Circ J doi: 10.1253/circj.CJ-18-0289 Advance Publication by-j-stage ORIGINAL ARTICLE Peripheral Vascular Disease Length and Cost of Hospital Stay in Poor-Risk Patients With Critical Limb Ischemia Undergoing Revascularization Mitsuyoshi Takahara, MD, PhD; Osamu Iida, MD; Yoshimitsu Soga, MD, PhD; Nobuyoshi Azuma, MD, PhD; Shinsuke Nanto, MD, PhD on behalf of the PRIORITY Investigators Background: The aim of the current study was to identify the distribution of length and cost of hospital stay and their associated risk factors in poor-risk Japanese critical limb ischemia (CLI) patients undergoing revascularization. Methods and Results: We analyzed prospectively collected data from 507 CLI patients who required assistance in their daily lives due to disability in activities of daily living and/or cognitive function impairment and who underwent revascularization. The median length and cost of hospital stay were 23 days (IQR, 9 52 days) and 2.25m (IQR, 1.33m 3.58m), respectively. Reduced albumin, tissue loss, infection, surgical reconstruction, and bilateral revascularization were associated with prolonged hospital stay (P=0.012, 0.019, <0.001, <0.001, and <0.001, respectively). Doubling the length of the hospital stay was associated with a 44% increase in hospital cost. Regular dialysis, surgical reconstruction, and bilateral revascularization were independently associated with an approximately 20% increase in the cost of hospital stay (all P<0.001). Conclusions: Length and cost of hospital stay varied considerably between patients. Low serum albumin, tissue loss, infection, surgical reconstruction, and bilateral revascularization were associated with longer hospital stay. Regular dialysis, surgical reconstruction, and bilateral revascularization were independently associated with an approximately 20% increase in the cost of hospital stay. Key Words: Critical limb ischemia; Hospital cost; Length of hospital stay; Poor risk Critical limb ischemia (CLI) is the most advanced form of peripheral arterial disease, with ischemic rest pain or ulcer/gangrene. CLI has an extremely poor prognosis, and clinical guidelines have recommended revascularization as the first-line treatment whenever feasible. 1,2 Revascularization requires substantial medical resources, 3 and comorbidity is in general known to raise medical costs. 4 To date, however, limited data are available on medical costs in poor-risk patients undergoing revascularization. In the current era of population aging, the number of elderly and frail patients with comorbidities is increasing. Poor-risk CLI patients are expected to have a high perioperative risk, leading to prolonged hospital stay and increased health-care costs. The aim of the current study was to identify the distribution of length and cost of hospital stay and their associated risk factors in poor-risk Japanese CLI patients undergoing revascularization. Methods Subjects This was a subanalysis of the Poor-Risk patients with and without RevascularIzation TherapY for critical limb ischemia (PRIORITY) registry, which was a prospective multicenter study registering CLI patients who required assistance in their daily lives because of disabilities in activities of daily living (ADL) and/or impairment of cognitive function. The details of the study protocol have been described previously. 5 The study was performed according to the Declaration of Helsinki and was approved by the ethics committee of each institution registering study subjects. Written informed consent was obtained from each participant or, if that was not feasible, from his/her family. A total of 551 patients underwent revascularization, and data on length and cost of hospital stay were available for 507 of the 551 patients (92%). The current study therefore analyzed data on the length and cost of hospital stay in 507 patients. Received March 12, 2018; revised manuscript received May 28, 2018; accepted June 5, 2018; released online July 14, 2018 Time for primary review: 18 days Department of Diabetes Care Medicine (M.T.), Department of Metabolic Medicine (M.T.), Osaka University Graduate School of Medicine, Suita; Cardiovascular Center, Kansai Rosai Hospital, Amagasaki (O.I.); Department of Cardiology, Kokura Memorial Hospital, Kitakyushu (Y.S.); Department of Vascular Surgery, Asahikawa Medical University, Asahikawa (N.A.); and Department of Cardiology, Nishinomiya Municipal Central Hospital, Nishinomiya (S.N.), Japan Mailing address: Osamu Iida, MD, Cardiovascular Center, Kansai Rosai Hospital, 3-1-69 Inabaso, Amagasaki 660-8511, Japan. E-mail: iida.osa@gmail.com ISSN-1346-9843 All rights are reserved to the Japanese Circulation Society. For permissions, please e-mail: cj@j-circ.or.jp
