Download PDF - Circulation: Cardiovascular Interventions

Structural Heart Disease
Thirty-Day Readmissions After Transcatheter Aortic
Valve Replacement in the United States
Insights From the Nationwide Readmissions Database
Dhaval Kolte, MD, PhD; Sahil Khera, MD; M. Rizwan Sardar, MD;
Neil Gheewala, MD, MPH; Tanush Gupta, MD; Saurav Chatterjee, MD;
Andrew Goldsweig, MD; Wilbert S. Aronow, MD; Gregg C. Fonarow, MD;
Deepak L. Bhatt, MD, MPH; Adam B. Greenbaum, MD; Paul C. Gordon, MD; Barry Sharaf, MD;
J. Dawn Abbott, MD
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Background—Readmissions after cardiac procedures are common and contribute to increased healthcare utilization and
costs. Data on 30-day readmissions after transcatheter aortic valve replacement (TAVR) are limited.
Methods and Results—Patients undergoing TAVR (International Classification of Diseases-Ninth Revision-CM codes 35.05
and 35.06) between January and November 2013 who survived the index hospitalization were identified in the Nationwide
Readmissions Database. Incidence, predictors, causes, and costs of 30-day readmissions were analyzed. Of 12 221 TAVR
patients, 2188 (17.9%) were readmitted within 30 days. Length of stay >5 days during index hospitalization (hazard ratio
[HR], 1.47; 95% confidence interval [CI], 1.24–1.73), acute kidney injury (HR, 1.23; 95% CI, 1.05–1.44), >4 Elixhauser
comorbidities (HR, 1.22; 95% CI, 1.03–1.46), transapical TAVR (HR, 1.21; 95% CI, 1.05–1.39), chronic kidney disease
(HR, 1.20; 95% CI, 1.04–1.39), chronic lung disease (HR, 1.16; 95% CI, 1.01–1.34), and discharge to skilled nursing
facility (HR, 1.16; 95% CI, 1.01–1.34) were independent predictors of 30-day readmission. Readmissions were because
of noncardiac causes in 61.8% of cases and because of cardiac causes in 38.2% of cases. Respiratory (14.7%), infections
(12.8%), bleeding (7.6%), and peripheral vascular disease (4.3%) were the most common noncardiac causes, whereas heart
failure (22.5%) and arrhythmias (6.6%) were the most common cardiac causes of readmission. Median length of stay and
cost of readmissions were 4 days (interquartile range, 2–7 days) and $8302 (interquartile range, $5229–16 021), respectively.
Conclusions—Thirty-day readmissions after TAVR are frequent and are related to baseline comorbidities, TAVR access
site, and post-procedure complications. Awareness of these predictors can help identify and target high-risk patients for
interventions to reduce readmissions and costs. (Circ Cardiovasc Interv. 2017;10:e004472. DOI: 10.1161/CIRCINTERVENTIONS.116.004472.)
Key Words: aortic stenosis ◼ costs and cost analysis ◼ length of stay ◼ readmission ◼ rehospitalization
◼ transcatheter aortic valve implantation ◼ transcatheter aortic valve replacement
T
in 2015, ≈17.8% of Medicare patients were still readmitted
to the hospital within 30 days.2 Unplanned readmissions are
associated with increased healthcare expenditure and costs.
Previous research has shown that readmission rates are influenced by patient and hospital characteristics, quality of inpatient and outpatient care, and local practice patterns.3
hirty-day readmission rates are considered a quality
performance measure. The Centers for Medicare and
Medicaid Services publicly reports 30-day readmission rates
for cardiovascular diagnoses and procedures such as acute
myocardial infarction, heart failure, percutaneous coronary intervention (PCI; as a pilot project between 2013 and
2014), and coronary artery bypass grafting.1 Initiatives such
as the Hospital Readmissions Reduction Program under the
Affordable Care Act have led to a significant decline in allcause 30-day readmission rates over the past few years.2 Yet,
See Editorial by Swaminathan and Rao
Since its initial Food and Drug Administration approval
in November 2011, the number of transcatheter aortic valve
Received September 3, 2016; accepted November 14, 2016.
From the Division of Cardiology, Brown University, Providence, RI (D.K., M.R.S., A.G., P.C.G., B.S., J.D.A.); Division of Cardiology, New York
Medical College, Valhalla (S.K., W.S.A.); Division of Cardiology, Northeast Ohio Medical University, Aultman Hospital, Canton (M.R.S.); Henry Ford
Hospital, Detroit, MI (N.G., A.B.G.); Division of Cardiology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY (T.G.); Division
of Cardiology, Temple University Hospital, Philadelphia, PA (S.C.); Division of Cardiology, University of California at Los Angeles (G.C.F.); and Division
of Cardiology, Brigham and Women’s Hospital Heart & Vascular Center and Harvard Medical School, Boston, MA (D.L.B.).
Guest Editor for this article was Sunil V. Rao, MD.
The Data Supplement is available at http://circinterventions.ahajournals.org/lookup/suppl/doi:10.1161/CIRCINTERVENTIONS.116.004472/-/DC1.
Correspondence to J. Dawn Abbott, MD, Division of Cardiology, Department of Medicine, Warren Alpert Medical School of Brown University, 593
Eddy St, RIH APC814, Providence, RI 02903. E-mail [email protected]
© 2016 American Heart Association, Inc.
Circ Cardiovasc Interv is available at http://circinterventions.ahajournals.org
1
DOI: 10.1161/CIRCINTERVENTIONS.116.004472
2 Kolte et al Thirty-Day Readmissions After TAVR
WHAT IS KNOWN
• Thirty-day,
all-cause readmission rates after transcatheter aortic valve replacement range from 14.6%
to 20.9%.
• Approximately 60% of the 30-day readmissions after
transcatheter aortic valve replacement are because of
noncardiac causes.
WHAT THE STUDY ADDS
• Length of stay >5 days during the index hospitaliza-
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
tion, acute kidney injury, presence of >4 Elixhauser
comorbidities, transapical access, chronic kidney
disease, chronic lung disease, and discharge to
skilled nursing facility are independent predictors of
30-day readmission after transcatheter aortic valve
replacement.
• Cost of readmission accounts for an average of
16.4% of the total cost of the episode of care in patients who are readmitted.
replacement (TAVR) procedures performed in the United
States has increased rapidly.4 TAVR is now approved for
patients with severe symptomatic aortic stenosis who are at
prohibitive-risk, high-risk, or intermediate-risk for surgical
aortic valve replacement (SAVR). This patient population is at
an increased risk for readmissions inherent to the high burden
of comorbidities and high rate of periprocedural complications. Thirty-day all-cause readmission rates after TAVR have
ranged from 14.6% to 20.9% in previous studies.5–8 However,
data on independent predictors of readmissions after TAVR in
a nationally representative cohort are lacking. Furthermore,
none of the previous studies provide information on costs of
readmission episodes, which has important healthcare policy implications and is critical in assessing the overall costeffectiveness of TAVR. Thus, the main objectives of our study
were to determine the incidence, predictors, causes, and costs
of 30-day readmissions after TAVR in a real-world, nationally representative patient population in the United States.
Knowledge of the independent predictors of 30-day readmissions can help identify and target high-risk patients for interventions to reduce rehospitalization and healthcare costs.
Methods
Data Source
We used the 2013 Nationwide Readmissions Database (NRD), a publicly available database of all-payer hospital inpatient stays developed
by the Agency for Healthcare Research and Quality as part of the
Healthcare Cost and Utilization Project. The NRD is drawn from the
State Inpatient Databases that contain reliable, verified patient linkage numbers that can be used to track a patient across hospitals within
a State, while adhering to strict privacy guidelines. The 2013 NRD
includes data from 21 states that are geographically dispersed and account for 49.3% of the total US resident population and 49.1% of all
hospitalizations. The NRD includes all discharge records of patients
treated in US community hospitals excluding rehabilitation and longterm acute care facilities. Discharge weights are provided to obtain
national estimates.
