Background: Cirrhosis is the permanent fibrosis of the liver and carries a prevalence of 600,000 cases per year. Hospitalizations due to cirrhosis have doubled from 2001 to 2011 with an average hospitalization cost of $15,000. Hospital readmissions for cirrhosis are frequent and have been noted to be up to 3 readmissions per year. Given the frequency of readmissions and the high cost of hospitalization related to cirrhosis efforts to reduce unnecessary testing are critical for high value care.

Methods:

A retrospective chart review was performed for patients discharged in 2015 with an ICD9 diagnosis of cirrhosis in a nine-month period.  An index consisting of 95 patients was created, noting laboratory evaluation of cirrhosis upon admission, and prior history of cirrhosis laboratory evaluation.

Results:

Ninety-five patients were evaluated with an average age of 48 years. Forty-nine patients were discharged with the diagnosis of alcoholic cirrhosis, 30 patients with alcoholic hepatitis, 14 patients with cirrhosis unspecified and two patients with autoimmune hepatitis (diagnosis known on admission). Sixty-one patients (64%) carried a diagnosis of cirrhosis prior to the hospital admission.

Table 1 reviews the results of screening tests performed in these patients, the frequency of positive results, and the number of repeated tests that were performed. Of note, further review performed on patients with positive screening tests showed that seven of nine patients with a ceruloplasmin below reference range had further work-up such us urinary copper testing and the single patient that had a low Alpha-1-antitrypsin level did not undergo further work-up. Also of note, there were two positive antinuclear antibody screenings at a titer of 1:320 or greater.  One patient had idiopathic cirrhosis and follow-up at an outside clinic was planned while the other patient had alcoholic hepatitis and this test was thought to be a false positive with plans to repeat testing in the future, which has not occurred at follow-up appointments.

Conclusions:

The diagnosis of cirrhosis should involve a thorough history and physical exam with appropriate evaluation for reversible or familial etiologies. Given high readmission rates, redundant work-up is common. An estimated 60 repeat laboratory tests were ordered in 95 patients. Furthermore, it was noted that abnormal lab values did not routinely receive further investigation. Several possibilities for this lack of follow up testing include the ordering of tests in patients with low pre-test probability of disease or high false-positive rate and the lack of algorithm to follow up positive results leading to possible lapses in patient care. Further quality improvement efforts will aim to target these areas to improve the value of inpatient cirrhosis work-up and management.