Continuous Renal Replacement Therapy Icd 10
Abstract
Background
Nationwide data for the prevalence and outcomes of patients receiving continuous renal replacement therapy (CRRT) in intensive care units (ICUs) are scarce. This study assessed the status of CRRT in Japanese ICUs using a nationwide administrative claim database.
Methods
Data were extracted from the Japanese Diagnosis Procedure Combination database for 2011. From a cohort of critically ill patients aged 12 years or older who were admitted to ICUs for 3 days or longer, acute kidney injury (AKI) patients treated with CRRT were identified. The period prevalence of CRRT and in-hospital mortality were calculated. Logistic regression analysis identified factors associated with in-hospital mortality.
Results
Of 165 815 ICU patients, 6478 (3.9%) received CRRT for AKI. The most frequent admission diagnosis category was diseases of the circulatory system (n = 3074). The overall in-hospital mortality rate of the CRRT-treated AKI patients was 50.6%. Clustering patients into four groups by background revealed the lowest in-hospital mortality rate of 41.5% for the cardiovascular surgery group (n = 1043) compared with 53.5% for the nonsurgical cardiovascular group (n = 2031), 51.7% for the sepsis group (n = 1863) and 51.6% for other cases (n = 1541). Multiple logistic regression analysis showed a significant association of these four group classifications with in-hospital mortality in addition to age, hospital characteristics (type and volume), time from hospital admission to CRRT initiation and interventions performed on the day of CRRT initiation.
Conclusions
Using a large Japanese nationwide database, this study revealed remarkably high in-hospital mortality of CRRT-treated AKI patients, although the period prevalence of CRRT for AKI in ICUs was low.
INTRODUCTION
Acute kidney injury (AKI), a major complication in intensive care units (ICUs), is associated with unacceptably high short-term and long-term mortality [1–3]. Without definitive effective pharmacotherapy for AKI, renal replacement therapy (RRT) is the mainstay for managing patients with severe AKI. According to a multinational observational study, the vast majority of patients with AKI received RRT in ICUs [2]. Dialysis-requiring AKI and associated deaths have been reportedly increasing recently [4].
Several modalities of RRT exist, including continuous renal replacement therapy (CRRT), intermittent renal replacement therapy (IRRT) and peritoneal dialysis. Among them, CRRT has been the preferred technique for critically ill patients, mainly because of its better hemodynamic tolerance [5, 6]. In fact, the Beginning and Ending Supportive Therapy for the Kidney (BEST Kidney) study showed CRRT as the most commonly used initial modality, accounting for 80% of all cases treated with RRT [2]. Hospital mortality of AKI patients receiving CRRT is reportedly very high, mainly because of critical conditions at the initiation of CRRT [7–11].
Reports on the period prevalence and mortality of AKI patients receiving CRRT showed wide variations in characteristics, depending on the participating hospitals and recruited patients [8, 12]. Although a more comprehensive survey is warranted to reduce selection bias, few large population-based studies have been conducted to date [4, 13, 14]. Moreover, potential contributing factors to dialysis-requiring AKI in ICUs have not been fully evaluated in large cohort studies. While cardiovascular surgery and sepsis are the most common contributing factors to AKI in ICUs [2], pathophysiological insults to the kidneys induced by these factors might engender different prognoses [15–17]. Cardiac involvement without surgical intervention must also be considered because cardiorenal syndrome has recently been recognized as another crucial contributing factor to AKI in ICUs [18].
The Diagnostic Procedure Combination (DPC) database is a nationwide administrative claim database including approximately half of all inpatient admissions to acute care hospitals in Japan. This large and comprehensive database may contribute to better elucidation of the role of CRRT for AKI in clinical practice in Japan. This study was designed to assess the prevalence, characteristics and hospital outcomes of AKI patients requiring CRRT in Japanese ICUs and to clarify the factors contributing to mortality in this population.
MATERIALS AND METHODS
Data source
The DPC is a Japanese case-mix classification system that was launched in 2002 by the Ministry of Health, Labor and Welfare of Japan (MHLW), and is linked with a lump-sum payment system. All 82 academic hospitals are obliged to adopt the DPC system, whereas community hospitals participate in the system voluntarily. An annual survey of these DPC hospitals has been conducted by the DPC Research Group in collaboration with the MHLW. A total of 1015 hospitals joined the DPC database in 2011. Both administrative claims data and complete discharge abstract data were collected for all inpatients discharged from participating hospitals. The number of patients surveyed was ∼500 000 per month, representing around half of all inpatient admissions to acute-care hospitals in Japan.
The database includes the following data: patient age and sex, the main diagnosis at admission, comorbidities at admission, post-admission complications coded by the 10th revision of the International Classification of Diseases (ICD-10) [19] and text data in Japanese. It also includes the procedures used, including types of surgery coded with the Japanese original procedure codes, with daily records of drugs, devices used and modality of RRT (CRRT, IRRT, maintenance hemodialysis and maintenance peritoneal dialysis). The database also includes data related to the type of hospital (academic or nonacademic), length of hospital stay and discharge status. Hospital volume was defined as the total number of patients, other than chronic dialysis patients, receiving CRRT for AKI in 2011. All clinical data for each patient were recorded by attending physicians using an electronically prepared data collection tool at the time of discharge. All patient identifiers were removed from this database for this study.
Study approval was obtained from the institutional review board of the University of Tokyo. Informed consent was waived because the data are anonymous.
Identification of CRRT-treated AKI
Our preliminary data screening suggested a significant underestimation of the true prevalence of CRRT-treated AKI when using only the ICD-10 codes of AKI (N17.0, N17.1, N17.2 and N17.9); a large number of patients on CRRT were given various diagnosis codes such as 'multiple organ failure' (ICD-10 code of R68.8) and 'unspecified kidney failure' (ICD-10 code of N19) along with Japanese texts pointing to a state of AKI superimposed on chronic kidney disease. Under the Japanese health insurance system, CRRT is approved for the following indications: (i) kidney failure including both AKI and end-stage renal disease (ESRD), (ii) acute liver failure and (iii) severe pancreatitis. For the latter two indications, CRRT can be used even without renal dysfunction for the purpose of eliminating inflammatory mediators from the blood [20–23]. Therefore, instead of using the ICD-10 codes of AKI, we chose the following strategy of identifying CRRT for AKI: (i) all patients receiving CRRT were identified; (ii) those with ESRD, acute liver failure and severe pancreatitis were excluded as described below and (iii) the remaining patients were judged to have received CRRT for AKI and were thus regarded as eligible for the study.
Patient selection
From the DPC database, we extracted patients aged 12 years or older who were admitted to ICUs (including emergency rooms, coronary care units and stroke care units) for 3 days or longer in 2011. This age limit of over 12 years is the same as that used in the BEST Kidney study [2]. Among the extracted patients, we identified those receiving CRRT irrespective of its duration, followed by exclusion of patients with the following conditions: (i) ESRD, as detected by the ICD-10 code of N18.0 at admission and/or the use of maintenance hemodialysis or peritoneal dialysis; (ii) acute liver failure, as detected by the ICD-10 code of K72.0 at any time during the hospitalization and (iii) severe pancreatitis according to the MHLW's severity scoring system for acute pancreatitis, which is recorded for all patients with the diagnosis of pancreatitis [24].
