- Email: [email protected]
Early Initiation of Continuous Renal Replacement Therapy Improves Clinical Outcomes in Patients With Acute Respiratory Distress Syndrome
CLINICAL INVESTIGATION
Early Initiation of Continuous Renal Replacement Therapy Improves Clinical Outcomes in Patients With Acute Respiratory Distress Syndrome Fang Han, MD, Renhua Sun, MD, Yin Ni, MD, Xiuping Hu, MD, Xv Chen, MD PhD, Lingzhi Jiang, MD, Aiping Wu, MD, Leilei Ma, MD, Minhua Chen, MD, Yunxiang Xv, MD and Yuexing Tu, MD
Abstract: Background: The acute respiratory distress syndrome (ARDS) is a common devastating syndrome in intensive care unit in critically ill patients. Continuous renal replacement therapy (CRRT) has been shown beneficial effects on oxygenation and survival in patients with ARDS. However, it is still controversial about the timing of initiation of CRRT. Methods: Fifty-three patients with ARDS admitted to intensive care unit in Zhejiang Provincial People’s Hospital, China from 2009 to 2013 were enrolled in the study. The authors compared ventilation parameter, including PaO2/FIO2, A-a gradient, positive endexpiratory pressure, plateau pressure, dynamic compliance and hemodynamic parameters, including central venous pressure, mean arterial pressure, cardiac index, extravascular lung water index, fluid balance between early initiation (within 12 hours after ARDS onset) and late initiation of CRRT (48 hours after ARDS onset) groups. The authors further investigated transforming growth factor (TGF)-b1 level changes in serum and bronchoalveolar lavage fluid (BALF) by enzyme-linked immunosorbent assay during 7 days of follow-up. Results: Significant improvement of oxygenation and shorter duration of mechanical ventilation were observed in early CRRT group during 7-day followup. In addition, TGF-b1 concentrations in serum and BALF were significantly decreased in patients with early initiation of CRRT compared to those with late initiation of CRRT on day 2 and day 7. Furthermore, patients who died of ARDS had higher levels of TGF-b1 in BALF than survivors. Conclusions: Our findings showed that early initiation of CRRT is associated with favorable clinical outcomes in ARDS patients, which might be due to the reduced serum and BALF TGF-b1 levels through CRRT. However, large multi-center studies are needed to make further recommendations as to the optimal use of CRRT in ARDS patient populations. Key Indexing Terms: Critically ill; ARDS; CRRT; Cytokines; TGF-b1. [Am J Med Sci 2015;349(3):199–205.]
T
he acute respiratory distress syndrome (ARDS) is a common devastating syndrome in critically ill patients in intensive care unit (ICU) with the mortality as high as 27% to 45%.1 It is characterized by increased permeability of alveolar-capillary barrier, which is composed of the microvascular endothelium and alveolar epithelium, then further resulting in extravascular accumulation of protein-rich edema fluid, leukocytes and erythrocytes into the alveolar space, as well as production of proinflammatory cytokines, such as tumor necrosis factor-a, From the Intensive Care Unit, Zhejiang Provincial People’s Hospital, Hangzhou, China. Submitted April 15, 2014; accepted in revised form October 6, 2014. Supported by the Grants from Science and Technology Department of Zhejiang Province, China (2011R50018-14) and Zhejiang Provincial Health Department, China (2012KYB021 and 2013RCA004). The authors have no conflicts of interest to disclose. Correspondence: Yuexing Tu, MD, Intensive Care Unit, Zhejiang Provincial People’s Hospital, 158 Shangtang Road, Hangzhou, Zhejiang 310014, China (E-mail: [email protected]).
The American Journal of the Medical Sciences
interleukin (IL)-1, IL-6, platelet-derived growth factor, etc.2–4 Clinically, it presents as refractory hypoxemia and noncardiogenic pulmonary edema. Currently, there is no available pharmacological therapy for ARDS and the main treatment is supportive.2 Measures aimed at reducing fluid overload, inflammation and especially the implementation of lung-protective ventilation strategy have been shown to improve clinical outcomes in patients with ARDS.1 Transforming growth factor (TGF)-b1 is a key mediator for developing ARDS. It is activated locally by integrin avb6 in cooperation with protease-activated receptor-15 to increase epithelial and endothelial permeability and promote alveolar flooding.6 It is a strong chemoattractant, not only for fibroblasts7 but also for T cells,8 macrophages and neutrophils.9 Moreover, it stimulates the expression of multiple cytokines, including tumor necrosis factor-a, IL-1 and platelet-derived growth factor, and inhibits expression of surfactant.10 Previous studies have reported a progressive and significant increase of serum TGF-b1 concentrations over time in patients with sepsis-induced ARDS11 and the elevated levels of TGF-b1 were correlated with decreases in PaO2/FIO2 ratio and survival.12 Furthermore, early studies in animal models have shown that blocking TGF-b with monoclonal antibody prevented hemorrhage-induced acute lung injury13 and bleomycin-induced lung fibrosis.14 Therefore, targeting TGF-b1 may represent a beneficial intervention in treating patients with ARDS. Continuous renal replacement therapy (CRRT) has been extensively used as renal support for critically ill patients in ICU. In recent years, it has also been extended to nonrenal indications, including sepsis, multiple organ dysfunction syndrome, congestive heart failure, ARDS, etc.15–19 Although CRRT has not been included in the standard therapy for ARDS, multiple studies have demonstrated that CRRT could improve survival in patients with ARDS due to different etiologies.20–25 However, it is still controversial about the timing of initiation of CRRT, which might exert significant influence on clinical outcomes. The current study, as a pilot project, aims to compare the effects of the timing of initiation of CRRT on clinical outcomes in patients with ARDS and to further investigate the changes of TGF-b1 levels in those patients.
MATERIALS AND METHODS Study Patients Selection Fifty-three patients aged between 25 and 65 years admitted to the ICU at Zhejiang Provincial People’s Hospital, China from March 2009 to March 2013 and who met the 1994 American-European Consensus definition for ARDS26 were enrolled in the study. Informed consent was obtained from patients or surrogates. The protocol was approved by the institutional review board of the hospital. Patients were excluded if they had immunodeficiency, autoimmune disease or cancer or
Volume 349, Number 3, March 2015
199
Han et al
were under any form of immunomodulating treatment. Patients were admitted to ICU once they were diagnosed with ARDS and were randomly assigned to early initiation of CRRT (within 12 hours of ICU admission) group or late initiation of CRRT (after 48 hour of ICU admission) group. Twenty healthy individuals matched for age and sex served as control group. Mechanical Ventilation All patients meeting ARDS criteria were ventilated with low tidal volumes (VT) of 6 mL/kg predicted body weight, inspiratory plateau pressure limited at 30 cm$H2O, initial ventilator rate of 30 breaths per minute adjusted to maintain a pH goal of 7.30 to 7.45, fraction of inspired oxygen (FIO2) ensuring PaO2 .60 mm Hg and positive end-expiratory pressure (PEEP) level that permitted the best oxygenation with lowest FIO2 without adverse hemodynamic effects. Patients were weaned and extubated according to the standard protocol described in the ARDSNet study.27 Continuous Renal Replacement Therapy The Aquarius system and Fresenius V600S polysulfone membrane hemofilters were used to deliver CRRT. The ultrafiltrate was removed at a rate of 250 mL/hr, blood flow was 150 to 200 mL/min. Heparin was used for anticoagulation of the circuit in patients without coagulopathy. Replacement solutions consist + of Na+ 147 mmol/L, Cl2 115 mmol/L, HCO2 3 76 mmol/L, Ca2 2.4 mmol/L, Mg2+ 0.7 mmol/L and Glu 200 mg/L. K+ was adjusted accordingly. Clinical Data Collection All patients were closely monitored from day 0 to day 7. Ventilator parameters, including PaO2, FIO2, VT, peak inspiratory pressure, PEEP, plateau pressure, etc, were recorded. PaO2/ FIO2 ratio and dynamic compliance (Cdyn) [VT/(peak inspiratory pressure 2 PEEP)] were calculated. Extravascular lung water index (EVLWI) was measured at the bedside using the PiCCO (Pulsion, Munich, Germany). Cardiac index is expressed in liters per minutes to body surface area (L$min21$m22). Mortality was defined as death occurring within 28 days after the patients’ enrollment. Specimen Collection All patients were intubated at the time of bronchoalveolar lavage (BAL). BAL was performed on days 0 (baseline), 2 and 7 of follow-up. The bronchoalveolar lavage fluid (BALF) was spun at 3,000 rpm for 10 minutes and the supernatant was obtained for TGF-b1 detection. Blood samples were drawn using an indwelling arterial catheter into sterilized, silicone-coated tubes at the time of ARDS diagnosis on days 0, 2 and 7 thereafter. Blood samples were collected in parallel from the healthy blood donors used as controls. Serum samples were obtained from clotted whole blood and frozen at 270°C for TGF-b1 detection. TGF-b1 Detection TGF-b1 concentrations in BALF and serum were determined by enzyme-linked immunosorbent assay kit purchased from Wuhan Boster Biological Engineering Co., Ltd. (Wuhan, China) according to the manufacturer’s instructions. Statistics All data are presented as mean 6 standard deviation. Descriptive statistics were computed to describe the demographic and clinical variables. Statistical analyses for differences between group means were conducted by unpaired Student’s t test.
