|
nif:isString
|
-
ANRS CO13 HEPAVIH is a multicenter observational cohort initiated in France in 2005 among HIV-HCV co-infected patients to characterize the natural history of co-infection in terms of both morbidity and mortality and their determinants, and to better understand the interactions between the two viruses and treatments [21]. ANRS CO13 HEPAVIH recruited outpatients from 32 clinical centers in France who were aged 18 or over, had both chronic hepatitis C and HIV infection, had detectable plasma HCV RNA, were positive for anti-HCV antibodies, and agreed to participate in the cohort (a signed letter of informed consent being provided). Patient recruitment for this cohort included three phases: The first phase between December 2005 and December 2008 recruited adult HIV-positive patients, who were also either chronically infected by HCV at enrollment or had a negative HCV RNA test by PCR six months after ending HCV treatment (PEG-IFN),The second phase, from September 2011 to March 2016, recruited both individuals who spontaneously cleared HCV before the enrolment, and HIV-HCV infected patients who initiated HCV treatment with triple therapy (PEG-IFN +ribavirin + either telaprevir or boceprevir),The third phase, from May 2014 to November 2015, was an additional recruitment phase, where patients who were already receiving, or were scheduled to receive DAA within six months, were recruited.In France, PEG-IFN+ ribavirin first became available in 2000. Boceprevir and Telaprevir became available in 2010. All three regimens were replaced from 2014 onwards with DAA. In the present study, we included all cohort patients who had received or were still receiving HCV treatment, whether PEG-IFN-based in the first phase of the cohort or DAA-based in the third phase (Fig 1).
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0199874.g001 Flow chart of the patients included in the study. We preferred to exclude Phase 2 patients from this comparison for two reasons: first, a very small sample were eligible for this study, mainly patients in a severe state of illness and patients who were very unresponsive to Phase 1 treatment (ie PEG-IFN+Ribavirin) due to the high toxicity of this treatment. In addition, patients in the Phase 2 did not complete the annual self-administered questionnaires.
For patients included in the first recruitment phase, clinical visits were scheduled annually for non-cirrhotic patients and every six months for cirrhotic patients. Collection of socio-behavioral data using self-administered questionnaires (SAQ) was scheduled annually. For patients initiating HCV treatment, additional visits were scheduled, before, during and after the end of treatment. For patients included in the third recruitment phase receiving DAA, clinical and biological data were collected at treatment initiation, at the end of treatment, and six months after the end of treatment, using medical questionnaires. These data included weight, blood pressure, clinical stage of HIV infection, blood count, immuno-virological data, hepatic and renal assessment values, HCV RNA PCR measurements, and data on HCV and HIV treatments (drug, dosage, indication, initiation date and date of treatment end). Socio-behavioral data were collected from SAQ at treatment initiation and at the end of treatment. The SAQ included items concerning patients’ socio-demographic characteristics and information on patients’ use of psychoactive drugs, tobacco, and alcohol (AUDIT-C).
Alcohol consumption was measured using the AUDIT-C questionnaire. This scale is composed of three questions each of which having four possible answers (i.e., a score ranging from 0 to 4; total scale score ranging from 0 to 12). Hazardous alcohol consumption was defined as having a AUDIT-C score ≥ 4 for men or ≥ 3 for women [22]. Binge drinking was defined as consuming six or more alcoholic drinks on any one occasion during the previous month.
Patients were asked about their consumption of the following psychoactive substances during the previous four weeks: cannabis, cocaine, crack, heroin, ecstasy, amphetamines, LSD. With regard to cannabis consumption, patients were classified into three categories: “No use”, “non-regular use” and “regular or daily use.” Tobacco smoking (smoking status, history of smoking, number of cigarettes per day) was assessed during face-to-face interviews with physicians for patients in the first and second recruitment phases and using self-administered questionnaires for patients from the third phase. Patients who were neither tenant nor owner of their housing were considered to have unstable housing. For the evaluation of liver fibrosis, we used the FIB4 index obtained by the following formula [23] Fib4index=Age(year)*AST(U/L)Plateletscount(109/L)*ALT(U/L) Severe fibrosis or cirrhosis was defined as a FIB-4 index score >3.25. A FIB-4 score between 1.45 and 3.25 corresponded to moderate fibrosis and <1.45 to no clinically significant fibrosis [24]. A FIB-4 score above 1.45 corresponded to significant fibrosis.
Statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Cary, North Carolina). The main characteristics of patients at HCV treatment initiation were compared between patients initiating PEG-IFN-based treatment (Phase 1) and those initiating DAA (Phase 3) (chi-square test for categorical variables, Student’s t test for continuous variables). We used data from the SAQ completed at the visit closest to the date of treatment initiation for patients receiving PEG-IFN treatment (Fig 1). For patients receiving DAA, we used data collected during the visit scheduled at the beginning of treatment. A multivariable analysis was then performed with logistic regression models, using a binary variable indicating the type of treatment received (PEG-IFN vs DAA) as the dependent variable. All significant variables with a p-value of <0.20 in the univariate analyses were introduced into the multivariable regression model to estimate the adjusted effect and derive the adjusted odds ratio (aOR) for each of the dependent variables. A 95% confidence level was also estimated for each aOR. In order to verify the robustness of the results and exclude possible recruitment phase bias, i.e. the possibility that patients recruited in the third phase had a different pattern of socio-behavioral characteristics from the patients in the first phase, a sensitivity analysis was conducted which included only patients followed up in clinical centers involved in all three phases of the cohort.
This study was approved by the "Comité de Protection des Personnes (CPP)" (Ref N° 2234 on April 19th, 2005) and the “Commission Nationale de l’Information et des Libertés de France (CNIL)” on September 6th, 2005. Participant provided signed, informed consent.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}