The eligible patients were then divided into four groups according to potential contributing factors: (i) a cardiovascular surgery group based on the admission diagnosis category of 'disease of the circulatory system' (ICD-10 code of I00–I99) with the Japanese procedure codes for cardiovascular surgery (vascular surgery here includes only that of the thoracic aorta); (ii) a nonsurgical cardiovascular group based on the admission diagnosis category of 'disease of the circulatory system' without the cardiovascular surgery defined above; (iii) a sepsis group based on the ICD-10 code of A40–41 as an admission diagnosis, comorbidity at admission or a post-admission complication and (iv) other cases. Patients with both the admission diagnosis of 'disease of the circulatory system' and a subsequent complication of sepsis were categorized into the cardiovascular group rather than the sepsis group.
Variables
First, we assessed the period prevalence of patients receiving CRRT for AKI among those admitted to ICUs according to the admission diagnosis category (ICD-10 code category from A to Z) and according to the potential contributing factor. We then compared the following characteristics among the four groups: patient demographics, days from hospital admission to CRRT initiation, days from ICU admission to CRRT initiation, hospital type, hospital volume as defined above and interventions performed on the day of CRRT initiation; vasoactive drugs including dopamine, dobutamine and noradrenaline or adrenaline; mechanical ventilation; transfusion of red cell concentrates, fresh frozen plasma and platelets; intra-aortic balloon pumping and percutaneous cardiopulmonary support (percutaneous veno-arterial extracorporeal membrane oxygenation).
The primary end point was overall in-hospital mortality. The duration of CRRT, the proportion of patients subsequently switched to IRRT and the length of hospital stay were also investigated according to discharge status (survivors versus non-survivors).
Statistical analysis
Descriptive statistics were presented as medians and interquartile range for age, days from hospital/ICU admission to CRRT initiation, duration of CRRT, length of ICU and hospital stay, and as percentages for other variables. One-way analysis of variance or χ 2 tests were used as appropriate to examine differences among the four groups. Uni- and multivariate logistic regression analyses were performed to identify independent predictors of in-hospital mortality. In addition to age and sex, we selected variables of statistically significant differences between survivors and non-survivors in a non-stepwise manner. A P-value of <0.05 was inferred as statistically significant. For comparison of in-hospital mortality among the four groups, when the χ 2 tests showed statistical significance, a Bonferroni test was subsequently conducted by setting the significance level at P < 0.008. All statistical analyses were conducted using IBM SPSS ver. 20 (IBM SPSS, Armonk, NY, USA).
RESULTS
In a cohort of 165 815 patients (age 73 [62–81], male 59.2%) who were admitted to the ICU for at least 3 days, 9905 (6.0%) patients received CRRT. Of those, we excluded (i) 2927 patients with ESRD at admission, (ii) 398 with acute liver failure and (iii) 102 with severe pancreatitis. The remaining 6478 were identified as having received CRRT because of AKI, accounting for 3.9% of the entire cohort (Figure1).
FIGURE 1:
Outline of patient selection and allocation. ICU, intensive care unit; CRRT, continuous renal replacement therapy.
Table1 presents the proportions of CRRT-treated AKI patients categorized by the admission diagnosis category. The prevalence of those receiving CRRT varied widely depending on the admission diagnosis. Although the absolute number of those receiving CRRT with the main admission diagnosis of 'disease of the circulatory system' (ICD-10 code of I00–I99) was the highest with 3074 patients, its proportion among those admitted to ICUs with the same diagnosis category was only 3.4%. In contrast, patients with the admission diagnosis category of 'certain infectious and parasitic diseases' (ICD-10 code of A00–B99) were the most likely to receive CRRT when admitted to ICUs, accounting for 14.3% of those admitted to ICUs with the same diagnosis category.
Table 1.
Prevalence of acute kidney injury patients receiving continuous renal replacement therapy in intensive care units
| All the patients admitted to ICUs (n) | CRRT-treated AKI, n (%) | |
|---|---|---|
| Main diagnosis (ICD-10 code category) at ICU admission | ||
| Diseases of the circulatory system (I00–I99) | 89 366 | 3074 (3.4) |
| Diseases of the digestive system (K00–K93) | 12 054 | 758 (6.3) |
| Neoplasms (C00–D48) | 14 455 | 705 (4.9) |
| Certain infectious and parasitic diseases (A00–B99) | 3686 | 526 (14.3) |
| Diseases of the respiratory system (J00–J99) | 12 970 | 369 (2.8) |
| Diseases of the genitourinary system (N00–N99) | 2887 | 322 (11.2) |
| Injury, poisoning and certain other consequences of external causes (S00–T98) | 15 131 | 263 (1.7) |
| Others | 15 266 | 461 (3.0) |
| Four dominant categories | ||
| Surgical cardiovascular groupa | 14 407 | 1043 (7.2) |
| Nonsurgical cardiovascular groupb | 74 959 | 2031 (2.7) |
| Sepsis groupc | 10 944 | 1863 (17.0) |
| Other cases | 65 505 | 1541 (2.4) |
| Total | 165 815 | 6478 (3.9) |
| All the patients admitted to ICUs (n) | CRRT-treated AKI, n (%) | |
|---|---|---|
| Main diagnosis (ICD-10 code category) at ICU admission | ||
| Diseases of the circulatory system (I00–I99) | 89 366 | 3074 (3.4) |
| Diseases of the digestive system (K00–K93) | 12 054 | 758 (6.3) |
| Neoplasms (C00–D48) | 14 455 | 705 (4.9) |
| Certain infectious and parasitic diseases (A00–B99) | 3686 | 526 (14.3) |
| Diseases of the respiratory system (J00–J99) | 12 970 | 369 (2.8) |
| Diseases of the genitourinary system (N00–N99) | 2887 | 322 (11.2) |
| Injury, poisoning and certain other consequences of external causes (S00–T98) | 15 131 | 263 (1.7) |
| Others | 15 266 | 461 (3.0) |
| Four dominant categories | ||
| Surgical cardiovascular groupa | 14 407 | 1043 (7.2) |
| Nonsurgical cardiovascular groupb | 74 959 | 2031 (2.7) |
| Sepsis groupc | 10 944 | 1863 (17.0) |
| Other cases | 65 505 | 1541 (2.4) |
| Total | 165 815 | 6478 (3.9) |
ICUs, intensive care units; CRRT, continuous renal replacement therapy; AKI, acute kidney injury; ICD-10, 10th revision of the International Classification of Diseases.
aBased on the surgery code of cardiovascular surgery (vascular surgery includes only that for thoracic aorta).
bBy the admission diagnosis category of 'disease of the circulatory system' (ICD-10 codes of I00–99) without cardiovascular surgery.
cBased on the ICD-10 codes of A40–41 as an admission diagnosis, comorbidity at admission or a post-admission complication.
Table 1.