200
P , 0.05 was considered statistically significant. SPSS 13.0 software (SPSS, Chicago, IL) and PRISM 5.0 (GraphPad Software, Inc., La Jolla, CA) were used to analyze the data.
RESULTS Patient Characteristics In the present study, the baseline patient characteristics are shown in Table 1. We analyzed a total number of 53 patients with ARDS who fulfilled the diagnostic criteria. All patients were treated according to the low-tidal-volume strategy previously reported.27 The mean age in patients with early CRRT was 50.36 years and 70.4% (19/27) was male while the mean age in patients with late CRRT was 53.12 years and 65.4% (17/26) was male. The mean age in healthy control group was 46.79 years and 70% was male. There were no differences in age and gender between 2 groups. Sepsis was the major cause of ARDS in 2 groups (40.7% and 38.5%, respectively). Other causes of ARDS included pneumonia, polytrauma, stroke and aspiration. Among the study subjects, 11.3% (3/27) in the early CRRT group and 7.7% (2/26) in the late CRRT group had increased creatinine levels. The average initial creatinine levels were 1.02 6 0.31 mg/dL and 0.94 6 0.34 mg/dL in the early CRRT and late CRRT group, respectively, which did not show a significant difference. Clinical Outcomes of the Study Patients As showed in Table 2, there were no significant differences in PaO2/FIO2, PEEP, plateau pressure and A-a gradient in patients between early CRRT and late CRRT at baseline of day 0. However, the trends of steady improvement of these parameters were observed in both groups over the 7 days of follow-up. There were significant improvements in PaO2/FIO2 at day 7, and PEEP and A-a gradient at days 4 and 7 in early CRRT group compared with late CRRT group. We further evaluated the effects of CRRT on Cdyn. Consistent with the changes of PaO2/FIO2, we observed similar trends of steady increase of Cdyn between early and late initiation of CRRT patients among 7 days of follow-up. On day 7 of follow-up, there was significant higher Cdyn in patients with early
TABLE 1. Subjects’ characteristics and clinical features Early Late CRRT CRRT Control P Patients Age Gender (M/F) Causes of ARDS Sepsis Pneumonia Polytrauma Stroke Aspiration AKI patient number Creatinine (mg/dL)
27 50.36 6 16.97 19/8
26 53.12 6 15.05 17/9
20 46.79 6 18.92 14/6
— —
40.7% (11) 18.5% (5) 18.5% (5) 14.8% (4) 7.4% (2) 11.1% (3)
38.5% (10) 23.1% (6) 15.4% (4) 15.4% (4) 15.4% (4) 7.7% (2)
— — — — — —
0.86533 0.68249 0.95046 0.95726 0.96463 —
1.02 6 0.31
0.94 6 0.34
—
0.35180
—
ARDS, acute respiratory distress syndrome; AKI, acute kidney injury; CRRT, continuous renal replacement therapy; F, female; M, male.
Volume 349, Number 3, March 2015
Early CRRT Improves Clinical Outcomes in ARDS
TABLE 2. Comparison of ventilatory function, hemodynamics and fluid balance Day 0 Day 2
PaO2/FIO2 PEEP (cm$H2O) Plateau pressure (cm$H2O) A-a gradient (mm Hg) Cdyn CVP (mm Hg) MAP (mm Hg) CI (L$min21$m23) EVLWI (mL/kg) Fluid balance (mL)
Early CRRT
Late CRRT
Early CRRT
121.45 6 46.39 13.45 6 5.78 31.85 6 5.62
134.07 6 38.86 12.98 6 5.23 30.02 6 5.39
139.40 6 37.49 8.90 6 4.49b 24.90 6 5.01b
46.18 6 11.12
44.09 6 10.92
17.18 6 11.12 19.09 14.09 6 3.78 14.78 69.23 6 25.1 71.15 3.13 6 1.27 2.95 15.26 6 5.69 16.07 163.31 6 32.67 1058.27
6 6 6 6 6 6
12.31 4.09 23.92 1.03 5.35 169.85a
Day 7
Late CRRT 126.95 6 36.39 11.76 6 5.07a 28.63 6 6.67
Early CRRT
Late CRRT
220.19 6 50.77 5.17 6 2.36b 19.97 6 3.83b
187.73 6 49.66a 7.59 6 2.68ab 23.88 6 4.41b
30.06 6 13.13b
39.33 6 15.71a
16.73 6 11.92b
6 6 6 6 6 6
6 6 6 6 6 6
6 6 6 6 6 6
20.06 11.90 75.71 3.28 10.26 50.38
13.13 18.76 3.21b 15.01 23.9 70.63 1.16 2.76 13.99 4.75b 23.20b 595.18
10.08 32.73 3.99a 9.91 25.33 80.32 1.12 3.53 3.87a 5.05 137.96ab 177.56
11.92 2.86b 27.5 1.20 2.33b 40.31
25.66 6 12.39ab 25.37 10.57 77.93 3.12 6.77 183.67
6 6 6 6 6 6
10.27a 3.73b 24.99 1.17 2.39ab 69.53b
P , 0.05 compared with early CRRT at the same time point. P , 0.05 compared with baseline of day 0. BALF, bronchoalveolar lavage fluid; Cdyn, dynamic compliance; CI, cardiac index; CVP, central venous pressure; EVLWI, extrtravascular lung water index; MAP, mean arterial pressure; PEEP, positive end-expiratory pressure. a b
CRRT compared with patients with late CRRT (32.73 6 11.92 versus 25.37 6 10.27, respectively, P 5 0.0199). EVLWI reflects pulmonary edema and correlates with outcome. An EVLWI between 3 and 7 mL/kg is considered normal and above 10 mL/kg is associated with clinical pulmonary edema.28,29 We observed significant improvements of EVLWI in both groups after initiation of CRRT. Early CRRT patients showed lower EVLWI than late CRRT patients at days 2 and 7 of follow-up (Table 2). In addition, we showed that patients with early CRRT required shorter duration of CRRT (81.39 6 20.57 hours versus 103.68 6 27.9 hours, respectively, P 5 0.00168) and mechanical ventilation compared with those with late CRRT (10.16 6 3.52 days versus 12.97 6 4.33 days, respectively, P 5 0.0123) (Figure 1). TGF-b1 Concentrations in Serum and BALF Compared with healthy volunteers, serum TGF-b1 concentrations were markedly increased in patients with ARDS. To further determine the effects of CRRT on the level of TGF-b1, we analyzed serum and BALF TGF-b1 concentration changes. As summarized in Table 3, there were no significant differences in serum TGF-b1 concentration noted among the patients with either early or late initiation of CRRT at baseline day 0. There was a steady reduction of serum TGF-b1 over the 7 days of follow-up in early CRRT patients. Serum TGF-b1 decreased 6% (from 14.25 6 5.92 ng/mL to 13.35 6 5.24 ng/mL) on day 2 and 57% (from 13.35 6 5.24 ng/mL to 5.79 6 1.28 ng/mL)
FIGURE 1. Comparison of duration of CRRT and mechanical ventilation between early initiation of CRRT and late initiation of CRRT groups. CRRT, continuous renal replacement therapy. Copyright © 2014 by the Southern Society for Clinical Investigation.
on day 7, respectively. However, serum TGF-b1 in late CRRT patients only showed reduction after initiation of CRRT on day 2 (62%, from 17.23 6 6.56 ng/mL to 6.56 6 1.39 ng/mL). In addition, serum TGF-b1 concentrations were significantly decreased in patients with early CRRT compared with those with late CRRT on day 2 (13.35 6 5.24 ng/mL versus 17.23 6 6.56 ng/mL, P 5 0.0209) and day 7 (5.79 6 1.28 ng/mL versus 6.56 6 1.39 ng/mL, P 5 0.0408) of follow-up. In addition, we also compared TGF-b1 level changes in BALF. BALF TGF-b1 concentrations were much higher than those in serum, indicating it might be a more sensitive parameter. We observed similar trends in BALF TGF-b1 concentration change as in serum over the 7-day follow-up. There was a steady reduction of BALF TGF-b1 over the 7 days of followup in early CRRT patients. However, BALF TGF-b1 in late CRRT patients only showed reduction after initiation of CRRT after day 2. In addition, BALF TGF-b1 concentrations were significantly lower in patients with early CRRT compared with those with late CRRT on day 2 (16.83 6 6.77 ng/mL versus 21.95 6 9.16 ng/mL, P 5 0.0244) and day 7 (9.18 6 3.19 ng/mL versus 12.83 6 3.66 ng/mL, P 5 0.0003) of follow-up.
TABLE 3. Comparison of TGF-b1 concentrations (ng/mL) in serum and BALF Group Day 0 Day 2 Day 7 Early CRRT Serum (ng/mL) BALF (ng/mL) Late CRRT Serum (ng/mL) BALF (ng/mL) Control Serum (ng/mL) a b
14.25 6 5.92 13.35 6 5.24 19.09 6 7.99 16.83 6 6.77
5.79 6 1.28 9.18 6 3.19
13.67 6 4.85 17.23 6 6.56a 6.56 6 1.39a 17.89 6 6.01 21.95 6 9.16b 12.83 6 3.66b 2.25 6 1.03
—
—
P , 0.05. P , 0.01, compared with early CRRT group.
