In patients with cirrhosis, vasoconstrictor therapy elicits splanchnic vasoconstriction and reduces the complications of portal hypertension.1 In the last decade, evidence has been gathered in support of the use of vasoconstrictors to stop active variceal bleeding2 and in the treatment of renal failure in patients with type 1 hepatorenal syndrome (HRS).3–11 In addition, preliminary studies suggest that vasoconstrictor administration may be a novel approach for the treatment of type 2 HRS3, 7, 12 or in the prevention of the hemodynamic alterations caused by large volume paracentesis [known as paracentesis-induced circulatory dysfunction (PICD)].13 Finally, preclinical evidence suggests that vasoconstrictor administration may be useful in the prevention of arterial hypotension induced by lipopolysaccharide, a gram-negative bacterial byproduct.14 This article focuses on the use of vasoconstrictors in circumstances other than variceal bleeding. Indeed, vasoconstrictor therapy for variceal bleeding has been the topic of many recent reviews (e.g., de Franchis2).
In patients with cirrhosis and type 1 hepatorenal syndrome (HRS), systemic vasodilation, which is mainly attributable to splanchnic vasodilation, plays a critical role in the activation of endogenous vasoconstrictor systems, resulting in renal vasoconstriction and functional renal failure. It has been suggested that the use of splanchnic (and systemic) vasoconstrictors such as terlipressin (a vasopressin analog) or alpha-1-adrenoceptor agonists (midodrine or noradrenaline) may improve renal function in patients with type 1 HRS. Six studies (with only one randomized study in a small series of patients) have shown that terlipressin improves renal function in these patients. However, there is evidence that terlipressin alone may be less effective than terlipressin combined with intravenous albumin in improving renal function. Future randomized studies should confirm this difference and evaluate the impact of terlipressin therapy (with or without intravenous albumin) on survival. Interestingly, in nonrandomized studies, the use of alpha-1 agonists combined with other therapies (octreotide and albumin for midodrine; furosemide and albumin for noradrenaline) has been shown to improve renal function in patients with type 1 HRS. The efficacy and safety of combined therapies including alpha-1 agonists should be confirmed in randomized studies. Finally, preliminary evidence suggests that vasoconstrictor administration may be a novel therapeutic approach targeting vasodilation involved in the mechanism of: (1) renal failure in type 2 HRS; (2) paracentesis-induced circulatory dysfunction; and (3) arterial hypotension induced by byproducts of gram-negative bacteria. Further studies are needed in all these fields. (HEPATOLOGY 2006;43:385–394.)
Type 1 HRS
Type 1 HRS is an acute functional renal failure that complicates end-stage cirrhosis.15–17 Renal failure is a result of renal hypoperfusion without glomerular or tubular lesions.16 At diagnosis of HRS, serum creatinine levels and creatinine clearance, both markers of glomerular filtration rate (GFR), are above 250 μmol/L and below 20 mL/min, respectively.17 In this case, renal failure is not responsive to the administration of 1.5 liters isotonic saline solution15 or to even more vigorous fluid therapy with intravenous albumin (e.g., 60-80g/d for at least 2 days).3, 7 Type 1 HRS is reversible if renal perfusion is rapidly restored by appropriate therapy.16 In HRS, spontaneous regression of renal failure is very rare.17 Type 1 HRS is a pre-ischemic state that may lead to acute ischemic tubular necrosis.16 HRS is a life-threatening complication of cirrhosis.17–23; the spontaneous median survival time is less than 2 weeks and the probability of survival at 1 month is very low (25%).17 Liver transplantation is the only treatment that can cure end-stage cirrhosis.18–22 However, patients who undergo transplantation with HRS have a lower probability of postoperative survival and a higher probability of developing postoperative complications than patients without HRS.20, 21 Therefore, measures that bridge patients to liver transplantation are needed. In addition, treatment for HRS is also needed for patients who are not candidates for liver transplantation.20
