Journal of the American College of Cardiology
Nationwide Trends of Hospital Admission and Outcomes Among Critical Limb Ischemia PatientsFrom 2003–2011
Author + information
- Shikhar Agarwal, MD, MPH,
- Karan Sud, MD and
- Mehdi H. Shishehbor, DO, MPH, PhD∗ (shishem{at}ccf.org)
- Department of Cardiovascular Medicine, Heart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio
- ↵∗Reprint requests and correspondence:
Dr. Mehdi H. Shishehbor, Department of Cardiovascular Medicine, Cleveland Clinic, 9500 Euclid Avenue, J3-05, Cleveland, Ohio 44195.
Abstract
Background Critical limb ischemia (CLI) continues to be a major cause of vascular-related morbidity and mortality in the United States.
Objectives The study sought to characterize the trends in hospitalization of U.S. patients with CLI from 2003 to 2011, using the Nationwide Inpatient Sample. We compared the cost utilization and in-hospital outcomes of endovascular and surgical revascularization procedures for CLI.
Methods CLI and revascularization procedures were identified using International Classification of Diseases-Ninth Edition-Clinical Modification codes. In-hospital mortality and amputation were coprimary outcomes. Length of stay (LOS) and cost of hospitalization were secondary outcomes.
Results We included a total of 642,433 admissions with CLI across 2003 to 2011. The annual rate of CLI admissions has been relatively constant across 2003 to 2011 (∼150 per 100,000 people in the United States). There has been a significant reduction in the proportion of patients undergoing surgical revascularization from 13.9% in 2003 to 8.8% in 2011, while endovascular revascularization has increased from 5.1% to 11.0% during the same time period. This was accompanied by a steady reduction in the incidence of in-hospital mortality and major amputation. Compared to surgical revascularization, endovascular revascularization was associated with reduced in-hospital mortality (2.34% vs. 2.73%, p < 0.001), mean LOS (8.7 days vs. 10.7 days, p < 0.001), and mean cost of hospitalization ($31,679 vs. $32,485, p < 0.001) despite similar rates of major amputation (6.5% vs. 5.7%, p = 0.75).
Conclusions While CLI admission rates have remained constant from 2003 to 2011, rates of surgical revascularization have significantly declined and endovascular revascularization procedures have increased. This has been associated with decreasing rates of in-hospital death and major amputation rates in the United States. Despite multiple adjustments, endovascular revascularization was associated with reduced in-hospital mortality compared to surgical revascularization during 2003 to 2011.
It has been estimated that 10 million individuals are currently suffering from critical limb ischemia (CLI) in the United States (1,2). In addition to pain, decreased quality of life, and immobility, amputation and death are major adverse effects of CLI (3). Previous data from the Nationwide Inpatient Sample (NIS) demonstrated a marked rise in endovascular revascularization during 1996 to 2005 and was temporally associated with a reduction in the rates of major amputation (4). However, this relationship is not causal and may be related to better medical and overall wound care. Importantly, the rise of endovascular procedures, in general, has raised some concerns (5). We conducted a comprehensive analysis using the large nationwide registry from 2003 to 2011 to expand on these previous findings and examine the overall CLI hospital admissions, cost utilization, and outcomes (mortality and major amputations) between surgical and endovascular procedures for CLI.
Methods
Data source
Data were obtained from the NIS from 2003 to 2011, which is sponsored by the Agency for Healthcare Research and Quality as a part of the Healthcare Cost and Utilization Project (HCUP). The NIS across 2003 to 2011 contains discharge level data from approximately 8 million hospitalizations annually from about 1,000 U.S hospitals. The NIS is designed to represent a 20% stratified sample of all hospitals in the country. Criteria used for stratified sampling of hospitals into the NIS include location (urban or rural), teaching status, geographic region, patient volume and hospital ownership. Every hospital has been classified into small, medium, and large size on the basis of the number of beds available for in-hospital admissions. The cutpoints for classification differ according to geographic location of the hospital and the teaching status (6).