201
Han et al
Comparison of Prognostic Value As showed above, early CRRT is beneficial to improve oxygenation in patients with ARDS; we further sought to compare the survival between 2 groups. There was a trend not reaching statistical difference (P 5 0.3167) in mortality rate reduction between early CRRT patients (22.2%, 6/27) and late CRRT patients (34.6%, 9/26). Next, we focused our investigation to address the prognostic value of serum and BALF TGF-b1 levels. Figure 2 depicts the comparison of the TGF-b1 levels between survivors and nonsurvivors over the 7-day follow-up. There were no significant differences in serum TGF-b1 at days 2 and 7 of follow-up between survivors and nonsurvivors in either group. However, on day 7 of follow-up, patients who died of ARDS had higher levels of TGF-b1 in BALF than survivors in both groups (early CRRT: 12.13 6 2.75 ng/mL versus 8.05 6 2.97 ng/mL, P 5 0.00588; late CRRT: 14.67 6 3.19 ng/mL versus 10.12 6 3.35 ng/mL, P 5 0.00268).
DISCUSSION In the current study, we have showed that early initiation of CRRT improves clinical outcomes in patients with ARDS, including improvement of oxygenation and decreased ventilator duration. Furthermore, our data suggested that it maybe associated with removal of lung water and inflammatory cytokine TGF-b1 in serum and BALF. In ARDS, the fluid clearance process is defective with compromised alveolar-capillary barrier function.30 It is widely believed that edema fluid must be cleared for patients with ARDS to survive.31 CRRT allows extracorporeal treatment in critically ill patients with fluid overload. Although it has not become a standard protocol for ARDS treatment, using CRRT to support ARDS treatment has been studied and reported in several medical centers for the past decade.23–25,32–35 Table 4 summarized the studies on the use of CRRT in these ARDS patients associated with the common causes, including sepsis, trauma, postoperative, pancreatitis, pneumonia, etc. These studies showed the therapeutic benefits of CRRT in ARDS patients. In addition, several reports also suggested improvement of oxygenation and survival benefit using CRRT in the treatment of ARDS under untypical medical conditions, including severe pneumonia after renal transplantation,20 bone morrow transplantation or chemotherapy21 and allogeneic hematopoietic stem cell transplantation.22 However, Hoste et al36 found neither benefit nor harm about CRRT on oxygenation in patients with ARDS and acute kidney injury (AKI). Our results confirmed the beneficial effects on the clinical outcomes with the initiation of CRRT in patients with ARDS caused by sepsis, pneumonia, polytrauma, stroke or aspiration. The optimal timing for CRRT initiation remains undetermined because the decision to start CRRT in critically ill patients must depend on individual patient’s condition. In this study, we compared early initiation (within 12 hours after
FIGURE 2. Comparison of serum (left) and BALF (right) TGF-b1 concentration changes between survivors and nonsurvivors. BALF, bronchoalveolar lavage fluid.
202
ARDS onset) and late initiation (48 hours after ARDS onset) of CRRT and observed significant improvement of clinical outcomes in early initiation of CRRT group. Our results are consistent with a number of previous studies, which suggested that early CRRT had a favorable outcome in critically ill patients. Oh et al37 demonstrated that in septic AKI patients, early CRRT treatment was independently associated with a lower mortality rate after adjustment for demographic characteristics, the severity of disease and causative organisms. Late starting CRRT in a multicenter cohort of ICU patients with severe AKI was associated with a longer stay in hospital and higher dialysis dependence.38 The meta-analysis by Seabra et al39 suggests that early initiation of CRRT in patients with AKI was associated with a statistically significant 28% mortality risk reduction (relative risk: 0.72; 95% confidence interval [CI]: 0.64–0.82; P , 0.001). A more recent meta-analysis also supported the early CRRT therapy due to a significant improvement in 28-day mortality (odds ratio: 0.45; 95% CI: 0.28–0.72).40 Consistently, we also showed a trend of reduction in mortality, although there was no statistical significance due to limited sample size. A larger cohort of patients will be needed to fully address this issue. Mechanisms other than removing fluid have been considered in an attempt to explain the beneficial effect of CRRT on ARDS. Besides hypothermia during CRRT, which decreases oxygen consumption and removing extravascular lung fluids, some researchers believe that CRRT is beneficial in eliminating inflammatory mediators that contribute to the pathogenesis of ARDS.17,34,41 Current understanding of the pathophysiology underlying ARDS highlights large number of neutrophil accumulation in the lungs and increased expression of proinflammatory cytokines, which cause damage in the integrity of the epithelial cells and endothelial cells.2,42 TGFb1, 1 of the 3 isoforms of TGF-b, plays a critical role in tissue repair after injury in multiple organs, including the lung.12 It has been extensively evaluated during the late fibroproliferative stage of ARDS.11,12,43 Now, researcher further realized that TGF-b1 actively participated in the early exudative stage of the ARDS pathogenesis. As a proinflammatory cytokine, it induces T-cell and natural killer cell suppression and monocyte/macrophage deactivation,12 and more recently, it has been shown to direct trafficking of the epithelial sodium channel and cause pulmonary edema.44 Therefore, it is critical to target TGF-b1 during the early treatment of patients with ARDS. Here, we showed that TGF-b1 concentrations in serum and BALF were significantly elevated in ARDS patients, which is consistent with previous reports by Budinger et al7 and Fahy et al.9 They showed the presence of TGF-b1 in lung tissue from patients with ARDS within the first 24 hours of diagnosis and increased TGF-b1 levels in lung fluids from patients with acute lung injury/ARDS, respectively. We further showed that after initiation of CRRT, both serum and BALF TGF-b1 concentrations were significantly reduced, which was associated with improvement of clinical outcomes, including 2 parameters PaO2/FIO2 and Cdyn, and a shorter duration on ventilation. Our results were in line with the previous studies by Dhainaut et al12 that TGF-b1 concentrations were correlated with decreases in the PaO2/FIO2 ratio and lower levels of TGF-b1 correlated with more ventilator-free and ICU-free days.7 In addition, several groups have been studying the efficacy of cytokines removal therapy with CRRT and showed promising results.34,45 Therefore, we believe that the observed clinical outcomes improvement in this study at least partially resulted from an attenuated inflammatory response especially by removal of TGF-b1 through CRRT. Volume 349, Number 3, March 2015
Early CRRT Improves Clinical Outcomes in ARDS
TABLE 4. Summary of the studies using CRRT therapy in patients with ARDS Age of No. of patients Associations of Reference no. patients (yr) ARDS CRRT strategies 23
65
24
14
25
12
34
32
43 6 13
Main outcomes
Sepsis: 8; 37 patients under CHVHF With prolonged time of postoperative: and other routine treatment, CHVHF 14; trauma: treatments; 28 patients group had significantly 12; shock: 15; under routine treatments elevated PaO2/FiO2, decreased EVLWI and pancreatitis: only as control group PaCO2 levels; duration 10; of mechanical ventilation pneumonia: 6 and ICU stay were shortened in CHVHF group 44.4 (range, Trauma: 10; CAVH-D for a mean of 11 patients had significant 14–68) postoperative: 65.2 hr (range, 12–140 improvement of 4; hr) respiratory function: pancreatitis: 1 mean FIO2 decreased from 0.73 to 0.45; PEEP from 14.3 to 8.9 cm; peak airway pressures fell from a mean of 60 to 45 mm Hg; no significant change in CI or wedge pressure, but oxygen consumption rose from a mean of 279 to 409 mL/m 50 (range, Sepsis CHVHF Significant decreases in 22–60) MPAP, PVR and TFC after 48 hours of treatment; oxygen delivery, oxygen consumption and oxygen extraction rate were stabilized at 72 hours with amelioration of PaO2, PaO2/FiO2 and peak airway pressure; significant decreases in blood levels of cytokines (tumor necrosis factor-a, IL-6 and IL-8) 61.5 6 15.0 Sepsis resulting Continuous Significantly decreased from hemodiafiltration; using blood levels of cytokines peritonitis, a PMMA-CHDF for and successful water pneumonia or renal replacement and removal without other cytokine removal for 3 changing central venous diseases: 26/ days versus treatment of pressure in the PMMA32; other: 6/ IHD and CHF in 19 CHDF group but not in 32 patients the IHD + CHF group; significant correlations between changes in blood levels of cytokines (IL-6 and IL-8) and changes in respiratory index were demonstrated in the PMMA-CHDF group
Survival rate or cases with severe side effects 28-day cumulative survival rate: CHVHF: 86.5%; control group: 71.4%
One patient bled from a loose tubing connection; 3 patients were grossly unstable when CAVH-D was begun
—
28-day cumulative survival rate: PMMA-CHDF: 68.8%; IDH + CHF: 36.8%
CAVHD, continuous arteriovenous countercurrent hemodialysis; CHF, continuous hemofiltration; CHVHF, continuous high-volume hemofiltration; IHD, intermittent hemodialysis; MPAP, mean pulmonary arterial pressure; PMMA-CHDF, polymethylmethacrylate membrane hemofilter; PVR, pulmonary vascular resistance; TFC, thoracic fluid content.