TAKAHARA M et al. Table 1. Baseline CLI Patient Characteristics n 507 Age (years) 76±10 Male 300 (59) Receiving welfare 46 (9) Diabetes mellitus 323 (64) Regular dialysis 234 (46) Ambulation Level 1: Bedridden 54 (11) Level 2: In a wheelchair with assistance 103 (20) Level 3: In a wheelchair without assistance 36 (7) Level 4: Ambulatory with the use of equipment 109 (22) Level 5: Ambulatory without any use of 204 (40) equipment Home-boundness Level 1: Leaving home once or more per week 287 (57) Level 2: Leaving the bedroom without 42 (8) assistance once or more per week Level 3: Leaving the bedroom with assistance 119 (24) once or more per week Level 4: Leaving the bedroom less than once 56 (11) per week Independence in ADL (Katz index) 3.9±2.4 Width of life space (LSA) 38±33 Cognitive impairment (GDS) 3.2±2.0 BMI (kg/m 2 ) 21.2±3.8 Hemoglobin (g/dl) 11.1±1.8 Total lymphocyte count (10 3 /μl) 1.2±0.6 Albumin (g/dl) 3.3±0.6 Cholinesterase (U/L) 208±74 LVEF (%) 61±12 Log(BNP (pg/ml)) 5.0±1.8 Tissue loss 425 (84) Infection 142 (28) Bilateral revascularization 135 (27) Surgical reconstruction 30 (6) Data given as mean ± SD or n (%). Data were unavailable on BMI in 8 patients (2%), ambulation in 1 patient (0.2%), home-boundness in 3 patients (0.6%), Katz index in 2 patients (0.4%), LSA in 3 patients (0.6%), total lymphocyte count in 15 patients (3%), serum albumin in 10 patients (2%), cholinesterase in 104 patients (21%), LVEF in 56 patients (11%), and BNP in 97 patients (19%). ADL, activities of daily living; BMI, body mass index; BNP, brain natriuretic peptide; CLI, critical limb ischemia; GDS, Global Deterioration Scale; LSA, Life-Space Assessment; LVEF, left ventricular ejection fraction. Definitions CLI was diagnosed if the patients met at least 1 of the following criteria: (1) chronic ischemic foot rest pain with ankle pressure <50 mmhg, toe blood pressure <30 mmhg, or skin perfusion pressure 30 mmhg; (2) ischemic foot ulcer or gangrene with ankle pressure <70 mmhg, toe blood pressure <50 mmhg, or skin perfusion pressure 40 mmhg; or (3) other modalities indicating apparent critical ischemia complicated by ischemic foot rest pain, ulcer, or gangrene. 1 ADL and mobility before CLI onset were assessed using the modified Katz index of independence, 6 an ambulation scale, a home-boundness scale, and the Life-Space Assessment (LSA). 7 The score of the modified Katz index, which assesses the number of activities (bathing, dressing, toileting, transferring, continence, and feeding) performed independently, ranges from 0, indicating total dependence, to 6, indicating full independence. Ambulation was classified into the following 5 levels: ambulatory without any use of equipment, level 5; ambulatory with the use of equipment, level 4; in a wheelchair without assistance (driving his/her wheelchair by him/herself), level 3; in a wheelchair with assistance (pushed in a wheelchair), level 2; and bed-ridden, level 1. The levels of home-boundness were defined as leaving home once or more per week, level 1; leaving the bedroom without assistance once or more per week, level 2; leaving the bedroom with assistance once or more per week, level 3; and leaving the bedroom less than once per week (regardless of assistance), level 4. The LSA score ranged from 0, totally bedbound; to 120, traveled out of town every day without assistance. Nutritional status was evaluated based on body mass index (BMI), hemoglobin, serum albumin, serum cholinesterase, and total lymphocyte count. Cognitive function was evaluated according to the Global Deterioration Scale, 8 with grades ranging from stage 1 (no dementia) to stage 7 (late-stage dementia). Cardiac function was measured using left ventricular ejection fraction and log-transformed circulating