The ideal treatment of type 1 HRS must fulfill at least three aims.20 First, it should induce reversal of HRS defined as a decrease in serum creatinine levels below 133 μmol/L (1.5 mg/dL). Second, there should be a slight prolongation of survival, which may increase the chance of receiving a liver transplant in appropriate candidates. Third, treatment should not induce any severe adverse events, especially because patients with type 1 HRS are in poor medical condition.5, 23
Mechanisms of Type 1 HRS
In patients with type 1 HRS, marked intrarenal, preglomerular vasoconstriction plays a major role in the decrease in GFR (Fig. 1A).15, 16 This renal vasoconstriction is mostly attributable to overactive endogenous renal vasoconstrictor systems (i.e., the renin-angiotensin system and the sympathetic nervous system).15, 16
In patients with type 1 HRS, intense systemic vasodilation, mainly caused by marked splanchnic vasodilation, plays a crucial role in decreasing effective arterial blood volume and arterial pressure, which stimulate the endogenous renal vasoconstrictor systems through arterial baroreceptor unloading (Fig. 1B).24–26 This model of the pathophysiology of HRS suggests that systemic and splanchnic vasoconstrictors may be useful for the treatment of renal failure in patients with type 1 HRS.16, 20–22
Recently, results of hemodynamic studies have suggested that mechanisms other than systemic vasodilatation may be involved in the development of type 1 HRS27 (Fig. 1B). Indeed, the development of this syndrome was found to be associated with significant reductions in both cardiopulmonary pressures (i.e., right atrial and pulmonary capillary wedge pressures) and stroke volume.27 Because decreases in these pressures are known to induce cardiopulmonary baroreceptor unloading, resulting in an increased activity of renal vasoconstrictor systems, reductions in cardiopulmonary pressures may contribute to renal vasoconstriction in patients with type 1 HRS.27 Patients who develop type 1 HRS have a reduction in stroke volume, which is not compensated by a rise in heart rate.27 As a result, cardiac output decreases,24 which may cause arterial underfilling and activation of endogenous renal vasoconstrictor systems.27 These findings suggest that type 1 HRS is the result of a decrease in cardiac output during marked systemic vasodilation.27
The mechanisms causing decreased stroke volume are unclear. It may attributable to a reduction in cardiac preload suggested by the decrease in cardiopulmonary pressures. In this case, stroke volume should increase in response to optimization of cardiac preload with intravascular fluid loading. Interestingly, plasma volume expansion is mandatory for the diagnosis of type 1 HRS.15 In addition, left ventricular dysfunction, a complication found in certain patients with cirrhosis,28 may contribute to decreased stroke volume. Investigations of left ventricular function (e.g., measurements of ejection fraction) should be performed at the onset of type 1 HRS. Indeed, patients who develop type 1 HRS and receive vasoconstrictor therapy may have effects on cardiac function as a result of this treatment.
The Use of Vasoconstrictors for Type 1 HRS
Vasoconstrictors evaluated in type 1 HRS include vasopressin analogs and alpha-1-adrenoceptor agonists.
The analogs used were ornipressin and terlipressin. These drugs elicit arteriolar constriction by stimulating V1a vasopressin receptors expressed at the surface of vascular smooth muscle cells. Vasopressin analogs are not universally available; for example, terlipressin is not licensed for use in North America.
Acute Hemodynamic Responses to Vasopressin Analogs in Patients With HRS.
Acute administration of vasopressin analogs induces vasoconstriction in systemic and splanchnic arterioles.29 Acute hemodynamic studies of the effects of intravenous ornipressin administration (6 IU/h over a period of 4 hours)29 noted the presence of splanchnic vasoconstriction was associated with a significant systemic vasopressor effect (i.e., increased mean arterial and pulmonary capillary wedge pressures) and coupled with a significant decrease in endogenous renal vasoconstrictor system activity29 (Fig. 2). The latter may be explained in part by arterial and cardiopulmonary baroreceptor uploading caused by the ornipressin-induced vasopressor effect. The overall response to ornipressin was associated with significant increases in renal plasma flow and GFR.29
Acute ornipressin administration decreases cardiac output response.29 In the classic paradigm of vasoconstrictor therapy for HRS, the decrease in cardiac output is viewed as a mere consequence of the bradycardia elicited by ornipressin-induced arterial baroreceptor uploading.