Prevalence of acute kidney injury patients receiving continuous renal replacement therapy in intensive care units
| All the patients admitted to ICUs (n) | CRRT-treated AKI, n (%) | |
|---|---|---|
| Main diagnosis (ICD-10 code category) at ICU admission | ||
| Diseases of the circulatory system (I00–I99) | 89 366 | 3074 (3.4) |
| Diseases of the digestive system (K00–K93) | 12 054 | 758 (6.3) |
| Neoplasms (C00–D48) | 14 455 | 705 (4.9) |
| Certain infectious and parasitic diseases (A00–B99) | 3686 | 526 (14.3) |
| Diseases of the respiratory system (J00–J99) | 12 970 | 369 (2.8) |
| Diseases of the genitourinary system (N00–N99) | 2887 | 322 (11.2) |
| Injury, poisoning and certain other consequences of external causes (S00–T98) | 15 131 | 263 (1.7) |
| Others | 15 266 | 461 (3.0) |
| Four dominant categories | ||
| Surgical cardiovascular groupa | 14 407 | 1043 (7.2) |
| Nonsurgical cardiovascular groupb | 74 959 | 2031 (2.7) |
| Sepsis groupc | 10 944 | 1863 (17.0) |
| Other cases | 65 505 | 1541 (2.4) |
| Total | 165 815 | 6478 (3.9) |
| All the patients admitted to ICUs (n) | CRRT-treated AKI, n (%) | |
|---|---|---|
| Main diagnosis (ICD-10 code category) at ICU admission | ||
| Diseases of the circulatory system (I00–I99) | 89 366 | 3074 (3.4) |
| Diseases of the digestive system (K00–K93) | 12 054 | 758 (6.3) |
| Neoplasms (C00–D48) | 14 455 | 705 (4.9) |
| Certain infectious and parasitic diseases (A00–B99) | 3686 | 526 (14.3) |
| Diseases of the respiratory system (J00–J99) | 12 970 | 369 (2.8) |
| Diseases of the genitourinary system (N00–N99) | 2887 | 322 (11.2) |
| Injury, poisoning and certain other consequences of external causes (S00–T98) | 15 131 | 263 (1.7) |
| Others | 15 266 | 461 (3.0) |
| Four dominant categories | ||
| Surgical cardiovascular groupa | 14 407 | 1043 (7.2) |
| Nonsurgical cardiovascular groupb | 74 959 | 2031 (2.7) |
| Sepsis groupc | 10 944 | 1863 (17.0) |
| Other cases | 65 505 | 1541 (2.4) |
| Total | 165 815 | 6478 (3.9) |
ICUs, intensive care units; CRRT, continuous renal replacement therapy; AKI, acute kidney injury; ICD-10, 10th revision of the International Classification of Diseases.
aBased on the surgery code of cardiovascular surgery (vascular surgery includes only that for thoracic aorta).
bBy the admission diagnosis category of 'disease of the circulatory system' (ICD-10 codes of I00–99) without cardiovascular surgery.
cBased on the ICD-10 codes of A40–41 as an admission diagnosis, comorbidity at admission or a post-admission complication.
Next, we separated the study population into four groups: (i) 1043 patients in the cardiovascular surgery group; (ii) 2031 patients in the nonsurgical cardiovascular group; (iii) 1863 patients in the sepsis group and (iv) 1541 patients as other cases. The surgical and nonsurgical cardiovascular groups included 189 and 389 patients, respectively, with subsequent diagnosis of sepsis. Table1 summarizes the prevalence of CRRT in each group. The proportion of those receiving CRRT differed largely among the groups, with the highest being 17.0% in the sepsis group.
Table2 presents the clinical characteristics of all eligible patients categorized into the four groups described above. While the median age was similar, the male proportion of the CRRT-treated AKI patients was higher than that of all the patients admitted to the ICU. From comparison of the clinical characteristics among the four groups, significant differences were found particularly in the following. The number of academic hospitals and the hospital volume were higher, and more intensive treatment regimens (including mechanical ventilation, vasoactive drug administration and blood transfusion) were used on the day of CRRT initiation in the cardiovascular surgery group than in any of the other three groups. Intra-aortic balloon pumping and percutaneous cardiopulmonary support were frequently used in the nonsurgical cardiovascular group, accounting for 23.7 and 8.3% of cases, respectively.
Table 2.
Baseline characteristics of patients receiving continuous renal replacement therapy by the group
| All patients (n = 6478) | Cardiovascular surgery group (n = 1043) | Nonsurgical cardiovascular group (n = 2031) | Sepsis group (n = 1863) | Other cases (n = 1541) | P-value | |
|---|---|---|---|---|---|---|
| Age (years)a | 72 [63–79] | 73 [65–78] | 74 [64–80] | 73 [63–79] | 70 [58–78] | <0.001 |
| Sex (male, %) | 66.0 | 66.3 | 69.6 | 63.5 | 63.9 | <0.001 |
| Hospital characteristics | ||||||
| Hospital type (academic hospital, %) | 33.7 | 40.0 | 30.5 | 31.2 | 36.7 | <0.001 |
| Hospital volume (patients with CRRT for AKI/year)a | 27 [14–42] | 30 [19–51] | 27 [15–42] | 22 [13–36] | 27 [14–42] | <0.001 |
| Time from hospital admission to CRRT initiation (days)a | 3 [1–14] | 7 [2–15] | 2 [1–10] | 2 [1–10] | 5 [1–22] | <0.001 |
| Time from ICU admission to CRRT initiation (days)a | 1 [0–3] | 2 [1–5] | 1 [0–4] | 1 [0–2] | 1 [0–3] | <0.001 |
| Interventions on the day of CRRT initiation (%) | ||||||
| Mechanical ventilation | 69.1 | 82.6 | 66.7 | 68.3 | 64.3 | <0.001 |
| Vasoactive drugs | ||||||
| Dopamine | 60.4 | 76.9 | 59.5 | 60.5 | 50.2 | <0.001 |
| Dobutamine | 32.0 | 60.4 | 40.2 | 19.1 | 17.4 | <0.001 |
| Noradrenaline or adrenaline | 53.6 | 71.9 | 49.8 | 60.0 | 38.6 | <0.001 |
| Transfusion | ||||||
| Red cell concentrates | 52.3 | 80.7 | 46.0 | 45.9 | 48.9 | <0.001 |
| Platelets | 23.3 | 49.8 | 15.1 | 19.5 | 20.9 | <0.001 |
| Fresh frozen plasma | 34.2 | 57.1 | 23.7 | 35.2 | 31.4 | <0.001 |
| Intra-aortic balloon pumping | 12.1 | 22.7 | 23.7 | 1.3 | 2.9 | <0.001 |
| Percutaneous cardiopulmonary support | 4.3 | 5.1 | 8.3 | 1.0 | 2.3 | <0.001 |
| All patients (n = 6478) | Cardiovascular surgery group (n = 1043) | Nonsurgical cardiovascular group (n = 2031) | Sepsis group (n = 1863) | Other cases (n = 1541) | P-value | |
|---|---|---|---|---|---|---|
| Age (years)a | 72 [63–79] | 73 [65–78] | 74 [64–80] | 73 [63–79] | 70 [58–78] | <0.001 |
| Sex (male, %) | 66.0 | 66.3 | 69.6 | 63.5 | 63.9 | <0.001 |
| Hospital characteristics | ||||||
| Hospital type (academic hospital, %) | 33.7 | 40.0 | 30.5 | 31.2 | 36.7 | <0.001 |
| Hospital volume (patients with CRRT for AKI/year)a | 27 [14–42] | 30 [19–51] | 27 [15–42] | 22 [13–36] | 27 [14–42] | <0.001 |
| Time from hospital admission to CRRT initiation (days)a | 3 [1–14] | 7 [2–15] | 2 [1–10] | 2 [1–10] | 5 [1–22] | <0.001 |
| Time from ICU admission to CRRT initiation (days)a | 1 [0–3] | 2 [1–5] | 1 [0–4] | 1 [0–2] | 1 [0–3] | <0.001 |
| Interventions on the day of CRRT initiation (%) | ||||||
| Mechanical ventilation | 69.1 | 82.6 | 66.7 | 68.3 | 64.3 | <0.001 |
| Vasoactive drugs | ||||||
| Dopamine | 60.4 | 76.9 | 59.5 | 60.5 | 50.2 | <0.001 |
| Dobutamine | 32.0 | 60.4 | 40.2 | 19.1 | 17.4 | <0.001 |
| Noradrenaline or adrenaline | 53.6 | 71.9 | 49.8 | 60.0 | 38.6 | <0.001 |
| Transfusion | ||||||
| Red cell concentrates | 52.3 | 80.7 | 46.0 | 45.9 | 48.9 | <0.001 |
| Platelets | 23.3 | 49.8 | 15.1 | 19.5 | 20.9 | <0.001 |
| Fresh frozen plasma | 34.2 | 57.1 | 23.7 | 35.2 | 31.4 | <0.001 |
| Intra-aortic balloon pumping | 12.1 | 22.7 | 23.7 | 1.3 | 2.9 | <0.001 |
| Percutaneous cardiopulmonary support | 4.3 | 5.1 | 8.3 | 1.0 | 2.3 | <0.001 |
ICU, intensive care unit; CRRT, continuous renal replacement therapy; AKI, acute kidney injury.
aContinuous variables are illustrated with median [interquartile range].
