The role of the hepatopathologist in the assessment of drug-induced liver injury

Hepatopathologists are sometimes asked whether they can tell that drug-induced liver injury (DILI; throughout this review, the term DILI is used to include liver injury from prescribed and over-the-counter medications, as well as injury from herbals and dietary supplements) is present simply by the pattern of changes in the liver biopsy, without any additional clinical information. The question misunderstands the role of the liver biopsy and the role of the hepatic pathologist as an interpreter of biopsy findings. In reality, hepatopathologists are sophisticated interpreters of the tissue injury and can synthesize the information obtained from a liver biopsy with the clinical information to provide an expert opinion, not only of the likelihood of drug injury but of other important information from the biopsy.1

Table 1 outlines the role of the hepatopathologist when evaluating a biopsy for potential DILI. The pathologist has an important part in both the assessment of the pattern of injury, which is used to establish the histological differential diagnosis, and the severity of injury, which has bearing on prognosis. The determination of causality mainly depends on the pattern of injury,1 although severity does play a lesser role in that very severe injuries, with extensive necrosis or marked inflammation, should be investigated for DILI even if DILI was not previously suspected.

Determination of the Pattern of Injury

When reviewing any biopsy or larger tissue specimen, the pathologist approaches the evaluation in a blinded fashion, initially leaving aside consideration of clinical information. This method allows the eyes and the mind to be open to the histological changes without being biased by what is known or assumed about the patient. When injured by either disease or some external or iatrogenic cause, tissues have characteristic ways of responding: the patterns of injury. For example, in chronic viral infection by hepatitis B or C, there is accumulation of lymphocytes and macrophages mainly within and at the edges of portal areas. There are also small foci of inflammation scattered in the parenchyma, but from low magnification the impression is one of predominant inflammatory infiltration and expansion of the portal areas. This pattern is termed “chronic hepatitis,” after the viral diseases. Further examination is required to exclude histological characteristics, such as cholestasis, that would take the pattern classification away from chronic hepatitis. Once the pattern of injury is defined, the non-DILI differential diagnosis is also defined: chronic viral hepatitis, autoimmune hepatitis, early primary biliary cholangitis (PBC), and a host of less common causes (Table 2). The pattern of injury also defines likely DILI causes. For example, minocycline causes a chronic hepatitis pattern, whereas azithromycin does not.

We have used modified lists of patterns of injury associated with DILI originally developed by Popper2 and by Zimmerman3 in our evaluation of cases of suspected DILI collected by the Drug-Induced Liver Injury Network.4 Detailed descriptions of the various patterns of injury are beyond the scope of this short review and may be found elsewhere.5 Once the objective assessment of the pattern and severity of injury is complete, the pathologist turns to consideration of the histological changes in light of the clinical information. Knowledge of the history may prompt further review of the biopsy, with an eye toward identifying particular lesions of interest or providing more detailed observations tailored to the clinical situation. But when the histological evaluation is complete, the next job is to interpret the findings in light of the clinical history, laboratory tests, and imaging.

Clinicopathological Correlation in Causality Determination

Long before there was codification of DILI causality by the use of tools like RUCAM or expert evaluation, Dr. Irey at the Armed Forces Institute of Pathology wrote a monograph to give pathologists a method for evaluating potential cases of toxic tissue injury (Table 3).6 Two of these principles are particularly relevant in the pathologist's evaluation of potential cases of DILI. First, a pathologist can use the biopsy findings to both exclude diseases (and sometime DILI) from consideration, as well as suggest other diseases that may require further testing to exclude. For example, a patient with known fatty liver disease presents with a positive anti-nuclear antibody test and higher than expected aminotransferase levels. If the biopsy shows only severe steatohepatitis, then the findings provide an explanation for the laboratory abnormalities and do not support an additional diagnosis of DILI. On the other hand, if the biopsy demonstrates hepatocyte necrosis in zone 3 and substantial portal inflammation, then the biopsy findings support a second cause of liver injury, possibly related to DILI.

The principle of precedent for the injury pattern is also a place for the pathologist to provide informed advice. Knowledge of reported patterns from the various suspect medications can be correlated with the observed injury, allowing the pathologist to further refine the likelihood of diagnostic DILI. Precedent for certain specialized patterns of injury can allow the pathologist to suggest DILI even when DILI is not in the clinical differential diagnosis. As noted earlier, extensive necrosis and/or severe hepatitis are unusual patterns in typical practice and should prompt a search for a toxic cause.7 Similarly, the pattern of cholestatic hepatitis has only a limited histological differential diagnosis that is topped by DILI. Granulomas (particularly after exclusion of sarcoidosis and infection) and eosinophils8 are components of hypersensitivity reactions in all organs and may also suggest DILI.

These considerations emphasize the fact that the pathologist's expertise is best used when paired with clinical information. If the pathologist does not have ready access to an electronic medical record that holds all the relevant data, it is important to supply that information. This information should include the medication and dietary supplement history, preferably over the preceding 6 months, as well as the results of serial liver-associated enzyme levels, serological and virological tests, and imaging results.

Summary

The hepatopathologist is a medical consultant whose expertise bridges the gap between the interpretation of biopsy findings and clinical liver disease. Evaluation of the biopsy begins with objective assessment of the pattern of injury and gradation of its severity. The hepatopathologist's knowledge of how disease affects the liver under different clinical conditions is then combined with the clinical information to clarify the contributions of known diseases, as well as to identify potential causes of additional injury, either caused by DILI or other, as yet undefined, causes.