This study was deemed exempt by the Lifespan-Rhode Island
Hospital Institutional Review Board as the NRD is a publicly available database that contains deidentified patient information.
Study Population
We used the International Classification of Diseases, Ninth Edition,
Clinical Modification procedure codes 35.05 and 35.06 to identify all
TAVR patients aged ≥18 years included in the 2013 NRD (n=14 071
across 219 centers). We excluded records of patients discharged in
December 2013 (n=1239) because of unavailability of 30-day followup data on these cases. We further excluded patients who died during
the index hospitalization (n=602 [in-hospital mortality, 4.7%]) and
those with missing discharge disposition (n=2). Finally, if a patient
underwent repeat TAVR within 30 days (n=7), the discharge record
for the second procedure was excluded from the index cases to avoid
duplication (these records were included as 30-day readmissions).
For patients who underwent repeat TAVR after 30 days (n=14), the
second discharge record was considered as a separate index procedure. This gave us a final sample size of 12 221 index TAVR procedures performed at 210 centers.
Patient and Hospital Characteristics
Baseline patient characteristics included were age, sex, primary expected payer, median household income, and relevant comorbidities
(smoking, dyslipidemia, hypertension, diabetes mellitus, obesity,
heart failure, known coronary artery disease, previous myocardial
infarction, previous PCI, previous coronary artery bypass grafting,
carotid artery disease, previous transient ischemic attack/stroke,
atrial fibrillation, previous permanent pacemaker [PPM], previous
implantable cardioverter defibrillator, peripheral vascular disease
[PVD], anemia, chronic kidney disease [CKD], chronic lung disease
[CLD], liver disease, coagulopathy, dementia, depression, hypothyroidism, fluid and electrolyte disorders, other neurological disorders,
pulmonary circulation disorders, and cancer). Hospital characteristics
such as location, teaching status, and bed size were also included.
Other variables extracted were TAVR access site (endovascular versus transapical [TA]), in-hospital procedures (coronary angiography,
PCI, and mechanical circulatory support), and in-hospital complications (conversion to SAVR, complete heart block, PPM placement,
transient ischemic attack/stroke, acute myocardial infarction, cardiogenic shock, cardiac arrest, acute kidney injury [AKI], major bleeding, and vascular complications). The Healthcare Cost and Utilization
Project Clinical Classification Software (CCS) and International
Classification of Diseases, Ninth Edition, Clinical Modification
codes used to define these variables are listed in Table I in the Data
Supplement.
Outcomes Measured
Our primary outcome of interest for this study was 30-day, all-cause
readmission. Readmissions were identified according to the methodology outlined by Healthcare Cost and Utilization Project.9 For
patients who had multiple readmissions within 30 days, only the
first readmission was included. Transfer to another hospital was not
considered as a readmission. Both unplanned and planned readmissions were included. Planned readmissions after TAVR are expected
to be extremely rare because unlike PCI, there is no staged procedure involved. Time to readmission was calculated as the number of
days between hospital discharge after index TAVR procedure and
the first day of hospital readmission. Two authors (D.K. and S.K.)
independently reviewed the primary diagnosis of each readmission
record and grouped them into clinically meaningful categories to determine the main cause of readmission. Discrepancies were resolved
by mutual agreement. Causes of readmissions were classified as cardiac and noncardiac. Cardiac causes included heart failure, arrhythmias, conduction disorders, valve disorders including endocarditis,
prosthesis-related mechanical complications, and aortic valve reintervention, hypertension/hypotension, pericarditis, coronary artery
disease, acute myocardial infarction, and others. Noncardiac causes
included respiratory (including pneumonia), infectious, bleeding,
3 Kolte et al Thirty-Day Readmissions After TAVR
PVD, transient ischemic attack/stroke, renal, gastrointestinal, trauma,
hematologic/neoplasms, endocrine/metabolic, neuropsychiatric, and
others. Primary diagnosis categories and the corresponding CCS
and International Classification of Diseases, Ninth Edition, Clinical
Modification codes are listed in Table II in the Data Supplement.
Secondary outcomes examined were length of stay (LOS) and total
hospital costs. Costs were inflation adjusted using the US Bureau of
Labor Statistics Consumer Price Index, with 2016 as the index base.10
Statistical Analysis
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Baseline patient and hospital characteristics, in-hospital procedures,
and in-hospital complications were compared between patients with
and without 30-day readmission using Pearson χ2 test for categorical
variables and Student t test for continuous variables. Cox proportional
hazards regression was used to identify predictors of 30-day readmission. Variables with a P<0.2 on univariable analysis were entered into a
multivariable regression model (method=ENTER) to identify independent predictors of 30-day readmission. A random-effect term for hospital was also added to the model. Kaplan–Meier curve and log-rank
test were used to compare 30-day readmission rates between groups.
Data were complete for all variables except primary expected
payer (0.1% missing) and median household income (1.1% missing).
Missing values were replaced with the dominant category for these
variables. Data on cost was missing for 166 index hospitalizations
and 35 readmissions. Therefore, results of cost analysis are based on
a sample size of 12 055 index hospitalizations and 2153 readmissions.
Statistical analysis was performed with IBM SPSS Statistics 20.0
(IBM Corp, Armonk, NY). All P values were 2 sided with a significance threshold of P<0.05. Categorical variables are expressed as
percentages and continuous variables as mean±SD or median (interquartile range [IQR]) as appropriate. Hazard ratio (HR) and 95%
confidence interval (CI) are used to report the results of Cox regression analyses.
Results
Baseline Characteristics
Baseline patient and hospital characteristics, in-hospital procedures, and in-hospital outcomes of TAVR patients who survived the index hospitalization (n=12 221) are summarized in
Table 1. Mean age of the study cohort was 81.5±8.4 years with
49.1% women. The majority of the TAVR procedures were
performed in large, urban, teaching hospitals. Compared with
patients who were not readmitted within 30 days, those readmitted were older, had higher prevalence of diabetes mellitus,
previous PCI, atrial fibrillation, anemia, CKD, CLD, and fluid/
electrolyte disorders, and lower prevalence of dyslipidemia,
hypertension, previous coronary artery bypass grafting, and
carotid artery disease. Patients who were readmitted within
30 days were more likely to have undergone TA-TAVR, had a
higher incidence of certain in-hospital complications such as
PPM placement, cardiac arrest, AKI, and major bleeding, had
longer LOS and higher total hospital costs, and were more
likely to have been discharged to SNF (Table 1).
Incidence, Hospital Variation, and Predictors
of 30-Day Readmission
Of 12 221 TAVR patients who survived the index hospitalization, 2188 (17.9%) were readmitted within 30 days. Median
time to readmission was 10 days (IQR, 4–18 days). Figure
IA in the Data Supplement shows the distribution of hospitals
according to number of TAVR procedures performed during
the study period. For hospital-level variation in readmission
rates, we restricted our analyses to hospitals performing ≥5
TAVRs. Thirty-day readmission rates varied significantly
across hospitals (median, 17.6%; IQR, 11.8%–22.9%, and
range 0%–50%) (Figure IIA in the Data Supplement). Similar
results were seen when the analysis was restricted to hospitals
performing ≥10 TAVRs (median, 17.6%; IQR, 12.6%–23.4%,
and range 0%–50%; Figure IIB in the Data Supplement).