No study has been published of the effects of acute terlipressin administration alone on systemic hemodynamics in patients with type 1 HRS. In these patients, however, the terlipressin-induced response is probably very similar to that elicited by ornipressin or by terlipressin in nonazotemic patients with cirrhosis. In the latter patients, the acute administration of terlipressin alone (1-2 mg, intravenous bolus) evokes systemic vasoconstrictor and vasopressor effects associated with bradycardia and a reduction in cardiac output.30, 31 Finally, 2-day terlipressin administration is known to induce an arterial vasopressor effect associated with significant decreases in plasma renin concentrations and significant increases in GFR.32
The effects of 1 to 2 weeks' administration of ornipressin or terlipressin alone on systemic hemodynamics are unknown.
Clinical Use of Ornipressin.
Administration of ornipressin for 1 to 2 weeks, in combination with albumin33 or dopamine,34 results in increased GFR. However, ornipressin administration was found to induce severe ischemic adverse events that prevent the clinical use of this drug.33, 34
Clinical use of Terlipressin.
Several clinical studies evaluating the efficacy of long-term terlipressin administration in patients with HRS have been published3–8 (Table 1). In these studies, the diagnosis of HRS was based on the criteria proposed by the International Ascites Club.15 The dose of terlipressin ranged from 2 to 6 mg/day. In prospective studies, terlipressin was given until serum creatinine values decreased to below 130 μmol/L or for a maximum of 15 days. The duration of terlipressin therapy ranged from 7 to 15 days.
|Study Characteristics||First Author (reference)|
|Uriz et al.3||Mulkay et al.4||Moreau et al.5||Halimi et al.6||Ortega et al.7||Solanki et al.8|
|Patients with type 1 HRS, n||6||12||99||16||16||24|
|Patients with type 2 HRS, n||3||0||0||2||5||0|
|Study design||Prospective, nonrandomized||Prospective, nonrandomized||Retrospective||Retrospective||Prospective, nonrandomized —13 patients received terlipressin plus albumin −8 patients received terlipressin alone||Single-blind, randomized, placebo-controlled —12 patients received terlipressin —12 patients received placebo|
|Terlipressin dose IV||0.5-2 mg/4 h||2 mg IV every 8 h or 12 h||3.2 ± 1.3 mg/day*||4 mg/day (1.5-12 mg/day)†||0.5-2 mg/4 h||1 mg/12 h|
|Concomitant IV albumin administration||1 g/kg on day 1 followed by 20-40g daily||NA||75% of patients received albumin at a mean dose of 40 g/day||NA||1 g/kg on day 1 followed by 20-40g daily||All patients received albumin during follow-up, dose not specified|
|Duration of therapy||5-15 days||1 week-2 months||11 ± 12 days*||7 days (2-16 days)†||4-14 days||4-15 days|
|Study end point||Reduction of serum creatinine levels to <133 μmol/L||NA||Reduction of serum creatinine levels to <130 μmol/L or a decrease of at least 20% at the end of therapy||Reduction of serum creatinine levels of at least 30% at day 5||Reduction of serum creatinine levels to <133 μmol/L||Reduction of serum creatinine levels to <133 μmol/L|
|Proportion of patients reaching the study end point (%)||78||NA||58||72||“Terlipressin plus albumin”: 77 “Terlipressin alone”: 25||Terlipressin: 42 Placebo: 0|
|Serum creatinine levels (μmol/L) for the whole group|
|At entry||441 ± 79‡||384§||254 ± 99*¶||298 ± 124‡||NA**||NA††|
|At the end of therapy||170 ± 23‡||203§ (at day 7)||NA¶||145 ± 85‡||NA**||NA††|
|Proportion of patients with adverse events (%)§§||11||0||18||22||5||25|
One weakness in these studies is their methodology. The only randomized, controlled trial was single-blinded and performed in a small series of patients.8 The others were prospective but nonrandomized3, 4, 7 or retrospective.5, 6 The only large study was retrospective.5 Certain prospective studies included both type 1 and type 2 HRS,3, 7 conditions that differ with regard to survival.20 Nevertheless, because renal failure does not spontaneously improve in patients with HRS,17 terlipressin obviously has beneficial effects on renal function. Indeed, studies using this drug have shown a reversal of HRS in 25% to 80% of patients (Table 1).