Table 2.
Baseline characteristics of patients receiving continuous renal replacement therapy by the group
| All patients (n = 6478) | Cardiovascular surgery group (n = 1043) | Nonsurgical cardiovascular group (n = 2031) | Sepsis group (n = 1863) | Other cases (n = 1541) | P-value | |
|---|---|---|---|---|---|---|
| Age (years)a | 72 [63–79] | 73 [65–78] | 74 [64–80] | 73 [63–79] | 70 [58–78] | <0.001 |
| Sex (male, %) | 66.0 | 66.3 | 69.6 | 63.5 | 63.9 | <0.001 |
| Hospital characteristics | ||||||
| Hospital type (academic hospital, %) | 33.7 | 40.0 | 30.5 | 31.2 | 36.7 | <0.001 |
| Hospital volume (patients with CRRT for AKI/year)a | 27 [14–42] | 30 [19–51] | 27 [15–42] | 22 [13–36] | 27 [14–42] | <0.001 |
| Time from hospital admission to CRRT initiation (days)a | 3 [1–14] | 7 [2–15] | 2 [1–10] | 2 [1–10] | 5 [1–22] | <0.001 |
| Time from ICU admission to CRRT initiation (days)a | 1 [0–3] | 2 [1–5] | 1 [0–4] | 1 [0–2] | 1 [0–3] | <0.001 |
| Interventions on the day of CRRT initiation (%) | ||||||
| Mechanical ventilation | 69.1 | 82.6 | 66.7 | 68.3 | 64.3 | <0.001 |
| Vasoactive drugs | ||||||
| Dopamine | 60.4 | 76.9 | 59.5 | 60.5 | 50.2 | <0.001 |
| Dobutamine | 32.0 | 60.4 | 40.2 | 19.1 | 17.4 | <0.001 |
| Noradrenaline or adrenaline | 53.6 | 71.9 | 49.8 | 60.0 | 38.6 | <0.001 |
| Transfusion | ||||||
| Red cell concentrates | 52.3 | 80.7 | 46.0 | 45.9 | 48.9 | <0.001 |
| Platelets | 23.3 | 49.8 | 15.1 | 19.5 | 20.9 | <0.001 |
| Fresh frozen plasma | 34.2 | 57.1 | 23.7 | 35.2 | 31.4 | <0.001 |
| Intra-aortic balloon pumping | 12.1 | 22.7 | 23.7 | 1.3 | 2.9 | <0.001 |
| Percutaneous cardiopulmonary support | 4.3 | 5.1 | 8.3 | 1.0 | 2.3 | <0.001 |
| All patients (n = 6478) | Cardiovascular surgery group (n = 1043) | Nonsurgical cardiovascular group (n = 2031) | Sepsis group (n = 1863) | Other cases (n = 1541) | P-value | |
|---|---|---|---|---|---|---|
| Age (years)a | 72 [63–79] | 73 [65–78] | 74 [64–80] | 73 [63–79] | 70 [58–78] | <0.001 |
| Sex (male, %) | 66.0 | 66.3 | 69.6 | 63.5 | 63.9 | <0.001 |
| Hospital characteristics | ||||||
| Hospital type (academic hospital, %) | 33.7 | 40.0 | 30.5 | 31.2 | 36.7 | <0.001 |
| Hospital volume (patients with CRRT for AKI/year)a | 27 [14–42] | 30 [19–51] | 27 [15–42] | 22 [13–36] | 27 [14–42] | <0.001 |
| Time from hospital admission to CRRT initiation (days)a | 3 [1–14] | 7 [2–15] | 2 [1–10] | 2 [1–10] | 5 [1–22] | <0.001 |
| Time from ICU admission to CRRT initiation (days)a | 1 [0–3] | 2 [1–5] | 1 [0–4] | 1 [0–2] | 1 [0–3] | <0.001 |
| Interventions on the day of CRRT initiation (%) | ||||||
| Mechanical ventilation | 69.1 | 82.6 | 66.7 | 68.3 | 64.3 | <0.001 |
| Vasoactive drugs | ||||||
| Dopamine | 60.4 | 76.9 | 59.5 | 60.5 | 50.2 | <0.001 |
| Dobutamine | 32.0 | 60.4 | 40.2 | 19.1 | 17.4 | <0.001 |
| Noradrenaline or adrenaline | 53.6 | 71.9 | 49.8 | 60.0 | 38.6 | <0.001 |
| Transfusion | ||||||
| Red cell concentrates | 52.3 | 80.7 | 46.0 | 45.9 | 48.9 | <0.001 |
| Platelets | 23.3 | 49.8 | 15.1 | 19.5 | 20.9 | <0.001 |
| Fresh frozen plasma | 34.2 | 57.1 | 23.7 | 35.2 | 31.4 | <0.001 |
| Intra-aortic balloon pumping | 12.1 | 22.7 | 23.7 | 1.3 | 2.9 | <0.001 |
| Percutaneous cardiopulmonary support | 4.3 | 5.1 | 8.3 | 1.0 | 2.3 | <0.001 |
ICU, intensive care unit; CRRT, continuous renal replacement therapy; AKI, acute kidney injury.
aContinuous variables are illustrated with median [interquartile range].
The first screened entire cohort of patients who were admitted to the ICU for 3 days and longer showed in-hospital mortality of 14.0% (23 138 of 165 815), although the study population of CRRT-treated AKI showed a much higher in-hospital mortality rate of 50.6% (3277 of 6478). When the CRRT-treated AKI patients were categorized into four groups as described above, in-hospital mortality rate was 41.5% (433 of 1043) for the cardiovascular surgery group, 53.5% (1086 of 2031) for the nonsurgical cardiovascular group, 51.7% (963 of 1863) for the sepsis group and 51.6% (795 of 1541) for other cases. Bonferroni tests revealed that in-hospital mortality rate was significantly lower in the cardiovascular surgery group than in any of the other three groups, while no significant difference was found among these latter three groups. Table3 presents the outcomes of hospital survivors and non-survivors. The duration of CRRT was shorter in the survivors than in the non-survivors, with the median duration of 4 and 6 days, respectively. Differences among the four groups were small in both the survivors and non-survivors. Around 20 and 10% of the survivors and non-survivors, respectively, were switched to IRRT.
Table 3.