Limitations This study has several limitations. First, instead of using the most current Berlin ARDS definition, 1994 AmericanEuropean Consensus definition for ARDS was used in our Copyright © 2014 by the Southern Society for Clinical Investigation.
study during the enrollment period. Second, we only included patients with ARDS on ICU admission and the number of patient is relatively small. Third, our study only investigated 1 particular cytokine—TGF-b1. Several other important cytokines
203
Han et al
that have been shown to participate in the pathogenesis of ARDS were not assessed. Fourth, defining “early” versus “late” in the initiation of CRRT in critically ill patients is very challenging. Whether timing of CRRT should be measured from ICU admission, from the onset of ARDS or by use of serum biomarkers remains controversial.
CONCLUSIONS
Taken together, our findings suggested that early initiation of CRRT is associated with favorable clinical outcomes in patients with ARDS, which might be due to the reduced serum and BALF TGF-b1 levels through CRRT. However, large multicenter studies are needed to make further recommendations as to the optimal use of CRRT in ARDS patient populations. ACKNOWLEDGMENTS The authors thank Dr. Ziqiang Zhu at the National Institutes of Health for the critical review of the manuscript. REFERENCES 1. Ranieri VM, Rubenfeld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012;307:2526–33. 2. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342:1334–49. 3. Bartram U, Speer CP. The role of transforming growth factor beta in lung development and disease. Chest 2004;125:754–65. 4. Hamacher J, Lucas R, Lijnen HR, et al. Tumor necrosis factor-alpha and angiostatin are mediators of endothelial cytotoxicity in bronchoalveolar lavages of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;166:651–6. 5. Pittet JF, Griffiths MJ, Geiser T, et al. TGF-beta is a critical mediator of acute lung injury. J Clin Invest 2001;107:1537–44. 6. Jenkins RG, Su X, Su G, et al. Ligation of protease-activated receptor 1 enhances alpha(v)beta6 integrin-dependent TGF-beta activation and promotes acute lung injury. J Clin Invest 2006;116:1606–14. 7. Budinger GR, Chandel NS, Donnelly HK, et al. Active transforming growth factor-beta1 activates the procollagen I promoter in patients with acute lung injury. Intensive Care Med 2005;31:121–8. 8. Adams DH, Hathaway M, Shaw J, et al. Transforming growth factorbeta induces human T lymphocyte migration in vitro. J Immunol 1991; 147:609–12. 9. Fahy RJ, Lichtenberger F, McKeegan CB, et al. The acute respiratory distress syndrome: a role for transforming growth factor-beta 1. Am J Respir Cell Mol Biol 2003;28:499–503. 10. Beers MF, Solarin KO, Guttentag SH, et al. TGF-beta1 inhibits surfactant component expression and epithelial cell maturation in cultured human fetal lung. Am J Physiol 1998;275:L950–60. 11. de Pablo R, Monserrat J, Reyes E, et al. Sepsis-induced acute respiratory distress syndrome with fatal outcome is associated to increased serum transforming growth factor beta-1 levels. Eur J Intern Med 2012; 23:358–62. 12. Dhainaut JF, Charpentier J, Chiche JD. Transforming growth factorbeta: a mediator of cell regulation in acute respiratory distress syndrome. Crit Care Med 2003;31(suppl 4):S258–64. 13. Shenkar R, Coulson WF, Abraham E. Anti-transforming growth factor-beta monoclonal antibodies prevent lung injury in hemorrhaged mice. Am J Respir Cell Mol Biol 1994;11:351–7. 14. Giri SN, Hyde DM, Hollinger MA. Effect of antibody to transforming growth factor beta on bleomycin induced accumulation of lung collagen in mice. Thorax 1993;48:959–66.
204
15. Bellomo R, Honore PM, Matson J, et al. Extracorporeal blood treatment (EBT) methods in SIRS/Sepsis. Int J Artif Organs 2005;28:450–8. 16. Dirkes S, Hodge K. Continuous renal replacement therapy in the adult intensive care unit: history and current trends. Crit Care Nurse 2007;27: 61–6; 68–72, 74–80; quiz 81. 17. Ronco C, Bonello M, Bordoni V, et al. Extracorporeal therapies in non-renal disease: treatment of sepsis and the peak concentration hypothesis. Blood Purif 2004;22:164–74. 18. Abdo AA, Castellanos R, Rocha M, et al. Continuous venovenous hemodiafiltration in patients with multiple organ dysfunction syndrome in an intensive care unit. MEDICC Rev 2012;14:26–30. 19. Marenzi G, Lauri G, Grazi M, et al. Circulatory response to fluid overload removal by extracorporeal ultrafiltration in refractory congestive heart failure. J Am Coll Cardiol 2001;38:963–8. 20. Sun Q, Liu ZH, Chen J, et al. An aggressive systematic strategy for acute respiratory distress syndrome caused by severe pneumonia after renal transplantation. Transpl Int 2006;19:110–16. 21. DiCarlo JV, Alexander SR, Agarwal R, et al. Continuous venovenous hemofiltration may improve survival from acute respiratory distress syndrome after bone marrow transplantation or chemotherapy. J Pediatr Hematol Oncol 2003;25:801–5. 22. Elbahlawan L, West NK, Avent Y, et al. Impact of continuous renal replacement therapy on oxygenation in children with acute lung injury after allogeneic hematopoietic stem cell transplantation. Pediatr Blood Cancer 2010;55:540–5. 23. Zhang JC, Chu YF, Zeng J, et al. Effect of continuous high-volume hemofiltration in patients with severe acute respiratory distress syndrome [in Chinese]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2013; 25:145–8. 24. Garzia F, Todor R, Scalea T. Continuous arteriovenous hemofiltration countercurrent dialysis (CAVH-D) in acute respiratory failure (ARDS). J Trauma 1991;31:1277–84; discussion 1284–5. 25. Liu CW, Zheng YK, Lu J, et al. Effect of high volume hemofiltration on respiration, hemodynamics and oxygen metabolism in sepsis with acute respiratory distress syndrome [in Chinese]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2006;18:657–60. 26. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994;149:818–24. 27. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000;342:1301–8. 28. Kuzkov VV, Kirov MY, Sovershaev MA, et al. Extravascular lung water determined with single transpulmonary thermodilution correlates with the severity of sepsis-induced acute lung injury. Crit Care Med 2006;34:1647–53. 29. Michard F. Bedside assessment of extravascular lung water by dilution methods: temptations and pitfalls. Crit Care Med 2007;35:1186–92. 30. Matthay MA, Ware LB, Zimmerman GA. The acute respiratory distress syndrome. J Clin Invest 2012;122:2731–40. 31. Sznajder JI. Alveolar edema must be cleared for the acute respiratory distress syndrome patient to survive. Am J Respir Crit Care Med 2001; 163:1293–4. 32. Grasselli G, Foti G, Patroniti N, et al. A case of ARDS associated with influenza A—H1N1 infection treated with extracorporeal respiratory support. Minerva Anestesiol 2009;75:741–5. 33. Schetz M. Non-renal indications for continuous renal replacement therapy. Kidney Int Suppl 1999;72:S88–94.
Volume 349, Number 3, March 2015
Early CRRT Improves Clinical Outcomes in ARDS
34. Matsuda K, Moriguchi T, Oda S, et al. Efficacy of continuous hemodiafiltration with a cytokine-adsorbing hemofilter in the treatment of acute respiratory distress syndrome. Contrib Nephrol 2010; 166:83–92.
40. Karvellas CJ, Farhat MR, Sajjad I, et al. A comparison of early versus late initiation of renal replacement therapy in critically ill patients with acute kidney injury: a systematic review and meta-analysis. Crit Care 2011;15:R72.
35. Davenport A. Can modification of renal replacement therapy improve the outcome of patients with systemic inflammatory response syndrome? Blood Purif 2006;24:317–18.
41. Ronco C, Tetta C, Mariano F, et al. Interpreting the mechanisms of continuous renal replacement therapy in sepsis: the peak concentration hypothesis. Artif Organs 2003;27:792–801.
36. Hoste EA, Vanholder RC, Lameire NH, et al. No early respiratory benefit with CVVHDF in patients with acute renal failure and acute lung injury. Nephrol Dial Transplant 2002;17:2153–8.
42. Meduri GU, Kohler G, Headley S, et al. Inflammatory cytokines in the BAL of patients with ARDS. Persistent elevation over time predicts poor outcome. Chest 1995;108:1303–14.
37. Oh HJ, Shin DH, Lee MJ, et al. Early initiation of continuous renal replacement therapy improves patient survival in severe progressive septic acute kidney injury. J Crit Care 2012;27:e749–718.
43. Santana A, Saxena B, Noble NA, et al. Increased expression of transforming growth factor beta isoforms (beta 1, beta 2, beta 3) in bleomycininduced pulmonary fibrosis. Am J Respir Cell Mol Biol 1995;13:34–44.
38. Bagshaw SM, Uchino S, Bellomo R, et al. Timing of renal replacement therapy and clinical outcomes in critically ill patients with severe acute kidney injury. J Crit Care 2009;24:129–40.
44. Peters DM, Vadasz I, Wujak L, et al. TGF-beta directs trafficking of the epithelial sodium channel ENaC which has implications for ion and fluid transport in acute lung injury. Proc Natl Acad Sci U S A 2013;111: E374–83.
39. Seabra VF, Balk EM, Liangos O, et al. Timing of renal replacement therapy initiation in acute renal failure: a meta-analysis. Am J Kidney Dis 2008;52:272–84.
45. Oda S, Sadahiro T, Hirayama Y, et al. Non-renal indications for continuous renal replacement therapy: current status in Japan. Contrib Nephrol 2010;166:47–53.
