Abstract
Background
Recent improvement in the care of patients with myocardial infarction should lead to better outcome. The goal of this study was to evaluate the incidence of all cause cardiogenic shock (CS) and CS occurring in the setting of ST elevation myocardial infarction (STEMI) in the United States.
Method
The Nationwide Inpatient Sample (NIS) database was utilized to calculate the age-adjusted incident rate of CS from 1996 to 2006 based on ICD-9 coding in the setting of STEMI. Furthermore, we evaluated this trend based on race and gender.
Results
A total population of 52,784,917 patients was available between 1996 and 2006. We found that the incidence of all cause CS has not changed over time. However, in the setting of STEMI, CS has been declining slowly over the last 10 years. The age-adjusted rate for CS was 4.3 per 100,000 in 1996 which remained steady with an incidence of 3.1 per 100.000 in 2006 (p < 0.01). This decline was persistent across different race or gender. However, African Americans and female gender had persistently lower rate of CS.
Conclusion
Advancement in the treatment of acute STEMI has led to gradual reduction in the incidence of STEMI related cardiogenic shock irrespective of ethnicities or gender suggesting improving outcome of patients presenting with STEMI in recent years.
Highlights
- •
We found that the incidence of cardiogenic shock (CS) in the setting of STEMI has been declining slowly over the last 10 years.
- •
This decline was persistent across different race or gender. However, female gender had persistently lower incident of age adjusted STEMI associated cardiogenic shock in comparison to male gender.
- •
This decline occurred regardless of race with persistently lower incidence of STEMI associated CS in African Americans similar to female gender.
- •
On the other hand, total incidence of all cause cardiogenic shocks has not changed over the years studied irrespective of race or gender.
1
Background
Cardiogenic shock (CS) is a serious complication of acute ST elevation myocardial infarction (STEMI) . Many cases of CS in the setting of STEMI are secondary to mechanical complications such as severe left ventricular dysfunction, papillary muscle rupture, inter-ventricular septal defect or free wall rupture. Right ventricular infarction is another serious cause of cardiogenic shock in the setting of STEMI. CS develops in about 5–8% of STEMI cases and 2.5% of non-STEMI patients . The overall in-hospital mortality remains very high ranging from 55 to72% . Over the last two decades, there has been a growing evidence that early reperfusion lead to better survival and lower rate of complications in patients with STEMI . Guidelines were developed to improve outcome in patients with STEMI. However, little is known about the true implementation and benefits of these recommendations. Previous studies have shown that the development of guidelines does not necessarily mean significant changes in clinical practice . With rapid reperfusion, we expect lower incidence of CS in the setting of STEMI that should lead to improvement in outcome. In order to study whether the implementation of new evidence based guidelines resulted in reduced incidence of CS in the setting of STEMI, we retrospectively analyzed a large database. Our hypothesis is that the recent improvements in the care of STEMI patients should lead to a reduction in the incidence of CS in the recent years.
2
Methods
2.1
Data collection and data sources
In order to study trends in the incidence of CS in the United States, we decided to analyze the Nationwide Inpatient Sample (NIS) database over a ten year period (1996 to 2006). The NIS is a large inpatient database developed by the Healthcare Cost and Utilization Project (HCUP). NIS database is used by researchers to identify and analyze national trends in health care utilization and patient outcomes. The NIS is one of the largest inpatient database containing over 8 million hospital stays each year in the United States. Using the available NIS database from the years 1996 to 2006, we analyzed the trends of CS over time. The NIS contains information on all patients regardless of insurance including Medicare, Medicaid or private insurance companies. It contains approximately a 20-percent stratified sample of patients with primary and secondary diagnosis.
2.2
Description of the data
For our study we utilized the International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM) primary code for cardiogenic shock (785.51). Furthermore, regarding STEMI related CS, we utilized ICD-9 codes that were consistent with STEMI in the NIS database. We used age over 40 in order to reduce the number of non-atherosclerotic causes of STEMI. We used the following codes consistent with the diagnosis of STEMI: True posterior wall infarction (410.61), acute myocardial infarction (AMI) of anterolateral wall (410.01), infarction of other anterior wall (410.11), infarction of infero-lateral wall (410.21), infarction of infero-posterior wall (410.31), and other inferior wall (410.41)/lateral wall (410.51) infarction; also various specified sites including infarction of atrium, papillary muscle, septum alone or STEMI in other specified sites (410.81) were present. Furthermore, we obtained demographic information, such as age and gender in order to calculate the age adjusted incidence of CS per 100,000 patients for the years 1996–2006. Furthermore, we calculated the total number of CS over the years studied.
