PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Hepatology. Author manuscript; available in PMC 2010 June 7.
Published in final edited form as:
PMCID: PMC2881483
NIHMSID: NIHMS116247

Evaluation of the Patient with Hepatitis B

Abstract

The initial evaluation of a patient with hepatitis B virus (HBV) infection should attempt to assess the disease activity and stage in the context of the known natural history of this infection and to properly assess the needs for treatment and surveillance. In addition to a medical history and focused physical examination, the initial evaluation usually requires serological, biochemical and virological tests to confirm the diagnosis as well as an imaging study to establish a baseline for future monitoring. A liver biopsy is generally not needed but can provide useful information on prognosis, need for surveillance for hepatocellular carcinoma (HCC) and whether to recommend therapy.

Follow-up monitoring is aimed at determining disease progression, development of complications and reassessing the need for treatment. Monitoring frequency should be determined based upon the activity and stage of disease. Initiation of screening for hepatocellular carcinoma (HCC) should be based on age, race, sex, family history and stage and duration of disease. The current recommended method of screening and surveillance for HCC is by ultrasonography and alpha-fetoprotein measurements every 6–12 months.

Prospective studies are needed to evaluate the role of longitudinal application of non-invasive assays of fibrosis, such as serum fibrosis markers and transient elastography. Better biomarkers and imaging modalities are needed for early detection of HCC. Finally, studies are needed to better refine the indications and to balance the risks and benefits of antiviral therapy.

Keywords: hepatitis B, fibrosis, cirrhosis, hepatocellular carcinoma, liver biopsy, antiviral therapy

Initial evaluation

The initial evaluation of the patient found to have hepatitis B surface antigen (HBsAg) in serum should attempt to establish the diagnosis and assess the grade (activity) and stage of disease in relationship to the complex and dynamic natural history of hepatitis B. Hepatitis B virus (HBV) infection evolves through several phases, the pattern of which may vary by viral strain, mode of infection, ethnicity and presence of co-morbidities (1, 2). These phases include (1) an early, immune-tolerant phase that typically occurs in children and is marked by high levels of HBV DNA, presence of HBeAg, but normal serum aminotransferase levels and absence of disease activity; (2) immune activation phase of chronic hepatitis B, marked by moderate to high levels of HBV replication, continued or fluctuating disease activity and aminotransferase elevations and during which progressive liver damage can occur; and (3) an inactive carrier state, marked by low levels of HBV replication, absence of HBeAg and normal serum aminotransferase levels with inactive liver disease. Chronic hepatitis B is a dynamic disease, and patients can pass from one phase to another and back, from disease activity to inactivity. During periods of activity, progressive liver damage can occur, resulting in cirrhosis and its complications, as well as hepatocellular carcinoma (HCC). Thus, the presence of HBsAg in serum may have a range of implications, indicating acute or chronic hepatitis B that can be mild or severe and progressive. The goals of the initial evaluation should be to determine the phase of HBV infection, the severity of disease, and the need for monitoring and treatment.

An initial evaluation should include a thorough medical history with specific focus on possible modes infection and their timing. Important elements of the history should be any exposure to blood or blood products, injection drug abuse, sexual exposure, occupational exposure, and family history of hepatitis B (3). If acute hepatitis B is suspected, recent history of medical procedures should be elicited, particularly if done in outpatient surgical suites (4). Immigrants from countries with a high prevalence of HBV infection should be questioned about family history of hepatitis, cirrhosis or liver cancer as well as about other exposure modes such as ritual scarification, piercing, and traditional or unsafe medical procedures (5).

The time of onset of infection in patients presenting with chronic hepatitis B is often difficult to define. Most patients with chronic hepatitis B do not report a history of acute, symptomatic onset, and the timing of exposures may be unclear. A history of blood donation can be useful in that all blood donations in the United States have been screened for HBsAg since 1972 and for antibodies to hepatitis B core (anti-HBc) since 1987. As in all patients with liver disease, a detailed medication history including past and present medications with specific attention to over-the-counter, herbal and alternative medications should be elicited and personal habits regarding smoking, alcohol consumption, drug abuse and sexual activity need to be documented. A history of previous treatment for HBV infection is important, and in patients coinfected with HIV, the exact anti-retroviral regimens used in the present and past should be determined, as these can promote viral resistance and affect treatment choices.

