Causes and Predictors of 30-Day Readmission in Patients with Cardiogenic Shock underwent Extra corporeal ECMO.
Abstract:
Background:
Cardiogenic shock is a complication that has significant mortality, morbidity and cost on the health care system. One of the strategies managing cardiogenic shock is the extra corporeal ECMO. Despite the hemodynamic support provided by ECMO the mortality readmission rate is still high and the data to predict outcomes still limited.
Methods and results:
We derived our study cohort of patient with cardiogenic shock who underwent ECMO at the same admission from 2016 to 2017. Healthcare Cost and Utilization Project National Readmission Database. Incidence, mortality, readmission rate and causes of 30-day readmissions were analyzed. From (n=69040) admissions, (n=1641, 2.37%) underwent VA ECMO, (n=907, 55.27%) survived to discharge. (n=661) patient who survived their follow up was available, (n=158, 23.9%) were readmitted within less than 30 days. More than 50% of these readmissions happened within the first 11 days, (7.4%) were died during the readmission hospitalization. Median cost of stay ($ 586,727), the median admission day to ecmo placement was 2 days. Leading cause of readmission was Cardiovascular 31.6% (Heart Failure 24.1%, Arrythmia 20.6%, Neurovascular 10.3%, Hypertension 10.3%, Endocarditis 6.8%), The next leading cause of readmission was the complications of medical/ device care accounted for (17.7%), followed by infection (11.3%), Gastroenteric and liver (10.1%) complications. Discharging the patient to SNF (odds ratio [OR], 3.253; 95% confidence interval [CI], 1.655–6.392; P<0.01), Long term ventricular assisted device (odds ratio [OR], 2.845; 95% confidence interval [CI], 1.517–5.33; P<0.01), Discharging the patient with Home Health Care (odds ratio [OR], 2.466; 95% confidence interval [CI], 1.312–4.634; P<0.05), Vascular Complication (odds ratio [OR], 1.628; 95% confidence interval [CI], 1.007–2.634; P<0.04), Liver disease (odds ratio [OR], 1.575; 95% confidence interval [CI], 1.005–2.469; P<0.01), were among the independent predictors of 30-day readmission. The Length of stay > 8 days (odds ratio [OR], 0.156; 95% confidence interval [CI], 0. 11–0.20; P<0.01), Acute DVT (odds ratio [OR], 0.561; 95% confidence interval [CI], 0.38–0.836; P<0.05), Myocarditis (odds ratio [OR], 0.474; 95% confidence interval [CI], 0.23–0.963; P<0.03), Medicaid as payor (odds ratio [OR], 0.6; 95% confidence interval [CI], 0.42–0.86; P<0.02), Heart Transplant (odds ratio [OR], 0.19; 95% confidence interval [CI], 0.08–0.46; P<0.01) were among the independent predictors of survival. While Age >49 (odds ratio [OR], 2.10; 95% confidence interval [CI], 1.695–2.61; P<0.01), Peripheral Vascular disease (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.03–1.93; P<0.01), Coagulopathy including DIC (odds ratio [OR], 1.35; 95% confidence interval [CI], 1.04–1.76; P<0.02), Liver disease (odds ratio [OR], 1.81; 95% confidence interval [CI], 1.38–2.39; P<0.01), Acute Kidney injury (odds ratio [OR], 2.68; 95% confidence interval [CI], 1.92–3.74; P<0.01), Chronic Kidney Disease (odds ratio [OR], 1.78; 95% confidence interval [CI], 1.30–1.2.44; P<0.01), Hemorrhagic Stroke (odds ratio [OR], 2.05; 95% confidence interval [CI], 1.09–3.84; P<0.02), Hemoptysis (odds ratio [OR], 1.88; 95% confidence interval [CI], 1.21–2.94; P<0.02) were among the independent predictors of mortality.
Conclusion:
Patients with cardiogenic shock underwent ECMO has significant mortality and readmission rate. Yet there is no perfect model that can predict the readmission and mortality for patients with cardiogenic shock who underwent ECMO. Identifying patient at high risk of readmission and mortality might help with improving outcomes.
