- Email: [email protected]
High versus low mean arterial pressures in hepatorenal syndrome: A randomized controlled pilot trial
Journal of Critical Care 52 (2019) 186–192
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.journals.elsevier.com/journal-of-critical-care
High versus low mean arterial pressures in hepatorenal syndrome: A randomized controlled pilot trial Benadin Varajic b, Rodrigo Cavallazzi a, Jason Mann a, Stephen Furmanek c, Juan Guardiola a, Mohamed Saad a a b c
Department of Pulmonary, Critical Care, and Sleep Disorders Medicine, University of Louisville, USA Department of Internal Medicine, University of Louisville, USA Department of Infectious Disease, University of Louisville, USA
a r t i c l e
i n f o
a b s t r a c t There is controversy regarding the mean arterial pressure (MAP) goals that should be targeted in the treatment of hepatorenal syndrome (HRS.) We conducted a study to assess different MAP targets in HRS in the intensive care unit (ICU). Materials and methods: This is a prospective randomized controlled pilot trial. ICU patients had target mean arterial pressure (MAP) ≥ 85 mmHg (control arm) or 65–70 mmHg (study arm). Urine output and serum creatinine were trended and recorded. Results: A total of 18 patients were enrolled. The day four urine output in the high and low MAP group was 1194 (SD = 1249) mL/24 h and 920 (SD = 812) mL/24 h, respectively. The difference in day four – day one urine output was −689 (SD = 1684) mL/24 h and 272 (SD = 582) mL/24 h for the high and low MAP groups. The difference in serum creatinine at day four – day one was −0.54 (SD = 0.63) mg/dL and − 0.77 (SD = 1.14) mg/dL in the high and low MAP groups, respectively. Conclusion: In this study, we failed to prove non-inferiority between a low and high target MAP in patients with HRS. Trial registration: This trial was registered with and approved by the University of Louisville Internal Review Board and hospital research review committees (IRB # 14.1190). The trial was registered with ClinicalTrials. gov (ID # NCT02789150). The IRB committee roster 7/21/2014–2/26/2015 is registered with IORG (IORG # IORG0000147; OMB # 0990–0279) and is available at http://louisville.edu/research/humansubjects/about-theirb/rosters/RosterEffective20140721thru20150226.pdf © 2019 Elsevier Inc. All rights reserved.
1. Introduction Hepatorenal Syndrome (HRS) is a recognized complication of liver cirrhosis that is characterized by progressive renal injury in the setting of underlying liver disease through several mechanisms causing intense intra-renal vasoconstriction [1]. Prior to the 1996 adoption of the International Ascites Club criteria for HRS diagnosis, the incidence was believed to be 18% at one year and 39% at five years of all patients with cirrhosis [2]. More recent studies show an annual incidence of 7.1% and prevalence of 13–45% of patients with liver cirrhosis that present with AKI [3,4]. Reported mortality varies between 20 and 80% depending on treatment modality [2,5-8]. Although liver transplant appears to be the only definite treatment, current therapies include the use of vasopressors, vasopressor sparing agents and albumin [9]. Although terlipressin is a commonly used treatment worldwide, it lacks FDA approval [10]. In the United States, HRS is
E-mail address: [email protected] (B. Varajic).
https://doi.org/10.1016/j.jcrc.2019.04.006 0883-9441/© 2019 Elsevier Inc. All rights reserved.
treated with oral midodrine, subcutaneous octreotide, intravenous albumin transfusions and vasopressors [10,11]. The Acute Dialysis Quality Initiative (ADQI) group recommends target mean arterial pressures (MAP) in HRS type I of 10 mmHg above the patients' baseline [7,12]. This is also in line with a 2016 multidisciplinary perspective statement by experts, which recommends an increase in MAP by 10–15 mmHg from baseline [10]. The ADQI also recommends that treatment be continued for 4 days after which the vasoconstrictor can either be continued in partial responders (i.e. decrease in serum creatinine) or discontinued in non-responders (i.e. no change in serum creatinine, or worsening of renal function) [12]. Because baseline MAPs are often unavailable, aiming for a MAP of 10 mmHg above baseline is impractical in many cases. In our institution, we commonly target MAP ≥85 mmHg based on the expert opinion of our clinicians. In this study we aimed to test if a lower MAP target is non-inferior to a higher MAP target. The higher pressure target is based upon anecdotal evidence and to our knowledge there are no published studies that validate the use of any specific blood pressure target in patients with HRS. In sepsis, which presents as similar state of peripheral vasodilation, the
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
2016 update to the surviving sepsis campaign recommends “…an initial target mean arterial pressure (MAP) of 65mm Hg in patients with septic shock requiring vasopressors (strong recommendation, moderate quality of evidence).” [13] Other experts have recommended a MAP N60 mmHg in patients with liver cirrhosis with shock [10]. In HRS it is not clear that supra-physiologic MAP targets provide benefit. 2. Methods
187
UOP and change in serum creatinine, were similar in patients with HRS treated with different targets of MAPs. This study was approved by the University Louisville Institutional Review Board– Human Subjects Protection Program Office and the hospital research review committees. All participants or their proxy provided informed consent. There are no disclosures, sponsors or funding. This research was supported by the department of medicine through the divisions of infectious diseases and pulmonary, critical care, and sleep medicine at the University of Louisville.
2.1. Study design and oversight 2.2. Participants This is a prospective, non-blinded, randomized, two-arm treatment, non-inferiority, pilot study. We aimed to test if MAP target of 65–70 mmHg is non-inferior to MAP ≥85 mmHg. We proposed that there would be no difference in clinical outcomes between the two MAP groups. We chose the “high” MAP target of ≥85 mmHg as this characterizes a practical and realistic pressure and represents a pressure 10 mmHg above baseline in most patients with cirrhosis [14] This pressure target is based upon anecdotal evidence and to our knowledge there are no published studies known to us that validate the use of any specific blood pressure target. Patients underwent simple randomization to receive either a target MAP ≥85 mmHg or 65–70 mmHg. Sealed envelopes were used for randomization. The study was designed to determine if renal outcomes, measured by day four UOP, change in
The study population consisted of patients that met the clinical diagnosis of HRS per the International Ascites Club at the time of study enrollment. The study was conducted in the intensive care unit (ICU) of University of Louisville Hospital and Jewish Hospital, both in Louisville, KY, USA. 2.2.1. Inclusion criteria 1. Admission to ICU 2. Age ≥ 18 years old 3. Fulfilling of the International Ascites Club diagnostic criteria for HRS 4. Lack of exclusion criteria
Fig. 1. Patient enrollment flowchart.