brain natriuretic peptide (BNP). Statistical Analysis Data are presented as mean ± SD for continuous variables and as percentage for categorical variables unless otherwise indicated. P<0.05 was considered statistically significant. The length and cost of hospital stay, as well as circulating BNP, had right-skewed distributions and were therefore log-transformed. The correlation between length and cost of hospital stay was assessed with Pearson s correlation coefficient (r) after log-transformation. We explored the association of baseline characteristics with prolonged hospital stay and increased hospital cost. Given that the CLI management strategy, which would considerably influence the length of hospital stay and hospital cost, was expected to differ between the participating institutions, we adopted the linear mixed model in which inter-institution variability was entered as a random effect. First, the crude association of baseline characteristics with length of hospital stay was investigated using a linear mixed model that included the log-transformed length of hospital stay as the dependent variable, each of the baseline characteristics as the fixed effect, and the inter-institution variability as the random effect (the univariate model). Baseline characteristics that had a significant association in the univariate model were then entered as fixed effects in a linear mixed model with the log-transformed length of hospital stay as the dependent variable and the inter-institution variability as the random effect (the multivariate model). The association of baseline characteristics with cost of hospital stay was similarly investigated. Given that the cost of hospital stay was expected to have a considerable correlation with the length of hospital stay, the logtransformed length of hospital stay was additionally entered as a fixed effect in the linear mixed model. The impact of each explanatory variable on the length and cost of hospital stay is presented as the fold difference of the outcome corresponding to a 1-unit change in the explanatory variable. The fold difference was calculated from the exponential transformation of the regression estimate in each linear mixed model because the dependent variable was the log-transformed value of the outcome. Multiple imputation was adopted for missing data on explanatory
Hospital Stay and Cost of Poor-Risk CLI variables. All statistical analysis was performed using R version 3.1.0 (R Development Core Team, Vienna, Austria). Results The subject baseline characteristics are listed in Table 1. Mean patient age was 76±10 years old, and 59% were male. As shown in Figure, the length and cost of hospital stay varied considerably among patients. The median length and cost of hospital stay were 23 days (IQR, 9 52 days) and 2.25m (IQR, 1.33m 3.58m), respectively. The length and cost of hospital stay were strongly correlated with each other (r=0.86, P<0.001). The associations of baseline characteristics and length of hospital stay are listed in Table 2. On multivariate modeling, serum albumin, tissue loss, infection, surgical reconstruction, and bilateral revascularization were significantly and independently associated with longer hospital stay. A 1-g/dL increase in serum albumin was associated with a 21% shorter hospital stay (P=0.012), whereas tissue loss, infection, surgical reconstruction, and bilateral revascularization were associated with a 39% (P=0.019), 75% (P<0.001), 121% (P<0.001), and 55% longer hospital stay (P<0.001), respectively. The associations of baseline characteristics with cost of hospital stay are given in Table 3. Doubling the length of the hospital stay was associated with a 44% increase in hospital cost. Regular dialysis, surgical reconstruction, and bilateral revascularization were independently associated with an approximately 20% increase in the cost of hospital stay (all P<0.001). The association was adjusted for the length of hospital stay, indicating that these 3 risk factors were associated with increased cost per length of hospital stay. Diabetes mellitus was not identified as an independent risk factor for prolonged hospital stay or increased hospital cost. Discussion The current study has analyzed the distribution of