BACKGROUD:
In progress
METHODE:
The study cohort was derived from Healthcare Cost and Utilization Project National Readmission Database (NRD) of 2016, supported by the Agency for Healthcare Research and Quality. The Nationwide Readmissions Database (NRD) is part of a family of databases and software tools developed for the Healthcare Cost and Utilization Project (HCUP).
The NRD contains data from approximately 17 million discharges each year. Weighted, it estimates roughly 36 million discharges.
The patients can be tracked using linkage number and their admissions, length of stay, comorbidities, procedures are reliably linked in the database.
We queried the NRD database using International Classification of disease 10th revision. We used the diagnosis code of cardiogenic shock (ICD10 CM R57.0) to query only the index admission for the first step and we queried the VA-ECMO procedure code (ICD 10 PSC of 5A1522G. We excluded patients with age ≤18 years and those missing data for age, sex, or mortality. We also excluded patients who were discharged on the month of December as we would not have their readmission information. We studied these patients further and characterized their morbidity and procedure underwent during the index admission and the main diagnosis of readmission using ICD10 CM and ICD10 PSC (available with the supplement). The variables age, sex, discharge disposition, zip code, primary payer, admission status, length of stay, cost of stay, admission day were available from the NRD database.
The primary outcome was 30-day readmission post admission index. We studied the previous factors with their relation to readmission and mortality. Patient who expired during the index admission were excluded from the 30-days readmission.
We used IBM SPSS version 24.0. We expressed the categorical data as percentage and count. Continuous variables as mean ±1 SD. We tested the categorical variables using Chi Square test and the continuous using t test. The demographics data were listed in (table1).
RESULTS:
Baseline Characteristics:
A total of (n=69040) admission identified their index hospitalization complicated by cardiogenic shock, (n=1641, 2.37%) of these admissions had ECMO. Out of these patients
who underwent ECMO, (n=734, 44.7%) survived their index admission.
Mean age of patients was (55.74 ±14.8) years, Median Length of stay was 17 days [6-36] days. (21.7%) of admissions were happened on weekend day, (33.6%) were females, (15.1%) were elective admissions.
Median cost of stay ($ 586,727), the median admission day to ECMO placement was 2 days.
(n=1039, 63.3%) of Cardiogenic shock etiology required ECMO support were ischemic, (n=57, 3.5%) were Myocarditis.
Mortality and predictor of mortality:
Mortality rate was higher among people who were admitted emergently (16.6% vs. 13.2%) p=0.031, underwent CABG (14.3 vs. 9.3%), p=0.002, had AKI (81.1% vs. 72.2%), p<0.001, acute MI (66% vs. 59.9%), p=0.012, Chronic pulmonary disease (19.6% vs. 14.2%), p=0.004, Coagulopathy including DIC (57.3% vs. 51.5%), p=0.019, Liver disease (45.1% vs. 34.9%), p<0.001, Chronic Kidney Disease (30.5% vs. 25.1%), p<0.015, Length of stay > 8 days (44.7% vs. 36.9%), p<0.001,
Mortality rate was lower in patients who had acute DVT (7.5% vs. 20.8%) p<0.001, Myocarditis (2.1% vs. 5.2%), p<0.001.
(n=267, 16.3%) had Impella support, (n=512, 31.2%) had Balloon pump support, (n=164, 10%) underwent Left Ventricular assisted device (LVAD), (n=67, 4.1%) underwent Heart Transplant, (n=198, 12.1%) had CABG. Mortality was lower among patients who underwent LVAD (6.4% vs. 14.4%) p<0.001. Same with patients who underwent heart transplant (1.4% vs. 7.4%) p<0.001. There was slight difference favor Impella (17% vs. 15.4%) but was not statistically significant, for IABP there was no difference in mortality.
