- Meet Consumers/Patients
|Newsletters: The Cleveland Clinic’s Experience with HPN-Associated Liver Disease|
The Cleveland Clinic’s Experience with Home Parenteral Nutrition Associated Liver Disease
Therese Austin, MS, RD, LD, CNSD; Douglas L. Seidner, MD, FACG, CNSP
Parenteral nutrition (PN) is a lifesaving treatment for patients with intestinal failure who are unable to ingest or absorb an adequate amount of nutrition, fluid and electrolytes from their diet. While the benefits of this therapy are obvious, the complex nature of providing intravenous nutrition is accompanied by a variety of problems, the most common being infections and mechanical complications associated with the vascular access device used to receive this therapy. Metabolic complications include fluid and electrolyte abnormalities which are usually associated with the underlying disease. One of the more serious metabolic complications is the development of liver disease. In this article, we will discuss the experience of the Home Parenteral Nutrition Service at the Cleveland Clinic. These results were recently published in the Journal of Parenteral and Enteral Nutrition, a professional publication of the American Society of Parenteral and Enteral Nutrition.
How Common Is Liver Disease?
Numerous studies have been conducted to investigate how many patients on long-term PN (usually defined as 6 months or more) develop liver test abnormalities and significant liver disease to help determine the cause of this problem. Some of these studies have shown an incidence of liver test abnormalities ranging from 25 to 100 percent and that advanced liver disease may occur in 15 to 40 percent of these patients.
The Home Parenteral Nutrition (HPN) program at the Cleveland Clinic is one of the largest HPN programs in the country. A computerized medical record has been used to help us manage consumers who require long-term PN. We conducted a study to investigate the prevalence of abnormal liver enzymes and advanced liver disease in an attempt to identify risk factors that might be associated with these abnormalities. The study was done by querying our computerized medical record for consumers who had received HPN for at least 6 months from July 1991 through June 2002.
For the purpose of our study we classified liver tests into two main categories; either liver associated enzymes (LAEs) or liver function tests (LFTs). LAEs include aspartate aminotransferase (AST), which reflectsinflammation or injury to hepatocytes (one of the major cells within the liver) and alkaline phosphatase (AP), which reflects inflammation or obstruction of the biliary system of the liver. We categorized the degree of abnormality for these labs as mild, moderate or severe depending on the degree of elevation (see Table 1 - click here for a larger version). LFTs were defined as blood tests that reflect the metabolic capacity of the liver and include bilirubin, albumin and prothrombin time. These labs were combined to determine overall liver function. This measure was only felt to be important when the bilirubin was over 3 mg/dL, the albumin was less than 3.2 mg/dL and the prothrombin time was more than 3 seconds prolonged. Patients with these results were defined as having severe liver dysfunction. (Note: these values vary slightly from table 1 because they are based on a Childs-Turcotte-Pugh score.) Patients with a known liver disease, drug induced liver injury, alcoholism and cancer, all of which can cause an elevation in LAEs, were excluded from the study. Patients with a transient elevation in LAEs were excluded by only studying patients who had lab abnormalities on more than one occasion over a period of at least 6 months. In addition to these laboratory results we gathered information on the patient’s age, gender, underlying disease, indication for PN and PN formula.
There were a total of 162 patients that constituted the study group. The clinical characteristics of these patients are shown in Table 2 (click here for a larger version). The average duration of PN in these patients was 2.14 years. While this may seem relatively brief, this reflects the fact that many patients are able to discontinue PN within one year. Many patients were on PN for more than two years and one patient was on therapy for more than 10 years. We found that 92 (57%) patients had chronically abnormal LAEs (AST and/or AP), 161 (99%) had abnormal LFTs (bilirubin, albumin and/or PT) and only one patient had normal liver tests. Closer inspection on an individual basis revealed that most abnormal liver function tests were in the mild or moderate range. The high rate of LFT abnormalities reflects the fact that one of the measures used to define LFTs is albumin, which is also a marker of nutritional status. When albumin was removed from our definition of LFTs the rate of abnormality dropped to 57%.
Despite this seemingly high rate of liver test abnormality, only seven of the 162 (4.3%) patients had severe liver dysfunction. After reviewing the records of these patients we found only one instance where severe liver dysfunction could be attributed solely to the use of PN. We also found that female gender was associated with a greater likelihood of developing liver failure. Because only seven patients developed severe liver dysfunction and six happened to be women, it is possible that the association between gender and liver failure was by chance alone and not a true result.
What to Look For
The study also assessed individual patient’s PN formula. They received an average of 24.7 total calories per kilogram per day and 18.5 calories per kilogram per day of dextrose. Approximately 10% of total energy was supplied as intravenous (IV) lipid emulsion. It is our practice to provide patients with IV lipid only one or two times per week to meet their essential fatty acid requirement. Daily IV lipid emulsion is reserved for patients with diabetes or elevated blood sugar levels that are not easily controlled by adding regular insulin to the PN formula. In our study there was a trend for a higher dose of PN calories, dextrose calories and a greater duration of PN to be associated with worsening liver enzyme abnormalities. This observation was not statistically significant; however, other studies have found an association between these factors and liver failure. It is possible that these factors did not reach statistical significance because our patients were not exposed to PN for as long as patients in these other studies.
There are several theories as to why patients on PN have abnormal liver associated enzymes or liver function tests. It may be due to a nutrient deficiency or toxicity in the PN formula.
A list of nutrient deficiencies and toxicities that may lead to liver disease is shown in Table 3 (click here for a larger version). Other possibilities include small bowel bacterial overgrowth that is associated with extensive bowel resection and mucosal disease; patients with short bowel syndrome; the presence of an underlying inflammatory disorder such as Crohn’s disease, cancer or pancreatitis as well as patients who are unable to take any nutrition by mouth. Another theory is that frequent sepsis from the patient’s catheter or another source may lead to abnormal liver tests.
When a patient does have a liver function test abnormality it is important for the physician managing the PN formula to diagnose and treat any other causes of hepatobiliary disease such as gallstones, viral or autoimmune hepatitis, Wilson’s disease, and hemochromatosis. A review of a patient’s current list of medications should also be conducted to eliminate any that may be toxic to the liver. This includes H2 antagonists such as famotidine (Pepcid) or ranitidine (Zantac), which are commonly used to reduce fluid loss in high output short bowel patients, and should be discontinued if there is significant liver dysfunction. (Note: The authors have never found that withdrawal of H2 antagonists has helped and always resume the therapy because of patient need.) Another medication to consider eliminating is octreotide (Sandostatin) which is used to inhibit gastric, pancreatic or biliary secretions and intestinal motility. PN consumers should also not drink alcohol in excess.
If no reversible cause for the abnormal liver tests are found, the managing physician should follow the labs closely and consider referral to a liver specialist when they remain elevated for more than six months, or if they are elevated to a moderate or greater degree (see Table 1). A rapid increase in liver tests may necessitate the withdrawal of PN and the provision of regular intravenous fluids to maintain hydration until the cause of the abnormal liver test can be determined. Patients with progressive liver dysfunction or liver failure should discuss with their physician the need for referral to a transplant center where both small bowel and liver transplantation can be performed. Hopefully future studies will help better define the cause of PN associated liver disease so that physicians can better manage and treat this condition.