188
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
2.2.2. Exclusion criteria 1. Pre-existing continuous renal replacement therapy or those initiated on dialysis during their hospital stay prior to enrollment in the study. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Artificial liver support therapies Ongoing gastrointestinal bleeding Active medical disease requiring surgical intervention Pre-existing or in-hospital placement of transjugular intrahepatic portosystemic shunt Long standing hypertension (on active treatment) Improvement in renal function after central blood volume expansion Contraindications to norepinephrine (active myocardial event, ventricular arrhythmia, obstructive physiology, limb ischemia) Pregnancy Treating physicians refusal to enroll patients
Table 1 Patient characteristics.
n Age (median [IQR]) Meld Score (median [IQR]) Sex: Male (%) Race: White (%) HRS Type 1 (%) HRS Type 2 (%) Cause of Liver Cirrhosis Hepatitis B (%) Hepatitis C (%) Alcoholism (%) NASH (%) Idiopathic (%)
Study group (Low MAP)
Control group (High MAP)
8 53 [51, 60] 37 [34, 39] 3 (38) 8 (100) 5 (63) 3 (38)
9 53 [50, 62] 30 [28, 39] 8 (89) 8 (89) 5 (56) 4 (44)
0 (0) 2 (25) 4 (50) 3 (38) 1 (13)
2 (22) 3 (33) 5 (56) 1 (11) 0 (0)
IQR: interquartile range; MELD: Model of End-stage Liver Disease; NASH: Non-Alcoholic Steatohepatitis.
2.3. Protocol and interventions According to the protocol, all patients should receive oral midodrine, octreotide, and intravenous albumin solution. Urinary bladder catheter was inserted and urine output recorded every 24 h. The ICU team assessed volume status clinically and euvolemia was targeted in all patients. Serum creatinine was monitored daily per ICU protocols. Arterial and central venous catheters were placed for accurate monitoring of the mean arterial pressure and delivering vasopressor support. Vasopressor choice was deferred to the treating ICU team, but norepinephrine infusion was the vasopressor of choice. Vasopressor titration was performed per hospital ICU protocol. The use of flow monitoring was not protocoled per the study, but was allowed to be used if the treating team felt it necessary. In the high MAP group, vasopressor infusion was titrated every 30 min with the goal of keeping the MAP above ≥85 mmHg. This singular target MAP is protocol for patients with HRS in our hospitals. In the low MAP group, vasopressor infusion was titrated every 30 min with the goal of keeping the MAP 65–70 mmHg. If in the low MAP group the MAP was spontaneously above 70 mmHg, no intervention was performed to lower the MAP. If patients in either group required more than one vasopressor, additional vasopressor agents (phenylephrine and/or vasopressin) were added with the same goal MAPs per the ICU team. MAP data was obtained from arterial pressure tracing by measuring the area under the curve and dividing this area by the time interval. This calculation is performed automatically by the bedside monitor's proprietary software. 2.4. Randomization All enrolled patients underwent simple randomization into either the MAP 65–70 mmHg or the ≥85 mmHg group through a sealed envelope system. 2.5. Endpoints The primary endpoints were the following: 1. Urine output at day four. 2. Change in urine output (day 4 - day 1). 3. Change in serum creatinine (day 4 - day 1). Secondary endpoints were the following: 1. Death 2. Cardiac events (i.e. arrhythmias or infarction) 3. Vascular events (i.e. limb or intestinal ischemia). Importantly, if a patient required renal replacement therapy or died before day 4, the urine output was considered 0 mL/24 h and the most recent creatinine was carried over.
2.6. Sample size estimation Sample size was based on a prior HRS clinical trial that had similar endpoints [15]. In this trial patients with HRS who received terlipressin had a urine output of 960 (SD = 139) mL/24 h. With a non-inferiority limit of 20% in 24 h urine output at 96 h (day 4) with 90% power using a 0.05-level two-sided test we required at least 9 patients per group, for a total of 18 patients. 2.7. Statistical analysis Descriptive statistics were performed. Continuous variables are presented as median and interquartile range (IQR). Categorical variables are presented as frequency and percent. Mann-Whitney U tests were performed to compare continuous patient characteristics among patient groups. Fisher's exact test was used to compare categorical patient characteristics among patient groups. MAP was recorded every hour and then averaged over the time the patient spent enrolled in the study. Primary endpoints were compared using t-tests with non-inferiority margins of 20%. Secondary endpoints were compared using Fisher's exact test. Analysis was performed with statistical software R Version 3.3.2 [16]. 3. Results 3.1. Characteristics of patients Patients were recruited between Feb 2015 and March 2017. A total of 47 patients were screened that met inclusion criteria. Of those, 10 patients (21%) met ≥1 exclusion criteria. We were unable to obtain
Table 2 Percent of blood pressure at goal in high and low MAP groups. Study group (Low MAP)
Control group (High MAP)
Median MAP
At Goal (%)
Median MAP
At Goal (%)
86 76 74 68 68 67 67 66
7 9 27 38 28 28 44 31
71a
27a,b
103 86 86 86 85 83 81 80 77 84a
86 69 56 56 52 40 31 35 13 48a,b
a b
Median MAP for all members of group across all time points. Total percent at goal for group across all time points.
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
189
Fig. 3. Median daily dose of norepinephrine between groups. Fig. 2. Median MAP over time with interquartile range.
informed consent from 3 patients (6%) and 16 patients (34%) declined enrollment. The remaining 18 patients (38%) were enrolled and one withdrew from the study after the randomization phase, leaving nine in the high MAP group and eight in the low MAP group (see Fig. 1). Overall, the median age was 53 (IQR: 50–60) years old, and the median Model for End-Stage Liver Disease score was 37 (IQR: 34–39) in the low MAP group and 30 (IQR: 28–39) in the high MAP group. No patients received beta blockers for the treatment of portal hypertension throughout the study period. Table 1 shows baseline patient characteristics. The median MAP per group and percentage of time at goal is shown in Table 2. The high MAP group had a higher percentage of time of MAP at goal as compared with the low MAP group (48% vs. 27%; p b 0.001). Median MAP over time between groups is shown in Fig. 2. There was no difference between the two groups in the percentage of patients who received norepinephrine, vasopressin, or phenylephrine (see Table 3). Fig. 3 shows daily norepinephrine dose by group. All patients received albumin per the study protocol. The overall fluid balance by group is shown in Fig. 4. The exact dose of every type of fluid for each patient is not known. 3.2. Primary outcomes 3.2.1. Urine output at day four The mean urine output at day four in the high MAP group was 1194 (SD = 1249; 95% CI: -1255 - 3642) mL/24 h. The mean urine output at day four for the low MAP group was 920 (SD = 812) mL/24 h. The difference is 274 mL, which is outside the preset 20% margin of noninferiority (239 mL). Based on these results, we failed to show that urine output in the low MAP group was non-inferior (p = 0.513) (see Fig. 5).
low MAP group was non-inferior because the confidence interval is beyond the non-inferiority region (p = 0.198). 3.2.3. Change in serum creatinine (day 4 – day 1) The mean change in serum creatinine of the high MAP group was −0.54 (SD = 0.63) mg/dL. The mean change the low MAP group serum creatinine was −0.77 (SD = 1.14) mg/dL. As can be seen in Fig. 7, we failed to show that urine output in the low MAP was noninferior because the confidence interval is beyond the non-inferiority region (p = 0.553). 3.2.3.1. Secondary outcomes. Death: During the study period, two (22%) patients died in the high MAP group and one (13%) patient died in the low MAP group (p N 0.999). However, the in-hospital mortality was six (67%) and six (75%) in the high MAP group and low MAP groups respectively (p N 0.999). Cardiac events: No arrhythmias were found the low MAP group versus three (33%) patients in the high MAP group. This difference in arrhythmias was not found to be statistically significant (p = 0.206). Fatal arrhythmias: No fatal arrhythmias occurred in any patients during this trial. Myocardial infarction: No fatal myocardial infarctions were seen in any patients during this trial. Vascular events: one patient had an episode of intestinal ischemia in the high MAP group and no patients had any vascular events in the low MAP group during the study period.
