Difference in the neurocognitive functions of WLWH and MLWH in an Italian cohort of people living with HIV

There are several issues relating to the HIV care that make it important to differentiate women living with HIV (WLWH) from men living with HIV (MLWH) (Geretti et al. 2017). Indeed, sex-specific factors may impact on behaviours and conditions of PLWH, affecting access to diagnosis and treatment, clinical management, treatment response, and the overall wellbeing (Rosin et al. 2015; Lundgren et al. 2015; Soon et al. 2012; Burch et al. 2016; Ribaudo et al. 2016). For instance, WLWH demonstrated lower adherence to antiretroviral therapy (Puskas et al. 2011) and, despite a lower viral load, they did not show appreciable benefits in terms of disease progression compared to MLWH (Perez-Elias et al. 2014; Law et al. 2015; Napravnik et al. 2002). Regarding studies on HIV and cognition, neurocognitive studies focusing on WLWH are increased since the introduction of effective combined antiretroviral therapy (cART) (Maki et al. 2018; Sundermann et al. 2018; Rubin et al. 2020).

Available data suggest that WLWH have some cognitive and emotional disadvantages compared to MLWH. A recent systematic American review assessed whether WLWH are more cognitively vulnerable than MLWH (Rubin et al. 2019) and concluded that few HIV studies are adequately powered to address sex differences in prevalence and pattern of HAND. However, several studies indicate that WLWH have a higher prevalence of HAND (Sundermann et al. 2018; Royal et al. 2016) and a lower performance or higher impairment of memory and learning (Kabuba et al. 2016; Royal et al. 2016), speed of information processing (Maki et al. 2018; Do et al. 2018) and motor abilities (Maki et al. 2018; Burlacu et al. 2018) than MLWH. Only one study reported lower performance of WLWH in executive functions (Maki et al. 2018), and no difference was observed for fluency or attention/working memory (Do et al. 2018; Sundermann et al. 2018; Kabuba et al. 2016; Burlacu et al. 2018; Royal et al. 2016; Behrman-Lay et al. 2016).

Several factors seem to contribute to documented sex differences. Among biological factors (e.g., inflammation, hormones, genetics) (Rubin et al. 2019), there is evidence of an association between testosterone insufficiency and cognitive complaints (Laan et al. 2019) and of sex differences in inflammatory biomarkers (Ticona et al. 2015; Fitch et al. 2013; Martin et al. 2013; Looby et al. 2016). The higher prevalence of HAND in WLWH compared to MLWH could also be linked to differences in other variables, such as socio-economic conditions and education (Basso and Bornstein 2000; Farinpour et al. 2003). These factors could be some of the cognitive reserve determinants prior to HIV infection and might contribute to greater cognitive vulnerability after infection (Tsai and Burns 2015; Singer 1994).

In fact, in a recent study, Rubin et al. (2020) suggested that despite different cognitive profiles according to sex, the most discriminative factors were similar between men and women and included reading level (cognitive reserve), current and nadir CD4 count, plasma HIV viral load, duration of HIV disease, age, depressive symptoms, and race/ethnicity.

Furthermore, mental health disorders might actually contribute to higher rates of HAND in WLWH versus MLWH. Indeed, in a recent review, Geretti et al. (2017) suggested that WLWH experience a disproportionate burden of mental health issues (Orza et al. 2015) and face very different emotional challenges upon diagnosis (Loutfy et al. 2013). Specifically, the higher frequency of cognitive impairment in WLWH compared to MLWH might be related to the higher prevalence of depression (Bayón et al. 2012; Morrison et al. 2002; Rabkin et al. 1991; Semple et al. 1996; Aljassem et al. 2016). In fact, depression is negatively associated with cognitive functioning in PLWH and both a history of depression and current depression are associated with HAND in PLWH (Yousuf et al. 2019). Several international studies suggest that depression is almost twice as common in women as in men (Yousuf et al. 2019; Fantahun et al. 2018; Antelman et al. 2007) and that WLWH may be more cognitively susceptible than MLWH to the effects of mental health factors (Rubin et al. 2019, 2020). In particular, Rubin et al. (2020) indicated that HIV comorbid with depression affects specific cognitive domains, including cognitive control, and that these effects are largest in women. Other studies reported that depression is associated with deficits in learning, memory, and attention in WLWH (Maki et al. 2018; Rubin et al. 2016, 2014, 2015) and that women with elevated depressive symptoms seem to have a three times greater risk of impairment in executive control/inhibition than MLWH with elevated depressive symptoms (Rubin et al. 2019).

However, to date, few studies have been published on this topic, especially in Italy where literature recounts inconsistent results in PLWH. For example, Focà et al. (2016) reported, in contrast with other previous studies, that MLWH might be most at risk for neurocognitive disorders than WLWH.

