Use of N‐Acetyl Cysteine for the Treatment of Parenteral Nutrition–induced Liver Disease in Children Receiving Home Parenteral Nutrition

Abstract

The use of long-term parenteral nutrition (PN) in children with significant gut failure has been associated with progressive cholestatic liver disease (1–4). Progression to liver cirrhosis and end-stage liver disease are well-known consequences of long-term dependence on PN. The etiology of liver disease related to PN administration seems to be multifactorial (1–3). There is evidence linking the macronutrient and micronutrient content of PN solutions to the onset and severity of PN-induced liver cholestasis (4). Recent studies in animal models, in particular, have demonstrated an association between the onset of cholestatic liver disease and intravenous (IV) methionine (MET) intake (4–6). Most commercial PN solutions have relatively higher amounts of MET and little or no cysteine (CYS) because of the instability of CYS in solution (7). Inasmuch as CYS is an important precursor of glutathione (GSH), it was hypothesized that provision of parenteral CYS could potentially counteract the oxidative stress that occurs in liver cholestasis (8). Recently, our group showed that N-acetyl cysteine (NAC) can be used as a source of CYS in PN solutions in piglets (9). We supplemented 2 infants and 1 child with PN-induced liver disease who were receiving PN at home with IV NAC as an adjunctive therapy to minimize further liver damage induced by PN. CASE 1 A 13-year-old boy with long-segment Hirschsprung disease demonstrated increasing clinical and biochemical evidence of PN-induced cholestasis secondary to progressive gut failure and increasing dependence on PN. Small bowel length was approximately 100 cm, with no colon in situ. Previous attempts at gastrostomy and gastrojejunal feedings were limited by severe and repetitive episodes of dysmotility and bowel obstruction. This resulted in marked reduction in enteral feeding. Medical therapy included IV pantoprazole, oral ursodeoycholic acid, vitamins E and C, low-dose prednisone and antibiotics (metronidazole) for treatment of gut inflammation. In addition, he had been receiving both domperidone and tegaserod to help promote gastric emptying. He had persistently elevated aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (AP), and γ-glutamyl transpeptidase (γ-GT). He had elevated ferritin, >1500 μmol/L, for more than 18 months before IV NAC was begun. Intravenous NAC was started at an initial dosage of 20 mg/kg/day on the basis of previous work in our laboratory (10,11) and increased in increments of 10 mg/kg/day every 1 to 2 months. NAC was added to the patient's IV hydration solution and delivered over a 12-h infusion period. Liver biochemistries (ALT, AST, conjugated bilirubin, unconjugated bilirubin, γ-GT, bile acids), ferritin, and red blood cell (RBC) GSH levels were measured at baseline and when NAC dosing was changed (Table 1). Increasing the dosage of IV NAC was associated with reductions in liver biochemistries and ferritin that ranged between 53% and 95% and an increase in RBC GSH levels of 29% over 18 months of daily supplementation (Fig. 1). The current IV NAC dosing remains at 135 mg/kg/day. No significant changes in medications occurred over the supplementation period.TABLE 1: Plasma liver biochemistries and ferritin levels in 3 patients treated with intravenous N-acetylcysteineFIG. 1: Impact of IV NAC supplementation on RBC GSH levels in an 11-year-old boy with long-segment Hirschsprung disease.CASE 2 A 9-month-old infant girl (7.5 months corrected age) born at 32 weeks gestation had short gut secondary to a midgut volvulus and internal hernia. After bowel resection, the infant was left with 27 cm of proximal jejunum, 3 cm of terminal ileum, and a descending loop sigmoid colostomy (15-cm colon) and was dependent on PN. The patient had an abrupt deterioration in liver function, with a bilirubin level exceeding 200 μmol/L and impaired synthetic function with hypoalbuminemia and coagulopathy. The results of abdominal ultrasonography were consistent with fatty liver, hepatomegaly, and splenomegaly. The results of septic investigation were negative. The patient was discharged home to receive IV NAC (20 mg/kg/day) administered over 12 hours in Ringer's lactate. Medical therapy included oral ursodeoxcholic acid, fat-soluble vitamins (A, D, E, and K), and cyclic antibiotic therapy for treatment of gut overgrowth (metronidazole and gentamycin). With the exception of dosage changes (to maintain dosage on a per-kilogram basis), no other significant changes in medical therapy were initiated during the NAC supplementation. The nutritional therapy consisted of PN (30% total energy intake, or ∼60 kcal/kg) and enteral gastrostomy feeds (70% of total energy intake, or ∼130 kcal/kg). Total energy intake on a per-kilogram body weight basis via the parenteral and enteral routes did not change over the 10 months that NAC was administered and did not significantly differ from energy distribution from these routes in the months preceding NAC supplementation. Liver biochemistries and ferritin were measured at baseline and once every month in response to incremental increases in NAC dosing (10 mg/kg/day). Increasing the dosage of IV NAC was associated with reductions in liver biochemistries and ferritin that ranged between 35% and 90% during the supplementation period (Table 1). The IV NAC dosage for this infant remained at 120 mg/kg/day for more than 4 months, with stable liver function parameters in the ambulatory setting until liver and small bowel transplantation were performed. CASE 3 A 4-month-old infant girl had a history of left-sided diaphragmatic hernia that resulted in a small bowel infarction. After