3.2.2. Change in urine output (day 4 – day 1) The mean change in urine output in the high MAP group was −689 (SD = 1684) mL/24 h. The mean change in urine output of the low MAP group was 272 (SD = 582) mL/24 h. The difference is −961 mL/24 h. As can be seen in Fig. 6, we failed to show that change in urine output in the Table 3 Number and percentage of patients that received vasopressors according to group.
Norepinephrine (%) Vasopressin (%) Phenylephrine (%)
High MAP
Low MAP
p
7 (78) 2 (22) 1 (11)
7 (88) 2 (25) 1 (13)
N0.999 N0.999 N0.999
Fig. 4. Median and interquartile range of daily fluid balance according to group.
190
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
Non inferiority region 95% Confidence Interval Mean: High MAP
Mean: Low MAP
500
0
500
1000
1500
2000
2500
Urine output Day 4 Fig. 5. Mean urine output at day 4 for the high and low mean arterial pressure groups with non-inferiority region and 95% confidence interval for the low mean arterial pressure group.
4. Discussion The main finding of our study is that all three primary endpoint results do not allow us to claim non-inferiority of the low MAP group compared to high MAP group. With regards to total urine output at day four, change in urine output and change in serum creatinine, all confidence intervals obtained were outside the 20% margin of non-inferiority. The statistical analysis of this study shows the study power to be 13.99%, indicating that our study is significantly underpowered and unable to prove non-inferiority. In order to show non-inferiority in a similarly constructed study, the results of this pilot trial suggest that 474 subjects would be required in each arm. To our knowledge, there are no clinical trials in HRS with this number of subjects. The pathophysiologic state of HRS is one of renal hypoperfusion and thus there is a rationale in attempting reversal of HRS through augmentation of blood pressure to a supraphysiologic level. Although the same rationale can be used for any state of hypotension causing a kidney injury, studies in sepsis have failed to demonstrate a benefit to maintaining high MAP targets [17]. Patients with end-stage liver disease develop cardio-circulatory dysfunction characterized by hyperdynamic syndrome and cirrhotic cardiomyopathy [18]. This may be responsible for some of the poor vasopressor response that we witnessed.
Vasopressor sparing agents have shown to be efficacious in the treatment of HRS. Commonly used agents include terlipressin and midodrine. Terlipressin is perhaps the most widely studied and used agent in the treatment of HRS, internationally. Its use has been shown to be superior to placebo in reversal of HRS [8,15,19,20]. However, terlipressin does not have approval from the Food and Drug Administration in the United States(US). In the US, HRS is commonly treated with midodrine, octreotide, albumin and vasopressors [10,21]. Several studies have compared norepinephrine with terlipressin for the treatment of HRS [20,22,23]. In a randomized controlled trial of 46 patients with HRS type II, norepinephrine had similar efficacy to terlipressin in reversal of HRS. However, this study assessed specifically HRS type II, which is less severe and has more favorable outcomes compared to HRS type I [22]. Nanda et al. performed a systemic review and meta-analysis of pharmacologic therapies in HRS and found that norepinephrine is an effective alternative to terlipressin with fewer adverse effects [21]. Mattos et al. performed a systemic review and meta-analysis of 4 studies comparing norepinephrine with terlipressin in HRS. There was no significant difference in 30-day survival between the 2 groups [23]. Duvoux et al. demonstrated that most patients with HRS type I had a reversal of their disease when treated with norepinephrine combined with albumin and furosemide [7]. The study, however, was
Non inferiority region 95% Confidence Interval Mean Change: High MAP
Mean Change: Low MAP
1000
500
0
500
Change in urine output (Day 4
1000
1500
Day 1)
Fig. 6. Mean change in urine output (day 4 – day 1) for the high and low mean arterial pressure groups with non-inferiority region and 95% confidence interval for the low mean arterial pressure group.
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
191
Non−inferiority region 95% Confidence Interval Mean Change: High MAP
Mean Change: Low MAP
−3
−2
−1
0
1
2
Change in creatinine (Day 4 − Day 1) Fig. 7. Mean change in serum creatinine (day 4 – day1) for the high and low mean arterial pressure groups with non-inferiority region and 95% confidence interval for the low mean arterial pressure group.
limited by a small sample size (only 12 patients) and lack of a comparator arm. In a systematic review, Gluud et al. found that the use of a vasoconstrictor reduced mortality by 18% in HRS. Despite the inclusion of 10 randomized trials, the review only included 376 patients and the confidence interval for the outcome was wide [24]. Our study has a number of limitations. Although we calculated a sample size a priori, our study was underpowered to demonstrate non-inferiority. Our study only included 2 hospitals, limiting its external generalizability. The percentage of time at MAP goal was significantly different per group. The high MAP group spent more time at target, but there was wide variability of MAP target achievement even within a group. As an example, in the high MAP group, time at target was as low as 31% and as high as 86%. In the low MAP group, these figures ranged from 7% to 44%. This demonstrates the difficulty of maintaining MAP at a narrow range (65–70 mmHg). This may be, in part, due to us not artificially lowering the MAP to target it MAP in the low MAP group were spontaneously above goal. The fluid balance differed between the two groups on day one (see Fig. 4). The use of cardiac output monitoring was not protocolized in this study. Although cardiac output monitoring was used to aid with volume assessment in some patients, it was not used in all patients. We do not have access to how many patients had cardiac output monitoring. Volume status assessment was made clinically by the treating ICU team. One patient that had a history of hypertension that was actively undergoing treatment for hypertension at the time of study recruitment ended up being inadvertently enrolled. Additionally, although the study was randomized, it was not blinded. For an ideal trial, MAP data monitoring is paramount. We checked MAP hourly, and there was considerable difficulty in keeping the MAP of each patient in the goal range for each group—only 27% of the time patients in the low MAP group, and 48% of the time in the high MAP group. From an analysis standpoint, even if we were to have had shown non-inferiority, with such difficulty keeping patients in the MAP goal, it would be difficult to justify such a conclusion. Many patients randomized to the low MAP group had spontaneously higher MAP than goal; it may be unethical to artificially lower their MAP. One way an ideal study could address this would be to expand the range of the low MAP goal. In addition, our clinical endpoints of interest (urine output and serum creatinine) were considerably more varied in our patients than literature had suggested. This may be due to the fact that we used the patients' worst outcome for these endpoints. Additionally, simple randomization was used for this trial. An adaptive trial design, such as minimization, may allow better balance between potentially confounding factors between groups.
