Hypertransaminasemia in patients with negative viral markers

A. Cuadrado and J. Crespo

Service of Digestive Diseases. University Hospital Marqués de Valdecilla. Santander. Spain

 

---

Cuadrado A, Crespo J. Hypertransaminasemia in patients with negative viral markers. Rev Esp Enferm Dig 2004; 96: 484-500.

---

Recibido: 13-11-03.
Aceptado: 19-11-03.

Correspondencia: Javier Crespo. Servicio de Aparato Digestivo. Hospital Universitario Marqués de Valdecilla. Avda. de Valdecilla, s/n. 39008 Santander.
e-mail: javiercrespo@teleline.es

 

INTRODUCTION

The inclusion of aminotransferase and alkaline phosphatase (AP) determination within routine laboratory tests for patients in both primary care and hospital care has brought about an increase in the detection of symptomatic-stage liver disturbances.

Viral hepatitis is the most common cause of aminotransferase elevation, and accounts for more than 90% of acute hepatitis. Diagnosis is easily confirmed through serologic immune marker determination and even genetic techniques. However, a small percentage of viral hepatitis remain undetected by these methods. On the other hand, other causes that may induce transaminase elevation include alcohol, and metabolic, toxic, autoimmune, infectious, cholostatic or endocrine disturbances, many of which may respond to either specific or symptomatic treatment.

PRACTICAL ASPECTS OF TRANSAMINASES

Not less than 60 transamination reactions have been detected in the liver, but only aspartate aminotransferase or glutamic-oxalacetic transaminase (ASAT or GOT) and alanine aminotransferase or glutamic-pyruvic transaminase (ALAT or GPT) are of clinical value. None of these enzymes are liver-specific, and they are widely distributed throughout the body. GOT is primarily found in the heart, liver, skeletic muscle, and kidney, with lesser concentrations in the heart and skeletic muscle (1). Highest GPT levels are found in the liver, and are therefore more specific as liver damage markers. GPT is an exclusively cytoplasmatic enzyme, while both mitochondrial and cytoplasmatic forms of GOT have been encountered in all cells. Thus, liver disease is the most important cause of increased GPT activity, and a common cause of increased GOT activity.

In most types of liver disease GPT activity is higher than GOT's. Alcoholic hepatitis is an exception to the rule, and this may be accounted for by a number of reasons: alcohol increases plasmatic GOT activity in contrast with other hepatitis forms; most forms of liver damage decrease liver cell activity for both GOT forms, whereas alcohol reduces only cytosolic activity; pyridoxine deficiency, which is common in alcoholics, reduces GOT activity, and eventually alcohol induces mitochondrial GOT release from cells with no apparent damage (1,2).

Both aminotransferases are usually present in the serum at low concentrations, below 30 to 40 U/L, and their normal range oscillates amongst laboratories (2). Although unusual in clinical practice, gender-specific and age-specific upper limits are recommended for reference ranges (GOT and GPT activities are significantly higher in males versus females), specifically regarding children and adults older than 60 years (among whom differences may be higher when compared to the relative homogeneity found within the 25-60-year group) (1,2).

Abnormal aminotransferase results -usually <2-fold increases above normal upper limits- must prompt repeat tests before an study may be initiated, since they will probably be back to normal. Various factors may alter GOT and GPT activity besides liver damage (1) (Table I). In this study, strenuous exercise or intense work-out may result in relevant transaminase increase; an increased body mass index may account for up to 40-50% increases in transaminase levels. Other causes such as muscular conditions or the presence of hemolytic anemia may significantly increase primarily GOT. Finally, differences induced by variability regarding time of day when samples are collected or between different days may also exist.

On the other hand, liver cell necrosis is not mandatory for aminotransferase release. In fact, a poor relationship exists between liver damage extent and aminotransferase level (2). Furthermore, the magnitude of transaminase elevation is not predictive of acute hepatitis outcome (3).

