HCV-infected women were older than HIV-monoinfected women, and both groups were older than women with neither infection. More than half of all women were African American, with the largest proportion among the HCV-infected group. Compared with HIV-infected women and those with neither infection, HCV-infected women were more likely to report being in menopause, having a history of injection drug use, and smoking currently and also more likely to have diabetes and to have a higher homeostasis model assessment–estimated insulin resistance value but a lower low-density lipoprotein level. HIV-infected women had lower high-density lipoprotein levels and higher triglyceride levels than HIV-uninfected women. Compared with HIV-monoinfected women, HIV/HCV-coinfected women had slightly lower CD4 counts and higher current HIV viral loads.To our knowledge, our study is the first to examine liver fibrosis using ultrasound-based TE measurement of liver stiffness in a geographically and ethnically diverse cohort of US women with well-characterized metabolic parameters. We found that HIV/HCV-coinfected women but not HIV-monoinfected women had higher liver stiffness values than those with neither infection. Greater waist circumference, a marker of central obesity, was associated with liver stiffness in HCV-infected women, but not HIV-monoinfected women. Among HIV-infected women , a history of AIDS was associated with greater liver stiffness. Our finding that HIV-monoinfected women did not seem to have higher liver stiffness values than uninfected women was unexpected.
HIV infection per se has been shown to infect hepatic stellate cells ,metal greenhouse benches which in turn can lead to hepatic injury. The authors of that study suggested that HSC activation would be required in order to infect HSCs and stimulate expression of collagen and proinflammatory cytokines that promote fibrosis. In our study, >50% of our women had undetectable HIV RNA levels. Furthermore, in analyzing HIV-moninfected women only, we observed an association of liver stiffness with higher HIV RNA levels and a history of clinical AIDS . There are 2 possible reasons for the difference between our finding of a lack of association between HIV monoinfection and liver fibrosis and findings of other studies directly comparing HIV-monoinfected adults with adults with neither HIV nor HCV infection. First, Price et al studied HIVmonoinfected men not receiving highly active ART, and Blackard et al studied HIV-monoinfected women seen in the HERS cohort from 1993 to 2000. These studies therefore included patients that were more likely to have uncontrolled viral replication and lower CD4 counts and to be receiving antiretroviral agents associated with hepatic steatosis or liver injury, such as stavudine, didanosine, and zidovudine. In both studies, when the analysis was limited to those with HIV monoinfection only, greater immunosuppression was associated with greater fibrosis. Second, those studies used indirect serum markers of fibrosis APRI and FIB-4, which are derived from clinical laboratory values that might be altered by HIV infection itself. Vermehren et al found that whereas 37% of HIV-monoinfected patients had significant liver fibrosis or cirrhosis, as indicated by the serum fibrosis marker Fibrotest , only 21% had significant fibrosis as predicted by TE.
That study concluded that Fibrotest estimated much higher levels of fibrosis than TE in HIV-monoinfected patients. Examination of the association of HIV and HCV status with indirect serum markers of fibrosis and direct serum markers relative to TE measured liver stiffness are currently underway. It is noteworthy that waist circumference was associated with liver stiffness in our HCV-infected patients but not in HIVmonoinfected patients. Studies in HCV-monoinfected persons have shown that central obesity is a strong predictor of fibrosis, a stronger predictor than BMI, consistent with our findings. In HIV-monoinfected patients, DallaPiazza et al did not find an association of obesity with liver fibrosis estimated by the APRI. That study did not examine the association of waist circumference with liver fibrosis. Taken together, our findings and others suggest that central obesity further worsens underlying HCV-associated liver injury. By contrast, in the setting of little injury to the liver, the adverse effects of obesity may be less apparent. One question that needs study is whether hepatic steatosis, thought to be a consequence of excess visceral adipose tissue, is a mechanism by which rapid fibrosis progression occurs in the setting of HCV infection. In our study we found that daily marijuana use was associated with less liver fibrosis, with associations reaching statistical significance in HIV-monoinfected and control participants. By contrast, previous studies in HCV-monoinfected patients found daily marijuana use to be a strong risk factor for fibrosis progression. This discrepancy may be partly explained by the fact that our study included few women with significant fibrosis, and marijuana use was more common in women without HCV infection.
Findings of prior studies have suggested that cannabis may have little or no influence on the initiation of fibrosis but may only be an important cofactor in fibrosis progression when fibrosis is already present. Although we also found that Hispanic race was associated with greater liver stiffness in HIV-monoinfected women, the number of women of Hispanic race in our cohort was low, and therefore our findings should be interpreted with caution. Nevertheless, studies in the general population have shown that, compared with African Americans and whites, Hispanics have a greater prevalence of hepatic steatosis and are at greater risk of nonalcoholic fatty liver disease–associated cirrhosis. Our study has some important limitations. First, the cross sectional design limits our study to evaluation of exposures and disease status at one point in time. It is theoretically possible that liver fibrosis preceded our studied exposures; this is especially true of metabolic markers. Second, the proportion of patients with significant fibrosis in our HIV-monoinfected patients was small and limits our statistical power for identifying potential risk factors. However, the advantage of TE in providing a continuous measure of liver stiffness allows for broader study of fibrosis progression, producing a more robust evaluation of the factors associated with liver disease progression. Finally, the use of TE in our cohort has some limitations, including the exclusion of women that were more often obese. The exclusion of women with a BMI >35 kg/m2 may also have limited our ability to detect an effect of central obesity with fi- brosis in the HIV-monoinfected and control women. A detailed analysis comparing serum biomarkers of fibrosis with TE in the same study population is underway. Using TE to estimate fibrosis in a large cohort of women with HIV monoinfection, HIV/HCV coinfection, HCV monoinfection, and neither infection, we conclude that HCV but not HIV infection is associated with increased liver fibrosis. Central obesity further worsens liver fibrosis in HCV-infected women, but its effect in the absence of HCV-associated liver injury seems to be lessened. The mechanisms by which central obesity may worsen liver fibrosis in the setting of underlying HCVassociated liver injury need further investigation, including study of the contribution of hepatic steatosis and inflammation to liver fibrosis.