length and cost of hospital stay and their associated risk factors in poor-risk CLI patients undergoing revascularization in Japan. Length and cost of hospital stay varied considerably Figure. Correlation between length and cost of hospital stay. Pearson s correlation coefficient between the log-transformed length and cost of hospital stay was 0.86 (95% CI: 0.84 0.88, P<0.001). Table 2. Significant Indicators of Longer Hospital Stay in CLI Patients Crude fold difference (univariate model), (95% CI) (P-value) Adjusted fold difference (multivariate model), (95% CI) (P-value) Age (per 10-year increase) 0.95 (0.86 1.05) (0.353) NI Male sex 1.07 (0.88 1.31) (0.488) NI Receiving welfare 0.75 (0.53 1.06) (0.101) NI Diabetes mellitus 1.15 (0.93 1.41) (0.190) NI Regular dialysis 0.93 (0.75 1.14) (0.462) NI Ambulation (per 1-level increase) 0.99 (0.92 1.06) (0.821) NI Katz index (per 1-point increase) 1.02 (0.98 1.07) (0.311) NI Home-boundness (per 1-level increase) 0.98 (0.89 1.07) (0.638) NI LSA (per 10-point increase) 0.99 (0.96 1.02) (0.365) NI GDS (per 1-point increase) 0.99 (0.94 1.04) (0.637) NI BMI (per 5-kg/m 2 increase) 0.90 (0.79 1.03) (0.140) NI Hemoglobin (per 2-g/dL increase) 0.91 (0.87 0.96) (0.001) 0.98 (0.93 1.04) (0.585) Total lymphocyte count (per 10 3 /μl increase) 1.08 (0.90 1.30) (0.388) NI Albumin (per 1-g/dL increase) 0.65 (0.55 0.76) (<0.001) 0.79 (0.65 0.95) (0.012) Cholinesterase (per 100-U/L increase) 0.80 (0.69 0.93) (0.003) 0.91 (0.77 1.07) (0.254) LVEF (per 10% increase) 0.91 (0.83 0.99) (0.034) 0.96 (0.88 1.04) (0.320) Log(BNP) (per 1-unit increase) 1.06 (0.99 1.12) (0.095) NI Tissue loss 1.90 (1.44 2.51) (<0.001) 1.39 (1.06 1.83) (0.019) Infection 1.99 (1.61 2.47) (<0.001) 1.75 (1.40 2.17) (<0.001) Surgical reconstruction 2.10 (1.34 3.30) (0.001) 2.21 (1.45 3.36) (<0.001) Bilateral revascularization 1.48 (1.18 1.84) (0.001) 1.55 (1.26 1.91) (<0.001) NI, not included in the multivariate model. Other abbreviations as in Table 1.
TAKAHARA M et al. Table 3. Significant Indicators of Higher Cost of Hospital Stay in CLI Patients Crude fold difference (univariate model), (95% CI) (P-value) Adjusted fold difference (multivariate model), (95% CI) (P-value) Age (per 10-year increase) 0.95 (0.89 1.01) (0.085) NI Male sex 1.07 (0.94 1.22) (0.304) NI Receiving welfare 0.89 (0.71 1.11) (0.313) NI Diabetes mellitus 1.18 (1.04 1.35) (0.013) 1.05 (0.98 1.13) (0.132) Regular dialysis 1.15 (1.01 1.31) (0.038) 1.17 (1.09 1.26) (<0.001) Ambulation (per 1-level increase) 0.99 (0.95 1.04) (0.741) NI Katz index (per 1-point increase) 1.02 (0.99 1.05) (0.137) NI Home-boundness (per 1-level increase) 0.97 (0.92 1.03) (0.350) NI LSA (per 10-point increase) 1.00 (0.98 1.02) (0.927) NI GDS (per 1-point increase) 0.99 (0.95 1.02) (0.368) NI BMI (per 5-kg/m 2 increase) 0.95 (0.87 1.04) (0.246) NI Hemoglobin (per 2-g/dL increase) 0.94 (0.91 0.97) (<0.001) 1.00 (0.98 1.02) (0.902) Total lymphocyte count (per 10 3 /μl increase) 1.03 (0.92 1.16) (0.612) NI Albumin (per 1-g/dL increase) 0.74 (0.67 0.82) (<0.001) 0.95 (0.89 1.01) (0.083) Cholinesterase (per 100-U/L increase) 0.85 (0.78 0.93) (0.001) 0.99 (0.94 1.04) (0.681) LVEF (per 10% increase) 0.92 (0.87 0.97) (0.002) 0.98 (0.95 1.00) (0.081) Log(BNP) (per 1-unit increase) 1.05 (1.01 1.09) (0.017) 0.99 (0.97 1.01) (0.436) Tissue loss 1.46 (1.22 1.74) (<0.001) 0.97 (0.88 1.07) (0.523) Infection 1.54 (1.34 1.77) (<0.001) 1.06 (0.98 1.14) (0.162) Surgical reconstruction 1.80 (1.36 2.40) (<0.001) 1.18 (1.03 1.36) (0.018) Bilateral revascularization 1.47 (1.28 1.69) (<0.001) 1.20 (1.11 1.28) (<0.001) Length of hospital stay (per 2-fold increase) 1.47 (1.43 1.50) (<0.001) 1.44 (1.41 1.47) (<0.001) Abbreviations as in Tables 1,2. in a longer hospital stay. The association of these features with prolonged hospital stay could be a target for intervention. Multidisciplinary approaches have recently been suggested as being effective for the management of CLI with tissue loss. 10,11 Introduction of multidisciplinary care in these patients could better control their wounds and shorten their wound healing time, leading to shorter hospital stay in the future. In addition, hospital stay could be shortened in advance by avoiding complications due to infection. Leaving