5. Conclusion Hepatorenal syndrome is a deadly condition that carries a high mortality, despite aggressive medical intervention. In this study, we assessed the efficacy of maintaining high target MAPs in patients with HRS in the ICU. Our findings do not allow concluding that lower MAP targets are non-inferior to higher MAP targets because this study is underpowered. Nonetheless, this study can be used as a basis for the planning of well-powered, multicenter, randomized clinical trials assessing clinically relevant outcomes in HRS. Acknowledgements Benadin Varajic, MD has contributed to subject recruitment, data collection, data processing, manuscript authorship, manuscript revision, and journal submission. Rodrigo Cavallazzi, MD has contributed to data processing, manuscript authorship, manuscript revision and journal submission. Jason C Mann, DO has contributed to study design, subject recruitment, data collection. Stephen Furmanek, PhD has contributed to data processing, manuscript authorship, and manuscript revision. Juan Guardiola, MD has contributed to study design and manuscript authorship. Mohamed Saad MD, is the primary investigator of this study; he has contributed to every step of the study from inception to manuscript submission; including: study design, research protocols, subject recruitment, and manuscript authorship and submission. Drs. Varajic and Saad take responsibility for the integrity of this study. Funding was provided by the University of Louisville School of Medicine Departments of Internal Medicine: Division of Medical Education, Division of Pulmonary, Critical Care and Sleep Disorders Medicine, and Division of Infectious Disease Medicine. Conflict of interest The authors of this study do not have any conflicts of interest to disclose. References [1] Wadei HM, Mai ML, Ahsan N, Gonwa TA. Hepatorenal syndrome: pathophysiology and management. Clin J Am Soc Nephrol 2006;1:1066–79.
192
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
[2] Gines A, Escorsell A, Gines P, et al. Incidence, predictive factors, and prognosis of the hepatorenal syndrome in cirrhosis with ascites. Gastroenterology 1993;105:229–36. https://doi.org/10.1016/0016-5085(93)90031-7. [3] Salerno F, Cazzaniga M, Merli M, et al. Diagnosis, treatment and survival of patients with hepatorenal syndrome: a survey on daily medical practice. J Hepatol 2011;55: 1241–8. https://doi.org/10.1016/j.jhep.2011.03.012. [4] Martin-Llahi M, Guevara M, Torre A, et al. Prognostic importance of the cause of renal failure in patients with cirrhosis. Gastroenterology 2011;140:488–96 [e4] https://doi.org/10.1053/j.gastro.2010.07.043. [5] Guevara M, Gines P, Bandi JC, et al. Transjugular intrahepatic portosystemic shunt in hepatorenal syndrome: effects on renal function and vasoactive systems. Hepatology (Baltimore, Md) 1998;28:416–22. https://doi.org/10.1002/hep. 510280219. [6] Angeli P, Volpin R, Gerunda G, et al. Reversal of type 1 hepatorenal syndrome with the administration of midodrine and octreotide. Hepatology (Baltimore, Md) 1999;29:1690–7. https://doi.org/10.1002/hep.510290629. [7] Duvoux C, Zanditenas D, Hezode C, et al. Effects of noradrenalin and albumin in patients with type I hepatorenal syndrome: a pilot study. Hepatology (Baltimore, Md) 2002;36:374–80. https://doi.org/10.1053/jhep.2002.34343. [8] Moreau R, Durand F, Poynard T, et al. Terlipressin in patients with cirrhosis and type 1 hepatorenal syndrome: a retrospective multicenter study. Gastroenterology 2002; 122:923–30. https://doi.org/10.1053/gast.2002.32364. [9] Runyon BA. Introduction to the revised American Association for the Study of Liver Diseases practice guideline management of adult patients with ascites due to cirrhosis 2012. Hepatology (Baltimore, Md) 2013;57:1651–3. https://doi.org/10.1002/hep. 26359. [10] Nadim MK, Durand F, Kellum JA, et al. Management of the critically ill patient with cirrhosis: a multidisciplinary perspective. J Hepatol 2016;64:717–35. https://doi. org/10.1016/j.jhep.2015.10.019. [11] de Mattos AZ, de Mattos AA, Mendez-Sanchez N. Hepatorenal syndrome: current concepts related to diagnosis and management. Ann Hepatol 2016;15: 474–81. [12] Nadim MK, Kellum JA, Davenport A, et al. Hepatorenal syndrome: the 8th international consensus conference of the Acute dialysis Quality Initiative (ADQI) Group. Crit Care 2012;16:R23. https://doi.org/10.1186/cc11188.
[13] Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Crit Care Med 2017; 45:486–552. https://doi.org/10.1097/CCM.0000000000002255. [14] Wagener G, Kovalevskaya G, Minhaz M, Mattis F, Emond JC, Landry DW. Vasopressin deficiency and vasodilatory state in end-stage liver disease. J Cardiothorac Vasc Anesth 2011;25:665–70. https://doi.org/10.1053/j.jvca.2010.09.018. [15] Solanki P, Chawla A, Garg R, Gupta R, Jain M, Sarin SK. Beneficial effects of terlipressin in hepatorenal syndrome: a prospective, randomized placebocontrolled clinical trial. J Gastroenterol Hepatol 2003;18:152–6. https://doi.org/10. 1046/j.1440-1746.2003.02934.x. [16] Team RC. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2016. [17] Asfar P, Meziani F, Hamel JF, et al. High versus low blood-pressure target in patients with septic shock. N Engl J Med 2014;370:1583–93. https://doi.org/10.1056/ NEJMoa1312173. [18] Fede G, Privitera G, Tomaselli T, Spadaro L, Purrello F. Cardiovascular dysfunction in patients with liver cirrhosis. Ann Gastroenterol 2015;28:31–40. [19] Rajekar H, Chawla Y. Terlipressin in hepatorenal syndrome: evidence for present indications. J Gastroenterol Hepatol 2011;26(Suppl. 1):109–14. [20] Wang H, Liu A, Bo W, Feng X, Hu Y. Terlipressin in the treatment of hepatorenal syndrome: a systematic review and meta-analysis. Medicine (Baltimore) 2018;97: e0431. https://doi.org/10.1097/MD.0000000000010431. [21] Nanda A, Reddy R, Safraz H, Habeeb S, Singal AK. Pharmacological therapies for hepatorenal syndrome: a systematic review and meta-analysis. J Clin Gastroenterol 2018 Apr;52(4):360–7. https://doi.org/10.1097/MCG.0000000000000913. [22] Ghosh S, Choudhary NS, Sharma AK, et al. Noradrenaline vs terlipressin in the treatment of type 2 hepatorenal syndrome: a randomized pilot study. Liver Int 2013;33: 1187–93. https://doi.org/10.1111/liv.12179. [23] Mattos AZ, Mattos AA, Ribeiro RA. Terlipressin versus noradrenaline in the treatment of hepatorenal syndrome: systematic review with meta-analysis and full economic evaluation. Eur J Gastroenterol Hepatol 2016;28:345–51. https://doi.org/10.1097/ MEG.0000000000000537. [24] Gluud LL, Christensen K, Christensen E, Krag A. Systematic review of randomized trials on vasoconstrictor drugs for hepatorenal syndrome. Hepatology (Baltimore, Md) 2010;51:576–84. https://doi.org/10.1002/hep.23286.