The Length of stay > 8 days (odds ratio [OR], 0.156; 95% confidence interval [CI], 0. 11–0.20; P<0.01), Acute DVT (odds ratio [OR], 0.561; 95% confidence interval [CI], 0.38–0.836; P<0.05), Myocarditis (odds ratio [OR], 0.474; 95% confidence interval [CI], 0.23–0.963; P<0.03), Medicaid as payor (odds ratio [OR], 0.6; 95% confidence interval [CI], 0.42–0.86; P<0.02), Heart Transplant (odds ratio [OR], 0.19; 95% confidence interval [CI], 0.08–0.46; P<0.01) were among the independent predictors of survival. While Age >49 (odds ratio [OR], 2.10; 95% confidence interval [CI], 1.695–2.61; P<0.01), Peripheral Vascular disease (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.03–1.93; P<0.01), Coagulopathy including DIC (odds ratio [OR], 1.35; 95% confidence interval [CI], 1.04–1.76; P<0.02), Liver disease (odds ratio [OR], 1.81; 95% confidence interval [CI], 1.38–2.39; P<0.01), Acute Kidney injury (odds ratio [OR], 2.68; 95% confidence interval [CI], 1.92–3.74; P<0.01), Chronic Kidney Disease (odds ratio [OR], 1.78; 95% confidence interval [CI], 1.30–1.2.44; P<0.01), Hemorrhagic Stroke (odds ratio [OR], 2.05; 95% confidence interval [CI], 1.09–3.84; P<0.02), Hemoptysis (odds ratio [OR], 1.88; 95% confidence interval [CI], 1.21–2.94; P<0.02) were among the independent predictors of mortality. (figuerx)
30 days readmission, Cause, length of stay
661 patients who survived their admission were discharged before December of 2016 that we were able to identify their readmission status. (n=158, 23.9%) of these patients readmitted within 30 days.
More than 50% of these readmissions happened within the first 11 days (figure 1), (n= 12, 7.4%) were died during the readmission hospitalization.
Mean age of patients readmitted was (51.3 ± 14.2) years, Median length of stay for the readmission 7 [4-14] days, (12.7%) of these readmissions were electives, (7.6%) of patients readmitted died during the readmission. (34.2%) were females. Median cost of stay for readmission was ($72,715).
Leading causes of readmission were summarized in (figure 2) as the following, Cardiovascular 31.6%, the next leading cause of readmission was the complications of medical/ device care accounted for (17.7%), followed by infection (11.3%), Gastroenteric and liver (10.1%) complications.
The cardiovascular readmission causes summarized in (figure 3) as the following with the leading cause of Heart Failure 24.1%, Arrythmia 20.6%, Neurovascular 10.3%, Hypertension 10.3%, Endocarditis 6.8%.
The Table shows baseline characteristics among (n = 661) patients who survived the index hospitalization for CS underwent ECMO based on whether they were subsequently readmitted within 30 days of discharge compared with patients who were not readmitted.
Patients who their index admissions longer than 30 days, underwent LVAD, has vascular complications, discharged to Skilled Nursing facility or Home Care with agency were more likely to be readmitted. There was no difference between patients who had Mechanical support, chronic medical conditions, ischemic vs. non-ischemic, sex, age or zip codes.
Predictor of 30-days readmission:
On multivariate analysis among those who survived to discharge, discharging the patient to SNF (odds ratio [OR], 3.253; 95% confidence interval [CI], 1.655–6.392; P<0.01), Long term ventricular assisted device (odds ratio [OR], 2.845; 95% confidence interval [CI], 1.517–5.33; P<0.01), Discharging the patient with Home Health Care (odds ratio [OR], 2.466; 95% confidence interval [CI], 1.312–4.634; P<0.05), Vascular Complication (odds ratio [OR], 1.628; 95% confidence interval [CI], 1.007–2.634; P<0.04), Liver disease (odds ratio [OR], 1.575; 95% confidence interval [CI], 1.005–2.469; P<0.01), Patient state at the same hospital state (odds ratio [OR], 2.08; 95% confidence interval [CI], 1.05–4.12; P<0.03) were among the independent predictors of 30-day readmission.
