Deaths from liver-related causes remained static between 1979 and 2008, according to updated estimates from the Rochester Epidemiology Project and the National Death Index.
The finding represents a dramatic correction to previous estimates from the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention, which used a very narrow definition of liver-related mortality, Dr. Sumeet K. Asrani and colleagues reported. The results were published in the August issue of Gastroenterology.
Courtesy: American Gastroenterological Association
According to Dr. Asrani of the Mayo Clinic, Rochester, Minn., "current estimates [from the NCHS] are solely based on one diagnostic category, namely chronic liver disease and cirrhosis, which fails to capture deaths attributed to other uniquely liver-related descriptors, such as hepatic encephalopathy or hepatorenal syndrome."
Deaths due to viral hepatitis or liver cancer are also not included.
"For example, the current estimates of liver-related deaths might not include the demise of a person with hepatitis C cirrhosis who died of hepatorenal syndrome."
In their updated assessment, the researchers looked at data from the Rochester Epidemiology Project, in which death records of Olmsted County residents are tracked from multiple sources, including county and state vital records, as well as individual medical charts.
The researchers found that when the CDC definition was used, which includes causes attributable to a single ICD-9 code of 571 and ICD-10 codes of K70, K73, and K74, there were 71 liver-related deaths among Olmsted County residents between 1999 and 2008.
In contrast, when the updated definition was applied, which included additional diagnoses specific to liver disease yet excluded in the CDC definition, as well as viral hepatitis and malignant neoplasm of the liver and intrahepatic bile ducts, there were 261 liver-related deaths in the county between 1999 and 2008.
That included 85 deaths (32.6%) from viral hepatitis and 70 deaths (26.8%) from hepatobiliary malignancies.
The researchers then looked at national mortality rates.
According to the restrictive CDC estimates, in 2008, there were 29,951 deaths due to liver disease in the United States, for a death rate of 11.7 per 100,000 persons (16.2 for men and 7.6 for women).
Using the updated definition, however, Dr. Asrani tallied 66,007 deaths, including 10,256 from the expanded liver disease diagnosis, 7,625 from viral hepatitis, and 18,175 from hepatobiliary malignancies, for a death rate of 25.7 per 100,000 persons (35.7 for men and 16.8 for women).
Finally, the investigators compared longitudinal trends in death rates.
They found that when the CDC definition was used, deaths due to liver disease (per 100,000 persons) decreased from 16.5 in the era encompassing the years 1979 to 1988 to 11.7 in the years between 1999 and 2008, a reduction of 38%.
"However, when the updated definition was applied, the downward trend disappeared," they wrote.
Indeed, liver-related mortality per 100,000 persons was basically unchanged over 30 years: 23.9 for the era between 1979 and1988 and 24.4 for the period from 1999 to 2008.
"This discrepancy was accounted for by deaths due to viral hepatitis, which increased from 0.5 to 2.5 per 100,000 and hepatobiliary cancer, which increased from 3.1 to 6.4 per 100,000," wrote the authors.
"These data support that deaths due to viral hepatitis and hepatobiliary cancers should be included in the enumeration of liver-related deaths to accurately represent the burden of chronic liver disease."
"Underappreciation of the prevalence and natural history of liver disease can lead to suboptimal care," they concluded.
The authors disclosed no conflicts of interest related to this study, which was partially supported by grants from the National Institutes of Health.