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.journals.elsevier.com/journal-of-critical-care
High versus low mean arterial pressures in hepatorenal syndrome: A randomized controlled pilot trial Benadin Varajic b, Rodrigo Cavallazzi a, Jason Mann a, Stephen Furmanek c, Juan Guardiola a, Mohamed Saad a a b c
Department of Pulmonary, Critical Care, and Sleep Disorders Medicine, University of Louisville, USA Department of Internal Medicine, University of Louisville, USA Department of Infectious Disease, University of Louisville, USA
a r t i c l e
i n f o
a b s t r a c t There is controversy regarding the mean arterial pressure (MAP) goals that should be targeted in the treatment of hepatorenal syndrome (HRS.) We conducted a study to assess different MAP targets in HRS in the intensive care unit (ICU). Materials and methods: This is a prospective randomized controlled pilot trial. ICU patients had target mean arterial pressure (MAP) ≥ 85 mmHg (control arm) or 65–70 mmHg (study arm). Urine output and serum creatinine were trended and recorded. Results: A total of 18 patients were enrolled. The day four urine output in the high and low MAP group was 1194 (SD = 1249) mL/24 h and 920 (SD = 812) mL/24 h, respectively. The difference in day four – day one urine output was −689 (SD = 1684) mL/24 h and 272 (SD = 582) mL/24 h for the high and low MAP groups. The difference in serum creatinine at day four – day one was −0.54 (SD = 0.63) mg/dL and − 0.77 (SD = 1.14) mg/dL in the high and low MAP groups, respectively. Conclusion: In this study, we failed to prove non-inferiority between a low and high target MAP in patients with HRS. Trial registration: This trial was registered with and approved by the University of Louisville Internal Review Board and hospital research review committees (IRB # 14.1190). The trial was registered with ClinicalTrials. gov (ID # NCT02789150). The IRB committee roster 7/21/2014–2/26/2015 is registered with IORG (IORG # IORG0000147; OMB # 0990–0279) and is available at http://louisville.edu/research/humansubjects/about-theirb/rosters/RosterEffective20140721thru20150226.pdf © 2019 Elsevier Inc. All rights reserved.
1. Introduction Hepatorenal Syndrome (HRS) is a recognized complication of liver cirrhosis that is characterized by progressive renal injury in the setting of underlying liver disease through several mechanisms causing intense intra-renal vasoconstriction [1]. Prior to the 1996 adoption of the International Ascites Club criteria for HRS diagnosis, the incidence was believed to be 18% at one year and 39% at five years of all patients with cirrhosis [2]. More recent studies show an annual incidence of 7.1% and prevalence of 13–45% of patients with liver cirrhosis that present with AKI [3,4]. Reported mortality varies between 20 and 80% depending on treatment modality [2,5-8]. Although liver transplant appears to be the only definite treatment, current therapies include the use of vasopressors, vasopressor sparing agents and albumin [9]. Although terlipressin is a commonly used treatment worldwide, it lacks FDA approval [10]. In the United States, HRS is
E-mail address: [email protected] (B. Varajic).
https://doi.org/10.1016/j.jcrc.2019.04.006 0883-9441/© 2019 Elsevier Inc. All rights reserved.
treated with oral midodrine, subcutaneous octreotide, intravenous albumin transfusions and vasopressors [10,11]. The Acute Dialysis Quality Initiative (ADQI) group recommends target mean arterial pressures (MAP) in HRS type I of 10 mmHg above the patients' baseline [7,12]. This is also in line with a 2016 multidisciplinary perspective statement by experts, which recommends an increase in MAP by 10–15 mmHg from baseline [10]. The ADQI also recommends that treatment be continued for 4 days after which the vasoconstrictor can either be continued in partial responders (i.e. decrease in serum creatinine) or discontinued in non-responders (i.e. no change in serum creatinine, or worsening of renal function) [12]. Because baseline MAPs are often unavailable, aiming for a MAP of 10 mmHg above baseline is impractical in many cases. In our institution, we commonly target MAP ≥85 mmHg based on the expert opinion of our clinicians. In this study we aimed to test if a lower MAP target is non-inferior to a higher MAP target. The higher pressure target is based upon anecdotal evidence and to our knowledge there are no published studies that validate the use of any specific blood pressure target in patients with HRS. In sepsis, which presents as similar state of peripheral vasodilation, the
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
2016 update to the surviving sepsis campaign recommends “…an initial target mean arterial pressure (MAP) of 65mm Hg in patients with septic shock requiring vasopressors (strong recommendation, moderate quality of evidence).” [13] Other experts have recommended a MAP N60 mmHg in patients with liver cirrhosis with shock [10]. In HRS it is not clear that supra-physiologic MAP targets provide benefit. 2. Methods
187
UOP and change in serum creatinine, were similar in patients with HRS treated with different targets of MAPs. This study was approved by the University Louisville Institutional Review Board– Human Subjects Protection Program Office and the hospital research review committees. All participants or their proxy provided informed consent. There are no disclosures, sponsors or funding. This research was supported by the department of medicine through the divisions of infectious diseases and pulmonary, critical care, and sleep medicine at the University of Louisville.
2.1. Study design and oversight 2.2. Participants This is a prospective, non-blinded, randomized, two-arm treatment, non-inferiority, pilot study. We aimed to test if MAP target of 65–70 mmHg is non-inferior to MAP ≥85 mmHg. We proposed that there would be no difference in clinical outcomes between the two MAP groups. We chose the “high” MAP target of ≥85 mmHg as this characterizes a practical and realistic pressure and represents a pressure 10 mmHg above baseline in most patients with cirrhosis [14] This pressure target is based upon anecdotal evidence and to our knowledge there are no published studies known to us that validate the use of any specific blood pressure target. Patients underwent simple randomization to receive either a target MAP ≥85 mmHg or 65–70 mmHg. Sealed envelopes were used for randomization. The study was designed to determine if renal outcomes, measured by day four UOP, change in
The study population consisted of patients that met the clinical diagnosis of HRS per the International Ascites Club at the time of study enrollment. The study was conducted in the intensive care unit (ICU) of University of Louisville Hospital and Jewish Hospital, both in Louisville, KY, USA. 2.2.1. Inclusion criteria 1. Admission to ICU 2. Age ≥ 18 years old 3. Fulfilling of the International Ascites Club diagnostic criteria for HRS 4. Lack of exclusion criteria
Fig. 1. Patient enrollment flowchart.