DISCUSSION:
Characteristics
Overall (n=661) 30-d Readmission
P Value
Yes (n=158) No (n=503)
Age, y, mean±SD 52.2 ±14.2 51.3 ± 14.2 52.5 ± 15.0 0.616
Age, y (categories) 0.414
≤49 38.0 39.9 37.4
50–64 41.3 42.4 41.0
65–79 19.7 17.7 20.3
≥80 1.1 0.0 1.4
Female sex, % 31.5
34.2
30.6
0.432
Weekend admission, % 23.8
23.4
23.9
1
Elective admission, % 13.6
10.8
14.5
0.287
Payer information, % 0.209
Medicare 28.1 28.5 28.0
Medicaid 16.2 20.3 14.9
Private 49.9 47.5 50.7
Self-pay 1.7 0.0 2.2
Other 4.1 3.8 4.2
Cost of hospitalization in USD (mean) 1067627 1219047 0.601
Median cost of hospitalization in USD (IQR) 728,754
(387944–1367464) 930,787
(498040-1625149) 667,730
(351067-1306670) n/a
Length of stay, d, mean±SD 40.3 ± 37.5 50.8 ± 45.6 37.0 ± 33.9 <0.001
Median length of stay, d (IQR) 31 (18–52) 38 (24-62) 30 (15-49) n/a
Length of stay categories, d <0.02
≤2 3.6 1.9 4.2
3–4 2.1 0.6 2.6
5–7 3.6 0.6 4.6
≥8 36.8 35.4 37.2
Length of stay >30 d 52.6 60.8 50.1
Median household income category for patient’s zip code* (percentile) 0.626
0–25th 23.8 22.8 24.1
26–50th 22.5 22.8 22.5
51–75th 26.2 22.8 27.2
76–100th 26.6 29.7 25.6
Comorbidities
Hypertension with and without complications, % 44.6 41.8 45.5 0.408
Diabetes mellitus with and without complications, % 26.6 31.0 25.2 0.18
Dyslipidemia, % 31.0 32.3 30.6 0.74
Chronic pulmonary disease, % 14.4
17.7
13.3
0.694
Pulmonary hypertension, % 26.6
9.4 31.6
25.0
0.121
Current or past smoker, % 16.9
.8 15.2
.9 17.5
0.545
History of stroke or TIA, % 4.7
4.7
4.4
4.8
1
History of myocardial infarction, % 11.0
9.5
11.5
0.561
Drug abuse, % 6.5
6.3
6.6
1
Alcohol abuse, % 4.8
3.8
5.2
0.671
Peripheral vascular disorders, % 22.5
24.1
22.1
0.587
Coagulopathy, % 51.9
44.9
54.1
0.055
Deficiency anemia, % 2.0
1.3
2.2
0.743
Chronic blood loss anemia, % 2.3
1.9
2.4
1
Collagen vascular disease or rheumatoid arthritis, % 2.3
1.9
2.4
1
Hypothyroidism, % 8.0
7.0
8.3
0.737
Liver disease, % 35.7 39.2 34.6 0.296
Fluid and electrolytes disorders, % 75.9 75.3 76.1 0.832
Obesity, % 17.4
16.5
17.7
0.81
Obstructive sleep apnea, % 8.9
9.5
8.7
0.751
Atrial fibrillation or flutter, % 38.0
37.3
34.3 38.2
0.925
Acute kidney injury, % 72.5
68.4
73.8
0.186
Chronic kidney disease, % 24.8
27.8
23.9
0.342
Depression, % 13.5
18.4
11.9
0.045
Psychoses, % 0.3
0.0
0.4
0.001
Valvular heart disease, % 25.6
25.9
25.4
1
Vasopressor use, % 7.9
6.3
8.3
0.49
Ventilator use, % 51.0
50.0
51.3
0.785
Ischemic stroke, % 7.3
7.6
7.2
0.861
Hemorrhagic stroke, % 3.6
3.2
3.8
1
Systemic thromboembolic event, % 5.7
3.0 7.0
5.4
0.438
Gastrointestinal bleeding, % 7.9
8.9
7.6
0.612
Hemoptysis, % 8.3
7.6
8.5
0.869
Prior venous thromboembolic event, % 3.3
3.8
.4 3.2
0.799
Acute deep venous thrombosis, % 21.5
24.7
4.5 20.5
0.268
Acute pulmonary embolism, % 3.6
3.2
3.8
1
Non–ST-segment–elevation myocardial infarction, % 14.2
13.3
14.5
0.794
ST-segment–elevation myocardial infarction, % 26.3
27.2
26.0
0.757
Blood product transfusion, % 14.2
13.3
14.5
0.794
Right heart catheterization, % 18.3
20.9
17.5
0.347
Percutaneous coronary intervention, % 17.4
15.2
18.1
0.471
Coronary artery bypass grafting, % 9.1
10.1
8.7
0.634
Durable ventricular assist device, % 15.0
26.6
11.3
<0.001
Intraaortic balloon pump, % 30.4
30.4
30.4
34.6 0.539
Impella/TandemHeart, % 14.8
17.7
13.9
0.249
Vascular complications, % 24.2
29.1
22.7
0.11
Major bleeding event, % 57.0
56.3
57.3
0.854
Discharge disposition, % <0.001
Home 20.1
13.3
22.3
Short-term hospital 15.9
7.0
18.7
SNF 31.8
40.5
29.0
Home with HHC 31.6
38.6
29.4
Causes and Predictors of 30-Day Readmission in Patients with Cardiogenic Shock underwent Extra corporeal ECMO.