Outcomes of hospital survivors and non-survivors by the group
| All patients | Cardiovascular surgery group | Nonsurgical cardiovascular group | Sepsis group | Other cases | |
|---|---|---|---|---|---|
| Hospital survivors (number of patients) | 3201 | 610 | 945 | 900 | 746 |
| Duration of CRRT (days)a | 4 [2–7] | 4 [3–8] | 4 [2–7] | 4 [2–7] | 4 [2–6] |
| Proportion of patients switched to IRRT (%) | 22.6 | 28.9 | 22.0 | 16.6 | 25.7 |
| Length of hospital stay (days)a | 51 [30–87] | 53 [34–82] | 46 [29–82] | 55 [32–95] | 50 [26–91] |
| Hospital non-survivors (number of patients) | 3277 | 433 | 1086 | 963 | 795 |
| Duration of CRRT (days)a | 6 [3–14] | 8 [4–18] | 6 [3–14] | 6 [3–14] | 6 [3–13] |
| Proportion of patients switched to IRRT (%) | 11.2 | 9.5 | 9.9 | 13.2 | 11.6 |
| Length of hospital stay (days)a | 26 [11–53] | 29 [13–55] | 21 [7–43] | 25 [13–52] | 33 [14–67] |
| All patients | Cardiovascular surgery group | Nonsurgical cardiovascular group | Sepsis group | Other cases | |
|---|---|---|---|---|---|
| Hospital survivors (number of patients) | 3201 | 610 | 945 | 900 | 746 |
| Duration of CRRT (days)a | 4 [2–7] | 4 [3–8] | 4 [2–7] | 4 [2–7] | 4 [2–6] |
| Proportion of patients switched to IRRT (%) | 22.6 | 28.9 | 22.0 | 16.6 | 25.7 |
| Length of hospital stay (days)a | 51 [30–87] | 53 [34–82] | 46 [29–82] | 55 [32–95] | 50 [26–91] |
| Hospital non-survivors (number of patients) | 3277 | 433 | 1086 | 963 | 795 |
| Duration of CRRT (days)a | 6 [3–14] | 8 [4–18] | 6 [3–14] | 6 [3–14] | 6 [3–13] |
| Proportion of patients switched to IRRT (%) | 11.2 | 9.5 | 9.9 | 13.2 | 11.6 |
| Length of hospital stay (days)a | 26 [11–53] | 29 [13–55] | 21 [7–43] | 25 [13–52] | 33 [14–67] |
CRRT, continuous renal replacement therapy; IRRT, intermittent renal replacement therapy.
aContinuous variables are illustrated with median [interquartile range].
Table 3.
Outcomes of hospital survivors and non-survivors by the group
| All patients | Cardiovascular surgery group | Nonsurgical cardiovascular group | Sepsis group | Other cases | |
|---|---|---|---|---|---|
| Hospital survivors (number of patients) | 3201 | 610 | 945 | 900 | 746 |
| Duration of CRRT (days)a | 4 [2–7] | 4 [3–8] | 4 [2–7] | 4 [2–7] | 4 [2–6] |
| Proportion of patients switched to IRRT (%) | 22.6 | 28.9 | 22.0 | 16.6 | 25.7 |
| Length of hospital stay (days)a | 51 [30–87] | 53 [34–82] | 46 [29–82] | 55 [32–95] | 50 [26–91] |
| Hospital non-survivors (number of patients) | 3277 | 433 | 1086 | 963 | 795 |
| Duration of CRRT (days)a | 6 [3–14] | 8 [4–18] | 6 [3–14] | 6 [3–14] | 6 [3–13] |
| Proportion of patients switched to IRRT (%) | 11.2 | 9.5 | 9.9 | 13.2 | 11.6 |
| Length of hospital stay (days)a | 26 [11–53] | 29 [13–55] | 21 [7–43] | 25 [13–52] | 33 [14–67] |
| All patients | Cardiovascular surgery group | Nonsurgical cardiovascular group | Sepsis group | Other cases | |
|---|---|---|---|---|---|
| Hospital survivors (number of patients) | 3201 | 610 | 945 | 900 | 746 |
| Duration of CRRT (days)a | 4 [2–7] | 4 [3–8] | 4 [2–7] | 4 [2–7] | 4 [2–6] |
| Proportion of patients switched to IRRT (%) | 22.6 | 28.9 | 22.0 | 16.6 | 25.7 |
| Length of hospital stay (days)a | 51 [30–87] | 53 [34–82] | 46 [29–82] | 55 [32–95] | 50 [26–91] |
| Hospital non-survivors (number of patients) | 3277 | 433 | 1086 | 963 | 795 |
| Duration of CRRT (days)a | 6 [3–14] | 8 [4–18] | 6 [3–14] | 6 [3–14] | 6 [3–13] |
| Proportion of patients switched to IRRT (%) | 11.2 | 9.5 | 9.9 | 13.2 | 11.6 |
| Length of hospital stay (days)a | 26 [11–53] | 29 [13–55] | 21 [7–43] | 25 [13–52] | 33 [14–67] |
CRRT, continuous renal replacement therapy; IRRT, intermittent renal replacement therapy.
aContinuous variables are illustrated with median [interquartile range].
Table4 presents factors associated with in-hospital mortality, as deduced from results of multiple logistic regression analysis. All the factors selected into the model were significantly associated with in-hospital mortality after adjusting for confounding factors. Among them, assignment to the nonsurgical cardiovascular, the sepsis and the other groups (the cardiovascular surgery group serving as the reference group) as well as the use of percutaneous cardiopulmonary support led to high in-hospital mortality with an odds ratio higher than 2.0.
Table 4.
Logistic regression analysis examining factors associated with in-hospital mortality in patients receiving continuous renal replacement therapy
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age (every 1 year) | 1.017 | 1.014–1.021 | <0.001 | 1.022 | 1.018–1.026 | <0.001 |
| Sex | ||||||
| Male | Ref. | Ref. | ||||
| Female | 0.915 | 0.825–1.014 | 0.089 | 0.885 | 0.792–0.990 | 0.033 |
| Group by a potential contributing factor | ||||||
| Cardiovascular surgery group | Ref. | Ref. | ||||
| Nonsurgical cardiovascular group | 1.619 | 1.392–1.883 | <0.001 | 2.554 | 2.151–3.032 | <0.001 |
| Sepsis group | 1.507 | 1.293–1.757 | <0.001 | 2.612 | 2.186–3.120 | <0.001 |
| Other cases | 1.501 | 1.281–1.759 | <0.001 | 2.982 | 2.471–3.599 | <0.001 |
| Hospital type | ||||||
| Nonacademic hospitals | Ref. | Ref. | ||||
| Academic hospitals | 0.791 | 0.713–0.877 | <0.001 | 0.849 | 0.748–0.964 | 0.011 |
| Hospital volume (every 1 patient/year)a | 0.991 | 0.989–0.994 | <0.001 | 0.994 | 0.991–0.996 | <0.001 |
| Time from admission to CRRT initiation (every 1 day) | 1.016 | 1.014–1.019 | <0.001 | 1.015 | 1.013–1.018 | <0.001 |
| Interventions on the day of CRRT initiation (yes/no) | ||||||
| Mechanical ventilation | 1.942 | 1.744–2.162 | <0.001 | 1.609 | 1.429–1.811 | <0.001 |
| Dopamine | 2.056 | 1.858–2.275 | <0.001 | 1.695 | 1.514–1.897 | <0.001 |
| Dobutamine | 1.465 | 1.319–1.628 | <0.001 | 1.178 | 1.037–1.337 | 0.011 |
| Noradrenaline or adrenaline | 1.856 | 1.681–2.049 | <0.001 | 1.474 | 1.313–1.655 | <0.001 |