Copyright © 2014 by the Southern Society for Clinical Investigation.
205
Early Initiation of Continuous Renal Replacement Therapy Improves Clinical Outcomes in Patients With Acute Respiratory Distress Syndrome Fang Han, MD, Renhua Sun, MD, Yin Ni, MD, Xiuping Hu, MD, Xv Chen, MD PhD, Lingzhi Jiang, MD, Aiping Wu, MD, Leilei Ma, MD, Minhua Chen, MD, Yunxiang Xv, MD and Yuexing Tu, MD
Abstract: Background: The acute respiratory distress syndrome (ARDS) is a common devastating syndrome in intensive care unit in critically ill patients. Continuous renal replacement therapy (CRRT) has been shown beneficial effects on oxygenation and survival in patients with ARDS. However, it is still controversial about the timing of initiation of CRRT. Methods: Fifty-three patients with ARDS admitted to intensive care unit in Zhejiang Provincial People’s Hospital, China from 2009 to 2013 were enrolled in the study. The authors compared ventilation parameter, including PaO2/FIO2, A-a gradient, positive endexpiratory pressure, plateau pressure, dynamic compliance and hemodynamic parameters, including central venous pressure, mean arterial pressure, cardiac index, extravascular lung water index, fluid balance between early initiation (within 12 hours after ARDS onset) and late initiation of CRRT (48 hours after ARDS onset) groups. The authors further investigated transforming growth factor (TGF)-b1 level changes in serum and bronchoalveolar lavage fluid (BALF) by enzyme-linked immunosorbent assay during 7 days of follow-up. Results: Significant improvement of oxygenation and shorter duration of mechanical ventilation were observed in early CRRT group during 7-day followup. In addition, TGF-b1 concentrations in serum and BALF were significantly decreased in patients with early initiation of CRRT compared to those with late initiation of CRRT on day 2 and day 7. Furthermore, patients who died of ARDS had higher levels of TGF-b1 in BALF than survivors. Conclusions: Our findings showed that early initiation of CRRT is associated with favorable clinical outcomes in ARDS patients, which might be due to the reduced serum and BALF TGF-b1 levels through CRRT. However, large multi-center studies are needed to make further recommendations as to the optimal use of CRRT in ARDS patient populations. Key Indexing Terms: Critically ill; ARDS; CRRT; Cytokines; TGF-b1. [Am J Med Sci 2015;349(3):199–205.]
T
he acute respiratory distress syndrome (ARDS) is a common devastating syndrome in critically ill patients in intensive care unit (ICU) with the mortality as high as 27% to 45%.1 It is characterized by increased permeability of alveolar-capillary barrier, which is composed of the microvascular endothelium and alveolar epithelium, then further resulting in extravascular accumulation of protein-rich edema fluid, leukocytes and erythrocytes into the alveolar space, as well as production of proinflammatory cytokines, such as tumor necrosis factor-a, From the Intensive Care Unit, Zhejiang Provincial People’s Hospital, Hangzhou, China. Submitted April 15, 2014; accepted in revised form October 6, 2014. Supported by the Grants from Science and Technology Department of Zhejiang Province, China (2011R50018-14) and Zhejiang Provincial Health Department, China (2012KYB021 and 2013RCA004). The authors have no conflicts of interest to disclose. Correspondence: Yuexing Tu, MD, Intensive Care Unit, Zhejiang Provincial People’s Hospital, 158 Shangtang Road, Hangzhou, Zhejiang 310014, China (E-mail: [email protected]).
The American Journal of the Medical Sciences
interleukin (IL)-1, IL-6, platelet-derived growth factor, etc.2–4 Clinically, it presents as refractory hypoxemia and noncardiogenic pulmonary edema. Currently, there is no available pharmacological therapy for ARDS and the main treatment is supportive.2 Measures aimed at reducing fluid overload, inflammation and especially the implementation of lung-protective ventilation strategy have been shown to improve clinical outcomes in patients with ARDS.1 Transforming growth factor (TGF)-b1 is a key mediator for developing ARDS. It is activated locally by integrin avb6 in cooperation with protease-activated receptor-15 to increase epithelial and endothelial permeability and promote alveolar flooding.6 It is a strong chemoattractant, not only for fibroblasts7 but also for T cells,8 macrophages and neutrophils.9 Moreover, it stimulates the expression of multiple cytokines, including tumor necrosis factor-a, IL-1 and platelet-derived growth factor, and inhibits expression of surfactant.10 Previous studies have reported a progressive and significant increase of serum TGF-b1 concentrations over time in patients with sepsis-induced ARDS11 and the elevated levels of TGF-b1 were correlated with decreases in PaO2/FIO2 ratio and survival.12 Furthermore, early studies in animal models have shown that blocking TGF-b with monoclonal antibody prevented hemorrhage-induced acute lung injury13 and bleomycin-induced lung fibrosis.14 Therefore, targeting TGF-b1 may represent a beneficial intervention in treating patients with ARDS. Continuous renal replacement therapy (CRRT) has been extensively used as renal support for critically ill patients in ICU. In recent years, it has also been extended to nonrenal indications, including sepsis, multiple organ dysfunction syndrome, congestive heart failure, ARDS, etc.15–19 Although CRRT has not been included in the standard therapy for ARDS, multiple studies have demonstrated that CRRT could improve survival in patients with ARDS due to different etiologies.20–25 However, it is still controversial about the timing of initiation of CRRT, which might exert significant influence on clinical outcomes. The current study, as a pilot project, aims to compare the effects of the timing of initiation of CRRT on clinical outcomes in patients with ARDS and to further investigate the changes of TGF-b1 levels in those patients.
MATERIALS AND METHODS Study Patients Selection Fifty-three patients aged between 25 and 65 years admitted to the ICU at Zhejiang Provincial People’s Hospital, China from March 2009 to March 2013 and who met the 1994 American-European Consensus definition for ARDS26 were enrolled in the study. Informed consent was obtained from patients or surrogates. The protocol was approved by the institutional review board of the hospital. Patients were excluded if they had immunodeficiency, autoimmune disease or cancer or
Volume 349, Number 3, March 2015
199
Han et al
were under any form of immunomodulating treatment. Patients were admitted to ICU once they were diagnosed with ARDS and were randomly assigned to early initiation of CRRT (within 12 hours of ICU admission) group or late initiation of CRRT (after 48 hour of ICU admission) group. Twenty healthy individuals matched for age and sex served as control group. Mechanical Ventilation All patients meeting ARDS criteria were ventilated with low tidal volumes (VT) of 6 mL/kg predicted body weight, inspiratory plateau pressure limited at 30 cm$H2O, initial ventilator rate of 30 breaths per minute adjusted to maintain a pH goal of 7.30 to 7.45, fraction of inspired oxygen (FIO2) ensuring PaO2 .60 mm Hg and positive end-expiratory pressure (PEEP) level that permitted the best oxygenation with lowest FIO2 without adverse hemodynamic effects. Patients were weaned and extubated according to the standard protocol described in the ARDSNet study.27 Continuous Renal Replacement Therapy The Aquarius system and Fresenius V600S polysulfone membrane hemofilters were used to deliver CRRT. The ultrafiltrate was removed at a rate of 250 mL/hr, blood flow was 150 to 200 mL/min. Heparin was used for anticoagulation of the circuit in patients without coagulopathy. Replacement solutions consist + of Na+ 147 mmol/L, Cl2 115 mmol/L, HCO2 3 76 mmol/L, Ca2 2.4 mmol/L, Mg2+ 0.7 mmol/L and Glu 200 mg/L. K+ was adjusted accordingly. Clinical Data Collection All patients were closely monitored from day 0 to day 7. Ventilator parameters, including PaO2, FIO2, VT, peak inspiratory pressure, PEEP, plateau pressure, etc, were recorded. PaO2/ FIO2 ratio and dynamic compliance (Cdyn) [VT/(peak inspiratory pressure 2 PEEP)] were calculated. Extravascular lung water index (EVLWI) was measured at the bedside using the PiCCO (Pulsion, Munich, Germany). Cardiac index is expressed in liters per minutes to body surface area (L$min21$m22). Mortality was defined as death occurring within 28 days after the patients’ enrollment. Specimen Collection All patients were intubated at the time of bronchoalveolar lavage (BAL). BAL was performed on days 0 (baseline), 2 and 7 of follow-up. The bronchoalveolar lavage fluid (BALF) was spun at 3,000 rpm for 10 minutes and the supernatant was obtained for TGF-b1 detection. Blood samples were drawn using an indwelling arterial catheter into sterilized, silicone-coated tubes at the time of ARDS diagnosis on days 0, 2 and 7 thereafter. Blood samples were collected in parallel from the healthy blood donors used as controls. Serum samples were obtained from clotted whole blood and frozen at 270°C for TGF-b1 detection. TGF-b1 Detection TGF-b1 concentrations in BALF and serum were determined by enzyme-linked immunosorbent assay kit purchased from Wuhan Boster Biological Engineering Co., Ltd. (Wuhan, China) according to the manufacturer’s instructions. Statistics All data are presented as mean 6 standard deviation. Descriptive statistics were computed to describe the demographic and clinical variables. Statistical analyses for differences between group means were conducted by unpaired Student’s t test.