2.3
Statistical analysis
The average age adjusted incidence rates from all cause and STEMI related CS were calculated by multiplying age specific mortality rates with age specific weights. The age specific weights were obtained from the year 2000 data, approximating the US population age brackets. The weighted rates were summed across age groups to give the age adjusted rate for each year from 1996 to 2006. We utilized independent samples t-test for our analysis. The Statistical Package for Social Sciences (SPSS) software was used for our calculation. We used stepwise multiple regression analysis for multivariate adjustment. A p value of less than 0.05 was accepted as statistically significant.
2
Methods
2.1
Data collection and data sources
In order to study trends in the incidence of CS in the United States, we decided to analyze the Nationwide Inpatient Sample (NIS) database over a ten year period (1996 to 2006). The NIS is a large inpatient database developed by the Healthcare Cost and Utilization Project (HCUP). NIS database is used by researchers to identify and analyze national trends in health care utilization and patient outcomes. The NIS is one of the largest inpatient database containing over 8 million hospital stays each year in the United States. Using the available NIS database from the years 1996 to 2006, we analyzed the trends of CS over time. The NIS contains information on all patients regardless of insurance including Medicare, Medicaid or private insurance companies. It contains approximately a 20-percent stratified sample of patients with primary and secondary diagnosis.
2.2
Description of the data
For our study we utilized the International Classification of Diseases, ninth revision, Clinical Modification (ICD-9-CM) primary code for cardiogenic shock (785.51). Furthermore, regarding STEMI related CS, we utilized ICD-9 codes that were consistent with STEMI in the NIS database. We used age over 40 in order to reduce the number of non-atherosclerotic causes of STEMI. We used the following codes consistent with the diagnosis of STEMI: True posterior wall infarction (410.61), acute myocardial infarction (AMI) of anterolateral wall (410.01), infarction of other anterior wall (410.11), infarction of infero-lateral wall (410.21), infarction of infero-posterior wall (410.31), and other inferior wall (410.41)/lateral wall (410.51) infarction; also various specified sites including infarction of atrium, papillary muscle, septum alone or STEMI in other specified sites (410.81) were present. Furthermore, we obtained demographic information, such as age and gender in order to calculate the age adjusted incidence of CS per 100,000 patients for the years 1996–2006. Furthermore, we calculated the total number of CS over the years studied.
2.3
Statistical analysis
The average age adjusted incidence rates from all cause and STEMI related CS were calculated by multiplying age specific mortality rates with age specific weights. The age specific weights were obtained from the year 2000 data, approximating the US population age brackets. The weighted rates were summed across age groups to give the age adjusted rate for each year from 1996 to 2006. We utilized independent samples t-test for our analysis. The Statistical Package for Social Sciences (SPSS) software was used for our calculation. We used stepwise multiple regression analysis for multivariate adjustment. A p value of less than 0.05 was accepted as statistically significant.
3
Results
A total population of 52,784,917 patients was available between 1996 and 2006. We found that the incidence of cardiogenic shock in the setting of STEMI has been declining slowly over the last 10 years. The age-adjusted rate for cardiogenic shock was 4.3 per 100,000 in 1996 which declined steadily with an incidence of 3.1 per 100.000 in 2006 (p < 0.01, Fig. 1 ). This decline was persistent across different race or gender. However, female gender had persistently lower incident of age adjusted STEMI associated cardiogenic shock in comparison to male gender (incidence of STEMI associated CS in female gender in 2006 was 3.2 per 100,000 with persistent decline to lowest level in 2006 to 1.9 per 100,000. For male gender, incidence of STEMI associated CS declined from 5.4 in 100,000 in 1996 to 4.4 per 100,000 in 2006, p < 0.01, Fig. 1 ). This decline occurred regardless of race with persistently lower incidence of STEMI associated CS in African Americans similar to female gender ( Fig. 2 ). On the other hand, total incidence of all cause cardiogenic shocks has not changed over the years studied irrespective of race or gender (total incidence for all cause CS was 10.08 per 100,000 in 1996 with no significant changes over the 10 years period with incidence of all cause CS of 11.8 per 100,000 in year 2006. However, similar to STEMI associated CS, African American race and female gender had always lower incidence of all cause CS ( Figs. 3–4 ).