The pattern of transaminase increase may suggest etiology (as already noted, a greater increase of GOT versus GPT may indicate alcoholic hepatitis); similarly, the magnitude of their increase may suggest an acute or chronic condition. In this sense, in the presence of a 10-fold or greater increase of GPT above the upper reference limit an acute liver injury may be diagnosed, with 90% of cases resulting from viral acute hepatitis - although other causes should be investigated (1,4) (Table II). In contrast, mild to moderate hypertransaminasemia, below 10 times the upper limit of the normal range, is more suggestive of chronic liver injury (Table II). Lastly, a determination of other liver enzymes such as alkaline phosphatase (AP) and gamma-glutamyltransferase (GGT) may be useful to provide guidance on the condition's etiology (for a cholestatic condition, for instance). Figures 1A and 1B outline the diagnostic approach to transaminase increase based on increase patterns.

CAUSES OF HYPERTRANSAMINASEMIA IN PATIENTS WITH NEGATIVE VIRAL MARKERS

Table II lists the most common causes of acute and chronic transaminase elevation. Some of the most common non-viral causes of hypertransaminasemia are discussed below, with emphasis on the most relevant aspects pointing towards either disease.

Alcoholic liver disease

This is the most common cause of cirrhosis in the Western world. The spectrum of alcohol-related liver disease ranges from fatty liver (steatosis) to cirrhosis, of which steatohepatitis and fibrosis are intermediate stages (5). From 20 to 40% of severe, chronic drinkers will develop advanced liver disease. Realizing that from a histologic standpoint advanced liver disease may be a symptom-free condition with minimal -if any- liver enzyme disturbance is important. History taking and physical examination are essential for diagnostic guidance. How-ever, this may be supplemented by observing a GOT:GPT ratio of at least 2:1. Thus, a correlation between histologic findings corresponding to alcoholic liver disease and the presence of GOT:GPT ratios equal to or greater than 2:1 exists (2).

In the absence of other conditions, the concentration of aminotransferases in patients with alcoholic hepatitis is moderate, usually below 300 IU (6), and its correlation to disease severity is poor.

Furthermore, GGT determination may help in the diagnosis of alcohol abuse. GGT is found in liver cells and bile-duct epithelial cells. It is a really sensitive though poorly specific marker of liver disease that may be risen in pancreatic conditions, acute miocardial infarction, renal failure, diabetes, hyperthyroidism, chronic obstructive pulmonary disease, and alcoholism. Exposure to drugs and other substances inducing mitochondrial hyperactivity (enzymatic induction) such as a number of insecticides, toxics such as alcohol, and drugs such as phenobarbital, phenitoin and carbamazepine brings about significant rises in GGT concentration. Therefore, GGT has a poor positive predictive value (32%) for liver disease (1). Nevertheless, the presence of 2-fold increased GGT concentrations above the upper limit of normality in patients with a GOT:GPT ratio of at least 2:1 is strongly suggestive of alcohol abuse. Significant overlap exists with viral hepatitis and postnecrotic cirrhosis in the range from 1 to 2.

Thus, the presence of moderate hypertransaminasemia (≤ 300 U/L of GPT and GOT, a GOT:GPT ratio ≥ 2, and GGT above 2 times the upper limit of the normal range, together with the presence of macrocytic anemia and leukocytosis with 12,400 cells per mm³ on average are parameters strongly suggestive of alcoholic liver disease. On the other hand, bilirubin concentration and prothrombin time in seconds are factors predictive of alcoholic liver disease severity, and are a part of Maddrey Index or “discriminating function”, which identifies patients with significant short-term mortality (7):

Discriminating function = 4.6 x [PT (seconds) - control PT] + Bilirubin (mg/dL)

A value equal to or greater than 32 predicts high short-term mortality.

More recently the MELD (Model for End-stage Liver Disease) model has been introduced because of its usefulness as a predictive model for survival in patients with chronic liver disease in various circumstances. Thus, for instance, it has been used for prognostic assessment in patients with acute alcoholic hepatitis, where it may predict mortality as much effectively as Maddrey's discriminating index (8,9). This model uses serum bilirubin concentration and prothrombin time INR (International Normalized Ratio) as variables, with values above 18 being representative of high risk with a median survival below 3 months.

