Prediction of early mortality for alcohol-related hepatitis

Researchers have proven that a new scoring system is accurate in predicting which patients with alcohol-related hepatitis are at increased risk of short-term death.

In the derived cohort, the mortality index of the so-called alcohol-related hepatitis (MIAAH) model, which incorporates patient age, blood urea nitrogen, albumin, bilirubin, and the International Sensitivity Ratio (INR), achieved C statistics. 0.86 (95% CI 0.82-0.91, P= 0.002) To predict mortality, Dr. Douglas Simonet of the Mayo Clinic, Rochester, Minnesota, and a colleague wrote: Mayo Clinic Minutes..

In the validation cohort, the accuracy of MIAAH is still reduced to 0.73 (95% CI 0.65-0.82), which is “clinically useful”, superior to the Maddrey Discriminant Function (MDF) model, but End-Stage Liver Disease (MELD). ) Better than the model. :

• MDF: C statistic 0.64 (95% CI 0.55-0.74, P= 0.01)
• MELD: C Statistics 0.72 (95% CI 0.63-0.80, P= 0.61)

“Given the significant mortality rates seen in patients with alcohol-related hepatitis, it is important to assess the severity and prognosis of the disease,” Simonet said in a press release. “We believe that MIAAH will probably be refined over time, perhaps in combination with existing models, but our study found that MIAAH is a useful tool for assessing mortality risk. “

Alcohol-related hepatitis is associated with high short-term mortality, especially in severe cases for which no medical treatment has been shown to reduce the risk of death for 90 days. Modeling may help determine the eligibility and timing of liver transplants, the researchers said.

For their derivative cohort, Simonetto’s group examined health record data from 266 patients with alcohol-related hepatitis ( National Institute of Alcohol Abuse and Alcoholism Criteria) Treated at Mayo Clinic from 1998 to 2018. The average age of the patients was 48 years, almost all were Caucasian, 38% were female, and the average obesity index was 29.2.

Here, the 30-day mortality rate was 19.2%, and multivariate analysis showed that the following variables were significantly associated with mortality.

• INR per unit: or 2.80 (95% CI 1.51-5.21)
• Age per 10 years: or 1.82 (95% CI 1.26-2.64)
• Bilirubin, per 1 mg / dL: or 1.04 (95% CI 1.01-1.08)
• Blood urea nitrogen per 1 mg / dL: or 1.02 (95% CI 1.01-1.04)
• Albumin, per 0.1 g / dL: or 0.91 (95% CI 0.85-0.98)

The validation cohort surveyed 249 patients treated at the University of South Dakota and University of Kansas Healthcare Centers from 2009 to 2018. The 30-day mortality rate for this group was 12.9%, at which point two patients had a liver transplant. point. After adjusting for age, no significant difference in 30-day mortality was observed between the groups.

Two cohorts (deriving and validation, respectively) have similar MELD scores (18.5 vs 19.3) and bilirubin (15.4 vs 13.7 mg / dL), albumin (2.9 vs 2.8 g / dL), and creatinine (1.5 vs 1.4 mg) levels. Had / dL), and blood urea nitrogen (21.9 vs 19.6 mg / dL).

The derived cohort had a significantly lower median MDF score (44.5 vs 59.0), an INR value (1.8 vs 2.1), and fewer patients with corticosteroids (21.4% vs 29.7%). P<0.05).

Camille Kezer, MD, co-author of the study, also at Mayo Clinic, states that the MIAAH model “has the advantage of working well in patients, whether or not they are treated with steroids.”

The authors acknowledged some restrictions on the data, including the use of INR. This is because the INR is tailored to the limits of vitamin K antagonists. The use of creatinine as a functional renal marker also has its drawbacks. Finally, no dynamic evaluation of bilirubin levels was performed.