188
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
2.2.2. Exclusion criteria 1. Pre-existing continuous renal replacement therapy or those initiated on dialysis during their hospital stay prior to enrollment in the study. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Artificial liver support therapies Ongoing gastrointestinal bleeding Active medical disease requiring surgical intervention Pre-existing or in-hospital placement of transjugular intrahepatic portosystemic shunt Long standing hypertension (on active treatment) Improvement in renal function after central blood volume expansion Contraindications to norepinephrine (active myocardial event, ventricular arrhythmia, obstructive physiology, limb ischemia) Pregnancy Treating physicians refusal to enroll patients
Table 1 Patient characteristics.
n Age (median [IQR]) Meld Score (median [IQR]) Sex: Male (%) Race: White (%) HRS Type 1 (%) HRS Type 2 (%) Cause of Liver Cirrhosis Hepatitis B (%) Hepatitis C (%) Alcoholism (%) NASH (%) Idiopathic (%)
Study group (Low MAP)
Control group (High MAP)
8 53 [51, 60] 37 [34, 39] 3 (38) 8 (100) 5 (63) 3 (38)
9 53 [50, 62] 30 [28, 39] 8 (89) 8 (89) 5 (56) 4 (44)
0 (0) 2 (25) 4 (50) 3 (38) 1 (13)
2 (22) 3 (33) 5 (56) 1 (11) 0 (0)
IQR: interquartile range; MELD: Model of End-stage Liver Disease; NASH: Non-Alcoholic Steatohepatitis.
2.3. Protocol and interventions According to the protocol, all patients should receive oral midodrine, octreotide, and intravenous albumin solution. Urinary bladder catheter was inserted and urine output recorded every 24 h. The ICU team assessed volume status clinically and euvolemia was targeted in all patients. Serum creatinine was monitored daily per ICU protocols. Arterial and central venous catheters were placed for accurate monitoring of the mean arterial pressure and delivering vasopressor support. Vasopressor choice was deferred to the treating ICU team, but norepinephrine infusion was the vasopressor of choice. Vasopressor titration was performed per hospital ICU protocol. The use of flow monitoring was not protocoled per the study, but was allowed to be used if the treating team felt it necessary. In the high MAP group, vasopressor infusion was titrated every 30 min with the goal of keeping the MAP above ≥85 mmHg. This singular target MAP is protocol for patients with HRS in our hospitals. In the low MAP group, vasopressor infusion was titrated every 30 min with the goal of keeping the MAP 65–70 mmHg. If in the low MAP group the MAP was spontaneously above 70 mmHg, no intervention was performed to lower the MAP. If patients in either group required more than one vasopressor, additional vasopressor agents (phenylephrine and/or vasopressin) were added with the same goal MAPs per the ICU team. MAP data was obtained from arterial pressure tracing by measuring the area under the curve and dividing this area by the time interval. This calculation is performed automatically by the bedside monitor's proprietary software. 2.4. Randomization All enrolled patients underwent simple randomization into either the MAP 65–70 mmHg or the ≥85 mmHg group through a sealed envelope system. 2.5. Endpoints The primary endpoints were the following: 1. Urine output at day four. 2. Change in urine output (day 4 - day 1). 3. Change in serum creatinine (day 4 - day 1). Secondary endpoints were the following: 1. Death 2. Cardiac events (i.e. arrhythmias or infarction) 3. Vascular events (i.e. limb or intestinal ischemia). Importantly, if a patient required renal replacement therapy or died before day 4, the urine output was considered 0 mL/24 h and the most recent creatinine was carried over.
2.6. Sample size estimation Sample size was based on a prior HRS clinical trial that had similar endpoints [15]. In this trial patients with HRS who received terlipressin had a urine output of 960 (SD = 139) mL/24 h. With a non-inferiority limit of 20% in 24 h urine output at 96 h (day 4) with 90% power using a 0.05-level two-sided test we required at least 9 patients per group, for a total of 18 patients. 2.7. Statistical analysis Descriptive statistics were performed. Continuous variables are presented as median and interquartile range (IQR). Categorical variables are presented as frequency and percent. Mann-Whitney U tests were performed to compare continuous patient characteristics among patient groups. Fisher's exact test was used to compare categorical patient characteristics among patient groups. MAP was recorded every hour and then averaged over the time the patient spent enrolled in the study. Primary endpoints were compared using t-tests with non-inferiority margins of 20%. Secondary endpoints were compared using Fisher's exact test. Analysis was performed with statistical software R Version 3.3.2 [16]. 3. Results 3.1. Characteristics of patients Patients were recruited between Feb 2015 and March 2017. A total of 47 patients were screened that met inclusion criteria. Of those, 10 patients (21%) met ≥1 exclusion criteria. We were unable to obtain
Table 2 Percent of blood pressure at goal in high and low MAP groups. Study group (Low MAP)
Control group (High MAP)
Median MAP
At Goal (%)
Median MAP
At Goal (%)
86 76 74 68 68 67 67 66
7 9 27 38 28 28 44 31
71a
27a,b
103 86 86 86 85 83 81 80 77 84a
86 69 56 56 52 40 31 35 13 48a,b
a b
Median MAP for all members of group across all time points. Total percent at goal for group across all time points.
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
189
Fig. 3. Median daily dose of norepinephrine between groups. Fig. 2. Median MAP over time with interquartile range.
informed consent from 3 patients (6%) and 16 patients (34%) declined enrollment. The remaining 18 patients (38%) were enrolled and one withdrew from the study after the randomization phase, leaving nine in the high MAP group and eight in the low MAP group (see Fig. 1). Overall, the median age was 53 (IQR: 50–60) years old, and the median Model for End-Stage Liver Disease score was 37 (IQR: 34–39) in the low MAP group and 30 (IQR: 28–39) in the high MAP group. No patients received beta blockers for the treatment of portal hypertension throughout the study period. Table 1 shows baseline patient characteristics. The median MAP per group and percentage of time at goal is shown in Table 2. The high MAP group had a higher percentage of time of MAP at goal as compared with the low MAP group (48% vs. 27%; p b 0.001). Median MAP over time between groups is shown in Fig. 2. There was no difference between the two groups in the percentage of patients who received norepinephrine, vasopressin, or phenylephrine (see Table 3). Fig. 3 shows daily norepinephrine dose by group. All patients received albumin per the study protocol. The overall fluid balance by group is shown in Fig. 4. The exact dose of every type of fluid for each patient is not known. 3.2. Primary outcomes 3.2.1. Urine output at day four The mean urine output at day four in the high MAP group was 1194 (SD = 1249; 95% CI: -1255 - 3642) mL/24 h. The mean urine output at day four for the low MAP group was 920 (SD = 812) mL/24 h. The difference is 274 mL, which is outside the preset 20% margin of noninferiority (239 mL). Based on these results, we failed to show that urine output in the low MAP group was non-inferior (p = 0.513) (see Fig. 5).
low MAP group was non-inferior because the confidence interval is beyond the non-inferiority region (p = 0.198). 3.2.3. Change in serum creatinine (day 4 – day 1) The mean change in serum creatinine of the high MAP group was −0.54 (SD = 0.63) mg/dL. The mean change the low MAP group serum creatinine was −0.77 (SD = 1.14) mg/dL. As can be seen in Fig. 7, we failed to show that urine output in the low MAP was noninferior because the confidence interval is beyond the non-inferiority region (p = 0.553). 3.2.3.1. Secondary outcomes. Death: During the study period, two (22%) patients died in the high MAP group and one (13%) patient died in the low MAP group (p N 0.999). However, the in-hospital mortality was six (67%) and six (75%) in the high MAP group and low MAP groups respectively (p N 0.999). Cardiac events: No arrhythmias were found the low MAP group versus three (33%) patients in the high MAP group. This difference in arrhythmias was not found to be statistically significant (p = 0.206). Fatal arrhythmias: No fatal arrhythmias occurred in any patients during this trial. Myocardial infarction: No fatal myocardial infarctions were seen in any patients during this trial. Vascular events: one patient had an episode of intestinal ischemia in the high MAP group and no patients had any vascular events in the low MAP group during the study period.