On multivariable analysis, LOS >5 days during the index
hospitalization (HR, 1.47; 95% CI, 1.24–1.73), AKI (HR,
1.23; 95% CI, 1.05–1.44), >4 Elixhauser comorbidities (HR,
1.22; 95% CI, 1.03–1.46), TA-TAVR (HR, 1.21; 95% CI,
1.05–1.39), CKD (HR, 1.20; 95% CI, 1.04–1.39), CLD (HR,
1.16; 95% CI, 1.01–1.34), and discharge to SNF (HR, 1.16;
95% CI, 1.01–1.34) were identified as independent predictors of 30-day readmission (Table 2; Figure III in the Data
Supplement).
Causes, LOS, and Costs of 30-Day Readmissions
Of 2188 readmissions, 1352 (61.8%) were because of noncardiac causes and 836 (38.2%) were because of cardiac causes.
Noncardiac causes included respiratory (14.7%), infections
(12.8%), bleeding (7.6%), PVD (4.3%), renal (4.0%), gastrointestinal (3.9%), transient ischemic attack/stroke (3.6%),
trauma (3.2%), hematologic/neoplasms (2.3%), endocrine/
metabolic (2.0%), neuropsychiatric (1.5%), and other (2.1%;
Figure 1A). Heart failure (22.5%), arrhythmias (6.6% [atrial
fibrillation 4.3%]), valve disorders including endocarditis,
prosthesis-related mechanical complications, and aortic valve
reintervention (2.1%), hypertension/hypotension (1.7%), conduction disorders (1.3%), coronary artery disease (0.9%),
pericarditis (0.8%), and acute myocardial infarction (0.7%)
were the most common cardiac causes of 30-day readmission
(Figure 1B). Specifically, 0.6% patients had aortic valve reintervention and 2.9% underwent PPM placement during the
readmission. Among patients readmitted within 30 days, 109
patients (5.0%) died.
The mean LOS for readmissions was 5.9±5.5 days with
a median of 4 days (IQR, 2–7 days). Total hospital cost of
all 30-day readmissions was $29 115 721. Mean total hospital
cost per readmission was $13 528±15 285 with a median of
$8302 (IQR, $5229–16 021; Figure 2). Cost of readmission
accounted for an average of 16.4%±12.2% of the total cost of
the episode of care (index+readmission).
Discussion
In this comprehensive analysis of 30-day all-cause readmissions after TAVR in the United States, we report several
important findings: (1) 17.9% of patients undergoing TAVR
were readmitted within 30 days, and readmission rates varied significantly across hospitals, (2) LOS >5 days during the
index hospitalization, AKI, presence of >4 Elixhauser comorbidities, TA-TAVR, CKD, CLD, and discharge to SNF were
independent predictors of 30-day readmission, (3) consistent
with previous data, more than three fifths of the readmissions
were because of noncardiac causes,8 and (4) cost of readmission accounted for 16.4% of the total cost of the episode of
care in patients who were readmitted.
Thirty-day readmissions are considered a quality performance measure by the Centers for Medicare and Medicaid
Services. Previous research has shown that 30-day readmission
4 Kolte et al Thirty-Day Readmissions After TAVR
Table 1. Baseline Characteristics and In-Hospital Outcomes of TAVR Patients
Discharged Alive After Index Hospitalization
30-d Readmission
Overall
No (n=10 033)
Yes (n=2188)
P Value
Age, y
81.5±8.4
81.3±8.4
82.0±8.3
0.001
<85
55.4%
56.1%
51.8%
<0.001
≥85
44.6%
43.9%
48.2%
49.1
48.9
50.0
Women, %
Primary expected payer
0.371
<0.001
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Medicare
91.6%
91.2%
93.5%
Medicaid
0.7%
0.7%
0.8%
Private
5.5%
5.8%
4.4%
Uninsured
0.7%
0.7%
0.7%
Other
1.5%
1.7%
0.6%
0–25th percentile
20.4%
20.5%
19.8%
26th–50th percentile
24.5%
24.5%
24.3%
51st–75th percentile
26.5%
26.4%
27.0%
76th–100th percentile
28.6%
28.6%
28.9%
Smoking
27.5%
27.7%
26.2%
0.133
Dyslipidemia
63.3%
64.2%
59.2%
<0.001
Hypertension
79.7%
80.4%
76.4%
<0.001
Diabetes mellitus
33.9%
33.5%
35.9%
0.027
Obesity
14.9%
14.6%
15.9%
0.146
Heart failure
72.6%
72.5%
73.3%
0.413
Known CAD
68.6%
68.7%
68.1%
0.558
Previous MI
12.8%
13.0%
11.9%
0.171
Previous PCI
20.1%
19.6%
22.7%
0.001
Previous CABG
22.7%
23.4%
19.5%
<0.001
Median household income
0.818
Comorbidities
Carotid artery disease
7.2%
7.4%
6.0%
0.021
Previous TIA/stroke
12.9%
13.0%
12.5%
0.580
Atrial fibrillation
45.9%
44.8%
50.7%
<0.001
Previous PPM
10.1%
10.2%
9.9%
0.737
Previous ICD
3.4%
3.5%
3.2%
0.516
Peripheral vascular disease
30.8%
30.6%
31.6%
0.355
Anemia
26.0%
25.5%
28.7%
0.002
Chronic kidney disease
35.3%
33.7%
42.7%
<0.001
Chronic lung disease
34.5%
33.6%
39.0%
<0.001
Liver disease
2.2%
2.1%
2.6%
0.166
Coagulopathy
24.4%
24.2%
25.5%
0.184
Dementia
7.8%
7.7%
8.5%
0.230
Depression
7.5%
7.7%
6.8%
0.165
Hypothyroidism
19.9%
19.7%
21.0%
0.149
Fluid and electrolyte disorders
29.7%
28.5%
35.4%
<0.001
Other neurological disorders
6.8%
6.6%
7.6%
0.105
Pulmonary circulation disorders
0.9%
0.9%
1.2%
0.129
Cancer
3.6%
3.7%
3.0%
0.126
(Continued )
5 Kolte et al Thirty-Day Readmissions After TAVR
Table 1. Continued
30-d Readmission
Overall
No (n=10 033)
Yes (n=2188)
P Value
No. of Elixhauser comorbidities
4.4±1.9
4.3±1.9
4.7±2.0
<0.001
≤4
54.6%
56.3%
46.7%
<0.001
>4
45.4%
43.7%
53.3%
Hospital characteristics
Bed size
0.039
Small
3.0%
2.8%
3.7%
Medium
12.6%
12.8%
11.7%
Large
84.4%
84.4%
84.5%
Location
0.040
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Rural
0.8%
0.7%
1.1%
Urban
99.2%
99.3%
98.9%
Nonteaching
11.3%
11.4%
10.8%
Teaching
88.7%
88.6%
89.2%
Endovascular
68.9%
70.4%
62.4%
Transapical
31.1%
29.6%
37.6%
Coronary angiography
25.5%
25.4%
25.8%
0.714
PCI
3.5%
3.4%
4.0%
0.125
Mechanical circulatory support
1.9%
1.8%
2.4%
0.076
Surgical AVR
0.3%
0.2%
0.3%
0.557
Complete heart block
7.8%
7.8%
8.1%
0.558
PPM placement
9.2%
8.8%
11.3%
<0.001
TIA/stroke
2.6%
2.6%
2.7%
0.821
AMI
2.8%
2.9%
2.5%
0.323
Cardiogenic shock
2.1%
2.2%
2.0%
0.658
Cardiac arrest
2.8%
2.5%
4.3%
<0.001
AKI
18.2%
16.4%
26.5%
<0.001
Major bleeding
17.1%
16.6%
19.5%
0.001
Teaching status
0.462
In-hospital procedures
TAVR access
<0.001
In-hospital outcomes
Vascular complications
5.0%
4.8%
5.7%
LOS, d
9.0±8.6
8.5±8.0
11.3±10.9
<0.001
≤5
39.2%
41.9%
27.0%
<0.001
>5
60.8%
58.1%
73.0%
Home (self-care)
29.3%
30.9%
22.1%
Short-term hospital
1.2%
1.2%
1.1%
Skilled nursing facility
32.0%
30.1%
40.7%
Home healthcare
37.5%
37.8%
36.2%
59 877±31 668
58 577±30 669
65 840±35 298
Discharge disposition
Total hospital cost, US$
0.109
<0.001
<0.001
AMI indicates acute myocardial infarction; AKI, acute kidney injury; AVR, aortic valve replacement; CABG,
coronary artery bypass grafting; CAD, coronary artery disease; ICD, implantable cardioverter defibrillator; LOS,
length of stay; MI, myocardial infarction; PCI, percutaneous coronary intervention; PPM, permanent pacemaker;
TAVR, transcatheter aortic valve replacement; TIA, transient ischemic attack; and US$, United States dollar.