Liver disease from fat deposition: non-alcoholic liver steatosis and steatohepatitis

Non-alcoholic liver disease from fat deposition is a condition involving 10 to 24% of the general population with an increasing prevalence that rises to 58-75% in obese individuals (10). It is the most common enzymatic disturbance of the liver in the USA, underlying in up to 66-90% of asymptomatic hypertransaminasemias (10,11). It includes a range of lesions oscillating between steatosis and steatohepatitis, fibrosis and cirrhosis. Obesity is the primary etiologic factor in this liver condition, followed by diabetes mellitus and hyperlipemia in frequency order, but other causes such as abdominal bypass surgery for morbid obesity such as gastroplasty and intestinal bypass, drugs, total parenteral nutrition, etc., have been described as well.

The most commonly found -sometimes the only- laboratory disturbance is a mild to moderate (usually inferior to 4 times the upper limit of normality) increase of GOT, GPT or both, with a ratio (GOT:GPT) usually below 1, even though this increases with fibrosis progression (10). In many patients, GGT and AP are also increased, but to an extent below that encountered in alcoholic hepatitis. Imaging techniques (ultrasonography, CT) may be suggestive of fatty deposits, and magnetic resonance allows a quantitative assessment of fat infiltration in the liver. Once other potential causes of liver disease have been excluded, which requires that a minimum use of 20 g alcohol per day in women and 30 g alcohol per day in men be also ruled out, a clinical suspicion of disease and disease severity may only be confirmed using liver biopsy, which will also provide prognostic information (10). This gives greater prognostic information in patients older than 45 years of age with obesity or type 2 diabetes mellitus, or data suggestive of advanced liver disease (GOT:GPT ratio >1, cytopenia, skin stigmata, etc.), which are indicative of greater progression towards fibrosis, reason why the indication of liver biopsy is usually restricted to these groups (10,12,13).

Autoimmune hepatitis

Autoimmune hepatitis is a liver inflammation of unknown etiology with an incidence of 1.9 per 100.000 among northern European caucasians (14). It typically develops in young or middle-aged women (ratio 3.6:1 females:males), although all age groups are susceptible. Although liver biopsy is essential for diagnosis, severity assessment and need-to-treat determination, diagnostic criteria suggesting the disease require that other causes of liver disease (viral, alcoholic, toxic, etc.) be excluded, the presence of autoantibodies -conventional markers include antinuclear antibodies (ANA), anti-smooth muscle antibodies (AML) and anti-liver/kidney microsome type 1 antibodies (antiLKM1)-, the presence of hypergammaglobulinemia -present in more than 80% of patients, which is more suggestive of disease when elevations are greater than 2-fold above the normal upper limit of plyclonal immunoglobulins- and a predominant aminotransferase rise (14). Acute disease onset is common (40%), and a fulminant presentation may develop that is characterized by liver encephalopathy within 8 weeks from disease onset. Autoantibodies are neither patognomonic nor specific for this disease, and their titration only reflects the intensity of immune response; they cannot be used to assess treatment response, and so they need no monitoring.

Finally, the extent of aminotransferase increase, specifically GOT, provides guidance for treatment indication, and treatment is absolutely indicated for increases ≥ 10 times the upper limit of normality -even ≥ 5 times- and gammaglobulin levels ≥ 2 times the normal value (14); it also provides good guidance to assess treatment response.