tissue loss untreated for months is assumed to increase the risk of infection. Earlier referral and consultation after CLI onset could avoid the unnecessary complication of infection. Non-ambulatory status, diabetes mellitus and regular dialysis, all of which are known risk factors for the progression of subclinical CLI to CLI, 12 were all highly prevalent in the present patients. Promoting general practitioner awareness of the risk of CLI and subclinical CLI in these patients could lead to prompt referral and consultation after CLI onset and before infection, possibly shortening hospital stay. Low serum albumin, which is a marker of malnutrition, was also associated with prolonged hospital stay. Malnutrition is well known to delay wound healing. 13,14 Patients with malnutrition likely stay in hospital longer until their wounds are under control. The association of low serum albumin with prolonged hospital stay might simply reflect these aspects. In the present study, however, albumin level had a stronger association with prolonged hospital stay than any other nutrition-related marker. Although the true reason for this remains unknown, a possible explanation is that low albumin is affected not only by malnutrition but also by other unfavorable pathogeneses such as inflammatory status. 15 Consequently, low albubetween patients; the median values were 23 days (IQR, 9 52 days) and 2.25m (IQR, 1.33m 3.58m), respectively. Patients with low albumin, tissue loss, infection, surgical reconstruction, and bilateral revascularization were likely to have a longer hospital stay. Regular dialysis, surgical reconstruction, and bilateral revascularization were independently associated with an approximately 20% increase in the cost of hospital stay. Poor-risk CLI patients have an extremely high risk of 1-year mortality. 5 A prolonged hospital stay in this group means that patients would spend most of their remaining life in hospital, which would considerably interfere with their quality of life. In addition, a prolonged hospital stay is costly and imposes a major burden on the health-care system. Indeed, in the present study, doubling the length of hospital stay was associated with a 44% increase in hospital cost. It is clinically important to identify patients who are at high risk of prolonged hospital stay. In the present study, surgical reconstruction and bilateral revascularization were identified as independent risk factors for prolonged hospital stay. Compared with endovascular therapy, surgical reconstruction is more invasive. Performing surgical reconstruction in poor-risk patients might increase the risk of perioperative complications, requiring longer in-hospital management. The association of bilateral revascularization with longer hospital stay would be explained by its requirement of more time to complete than unilateral revascularization. Other risk factors for prolonged hospital stay included tissue loss (presence of unhealed wounds) and infection, both of which indicate a more severe stage of CLI. 9 The presence of unhealed wounds and infection would require additional treatment for their control in hospital, resulting
Hospital Stay and Cost of Poor-Risk CLI min could reflect general patient vulnerability and disease severity, partly including malnutrition; this trend may explain the superior ability of albumin level, compared with other nutrition-related markers, to predict the difficulty of wound control. 16 The association of malnutrition with prolonged hospital stay could provide another target for future interventions; effective medical nutritional management could reduce the length of hospital stay. Further studies will be needed to confirm the efficacy of nutritional management in reducing the length of hospital stay in CLI patients. The present study also identified risk factors associated with increased hospital cost. As expected, prolonged hospital stay was strongly associated with increased hospital cost. This suggests that reducing the length of hospital stay could yield large cost savings. Given that low albumin, tissue loss, and infection had a significant association with the length of hospital stay, effective medical support for nutrition and intensive foot care against infected tissue loss would possibly reduce the length of hospital stay, in turn decreasing hospital cost. The current study was only an observational study and therefore future studies are needed to confirm these beneficial effects. Additionally, the current study identified regular dialysis, surgical reconstruction, and bilateral revascularization as additional risk factors for increased hospital cost. The association of regular dialysis could be reasonably explained by the additional cost of the dialysis treatment itself, but also could be due to the high risk of perioperative adverse events. 