200
P , 0.05 was considered statistically significant. SPSS 13.0 software (SPSS, Chicago, IL) and PRISM 5.0 (GraphPad Software, Inc., La Jolla, CA) were used to analyze the data.
RESULTS Patient Characteristics In the present study, the baseline patient characteristics are shown in Table 1. We analyzed a total number of 53 patients with ARDS who fulfilled the diagnostic criteria. All patients were treated according to the low-tidal-volume strategy previously reported.27 The mean age in patients with early CRRT was 50.36 years and 70.4% (19/27) was male while the mean age in patients with late CRRT was 53.12 years and 65.4% (17/26) was male. The mean age in healthy control group was 46.79 years and 70% was male. There were no differences in age and gender between 2 groups. Sepsis was the major cause of ARDS in 2 groups (40.7% and 38.5%, respectively). Other causes of ARDS included pneumonia, polytrauma, stroke and aspiration. Among the study subjects, 11.3% (3/27) in the early CRRT group and 7.7% (2/26) in the late CRRT group had increased creatinine levels. The average initial creatinine levels were 1.02 6 0.31 mg/dL and 0.94 6 0.34 mg/dL in the early CRRT and late CRRT group, respectively, which did not show a significant difference. Clinical Outcomes of the Study Patients As showed in Table 2, there were no significant differences in PaO2/FIO2, PEEP, plateau pressure and A-a gradient in patients between early CRRT and late CRRT at baseline of day 0. However, the trends of steady improvement of these parameters were observed in both groups over the 7 days of follow-up. There were significant improvements in PaO2/FIO2 at day 7, and PEEP and A-a gradient at days 4 and 7 in early CRRT group compared with late CRRT group. We further evaluated the effects of CRRT on Cdyn. Consistent with the changes of PaO2/FIO2, we observed similar trends of steady increase of Cdyn between early and late initiation of CRRT patients among 7 days of follow-up. On day 7 of follow-up, there was significant higher Cdyn in patients with early
TABLE 1. Subjects’ characteristics and clinical features Early Late CRRT CRRT Control P Patients Age Gender (M/F) Causes of ARDS Sepsis Pneumonia Polytrauma Stroke Aspiration AKI patient number Creatinine (mg/dL)
27 50.36 6 16.97 19/8
26 53.12 6 15.05 17/9
20 46.79 6 18.92 14/6
— —
40.7% (11) 18.5% (5) 18.5% (5) 14.8% (4) 7.4% (2) 11.1% (3)
38.5% (10) 23.1% (6) 15.4% (4) 15.4% (4) 15.4% (4) 7.7% (2)
— — — — — —
0.86533 0.68249 0.95046 0.95726 0.96463 —
1.02 6 0.31
0.94 6 0.34
—
0.35180
—
ARDS, acute respiratory distress syndrome; AKI, acute kidney injury; CRRT, continuous renal replacement therapy; F, female; M, male.
Volume 349, Number 3, March 2015
Early CRRT Improves Clinical Outcomes in ARDS
TABLE 2. Comparison of ventilatory function, hemodynamics and fluid balance Day 0 Day 2
PaO2/FIO2 PEEP (cm$H2O) Plateau pressure (cm$H2O) A-a gradient (mm Hg) Cdyn CVP (mm Hg) MAP (mm Hg) CI (L$min21$m23) EVLWI (mL/kg) Fluid balance (mL)
Early CRRT
Late CRRT
Early CRRT
121.45 6 46.39 13.45 6 5.78 31.85 6 5.62
134.07 6 38.86 12.98 6 5.23 30.02 6 5.39
139.40 6 37.49 8.90 6 4.49b 24.90 6 5.01b
46.18 6 11.12
44.09 6 10.92
17.18 6 11.12 19.09 14.09 6 3.78 14.78 69.23 6 25.1 71.15 3.13 6 1.27 2.95 15.26 6 5.69 16.07 163.31 6 32.67 1058.27
6 6 6 6 6 6
12.31 4.09 23.92 1.03 5.35 169.85a
Day 7
Late CRRT 126.95 6 36.39 11.76 6 5.07a 28.63 6 6.67
Early CRRT
Late CRRT
220.19 6 50.77 5.17 6 2.36b 19.97 6 3.83b
187.73 6 49.66a 7.59 6 2.68ab 23.88 6 4.41b
30.06 6 13.13b
39.33 6 15.71a
16.73 6 11.92b
6 6 6 6 6 6
6 6 6 6 6 6
6 6 6 6 6 6
20.06 11.90 75.71 3.28 10.26 50.38
13.13 18.76 3.21b 15.01 23.9 70.63 1.16 2.76 13.99 4.75b 23.20b 595.18
10.08 32.73 3.99a 9.91 25.33 80.32 1.12 3.53 3.87a 5.05 137.96ab 177.56
11.92 2.86b 27.5 1.20 2.33b 40.31
25.66 6 12.39ab 25.37 10.57 77.93 3.12 6.77 183.67
6 6 6 6 6 6
10.27a 3.73b 24.99 1.17 2.39ab 69.53b
P , 0.05 compared with early CRRT at the same time point. P , 0.05 compared with baseline of day 0. BALF, bronchoalveolar lavage fluid; Cdyn, dynamic compliance; CI, cardiac index; CVP, central venous pressure; EVLWI, extrtravascular lung water index; MAP, mean arterial pressure; PEEP, positive end-expiratory pressure. a b
CRRT compared with patients with late CRRT (32.73 6 11.92 versus 25.37 6 10.27, respectively, P 5 0.0199). EVLWI reflects pulmonary edema and correlates with outcome. An EVLWI between 3 and 7 mL/kg is considered normal and above 10 mL/kg is associated with clinical pulmonary edema.28,29 We observed significant improvements of EVLWI in both groups after initiation of CRRT. Early CRRT patients showed lower EVLWI than late CRRT patients at days 2 and 7 of follow-up (Table 2). In addition, we showed that patients with early CRRT required shorter duration of CRRT (81.39 6 20.57 hours versus 103.68 6 27.9 hours, respectively, P 5 0.00168) and mechanical ventilation compared with those with late CRRT (10.16 6 3.52 days versus 12.97 6 4.33 days, respectively, P 5 0.0123) (Figure 1). TGF-b1 Concentrations in Serum and BALF Compared with healthy volunteers, serum TGF-b1 concentrations were markedly increased in patients with ARDS. To further determine the effects of CRRT on the level of TGF-b1, we analyzed serum and BALF TGF-b1 concentration changes. As summarized in Table 3, there were no significant differences in serum TGF-b1 concentration noted among the patients with either early or late initiation of CRRT at baseline day 0. There was a steady reduction of serum TGF-b1 over the 7 days of follow-up in early CRRT patients. Serum TGF-b1 decreased 6% (from 14.25 6 5.92 ng/mL to 13.35 6 5.24 ng/mL) on day 2 and 57% (from 13.35 6 5.24 ng/mL to 5.79 6 1.28 ng/mL)
FIGURE 1. Comparison of duration of CRRT and mechanical ventilation between early initiation of CRRT and late initiation of CRRT groups. CRRT, continuous renal replacement therapy. Copyright © 2014 by the Southern Society for Clinical Investigation.
on day 7, respectively. However, serum TGF-b1 in late CRRT patients only showed reduction after initiation of CRRT on day 2 (62%, from 17.23 6 6.56 ng/mL to 6.56 6 1.39 ng/mL). In addition, serum TGF-b1 concentrations were significantly decreased in patients with early CRRT compared with those with late CRRT on day 2 (13.35 6 5.24 ng/mL versus 17.23 6 6.56 ng/mL, P 5 0.0209) and day 7 (5.79 6 1.28 ng/mL versus 6.56 6 1.39 ng/mL, P 5 0.0408) of follow-up. In addition, we also compared TGF-b1 level changes in BALF. BALF TGF-b1 concentrations were much higher than those in serum, indicating it might be a more sensitive parameter. We observed similar trends in BALF TGF-b1 concentration change as in serum over the 7-day follow-up. There was a steady reduction of BALF TGF-b1 over the 7 days of followup in early CRRT patients. However, BALF TGF-b1 in late CRRT patients only showed reduction after initiation of CRRT after day 2. In addition, BALF TGF-b1 concentrations were significantly lower in patients with early CRRT compared with those with late CRRT on day 2 (16.83 6 6.77 ng/mL versus 21.95 6 9.16 ng/mL, P 5 0.0244) and day 7 (9.18 6 3.19 ng/mL versus 12.83 6 3.66 ng/mL, P 5 0.0003) of follow-up.
TABLE 3. Comparison of TGF-b1 concentrations (ng/mL) in serum and BALF Group Day 0 Day 2 Day 7 Early CRRT Serum (ng/mL) BALF (ng/mL) Late CRRT Serum (ng/mL) BALF (ng/mL) Control Serum (ng/mL) a b
14.25 6 5.92 13.35 6 5.24 19.09 6 7.99 16.83 6 6.77
5.79 6 1.28 9.18 6 3.19
13.67 6 4.85 17.23 6 6.56a 6.56 6 1.39a 17.89 6 6.01 21.95 6 9.16b 12.83 6 3.66b 2.25 6 1.03
—
—
P , 0.05. P , 0.01, compared with early CRRT group.