6 Kolte et al Thirty-Day Readmissions After TAVR
Table 2. Independent Predictors of 30-Day Readmission After TAVR
Univariable
Multivariable
Predictors
HR (95% CI)
P Value
HR (95% CI)
P Value
LOS >5 d
1.91 (1.66–2.21)
<0.001
1.47 (1.24–1.73)
<0.001
Acute kidney injury
1.69 (1.46–1.95)
<0.001
1.23 (1.05–1.44)
0.011
>4 Elixhauser comorbidities
1.47 (1.29–1.66)
<0.001
1.22 (1.03–1.46)
0.026
Transapical TAVR
1.40 (1.23–1.60)
<0.001
1.21 (1.05–1.39)
0.008
Chronic kidney disease
1.41 (1.24–1.60)
<0.001
1.20 (1.04–1.39)
0.014
Chronic lung disease
1.23 (1.08–1.40)
0.002
1.16 (1.01–1.34)
0.034
Discharge to SNF
1.53 (1.34–1.74)
<0.001
1.16 (1.01–1.34)
0.038
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Cardiac arrest
1.60 (1.15–2.22)
0.005
Fluid and electrolyte disorders
1.40 (1.22–1.60)
<0.001
Major bleeding
1.29 (1.10–1.51)
0.002
Atrial fibrillation
1.23 (1.09–1.40)
0.001
Anemia
1.21 (1.05–1.39)
0.008
Heart failure
1.19 (1.03–1.38)
0.020
Age ≥85 y
1.14 (1.00–1.29)
0.045
Hypertension
0.86 (0.74–1.00)
0.044
Previous CABG
0.81 (0.69–0.96)
0.012
Dyslipidemia
0.80 (0.70–0.91)
0.001
CABG indicates coronary artery bypass grafting; CI, confidence interval; HR, hazard ratio; LOS, length of stay;
SNF, skilled nursing facility; and TAVR, transcatheter aortic valve replacement.
rates after cardiac procedures such as PCI, coronary artery
bypass grafting, and SAVR range from 12.4% to 19.2%.11–14
In our study of 12 221 patients who underwent TAVR between
January and November 2013, the 30-day all-cause readmission rate was 17.9%. Our results are comparable to data
from the Society of Thoracic Surgeons/American College of
Cardiology Transcatheter Valve Therapy Registry in which
30-day readmission rate was 17.4% among 12 182 Medicare
patients who underwent TAVR between November 2011 and
June 2013.6 Similarly, Murugiah et al7 and Nombela-Franco
et al8 reported 30-day readmission rates of 20.9% and 14.6%
after TAVR in cohorts of 14 722 Medicare fee-for-service beneficiaries and 720 patients at 2 centers, respectively.
TAVR is currently approved for patients with severe symptomatic aortic stenosis who are at prohibitive-risk, high-risk,
or intermediate-risk for SAVR. This patient population is at an
increased risk for readmissions inherent to the high burden of
comorbidities and high rate of periprocedural complications.
Indeed, the presence of >4 Elixhauser comorbidities, specifically CKD and CLD, was identified as an independent predictor of 30-day readmission after TAVR in our study. TAVR
has been shown to be at least as safe and effective as SAVR
in patients with CKD and CLD.15–18 However, patients with
CKD and CLD undergoing TAVR are at increased risk of procedural complications, mortality, and rehospitalization when
compared with those without these conditions.17,18
In addition to patient comorbidities, TA-TAVR and AKI
were also identified as independent predictors of 30-day
readmission. TA-TAVR is an accepted approach in patients
in whom vascular anatomy or severe PVD precludes safe
transfemoral (TF) access. However, TA-TAVR is associated
with higher mortality and prolonged recovery when compared
with TF-TAVR, and unlike TF-TAVR, health-related quality
of life after TA-TAVR is no better than SAVR.19,20 Our study
showed that TA-TAVR is also associated with an increased risk
of 30-day readmission. This association may be influenced
by residual confounding as patients undergoing TA-TAVR
are typically sicker with more complex PVD that cannot be
captured completely in an administrative database such as
the NRD. Nonetheless, alternate access such as direct aortic,
subclavian, transcaval, and transcarotid should be considered
in patients not suitable for TF-TAVR.21–23 Postprocedural AKI
occurs in 12% to 57% of patients after TAVR and is associated with increased mortality.24,25 AKI also predicted increased
30-day readmission after TAVR in our study. Discharge to
SNF and prolonged LOS, which may reflect frailty and high
burden of comorbidities and in-hospital complications, were
also associated with increased rates of 30-day readmission.26,27
Strategies such as multidisciplinary evaluation of TAVR
patients in collaboration with pulmonologists and nephrologists, consideration of alternate access other than TA-TAVR,
prevention or reduction of AKI with intravenous hydration,
and early discharge to home whenever possible, may help
reduce readmission rates.
The causes of 30-day readmission after TAVR were
similar to those previously reported with 61.8% of patients
being rehospitalized for noncardiac and 38.2% for cardiac
diagnoses.8 Respiratory disorders (including pneumonia)
and infections accounted for majority of the noncardiac
causes of readmission. Previous studies have shown that
7 Kolte et al Thirty-Day Readmissions After TAVR
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Figure 1. Causes of 30-day readmissions after transcatheter
aortic valve replacement. Noncardiac (A) and cardiac (B) causes
of 30-d readmissions after TAVR. AMI indicates acute myocardial
infarction; CAD, coronary artery disease; PVD, peripheral vascular
disease; and TIA, transient ischemic attack.
infectious complications after TAVR are mainly associated
with procedural factors such as surgical cut-down of the
femoral artery, life-threatening bleeding, prolonged intensive care unit stay, and potentially the use of general anesthesia (compared with deep sedation).28,29 Thus, attention to these
factors and implementation of effective strategies for reducing
hospital-acquired infections may help lower readmission rates
after TAVR.
Approximately two fifths of readmissions after TAVR were
because of cardiac causes, with heart failure being the most
common reason accounting for 22.5% of all readmissions.
Figure 2. Mean costs of index hospitalization and 30-day readmission after transcatheter aortic valve replacement (TAVR).
Mean costs of index hospitalization of TAVR patients with and
without 30-d readmission (blue) and mean costs of 30-d readmission (red) are shown.