surgery, the infant had 66 cm of small bowel, the ileocecal valve, and an intact colon. The infant was discharged from hospital to receive PN and IV NAC added to her IV hydration solution at an initial dosage of 50 mg/kg/day administered over 12 to 14 hours. Medical therapy included oral ursodeoxcholic acid, fat-soluble vitamins (A, D, E, and K), and cyclic antibiotic therapy for treatment of gut overgrowth (metronidazole and gentamycin). No significant changes in medical therapy occurred during the NAC supplementation. Nutritional therapy consisted of PN (∼40 kcal/kg) and gastrostomy/oral feeds (∼64 kcal/kg) at the time of IV NAC administration. Energy intake via parenteral sources on a per-kilogram body weight basis remained constant over the IV NAC administration period, although enteral sources of energy increased to 95 kcal/kg. Liver biochemistries and ferritin were measured at baseline and once every month in response to incremental increases in IV NAC dosing (10–20 mg/kg/day). Increasing the dosage of IV NAC was associated with reductions in liver biochemistries and ferritin that ranged between 65% and 90% during the 6-month supplementation period (Table 1). IV dosing of NAC remained at 70 mg/kg/day, with stable liver function until PN was discontinued 3 months later. METHODS Intravenous NAC was obtained from Coram Health Care in the ambulatory setting and was dispensed in quantities of 100 mg/mL in a sterile syringe prepared aseptically in a laminar air flow hood. Written consent was obtained from the patients' responsible caregivers. All of the study procedures were approved by the Research Ethics Board of the Hospital for Sick Children in Toronto. GSH analysis was conducted in our laboratory by use of liquid chromatography tandem mass spectrometry, adapted from the method of Jahoor et al (12). DISCUSSION The rationale behind the administration of NAC in these children given PN was to increase CYS availability for GSH synthesis, inasmuch as evidence has suggested that oxidative damage is an important component of PN-associated liver disease (13). There is also evidence that elevated levels of MET have been associated with increasing severity of cholestasis caused by a defect in the transsulfuration pathway of the MET (4–6,14). Timbrell and Waterfield (13) showed that NAC infusion in rat models was associated with increased intracellular GSH biosynthesis, whereas others have shown that NAC supplementation improves the RBC GSH synthesis rate in children with edematous malnutrition (15). Recently, our group showed that NAC can be used as a source of CYS in PN solutions and, when added to PN fed to neonatal piglets, results in similar rates of nitrogen retention and growth in comparison with neonatal piglets fed equimolar IV amounts of L-cysteine (9). This suggests not only that NAC is an effective precursor of CYS for growth and protein deposition but also that it may confer the additional benefit of promoting an improved antioxidant status in parenterally fed animals and hence may be an effective clinical strategy to temporize the oxidative damage to the liver induced by PN. Given that commercial PN solutions have little or no CYS and relatively higher amounts of MET, IV NAC was provided to these children at varying dosages. Dosing was based upon work in our group in children and adults examining total sulfur amino acid requirements (10,11,16,17). We initiated NAC dosing at estimated maintenance requirements, and then empirically increased dosages by 10 mg/kg/day to a maximum of 135 mg/kg/day in case 1. NAC dosing provided approximately 7.1% of total amino acid intake (g/100 g total amino acid) at this level of intake and was added to the existing IV hydration solution that was provided to the patient. We used the same approach in the subsequent 2 cases, in which maximum dosages of NAC of 70 mg/kg/day (patient 3) and 120 mg/kg/day (patient 2) were administered. These dosages are still below the typical NAC dosage used to treat acetaminophen toxicity (140 mg/kg/day). Potential adverse side effects for NAC include the potential for generalized urticaria (rare), stomatitis, and rhinorrhea and are typically reported at dosages in excess of 140 mg/kg/day. None of the children experienced any of these potential side effects of NAC administration. Although NAC supplementation was associated with reductions in liver biochemical function, in the presence of stable nutritional and medical management, these variables may have also contributed to the improvements in liver biochemical function that were observed in these cases. CONCLUSIONS All of the patients studied demonstrated significant reductions in serum ferritin and in liver biochemistries when given supplementation IV NAC. In addition, RBC GSH levels actually returned to normal with NAC supplementation in 1 patient when it was measured. Inasmuch as RBC GSH reflects the balance between hepatic uptake and renal excretion, it is likely that these changes reflected changes in hepatic GSH balance in this patient (12,15,18). Reduction in serum ferritin was consistent in all 3 cases, which suggests that reductions in oxidative stress also contributed to a more favorable metabolic environment, resulting in improvements in liver function. These improvements occurred at the higher NAC dosages, which suggests that requirements for NAC to promote increased GSH biosynthesis to counteract liver dysfunction may be higher in children with liver disease who receive PN. In summary, IV NAC supplementation is a safe adjunctive therapy in children with PN-induced liver disease. Acknowledgments The authors thank Mahroukh Raffi for the measurement of RBC GSH and Sahar Whelan, RPh, BScPharm, MScPharm, and Shawn Launder, RPh, BScPharm, for the preparation of IV NAC.