| Blood transfusion | 1.767 | 1.598–1.954 | <0.001 | 1.522 | 1.356–1.709 | <0.001 |
| Intra-aortic balloon pumping | 1.390 | 1.195–1.616 | <0.001 | 1.290 | 1.081–1.539 | 0.005 |
| Percutaneous cardiopulmonary support | 3.252 | 2.456–4.308 | <0.001 | 3.387 | 2.495–4.599 | <0.001 |
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age (every 1 year) | 1.017 | 1.014–1.021 | <0.001 | 1.022 | 1.018–1.026 | <0.001 |
| Sex | ||||||
| Male | Ref. | Ref. | ||||
| Female | 0.915 | 0.825–1.014 | 0.089 | 0.885 | 0.792–0.990 | 0.033 |
| Group by a potential contributing factor | ||||||
| Cardiovascular surgery group | Ref. | Ref. | ||||
| Nonsurgical cardiovascular group | 1.619 | 1.392–1.883 | <0.001 | 2.554 | 2.151–3.032 | <0.001 |
| Sepsis group | 1.507 | 1.293–1.757 | <0.001 | 2.612 | 2.186–3.120 | <0.001 |
| Other cases | 1.501 | 1.281–1.759 | <0.001 | 2.982 | 2.471–3.599 | <0.001 |
| Hospital type | ||||||
| Nonacademic hospitals | Ref. | Ref. | ||||
| Academic hospitals | 0.791 | 0.713–0.877 | <0.001 | 0.849 | 0.748–0.964 | 0.011 |
| Hospital volume (every 1 patient/year)a | 0.991 | 0.989–0.994 | <0.001 | 0.994 | 0.991–0.996 | <0.001 |
| Time from admission to CRRT initiation (every 1 day) | 1.016 | 1.014–1.019 | <0.001 | 1.015 | 1.013–1.018 | <0.001 |
| Interventions on the day of CRRT initiation (yes/no) | ||||||
| Mechanical ventilation | 1.942 | 1.744–2.162 | <0.001 | 1.609 | 1.429–1.811 | <0.001 |
| Dopamine | 2.056 | 1.858–2.275 | <0.001 | 1.695 | 1.514–1.897 | <0.001 |
| Dobutamine | 1.465 | 1.319–1.628 | <0.001 | 1.178 | 1.037–1.337 | 0.011 |
| Noradrenaline or adrenaline | 1.856 | 1.681–2.049 | <0.001 | 1.474 | 1.313–1.655 | <0.001 |
| Blood transfusion | 1.767 | 1.598–1.954 | <0.001 | 1.522 | 1.356–1.709 | <0.001 |
| Intra-aortic balloon pumping | 1.390 | 1.195–1.616 | <0.001 | 1.290 | 1.081–1.539 | 0.005 |
| Percutaneous cardiopulmonary support | 3.252 | 2.456–4.308 | <0.001 | 3.387 | 2.495–4.599 | <0.001 |
OR, odds ratio; CI, confidence interval; CRRT, continuous renal replacement therapy.
aHospital volume was defined as the total number of patients receiving CRRT for acute kidney injury in 2011.
Table 4.
Logistic regression analysis examining factors associated with in-hospital mortality in patients receiving continuous renal replacement therapy
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age (every 1 year) | 1.017 | 1.014–1.021 | <0.001 | 1.022 | 1.018–1.026 | <0.001 |
| Sex | ||||||
| Male | Ref. | Ref. | ||||
| Female | 0.915 | 0.825–1.014 | 0.089 | 0.885 | 0.792–0.990 | 0.033 |
| Group by a potential contributing factor | ||||||
| Cardiovascular surgery group | Ref. | Ref. | ||||
| Nonsurgical cardiovascular group | 1.619 | 1.392–1.883 | <0.001 | 2.554 | 2.151–3.032 | <0.001 |
| Sepsis group | 1.507 | 1.293–1.757 | <0.001 | 2.612 | 2.186–3.120 | <0.001 |
| Other cases | 1.501 | 1.281–1.759 | <0.001 | 2.982 | 2.471–3.599 | <0.001 |
| Hospital type | ||||||
| Nonacademic hospitals | Ref. | Ref. | ||||
| Academic hospitals | 0.791 | 0.713–0.877 | <0.001 | 0.849 | 0.748–0.964 | 0.011 |
| Hospital volume (every 1 patient/year)a | 0.991 | 0.989–0.994 | <0.001 | 0.994 | 0.991–0.996 | <0.001 |
| Time from admission to CRRT initiation (every 1 day) | 1.016 | 1.014–1.019 | <0.001 | 1.015 | 1.013–1.018 | <0.001 |
| Interventions on the day of CRRT initiation (yes/no) | ||||||
| Mechanical ventilation | 1.942 | 1.744–2.162 | <0.001 | 1.609 | 1.429–1.811 | <0.001 |
| Dopamine | 2.056 | 1.858–2.275 | <0.001 | 1.695 | 1.514–1.897 | <0.001 |
| Dobutamine | 1.465 | 1.319–1.628 | <0.001 | 1.178 | 1.037–1.337 | 0.011 |
| Noradrenaline or adrenaline | 1.856 | 1.681–2.049 | <0.001 | 1.474 | 1.313–1.655 | <0.001 |
| Blood transfusion | 1.767 | 1.598–1.954 | <0.001 | 1.522 | 1.356–1.709 | <0.001 |
| Intra-aortic balloon pumping | 1.390 | 1.195–1.616 | <0.001 | 1.290 | 1.081–1.539 | 0.005 |
| Percutaneous cardiopulmonary support | 3.252 | 2.456–4.308 | <0.001 | 3.387 | 2.495–4.599 | <0.001 |
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Age (every 1 year) | 1.017 | 1.014–1.021 | <0.001 | 1.022 | 1.018–1.026 | <0.001 |
| Sex | ||||||
| Male | Ref. | Ref. | ||||
| Female | 0.915 | 0.825–1.014 | 0.089 | 0.885 | 0.792–0.990 | 0.033 |
| Group by a potential contributing factor | ||||||
| Cardiovascular surgery group | Ref. | Ref. | ||||
| Nonsurgical cardiovascular group | 1.619 | 1.392–1.883 | <0.001 | 2.554 | 2.151–3.032 | <0.001 |
| Sepsis group | 1.507 | 1.293–1.757 | <0.001 | 2.612 | 2.186–3.120 | <0.001 |
| Other cases | 1.501 | 1.281–1.759 | <0.001 | 2.982 | 2.471–3.599 | <0.001 |
| Hospital type | ||||||
| Nonacademic hospitals | Ref. | Ref. | ||||
| Academic hospitals | 0.791 | 0.713–0.877 | <0.001 | 0.849 | 0.748–0.964 | 0.011 |
| Hospital volume (every 1 patient/year)a | 0.991 | 0.989–0.994 | <0.001 | 0.994 | 0.991–0.996 | <0.001 |
| Time from admission to CRRT initiation (every 1 day) | 1.016 | 1.014–1.019 | <0.001 | 1.015 | 1.013–1.018 | <0.001 |
| Interventions on the day of CRRT initiation (yes/no) | ||||||
| Mechanical ventilation | 1.942 | 1.744–2.162 | <0.001 | 1.609 | 1.429–1.811 | <0.001 |
| Dopamine | 2.056 | 1.858–2.275 | <0.001 | 1.695 | 1.514–1.897 | <0.001 |
| Dobutamine | 1.465 | 1.319–1.628 | <0.001 | 1.178 | 1.037–1.337 | 0.011 |
| Noradrenaline or adrenaline | 1.856 | 1.681–2.049 | <0.001 | 1.474 | 1.313–1.655 | <0.001 |
| Blood transfusion | 1.767 | 1.598–1.954 | <0.001 | 1.522 | 1.356–1.709 | <0.001 |
| Intra-aortic balloon pumping | 1.390 | 1.195–1.616 | <0.001 | 1.290 | 1.081–1.539 | 0.005 |
| Percutaneous cardiopulmonary support | 3.252 | 2.456–4.308 | <0.001 | 3.387 | 2.495–4.599 | <0.001 |
OR, odds ratio; CI, confidence interval; CRRT, continuous renal replacement therapy.
aHospital volume was defined as the total number of patients receiving CRRT for acute kidney injury in 2011.