Heart failure is frequent in patients with severe aortic stenosis
and remains the most common cause of rehospitalization and
death after TAVR.30,31 In previous studies, 22.8% to 30.4% of
all 30-day readmissions were because of heart failure.5,6,8 The
high rate of heart failure rehospitalizations after TAVR may
also be related to older age and high prevalence of coexisting
comorbidities such as hypertension, coronary artery disease,
anemia, atrial fibrillation, and CKD in patients with severe
aortic stenosis. Early postdischarge follow-up and other interventions shown to be effective in reducing heart failure rehospitalizations may also be helpful in lowering 30-day heart
failure readmissions after TAVR.32 Tachyarrhythmias were the
second most frequent cardiac cause of 30-day readmission,
with atrial fibrillation being the most common. In our study,
45.9% of patients had either preexisting or new-onset atrial
fibrillation during the index hospitalization, both of which
have been shown to be associated with worse outcomes in
patients undergoing TAVR.33,34 Rates of aortic valve reintervention and PPM placement during 30-day readmission were
relatively low.
Cost-effectiveness analyses of data from the Placement
of Aortic Transcatheter Valve (PARTNER) Cohort A and
CoreValve US High Risk Pivotal trials have shown that
in patients with severe aortic stenosis who are at high risk
for surgery, TAVR (especially via TF approach) is an economically attractive strategy as compared with SAVR, with
acceptable incremental costs.35,36 The estimated mean cost of
follow-up hospitalizations was $18 122±58 142 (TF-TAVR) and
11 733±31 924 (TA-TAVR) for the SAPIEN heart valve system
and $12 208±22 315 for the CoreValve system. Similarly, in
the current study, the mean hospital cost of readmission was
$13 528±15 285, which accounted for 16.4% of the total cost
of the episode of care. Although reductions in the cost of index
TAVR admissions can improve overall cost-effectiveness of
TAVR when compared with SAVR, efforts to reduce unplanned
rehospitalizations can help lower costs even further.36
Limitations
Our study has certain limitations. First, our analysis is based
on administrative data, which lacks clinical and laboratory
variables. Information on STS score, valve type and size,
echocardiographic findings such as aortic regurgitation/
paravalvular leak, and medications (eg, dual-antiplatelet
therapy, anticoagulation) was not available. Second, causes
of readmission were identified using the primary discharge
diagnosis codes. However, most previous studies on 30-day
readmission after cardiac procedures have also utilized a
similar approach.5,7 Third, patients who are hospitalized in
1 state and readmitted to a hospital in another state are not
tracked in the NRD; however, we expect that to be uncommon in patients undergoing TAVR. Last, because of unavailability of mortality data on TAVR patients who died outside
the hospital or in the emergency room, we were unable to
account for the influence of mortality on the rates of readmission. Despite these limitations, 30-day readmission rates in
our study are comparable to those reported in the Society of
Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy registry. The NRD provides valuable
information not completely captured in the TVT registry such
8 Kolte et al Thirty-Day Readmissions After TAVR
as causes of noncardiac readmissions and costs of readmissions after TAVR.
Conclusions
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
In a real-world, all-payer, nationally representative cohort of
TAVR patients in the United States, 30-day readmission rates
were 17.9% with significant variation across hospitals. High
burden of comorbidities, particularly CKD and CLD, TATAVR, AKI, discharge to SNF, and longer LOS during the
index hospitalization were associated with increased risk of
30-day readmission. Heart failure was the most common overall cause of readmission; however, more than three fifths of
readmissions were because of noncardiac causes. Cost of readmission accounted for 16.4% of the total cost of the episode
of care in patients who were readmitted. As the use of TAVR
expands with the recent approval for intermediate-risk patients,
and potentially even for low-risk patients in the near future,
implementation of strategies to reduce readmissions would be
crucial to reduce healthcare burden and maintain cost-effectiveness of TAVR. Utilization of 30-day risk-standardized
readmission rates after TAVR as a performance metric may
help incentivize hospitals to develop and implement institution-specific strategies to reduce readmissions. However, public reporting of such measures for TAVR should be avoided as
this may lead to inadvertent risk-averse behavior among operators and institutions as seen with PCI.37 Our findings do support future inclusion of TAVR in the bundled payment models
proposed by the Centers for Medicare and Medicaid Services
to promote high-quality cardiac care at a lower cost.
Disclosures
Dr Bhatt discloses the following relationships—Advisory Board:
Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology,
Regado Biosciences; Board of Directors: Boston VA Research
Institute, Society of Cardiovascular Patient Care; Chair: American
Heart Association Quality Oversight Committee; Data Monitoring
Committees: Duke Clinical Research Institute, Harvard Clinical
Research Institute, Mayo Clinic, Population Health Research Institute;
Honoraria: American College of Cardiology (Senior Associate Editor,
Clinical Trials and News, ACC.org), Belvoir Publications (Editor
in Chief, Harvard Heart Letter), Duke Clinical Research Institute
(clinical trial steering committees), Harvard Clinical Research
Institute (clinical trial steering committee), HMP Communications
(Editor in Chief, Journal of Invasive Cardiology), Journal of the
American College of Cardiology (Guest Editor; Associate Editor),
Population Health Research Institute (clinical trial steering committee), Slack Publications (Chief Medical Editor, Cardiology Today’s
Intervention), Society of Cardiovascular Patient Care (Secretary/
Treasurer), WebMD (CME steering committees); Other: Clinical
Cardiology (Deputy Editor), NCDR-ACTION Registry Steering
Committee (Chair), VA CART Research and Publications Committee
(Chair); Research Funding: Amarin, Amgen, AstraZeneca, BristolMyers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix,
Lilly, Medtronic, Pfizer, Roche, Sanofi Aventis, The Medicines
Company; Royalties: Elsevier (Editor, Cardiovascular Intervention:
A Companion to Braunwald’s Heart Disease); Site Co-Investigator:
Biotronik, Boston Scientific, St. Jude Medical; Trustee: American
College of Cardiology; Unfunded Research: FlowCo, PLx Pharma,
Takeda. The other authors report no conflicts.
References
1. Medicare.gov. 30-day unplanned readmission and death measures. 2016.
https://www.medicare.gov/hospitalcompare/Data/30-day-measures.html.
Accessed August 7, 2016.
2. Obama B. United States health care reform: progress to date and next
steps. JAMA. 2016;316:525–532. doi: 10.1001/jama.2016.9797.
3. Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients
in the Medicare fee-for-service program. N Engl J Med. 2009;360:1418–
1428. doi: 10.1056/NEJMsa0803563.
4.Holmes DR Jr, Nishimura RA, Grover FL, Brindis RG, Carroll JD,
Edwards FH, Peterson ED, Rumsfeld JS, Shahian DM, Thourani VH,
Tuzcu EM, Vemulapalli S, Hewitt K, Michaels J, Fitzgerald S, Mack MJ;
STS/ACC TVT Registry. Annual outcomes with transcatheter valve therapy: from the STS/ACC TVT Registry. J Am Coll Cardiol. 2015;66:2813–
2823. doi: 10.1016/j.jacc.2015.10.021.
5. Hannan EL, Samadashvili Z, Jordan D, Sundt TM 3rd, Stamato NJ, Lahey
SJ, Gold JP, Wechsler A, Ashraf MH, Ruiz C, Wilson S, Smith CR. Thirtyday readmissions after transcatheter aortic valve implantation versus surgical aortic valve replacement in patients with severe aortic stenosis in
New York State. Circ Cardiovasc Interv. 2015;8:e002744. doi: 10.1161/
CIRCINTERVENTIONS.115.002744.
