To the Editors: Coinfection by hepatitis C virus (HCV) is frequent in patients infected with human immunodeficiency virus (HIV) because both viruses share routes of transmission.1 Progression of chronic hepatitis C to cirrhosis and end-stage liver disease is more frequent and rapid among HIV-infected individuals than in those not infected by HIV.2 The major determinants of this accelerated fibrogenesis in HIV/HCV-coinfected patients are immunosuppression and alcohol consumption, although direct action of HIV may also be a contributing factor.3 Therefore, liver disease is a leading cause of hospital admissions and death after the introduction of combination antiretroviral therapy in areas where the prevalence of HIV/HCV coinfection is high.4 Decompensation in cirrhosis results primarily from portal hypertension, which is defined as a pathologic increase in the hepatic venous pressure gradient (HVPG), that is, the pressure gradient between the portal vein and the inferior vena cava.5 HVPG is the most accurate predictor of liver decompensation and hepatocellular carcinoma in patients with compensated cirrhosis.6 In addition, HVPG is an independent predictor of death in patients with decompensated cirrhosis.7 Therefore, HVPG is considered the best surrogate marker of the natural history of cirrhosis. Sustained virologic response (SVR) after anti-HCV therapy has been associated with improvements in liver fibrosis and with reversion of cirrhosis in one-third to half of patients with baseline cirrhosis included in randomized clinical trials of interferon plus ribavirin with pretreatment and posttreatment liver biopsies.8 SVR has also been associated with improved survival and reduced liver decompensation and hepatocellular carcinoma in patients with chronic hepatitis C with or without HIV infection.9,10 Although these findings support the notion that eradication of HCV can reduce HVPG in patients with HCV-related cirrhosis, this area has received little attention,11 particularly in coinfected patients. Our objective was to assess the effect of eradication of HCV after anti-HCV therapy on HVPG in HIV-infected patients with compensated HCV-related cirrhosis. The study population comprised all HIV/HCV-coinfected patients with compensated liver cirrhosis whose HVPG was measured before and after anti-HCV therapy between 2007 and 2012. HVPG was measured after an overnight fast as described elsewhere.12 The HVPG was calculated as the wedged hepatic venous pressure minus the free hepatic venous pressure. All patients signed an informed consent document before the portal hemodynamic study. Transient elastography (TE) was performed immediately before the determination of HVPG using a FibroScan device (Echosens, Paris, France). The median value [expressed in kilopascals (kPa)] of 10 successful acquisitions was considered the representative measurement of liver stiffness.13 We considered a cutoff value of liver stiffness ≥14 kPa as diagnostic of liver cirrhosis.14 Between 2007 and 2012, determination of HVPG and TE was performed in 60 HIV/HCV–coinfected patients with compensated cirrhosis confirmed by liver biopsy and/or TE. During this period, 27 of these 60 patients were treated with pegylated interferon (Peg-IFN) alpha-2a or alpha-2b plus ribavirin (programmed for 48 weeks), whereas 33 were not. A total of 15 treated patients achieved SVR, defined as undetectable HCV-RNA in serum at 24 weeks after discontinuation of anti-HCV therapy. The median [interquartile range (IQR)] baseline HVPG was 13 (10–15) mm Hg for untreated patients, 15 (10–18) mm Hg for treated patients with SVR, and 10 (9–14) mm Hg for nonresponders. The median (IQR) baseline liver stiffness was 30.4 (15.7–43.9) kPa for untreated patients, 23.9 (15.1–38.0) kPa for treated patients with SVR, and 14.6 (12.2–24.6) kPa for nonresponders. Eight of the 15 patients with SVR consented to have a second determination of HVPG and are the focus of this study. All 8 patients had clinically significant portal hypertension (HVPG ≥ 10 mm Hg), and 6 had severe portal hypertension (HVPG > 12 mm Hg). The second HVPG and TE determinations were performed a median of 18 (10–26) months after the discontinuation of anti-HCV therapy. The median (IQR) HVPG was 15.50 (11.25–17.62) mm Hg at baseline and 8.75 (6.25–12.00) mm Hg after SVR (P = 0.01) (Figure 1). After SVR, 7 patients experienced a reduction in HVPG from baseline of ≥20%, and all 7 had HVPG values ≤12 mm Hg. Likewise, the median (IQR) liver stiffness was 27.75 (18.20–38.60) kPa at baseline and 12.80 (7.30–17.05) kPa after SVR (P = 0.01) (Figure 1). After