Pharmacotoxicologic disease

Drugs may affect the liver via several mechanisms, and liver drug-induced reactions may mimic any liver disease; this is why their manifestations are variable and range from asymptomatic reversible hypertransaminasemia to fulminant liver failure (15). Liver lesions caused by drugs have been associated to more than 800 drugs, and have been estimated to be responsible for 1 in every 600-3500 hospital admissions (16), thus representing 10 to 15% of fulminant hepatitis cases. Given their absence of specific clinical, laboratory, and pathological characteristics, the diagnosis of drug-induced liver toxicity is based on clinical suspicion, evidence of exposure to drugs, and the exclusion of other causes of liver disease (17). The chronologic relationship between treatment onset and liver lesion resolution are a most relevant aspect of diagnosis (18). A previous period of three months is considered the recommended time deadline for investigation, even though the latency period between toxic ingestion and symptom development is very variable (17); a latency period shorter than 1 week is consistent with a hypersensitivity reaction, whereas longer periods are consistent with cumulative toxic metabolites. Occasionally, symptoms may develop even weeks after treatment discontinuation, as is the case with the administration of amoxicillin-clavulanic acid (19). The presence of extrahepatic symptoms such as skin rash, eosinophilia or other organ involvement may suggest a drug-induced adverse reaction.

A number of specific algorithms for an estimation of causality in cases of drug-induced liver lesion have been developed, and the Council for the International Organization of Medical Sciences (CIOMS) scale is most widely accepted to assess causality in a comparative analysis versus other methods (20). Basically, this scale assesses 7 variables: time from treatment onset to adverse reaction initiation, outcome after treatment discontinuation, risk factors, presence of concurrent ther-apies, presence of causes other than drugs, previous understanding of the drug's hepatotoxic potential, and response to the drug's readministration. Depending on the values assigned to each variable, a score develops in such a way that the sum of all scores results in a final score related to the causality that is attributable to a given drug.

Hemochromatosis

This is the most common genetic disease amongst Caucasians. Although of worldwide distribution, it mainly involves northern European individuals (between 1/200 and 1/400 of individuals). It is of autosomal recessive inheritance, and the gene for hemochromatosis (HFE) is located on the short arm of chromosome 6, hemochromatosis phenotype being primarily linked to the presence of two mutations: C282Y and H63D. In this respect, the homozygous state in which both chromosome 6 alleles have mutation C282Y (found in more than 90% of patients with hemochromatosis), and the heterozygous composite state in which one chromosome has mutation C282Y and its counterpart has mutation H63D (found in 3-5% of cases) are the main genetic abnormalities found in patients with a hemochromatosis phenotype (21). However, it is fairly likely that genes other than HFE may bring about iron deposition in a way similar to hemochromatosis, since a varying percentage of hemochromatosis cases -between 0% and 30% in the various series- have no HFE gene mutations despite being phenotypically similar to positive patients. Furthermore, it should be born in mind that a complete phenotypical expression (progressive tissue iron overload) of said genetic disturbance -particularly of C282Y homozygosis- only ever occurs in 58% of patients (21).

HFE gene-associated hemochromatosis is characterized by increased gastrointestinal iron absorption and the ensuing iron deposition in the liver, heart, pancreas, other endocrine organs, joints and skin (22). Hemochromatosis must be suspected in the presence of persistent hypertransaminasemia, disturbed iron parameters, chronic asthenia, joint pain of undetermined origin, diabetes mellitus, myocardiopathy and arrhythmias, or infertility and impotence, among other clinical evidence.

Hemochromatosis screening must start by determining iron metabolism parameters. Specifically, transferrin saturation index (TSI: serum iron / transferrin or total iron binding capacity x 100%) and ferritin are the most informative parameters, and they are both altered in symptomatic hemochromatosis. However, TSI is the earliest phenotypical marker to become altered, and may be elevated even with still normal ferritin levels in younger individuals (23). Population-based studies have shown that approximately 36% of C282Y homozygotes have a TSI ≤ 45%, and whether or not they will develop symptomatic iron overload cannot be forecast (23). On the other hand, ferritin may be abnormal in approximately 50% of patients with alcoholic liver disease, non-alcoholic steatohepatitis or chronic hepatitis C in the absence of hemochromatosis, as well as in other inflammatory and neoplastic conditions. Should TSI be ≥ 45 in two fasting determinations, a genetic study must be initiated to establish the existing mutation type. This study should be directly performed -regardless of TSI- in adult first-degree relatives of patients with hemochromatosis (21). In C282Y mutation homozygotes a diagnosis is firmly established and biopsy may only be recommended -solely for predictive purposes- in patients older than 40 years and/or with ferritin >1000 or some evidence of liver involvement (e. g., hepatomegaly and/or altered liver enzymes) with increased risk for fibrosis and cirrhosis (21,24). In addition to prognostic value in the assessment of liver damage and the staging of potential cirrhosis, liver biopsy is of diagnostic value primarily in patients with a characteristic phenotype but none of the mutations potentially responsible for the disease. The most significant parameter from biopsy performance is hepatic iron index (HII), which represents iron concentration in dry liver tissue as divided by patient age in years. An HII ≥ 1.9 has been considered virtually diagnostic for hemochromatosis, even though a significant number of C282Y homozygotes are known to have an HII below 1.9, whereas C282Y-negative patients with advanced liver disease may have an HII above 1.9 (24). In patients exhibiting composed heterozygosis (C282Y/H63D) liver biopsy should also be considered.