Study population
The NIS provides up to 15 diagnoses and 15 procedures for each hospitalization record for the years 2003 to 2009. The number of diagnoses coded in the database was expanded to 25 for the years 2010 to 2011. All these have been coded using the standard International Classification of Diseases-Ninth Edition-Clinical Modification (ICD-9-CM) codes. All adult hospitalizations (>18 years of age) with a diagnosis code corresponding to CLI were included in our study. The list of diagnosis codes used to identify patients with CLI and peripheral artery disease (PAD) is shown in Online Table 1. The first diagnosis in the database is referred to as the principal diagnosis and is considered the primary reason for admission to the hospital. The ICD-9-CM codes for surgical and endovascular procedures performed during the hospitalization are shown in Online Table 2. Sequential revascularization was defined as both endovascular and surgical revascularization performed during a single hospital admission. We used the Charlson Comorbidity Score to quantitate the comorbidity of each admitted patient on the basis of 17 categories of diagnoses (7). In addition, the NIS database provides 29 Elixhauser comorbidities on each hospital admission, on the basis of standard ICD-9-CM codes (8). These were used to derive the prevalence of hypertension, diabetes, obesity, and chronic kidney disease in our population.
Study outcomes
Our study aimed to evaluate the trend in the CLI hospitalization of adults across 2003 to 2011. To calculate the annual rate of admissions among CLI patients, we divided the total of number of hospitalizations in a given year by the U.S. census population for that year and expressed it as a rate per 100,000 U.S. population. This methodology has been described and validated by HCUP and other recent studies (9,10). We also evaluated the changes in the distribution of demographic and clinical characteristics among these patients. In-hospital mortality and major amputation (amputation above the ankle) were coprimary endpoints of our study. Secondary endpoints included resource utilization, which was assessed by evaluating the trends in revascularization procedures, length of stay (LOS), and total cost of hospitalization across the study period. The NIS database provides the total charges associated with each hospital stay that was claimed by the respective hospital. The total charges of each hospital stay were converted to cost estimates using the group average all-payer in-hospital cost and charge information from the detailed reports by hospitals to the Centers for Medicare and Medicaid Services. All costs and charges were then converted to projected estimates for the year 2015, after accounting for annual inflation rates on the basis of consumer price index data available from the Bureau of Labor Statistics (11). Furthermore, we compared the primary and secondary outcomes between the endovascular, surgical and sequential revascularization in the study cohort, across 2003 to 2011.
Statistical analysis
Survey statistics traditionally used to analyze complex semi-random survey designs were employed to analyze these data (12,13). Because the data from NIS represent a collection of scattered hospital clusters, analysis was structured to account for a complex, multistage, probability sampling. NIS recommends the use of “strata” for constructing analysis clusters, which include geographic census region, hospital ownership, teaching status, urban/rural location and bed size. Furthermore, the analysis is further stratified into individual hospitals, which serve as primary sampling units for the analysis. In the NIS, each hospital admission is linked to a “discharge weight” that can be utilized to calculate projected national estimates for all hospital-related outcomes, after accounting for the hierarchical structure of the dataset. To facilitate analysis of trends across multiple years, the Agency for Healthcare Research and Quality has developed new discharge trend weights (TRENDWT) for NIS data to create national estimates for trends analysis (14). Although TRENDWT was developed for trends analysis, it can be used for all analyses as per HCUP. The nonparametric test for trend across ordered groups by Cuzick (15) was used to determine the statistical significance of differences in CLI prevalence across the study period.