Certain patients with type 2 HRS (i.e., with the chronic form of HRS, see below) develop type 1 HRS.21 In these patients, terlipressin therapy induces a decrease in, but not a “normalization” of, serum creatinine levels, which remain above 133 μmol/L at the end of treatment. Thus, in patients with “acute-on-chronic” HRS treated with terlipressin, an end point other than a decrease in serum creatinine level below 133 μmol/L should be used to define renal responders.
Type 1 HRS (like other causes of functional renal failure) and ischemic acute tubular necrosis represent a continuum, with the former leading to the latter when blood flow is sufficiently compromised to result in the death of tubular cells.16 Even when International Ascites Club diagnostic criteria15 are used at enrollment, it may be difficult to distinguish patients with true type 1 HRS from patients who have already developed “HRS-induced” ischemic acute tubular necrosis.16 At enrollment, none of the studies3–8 measured the usual markers of tubular damage, including urinary excretion of beta2-microglobulin16, 35 (Table 2). In other words, these studies may have enrolled patients with true “HRS-induced” ischemic acute tubular necrosis. This is important because renal function does not improve in patients with tubular necrosis in response to terlipressin therapy.16
|Type 1 HRS||Acute tubular necrosis|
|History of recent shock*||No||Frequent|
|History of recent use of nephrotoxic drugs*||No||Frequent|
|Sodium concentrations (mmol/L)||<20||>40|
|Fractional excretion of sodium (%)||<1||>1|
|Urine osmolality (mOsm/kg)||<500||>350|
|Renal pathology†||No cellular lesion||Necrotic renal tubules|
Are there patients in whom terlipressin is ineffective? The terlipressin-induced improvement of renal function is thought to be mediated by the systemic and splanchnic hemodynamic response to this drug.20–23 However, because no studies have been done on systemic hemodynamics after several days of terlipressin in patients with type 1 HRS, the exact pattern of “good” or “poor” terlipressin-induced hemodynamic response (i.e., the response correlated with the improvement or deterioration of renal function) is unknown. Theoretically, the following mechanisms may explain the lack of improvement of renal function in response to terlipressin. First, the vasoconstrictor/vasopressor response to terlipressin, which triggers improvement of renal function, may not occur in certain patients with type 1 HRS and severe liver failure, even when maximum doses of terlipressin are used. Indeed, severe liver failure is known to induce systemic and splanchnic arterial hyporeactivity to exogenous vasoconstrictors.36 Interestingly, a Child-Pugh score above 13 is predictive of a lack of beneficial effect of terlipressin on renal function.5 Second, terlipressin therapy may elicit a reduction in cardiac output by mechanisms other than physiological bradycardia, for example, by inducing coronary vasoconstriction37 responsible for left ventricular dysfunction and subsequent decreased stroke volume and arterial underfilling. This latter mechanism may limit improvement of renal function in response to terlipressin-induced splanchnic and systemic vasoconstriction. However, no evidence of arterial underfilling was found in an acute hemodynamic study on the effects of the administration of 2 mg terlipressin in patients with cirrhosis.31 Clearly, hemodynamic studies after 1 to 2 weeks' terlipressin administration are needed.
An independent predictive factor of improvement of renal function in response to terlipressin is younger age.5 The reasons for this are unclear. In addition, the nature of the renal response to terlipressin may depend on the intravenous administration of albumin. Results of a prospective, nonrandomized study in patients with HRS show that renal function improves more frequently in patients treated with terlipressin and intravenous albumin than in those treated with terlipressin alone (77% and 25%, respectively).7 Thus, albumin seems to accentuate the renal response to terlipressin. However, the mechanisms for this beneficial effect are unknown. Albumin alone does not have any significant effects on renal function in patients with type 1 HRS.3, 7, 38 Certain patients treated with albumin may become “good hemodynamic responders” to terlipressin and thus become “renal responders” to this drug. Again hemodynamic studies are needed. Conversely, the study comparing terlipressin alone with terlipressin plus albumin7 had many limitations. For example, among patients with type 1 HRS, the proportion of renal responders and nonresponders in the terlipressin alone group and in the terlipressin plus albumin group was not given.7 Moreover, a discrepancy was seen between this study7 and a retrospective study conducted in a large series of patients treated with terlipressin,5 which did not show any significant difference between patients with improved renal function and those without, in terms of the number of patients receiving albumin and the dose of albumin. Thus, the role of albumin in the terlipressin-induced improvement of renal function should be clarified in randomized, comparative studies.
In certain studies, terlipressin was given at fixed doses (1 mg every 8 or 12 hours).5, 8 However, the effect of a dose of terlipressin may differ from one patient to another, especially according to the degree of liver failure. The higher the Child-Pugh score, the greater the dose of terlipressin.7 Interestingly, other studies used goal-directed terlipressin therapy3, 7; terlipressin was initially given at a dose of 0.5 mg/4 h and increased in a stepwise fashion every 3 days to 1 mg/4 h and 2 mg/4 h if a significant reduction in serum creatinine (of at least 88 μmol/L (1 mg/dL)) was not observed.3, 7 Although a 3-day delay is reasonable for obtaining an 88 μmol/L decrease in serum creatinine, it also may favor a shift from terlipressin-sensitive functional renal failure to terlipressin-resistant tubular necrosis. The impact of more rapid increases in doses of terlipressin according to goals other than a 3-day decrease in serum creatinine levels has not yet been studied. Because the aim of terlipressin therapy is to inhibit systemic hemodynamic alterations that play a role in renal failure, hemodynamic end points could be useful to rapidly adjust terlipressin doses. These hemodynamic end points must be defined. Several candidates exist, including cardiopulmonary pressures, arterial pressure, cardiac output, and left ventricular ejection fraction.
Recurrence of HRS after treatment withdrawal is observed in approximately 50% of patients.39, 40 Predictive factors of recurrence are unknown. Retreatment of recurrence with terlipressin is effective.40
In the largest study the median survival was 3 weeks and the chance of survival was 60%, 40%, and 28% at 15 days, 1 month, and 2 months, respectively.5 After the first 3 months, the probability of survival remained stable, at 19% for several months. In a previous study investigating the natural history of HRS, the median survival time was 1.7 weeks, and the probability of survival was 40%, 25%, and 18% at 15 days, 1 month, and 2 months, respectively.17 These findings suggest that terlipressin may improve survival in patients with type 1 HRS. Moreover, the large retrospective study also showed that the independent predictive factors of survival were improved renal function during terlipressin therapy and a Child-Pugh score of 11 or less at inclusion.5 Thus, in patients with type 1 HRS treated with terlipressin, the improvement in renal function may have a beneficial effect on survival that is independent of the severity of the underlying liver dysfunction.
In the previously mentioned large retrospective study, a significant proportion of patients (13%) underwent liver transplantation.5 Renal function improved in most of these patients during terlipressin therapy. Patients with HRS treated with terlipressin before transplantation have a low incidence of posttransplantation complications, similar to patients undergoing transplantation without HRS.41 Thus, terlipressin may be used as a bridge to liver transplantation.42
The frequency of ischemic side effects requiring the discontinuation of terlipressin was approximately 15% (Table 1). Overall, terlipressin administration was well tolerated in the studies, probably because significant efforts were made to exclude patients with contraindications to treatment (Table 3).
|History of coronary artery disease|
|Dilated and nondilated cardiomyopathies|
|Obliterative arterial disease of the lower limbs|
|Asthma, chronic obstructive pulmonary disease|
|Age > 70 years|
To summarize, evidence suggests that terlipressin is a good candidate for the pharmacological treatment of renal failure in patients with type 1 HRS. Unresolved issues remain, however. First, the finding that terlipressin combined with albumin improves renal function better than terlipressin alone should be confirmed. Second, more information is needed on the hemodynamic responses to terlipressin therapy (with or without albumin) in these patients. Third, the beneficial effect of terlipressin therapy on survival and the good tolerance to this treatment should be confirmed in randomized studies. A double-blind, randomized, placebo-controlled trial conducted in patients with type 1 HRS is ongoing in the United States and Germany.
Alpha-1 agonists have not been used alone in patients with type 1 HRS. When used alone, alpha-1 agonists elicit systemic vasoconstriction. However, the effects of these drugs on the splanchnic circulation is unknown in patients with cirrhosis, and there is no evidence that alpha-1 agonist–induced vasoconstriction per se is associated with improved renal function in patients with type 1 HRS. In patients with type 2 HRS, acute midodrine administration was found to elicit systemic vasoconstriction; however, this effect was not accompanied by any improvement of renal function.43 In fact, in patients with type 1 HRS, alpha-1 agonists have only been used in combination with other therapies (discussed later).
Clinical Use of Midodrine.
Two nonrandomized studies evaluated a two-drug combined therapy with midodrine and octreotide.9, 10 This combination was used because midodrine per se was known to induce a systemic vasoconstrictor effect, and octreotide is known to inhibit the secretion of glucagon (a vasodilator that may play a role in cirrhosis-associated vasodilation).9 However, in patients with HRS, octreotide administration alone does not elicit any vasoconstrictor/vasopressor effect and does not improve renal function.38 In addition, intravenous albumin was administered in both studies.9, 10 The first study included 5 patients.9 The doses of midodrine and octreotide were titrated to obtain an increase in mean arterial pressure of at least 15 mm Hg (Table 4). The dose was increased from 20 g/day to 40 g/day when the central venous pressure was less than 12 mm Hg or plasma renin activity was not reduced by more than 50% of basal value after 3 days of therapy. After 20 days of therapy, mean arterial pressure was significantly increased, plasma renin activity significantly decreased, and all patients had serum creatinine levels below 177 μmol/L (2 mg/dL).9 In the second study, 14 patients received the combined therapy (Table 4).10 Midodrine was temporary withheld if the patient's systolic blood pressure was above 120 mm Hg. Therapy was administered until serum creatinine was below 135 μmol/L for at least 3 days. This end point was achieved in 70% of patients. In both studies tolerance to treatment was good.
|Study Characteristics||First Author (reference)|
|Angeli et al.9||Wong et al.10||Duvoux et al.11|
|Study design||Prospective, nonrandomized||Prospective, nonrandomized||Prospective, nonrandomized|
|Alpha-1-agonist (dose)||Midodrine (7.5 mg orally 3 times daily, increased to 12.5 mg 3 times daily if needed)*||Midodrine (2.5 mg/day orally)*||Noradrenaline (0.5-3mg/h intravenously)*|
|Concomitant drug administration (dose)||Octreotide (100 μg subcutaneously three times daily, increased to 200 μg 3 times daily if needed)*||Octreotide (iv bolus of 25 μg followed by 25 μg/h continuous IV administration)||Furosemide (intravenous bolus of 120 mg if needed)*|
|Concomitant intravenous albumin administration||20-40 g daily*||50 g/day||If needed (22 ± 15 g/day were administered)*|
|Duration of therapy||20 days||14 ± 3 days†||10 ± 3 days‡|
|Study end point||NA||Reduction of serum creatinine levels to < 135 μmol/L for at least 3 days||Reduction of serum creatinine levels to < 133 μmol/L|
|Proportion of patients reaching the study end point (%)||NA||71||83|
|Serum creatinine levels (μmol/L) for the whole group|
|At entry||440 ± 71†||NA§||358 ± 161‡|
|At the end of therapy||159 ± 11† (at day 20)||NA§||145 ± 78‡|
|Proportion of patients with adverse events (%)¶||0||0||17**|
Midodrine is a prodrug; it is absorbed from the gastrointestinal tract and metabolized by the liver into an active metabolite, desglymidodrine.44 Elimination is via the urine.44 No pharmacokinetic evaluation of midodrine and its metabolite in patients with type 1 HRS has been performed. Previous clinical studies conducted in these patients used different doses of midodrine (Table 4). Future studies are needed to investigate the pharmacokinetics of midodrine and desglymidodrine in patients with type 1 HRS.
Clinical Use of Noradrenaline.
Noradrenaline has been used in combination with intravenous albumin and furosemide in 12 patients.11 Continuous intravenous administration of noradrenaline was started at 0.5 mg/h, designed to achieve an increase in mean arterial pressure of at least 10 mm Hg or an increase in 4-hour urine output to more than 200 mL. When one of these goals was not achieved, the noradrenaline dose was increased every 4 hours in steps of 0.5 mg/h, up to a maximum dose of 3 mg/h. During noradrenaline therapy, intravenous albumin (20 g) and furosemide (bolus of 120 mg) were used (and repeated if needed) to maintain central venous pressure between 4 and 10 mm Hg (Table 4). In addition, patients underwent a 100% compensation of urine output with crystalloids on a 4-hour basis. Noradrenaline was given either until serum creatinine levels decreased to below 133 μmol/L or for 15 days. Noradrenaline was given at a mean dose of 0.8 mg/h for a mean duration of 10 days. Hemodynamics were measured at the start of therapy and on day 5. At day 5, a significant increase occurred in mean arterial pressure (attributable to slight increases in both cardiac output and systemic vascular resistance) as did a significant decrease in plasma renin concentrations. HRS reversal occurred in 83% of the patients. The drug was well tolerated.
No study has compared terlipressin therapy and combined therapies including alpha-1 agonists.
Thus, studies in a small series of patients with type 1 HRS treated by combined therapies including alpha-1 agonists provide promising results in terms of improvement of renal function. Interestingly, treatment doses were adjusted to achieve prespecified values of arterial pressure,9–11 central venous pressure,9, 11 plasma renin activity,9 and urine output.11 However, the impact of combined therapies using midodrine or noradrenaline on renal function, survival, and safety should be investigated in larger series of patients.
The Use of Vasoconstrictors Beyond Type 1 HRS
Type 2 HRS
Type 2 HRS is “chronic” functional renal failure in patients with refractory ascites.15 Patients with type 2 HRS are often candidates for liver transplantation, and there is evidence that renal function should be improved before transplantation.42 However, no established treatment exists for renal failure in patients with type 2 HRS.21 These patients generally receive standard medical therapy of refractory ascites, that is, repeated large-volume paracentesis plus intravenous albumin.45 Because the mechanisms of type 2 HRS seem to be similar to those of type 1 HRS,27 vasoconstrictor therapy may be effective on renal failure associated with type 2 HRS. Nonrandomized studies that enrolled a small number of patients with type 2 HRS show that 1 to 2 weeks of terlipressin administration (with3, 7 or without12 albumin) improves renal function in these patients. However, no information exists on the long-term efficacy and safety of terlipressin therapy in type 2 HRS. Finally, combined therapies including alpha-1 agonists have not yet been investigated in type 2 HRS. Thus, studies are needed to evaluate the usefulness of vasoconstrictors in the treatment of patients with type 2 HRS.
Large-volume paracentesis, which removes 5 liters of ascites or more, is used in patients with cirrhosis and large ascites.45, 46 When this treatment is used alone, it induces systemic vasodilation, a decrease in effective arterial blood volume, and an increased activity in the renin-angiotensin system in approximately 80% of cases.45 (Fig. 3A). As mentioned earlier, these circulatory alterations are called PICD.47 PICD starts after the end of paracentesis and may persist for months and induce clinically significant complications.46 Intravenous albumin administration after paracentesis has been shown to prevent the decrease in effective arterial blood volume and the resulting increase in the activity of the renin-angiotensin system46 (Fig. 3B). However, the intravenous administration of albumin after paracentesis does not prevent paracentesis-induced systemic vasodilation (i.e, the trigger of PICD) (Fig. 3B). Thus, the administration of a vasoconstrictor may limit paracentesis-induced systemic vasodilation and have an effect on the development of PICD (Fig. 3C). A randomized pilot study comparing the effects of terlipressin and albumin on PICD has been performed in patients with cirrhosis and large ascites treated with paracentesis.13 Twenty patients were randomly assigned to be treated with either paracentesis and terlipressin or paracentesis and albumin. Terlipressin (1 mg intravenous bolus just before paracentesis, 1 mg intravenous bolus 8 and 16 hours after the end of paracentesis) or a 20% solution of human albumin (8 g/L removed ascites) were intravenously administered on the day of paracentesis. The occurrence of PICD was assessed by measuring plasma renin concentrations, at baseline and on the day of hospital discharge (4-6 days after treatment). PICD was defined as an increase in plasma renin concentrations on the day of hospital discharge of more than 50% of the baseline value. The proportion of patients who developed PICD was similar in the two treatment groups. Good tolerance to terlipressin administration was noted. Together, the results of this pilot study suggest that terlipressin therapy and albumin administration are two different approaches to PICD prevention. Therefore, terlipressin therapy may be a supplement to intravenous albumin in the setting of PICD prevention. The long-term use of terlipressin cannot be recommended to prevent PICD in patients with large ascites treated with paracentesis followed by intravenous albumin administration. Indeed, no information is available on the efficacy and safety of terlipressin in these patients. Thus, large randomized studies should be performed to evaluate the effects of terlipressin therapy on the outcome of patients with ascites treated by paracentesis and intravenous albumin. The effects on PICD of treatments using alpha-1 agonists have not yet been studied.
In patients with septic shock, arterial vasodilation and hypotension contribute to tissue hypoperfusion, end-organ dysfunction, and death.48 Vasopressin therapy may be used in patients without cirrhosis with septic shock. Indeed, these patients have marked decreases in the plasma concentrations of endogenous vasopressin.49 Moreover, in patients with refractory septic shock (i.e., with arterial hypotension refractory to exogenous catecholamine administration), the intravenous administration of vasopressin has been shown to increase arterial pressure so that catecholamine treatment may be discontinued.49 Septic shock is more severe in patients with cirrhosis than in patients without cirrhosis.50 In patients with cirrhosis with septic shock, the plasma concentrations of endogenous vasopressin have not yet been measured, and the effects of vasopressin therapy are unknown. Interestingly, in rats with cirrhosis challenged with the gram-negative bacteria byproduct lipopolysaccharide, terlipressin administration was found to improve lipopolysaccharide-induced arterial alterations.14 These findings suggest that vasopressin therapy may be a novel approach in the treatment of patients with cirrhosis and septic shock. Clinical studies are needed in this field.
The paradigm of medical treatment for type 1 HRS is vasoconstrictor therapy. Studies have shown that terlipressin therapy induces an improvement in renal function in these patients. However, terlipressin is not universally available. Moreover, evidence indicates that terlipressin alone may be less effective than terlipressin combined with intravenous albumin to improve renal function. Future randomized studies are needed to confirm this difference and evaluate the impact of terlipressin (with or without intravenous albumin) on survival. The use of alpha-1 agonists (midodrine or noradrenaline) combined with other therapies (octreotide and albumin for midodrine; furosemide and albumin for noradrenaline) is a promising approach. The efficacy and safety of combined therapies including alpha-1 agonists need to be confirmed in large series of patients. In the future, a shift may occur in the paradigm for HRS treatment from “pure vasoconstrictor” therapy to combined therapies. Finally, studies should evaluate the use of vasoconstrictors in the treatment of type 2 HRS, the prevention of PICD, or the prevention of lipopolysaccharide-induced arterial hypotension.