a median follow-up time of 60 months, all 8 patients were alive and free from decompensation and hepatocellular carcinoma.FIGURE 1: A, Median and IQR values and individual values of HVPG at baseline and after SVR. B, Median and IQR values and individual TE values at baseline and after SVR.The effect of anti-HCV therapy on HVPG in patients with HCV-related cirrhosis has received little attention hitherto. In one study, HVPG was measured in 5 patients immediately before and immediately after the discontinuation of a 48-week course of interferon monotherapy and in 5 control patients.15 In treated patients, HVPG decreased from a mean of 13.8 mm Hg at baseline to 11.1 mm Hg at the end of therapy, whereas in controls, it increased from a mean of 13.2 to 14.6 mm Hg; neither of these changes was statistically significant.15 In a second study, HVPG measurements were performed immediately before and immediately after discontinuation of Peg-IFN plus ribavirin in 20 patients with compensated chronic hepatitis C and portal hypertension, defined as an HVPG > 5 mm Hg.16 The authors found a greater reduction in HVPG in patients who achieved end-of-treatment virologic response than in nonresponders. The effect of SVR on portal pressure in 33 patients with HCV-related cirrhosis was assessed at baseline and after 6 months of treatment-free follow-up in 33 patients with cirrhosis by Roberts et al,11 who found that after treatment, there was a significant decrease in HVPG in sustained responders (from a mean of 9.0 mm Hg to a mean of 6.9 mm Hg) compared with relapsers and primary nonresponders. Our findings and those of others bring a pathophysiological correlate to the observation of clinical benefits obtained with SVR in patients with HCV-related compensated liver cirrhosis with or without HIV infection.9,10 It is important to remark that the achievement of an adequate reduction in HVPG after pharmacological treatment of portal hypertension in patients with variceal bleeding reduces not only the probability of recurrent bleeding but also the probability of developing other severe manifestations of end-stage liver cirrhosis such as spontaneous bacterial peritonitis, hepatorenal syndrome, and even death.17 This adequate reduction in portal pressure has been estimated as a decrease in HVPG of ≥20% from baseline or an absolute reduction of HVPG to ≤12 mm Hg.17 We found that 7 of the 8 patients met both these requirements for reduction of portal pressure and, therefore, were at less risk of decompensation and death. The remaining patient had a reduction of 10% in HVPG from baseline, and his HVPG value after SVR was 13.5 mm Hg. However, he was also the patient with the shortest interval between drug discontinuation and the second HVPG determination (only 7 months). Roberts et al11 found that among patients with portal hypertension, a higher proportion of responders achieved a ≥20% reduction in HVPG compared with nonresponders (71% vs. 20%, respectively). According to our experience and that of others, most patients with HCV-related compensated cirrhosis and portal hypertension achieve adequate reductions in HVPG after SVR following anti-HCV therapy, although a small fraction do not achieve that goal. These patients probably remain at risk of developing complications related to portal hypertension despite eradication of HCV.10,18 HVPG determinations or even TE19 could help to identify patients with HCV-related cirrhosis who remain at risk for decompensation or hepatocellular carcinoma despite having achieved SVR. This issue should be addressed in prospective studies. The main limitation of this study is the small sample size and the lack of a control group comprising HIV-infected patients with HCV-related cirrhosis who are nonresponders to anti-HCV therapy and whose HVPG was determined before and after treatment. However, a key strength of our study is the long interval (18 months) between the discontinuation of interferon and ribavirin and the determination of the second portal pressure measurement. Previous studies on the subject addressed the effect of antiviral therapy on HVPG but not that of SVR on HVPG15,16 or they determined changes in portal pressure 6 months after treatment discontinuation11; therefore, the question of whether the reduction of portal pressure after SVR was maintained in the long term remains unresolved. In conclusion, the results of our study support the notion that eradication of HCV is associated with a marked and clinically significant reduction in portal pressure in most—but not all—HIV-infected patients with HCV-related cirrhosis.