Wilson's disease

Wilson's disease is an autosomal recessive condition involving copper metabolism, with excessive amounts of copper accumulating in various organs, particularly the liver and brain, whence it may result in clinical manifestations. Together with hemochromatosis, this is the only chronic liver disease with a highly effective treatment that may prevent severe neuropsychiatric sequelae from developing (25). When left untreated, it may end up in death at the age of 50. Its prevalence is 10 to 30 cases per million population. Its manifestations usually develop between 15 and 50 years of age, with most patients presenting with liver disease (usually earlier, at 10-13 years) that precedes neuropsychiatric complaints by 5-10 years, the latter developing at 19-20 years of age. Other clinical manifestations include hemolytic anemia, joint disease, and renal, heart, endocrine or skin involvement. Copper deposition within Descemet's membrane results in the so-called Kayser-Fleischer ring, which although non-specific for Wilson's disease does represent a characteristic finding of help in diagnosing the condition; it is commonly found in almost all patients with neurologic manifestations, but may be absent in a small percentage of cases with neuropsychiatric symptoms and in 15-50% of those with liver disease alone (25).

Although no single laboratory test is adequately sensitive and specific in reaching Wilson's disease diagnosis for all patients, serum ceruloplasmin below 20 mg/dL in a young individual (younger than 40 years) with altered transaminases represents a lead for a likely Wilson's disease. In fact, ceruloplasmin is the initial screening test for Wilson's disease (2). However, it is an acute phase reagent that may become altered in varying circumstances (malnutrition, inflammatory or infectious conditions, pregnancy, etc.), with 5-15% of patients exhibiting normal or even slightly low ceruloplasmin levels (25). In Wilson's disease, when hepatic copper storage capacity is exceeded, or when liver cell damage has already occurred, cell copper is released into the systemic circulation, and serum free copper (not bound to ceruloplasmin) rises, which facilitates copper deposition in tissues and urinary clearance. However, total serum copper (bound to ceruloplasmin) decreases as a result of reduced ceruloplasmin synthesis, and thus is scarcely of any value. Table III lists both normal and Wilson's disease-related chemical parameters. In the presence of clinical suspicion, and in the absence of abnormal ceruloplasminemia and Kayser-Fleischer ring, the presence of urinary copper clearance in 24-hour urine above 100 µg/day is suggestive of Wilson's disease. The diagnosis is usually confirmed by establishing copper concentration using atomic absorption spectrometry in a tissue sample obtained from liver biopsy. Patients with Wilson's disease have values above 250 µg/g of dry weight (normal, 15-55 µg/g).

Liver disease may present as active chronic hepatitis, cirrhosis or fulminant hepatitis. Wilson's disease must be considered in the differential diagnosis of any unexplained active chronic hepatopathy, particularly if the patient is younger than 40 years of age. A moderate transaminase increase usually does not reflect inflammation severity. It presents as fulminant hepatitis in some cases. Such patients usually exhibit disproportionately low transaminase and AP levels, while bilirubin may be increased from severe hemolytic anemia with a negative Coombs' test; GOT:GPT ratio also tends to be above 4.