DISCUSSION
Most epidemiological studies that have reported prevalence and outcomes of patients receiving CRRT for AKI were conducted either in a single center or in a few centers [25, 26]. Their results suggested that the patient characteristics, treatments and outcomes of AKI in epidemiological studies differed greatly among participating hospitals [12]. Therefore, a study involving numerous participating hospitals is warranted to more accurately estimate the prevalence of CRRT for AKI. Only a few such studies have been reported to date [4, 13, 14]. The present study is based on a nationwide administrative claim database, and clinical data were obtained from significantly greater numbers of hospitals compared with these previous reports.
Our study showed that the period prevalence of patients receiving CRRT for AKI in Japanese ICUs in 2011 was 3.9%, and that their in-hospital mortality rate was 50.6%. The mortality rate in this study was similar to that of multinational studies [9, 11], an Australian nationwide study [13] and a large randomized control trial that evaluated the effects of higher intensity of CRRT [27]. Our results also demonstrated that the prevalence of AKI patients receiving CRRT varied widely across their background: cardiovascular diseases, surgical versus nonsurgical and sepsis. Although cardiovascular involvement at admission with or without surgery was the most common reason for CRRT use, the probability of receiving CRRT was low among those admitted to ICUs with the same background diagnosis. In contrast, sepsis frequently led to the use of CRRT at the rate of nearly 1 in 5.
A previous multicenter study conducted using data from Australian ICUs showed that the in-hospital mortality rate of patients receiving RRT (CRRT comprising 89.6%) varied largely with background diagnosis [8]. Our study demonstrated that the mortality rate was significantly lower in the cardiovascular surgery group than in any of the other groups, even though these patients were treated using more intensive treatment regimens on the day of CRRT initiation. One possible reason for the more favorable prognosis in the cardiovascular surgery group is that most of these patients had scheduled surgeries. Their preoperative conditions were presumably better than those of the other groups. In addition, although cardiovascular surgery can damage the kidneys to a considerable degree, insult by the surgical procedure might be transient during surgery, whereas renal insults in the other groups were potentially still progressing at ICU admission. The mortality of the nonsurgical cardiovascular group was as high as that of the sepsis group, even after adjustment with multiple regression. This observation appears to differ from those reported from previous studies where septic AKI led to much higher mortality than non-septic AKI [17, 28]. The difference suggests that severe AKI caused by cardiovascular disease, i.e. cardiorenal syndrome, can show outcomes as poor as those shown by septic AKI.
Among the hospital survivors, the median duration of CRRT was 4 days and did not differ significantly between the groups. The non-survivors tended to receive CRRT for a longer period with the median duration of 6 days. This finding indicates that non-survivors might not have been able to withdraw from CRRT because of prolonged hemodynamic instability, a risk factor for increased mortality [29]. We also found that the rate of switching to IRRT was much lower in the non-survivors than in the survivors, suggesting that more patients in the non-survivors group might have died earlier before switching to IRRT was contemplated. Restoration of hemodynamic stability allowing early CRRT withdrawal with or without subsequent IRRT may be the key to survival up to the time of hospital discharge, although the lack of serum creatinine and urinary data precludes discussion about the appropriate timing of CRRT cessation [30].
A multiple logistic regression analysis demonstrated that in-hospital mortality was associated with a number of clinical factors as shown in Table4. Many of these factors have been suggested as predictive factors in previous studies [7, 8, 31–35]. In-hospital mortality was largely associated with the severity of illness at baseline as indicated by the use of mechanical ventilation, vasoactive supports and blood transfusion in our study. Of particular note is that later CRRT initiation was associated with higher mortality, which is in line with previous observational studies [36, 37], although comparison of early versus late initiation is hampered by a lack of laboratory results in our study [38]. In addition, our results suggest that hospital type and volume (defined as the number of patients on CRRT in a year) might influence mortality. A parameter of hospital volume has been associated with better patient outcomes in different healthcare settings, such as in surgical specialties [39] and medical ICU admissions [40]. To the best of our knowledge, there has been no study examining the influence of hospital type and volume on patient outcomes in the field of CRRT, while experts expressed the view that numerous factors are involved in CRRT management including organizational/logistical factors [41]. Although results of observational studies should be carefully interpreted due to possible residual confounding, our finding poses a hypothesis regarding the association between hospital characteristics and mortality of CRRT-treated patients.
This study was based on administrative claims data, mainly including data related to patient demographics, diagnosis, procedures and drugs used. Several limitations to this study must be acknowledged, despite the advantage of including a large study population. First, as described in the Materials and methods section, the diagnosis of AKI is not based on physiological measurements such as blood urea nitrogen, creatinine or urine volume. It is possible that the analyzed data include cases in which CRRT was used for patients who showed insufficient abnormality in serum creatinine and urine output for AKI diagnosis criteria, although 2012 KDIGO criteria defined the requirement of RRT as stage 3 AKI, irrespective of serum creatinine concentration [42]. Conversely, our strategy for identification of CRRT-treated AKI might have excluded patients with both acute liver failure and AKI as well as those with both severe pancreatitis and AKI. However, its influence on the true prevalence of the CRRT-treated AKI seems negligible because acute liver failure and severe pancreatitis were uncommon in our cohort. A lack of data on creatinine and urine volume also prevented us from making an accurate estimate of the period prevalence of AKI in this study cohort. A recent study reported that the use of billing codes for identifying AKI had low sensitivity compared with the 2012 KDIGO criteria [43]. Fourth, we separated the study population into groups according to the most common renal insult, as reported previously [2]. Only a small number of people had cardiovascular diagnosis at admission and subsequent septic diagnosis, where we were unable to confirm which condition was the critical reason for CRRT use. Other possible renal insult might arise from hypovolemia, drugs and urinary tract obstruction. Finally, details of CRRT including the use of dialysis and/or hemofiltration, dialysis intensity and exact duration per day were not available. In multiple logistic regression analyses for in-hospital mortality, our results might be biased by unmeasured confounders such as details of CRRT, laboratory results, vital signs and changes in the patients' general condition over time. Likewise, we were not able to retrieve the information needed to calculate severity scores for use in ICUs such as the Acute Physiology and Chronic Health Evaluation II score [44] and the Sequential Organ Failure Assessment score [45], which are shown to be significantly associated with outcomes [7, 46]. We therefore utilized the available information, including vasoactive drug use, mechanical ventilation and transfusion on the day of CRRT initiation, to adjust for the severity of illness.
In conclusion, results of this retrospective analysis using a large nationwide database demonstrate that the period prevalence of CRRT for AKI in ICUs was 3.9%, although it varied widely with admission diagnosis and the potential contributing factor to AKI. The overall in-hospital mortality rate was 50.6%. Background conditions were associated with in-hospital mortality in addition to age, hospital characteristics, time from hospital admission to CRRT initiation and interventions performed on the day of CRRT initiation. These findings are expected to be useful for future clinical trials on CRRT, requiring AKI for determination of the target population.
ACKNOWLEDGEMENTS
This study was funded by grants from the Ministry of Health, Labor and Welfare, Japan (Research on Policy Planning and Evaluation grant no. H22-Policy-031), Ministry of Education, Culture, Sports, Science and Technology, Japan (Scientific Research B, no. 22390131), Council for Science and Technology Policy, Japan (Funding Program for World-Leading Innovative R&D on Science and Technology, FIRST program—grant no. 0301002001001) and the Tokyo Society of Medical Sciences.