The initial physical examination should document the presence or lack of signs of chronic liver disease. For most patients with chronic hepatitis B, the physical examination is normal. Findings such as jaundice, hepatomegaly or a shrunken liver, splenomegaly, palmar erythema, spider angiomata, Dupuytren’s contracture, gynecomastia, ascites, peripheral edema, caput medusae, or asterixis are indicative of advanced disease. Signs of other disorders that could affect treatment decision (such as heart failure, renal function impairment, or thyroid disease) should be sought, as well as evidence of extrahepatic manifestations of HBV infection (rash, arthritis and neuropathy).

Laboratory tests on the initial visit (Table 1) should include a chemistry panel of serum enzymes, direct and total bilirubin, albumin, total protein and creatinine, complete blood count, coagulation tests and alpha-fetoprotein. HBV markers including HBsAg, hepatitis B e antigen (HBeAg) and antibody (anti-HBe) and HBV DNA will confirm the diagnosis and help determine the phase of disease (Table 2). Repeating the test for HBsAg is useful, as patients are sometimes told that they have hepatitis B based upon the presence of antibody to HBsAg (anti-HBs) or anti-HBc alone. Tests for anti-HBs and anti-HBc are usually not needed, although IgM anti-HBc testing may be appropriate if acute hepatitis B is considered possible.

Table 1
Laboratory and imaging studies on initial visit
Table 2
Interpretation of serologic testing

Chronic hepatitis B cannot be correctly classified with reliability on the basis of a single assessment, and serial testing for serum enzymes and HBV DNA is almost always helpful. This is particularly true for HBeAg-negative chronic hepatitis B, which is often marked by a fluctuating course of serum aminotransferase levels in response to fluctuating levels of HBV DNA (2). A single normal ALT value is often thought to imply an inactive carrier or immune tolerant state and can affect the decision to perform a liver biopsy or to initiate therapy. However, a single normal level cannot be viewed in isolation. Serial testing should be done before it is assumed that the ALT levels are persistently normal; furthermore a normal ALT associated with an elevated AST level, elevated serum bilirubin or decreases in serum albumin values may be indicative of advanced disease and even cirrhosis. It is also important to interpret assay results within the context of the assay characteristics. For example, with the increasing sensitivity of HBV DNA assays, some patients who were previously thought to be HBV DNA negative can be shown to have low-levels of viremia (6, 7). Similarly, the absolute level of ALT may be more important than its relation to the local normal range, as values in the high end of the currently accepted range are associated with a higher risk of histological damage and disease progression (8, 9), and probably should not be accepted as being fully normal in patients presumed to be carriers or immune-tolerant.

Other laboratory studies are important in defining the severity of disease and can provide clues for the presence of fibrosis or advanced liver disease. Laboratory evidence of hepatic decompensation such as low albumin, prolonged prothrombin time, and increased direct bilirubin should be sought. An elevated level of alpha fetoprotein can also hint at advanced disease. Although less accurate than a liver biopsy (see below) in determining actual histological stage, some common laboratory tests can be useful for detecting advanced disease or minimal disease (Table 3). Testing for these “red flags” or “danger signals” should be done routinely at initial evaluation. Additionally, testing for coinfection with hepatitis C virus, hepatitis delta virus and human immunodeficiency virus is mandated, as these significantly affect therapy and prognosis.

Table 3
Markers of advanced liver disease

Testing for the genotype of hepatitis B may provide some information on the expected natural history of the disease and on the risk for liver cancer. However, in most scenarios, this assay is not required in clinical practice and should be reserved for selected patients and research protocols. Similarly, testing for mutations conferring resistance to antiviral agents is not applicable to the initial evaluation in routine practice.

An imaging study should be included in the initial evaluation, preferably an abdominal ultrasound. The operator should be directed to assess for signs of advanced liver disease (such as irregular liver contour, splenomegaly, ascites, abnormal portal vein flow) and for liver masses. This will also serve as a useful baseline for future imaging studies, as part of the screening regimen for HCC.