Neurocognitive dysfunction is a hallmark feature of schizophrenia and, to a lesser extent, of other psychoses; a conceptualization originating roughly 100 years ago with Kraepelin and Bleuler. There is ample evidence of significant but milder impairments during the premorbid phase,rolling greenhouse tables greater deficits during the prodromal or clinical high risk period,culminating in relatively severe deficits in the first episode and chronic phases. This suggests an evolution of neurocognitive dysfunction in individuals developing psychosis, especially schizophrenia. The CHR20 period is of considerable interest because it offers a temporal window into the changes occurring during the “near-psychotic” state, before confounders such as chronicity and long-term medication use cloud the picture. A substantial body of neurocognitive research in CHR populations has been summarized in a number of meta-analyses. “Small-to-medium effect size impairments across most neurocognitive domains studied and small-to-large ESs in those who convert to psychosis ” have been reported14 . Verbal memory and processing speed have emerged as relatively strong predictors of psychosis. However, small samples, different measures, and variable reporting of sample characteristics limit the reliability of these findings. In this second phase of the North American Prodrome Longitudinal Study , we assessed the largest CHR sample to date. First, we sought to identify the key neurocognitive functions impaired in the CHR stage, especially in those who later convert to psychosis. Descriptions of schizophrenia place considerable emphasis on the centrality of dysfunctions in attention and working memory. Evidence of severe deficits in declarative memory has more recently emerged in first episode and CHR14, samples. Olfactory identification deficits have also been touted as a possible risk factor32,33 and processing speed and general cognitive ability have been shown to be robustly impaired in persons who later develop schizophrenia.We chose to provide extensive coverage of neurocognitive dimensions thought a priori to mark the evolution into frank psychosis. Second, we investigated if the neurocognitive profiles were characterized by a general deficit syndrome or specific impairments. This is of particular relevance for those individuals who transition to psychotic disorders as it provides critical information about the nature of neurocognition in the earliest phase of psychosis36. We hypothesized that the CHR+C group would be characterized by especially salient deficits against a background of general impairments. Third, we examined differences between medicated and unmedicated CHR individuals. Many of these young people take a range of medications including antipsychotics.Such medications could improve or impair cognition idiosyncratically. Prior CHR neurocognitive studies have not systematically addressed medication status.
The large sample in NAPLS-2 enabled an investigation of a sizeable subgroup of CHR+C individuals who have never been medicated, and thus help to identify an unadulterated picture of neurocognitive function. Finally, we explored the potential usefulness of neurocognition for predicting transition to psychosis. While it is unlikely that neurocognitive measures will be highly predictive by themselves of conversion to psychosis, in part because they are impaired in many neuropsychiatric disorders, knowing their relative sensitivities in combination with clinical features may help in the real-world prediction of psychosis or disability . NAPLS-2 is a consortium of eight programs studying the psychosis prodrome in North America, as in NAPLS-1. The methodology and clinical features of the NAPLS-2 study are detailed elsewhere. From a sample of 764 CHR participants and 279 healthy controls ranging in age from 12–35, 689 CHR and 264 HC participants provided baseline neurocognitive data. The study protocols and informed consents were approved by the ethical review boards of all sites, and all procedures comply with the ethical standards of the relevant committees on human experimentation and with the Helsinki Declaration, as revised in 2008. The CHR sample met the Criteria of Prodromal Syndromes 20, based on the Structured Interview for Psychosis Risk Syndromes 20, or if under age 19, criteria for schizotypal personality disorder or COPS. Participants were excluded if they had a lifetime Axis I psychotic disorder, estimated IQ’s < 70 on both measures of IQ, a central nervous system disorder, or DSM-IV substance dependence in the past 6 months. Other nonpsychotic DSM-IV disorders were not exclusionary unless they clearly caused or better accounted for prodromal symptoms. Antipsychotic medications were allowed, provided there was clear evidence that psychotic symptoms were not present when the medication was started. HCs could not meet criteria for any prodromal syndrome, current or past psychotic or Cluster A personality disorder, or have first-degree relatives with a history of psychotic disorder or psychotic symptoms. For some analyses, we used a rescaled sum of unusual thought content/ delusional ideas and suspiciousness/persecutory ideas items from the SIPS positive symptoms. Transition to psychosis was determined by meeting SIPS Presence of Psychotic Symptoms criteria. Assessments were at baseline, 12 and 24 months. Current alcohol and marijuana use was assessed with the Alcohol and Drug Use Scale . The Calgary Depression Scale for Schizophrenia was used to assess depression. The neuropsychological battery was designed to cover a range of functions using well established clinical neuropsychological tests, as well as experimental measures of sensory, perceptual, or cognitive functions hypothesized to be important indicators of CHR status or conversion to psychosis. These included the MATRICS battery, the Wechsler Abbreviated Scale of Intelligence for general intellectual ability and the Wide Range Achievement Test-4 reading task to estimate premorbid ability. Experimental measures included the Babble test , the University of Pennsylvania Smell Identification Test for olfactory identification, a visual and verbal paired associate memory test , and three auditory attention & working memory continuous performance tests .