 |
|
|
| |
Lower Glucose and HIV Load Linked to Fibrosis Regression in 5-Year Study
|
| |
| |
5th IAS Conference on HIV Pathogenesis, Treatment and Prevention, July 19-22, 2009, Cape Town
Mark Mascolini
"Multivariate statistical analysis determined that treatment for HCV infection nearly quadrupled chances of fibrosis regression"
A lower HIV load, lower glucose, and treatment of HCV infection correlated with regression of liver fibrosis in 632 people coinfected with HIV and a hepatitis virus [1]. Of these three factors, glucose control and HCV therapy independently predicted improving fibrosis in a multivariate analysis. Treatment with a nonnucleoside versus a protease inhibitor (PI) had no impact on fibrosis regression.
Clinicians at Madrid's Carlos III Hospital prospectively tracked 632 people who had at least two liver Fibroscans separated by 18 months or more from 2004 through 2008. Three quarters of the study group (78%) were men, 63% were injecting drug users, and 9% were alcohol abusers. When people entered the cohort, their median age was 45 years and median CD4 count 494. While 72% had a positive HCV RNA test, 10% had hepatitis B virus surface antigen (HBsAg).
Jose Fernandez-Montero and colleagues defined fibrosis regression as at least a 30% decrease in Fibroscan-measured liver stiffness in people with a METAVIR score of F3 or F4 ("significant fibrosis") on their first test. The first Fibroscan detected F3-F4 fibrosis in 193 people (30.5%), and the second Fibroscan (done a median of 27 months later) found F3-F4 fibrosis in 65 people (10%).
Comparing people with and without liver fibrosis regression, the Madrid team found that regressors included a lower proportion of men (67.7% versus 83.6%, P = 0.01), a lower proportion of people infected while injecting drugs (69.4% versus 84.3%, P = 0.02), a lower proportion of people with HCV (80.3% versus 91.5%, P = 0.03), and a higher proportion of people treated for HCV (81.3% versus 44.2%, P < 0.001). Among people treated for HCV, regressors included a higher proportion with a sustained virologic response (60% versus 25.6%, P = 0.001).
A lower proportion of regressors had never taken antiretrovirals, but that difference fell short of statistical significance (4.6% versus 13.3%, P = 0.06). Regressors had a significantly lower HIV load at their second Fibroscan (1.8 versus 2.1 log copies/mL, P < 0.001). CD4 count and treatment with a nonnucleoside versus a PI did not distinguish regressors from nonregressors.
People with fibrosis regression had a greater improvement in glycemia between the first and second Fibroscan (-3.2 versus +1.9 mg/dL, P = 0.04). Triglycerides also fell in regressors while rising in nonregressors, but this difference fell short of statistical significance (-31.57 versus +7.5 mg/dL, P = 0.08).
Multivariate statistical analysis determined that treatment for HCV infection nearly quadrupled chances of fibrosis regression (odds ratio 3.8, 95% confidence interval 1.2 to 12.1, P = 0.02). Better glycemic control independently improved changes of regression only 5%, but that correlation was significant (odds ratio 1.05, 95% confidence interval 1.03 to 1.09, P = 0.03). Every 10-fold lower HIV load at the second Fibroscan made regression more likely, but this correlation lacked statistical significance (P = 0.12).
The results confirm that control of HCV coinfection can reverse fibrosis in people with HIV, including those taking antiretrovirals. Indeed, the study found a trend toward fibrosis regression in people with a lower HIV load at the time of their second Fibroscan.
Reference
1. Fernandez Montero JV, Labarga P, Barreiro P, et al. Factors associated with improvement of liver fibrosis in a large cohort of HIV-infected patients followed for 5 years. 5th IAS Conference on HIV Pathogenesis, Treatment and Prevention. July 19-22, 2009. Cape Town. Abstract WEPEB218.
|
| |
|
|
|
|
|