The role of a liver biopsy as part of the evaluation of the patient with HBV is controversial. There is no absolute indication for a biopsy; however, a biopsy is still the gold standard for assessing necroinflammatory activity (grade) and degree of fibrosis (stage) and serves as a useful baseline for future follow-up. The most important role for the biopsy is to assist in deciding on the need for antiviral therapy. Thus, if a biopsy result is going to affect treatment decision, it is clearly indicated. This is especially true in patients with low or normal liver enzymes, in which a biopsy can differentiate active, progressive disease from the inactive carrier or immune-tolerant state. On the other hand, if the decision to treat is obvious prior to a liver biopsy, then the biopsy serves only as a baseline assessment. The specific indication should be discussed with the patient when the recommendation for a procedure is made.

Non-invasive methods of assessing liver histology are being developed. Combinations of serologic markers (10) or imaging studies such as transient elastography (11) can detect advanced fibrosis with a high degree of accuracy, and to a lesser degree, disease activity. However, the use of these tests to assist in deciding about treatment initiation has yet to be validated. Non-invasive studies may be more useful, when done serially and in combination with a baseline liver biopsy, to assess disease progression over time (12). It is advisable to utilize the available studies at the initial assessment, even if a biopsy is performed, as this could serve s a baseline for longitudinal assessment.

The initial evaluation appointment should also be used to educate patients on their disease, the prognosis, and lifestyle recommendations. Alcohol use should be discouraged. Vaccination against hepatitis A should be offered. Signs and symptoms of disease flares should be discussed. Specific inquiry should be made regarding testing and vaccination of family members or sexual partners and disclosure of infectious status to them. It is important to discuss child-bearing plans with female patients and to recommend monitoring during pregnancy. The risk of disease reactivation during, and after, treatment with immune-suppressant agents (13, 14) should be conveyed to all patients and reinforced periodically. Patients should be advised to contact their hepatologist immediately if the need for such a treatment arises. With the currently available treatment options, one can realistically provide patients with hope that their illness can be controlled, enabling them to lead long and normal lives. Patient fears, both cultural and individual, need to be addressed in an objective and honest manner. Finally, patients should be encouraged to educate themselves and keep up to date on the understanding of hepatitis B and its treatment. Several excellent websites are available with reliable information about hepatitis B (Table 4).

Table 4
Internet Resources for Patients

Monitoring

After an initial evaluation (which may require serial assessments), the goals of patient follow-up visits are to reassess disease activity and the need for treatment, to monitor for progression of liver disease and its complications, and to screen for liver cancer. Patients should be reminded that they have a chronic disease, and long-term if not life-long monitoring is required.

The frequency of patient monitoring for disease activity should be based on the temporal pattern of disease, on the importance of early detection, and the potential change in management as a result of follow-up. Carriers and patients with normal liver enzymes and without HBeAg can develop intermittent and transient flares of disease that typically last for a few weeks to several months. Monitoring liver enzymes every 3 months would detect most of the flares at a relatively early stage, although there is no evidence that early detection of a flare affects treatment success. Thus, monitoring patients every 3–6 months with serum aminotransferase determinations should be sufficient. Patients in the immune-tolerant phase do not typically develop hepatitis until the 3rd or 4th decade of life, and again, early detection may not affect management. Thus, in children with the immune tolerant state, monitoring every 6 to 12 months is probably adequate. In contrast, patients who have cirrhosis can decompensate abruptly and spontaneously and may not have the liver reserve to tolerate a disease flare, mandating early intervention. If not being treated, these patients should be monitored every 3 months.

The routine tests to accompany monitoring have not been fully established. Ideally, measuring HBV DNA levels concomitantly with liver enzymes would be optimal; however, this has to be balanced with cost and test availability. Most patients can be monitored using serum aminotransferase levels as a guide to disease activity. Non-viremic carriers should be tested occasionally for loss of HBsAg, a phenomenon that can occur spontaneously at an annual rate of 0.5–1.4% (1517) as well as for the development of anti-HBs. In patients with known chronic hepatitis who are not being treated, monitoring HBV DNA levels and enzymes is less important as it is not likely to affect management. Naturally, all patients should be educated on symptoms of disease flares or exacerbations and should be instructed to contact their physician if these occur.

Monitoring of patients during antiviral therapy depends upon the agent used and the initial disease severity and is discussed elsewhere in these proceedings. HBV DNA testing is more important during therapy because of the need to assess the possibility of antiviral resistance: genotypic and virological breakthrough precedes biochemical breakthrough, and early intervention with salvage therapy appearance to improve long-term response rates (18).