6.Holmes DR Jr, Brennan JM, Rumsfeld JS, Dai D, O’Brien SM,
Vemulapalli S, Edwards FH, Carroll J, Shahian D, Grover F, Tuzcu EM,
Peterson ED, Brindis RG, Mack MJ; STS/ACC TVT Registry. Clinical
outcomes at 1 year following transcatheter aortic valve replacement.
JAMA. 2015;313:1019–1028. doi: 10.1001/jama.2015.1474.
7.Murugiah K, Wang Y, Desai NR, Nuti SV, Krumholz HM. Hospital
variation in outcomes for transcatheter aortic valve replacement among
Medicare beneficiaries, 2011 to 2013. J Am Coll Cardiol. 2015;66:2678–
2679. doi: 10.1016/j.jacc.2015.10.008.
8. Nombela-Franco L, del Trigo M, Morrison-Polo G, Veiga G, JimenezQuevedo P, Abdul-Jawad Altisent O, Campelo-Parada F, Biagioni C,
Puri R, DeLarochellière R, Dumont E, Doyle D, Paradis JM, Quirós A,
Almeria C, Gonzalo N, Nuñez-Gil I, Salinas P, Mohammadi S, Escaned
J, Fernández-Ortiz A, Macaya C, Rodés-Cabau J. Incidence, causes, and
predictors of early (≤30 days) and late unplanned hospital readmissions
after transcatheter aortic valve replacement. JACC Cardiovasc Interv.
2015;8:1748–1757. doi: 10.1016/j.jcin.2015.07.022.
9.US Agency for Healthcare Research and Quality. Overview of Key
Readmission Measures and Methods. 2012. http://www.hcupus.ahrq.gov/
reports/methods/methods.jsp. Accessed August 7, 2016.
10. US Bureau of Labor Statistics, Division of Consumer Prices and Price
Indexes. Consumer Price Index - All Urban Consumers. 2016. http://data.
bls.gov/pdq/SurveyOutputServlet. Accessed May 20, 2016.
11. Curtis JP, Schreiner G, Wang Y, Chen J, Spertus JA, Rumsfeld JS, Brindis
RG, Krumholz HM. All-cause readmission and repeat revascularization
after percutaneous coronary intervention in a cohort of medicare patients.
J Am Coll Cardiol. 2009;54:903–907. doi: 10.1016/j.jacc.2009.04.076.
12.Hannan EL, Zhong Y, Lahey SJ, Culliford AT, Gold JP, Smith CR,
Higgins RS, Jordan D, Wechsler A. 30-day readmissions after coronary
artery bypass graft surgery in New York State. JACC Cardiovasc Interv.
2011;4:569–576. doi: 10.1016/j.jcin.2011.01.010.
13. Wasfy JH, Rosenfield K, Zelevinsky K, Sakhuja R, Lovett A, Spertus JA,
Wimmer NJ, Mauri L, Normand SL, Yeh RW. A prediction model to identify patients at high risk for 30-day readmission after percutaneous coronary intervention. Circ Cardiovasc Qual Outcomes. 2013;6:429–435. doi:
10.1161/CIRCOUTCOMES.111.000093.
14. Murugiah K, Wang Y, Dodson JA, Nuti SV, Dharmarajan K, Ranasinghe I,
Cooper Z, Krumholz HM. Trends in hospitalizations among medicare survivors of aortic valve replacement in the United States from 1999 to 2010.
Ann Thorac Surg. 2015;99:509–517. doi: 10.1016/j.athoracsur.2014.08.045.
15.Nguyen TC, Babaliaros VC, Razavi SA, Kilgo PD, Guyton RA,
Devireddy CM, Shults CC, Mavromatis K, Kanitkar M, Block P, Lerakis
S, Thourani VH. Impact of varying degrees of renal dysfunction on transcatheter and surgical aortic valve replacement. J Thorac Cardiovasc Surg.
2013;146:1399–1406, discussion 1406. doi: 10.1016/j.jtcvs.2013.07.065.
16.Suri RM, Gulack BC, Brennan JM, Thourani VH, Dai D, Zajarias A,
Greason KL, Vassileva CM, Mathew V, Nkomo VT, Mack MJ, Rihal CS,
Svensson LG, Nishimura RA, O’Gara PT, Holmes DR Jr. Outcomes of patients with severe chronic lung disease who are undergoing transcatheter
aortic valve replacement. Ann Thorac Surg. 2015;100:2136–2145, discussion 2145. doi: 10.1016/j.athoracsur.2015.05.075.
17.Szerlip M, Kim RJ, Adeniyi T, Thourani V, Babaliaros V, Bavaria J,
Herrmann HC, Anwaruddin S, Makkar R, Chakravarty T, Rovin J,
Creighton D, Miller DC, Baio K, Walsh E, Katinic J, Letterer R, Trautman
L, Herbert M, Farkas R, Rudolph J, Brown D, Holper EM, Mack M. The
outcomes of transcatheter aortic valve replacement in a cohort of patients
with end-stage renal disease. Catheter Cardiovasc Interv. 2016;87:1314–
1321. doi: 10.1002/ccd.26347.
9 Kolte et al Thirty-Day Readmissions After TAVR
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
18. Dvir D, Waksman R, Barbash IM, Kodali SK, Svensson LG, Tuzcu EM,
Xu K, Minha S, Alu MC, Szeto WY, Thourani VH, Makkar R, Kapadia S,
Satler LF, Webb JG, Leon MB, Pichard AD. Outcomes of patients with
chronic lung disease and severe aortic stenosis treated with transcatheter
versus surgical aortic valve replacement or standard therapy: insights from
the PARTNER trial (Placement of Aortic Transcatheter Valve). J Am Coll
Cardiol. 2014;63:269–279. doi: 10.1016/j.jacc.2013.09.024.
19.Blackstone EH, Suri RM, Rajeswaran J, Babaliaros V, Douglas PS,
Fearon WF, Miller DC, Hahn RT, Kapadia S, Kirtane AJ, Kodali SK,
Mack M, Szeto WY, Thourani VH, Tuzcu EM, Williams MR, Akin JJ,
Leon MB, Svensson LG. Propensity-matched comparisons of clinical
outcomes after transapical or transfemoral transcatheter aortic valve replacement: a Placement of Aortic Transcatheter Valves (PARTNER)-I
trial substudy. Circulation. 2015;131:1989–2000. doi: 10.1161/
CIRCULATIONAHA.114.012525.
20.Gada H, Kirtane AJ, Wang K, Lei Y, Magnuson E, Reynolds MR,
Williams MR, Kodali S, Vahl TP, Arnold SV, Leon MB, Thourani V,
Szeto WY, Cohen DJ; PARTNER Investigators. Temporal trends in
quality of life outcomes after transapical transcatheter aortic valve
replacement: a Placement of Aortic Transcatheter Valve (PARTNER)
trial substudy. Circ Cardiovasc Qual Outcomes. 2015;8:338–346. doi:
10.1161/CIRCOUTCOMES.114.001335.
21. Mack MJ. Access for transcatheter aortic valve replacement: which is the
preferred route? JACC Cardiovasc Interv. 2012;5:487–488. doi: 10.1016/j.
jcin.2012.03.009.
22. Greenbaum AB, O’Neill WW, Paone G, Guerrero ME, Wyman JF, Cooper
RL, Lederman RJ. Caval-aortic access to allow transcatheter aortic valve replacement in otherwise ineligible patients: initial human experience. J Am
Coll Cardiol. 2014;63(25 pt A):2795–2804. doi: 10.1016/j.jacc.2014.04.015.
23. Mylotte D, Sudre A, Teiger E, Obadia JF, Lee M, Spence M, Khamis
H, Al Nooryani A, Delhaye C, Amr G, Koussa M, Debry N, Piazza N,
Modine T. Transcarotid transcatheter aortic valve replacement: feasibility and safety. JACC Cardiovasc Interv. 2016;9:472–480. doi: 10.1016/j.
jcin.2015.11.045.