Abstract:
Background:
Cardiogenic shock is a complication that has significant mortality, morbidity and cost on the health care system. One of the strategies managing cardiogenic shock is the extra corporeal ECMO. Despite the hemodynamic support provided by ECMO the mortality readmission rate is still high and the data to predict outcomes still limited.
Methods and results:
We derived our study cohort of patient with cardiogenic shock who underwent ECMO at the same admission from 2016 to 2017. Healthcare Cost and Utilization Project National Readmission Database. Incidence, mortality, readmission rate and causes of 30-day readmissions were analyzed. From (n=69040) admissions, (n=1641, 2.37%) underwent VA ECMO, (n=907, 55.27%) survived to discharge. (n=661) patient who survived their follow up was available, (n=158, 23.9%) were readmitted within less than 30 days. More than 50% of these readmissions happened within the first 11 days, (7.4%) were died during the readmission hospitalization. Median cost of stay ($ 586,727), the median admission day to ecmo placement was 2 days. Leading cause of readmission was Cardiovascular 31.6% (Heart Failure 24.1%, Arrythmia 20.6%, Neurovascular 10.3%, Hypertension 10.3%, Endocarditis 6.8%), The next leading cause of readmission was the complications of medical/ device care accounted for (17.7%), followed by infection (11.3%), Gastroenteric and liver (10.1%) complications. Discharging the patient to SNF (odds ratio [OR], 3.253; 95% confidence interval [CI], 1.655–6.392; P<0.01), Long term ventricular assisted device (odds ratio [OR], 2.845; 95% confidence interval [CI], 1.517–5.33; P<0.01), Discharging the patient with Home Health Care (odds ratio [OR], 2.466; 95% confidence interval [CI], 1.312–4.634; P<0.05), Vascular Complication (odds ratio [OR], 1.628; 95% confidence interval [CI], 1.007–2.634; P<0.04), Liver disease (odds ratio [OR], 1.575; 95% confidence interval [CI], 1.005–2.469; P<0.01), were among the independent predictors of 30-day readmission. The Length of stay > 8 days (odds ratio [OR], 0.156; 95% confidence interval [CI], 0. 11–0.20; P<0.01), Acute DVT (odds ratio [OR], 0.561; 95% confidence interval [CI], 0.38–0.836; P<0.05), Myocarditis (odds ratio [OR], 0.474; 95% confidence interval [CI], 0.23–0.963; P<0.03), Medicaid as payor (odds ratio [OR], 0.6; 95% confidence interval [CI], 0.42–0.86; P<0.02), Heart Transplant (odds ratio [OR], 0.19; 95% confidence interval [CI], 0.08–0.46; P<0.01) were among the independent predictors of survival. While Age >49 (odds ratio [OR], 2.10; 95% confidence interval [CI], 1.695–2.61; P<0.01), Peripheral Vascular disease (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.03–1.93; P<0.01), Coagulopathy including DIC (odds ratio [OR], 1.35; 95% confidence interval [CI], 1.04–1.76; P<0.02), Liver disease (odds ratio [OR], 1.81; 95% confidence interval [CI], 1.38–2.39; P<0.01), Acute Kidney injury (odds ratio [OR], 2.68; 95% confidence interval [CI], 1.92–3.74; P<0.01), Chronic Kidney Disease (odds ratio [OR], 1.78; 95% confidence interval [CI], 1.30–1.2.44; P<0.01), Hemorrhagic Stroke (odds ratio [OR], 2.05; 95% confidence interval [CI], 1.09–3.84; P<0.02), Hemoptysis (odds ratio [OR], 1.88; 95% confidence interval [CI], 1.21–2.94; P<0.02) were among the independent predictors of mortality.