CONFLICT OF INTEREST STATEMENT
None declared. The results presented in this paper have not been published previously in whole or part, except in abstract format.
REFERENCES
1
et al.
Epidemiology and outcomes of acute renal failure in hospitalized patients: a national survey
.
Clin J Am Soc Nephrol
2006
;
1
:
43
–
51
2
et al.
Acute renal failure in critically ill patients: a multinational, multicenter study
.
JAMA
2005
;
294
:
813
–
818
3
et al.
Prognosis for long-term survival and renal recovery in critically ill patients with severe acute renal failure: a population-based study
.
Crit Care
2005
;
9
:
R700
–
R709
4
et al.
Temporal changes in incidence of dialysis-requiring AKI
.
J Am Soc Nephrol
2013
;
24
:
37
–
42
5
et al.
Management of renal replacement therapy in ICU patients: an international survey
.
Intensive Care Med
2013
;
39
:
101
–
108
6
et al.
Practice patterns in the management of acute renal failure in the critically ill patient: an international survey
.
Nephrol Dial Transplant
2006
;
21
:
690
–
696
7
et al.
Acute renal failure in intensive care units—causes, outcome, and prognostic factors of hospital mortality; a prospective, multicenter study. French Study Group on Acute Renal Failure
.
Crit Care Med
1996
;
24
:
192
–
198
8
et al.
Effect of acute renal failure requiring renal replacement therapy on outcome in critically ill patients
.
Crit Care Med
2002
;
30
:
2051
–
2058
9
et al.
Patient and kidney survival by dialysis modality in critically ill patients with acute kidney injury
.
Int J Artif Organs
2007
;
30
:
281
–
292
10
et al.
Survival by dialysis modality in critically ill patients with acute kidney injury
.
J Am Soc Nephrol
2006
;
17
:
3132
–
3138
11
et al.
Delivered dose of renal replacement therapy and mortality in critically ill patients with acute kidney injury
.
Crit Care
2009
;
13
:
R57
12
et al.
Spectrum of acute renal failure in the intensive care unit: the PICARD experience
.
Kidney Int
2004
;
66
:
1613
–
1621
13
.
Epidemiology, management, and outcome of severe acute renal failure of critical illness in Australia
.
Crit Care Med
2001
;
29
:
1910
–
1915
14
et al.
Acute renal failure requiring renal replacement therapy: incidence and outcome
.
QJM
2002
;
95
:
579
–
583
15
.
Agarwal, sepsis and acute kidney injury
.
J Am Soc Nephrol
2011
;
22
:
999
–
1006
16
et al.
Acute kidney injury in patients undergoing cardiac surgery
.
J Nephrol
2013
;
26
:
32
–
40
17
et al.
Septic acute kidney injury in critically ill patients: clinical characteristics and outcomes
.
Clin J Am Soc Nephrol
2007
;
2
:
431
–
439
18
et al.
Cardiorenal syndrome
.
J Am Coll Cardiol
2008
;
52
:
1527
–
1539
19
World Health Organization
.
International Classification of Diseases (ICD)-10, 2003 Version. 2nd edn
.
2003
.
20
.
Acute liver failure in Japan: definition, classification, and prediction of the outcome
.
J Gastroenterol
2012
;
47
:
849
–
861
21
et al.
Japanese-style intensive medical care improves prognosis for acute liver failure and the perioperative management of liver transplantation
.
Transplant Proc
2010
;
42
:
4109
–
4112
22
et al.
Continuous hemodiafiltration using a polymethyl methacrylate membrane hemofilter for severe acute pancreatitis
.
Contrib Nephrol
2010
;
166
:
54
–
63
23
et al.
Non-renal indications for continuous renal replacement therapy: current status in Japan
.
Contrib Nephrol
2010
;
166
:
47
–
53
24
et al.
Japanese severity score for acute pancreatitis well predicts in-hospital mortality: a nationwide survey of 17,901 cases
.
J Gastroenterol
2013
;
48
:
1384
–
1391
25
.
Diagnosis, epidemiology and outcomes of acute kidney injury
.
Clin J Am Soc Nephrol
2008
;
3
:
844
–
861
26
et al.
Epidemiology of acute kidney injury in the intensive care unit
.
Crit Care Res Pract
2013
;
2013
:
479730
27
et al.
Intensity of continuous renal-replacement therapy in critically ill patients
.
N Engl J Med
2009
;
361
:
1627
–
1638
28
et al.
Prognostic factors in acute renal failure due to sepsis. Results of a prospective multicentre study. The French Study Group on Acute Renal Failure
.
Nephrol Dial Transplant
1996
;
11
:
293
–
299
29
et al.
Survival to discharge among patients treated with continuous renal replacement therapy
.
Hemodial Int
2006
;
10
:
82
–
87
30
et al.
Discontinuation of continuous renal replacement therapy: a post hoc analysis of a prospective multicenter observational study
.
Crit Care Med
2009
;
37
:
2576
–
2582
31
et al.
Prognostic stratification in critically ill patients with acute renal failure requiring dialysis
.
Arch Intern Med
1995
;
155
:
1505
–
1511
32
.
Risk modeling in acute renal failure requiring dialysis: the introduction of a new model
.
Clin Nephrol
1996
;
46
:
206
–
211
33
et al.
Continuous renal replacement therapy: a worldwide practice survey. The beginning and ending supportive therapy for the kidney (B.E.S.T. kidney) investigators
.
Intensive Care Med
2007
;
33
:
1563
–
1570
34
et al.
Continuous renal replacement therapy outcomes in acute kidney injury and end-stage renal disease: a cohort study
.
Crit Care
2013
;
17
:
R109
35
et al.
Validity of low-intensity continuous renal replacement therapy
.
Crit Care Med
2013
;
41
:
2584
–
2591
36
et al.
Timing of renal replacement therapy initiation in acute renal failure: a meta-analysis
.
Am J Kidney Dis
2008
;
52
:
272
–
284
37
et al.
Timing of renal replacement therapy and clinical outcomes in critically ill patients with severe acute kidney injury
.
J Crit Care
2009
;
24
:
129
–
140
38
.
Clinical review: timing of renal replacement therapy
.
Crit Care
2011
;
15
:
223
39
et al.
Impact of hospital volume on operative mortality for major cancer surgery
.
JAMA
1998
;
280
:
1747
–
1751
40
et al.
Hospital volume-outcome relationships among medical admissions to ICUs
.
Chest
2005
;
128
:
1682
–
1689
41
et al.
A proposed algorithm for initiation of renal replacement therapy in adult critically ill patients
.
Crit Care
2009
;
13
:
317
42
Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group
.
KDIGO Clinical Practice Guideline for Acute Kidney Injury
.
Kidney Int
2012
;
2
(
Suppl
):
1
–
138
43
et al.
Performance and limitations of administrative data in the identification of AKI
.
Clin J Am Soc Nephrol
2014
;
9
:
682
–
689
44
et al.
APACHE II: a severity of disease classification system
.
Crit Care Med
1985
;
13
:
818
–
829
45
et al.
The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine
.
Intensive Care Med
1996
;
22
:
707
–
710
46
et al.
Outcome prediction of acute renal failure in medical intensive care
.
Intensive Care Med
1991
;
17
:
19
–
24
© The Author 2015. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
© The Author 2015. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
Source: https://academic.oup.com/ndt/article/30/6/988/2324934