To assess disease progression during follow-up visits, apart from liver enzymes, routine monitoring of synthetic function (albumin, prothrombin time), bilirubin, and complete blood count is required. Changes over time in these parameters, and especially a gradual decline in platelet count, may be more sensitive markers of progressing liver fibrosis than one-time measurements, especially for the parameters described in table 2. One example may be the AST-to-platelet ratio index (APRI = [AST value/ULN] ÷ [Platelet count/1000] * 100); although this index is only moderately accurate at predicting fibrosis on a single occasion, longitudinal monitoring may serve to detect progression of disease. Values above 1.5 are considered suspicious and above 2.0 highly suggestive of the presence of cirrhosis or advanced fibrosis. Continuously rising levels may be an early indication of progressive disease. This approach has not been validated specifically in hepatitis B, but was shown to be useful in patients with HCV-HIV coinfection (19). Similarly, repeated application of serum fibrosis marker panels (12) or transient elastography, can detect progression of fibrosis. Repeating these assays is especially relevant when monitoring for disease activity is infrequent, and disease flares may have been missed. If these non-invasive markers suggest progression of disease, even in a patient with normal ALT, a decision to treat may ensue. In this situation, a repeat liver biopsy may also be considered.

Screening and surveillance for HCC

Patients with hepatitis B infection are at risk for development of liver cancer. Specific risk factors are age, male sex, African or Asian origin, family history of liver cancer, genotype, core promoter mutations, and necroinflammatory activity, and high viral load (20). However, the major risk factor is cirrhosis. Thus, all cirrhotic patients, as well as patients with advanced necroinflammatory activity, should be screened. The current guidelines recommend initiating screening in all other chronically infected patients based on age, gender and race (Table 6). Recently, a risk score for the development of HCC in patients with chronic hepatitis B was described (21). If validated, especially in non-Asian patients, this score may serve as a guide for screening initiation. For specific patients, diversion from these guidelines may be justified based on individual risk factors. The recommended screening and surveillance method is by abdominal ultrasonography every 6–12 months (20). Alpha-fetoprotein (AFP) should not be used as a single screening test as its performance characteristics are markedly inferior to imaging studies. It can be used in combination with sonography, to increase sensitivity, albeit specificity is decreased concomitantly. However, repeated sonography and AFP measurement can also provide information on progression of disease beyond their applicability for cancer screening.

Table 6
Screening indications for HCC in patients with HBV

Special populations

Pregnant patients should be followed at least once per trimester. It is advisable to maintain contact with the patient’s obstetrician and recommend passive and active immunization of the newborn (22), especially in mothers who are positive for HBeAg. In patients with a high viral load, the benefits and hazards of treatment with a nucleoside analogue during the third trimester, to reduce the risk of vertical transmission, should be discussed with the patient, although the data to favor such an approach is limited (23).

Patients who are undergoing, or are about to undergo, profound immune-suppression should be evaluated, preferably prior to treatment initiation. A full serologic panel and HBV DNA level should be repeated for baseline. The decision to initiate prophylactic therapy should be individualized and based on the patient’s status of hepatitis B infection and projected degree and duration of immune-suppression. The patient and the treating physician should be educated about potential signs and symptoms of disease reactivation or exacerbation. Close monitoring of liver enzymes and HBV DNA is mandated, whether prophylaxis is given or not.

Conclusion and Needs for Future Research

The initial approach and subsequent monitoring of patients with hepatitis B should focus on assessing disease activity, severity, and need for treatment. Often, more than a single visit is required for correct diagnosis of disease status. Although much is still to be learnt about the natural history of hepatitis B, the main focus of future research should be to determine the optimal timing to initiate the modern-day effective, albeit costly and potentially life-long, antiviral medications. The usefulness of repeated longitudinal measurements of non-invasive assays of liver fibrosis (such as serum marker panels or transient elastography) for monitoring disease progression should be evaluated formally. Improving the sensitivity and specificity of screening methods for HCC should be attempted, and validated across races and geographic regions.

Table 5
Monitoring disease activity

Acknowledgments

Support: This work was supported in part by the Intramural Division of the NIDDK.