24. Généreux P, Kodali SK, Green P, Paradis JM, Daneault B, Rene G, Hueter
I, Georges I, Kirtane A, Hahn RT, Smith C, Leon MB, Williams MR.
Incidence and effect of acute kidney injury after transcatheter aortic valve
replacement using the new valve academic research consortium criteria.
Am J Cardiol. 2013;111:100–105. doi: 10.1016/j.amjcard.2012.08.057.
25. Arsalan M, Squiers JJ, Farkas R, Worley C, Herbert M, Stewart W, Brinkman
WT, Ungchusri E, Brown DL, Mack MJ, Holper EM. Prognostic usefulness of acute kidney injury after transcatheter aortic valve replacement. Am
J Cardiol. 2016;117:1327–1331. doi: 10.1016/j.amjcard.2016.01.037.
26.Lee DH, Buth KJ, Martin BJ, Yip AM, Hirsch GM. Frail patients are
at increased risk for mortality and prolonged institutional care after cardiac surgery. Circulation. 2010;121:973–978. doi: 10.1161/
CIRCULATIONAHA.108.841437.
27. Schoenenberger AW, Stortecky S, Neumann S, Moser A, Jüni P, Carrel
T, Huber C, Gandon M, Bischoff S, Schoenenberger CM, Stuck AE,
Windecker S, Wenaweser P. Predictors of functional decline in elderly
patients undergoing transcatheter aortic valve implantation (TAVI). Eur
Heart J. 2013;34:684–692. doi: 10.1093/eurheartj/ehs304.
28. van der Boon RM, Nuis RJ, Benitez LM, Van Mieghem NM, Perez S,
Cruz L, van Geuns RJ, Serruys PW, van Domburg RT, Dager AE, de
Jaegere PP. Frequency, determinants and prognostic implications of infectious complications after transcatheter aortic valve implantation. Am J
Cardiol. 2013;112:104–110. doi: 10.1016/j.amjcard.2013.02.061.
29. Tirado-Conte G, Freitas-Ferraz AB, Nombela-Franco L, Jimenez-Quevedo
P, Biagioni C, Cuadrado A, Nuñez-Gil I, Salinas P, Gonzalo N, Ferrera C,
Vivas D, Higueras J, Viana-Tejedor A, Perez-Vizcayno MJ, Vilacosta I,
Escaned J, Fernandez-Ortiz A, Macaya C. Incidence, causes, and impact
of in-hospital infections after transcatheter aortic valve implantation. Am J
Cardiol. 2016;118:403–409. doi: 10.1016/j.amjcard.2016.05.012.
30. Himbert D, Vahanian A. Transcatheter aortic valve replacement for patients with heart failure. Heart Fail Clin. 2015;11:231–242. doi: 10.1016/j.
hfc.2014.12.003.
31.Urena M, Webb JG, Eltchaninoff H, Muñoz-García AJ, Bouleti C,
Tamburino C, Nombela-Franco L, Nietlispach F, Moris C, Ruel M, Dager
AE, Serra V, Cheema AN, Amat-Santos IJ, de Brito FS, Lemos PA, Abizaid
A, Sarmento-Leite R, Ribeiro HB, Dumont E, Barbanti M, Durand E,
Alonso Briales JH, Himbert D, Vahanian A, Immè S, Garcia E, Maisano
F, del Valle R, Benitez LM, García del Blanco B, Gutiérrez H, Perin MA,
Siqueira D, Bernardi G, Philippon F, Rodés-Cabau J. Late cardiac death in
patients undergoing transcatheter aortic valve replacement: incidence and
predictors of advanced heart failure and sudden cardiac death. J Am Coll
Cardiol. 2015;65:437–448. doi: 10.1016/j.jacc.2014.11.027.
32.Bradley EH, Curry L, Horwitz LI, Sipsma H, Wang Y, Walsh MN,
Goldmann D, White N, Piña IL, Krumholz HM. Hospital strategies associated with 30-day readmission rates for patients with heart failure. Circ Cardiovasc Qual Outcomes. 2013;6:444–450. doi: 10.1161/
CIRCOUTCOMES.111.000101.
33.Chopard R, Teiger E, Meneveau N, Chocron S, Gilard M, Laskar M,
Eltchaninoff H, Iung B, Leprince P, Chevreul K, Prat A, Lievre M,
Leguerrier A, Donzeau-Gouge P, Fajadet J, Mouillet G, Schiele F;
FRANCE-2 Investigators. Baseline characteristics and prognostic implications of pre-existing and new-onset atrial fibrillation after transcatheter
aortic valve implantation: results from the FRANCE-2 Registry. JACC
Cardiovasc Interv. 2015;8:1346–1355. doi: 10.1016/j.jcin.2015.06.010.
34. Biviano AB, Nazif T, Dizon J, Garan H, Fleitman J, Hassan D, Kapadia
S, Babaliaros V, Xu K, Parvataneni R, Rodes-Cabau J, Szeto WY, Fearon
WF, Dvir D, Dewey T, Williams M, Mack MJ, Webb JG, Miller DC,
Smith CR, Leon MB, Kodali S. Atrial fibrillation is associated with increased mortality in patients undergoing transcatheter aortic valve replacement: insights from the Placement of Aortic Transcatheter Valve
(PARTNER) Trial. Circ Cardiovasc Interv. 2016;9:e002766. doi: 10.1161/
CIRCINTERVENTIONS.115.002766.
35. Reynolds MR, Magnuson EA, Lei Y, Wang K, Vilain K, Li H, Walczak J,
Pinto DS, Thourani VH, Svensson LG, Mack MJ, Miller DC, Satler LE,
Bavaria J, Smith CR, Leon MB, Cohen DJ; PARTNER Investigators. Costeffectiveness of transcatheter aortic valve replacement compared with
surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results of the PARTNER (Placement of Aortic Transcatheter
Valves) trial (Cohort A). J Am Coll Cardiol. 2012;60:2683–2692. doi:
10.1016/j.jacc.2012.09.018.
36. Reynolds MR, Lei Y, Wang K, Chinnakondepalli K, Vilain KA, Magnuson
EA, Galper BZ, Meduri CU, Arnold SV, Baron SJ, Reardon MJ, Adams
DH, Popma JJ, Cohen DJ; CoreValve US High Risk Pivotal Trial
Investigators. Cost-effectiveness of transcatheter aortic valve replacement
with a self-expanding prosthesis versus surgical aortic valve replacement.
J Am Coll Cardiol. 2016;67:29–38. doi: 10.1016/j.jacc.2015.10.046.
37. McCabe JM, Waldo SW, Kennedy KF, Yeh RW. Treatment and outcomes
of acute myocardial infarction complicated by shock after public reporting policy changes in New York. JAMA Cardiol. 2016;1:648–654. doi:
10.1001/jamacardio.2016.1806.