Conclusion:
Patients with cardiogenic shock underwent ECMO has significant mortality and readmission rate. Yet there is no perfect model that can predict the readmission and mortality for patients with cardiogenic shock who underwent ECMO. Identifying patient at high risk of readmission and mortality might help with improving outcomes.
BACKGROUD:
In progress
METHODE:
The study cohort was derived from Healthcare Cost and Utilization Project National Readmission Database (NRD) of 2016, supported by the Agency for Healthcare Research and Quality. The Nationwide Readmissions Database (NRD) is part of a family of databases and software tools developed for the Healthcare Cost and Utilization Project (HCUP).
The NRD contains data from approximately 17 million discharges each year. Weighted, it estimates roughly 36 million discharges.
The patients can be tracked using linkage number and their admissions, length of stay, comorbidities, procedures are reliably linked in the database.
We queried the NRD database using International Classification of disease 10th revision. We used the diagnosis code of cardiogenic shock (ICD10 CM R57.0) to query only the index admission for the first step and we queried the VA-ECMO procedure code (ICD 10 PSC of 5A1522G. We excluded patients with age ≤18 years and those missing data for age, sex, or mortality. We also excluded patients who were discharged on the month of December as we would not have their readmission information. We studied these patients further and characterized their morbidity and procedure underwent during the index admission and the main diagnosis of readmission using ICD10 CM and ICD10 PSC (available with the supplement). The variables age, sex, discharge disposition, zip code, primary payer, admission status, length of stay, cost of stay, admission day were available from the NRD database.
The primary outcome was 30-day readmission post admission index. We studied the previous factors with their relation to readmission and mortality. Patient who expired during the index admission were excluded from the 30-days readmission.
We used IBM SPSS version 24.0. We expressed the categorical data as percentage and count. Continuous variables as mean ±1 SD. We tested the categorical variables using Chi Square test and the continuous using t test. The demographics data were listed in (table1).
RESULTS:
Baseline Characteristics:
A total of (n=69040) admission identified their index hospitalization complicated by cardiogenic shock, (n=1641, 2.37%) of these admissions had ECMO. Out of these patients
who underwent ECMO, (n=734, 44.7%) survived their index admission.
Mean age of patients was (55.74 ±14.8) years, Median Length of stay was 17 days [6-36] days. (21.7%) of admissions were happened on weekend day, (33.6%) were females, (15.1%) were elective admissions.
Median cost of stay ($ 586,727), the median admission day to ECMO placement was 2 days.
(n=1039, 63.3%) of Cardiogenic shock etiology required ECMO support were ischemic, (n=57, 3.5%) were Myocarditis.
Mortality and predictor of mortality:
Mortality rate was higher among people who were admitted emergently (16.6% vs. 13.2%) p=0.031, underwent CABG (14.3 vs. 9.3%), p=0.002, had AKI (81.1% vs. 72.2%), p<0.001, acute MI (66% vs. 59.9%), p=0.012, Chronic pulmonary disease (19.6% vs. 14.2%), p=0.004, Coagulopathy including DIC (57.3% vs. 51.5%), p=0.019, Liver disease (45.1% vs. 34.9%), p<0.001, Chronic Kidney Disease (30.5% vs. 25.1%), p<0.015, Length of stay > 8 days (44.7% vs. 36.9%), p<0.001,
Mortality rate was lower in patients who had acute DVT (7.5% vs. 20.8%) p<0.001, Myocarditis (2.1% vs. 5.2%), p<0.001.
(n=267, 16.3%) had Impella support, (n=512, 31.2%) had Balloon pump support, (n=164, 10%) underwent Left Ventricular assisted device (LVAD), (n=67, 4.1%) underwent Heart Transplant, (n=198, 12.1%) had CABG. Mortality was lower among patients who underwent LVAD (6.4% vs. 14.4%) p<0.001. Same with patients who underwent heart transplant (1.4% vs. 7.4%) p<0.001. There was slight difference favor Impella (17% vs. 15.4%) but was not statistically significant, for IABP there was no difference in mortality.