3.2.2. Change in urine output (day 4 – day 1) The mean change in urine output in the high MAP group was −689 (SD = 1684) mL/24 h. The mean change in urine output of the low MAP group was 272 (SD = 582) mL/24 h. The difference is −961 mL/24 h. As can be seen in Fig. 6, we failed to show that change in urine output in the Table 3 Number and percentage of patients that received vasopressors according to group.
Norepinephrine (%) Vasopressin (%) Phenylephrine (%)
High MAP
Low MAP
p
7 (78) 2 (22) 1 (11)
7 (88) 2 (25) 1 (13)
N0.999 N0.999 N0.999
Fig. 4. Median and interquartile range of daily fluid balance according to group.
190
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
Non inferiority region 95% Confidence Interval Mean: High MAP
Mean: Low MAP
500
0
500
1000
1500
2000
2500
Urine output Day 4 Fig. 5. Mean urine output at day 4 for the high and low mean arterial pressure groups with non-inferiority region and 95% confidence interval for the low mean arterial pressure group.
4. Discussion The main finding of our study is that all three primary endpoint results do not allow us to claim non-inferiority of the low MAP group compared to high MAP group. With regards to total urine output at day four, change in urine output and change in serum creatinine, all confidence intervals obtained were outside the 20% margin of non-inferiority. The statistical analysis of this study shows the study power to be 13.99%, indicating that our study is significantly underpowered and unable to prove non-inferiority. In order to show non-inferiority in a similarly constructed study, the results of this pilot trial suggest that 474 subjects would be required in each arm. To our knowledge, there are no clinical trials in HRS with this number of subjects. The pathophysiologic state of HRS is one of renal hypoperfusion and thus there is a rationale in attempting reversal of HRS through augmentation of blood pressure to a supraphysiologic level. Although the same rationale can be used for any state of hypotension causing a kidney injury, studies in sepsis have failed to demonstrate a benefit to maintaining high MAP targets [17]. Patients with end-stage liver disease develop cardio-circulatory dysfunction characterized by hyperdynamic syndrome and cirrhotic cardiomyopathy [18]. This may be responsible for some of the poor vasopressor response that we witnessed.
Vasopressor sparing agents have shown to be efficacious in the treatment of HRS. Commonly used agents include terlipressin and midodrine. Terlipressin is perhaps the most widely studied and used agent in the treatment of HRS, internationally. Its use has been shown to be superior to placebo in reversal of HRS [8,15,19,20]. However, terlipressin does not have approval from the Food and Drug Administration in the United States(US). In the US, HRS is commonly treated with midodrine, octreotide, albumin and vasopressors [10,21]. Several studies have compared norepinephrine with terlipressin for the treatment of HRS [20,22,23]. In a randomized controlled trial of 46 patients with HRS type II, norepinephrine had similar efficacy to terlipressin in reversal of HRS. However, this study assessed specifically HRS type II, which is less severe and has more favorable outcomes compared to HRS type I [22]. Nanda et al. performed a systemic review and meta-analysis of pharmacologic therapies in HRS and found that norepinephrine is an effective alternative to terlipressin with fewer adverse effects [21]. Mattos et al. performed a systemic review and meta-analysis of 4 studies comparing norepinephrine with terlipressin in HRS. There was no significant difference in 30-day survival between the 2 groups [23]. Duvoux et al. demonstrated that most patients with HRS type I had a reversal of their disease when treated with norepinephrine combined with albumin and furosemide [7]. The study, however, was
Non inferiority region 95% Confidence Interval Mean Change: High MAP
Mean Change: Low MAP
1000
500
0
500
Change in urine output (Day 4
1000
1500
Day 1)
Fig. 6. Mean change in urine output (day 4 – day 1) for the high and low mean arterial pressure groups with non-inferiority region and 95% confidence interval for the low mean arterial pressure group.
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
191
Non−inferiority region 95% Confidence Interval Mean Change: High MAP
Mean Change: Low MAP
−3
−2
−1
0
1
2
Change in creatinine (Day 4 − Day 1) Fig. 7. Mean change in serum creatinine (day 4 – day1) for the high and low mean arterial pressure groups with non-inferiority region and 95% confidence interval for the low mean arterial pressure group.
limited by a small sample size (only 12 patients) and lack of a comparator arm. In a systematic review, Gluud et al. found that the use of a vasoconstrictor reduced mortality by 18% in HRS. Despite the inclusion of 10 randomized trials, the review only included 376 patients and the confidence interval for the outcome was wide [24]. Our study has a number of limitations. Although we calculated a sample size a priori, our study was underpowered to demonstrate non-inferiority. Our study only included 2 hospitals, limiting its external generalizability. The percentage of time at MAP goal was significantly different per group. The high MAP group spent more time at target, but there was wide variability of MAP target achievement even within a group. As an example, in the high MAP group, time at target was as low as 31% and as high as 86%. In the low MAP group, these figures ranged from 7% to 44%. This demonstrates the difficulty of maintaining MAP at a narrow range (65–70 mmHg). This may be, in part, due to us not artificially lowering the MAP to target it MAP in the low MAP group were spontaneously above goal. The fluid balance differed between the two groups on day one (see Fig. 4). The use of cardiac output monitoring was not protocolized in this study. Although cardiac output monitoring was used to aid with volume assessment in some patients, it was not used in all patients. We do not have access to how many patients had cardiac output monitoring. Volume status assessment was made clinically by the treating ICU team. One patient that had a history of hypertension that was actively undergoing treatment for hypertension at the time of study recruitment ended up being inadvertently enrolled. Additionally, although the study was randomized, it was not blinded. For an ideal trial, MAP data monitoring is paramount. We checked MAP hourly, and there was considerable difficulty in keeping the MAP of each patient in the goal range for each group—only 27% of the time patients in the low MAP group, and 48% of the time in the high MAP group. From an analysis standpoint, even if we were to have had shown non-inferiority, with such difficulty keeping patients in the MAP goal, it would be difficult to justify such a conclusion. Many patients randomized to the low MAP group had spontaneously higher MAP than goal; it may be unethical to artificially lower their MAP. One way an ideal study could address this would be to expand the range of the low MAP goal. In addition, our clinical endpoints of interest (urine output and serum creatinine) were considerably more varied in our patients than literature had suggested. This may be due to the fact that we used the patients' worst outcome for these endpoints. Additionally, simple randomization was used for this trial. An adaptive trial design, such as minimization, may allow better balance between potentially confounding factors between groups.