Finally, although Wilson's disease gene -called the ATP7B gene- has been identified in chromosome 13, more than 200 mutations of this gene have been described, and genetic studies are currently limited to family screening, which is used together with traditional clinical and chemical studies (2,25).

Alpha-1-antitrypsin deficiency

Alpha-1-antitrypsin (A1AT) deficiency is a hereditary disorder transmitted in an autosomal codominant fashion that manifests with panacinar emphysema and paniculitis, and is a rare cause of chronic liver disease in the adult (2). Human A1AT is primarily synthesized in the liver and represents the main protease inhibitor. Two codominant alleles (designated using alphabetic characters according to electrophoretic mobility) are usually inherited, which are within the same locus -called Pi for protease inhibitor- on chromosome 14q32.1. Most relevant variants include M (Pi M), present in 90% of the population, and the deficient S and Z (Pi S and Pi Z). In all, 95% of subjects with acute A1AT deficiency have a ZZ phenotype, with prevelence varying from 1/1575 in Scandinavia to 1/5000 in Mediterranean countries. Up to 40% of adult ZZ homozygotes may develop liver disease that progresses to cirrhosis and liver cell carcinoma, and up to 15% of adult ZZ phenotype homozygotes as well as some heterozygotes may have liver cell carcinoma even in the absence of cirrhosis (26). Low A1AT levels as directly or indirectly detected by the absence of an a-globulin peak at protein electrophoresis represents a positive lead. However, A1AT concentration may be increased in response to inflammation, and an acute phase reagent would induce a false negative result. Thus, the diagnosis is reached using phenotypical determination (2).

Cholestatic diseases, autoimmune cholangiopathies

Although cholestasis primarily translates into increased AP and GGT through liver-cell enhanced synthesis and release facilitation from the cell membrane -as well as through increased bilirubin to greater or lesser extents- aminotransferase increase to some extent is surely no wonder. History taking, physical examination, and pattern of liver enzyme increase may provide guidance towards a biliary-pancreatic condition whose origin may easily be confirmed with imaging techniques such as abdominal ultrasounds or magnetic resonance.

The presence of primary biliary cirrhosis (PBC) must be suspected when a woman (90%) in her twenties or thirties complains of asthenia, pruritus, jaundice, and xanthomas or xanthelasmas, or -if symptom-free- has cholestasis (elevated GGT and, particularly, elevated AP by 2-20 times the normal value). Moderate cytolysis may exist, with 1- to 5-fold transaminase increases (27). The primary marker of PBC is the presence of antimitochondrial antibodies (AMA), which reach a high titration (>1:40) and whose sensitivity and specificity regarding PBC are above 95% (28). Antinuclear (ANA) and anti-smooth muscle (AML) antibodies may be found in one third of patients with PBC. Hypergammaglobulinemia is also a predominant serologic characteristic, and IgM has a tendency to become earlier and more consistently increased (27,29). The presence of negative AMA and high ANA and/or AML titers in the presence of a clinical, chemical and histopathologic picture compatible with PBC has been dubbed autoimmune cholangitis (AIC) (28). Overall, the clinical and pathogenic aspects of AIC are identical to those of classic (AMA-positive) PBC, so that these two conditions may only be told apart by their serologic profile (27).

Primary sclerosing cholangitis (PSC) is a chronic cholestatic disease characterized by a progressive process that both scleroses and obliterates bile ducts, which results in an alteration of extrahepatic and intrahepatic ducts, or of either component separately (30). It mainly affects males around the fourth decade of life, and is associated with inflammatory bowel disease, particularly ulcerative colitis, in 40-80% of cases, with progressive asthenia, pruritus and jaundice being the most characteristic signs and symptoms (31). Although an increase in cholestasis-related enzymes is characteristic, with AP peaking at 2-3 times its normal value, 92% of patients in some series has up to 3-fold aminotransferase increases above the normal range (31). Hypergammaglobulinemia is seen in 45%, and anti-neutrophil cytoplasmic antibodies (ANCA) -specifically perinuclear antibodies (p-ANCA)- are found in the serum of patients with PSC with or without ulcerative colitis in a proportion of about 65 to 85% (32). Finally, ANA and AML may be present in up to 55 and 33%, respectively, of cases.