Thus, as significant knowledge gaps remain, additional studies are needed to deepen this subject and improve management of PLWH.

Our aim was to compare the cognitive performance of MLWH and WLWH in an Italian Cohort of People Living With HIV and to analyse factors potentially contributing to sex differences in cognitive function.

The aim of this study was to analyse the cognitive profile of MLWH and WLWH in an Italian cohort of PLWH and to examine the effect of mediating variables on the relationship between sex and cognitive function. Indeed, based on the available literature, WLWH seem to have cognitive and emotional disadvantages (Rubin et al. 2019; Bayón et al. 2012; Morrison et al. 2002). However, a significant knowledge gap remains on this topic especially in Italy where few studies have been published with inconsistent results (Focà et al. 2016).

Overall, the WLWH in our cohort of patients had a lower level of education and longer time from HIV diagnosis and from first cART compared to MWLH. In line with previous evidence (Sundermann et al. 2018; Royal et al. 2016), our findings show that WLWH demonstrate lower global cognitive performance than MLWH and a higher proportion of subjects with probable depression. However, our WLWH did not manifest a significantly higher frequency of HAND compared to MLWH, in agreement with the Frascati criteria (Antinori et al. 2007). This discrepancy could be explained by the fact that WLWH demonstrate mild impairment than MLWH across domains that does not typically meet the threshold of impairment contributing to a HAND diagnosis.

In this regard, discrepant data emerged from previous observations. For example, Sundermann et al. (2018) found a higher prevalence of HAND in WLWH compared to MLWH; however, this difference was not confirmed by adjusting for reading level and ethnicity. On the other hand, from Rubin et al.’s review (2019), it emerged that only 43% of the studies examined demonstrated that the prevalence of HAND was higher in WLWH. Considering cognitive domains individually, we found additional differences. First, WLWH performed worse than MLWH in the executive skills domain, confirming previous observations (Maki et al. 2018). Furthermore, our WLWH sample also performed worse than MLWH on the attention tasks; note, however, that in this domain, there are inconsistent results across studies (Do et al. 2018; Sundermann et al. 2018; Kabuba et al. 2016; Burlacu et al. 2018; Royal et al. 2016; Behrman-Lay et al. 2016) probably due to methodological differences.

However, we identified other demographic and clinical factors besides sex that could help explain the aforementioned difference between WLWH and MLWH. Indeed, our results showed that there was no statistically significant direct effect between sex and GCP, attention and executive skills. On the contrary, the associations between male sex and better GCP and attention skills were mediated by higher level of education and fewer depressive symptoms. Moreover, a lower Zung depression score mediated the association between male sex and better executive skills.

Thus, as suggested in previous studies, we confirm that depressive symptoms seem to be more prevalent in WLWH than in MLWH (Aljassem et al. 2016) and might contribute to higher cognitive impairment (Bayón et al. 2012). Furthermore, we confirm several data reported in the literature regarding predictive factors for cognitive performance in PLWH such as educational background (Tedaldi et al. 2015; Milanini et al. 2016).

Indeed, we corroborate the finding that adjusting for critical factors such as education, clinical variables, and mental health reduces the dimension of the sex difference and elucidates factors contributing to NCI in women (Rubin et al. 2019). Moreover, our findings support those of Sundermann et al. (2018), suggesting that discrepancies in biopsychosocial factors (e.g., reading level that reflect education quality) may explain the sex difference in NCI regarding PLWH.

We acknowledge that our study has some limitations. First, our conclusions are limited by the small sample size of the study. Indeed, due to the small number of patients who were submitted to a completed SNB in the dataset used for the retrospective analysis, our study reached a low statistical power. Thus, in future studies, it will be important to extend the sample to confirm the generalizability of the results achieved.

Second, this is a cross-sectional study and uncontrolled biases can occur in routine clinical practice. Thus, future longitudinal studies are needed to confirm our findings also given the fact that only one longitudinal study to date has examined the risk of neurocognitive decline in WLWH and MLWH (Heaton et al. 2015).

Lastly, in future investigations, it would be useful to include a healthy control group to compare differences in cognitive functions between women and men and between PLWH and the general population. In addition, our WLWH and MLWH subgroups were not matched for education, years from HIV diagnosis and years from first ARV; future investigations that include matched groups are needed to check the probable influences of these disparities on cognitive performance.

In conclusion, taken together, our findings suggest that in our sample, WLWH demonstrate lower global cognitive performance, attention, and executive skills compared to MLWH. However, other demographic and clinical factors might be able to explain these disadvantages in WLWH besides sex, in particular the lower level of education and more depressive symptoms. On these bases, in planning cognitive interventions, it seems appropriate to pay particular attention to WLWH, but it would also be helpful to monitor demographic, clinical, and mental health factors to identify the patients who are most in need and to detect cognitive impairments as soon as possible and intervene to improve the psychological wellbeing of PLWH.