Abbreviations

HBV
hepatitis B virus
HBsAg
hepatitis B surface antigen
anti-HBs
Hepatitis B s antibody
anti-HBc
hepatitis B core antibody
HBeAg
hepatitis B e Antigen
anti-HBe
hepatitis B e Antibody
HCC
hepatocellular carcinoma
AFP
Alpha-fetoprotein
NIDDK
National Institute of Diabetes and Digestive and Kidney Diseases

Footnotes

Potential Conflicts of Interest: None to declare

References

1. Feld JJ, Heathcote EJ. Hepatitis B e antigen-positive chronic hepatitis B: natural history and treatment. Semin Liver Dis. 2006;26:116–129. [PubMed]
2. Hadziyannis SJ, Papatheodoridis GV. Hepatitis B e antigen-negative chronic hepatitis B: natural history and treatment. Semin Liver Dis. 2006;26:130–141. [PubMed]
3. Shepard CW, Simard EP, Finelli L, Fiore AE, Bell BP. Hepatitis B virus infection: epidemiology and vaccination. Epidemiol Rev. 2006;28:112–125. [PubMed]
4. Thompson ND, Perz JF, Moorman AC, Holmberg SD. Nonhospital health care-associated hepatitis B and C virus transmission: United States, 1998–2008. Ann Intern Med. 2009;150:33–39. [PubMed]
5. Hauri AM, Armstrong GL, Hutin YJ. The global burden of disease attributable to contaminated injections given in health care settings. Int J STD AIDS. 2004;15:7–16. [PubMed]
6. Allice T, Cerutti F, Pittaluga F, Varetto S, Gabella S, Marzano A, Franchello A, et al. COBAS AmpliPrep-COBAS TaqMan hepatitis B virus (HBV) test: a novel automated real-time PCR assay for quantification of HBV DNA in plasma. J Clin Microbiol. 2007;45:828–834. [PMC free article] [PubMed]
7. Garbuglia AR, Angeletti C, Lauria FN, Zaccaro P, Cocca AM, Pisciotta M, Solmone M, et al. Comparison of Versant HBV DNA 3.0 and COBAS AmpliPrep-COBAS TaqMan assays for hepatitis B DNA quantitation: Possible clinical implications. J Virol Methods. 2007;146:274–280. [PubMed]
8. Yuen MF, Yuan HJ, Wong DK, Yuen JC, Wong WM, Chan AO, Wong BC, et al. Prognostic determinants for chronic hepatitis B in Asians: therapeutic implications. Gut. 2005;54:1610–1614. [PMC free article] [PubMed]
9. Lai M, Hyatt BJ, Nasser I, Curry M, Afdhal NH. The clinical significance of persistently normal ALT in chronic hepatitis B infection. J Hepatol. 2007;47:760–767. [PubMed]
10. Myers RP, Tainturier MH, Ratziu V, Piton A, Thibault V, Imbert-Bismut F, Messous D, et al. Prediction of liver histological lesions with biochemical markers in patients with chronic hepatitis B. J Hepatol. 2003;39:222–230. [PubMed]
11. Marcellin P, Ziol M, Bedossa P, Douvin C, Poupon R, de Ledinghen V, Beaugrand M. Non-invasive assessment of liver fibrosis by stiffness measurement in patients with chronic hepatitis B. Liver Int. 2009;29:242–247. [PubMed]
12. Poynard T, Zoulim F, Ratziu V, Degos F, Imbert-Bismut F, Deny P, Landais P, et al. Longitudinal assessment of histology surrogate markers (FibroTest-ActiTest) during lamivudine therapy in patients with chronic hepatitis B infection. Am J Gastroenterol. 2005;100:1970–1980. [PubMed]
13. Katz LH, Fraser A, Gafter-Gvili A, Leibovici L, Tur-Kaspa R. Lamivudine prevents reactivation of hepatitis B and reduces mortality in immunosuppressed patients: systematic review and meta-analysis. J Viral Hepat. 2008;15:89–102. [PubMed]
14. Loomba R, Rowley A, Wesley R, Liang TJ, Hoofnagle JH, Pucino F, Csako G. Systematic review: the effect of preventive lamivudine on hepatitis B reactivation during chemotherapy. Ann Intern Med. 2008;148:519–528. [PMC free article] [PubMed]