Other causes

Ruling out the presence of celiac disease underlying chronic aminotransferase elevation is important, since this cause has been implied in up to 10% of unexplained hypertransaminasemias (33,34); thus, a determination of anti-celiac antibodies is recommended in such circumstances (2).
Transaminase -mainly GOT- elevation may occur in disorders involving organs and tissues other than the liver, with skeletic muscle being most common (28). These disturbances may range from congenital disorders of muscle metabolism to acquired disorders such as polymiositis, including “use” disturbances such as strenuous exercise disorders. In case of a muscle-based disorder of liver enzymes, a determination of increased muscle enzymes such as creatine-kinase and aldolase is of diagnostic value.

Thyroid hormone determination may reveal hyperthyroidism as the cause of transaminase disturbance; similarly, a consistent clinical history together with increased urinary porphyrins may unveil the presence of liver porphyria.

If, in spite of all studies performed, the cause of transaminase increase may not be identified, liver biopsy should be considered. The criterion to decide on whether a biopsy should be undertaken is based on transaminase elevation extent (2,13). In this sense, when the increase is below 2 times the upper reference range limit, only follow-up is advised; if, on the contrary, the increase is equal to or greater than 2-fold in a persistent way, a biopsy should be performed, which will exclude severe liver disease; however it will neither provide relevant diagnostic information nor condition changes in patient management.

 

REFERENCES

1. Dufour DR, Lott JA, Nolte FS, et al. Diagnosis and monitoring of hepatic injury. II. Recommendations for use of laboratory tests in screening, diagnosis, and monitoring. Clin Chem 2000; 46: 2050-68.         [ Links ]

2. Pratt DS, Kaplan MM. Evaluation of abnormal liver-enzyme results in asymptomatic patients. N Engl J Med 2002; 342: 1266-71.         [ Links ]

3. Davern II TJ, Scharschmidt BF. Biochemical liver tests. En: Feldman M, Scharchmidt BF, Sleisenger MH, eds. Gastrointestinal and Liver Disease. Philadelphia: W.B. Saunders, 1998. p. 1112-22.         [ Links ]

4. Craxi A, Almasio P. Diagnostic approach to liver enzyme elevation. J Hepatol 1996; 25 (Supl. 1): 47-51.         [ Links ]

5. Stewart SF, Day CP. The manegement of alcoholic liver disease. J Hepatol 2003; 38: S2-S13.         [ Links ]

6. Maher JJ. Alcoholic liver disease. En: Feldman M, Scharchmidt BF, Sleisenger MH, eds. Gastrointestinal and Liver Disease. Philadelphia: W.B. Saunders, 1998. p. 1199-214.         [ Links ]

7. Maddrey WC, et al. Corticosteroid therapy of alcoholic hepatitis. Gastroenterology 1978; 75: 193-9.         [ Links ]

8. Malinchoc M, Kamath PS, Gordon FD, et al. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology 2000; 31: 864-71.         [ Links ]

9. Vargas V, Ortiz M. Modelos pronósticos en la cirrosis hepática. El modelo MELD. Gastroenterol Hepatol 2003; 26 (4): 257-9.         [ Links ]

10. Angulo P. Nonalcoholic fatty liver disease. N Eng J Med 2002; 346 (16): 1221-31.         [ Links ]

11. Daniel S, Ben-Menachem T, Vasudevan G, et al. Prospective evaluation of unexplained chronic liver transaminase abnormalities in asymptomatic and symptomatic patients. Am J Gastroenterol 1999; 94: 3010-4.         [ Links ]

12. García-Monzón C. Esteatohepatitis no alcohólica. Gastroenterología y Hepatología 2001; 24 (8): 395-402.         [ Links ]

13. Rodríguez C, Martín L. Estudio del paciente con elevación de transaminasas. GH continuada 2002; 1: 345-8.         [ Links ]