Downloaded from http://circinterventions.ahajournals.org/ by guest on June 18, 2017
Thirty-Day Readmissions After Transcatheter Aortic Valve Replacement in the United
States: Insights From the Nationwide Readmissions Database
Dhaval Kolte, Sahil Khera, M. Rizwan Sardar, Neil Gheewala, Tanush Gupta, Saurav
Chatterjee, Andrew Goldsweig, Wilbert S. Aronow, Gregg C. Fonarow, Deepak L. Bhatt, Adam
B. Greenbaum, Paul C. Gordon, Barry Sharaf and J. Dawn Abbott
Circ Cardiovasc Interv. 2017;10:
doi: 10.1161/CIRCINTERVENTIONS.116.004472
Circulation: Cardiovascular Interventions is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 2016 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-7640. Online ISSN: 1941-7632
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circinterventions.ahajournals.org/content/10/1/e004472
Data Supplement (unedited) at:
http://circinterventions.ahajournals.org/content/suppl/2017/01/16/CIRCINTERVENTIONS.116.004472.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Cardiovascular Interventions can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which
permission is being requested is located, click Request Permissions in the middle column of the Web page
under Services. Further information about this process is available in the Permissions and Rights Question and
Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Cardiovascular Interventions is online at:
http://circinterventions.ahajournals.org//subscriptions/
SUPPLEMENTAL MATERIAL
1
Supplemental Tables:
Table 1. International Classification of Diseases, Ninth Edition, Clinical Modification (ICD-9CM) Codes Used To Identify Baseline Comorbidities, Procedures and In-Hospital Outcomes
Variable
Code(s)
Comorbidities
Smoking
Dyslipidemia
V15.82, 30.51
272.0–4
Known CAD
Prior MI
Prior PCI
Prior CABG
414.00–07
412
V45.82
V45.81
Carotid artery disease
Prior TIA/stroke
Atrial fibrillation
Prior PPM
Prior ICD
433.10
V12.54, 438.x
427.31
V45.01
V45.02
Dementia
290.xx, 294.1x, 294.2x, 294.8, 331.0–2, 331.82, 797
In-Hospital Procedures
Endovascular TAVR
Transapical TAVR
Coronary angiography
PCI
Mechanical circulatory support
In-Hospital Outcomes
Surgical AVR
Complete heart block
PPM placement
TIA/Stroke
35.05
35.06
88.55, 88.56, 37.22, 37.23
00.66, 36.01, 36.02, 36.05, 36.06, 36.07
37.61, 37.68, 39.65
35.21, 35.22
42.60
37.80–83, 00.50
431, 433.x1, 434.x1, 435.x, 344.6x, 997.01
AMI
Cardiogenic shock
410.x1, 411.1
785.51
Cardiac arrest
AKI
427.5
584.x
2
Major bleeding
430, 431, 432.x, 336, 362.43, 362.81, 363.61,
363.62, 363.72, 364.41, 3736.32, 377.42, 379.23;
423.0 + 37.0;
923.x + 729.71, 924.x + 729.72, 922.2–9 + 729.73;
456.0, 456.20, 530.7, 530.82, 531.00, 531.01,
531.20, 531.21, 531.40, 531.41, 531.60, 531.61,
532.00, 532.01, 532.20, 532.21, 532.40, 532.41,
532.60, 532.61, 533.00, 533.01, 533.20, 533.21,
533.40, 533.41, 533.60, 533.61, 534.00, 534.01,
534.20, 534.21, 534.40, 534.41, 534.60, 534.61,
569.3, 578.0, 578.1, 578.9, 568.81, 599.70, 599.71,
719.1x, 784.7, 784.8, 459, 998.11, 998.12, 285.1
+ 998.00, 998.09, 785.50, 785.59, 276.52
+ 00.17
+ 99.0x
Vascular complications
900–904, 998.2, 999.2, 997.7, 447.0, 868.04
+ 39.31, 39.41, 39.49, 39.52, 39.56, 39.57, 39.59,
39.79
Hypertension, diabetes mellitus, obesity, heart failure, peripheral vascular disease, anemia,
chronic kidney disease, chronic lung disease, liver disease, coagulopathy, depression,
hypothyroidism, fluid and electrolyte disorders, other neurological disorders, pulmonary
circulation disorders, and cancer were identified from the 29 Elixhauser comorbidities included
in the Nationwide Readmissions Database.
CAD = coronary artery disease, MI = myocardial infarction, PCI = percutaneous coronary
intervention, CABG = coronary artery bypass grafting, TIA = transient ischemic attack, PPM =
permanent pacemaker, ICD = implantable cardioverter defibrillator, AVR = aortic valve
replacement, AKI = acute kidney injury
3
Table 2. Causes of 30-Day Readmissions Categorized According to Clinical Classifications
Software (CCS) and/or International Classification of Diseases, Ninth Edition, Clinical
Modification (ICD-9-CM) Codes in the Primary Diagnosis Position*
Causes of Readmission
CCS code(s)
ICD-9-CM code(s)
122, 124, 125, 127, 128, 129,
130, 131, 133, 134
415.19, 997.3x
Infections
2, 4, 135, 159, 197, 246
519.01, 780.62, 996.6x,
998.5x
Bleeding
60, 153
280.0, 537.83, 569.85–86,
599.7x, 998.1x
Peripheral vascular disease
114, 115, 116, 118, 248
415.11, 443.89, 997.2, 997.79
Genitourinary
156, 157, 158, 160, 161, 163,
165, 166
403.91, 458,21, 997.5, 996.73,
996.76
Gastrointestinal
138, 140, 141, 143, 145, 146,
147, 149, 151, 152, 154, 155,
250, 251
109, 112
996.82, 997.49
Trauma
226, 229, 230, 231, 233, 235,
239, 242, 244
997.99, 998.32, 998.83
Endocrine/metabolic
49, 50, 51, 52, 54, 55
—
Neuropsychiatric
83, 84, 85, 93, 95, 653, 660,
661
780.97
Hematological/neoplasms
59, 14, 32, 42, 44
790.92
Others
199, 204, 209, 211, 252, 253,
254, 257
—
108
402.91, 404.11, 404.91,
404.93, 429.4, 997.1
—
NON-CARDIAC
Respiratory
TIA/stroke
CARDIAC
Heart failure
Arrhythmias
106, 107
4
—
Conduction disorders
105
—
Valve disordersy
96
Hyper/hypotension
—
421.x, 996.02, 996.71, 996.72,
996.74
401.x, 458.0, 458.29, 458.9
Pericarditis
—
420.x, 423.x
CAD
101, 104
—
AMI
100
—
Others
102, 245
416.8
*CCS and ICD-9-CM codes listed are NOT all inclusive for each category, but represent the
primary diagnosis codes present in the discharge record of TAVR patients who experienced 30day readmission in this study.
y
including endocarditis, prosthesis-related mechanical complications, and aortic valve re-
intervention
TIA = transient ischemic attack, CAD = coronary artery disease, AMI = acute myocardial
infarction
5
Supplemental Figures and Figure Legends:
Figure 1. TAVR Volume
6
Figure 2. Hospital Variation in 30-Day Readmission Rates After TAVR
7
Figure 3. Kaplan-Meier Curves for 30-Day Readmissions for Different Risk Factors
8
Figure Legends:
Figure 1. TAVR Volume
Histogram showing distribution of hospitals according to number of TAVR procedures
performed during the study period
Figure 2. Hospital Variation in 30-Day Readmission Rates After TAVR
Hospitals performing ≥5 (A) or ≥10 (B) TAVRs were divided into deciles of 30-day readmission
rates. Box plots depict median, interquartile range and range of 30-day readmission rates for
hospitals in each decile.
Figure 3. Kaplan-Meier Curves for 30-Day Readmissions for Different Risk Factors
Kaplan-Meier curves showing 30-day readmission rates for TAVR patients with LOS >5 vs ≤5
days (A), AKI vs no AKI (B), >4 vs ≤4 Elixhauser comorbidities (C), TA vs endovascular
TAVR (D), CKD vs no CKD (E), CLD vs no CLD (F), and discharge to SNF vs home (G).
LOS = length of stay, AKI = acute kidney injury, CKD = chronic kidney disease, TA =
transapical, CLD = chronic lung disease, SNF = skilled nursing facility.
9