15. de Franchis R, Meucci G, Vecchi M, Tatarella M, Colombo M, Del Ninno E, Rumi MG, et al. The natural history of asymptomatic hepatitis B surface antigen carriers. Ann Intern Med. 1993;118:191–194. [PubMed]
16. Liaw YF, Sheen IS, Chen TJ, Chu CM, Pao CC. Incidence, determinants and significance of delayed clearance of serum HBsAg in chronic hepatitis B virus infection: a prospective study. Hepatology. 1991;13:627–631. [PubMed]
17. Villeneuve JP, Desrochers M, Infante-Rivard C, Willems B, Raymond G, Bourcier M, Cote J, et al. A long-term follow-up study of asymptomatic hepatitis B surface antigen-positive carriers in Montreal. Gastroenterology. 1994;106:1000–1005. [PubMed]
18. Lampertico P, Vigano M, Manenti E, Iavarone M, Lunghi G, Colombo M. Adefovir rapidly suppresses hepatitis B in HBeAg-negative patients developing genotypic resistance to lamivudine. Hepatology. 2005;42:1414–1419. [PubMed]
19. Al-Mohri H, Murphy T, Lu Y, Lalonde RG, Klein MB. Evaluating liver fibrosis progression and the impact of antiretroviral therapy in HIV and hepatitis C coinfection using a noninvasive marker. J Acquir Immune Defic Syndr. 2007;44:463–469. [PubMed]
20. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005;42:1208–1236. [PubMed]
21. Yuen MF, Tanaka Y, Fong DY, Fung J, Wong DK, Yuen JC, But DY, et al. Independent risk factors and predictive score for the development of hepatocellular carcinoma in chronic hepatitis B. J Hepatol. 2009;50:80–88. [PubMed]
22. Lee C, Gong Y, Brok J, Boxall EH, Gluud C. Effect of hepatitis B immunisation in newborn infants of mothers positive for hepatitis B surface antigen: systematic review and meta-analysis. BMJ. 2006;332:328–336. [PMC free article] [PubMed]
23. Xu WM, Cui YT, Wang L, Yang H, Liang ZQ, Li XM, Zhang SL, et al. Lamivudine in late pregnancy to prevent perinatal transmission of hepatitis B virus infection: a multicentre, randomized, double-blind, placebo-controlled study. J Viral Hepat. 2009;16:94–103. [PubMed]
24. Wang CC, Lim LY, Deubner H, Tapia K, Lau AW, Manansala J, Krows M, et al. Factors predictive of significant hepatic fibrosis in adults with chronic hepatitis B and normal serum ALT. J Clin Gastroenterol. 2008;42:820–826. [PubMed]
25. Hui CK, Leung N, Yuen ST, Zhang HY, Leung KW, Lu L, Cheung SK, et al. Natural history and disease progression in Chinese chronic hepatitis B patients in immune-tolerant phase. Hepatology. 2007;46:395–401. [PubMed]
26. Lok AS, McMahon BJ. Chronic hepatitis B. Hepatology. 2007;45:507–539. [PubMed]
27. Williams AL, Hoofnagle JH. Ratio of serum aspartate to alanine aminotransferase in chronic hepatitis. Relationship to cirrhosis. Gastroenterology. 1988;95:734–739. [PubMed]
28. Shin WG, Park SH, Jang MK, Hahn TH, Kim JB, Lee MS, Kim DJ, et al. Aspartate aminotransferase to platelet ratio index (APRI) can predict liver fibrosis in chronic hepatitis B. Dig Liver Dis. 2008;40:267–274. [PubMed]
29. Schmilovitz-Weiss H, Tovar A, Halpern M, Sulkes J, Braun M, Rotman Y, Tur-Kaspa R, et al. Predictive value of serum globulin levels for the extent of hepatic fibrosis in patients with chronic hepatitis B infection. J Viral Hepat. 2006;13:671–677. [PubMed]
30. Mohamadnejad M, Montazeri G, Fazlollahi A, Zamani F, Nasiri J, Nobakht H, Forouzanfar MH, et al. Noninvasive markers of liver fibrosis and inflammation in chronic hepatitis B-virus related liver disease. Am J Gastroenterol. 2006;101:2537–2545. [PubMed]