201
Han et al
Comparison of Prognostic Value As showed above, early CRRT is beneficial to improve oxygenation in patients with ARDS; we further sought to compare the survival between 2 groups. There was a trend not reaching statistical difference (P 5 0.3167) in mortality rate reduction between early CRRT patients (22.2%, 6/27) and late CRRT patients (34.6%, 9/26). Next, we focused our investigation to address the prognostic value of serum and BALF TGF-b1 levels. Figure 2 depicts the comparison of the TGF-b1 levels between survivors and nonsurvivors over the 7-day follow-up. There were no significant differences in serum TGF-b1 at days 2 and 7 of follow-up between survivors and nonsurvivors in either group. However, on day 7 of follow-up, patients who died of ARDS had higher levels of TGF-b1 in BALF than survivors in both groups (early CRRT: 12.13 6 2.75 ng/mL versus 8.05 6 2.97 ng/mL, P 5 0.00588; late CRRT: 14.67 6 3.19 ng/mL versus 10.12 6 3.35 ng/mL, P 5 0.00268).
DISCUSSION In the current study, we have showed that early initiation of CRRT improves clinical outcomes in patients with ARDS, including improvement of oxygenation and decreased ventilator duration. Furthermore, our data suggested that it maybe associated with removal of lung water and inflammatory cytokine TGF-b1 in serum and BALF. In ARDS, the fluid clearance process is defective with compromised alveolar-capillary barrier function.30 It is widely believed that edema fluid must be cleared for patients with ARDS to survive.31 CRRT allows extracorporeal treatment in critically ill patients with fluid overload. Although it has not become a standard protocol for ARDS treatment, using CRRT to support ARDS treatment has been studied and reported in several medical centers for the past decade.23–25,32–35 Table 4 summarized the studies on the use of CRRT in these ARDS patients associated with the common causes, including sepsis, trauma, postoperative, pancreatitis, pneumonia, etc. These studies showed the therapeutic benefits of CRRT in ARDS patients. In addition, several reports also suggested improvement of oxygenation and survival benefit using CRRT in the treatment of ARDS under untypical medical conditions, including severe pneumonia after renal transplantation,20 bone morrow transplantation or chemotherapy21 and allogeneic hematopoietic stem cell transplantation.22 However, Hoste et al36 found neither benefit nor harm about CRRT on oxygenation in patients with ARDS and acute kidney injury (AKI). Our results confirmed the beneficial effects on the clinical outcomes with the initiation of CRRT in patients with ARDS caused by sepsis, pneumonia, polytrauma, stroke or aspiration. The optimal timing for CRRT initiation remains undetermined because the decision to start CRRT in critically ill patients must depend on individual patient’s condition. In this study, we compared early initiation (within 12 hours after
FIGURE 2. Comparison of serum (left) and BALF (right) TGF-b1 concentration changes between survivors and nonsurvivors. BALF, bronchoalveolar lavage fluid.
202
ARDS onset) and late initiation (48 hours after ARDS onset) of CRRT and observed significant improvement of clinical outcomes in early initiation of CRRT group. Our results are consistent with a number of previous studies, which suggested that early CRRT had a favorable outcome in critically ill patients. Oh et al37 demonstrated that in septic AKI patients, early CRRT treatment was independently associated with a lower mortality rate after adjustment for demographic characteristics, the severity of disease and causative organisms. Late starting CRRT in a multicenter cohort of ICU patients with severe AKI was associated with a longer stay in hospital and higher dialysis dependence.38 The meta-analysis by Seabra et al39 suggests that early initiation of CRRT in patients with AKI was associated with a statistically significant 28% mortality risk reduction (relative risk: 0.72; 95% confidence interval [CI]: 0.64–0.82; P , 0.001). A more recent meta-analysis also supported the early CRRT therapy due to a significant improvement in 28-day mortality (odds ratio: 0.45; 95% CI: 0.28–0.72).40 Consistently, we also showed a trend of reduction in mortality, although there was no statistical significance due to limited sample size. A larger cohort of patients will be needed to fully address this issue. Mechanisms other than removing fluid have been considered in an attempt to explain the beneficial effect of CRRT on ARDS. Besides hypothermia during CRRT, which decreases oxygen consumption and removing extravascular lung fluids, some researchers believe that CRRT is beneficial in eliminating inflammatory mediators that contribute to the pathogenesis of ARDS.17,34,41 Current understanding of the pathophysiology underlying ARDS highlights large number of neutrophil accumulation in the lungs and increased expression of proinflammatory cytokines, which cause damage in the integrity of the epithelial cells and endothelial cells.2,42 TGFb1, 1 of the 3 isoforms of TGF-b, plays a critical role in tissue repair after injury in multiple organs, including the lung.12 It has been extensively evaluated during the late fibroproliferative stage of ARDS.11,12,43 Now, researcher further realized that TGF-b1 actively participated in the early exudative stage of the ARDS pathogenesis. As a proinflammatory cytokine, it induces T-cell and natural killer cell suppression and monocyte/macrophage deactivation,12 and more recently, it has been shown to direct trafficking of the epithelial sodium channel and cause pulmonary edema.44 Therefore, it is critical to target TGF-b1 during the early treatment of patients with ARDS. Here, we showed that TGF-b1 concentrations in serum and BALF were significantly elevated in ARDS patients, which is consistent with previous reports by Budinger et al7 and Fahy et al.9 They showed the presence of TGF-b1 in lung tissue from patients with ARDS within the first 24 hours of diagnosis and increased TGF-b1 levels in lung fluids from patients with acute lung injury/ARDS, respectively. We further showed that after initiation of CRRT, both serum and BALF TGF-b1 concentrations were significantly reduced, which was associated with improvement of clinical outcomes, including 2 parameters PaO2/FIO2 and Cdyn, and a shorter duration on ventilation. Our results were in line with the previous studies by Dhainaut et al12 that TGF-b1 concentrations were correlated with decreases in the PaO2/FIO2 ratio and lower levels of TGF-b1 correlated with more ventilator-free and ICU-free days.7 In addition, several groups have been studying the efficacy of cytokines removal therapy with CRRT and showed promising results.34,45 Therefore, we believe that the observed clinical outcomes improvement in this study at least partially resulted from an attenuated inflammatory response especially by removal of TGF-b1 through CRRT. Volume 349, Number 3, March 2015
Early CRRT Improves Clinical Outcomes in ARDS
TABLE 4. Summary of the studies using CRRT therapy in patients with ARDS Age of No. of patients Associations of Reference no. patients (yr) ARDS CRRT strategies 23
65
24
14
25
12
34
32
43 6 13
Main outcomes
Sepsis: 8; 37 patients under CHVHF With prolonged time of postoperative: and other routine treatment, CHVHF 14; trauma: treatments; 28 patients group had significantly 12; shock: 15; under routine treatments elevated PaO2/FiO2, decreased EVLWI and pancreatitis: only as control group PaCO2 levels; duration 10; of mechanical ventilation pneumonia: 6 and ICU stay were shortened in CHVHF group 44.4 (range, Trauma: 10; CAVH-D for a mean of 11 patients had significant 14–68) postoperative: 65.2 hr (range, 12–140 improvement of 4; hr) respiratory function: pancreatitis: 1 mean FIO2 decreased from 0.73 to 0.45; PEEP from 14.3 to 8.9 cm; peak airway pressures fell from a mean of 60 to 45 mm Hg; no significant change in CI or wedge pressure, but oxygen consumption rose from a mean of 279 to 409 mL/m 50 (range, Sepsis CHVHF Significant decreases in 22–60) MPAP, PVR and TFC after 48 hours of treatment; oxygen delivery, oxygen consumption and oxygen extraction rate were stabilized at 72 hours with amelioration of PaO2, PaO2/FiO2 and peak airway pressure; significant decreases in blood levels of cytokines (tumor necrosis factor-a, IL-6 and IL-8) 61.5 6 15.0 Sepsis resulting Continuous Significantly decreased from hemodiafiltration; using blood levels of cytokines peritonitis, a PMMA-CHDF for and successful water pneumonia or renal replacement and removal without other cytokine removal for 3 changing central venous diseases: 26/ days versus treatment of pressure in the PMMA32; other: 6/ IHD and CHF in 19 CHDF group but not in 32 patients the IHD + CHF group; significant correlations between changes in blood levels of cytokines (IL-6 and IL-8) and changes in respiratory index were demonstrated in the PMMA-CHDF group
Survival rate or cases with severe side effects 28-day cumulative survival rate: CHVHF: 86.5%; control group: 71.4%
One patient bled from a loose tubing connection; 3 patients were grossly unstable when CAVH-D was begun
—
28-day cumulative survival rate: PMMA-CHDF: 68.8%; IDH + CHF: 36.8%
CAVHD, continuous arteriovenous countercurrent hemodialysis; CHF, continuous hemofiltration; CHVHF, continuous high-volume hemofiltration; IHD, intermittent hemodialysis; MPAP, mean pulmonary arterial pressure; PMMA-CHDF, polymethylmethacrylate membrane hemofilter; PVR, pulmonary vascular resistance; TFC, thoracic fluid content.
Limitations This study has several limitations. First, instead of using the most current Berlin ARDS definition, 1994 AmericanEuropean Consensus definition for ARDS was used in our Copyright © 2014 by the Southern Society for Clinical Investigation.
study during the enrollment period. Second, we only included patients with ARDS on ICU admission and the number of patient is relatively small. Third, our study only investigated 1 particular cytokine—TGF-b1. Several other important cytokines
203
Han et al
that have been shown to participate in the pathogenesis of ARDS were not assessed. Fourth, defining “early” versus “late” in the initiation of CRRT in critically ill patients is very challenging. Whether timing of CRRT should be measured from ICU admission, from the onset of ARDS or by use of serum biomarkers remains controversial.