17 Surgical reconstruction has a higher cost than endovascular therapy, 3 whereas the additional cost for bilateral revascularization would be explained by the increased number of revascularization procedures. The association of these 2 interventional features with both prolonged hospital stay and increased hospital cost would be clinically worth discussing. Surgical reconstruction can improve blood flow more successfully, compared with endovascular treatment, and would be preferred for patients with severe limb status. 18 Bilateral revascularization would be required in patients with bilateral CLI. In this sense, there exists a subgroup in whom unilateral endovascular treatment could not be a substitute for surgical reconstruction or bilateral revascularization. One should select surgical reconstruction and bilateral revascularization for these patients but at the same time should bear in mind their risk of prolonged hospital stay and increased hospital cost. Effort would be required to shorten hospital stay and lower hospital cost in these patients. Given that surgical reconstruction and bilateral revascularization were associated with prolonged hospital stay and increased hospital cost, health-care strategies for shortening hospital stay and lowering hospital cost in patients undergoing these interventions would be more effective for the overall saving of medical resources. The current risk analysis adopted a linear mixed model in which inter-institution variability was treated as a random effect, because inter-institution difference in CLI management strategy, which would considerably influence hospital stay and cost, was expected. The impact of the inter-institution difference on hospital stay and cost could be estimated by calculating the marginal and conditional R 2 proposed by Nakagawa and Schielzeth. 19 Marginal R 2 indicates the proportion of variance explained by the fixed effects in a linear mixed model, whereas the conditional R 2 indicates the proportion of variance explained by both the fixed and random effects in the model. The difference between the marginal and conditional R 2 therefore reflects the amount of variability in the random effects. As the proposers admit, this is not a perfect solution, but it is currently believed to be less susceptible to the common problems that plague alternative measures of R 2. 19 In the current study, the marginal and conditional R 2 of the multivariate model for hospital length (Table 2) were calculated to be 0.16 and 0.29, respectively, indicating that inter-institution variability had a substantial impact. In contrast, the corresponding values for the multivariate model for hospital cost (multivariate model in Table 3) were 0.77 and 0.79, respectively. Given that the hospital cost was strongly correlated with hospital stay (r=0.86), it is no surprise that these R 2 of the model were considerably higher. Although the difference between the marginal and conditional R 2 of the model appeared relatively small, it would be of note that there still remained room for additional explanation of the total cost variance by the interinstitution variability. As noted, the statistics are not perfect, and therefore it is difficult to accurately describe the impact. These data, however, suggest a considerable impact of the inter-institution difference on hospital stay and cost. The current study did not collect data on the strategies for CLI management in individual institutions, which would considerably influence hospital stay and cost. Future studies are needed to clarify the diversity in CLI management strategies between domestic institutions. Study