5. Conclusion Hepatorenal syndrome is a deadly condition that carries a high mortality, despite aggressive medical intervention. In this study, we assessed the efficacy of maintaining high target MAPs in patients with HRS in the ICU. Our findings do not allow concluding that lower MAP targets are non-inferior to higher MAP targets because this study is underpowered. Nonetheless, this study can be used as a basis for the planning of well-powered, multicenter, randomized clinical trials assessing clinically relevant outcomes in HRS. Acknowledgements Benadin Varajic, MD has contributed to subject recruitment, data collection, data processing, manuscript authorship, manuscript revision, and journal submission. Rodrigo Cavallazzi, MD has contributed to data processing, manuscript authorship, manuscript revision and journal submission. Jason C Mann, DO has contributed to study design, subject recruitment, data collection. Stephen Furmanek, PhD has contributed to data processing, manuscript authorship, and manuscript revision. Juan Guardiola, MD has contributed to study design and manuscript authorship. Mohamed Saad MD, is the primary investigator of this study; he has contributed to every step of the study from inception to manuscript submission; including: study design, research protocols, subject recruitment, and manuscript authorship and submission. Drs. Varajic and Saad take responsibility for the integrity of this study. Funding was provided by the University of Louisville School of Medicine Departments of Internal Medicine: Division of Medical Education, Division of Pulmonary, Critical Care and Sleep Disorders Medicine, and Division of Infectious Disease Medicine. Conflict of interest The authors of this study do not have any conflicts of interest to disclose. References [1] Wadei HM, Mai ML, Ahsan N, Gonwa TA. Hepatorenal syndrome: pathophysiology and management. Clin J Am Soc Nephrol 2006;1:1066–79.
192
B. Varajic et al. / Journal of Critical Care 52 (2019) 186–192
[2] Gines A, Escorsell A, Gines P, et al. Incidence, predictive factors, and prognosis of the hepatorenal syndrome in cirrhosis with ascites. Gastroenterology 1993;105:229–36. https://doi.org/10.1016/0016-5085(93)90031-7. [3] Salerno F, Cazzaniga M, Merli M, et al. Diagnosis, treatment and survival of patients with hepatorenal syndrome: a survey on daily medical practice. J Hepatol 2011;55: 1241–8. https://doi.org/10.1016/j.jhep.2011.03.012. [4] Martin-Llahi M, Guevara M, Torre A, et al. Prognostic importance of the cause of renal failure in patients with cirrhosis. Gastroenterology 2011;140:488–96 [e4] https://doi.org/10.1053/j.gastro.2010.07.043. [5] Guevara M, Gines P, Bandi JC, et al. Transjugular intrahepatic portosystemic shunt in hepatorenal syndrome: effects on renal function and vasoactive systems. Hepatology (Baltimore, Md) 1998;28:416–22. https://doi.org/10.1002/hep. 510280219. [6] Angeli P, Volpin R, Gerunda G, et al. Reversal of type 1 hepatorenal syndrome with the administration of midodrine and octreotide. Hepatology (Baltimore, Md) 1999;29:1690–7. https://doi.org/10.1002/hep.510290629. [7] Duvoux C, Zanditenas D, Hezode C, et al. Effects of noradrenalin and albumin in patients with type I hepatorenal syndrome: a pilot study. Hepatology (Baltimore, Md) 2002;36:374–80. https://doi.org/10.1053/jhep.2002.34343. [8] Moreau R, Durand F, Poynard T, et al. Terlipressin in patients with cirrhosis and type 1 hepatorenal syndrome: a retrospective multicenter study. Gastroenterology 2002; 122:923–30. https://doi.org/10.1053/gast.2002.32364. [9] Runyon BA. Introduction to the revised American Association for the Study of Liver Diseases practice guideline management of adult patients with ascites due to cirrhosis 2012. Hepatology (Baltimore, Md) 2013;57:1651–3. https://doi.org/10.1002/hep. 26359. [10] Nadim MK, Durand F, Kellum JA, et al. Management of the critically ill patient with cirrhosis: a multidisciplinary perspective. J Hepatol 2016;64:717–35. https://doi. org/10.1016/j.jhep.2015.10.019. [11] de Mattos AZ, de Mattos AA, Mendez-Sanchez N. Hepatorenal syndrome: current concepts related to diagnosis and management. Ann Hepatol 2016;15: 474–81. [12] Nadim MK, Kellum JA, Davenport A, et al. Hepatorenal syndrome: the 8th international consensus conference of the Acute dialysis Quality Initiative (ADQI) Group. Crit Care 2012;16:R23. https://doi.org/10.1186/cc11188.
[13] Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Crit Care Med 2017; 45:486–552. https://doi.org/10.1097/CCM.0000000000002255. [14] Wagener G, Kovalevskaya G, Minhaz M, Mattis F, Emond JC, Landry DW. Vasopressin deficiency and vasodilatory state in end-stage liver disease. J Cardiothorac Vasc Anesth 2011;25:665–70. https://doi.org/10.1053/j.jvca.2010.09.018. [15] Solanki P, Chawla A, Garg R, Gupta R, Jain M, Sarin SK. Beneficial effects of terlipressin in hepatorenal syndrome: a prospective, randomized placebocontrolled clinical trial. J Gastroenterol Hepatol 2003;18:152–6. https://doi.org/10. 1046/j.1440-1746.2003.02934.x. [16] Team RC. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2016. [17] Asfar P, Meziani F, Hamel JF, et al. High versus low blood-pressure target in patients with septic shock. N Engl J Med 2014;370:1583–93. https://doi.org/10.1056/ NEJMoa1312173. [18] Fede G, Privitera G, Tomaselli T, Spadaro L, Purrello F. Cardiovascular dysfunction in patients with liver cirrhosis. Ann Gastroenterol 2015;28:31–40. [19] Rajekar H, Chawla Y. Terlipressin in hepatorenal syndrome: evidence for present indications. J Gastroenterol Hepatol 2011;26(Suppl. 1):109–14. [20] Wang H, Liu A, Bo W, Feng X, Hu Y. Terlipressin in the treatment of hepatorenal syndrome: a systematic review and meta-analysis. Medicine (Baltimore) 2018;97: e0431. https://doi.org/10.1097/MD.0000000000010431. [21] Nanda A, Reddy R, Safraz H, Habeeb S, Singal AK. Pharmacological therapies for hepatorenal syndrome: a systematic review and meta-analysis. J Clin Gastroenterol 2018 Apr;52(4):360–7. https://doi.org/10.1097/MCG.0000000000000913. [22] Ghosh S, Choudhary NS, Sharma AK, et al. Noradrenaline vs terlipressin in the treatment of type 2 hepatorenal syndrome: a randomized pilot study. Liver Int 2013;33: 1187–93. https://doi.org/10.1111/liv.12179. [23] Mattos AZ, Mattos AA, Ribeiro RA. Terlipressin versus noradrenaline in the treatment of hepatorenal syndrome: systematic review with meta-analysis and full economic evaluation. Eur J Gastroenterol Hepatol 2016;28:345–51. https://doi.org/10.1097/ MEG.0000000000000537. [24] Gluud LL, Christensen K, Christensen E, Krag A. Systematic review of randomized trials on vasoconstrictor drugs for hepatorenal syndrome. Hepatology (Baltimore, Md) 2010;51:576–84. https://doi.org/10.1002/hep.23286.