Table 5. Univariate analysis of variables for prediction of SVR in GT2 and GT3.
Genotype 2 | Genotype 3 | |||||||||
Baseline variables | p | OR | 95% CI for OR * | N | p | OR | 95% CI for OR * | N | ||
Gender | 0.805 | 391 | 0.156 | 1956 | ||||||
Age ≤45 vs.>45 years | 0.842 | 391 | 0.014 | 0.781 | 0.642 | 0.950 | 1956 | |||
BMI | 0.843 | 391 | 0.364 | 1956 | ||||||
Subtype of HCV genotype | 0.381 | 224 | 0.753 | 1332 | ||||||
Duration of infection (years) | 0.910 | 332 | 0.800 | 1743 | ||||||
HCV-RNA ≤400,000 vs.>400,000 IU/ml | 0.458 | 390 | 0.045 | 0.832 | 0.695 | 0.996 | 1935 | |||
Psychiatric co-morbidity | 0.972 | 391 | 0.022 | 0.754 | 0.591 | 0.960 | 1956 | |||
HIV co-infection | 0.752 | 391 | 0.558 | 1956 | ||||||
HBV co-infection | 0.406 | 391 | 0.166 | 1956 | ||||||
Current drug use | 0.388 | 391 | 0.063 | 1956 | ||||||
LDL cholesterol ≤85 vs.>85 mg/dl | 0.308 | 120 | 0.000 | 2.073 | 1.456 | 2.951 | 532 | |||
Cholesterol ≤130 vs.>130 mg/dl | 0.332 | 220 | 0.000 | 1.838 | 1.360 | 2.483 | 1037 | |||
Triglycerides (mg/dl) | 0.962 | 210 | 0.564 | 945 | ||||||
Serum ferritin (µg/l) | 0.075 | 150 | 0.206 | 657 | ||||||
Prothrombine time (%) | 0.007 | 1.032 | 1.009 | 1.055 | 245 | 0.138 | 1155 | |||
APRI score <0.5; 0.5-≤1.5; 1.5-≤2 # | 0.050 | 0.783 | 0.613 | 1.000 | 346 | 0.000 | 0.764 | 0.689 | 0.846 | 1780 |
γ-GT> vs. ≤3-times ULN | 0.017 | 0.392 | 0.182 | 0.844 | 370 | 0.000 | 0.480 | 0.347 | 0.663 | 1872 |
Variables under treatment | ||||||||||
RVR | 0.000 | 3.759 | 2.004 | 7.052 | 195 | 0.000 | 2.900 | 2.210 | 3.803 | 963 |
Duration of therapy <22 weeks - 22–26 weeks ->26 weeks | 0.247 | 359 | 0.732 | 1643** | ||||||
Change in dose PEG-interferon | 0.326 | 391 | 0.140 | 1953 | ||||||
RBV starting dose (mg/kg body weight) | 0.002 | 1.127 | 1.046 | 1.213 | 390 | 0.004 | 1.041 | 1.013 | 1.069 | 1950 |
* OR = Odds Ratio; CI = Confidence interval.
** With a planned treatment end or with treatment discontinuation for virological failure or adverse events.
# By univariate analysis baseline thrombocytes and GOT were also significant in GT3 in predicting SVR; in GT2 these variables were not significant. Since the APRI score which combines these two variables had a higher predictive value when compared with the two single variables only the APRI score was used for further analyses.