14. Czaja AJ, Freese DK. Diagnosis and treatment of autoinmune hepatitis. Hepatology 2002; 36(2): 479-93.         [ Links ]

15. Lewis JH. Drug-induced liver disease. Med Clin North Am 2000; 84: 1275-311.         [ Links ]

16. Dossing M, Sonne J. Drug-induced hepatic disorders. Incidence, management and avoidance. Drug Saf 1993; 9: 441-9.         [ Links ]

17. Larrey D. Drug-induced liver diseases. J Hepatol 2000; 32 (Supl. 1): 77-8.         [ Links ]

18. Fernández-Bermejo M, Robledo P, Mateos JM. Hepatotoxicidad por fármacos: diagnóstico y tratamiento. Gastroenterología Práctica 2003; 12: 4-6.         [ Links ]

19. Andrade RJ, Lucena MI, García-Escaño, et al. Hepatotoxicity in patients with cirrhosis, an often unrecognized problem. Lessons from a fatal case related with amoxicillin/clavulanic acid. Dig Dis Sci 2001; 46: 1416-9.         [ Links ]

20. Lucena MI, Camargo R, Andrade RJ, et al. Comparison of two clinical scales for causality assessment in hepatotoxicity. Hepatology 2001; 33: 123-30.         [ Links ]

21. Tavill AS. Diagnosis and management of hemochromatosis. Hepatology 2001; 33 (5): 1321-8.         [ Links ]

22. Harrison SA, Bacon BR. Hereditary hemochromatosis: update for 2003. J Hepatol 2003; 38: S14-S23.         [ Links ]

23. Bacon BR. Hemochromatosis: diagnosis and management. Gastroenterology 2001; 120: 718-25.         [ Links ]

24. Pardo A, Quintero E. Hemocromatosis hereditaria: diagnóstico y tratamiento. Gastroenterología Práctica 2002; 11 (3): 32-6.         [ Links ]

25. Pérez-Aguilar F. Enfermedad de Wilson: consideraciones fisiopatológicas, clínicas y terapéuticas. Gastroenterol Hepatol 2003; 26 (1): 42-51.         [ Links ]

26. Cosme A, Ojeda E, Torrado J, et al. Alteraciones hepáticas por déficit de alfa-1-antitripsina en adultos. Estudio de 5 pacientes y análisis de los casos publicados en la bibliografía española. Gastroenterol Hepatol 2002; 26 (4): 251-6.         [ Links ]

27. Moreno R, García-Buey L. Diagnóstico y tratamiento de las colangiopatías autoinmunes. Gastroenterol Hepatol 2003; 26 (Supl. 2): 11-5.         [ Links ]

28. Heathcote EJ. Management of primary biliary cirrhosis. Hepatology 2000; 31 (4): 1005-13.         [ Links ]

29. Lindgren S, Eriksson S. IgM in primary biliary corrhosis. Physicochemical and complement activating properties. J Lab Clin Med 1982; 99: 636-45.         [ Links ]

30. Wiesner RM, LaRusso NF. Clinicopathologic features of the syndrome of primary sclerosing cholangitis. Gastroenterology 1980; 79: 200-6.         [ Links ]

31. Jorge AD, Jorge OA. Colangitis esclerosante primaria. Revis Gastroenterol 2002; 4: 196-218.         [ Links ]

32. Vidrich A, Lee J, James E, et al. Segregation of pANCA antigenic recognition by Dnase treatment of neutrophils: ulcerative colitis, type 1 autoinmune hepatitis and primary sclerosin cholangitis. J Clin Inmunol 1995; 15: 239-9.         [ Links ]

33. Volta U, De Franceschi L, Lari F, et al. Coeliac disease hidden by cryptogenic hypertransaminasaemia. Lancet 1998; 352: 26-9.         [ Links ]

34. Bardella MT, Cecchi M, Conte D, et al. Chronic unexplained hypertransaminasemia may be caused by occult celiac disease. Hepatology 1999; 29: 654-7.         [ Links ]