Multivariable hierarchical logistic regression analysis with exchangeable matrix was used to determine independent predictors of in-hospital mortality in the study cohort. Covariates included age, sex, race, socioeconomic status (SES) quartile (assessed using median income of the residential zip code), history of prior amputation, revascularization during hospitalization, stump complication, emergency admission, hospital characteristics (region, size, teaching status), comorbidity score, reason for admission, and year of presentation. Because the variable “race” had ∼20% missing data, we used multiple imputation for missing data using ordered logistic regression accounting for the clustered nature of the dataset by incorporating hospital characteristics into the multiple imputation model. The comparison of primary and secondary endpoints between endovascular, surgical and sequential revascularization was performed using similar multivariable regression analysis and covariates. For regression modeling, LOS and cost of hospitalization were logarithm transformed to eliminate the rightward skew. To avoid the possibility of type I error due to multiple testing, we report 99% confidence intervals (CI) for all our regression analyses. All statistical analyses were performed using the statistical software Stata version 13.1 (StataCorp, College Station, Texas).
Subgroup analysis
Subgroup analyses were performed for death, major amputation, likelihood of any revascularization (vs. no revascularization), and likelihood of endovascular revascularization (vs. surgical revascularization) in the study cohort on the basis of a priori–defined subgroups. Subgroups were defined on the basis of age, sex, SES quartiles, race, hospital size, type, and location.
Results
Hospitalization trends
We included a total of 642,433 admissions with CLI across 2003 to 2011, which projected to an estimated population of 2,999,007 admissions nationally across the study duration. The annual rate of CLI admissions has been relatively constant across 2003 to 2011 (Figure 1A). This was despite an increase in the rate of PAD related admissions during 2003 to 2011 (Figure 1B). There was a progressive reduction in the mean age of admitted patients from 2003 to 2011 (Figure 2A). The percentage of patients with CLI <65 years of age admitted to the hospital increased from 34.2% in 2003 to 39.3% in 2011 (p < 0.001) (Figure 2A). Besides this, there was a gradual increase in the proportion of males with CLI admitted to the hospital with a corresponding decrease in the proportion of females hospitalized during the study period (Figure 2B). Table 1 shows the trend of prevalence of cardiovascular risk factors in the study cohort during 2003 to 2011. We noted a progressive increase in hypertension, obesity, diabetes mellitus, chronic kidney disease as well as prior amputation rates from 2003 to 2011.
Annual Rate of CLI and PAD Admissions in the United States
(A) Annual rate of critical limb ischemia (CLI) (B) and peripheral artery disease (PAD) admissions. Bars correspond to the y-axis on the left side of the graph and represent the projected annual number of admissions. The line corresponds to the y-axis on the right side of the figure and represents the admission rate expressed as rate per 100,000 U.S. population calculated using annual census estimates.
U.S. Trends in CLI Hospital Admissions and Outcomes
The salmon line (closed circles) shows the trend of critical limb ischemia (CLI) admissions in the United States during 2003 to 2011. These hospital admissions correspond to the y-axis on the left side of the figure and have been expressed as number of admissions per 100,000 United States population calculated using annual census estimates. All other lines correspond to the y-axis on the right side of the figure and represent the proportion of all CLI admissions expressed as a percentage. During the study period, there was a decline in the surgical revascularization (maroon line [diamonds]) with a corresponding increase in endovascular revascularization (purple line [triangles]). This was associated with a significant decrease in in-hospital mortality (orange line [open circles]) and major amputations (teal line [squares]).
Age and Sex Distribution Among CLI Patients
(A) Mean age (black line, right sided y-axis) as well as the proportion of patients <65 years of age (bars, left sided y-axis) and (B) proportion of males (blue squares) and females (red circles) among patients admitted with critical limb ischemia (CLI).
Trend of Cardiovascular Risk Factors in the Study Cohort Across 2003 to 2011
Figure 3 shows the primary reason for admission to the hospital among patients with CLI. Of all patients, 53.5% of the patients were admitted for primary CLI-related diagnosis. Diabetes mellitus with nonvascular complaints (8.6%), septicemia (5.6%), and post-procedure complications (5.2%) were the other significant reasons for admission to the hospital among these patients. There was also a progressive increase in the proportion of patients admitted on an “emergent” basis (rather than elective) from 68.7% in 2003 to 75.4% in 2011 (p for trend <0.001) (Online Figure 1).
Reasons for Admission
The primary reason for admission to the hospital among patients with critical limb ischemia (CLI). The number beside the each bar denotes the percent of patients presenting with the corresponding principal diagnosis. ARF = acute renal failure; CAD = coronary artery disease; CHF = congestive heart failure; CVA = cerebrovascular accident; MI = myocardial infarction; TIA = transient ischemic attack; VTE = venous thromboembolism.
Outcomes
Over the study duration, there was a steady reduction in the annual incidence of in-hospital mortality and major amputation rates (Figure 4A). The annual in-hospital mortality rate reduced from 5.4% in 2003 to 3.4% in 2011 (p for trend <0.001). Similarly, the annual major amputation rate reduced from 16.7% in 2003 to 10.8% in 2011 (p for trend <0.001). In addition, there was a significant reduction in the LOS (reported in the article as mean ± SE) from 10.0 ± 0.1 days in 2003 to 8.4 ± 0.1 days in 2011 (p for trend <0.001) (Figure 4B). However, there was no significant difference in the cost of hospitalization among CLI patients across the study duration (p for trend = 0.10) (Figure 4B).
Primary and Secondary Outcomes Among CLI Patients
(A) The trend of annual incidence of in-hospital mortality and major amputation rates among patients hospitalized with critical limb ischemia (CLI). (B) The trend of annual mean length of stay (black line, right sided y-axis) and mean cost of hospitalization (bars, left sided y-axis) among patients hospitalized with CLI.
Table 2 shows the predictors of in-hospital mortality derived from multivariable hierarchical logistic regression analysis. Older age and female sex were associated with increased in-hospital mortality. We did not observe any impact of race and SES upon in-hospital mortality. Furthermore, emergent admissions were associated with significantly higher mortality as compared to elective admissions. A primary presentation of septicemia, congestive heart failure, and respiratory disease were noted to be strongly predictive of in-hospital mortality. Presence of stump complication during hospital admission was also independently predictive of in-hospital mortality. Importantly, any revascularization was associated with significantly reduced in-hospital mortality in the study cohort. There was a significant impact of hospital characteristics on in-hospital mortality among CLI patients (Table 2).
Multivariable Hierarchical Logistic Regression Model for In-Hospital Mortality Among Patients With CLI Admitted to Hospitals During 2003 to 2011
Comparison of revascularization strategies
Figure 5 shows the trend of surgical revascularization (Figure 5A), endovascular revascularization (Figure 5B), sequential revascularization (Figure 5C), and revascularization procedures with minor amputations (Figure 5D) across the study period. There was a significant reduction in the proportion of patients undergoing surgical revascularization from 13.9% in 2003 to 8.8% in 2011 (p for trend <0.001). This was accompanied by a corresponding increase in endovascular revascularization from 5.1% in 2003 to 11.0% in 2011 (p for trend <0.001). Besides this, there was a small but statistically significant increase in the incidence of sequential revascularization procedures from 1.8% in 2003 to 2.2% in 2011 (p for trend = 0.01). However, there was no significant change in the incidence of revascularization procedures along with minor amputations during 2003 to 2011 (p for trend = 0.43). Figure 6 shows the trend of primary and secondary endpoints among patients undergoing endovascular, surgical, and sequential revascularization across 2003 to 2011. Compared to surgical revascularization, endovascular revascularization was associated with significantly reduced in-hospital mortality (2.34% vs. 2.73%; adjusted odds ratio [OR]: 0.69; 99% CI: 0.62 to 0.77) (Figure 6A). Both endovascular and surgical revascularization were associated with similar rates of major amputation (6.50% vs. 5.65%; adjusted OR: 0.99; 99% CI: 0.91 to 1.07) (Figure 6B). LOS among patients undergoing endovascular revascularization was 8.70 ± 0.12 days, which was significantly lower than the LOS in the surgical revascularization cohort (10.70 ± 0.13 days; adjusted OR: 0.80; 99% CI: 0.79 to 0.81) (Figure 6C). Similarly, the cost (mean ± SE) of hospitalization among patients undergoing endovascular revascularization was $31,679 ± $449, which was significantly lower than the cost of hospitalization in the surgical revascularization cohort ($32,485 ± $550; adjusted OR: 0.95; 99% CI: 0.94 to 0.96) (Figure 6D). Compared to endovascular and surgical revascularization, sequential revascularization was associated with significantly higher in-hospital mortality, major amputation, and cost of hospitalization (Figure 6). Although the mean LOS among patients undergoing sequential revascularization was higher than those undergoing endovascular revascularization, it appeared similar between the sequential and surgical revascularization groups (Figure 6).
Trends of Revascularization Strategies Among CLI Patients
Trends of (A) surgical revascularization, (B) endovascular revascularization, (C) sequential revascularization, and (D) revascularization procedures with minor amputations among patients hospitalized with critical limb ischemia (CLI) during 2003 to 2011. Bars denote the estimate expressed as a percent of all admissions and correspond to the left sided y-axis. Black line denotes the trend of the projected number of revascularization procedures performed annually and corresponds to the right-sided y-axis.
Trends in Primary and Secondary Outcomes
Comparison of primary and secondary outcomes among endovascular (circles), surgical (squares), and sequential revascularization (triangles) procedures. A to D demonstrate the trend of annual in-hospital mortality, major amputation, mean length of stay, and mean cost of hospitalization among patients undergoing endovascular, surgical, and sequential revascularization procedures, respectively. Only those patients who underwent revascularization during the hospitalization were included in this analysis. All adjusted estimates were calculated after adjusting for age, sex, imputed race, socioeconomic status, emergency admits, hospital location, hospital size, hospital teaching status, insurance status, Charlson Comorbidity Score, and primary reason for admission. 1) Adjusted effect estimate (with 99% CIs) comparing endovascular with surgical revascularization; 2) adjusted effect estimate (with 99% CIs) comparing endovascular with sequential revascularization; and 3) denotes adjusted effect estimate (with 99% CIs) comparing surgical with sequential revascularization. CI = confidence interval; OR = odds ratio.
Subgroup analysis
Table 3 shows the subgroup analyses for death, major amputation, likelihood of any revascularization and likelihood of endovascular revascularization on the basis of a priori–defined subgroups. Compared to younger patients, older patients had a higher risk of death or major amputation and were more likely to undergo a revascularization procedure. Females had higher in-hospital mortality but a lower rate of major amputation compared to males. Among those who underwent a revascularization procedure, females were more likely to undergo endovascular intervention as compared to males. Although there was no significant impact of race on in-hospital mortality, blacks and other races had significantly higher rates of major amputation and lower rates of revascularization, compared to whites. Similarly, higher SES was associated with lower rates of major amputation, compared to lower SES categories. There was a significant heterogeneity apparent in outcomes between the different hospitals. Urban teaching hospitals had significantly higher rates of in-hospital mortality, major amputation as well as revascularization rates compared to rural hospitals. Similarly, large-sized hospitals had higher rates of in-hospital mortality, major amputation and revascularization rates compared to small-sized hospitals. Furthermore, there were significant differences in hospital outcomes between various geographic regions, shown in Table 3.
Subgroup Analyses for Death, Major Amputation, Likelihood of Any Revascularization, and Likelihood of Endovascular Revascularization in the Study Cohort on the Basis of A Priori–Defined Subgroups
Online Figure 2 shows the changing trend of CLI admissions among blacks (Online Figure 2A), uninsured patients (Online Figure 2B), patients with Medicaid as primary insurance (Online Figure 2C) and those residing in lowest SES quartile zip codes (Online Figure 2D). During the course of the study duration, there has been an increase in the proportion of uninsured patients (p for trend <0.001) as well as patients with Medicaid as primary insurance (p for trend <0.001). However, there have been no significant changes in trends in admission of black patients (p for trend = 0.65) or those residing in lowest quartile SES zip codes (p for trend = 0.62).
Discussion
In the current study we evaluated and compared the overall trends, outcomes, and resource utilization among endovascular, surgical, and sequential revascularization procedures among patients admitted with CLI in the contemporary era using a large comprehensive nationwide database. We have several important findings (Central Illustration). First, the annual rate of CLI admissions has been relatively constant across 2003 to 2011, despite a progressive increase in the rate of PAD admissions. Second, besides primary CLI related diagnoses, diabetes mellitus with nonvascular complaints, septicemia and post-procedure complications were the other significant reasons for admission to the hospital among CLI patients. Third, there was a progressive increase in the proportion of patients admitted on an emergent basis during 2003 to 2011. Fourth, there has been a reduction in the proportion of patients undergoing surgical revascularization, accompanied by a corresponding increase in endovascular revascularization during the study duration. Fifth, there was a reduction in the incidence of in-hospital mortality, major amputation as well as mean LOS during the study duration. Last, despite similar rates of major amputation, the endovascular approach was associated with reduced in-hospital mortality, mean LOS and mean cost of hospitalization compared to surgical revascularization, after multiple adjustments.
Several studies have examined patterns in treatment of PAD in the past (16–20). Rowe et al. (4) demonstrated a progressive decline in surgical revascularization accompanied by an increase in the rates of endovascular revascularization during 1996 to 2005. Similarly, Nowygrod et al. (18) demonstrated an increased utilization of endovascular procedures combined with a decrease in open revascularization surgery and a decrease in major amputation rates from 1998 to 2003. Although both studies showed decreasing amputation rates with increasing endovascular procedures, this relationship has in general been considered noncausal and likely attributable to aggressive medical therapy and wound care. However, our results indicate that the annual rate of admission among CLI patients remained relatively constant. Interestingly, we noted that there was a significant increase in PAD related admissions during 2003 to 2011. This rise in PAD might be secondary to the rise in the prevalence of cardiovascular risk factors, as observed in our study. An increase in the rate of PAD related admissions, with a constant rate of CLI related admissions, might suggest early detection of PAD among population by their physicians, leading to improved management and the relative stabilization of the CLI rates. In addition, we noted a reduction in the amputation rates over time, despite a constant rate of CLI and an increasing rate of PAD related admissions. This suggests that other factors beyond medical therapy might be responsible for decreasing rates of amputation over this period. This may be secondary to increasing availability and improvement in endovascular revascularization techniques in the United States (21). We observed a progressive increase in the proportion of patients admitted on an emergent basis rather than elective basis across 2003 to 2011. This may be secondary to higher rates of revascularization in the elderly with more complex disease, changing patterns in presentation among CLI patients, or because of changing patterns in insurance coverage among patients making emergency room their primary source of healthcare. Regardless of the reasons, this underscores the increasing healthcare cost and burden imposed by these patients and the need for increased emphasis on quality of care (22).
Approximately one-half of the hospital admissions in our study population were related to primary CLI-related diagnoses. The remaining one-half of admissions was due to other non–CLI-related causes. In addition, we recognized that several of these non-CLI diagnoses such as acute myocardial infarction, cerebrovascular accident, respiratory disease, congestive heart failure, and acute kidney injury were associated with significantly elevated risk for in-hospital mortality. It must be emphasized that patients with CLI harbor an elevated atherosclerotic burden in multiple vascular beds, predisposing them to increased vascular morbidity and mortality (1,2). Multiple studies have demonstrated marked escalation of risk of short- and long-term mortality with coexisting CLI and other atherosclerotic diseases such as coronary artery disease or cerebrovascular disease (6,23–27).
There have been multiple studies in the last decade examining various aspects of PAD and CLI in the last decade (9,28–33). Similar to earlier studies examining trends among PAD and CLI patients, we observed a continued trend towards a decline in in-hospital mortality and major amputation rates among patients admitted with CLI. We observed impressive reduction in in-hospital mortality (relative reduction 37%) and major amputation (relative reduction 35%) over the course of the study duration. This period also witnessed an increase in the number of endovascular revascularization procedures accompanied by a simultaneous decline in the number of surgical revascularization procedures. Review of the current literature shows heterogeneous estimates for proportion of patients with CLI undergoing revascularization (34–36). Although several of these studies are robust in terms of included data and research conduct, the differences from our study might be secondary to different treatment practices in the United States versus other countries. In addition to international differences, there has been a well-documented regional variation in the use of lower extremity revascularization procedures for CLI in the United States (34,37). As noted in these studies, the intensity of vascular care including the percentage of revascularization procedures among CLI patients varied significantly among various U.S. regions. On an average, the proportion of Medicare beneficiaries with CLI (requiring amputation) who underwent revascularization was rather low during 2003 to 2006 (7.8% endovascular and 3.3% surgical).
The last decade also witnessed a substantial decrease in the mean LOS among patients admitted with CLI. Although the adjusted cost of hospitalization appeared to be relatively unchanged among the entire cohort during 2003 to 2011, endovascular revascularization procedures seemed to be associated with smaller cost of hospitalization as compared to surgical revascularization procedures. This finding was similar to that reported in the resource utilization analysis from the BASIL (Bypass versus Angioplasty in Severe Ischemia of the Leg) trial, where bypass surgery was associated with higher inpatient treatment costs as compared to balloon angioplasty at 1-year follow up (38). Patient-specific variables along with outcomes over a 30-day follow-up period, especially readmission rates, would be necessary to understand implications of these observations in a definitive manner. Further exploration of these matters would be vital with the widespread implementation of the Affordable Care Act and as the penalties for 30-day readmissions are invoked universally (39,40).
Despite similarities to earlier studies, our study has several unique features, which provide important insight into the contemporary management of CLI. Our study uses a large (n = 642,433) well-validated dataset with most contemporary (2003 to 2011) analyses providing insight into the latest trends in prevalence, outcomes, and resource utilization in CLI. Unlike several prior studies, our study is on the basis of an all insured as well as uninsured patients rather than being limited to Medicare or insured, patient populations only (28,30,33). The majority of studies in this area are heterogeneous with inclusion of patients with PAD and CLI (9,28,30,33). In contrast, our study provides insights and trends specific to patients with CLI. In addition, our manuscript has evaluated both cost and LOS associated with open surgery and endovascular revascularization among patients admitted with CLI. Furthermore, to the best of our knowledge, our study is the only one that has presented trends of utilization of sequential therapy as well as revascularization in conjunction with minor amputation. Sequential revascularization strategy was associated with higher in-hospital mortality, major amputation as well as higher cost of hospitalization as compared to endovascular or surgical revascularization. This was probably because of a higher disease burden in these patients, necessitating multiple treatment strategies for complete revascularization.
In addition to a detailed perspective on the trends in treatment and outcomes among CLI patients, our study also provides some additional data that might form the basis of future studies. There were sex-based disparities in treatment and outcomes evident in our analysis. Females were observed to have higher in-hospital mortality but a lower rate of major amputations compared to males. Among those who underwent a revascularization procedure, females were more likely to undergo endovascular intervention as compared to males. These observations are similar to those reported by Lo et al. (21) using NIS data from an earlier time period. Possibly, the inverse association of CLI admissions with improved medical therapy remains hypothetical. Other factors such as socioeconomic factors, access to health care system, and degree of people with adequate health insurance may be important in mediating these associations. We found racial and SES-based disparities evident in our cohort. Compared to white patients, blacks had a higher incidence of major amputations and lower incidence of revascularization procedures. Similarly, low-SES patients had higher rates of major amputation compared to high-SES patients. The impact of SES on medical care delivery and outcomes has been extensively explored in the area of coronary artery disease and cerebrovascular disease (12,13). It is very likely that several factors that govern this relationship in other atherosclerotic diseases also apply to CLI. However, these have yet to be comprehensively and systematically evaluated. Furthermore, our study demonstrated significant institutional heterogeneity in the treatments and outcomes in CLI patients. Disparities in outcomes between teaching and nonteaching hospitals as well as large and small hospitals can be attributed to increased complexity of the case mix at larger tertiary care hospitals. In addition, our study demonstrated significant differences in mortality, amputation rates and revascularization practices across various geographic regions. It is unclear if these differences were secondary to an increased prevalence of CLI in particular regions, differences in presentation/severity, differences in risk factor profiles, differences in regional patterns of care, or a combination of these factors.
Study limitations
Our study has several important limitations that are inherent to large administrative databases. First, there may be errors in coding of diseases or procedures; however, these would most likely impact both surgical and endovascular revascularization. Second, because the unit of analysis in the NIS database is “unique admission” rather than “unique patient,” it is possible that one patient might have been represented more than once, in case of repeat admission for recurrent admissions. Third, this is a retrospective observational study, which may be subject to traditional biases of observational studies such as selection bias. Fourth, the database does not capture the procedures that were performed on an outpatient visit. More recently, stand-alone outpatient facilities have increased, and more complex patients may be treated in these centers. Last, NIS does not provide details about anatomic characteristics that are important in deciding the mode of revascularization in CLI patients. It was also not possible to determine the type and invasiveness of the surgical or endovascular therapy using the NIS database. It is possible that simple lesions were preferentially treated with endovascular therapy, whereas more complex lesions were treated using surgical therapy, leading to obvious differences in outcomes. Alternatively, it may be likely that the findings underestimate the impact of endovascular therapy, as sicker patients with higher comorbidities and poor targets were more likely to undergo endovascular procedures. Although the comparison of outcomes was adjusted for Charlson Comorbidity Score, it is possible that differences might arise due to residual confounding. Furthermore, due to changes in coding practices during the study duration, the data on smoking has been traditionally felt to be unreliable in the NIS database. Hence, trends on smoking prevalence were not reported, as these were felt to be inaccurate.
Conclusions
CLI is a complex disease with significant morbidity and mortality. There has been a significant increase in the utilization of endovascular revascularization with a corresponding decline in surgical revascularization, accompanied by a decrease in in-hospital death and major amputation rates in the United States during 2003 to 2011. Despite similar rates of major amputation, endovascular revascularization was associated with reduced in-hospital mortality, mean LOS and mean cost of hospitalization compared to surgical revascularization. Although the results are encouraging, there remain significant disparities and gaps in the care of CLI patients that must be addressed.
COMPETENCY IN SYSTEMS-BASED PRACTICE: Despite an increase in hospital admissions for PAD, the number of admissions for CLI remained constant during the years 2003 to 2011. Among patients admitted with CLI over this period, there was an increase in endovascular revascularization, decline in surgical revascularization, decrease in major amputations and lower mortality.
TRANSLATIONAL OUTLOOK: Further studies are needed to understand the reasons for and consequences of disparities in risk factor profiles, severity of ischemia at presentation, patterns of care, and outcomes between various geographic regions and types of hospitals.
Appendix
Footnotes
All authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Listen to this manuscript's audio summary by JACC Editor-in-Chief Dr. Valentin Fuster.
- Abbreviations and Acronyms
- CI
- confidence interval
- CLI
- critical limb ischemia
- HCUP
- Healthcare Cost and Utilization Project
- ICD-9-CM
- International Classification of Diseases-Ninth Edition-Clinical Modification
- LOS
- length of stay
- NIS
- Nationwide Inpatient Sample
- OR
- odds ratio
- PAD
- peripheral artery disease
- SES
- socioeconomic status
- Received January 30, 2016.
- Revision received February 9, 2016.
- Accepted February 11, 2016.
- American College of Cardiology Foundation
References
Figures in Article
Figures
Supplementary Materials
- Online Tables 1 and 2 and Online Figures 1 and 2[S0735109716009086_mmc1.pdf]
Audio Summary