Limitations The current study has some other limitations. First, the sample size was limited. In the current analytic strategy for model building, not a few variables (a maximum of 22 variables; Tables 2,3) could be potentially simultaneously entered as fixed effects in 1 linear mixed model. As is well known, a small number of samples relative to the explanatory variables increases the risk of overfitting of developed models. In multiple linear regression analysis, 1 rule of thumb is to have at least 10 observations per term, which is well known as the 1 in 10 rule. This indicates that the current sample size (n=507) would enable the construction of a regression model including a maximum of 50 explanatory variables. There is no such convenient rule, however, for the linear mixed model. We therefore provisionally checked the risk of overfitting, by 10,000-iteration simulation analysis using 10,000 datasets containing as many samples from as many institutions as the current database. Consequently, the linear mixed model including 22 variables as fixed effects produced a significantly lower mean squared error in the 5-fold cross-validation than the multiple linear regression model with 50 explanatory variables (which would provide the upper limit of tolerance; P=0.005), suggesting that the risk of overfitting in the current strategy for model building would be within tolerance. Second, the current PRIORITY registry did not collected detailed information on revascularization: all the data available referred only to whether the revascularization was bilateral and whether it was surgical reconstruction. Of the 135 patients undergoing bilateral revascularization, 7 patients (5%) underwent surgical reconstruction, and the remaining 128 (95%) underwent endovascular treatment alone. Of the 372 patients undergoing unilateral revascularization, 23 patients (6%) underwent surgical reconstruction, and the remaining 349 (94%) underwent endovascular treatment alone. The proportion of patients surgical reconstruction was not significantly different with regard to bilateral and unilateral revascularization (P=0.84). We did not have
TAKAHARA M et al. more detailed information, however, regarding revascularization. Revascularization includes various procedures, and the variety could influence the hospital stay and cost. Unfortunately we were unable to assess this potential impact in the current study. Third, data were collected on the length and cost of hospital stay, but data after discharge were not obtained. We also did not collect data on indirect costs. Further studies are needed to evaluate these aspects. Conclusions The current study described the distribution of length and cost of hospital stay and their associated risk factors in poor-risk CLI patients undergoing revascularization in Japan. Length and cost of hospital stay varied considerably between patients; the median values were 23 days (IQR, 9 52 days) and 2.25m (IQR, 1.33m 3.58m), respectively. Patients with low albumin, tissue loss, infection, surgical reconstruction, and bilateral revascularization were likely to require a longer hospital stay. Regular dialysis, surgical reconstruction, and bilateral revascularization were independently associated with an approximately 20% increase in the cost of hospital stay. Funding The PRIORITY study is sponsored by the following companies (in alphabetical order): Boston Scientific Japan K.K., HOKUSHINMEDICAL Co., Ltd, Japan, Johnson & Johnson K.K., Orbusneich Foundation, Medicon Inc., Medtronic Japan Co., Ltd, MSD K.K., Otsuka Pharmaceutical Co., Ltd, Terumo Corp., and W.L. Gore & Associates, Co., Ltd. The funding companies played no role in the design of the study, selection of the enrolled patients, treatment strategy, revascularization procedures or equipment, or the collection, analysis, or interpretation of the data. All companies provided research funds, which did not exceed an annual total of 1,000,000 yen. Disclosures M.T., Y.S., N.A., and S.N. declare no conflicts of interest. O.I. has received remuneration for attending meetings (presentations) that exceeds an annual total of 500,000 yen, per company or organization from Cardinal Health Japan., Otsuka Pharmaceutical Co., Ltd., Sanofi KK, Boston Scientific Japan K.K., Terumo Corp., Cook Japan. References 1. Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG, et al. Inter-Society consensus for the management of peripheral arterial disease (TASC II). Eur J Vasc Endovasc Surg 2007; 33(Suppl 1): S1 S75. 2. Gerhard-Herman MD, Gornik HL, Barrett C, Barshes NR, Corriere MA, Drachman DE, et al. 2016 AHA/ACC guideline on the management of patients with lower extremity peripheral artery disease: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2017; 135: e726 e779. 3. Forbes JF, Adam DJ, Bell J, Fowkes FG, Gillespie I, Raab GM, et al. Bypass versus Angioplasty in Severe Ischaemia of the Leg (BASIL) trial: Health-related quality of life outcomes, resource utilization, and cost-effectiveness analysis. J Vasc Surg 2010; 51: 43S 51S. 4. Hustedt JW, Goltzer O, Bohl DD, Fraser JF, Lara NJ, Spangehl MJ. Calculating the cost and risk of comorbidities in total joint arthroplasty in the United States. J Arthroplasty 2017; 32: 355 361.e1. 5. Iida O, Takahara M, Soga Y, Azuma N, Nanto S, Uematsu M. Prognostic impact of revascularization in poor-risk patients with critical limb ischemia: The PRIORITY Registry (Poor-Risk Patients With and Without Revascularization Therapy for Critical Limb Ischemia). JACC Cardiovasc Interv 2017; 10: 1147 1157. 6. Wallace M, Shelkey M. Monitoring functional status in hospitalized older adults. Am J Nurs 2008; 108: 64 71; quiz-2. 7. Peel C, Sawyer Baker P, Roth DL, Brown CJ, Brodner EV, Allman RM. Assessing mobility in older adults: The UAB study of aging life-space assessment. Phys Ther 2005; 85: 1008 1119. 8. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry 1982; 139: 1136 1139. 9. Mills JL Sr, Conte MS, Armstrong DG, Pomposelli FB, Schanzer A, Sidawy AN, et al. The Society for Vascular Surgery lower extremity threatened limb classification system: Risk stratification based on wound, ischemia, and foot infection (WIfI). J Vasc Surg 2014; 59: 220 234.e1 2. 10. Chung J, Modrall JG, Ahn C, Lavery LA, Valentine RJ. Multidisciplinary care improves amputation-free survival in patients with chronic critical limb ischemia. J Vasc Surg 2015; 61: 162 169. 11. Mii S, Tanaka K, Kyuragi R, Ishimura H, Yasukawa S, Guntani A, et al. Aggressive wound care by a multidisciplinary team improves wound healing after infrainguinal bypass in patients with critical limb ischemia. Ann Vasc Surg 2017; 41: 196 204. 12. Takahara M, Iida O, Soga Y, Kodama A, Azuma N. Absence of preceding intermittent claudication and its associated clinical features in patients with critical limb ischemia. J Atheroscler Thromb 2015; 22: 718 725. 13. Shiraki T, Iida O, Takahara M, Soga Y, Yamauchi Y, Hirano K, et al. Predictors of delayed wound healing after endovascular therapy of isolated infrapopliteal lesions underlying critical limb ischemia in patients with high prevalence of diabetes mellitus and hemodialysis. Eur J Vasc Endovasc Surg 2015; 49: 565 573. 14. Azuma N, Uchida H, Kokubo T, Koya A, Akasaka N, Sasajima T. Factors influencing wound healing of critical ischaemic foot after bypass surgery: Is the angiosome important in selecting bypass target artery? Eur J Vasc Endovasc Surg 2012; 43: 322 328. 15. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999; 340: 448 454. 16. Montalcini T, Moraca M, Ferro Y, Romeo S, Serra S, Raso MG, et al. Nutritional parameters predicting pressure ulcers and short-term mortality in patients with minimal conscious state as a result of traumatic and non-traumatic acquired brain injury. J Transl Med 2015; 13: 305. 17. Nakano M, Hirano K, Iida O, Yamauchi Y, Soga Y, Kawasaki D, et al. Clinical efficacy of infrapopliteal endovascular procedures for hemodialysis patients with critical limb ischemia. Ann Vasc Surg 2015; 29: 1225 1234. 18. Iida O, Takahara M, Soga Y, Kodama A, Terashi H, Azuma N. Three-year outcomes of surgical versus endovascular revascularization for critical limb ischemia: The SPINACH Study (Surgical Reconstruction Versus Peripheral Intervention in Patients With Critical Limb Ischemia). Circ Cardiovasc Interv 2017; 10: e005531. 19. Nakagawa S, Schielzeth H. A general and simple method for obtaining R 2 from generalized linear mixed-effects models. Methods Ecol Evol 2013; 4: 133 142.