The Length of stay > 8 days (odds ratio [OR], 0.156; 95% confidence interval [CI], 0. 11–0.20; P<0.01), Acute DVT (odds ratio [OR], 0.561; 95% confidence interval [CI], 0.38–0.836; P<0.05), Myocarditis (odds ratio [OR], 0.474; 95% confidence interval [CI], 0.23–0.963; P<0.03), Medicaid as payor (odds ratio [OR], 0.6; 95% confidence interval [CI], 0.42–0.86; P<0.02), Heart Transplant (odds ratio [OR], 0.19; 95% confidence interval [CI], 0.08–0.46; P<0.01) were among the independent predictors of survival. While Age >49 (odds ratio [OR], 2.10; 95% confidence interval [CI], 1.695–2.61; P<0.01), Peripheral Vascular disease (odds ratio [OR], 1.4; 95% confidence interval [CI], 1.03–1.93; P<0.01), Coagulopathy including DIC (odds ratio [OR], 1.35; 95% confidence interval [CI], 1.04–1.76; P<0.02), Liver disease (odds ratio [OR], 1.81; 95% confidence interval [CI], 1.38–2.39; P<0.01), Acute Kidney injury (odds ratio [OR], 2.68; 95% confidence interval [CI], 1.92–3.74; P<0.01), Chronic Kidney Disease (odds ratio [OR], 1.78; 95% confidence interval [CI], 1.30–1.2.44; P<0.01), Hemorrhagic Stroke (odds ratio [OR], 2.05; 95% confidence interval [CI], 1.09–3.84; P<0.02), Hemoptysis (odds ratio [OR], 1.88; 95% confidence interval [CI], 1.21–2.94; P<0.02) were among the independent predictors of mortality. (figuerx)
30 days readmission, Cause, length of stay
661 patients who survived their admission were discharged before December of 2016 that we were able to identify their readmission status. (n=158, 23.9%) of these patients readmitted within 30 days.
More than 50% of these readmissions happened within the first 11 days (figure 1), (n= 12, 7.4%) were died during the readmission hospitalization.
Mean age of patients readmitted was (51.3 ± 14.2) years, Median length of stay for the readmission 7 [4-14] days, (12.7%) of these readmissions were electives, (7.6%) of patients readmitted died during the readmission. (34.2%) were females. Median cost of stay for readmission was ($72,715).
Leading causes of readmission were summarized in (figure 2) as the following, Cardiovascular 31.6%, the next leading cause of readmission was the complications of medical/ device care accounted for (17.7%), followed by infection (11.3%), Gastroenteric and liver (10.1%) complications.
The cardiovascular readmission causes summarized in (figure 3) as the following with the leading cause of Heart Failure 24.1%, Arrythmia 20.6%, Neurovascular 10.3%, Hypertension 10.3%, Endocarditis 6.8%.
The Table shows baseline characteristics among (n = 661) patients who survived the index hospitalization for CS underwent ECMO based on whether they were subsequently readmitted within 30 days of discharge compared with patients who were not readmitted.
Patients who their index admissions longer than 30 days, underwent LVAD, has vascular complications, discharged to Skilled Nursing facility or Home Care with agency were more likely to be readmitted. There was no difference between patients who had Mechanical support, chronic medical conditions, ischemic vs. non-ischemic, sex, age or zip codes.
Predictor of 30-days readmission:
On multivariate analysis among those who survived to discharge, discharging the patient to SNF (odds ratio [OR], 3.253; 95% confidence interval [CI], 1.655–6.392; P<0.01), Long term ventricular assisted device (odds ratio [OR], 2.845; 95% confidence interval [CI], 1.517–5.33; P<0.01), Discharging the patient with Home Health Care (odds ratio [OR], 2.466; 95% confidence interval [CI], 1.312–4.634; P<0.05), Vascular Complication (odds ratio [OR], 1.628; 95% confidence interval [CI], 1.007–2.634; P<0.04), Liver disease (odds ratio [OR], 1.575; 95% confidence interval [CI], 1.005–2.469; P<0.01), Patient state at the same hospital state (odds ratio [OR], 2.08; 95% confidence interval [CI], 1.05–4.12; P<0.03) were among the independent predictors of 30-day readmission.
DISCUSSION: