3. Country specific analysis (Exclusion of countries) – Time to intervention. FDR, facilitated data relay; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement