Role of Molecular Adsorbent Recycling System (MARS) in the treatment of patients with acute exacerbation of chronic liver failure

We demonstrated a 65.5 % overall survival rate in eight consecutive patients suffering from acute exacerbation of chronic liver disease following treatment with the MARS detoxication system. This study joins other reports concerning the utility of the MARS system in supporting patients with liver failure. In contrast to the regular methods of hemodiafiltration, this system was designed to remove both low and middle molecular weight water soluble substances as well as albumin bound molecules.

Indeed, serum levels of ammonia, as well as bilirubin were significantly reduced following the treatment.

The system is based on the dialysis of blood against a special membrance coated albumin but is impermeable to it. The albumin compounds with the free binding sites in the membrane, Compete with the carrier proteins for the toxins in the blood. The membrane transiently absorbs and holds the toxins normally carrier proteins for the toxins in the blood. The membrane transiently absorbs and holds the toxins normally carried on the albumin molecules in the blood (i. e. bilirubin and bile acids) that are being released upon contact with the membrane,according to the concentration gradient. After trapping the toxins, they are carried to the other side of the membrane where dialysis against a fluid rich in albumin completely separates the toxins. The dialysate is re-circulated against a charcoal column for the removal of lipophylic toxins such as bile acids and an anion exchange resin for the removal of bilirubin. The albumin containing dialysate is then completely renewed by dialysing it against a normal dialysis solution. In that way, water-soluble toxins are removed. At the end of the process the albumin-containing dialysate solution is ready for re-use thus, reducing the need to replenish the costly albumin solution. The MARS blood purification method minimizes the loss of vital compounds such as growth factors, hormones and vitamins, carried by proteins other than albumin which are not dialysed by this process. This prevents the negative side effects of plasma-sorbent contact, as in plasmapheresis. In addition, the dialysis step of the MARS system enables the removal of water-soluble toxins and maintenance of acid-base and electrolytes balance which are disturbed in patients with ALF due to concomitant loss of renal function. Ammonia is not removed by charcoal because, at physiological pH, it is ionized. However, ammonia is efficiently removed by the MARS system.

A considerable neurological improvement were observed in all patients. This is in contrast to other reports of 30 to 70 % improvement rates in mental status following the institution of liver support to patients with liver failure.

The improvement in the patients’ coma status was accompanied by a reduction of ICP and SjvO₂ values as measured in 3 patients. The initial ICP measurements were from 12 to 22 mmHg i. e. in the normal range. These measurements coincide with the chronic nature of patients’ liver disease.

Patients with chronic liver failure do not normally have elevated ICP. In addition, these sort of patients have a much lower incidence of brain edema. However, the post -treatment decrease in ICP values mach other reports in which bioartificial liver support devices based on pig hepatocytes were used to assist patients with ALF. Undoubtedly, the acute exacerbation of the chronic liver failure led to the comatose state which improved by the MARS treatment followed by a parallel decrease in ICP values. As no sedative were given, the improvement in state of consciousness could not be attributed to a possible removal of albumin-bound sedatives from the circulation by the treatment. The rapid reduction of ICP, occuring within an hour of treatment initiation, might be explained by a decrease in cerebral blood volume or a change in brain water content pathologically increased in patients with ALF or, alternatively, by rapid clearance of ammonia from the serum. Indeed, ammonia is currently thought to be an important factor in the pathogenesis of brain edema in liver failure. In children with urea cycle enzyme deficiencies, removal of ammonia by dialysis ameliorated brain edema and reduced intracranial pressure.

The SjvO₂ value generally reflects cerebral oxygen balance. The pretreatment elevated cerebral venous O₂ saturation value (>75 %) maj reflect compromized cerebral blood flow caused by cerebral edema. In this circumstance the increase in ICP was followed by a decrease in cerebral oxygen extraction. During the MARS treatment, the continuous decrease of SjvO₂ was associated of ICP and an improvement of neurological status. This could possibly be explained by a reduction in the cytotoxic brain edema, a devastating and often a lethal complication which lead to an increased ICP and brain tonsillar herniation. Reduced brain edema and ICP, a possible improvement in cerebral perfusion pressure was achieved leading to enhanced of cerebral cellular metabolism, reflected by elevated oxygen utilization and a reduction of the SjvO₂.

Portal hypertension, a familiar consequence of most liver diseases is manifested by high cardiac output and a low systemic vescular resistance. Although our patients did not represent the fulminant syndrome their hemodynamic pattern was nevertheless similar. This hemodynamic profile steadily with progression of the treatment, as was evident by a continuous elevation of blood pressure and systemic vascular resistence and a reduction of the cardiac output values. The pathophysiology responsible for the vasodilatation in the presence of liver dysfunction is not completely understood but probably related to lack of clearance of various vasodilating substances which accumulate in the blood. Recently, attention has been focused on the role of nitric oxide (NO) as the main substance responsible for the vasodilatation in liver failure. It might be possible that the NO accumulated in patients with liver failure is effwctively cleared from serum using the MARS system. Indeed, plasma NO is transported primarily bound to serum albumin as a s-nitrosothiol. Hence, the MARS system, based on albumin-mediated detoxification process, could remove the excess serum NO causing a decrease of vasodilatation.

During the MARS treatment no significant reduction of serum levels of liver enzymes was noted nor improvement in the coagulation profile. One might argue that the small patient group and the large variability in the liver enzyme levels are responsible for the lack of detection of significant changes in the enzyme levels. However, this is not surprising as the treatment with MARS replaced only excretoty functions of the failing liver and no metabolic recovery is expected to occur. Hence, the failure to correct coagulation abnormalities would be entirely expected. During treatment no additional reduction of platelets level was observed in contrast to that reported by Watanabe,et al, using the bioartificial liver.

Serum albumin levels of our patients was relatively normal. Albumin has a relatively long half life (21 days) and the albumin-containing resuscitative solutions undoubtedly contributed to the elevation of its serum levels. In addition, the reduction of serum albumin levels which was observed with the use of the bioartificial liver was prevented during the treatment with MARS. This might be explained by the greater hemodynamic stability conferred by the MARS, which saved the administration of additional quantities of fluids, thus preventing hemodilution.

Each treatment session lasted 8-9 hours, as a longer time period was associated with maximal filter saturation with loss of function. The Mars system is a relatively new technology for liver support. Nevertheless, its utility was already studied in 13 patients suffering from life threatening acute exacerbation of chronic liver disease whohad not responded to conventional therapy. The overall reported survival rate was 69 %. All patients showed positive response to therapy (reduction in encephalopathy). Furthermore, improvement of biochemical profile (reduction in blood levels of bilirubin, creatinine and bile acids) and liver synthetic function (elevation of level of coagulation factors) were also noted. It seems that the MARS membrane separation is a protein-impermeable safety barrier between the patients’ serum and any cellular or chemical agent of detoxification. In addition, the system is effective in reducing the blood levels of the protein-bound toxins accumulating in ALF.

In conclusion: the MARS treatment proved to be a promising technology in the treatment of patients with acute exacerbation of chronic liver disease. In view of these encouraging results, further randomized controlled trials are mandatory to determine indications, safety and optimal utilization of this technology in cases of ALF as well as in amelioration of acute decompensation of chronic liver disease.