CONCLUSIONS
Taken together, our findings suggested that early initiation of CRRT is associated with favorable clinical outcomes in patients with ARDS, which might be due to the reduced serum and BALF TGF-b1 levels through CRRT. However, large multicenter studies are needed to make further recommendations as to the optimal use of CRRT in ARDS patient populations. ACKNOWLEDGMENTS The authors thank Dr. Ziqiang Zhu at the National Institutes of Health for the critical review of the manuscript. REFERENCES 1. Ranieri VM, Rubenfeld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012;307:2526–33. 2. Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342:1334–49. 3. Bartram U, Speer CP. The role of transforming growth factor beta in lung development and disease. Chest 2004;125:754–65. 4. Hamacher J, Lucas R, Lijnen HR, et al. Tumor necrosis factor-alpha and angiostatin are mediators of endothelial cytotoxicity in bronchoalveolar lavages of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;166:651–6. 5. Pittet JF, Griffiths MJ, Geiser T, et al. TGF-beta is a critical mediator of acute lung injury. J Clin Invest 2001;107:1537–44. 6. Jenkins RG, Su X, Su G, et al. Ligation of protease-activated receptor 1 enhances alpha(v)beta6 integrin-dependent TGF-beta activation and promotes acute lung injury. J Clin Invest 2006;116:1606–14. 7. Budinger GR, Chandel NS, Donnelly HK, et al. Active transforming growth factor-beta1 activates the procollagen I promoter in patients with acute lung injury. Intensive Care Med 2005;31:121–8. 8. Adams DH, Hathaway M, Shaw J, et al. Transforming growth factorbeta induces human T lymphocyte migration in vitro. J Immunol 1991; 147:609–12. 9. Fahy RJ, Lichtenberger F, McKeegan CB, et al. The acute respiratory distress syndrome: a role for transforming growth factor-beta 1. Am J Respir Cell Mol Biol 2003;28:499–503. 10. Beers MF, Solarin KO, Guttentag SH, et al. TGF-beta1 inhibits surfactant component expression and epithelial cell maturation in cultured human fetal lung. Am J Physiol 1998;275:L950–60. 11. de Pablo R, Monserrat J, Reyes E, et al. Sepsis-induced acute respiratory distress syndrome with fatal outcome is associated to increased serum transforming growth factor beta-1 levels. Eur J Intern Med 2012; 23:358–62. 12. Dhainaut JF, Charpentier J, Chiche JD. Transforming growth factorbeta: a mediator of cell regulation in acute respiratory distress syndrome. Crit Care Med 2003;31(suppl 4):S258–64. 13. Shenkar R, Coulson WF, Abraham E. Anti-transforming growth factor-beta monoclonal antibodies prevent lung injury in hemorrhaged mice. Am J Respir Cell Mol Biol 1994;11:351–7. 14. Giri SN, Hyde DM, Hollinger MA. Effect of antibody to transforming growth factor beta on bleomycin induced accumulation of lung collagen in mice. Thorax 1993;48:959–66.
204
15. Bellomo R, Honore PM, Matson J, et al. Extracorporeal blood treatment (EBT) methods in SIRS/Sepsis. Int J Artif Organs 2005;28:450–8. 16. Dirkes S, Hodge K. Continuous renal replacement therapy in the adult intensive care unit: history and current trends. Crit Care Nurse 2007;27: 61–6; 68–72, 74–80; quiz 81. 17. Ronco C, Bonello M, Bordoni V, et al. Extracorporeal therapies in non-renal disease: treatment of sepsis and the peak concentration hypothesis. Blood Purif 2004;22:164–74. 18. Abdo AA, Castellanos R, Rocha M, et al. Continuous venovenous hemodiafiltration in patients with multiple organ dysfunction syndrome in an intensive care unit. MEDICC Rev 2012;14:26–30. 19. Marenzi G, Lauri G, Grazi M, et al. Circulatory response to fluid overload removal by extracorporeal ultrafiltration in refractory congestive heart failure. J Am Coll Cardiol 2001;38:963–8. 20. Sun Q, Liu ZH, Chen J, et al. An aggressive systematic strategy for acute respiratory distress syndrome caused by severe pneumonia after renal transplantation. Transpl Int 2006;19:110–16. 21. DiCarlo JV, Alexander SR, Agarwal R, et al. Continuous venovenous hemofiltration may improve survival from acute respiratory distress syndrome after bone marrow transplantation or chemotherapy. J Pediatr Hematol Oncol 2003;25:801–5. 22. Elbahlawan L, West NK, Avent Y, et al. Impact of continuous renal replacement therapy on oxygenation in children with acute lung injury after allogeneic hematopoietic stem cell transplantation. Pediatr Blood Cancer 2010;55:540–5. 23. Zhang JC, Chu YF, Zeng J, et al. Effect of continuous high-volume hemofiltration in patients with severe acute respiratory distress syndrome [in Chinese]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2013; 25:145–8. 24. Garzia F, Todor R, Scalea T. Continuous arteriovenous hemofiltration countercurrent dialysis (CAVH-D) in acute respiratory failure (ARDS). J Trauma 1991;31:1277–84; discussion 1284–5. 25. Liu CW, Zheng YK, Lu J, et al. Effect of high volume hemofiltration on respiration, hemodynamics and oxygen metabolism in sepsis with acute respiratory distress syndrome [in Chinese]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 2006;18:657–60. 26. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994;149:818–24. 27. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000;342:1301–8. 28. Kuzkov VV, Kirov MY, Sovershaev MA, et al. Extravascular lung water determined with single transpulmonary thermodilution correlates with the severity of sepsis-induced acute lung injury. Crit Care Med 2006;34:1647–53. 29. Michard F. Bedside assessment of extravascular lung water by dilution methods: temptations and pitfalls. Crit Care Med 2007;35:1186–92. 30. Matthay MA, Ware LB, Zimmerman GA. The acute respiratory distress syndrome. J Clin Invest 2012;122:2731–40. 31. Sznajder JI. Alveolar edema must be cleared for the acute respiratory distress syndrome patient to survive. Am J Respir Crit Care Med 2001; 163:1293–4. 32. Grasselli G, Foti G, Patroniti N, et al. A case of ARDS associated with influenza A—H1N1 infection treated with extracorporeal respiratory support. Minerva Anestesiol 2009;75:741–5. 33. Schetz M. Non-renal indications for continuous renal replacement therapy. Kidney Int Suppl 1999;72:S88–94.
Volume 349, Number 3, March 2015
Early CRRT Improves Clinical Outcomes in ARDS
34. Matsuda K, Moriguchi T, Oda S, et al. Efficacy of continuous hemodiafiltration with a cytokine-adsorbing hemofilter in the treatment of acute respiratory distress syndrome. Contrib Nephrol 2010; 166:83–92.
40. Karvellas CJ, Farhat MR, Sajjad I, et al. A comparison of early versus late initiation of renal replacement therapy in critically ill patients with acute kidney injury: a systematic review and meta-analysis. Crit Care 2011;15:R72.
35. Davenport A. Can modification of renal replacement therapy improve the outcome of patients with systemic inflammatory response syndrome? Blood Purif 2006;24:317–18.
41. Ronco C, Tetta C, Mariano F, et al. Interpreting the mechanisms of continuous renal replacement therapy in sepsis: the peak concentration hypothesis. Artif Organs 2003;27:792–801.
36. Hoste EA, Vanholder RC, Lameire NH, et al. No early respiratory benefit with CVVHDF in patients with acute renal failure and acute lung injury. Nephrol Dial Transplant 2002;17:2153–8.
42. Meduri GU, Kohler G, Headley S, et al. Inflammatory cytokines in the BAL of patients with ARDS. Persistent elevation over time predicts poor outcome. Chest 1995;108:1303–14.
37. Oh HJ, Shin DH, Lee MJ, et al. Early initiation of continuous renal replacement therapy improves patient survival in severe progressive septic acute kidney injury. J Crit Care 2012;27:e749–718.
43. Santana A, Saxena B, Noble NA, et al. Increased expression of transforming growth factor beta isoforms (beta 1, beta 2, beta 3) in bleomycininduced pulmonary fibrosis. Am J Respir Cell Mol Biol 1995;13:34–44.
38. Bagshaw SM, Uchino S, Bellomo R, et al. Timing of renal replacement therapy and clinical outcomes in critically ill patients with severe acute kidney injury. J Crit Care 2009;24:129–40.
44. Peters DM, Vadasz I, Wujak L, et al. TGF-beta directs trafficking of the epithelial sodium channel ENaC which has implications for ion and fluid transport in acute lung injury. Proc Natl Acad Sci U S A 2013;111: E374–83.
39. Seabra VF, Balk EM, Liangos O, et al. Timing of renal replacement therapy initiation in acute renal failure: a meta-analysis. Am J Kidney Dis 2008;52:272–84.
45. Oda S, Sadahiro T, Hirayama Y, et al. Non-renal indications for continuous renal replacement therapy: current status in Japan. Contrib Nephrol 2010;166:47–53.
Copyright © 2014 by the Southern Society for Clinical Investigation.
205