Neurocognitive Functioning is Impaired in Perinatally HIV-Infected Youth

1. Introduction

Perinatal Human Immunodeficiency Virus (PHIV)-infected youth appear associated with impairments in neurocognitive (Cohen et al., 2014; Nagarajan et al., 2012; Yadav et al., 2017), neurodevelopmental (Yadav et al., 2017), and psychiatric assessment performance (Chernoff et al., 2018; Willen et al., 2017). Global cognitive measures suggest that individuals diagnosed with PHIV perform below average in most cognitive domains including executive functioning, working memory, cognitive flexibility, motor performance, language, and speeded information processing (Cohen et al., 2014; Hermetet-Lindsay et al., 2017; Malee, Smith, & Mellins, 2016; Nagarajan et al., 2012; Robbins et al., 2019; Sirois et al., 2016; Willen et al., 2017). Compared to healthy controls, research suggests PHIV-infected-youth do not perform as well on measures of general cognition or visuospatial functioning (Laughton et al., 2013).

Imaging studies have identified neurological deficits and brain atrophy before the emergence of adolescence, when cognitive functioning, including memory, language, reading skills, processing speed, attention, concentration, executive control, planning, cognitive flexibility, problem-solving, and behavior rapidly develops (Laughton et al., 2013). Chronic exposure to a neurotropic infection during CNS development is a risk for PHIV (Smith & Wilkins, 2015). The more damage and dysregulation that occurs in the brain during illness onset, the more likely an individual’s cognitive reserve may be impacted (Basso & Bornstein, 2000). Although these young people continue to develop cognitively in the presence of illness (Ashby et al., 2015), they are subject to increased risk for developmental language impairment throughout adulthood (Laughton et al., 2013; Malee et al., 2016).

Research suggests PHIV-infected youth are at higher risk for psychiatric illness due to genomic, biological, environmental, psychosocial (Abrams et al., 2018; Laughton et al., 2013; Mellins & Malee, 2013; Smith & Wilkins, 2015), and cognitive issues (Hermetet-Lindsay et al., 2017; Laughton et al., 2013; Mellins & Malee, 2013) compared to undiagnosed individuals. Individuals with PHIV-infection often reported symptoms of depression, anxiety, behavioral difficulties, suicidal ideation/attempt, hyperactivity, attention difficulties, and sleep disturbance (Hermetet-Lindsay et al., 2017; Le Prevost et al., 2018; Malee et al., 2011; Mellins & Malee, 2013; Smith et al., 2019; Vreeman et al., 2015), which also result in high hospitalization rates for comorbid psychiatric disorders (Vreeman et al., 2015). Thus, depressive symptoms might impact developmental levels of functioning and neurocognitive development which might in turn impact functional outcomes into adulthood.

Evidence of neurodevelopmental, neurocognitive, and comorbid mental illness have been found in PHIV-infected youth. However, only a limited number of studies have used comprehensive assessment batteries that included measures of depression, sleep, neurocognitive deficits, and developmental level of intellectual functioning for PHIV-infected-youth with extensive treatment histories. Little is known about the long-term neurological effects of ART on brain development and functioning in PHIV-infected young adults (Laughton et al., 2013). Therefore, our study examined cognitive functioning and clinical features of viral pathogenesis and long-term early intervention ART treatment on PHIV-infected-youth.

A comprehensive neurocognitive battery was administered and included assessments of depression and sleep to assess the effects of these parameters on neurocognitive functioning. We studied young adults receiving long-term cART compared to age and gender matched healthy controls in an effort to better understand the long-term potential deficits of youth diagnosed with PHIV and to minimize potential confounding variables between groups. We hypothesized that PHIV-infected young adults would perform more poorly compared to healthy controls on most neurocognitive assessments and on assessments of mood and sleep. Also, we hypothesized that psychosocial impairments after a life-time of HIV infection would be substantial with many areas impacted by HIV and its treatment.

2. Methods

We report the findings from a study for which the primary aim was brain imaging in PHIV-infected-youth. As part of that study, a neurocognitive battery was administered using a cross-sectional, paired subject design.

3. Results

4. Discussion

Consistent with prior studies, our results demonstrate the existence of neurocognitive deficits between PHIV-infected-youth with early, prolonged, and current antiretroviral treatment and the paired healthy age and gender matched control group. Specifically, significant neurocognitive performance deficits emerged between groups on Overall MCCB Neurocognitive Composite Scores (p=0.003), Working Memory, Verbal Learning, Visual Learning, Social Cognition, Executive Functioning, Abstract Thinking, and a Global Deficit Score. After correction for multiple comparisons, both Verbal Learning and the Overall Neurocognitive Composite Score differences remained statistically significant (p<0.004). Our study corroborates previous findings that individuals diagnosed with PHIV have neurocognitive impairments (Cohen et al., 2014; Hermetet-Lindsay et al., 2017; Laughton et al., 2013; Malee, Smith, Mellins et al., 2016; Nagarajan et al., 2012; Nichols et al., 2015; Sarma et al., 2014; Sirois et al., 2016; Willen et al., 2017; Yadav et al., 2017). These deficiencies persist into adulthood despite antiretroviral therapy and most likely reflect the early infection of the central nervous system. Furthermore, additional support for neurocognitive deficits comes from the very large effect sizes for Speed of Information Processing, Verbal Learning, Social Cognition, Executive Functioning, Abstract Thinking, Overall Neurocognitive Composite Score, and Global Deficit Scores.

Our research indicated this is the first time a GDS score has been applied to detect neurocognitive impairments of PHIV-infected-youth. Although the GDS did not meet the multiple comparisons cutoff (Table 3) there was a large effect size which displayed significant differences between groups (d=1.32). Laughton et al. (2013), noted that global cognitive scores might overlook subtle individual domain impairments. A strength of the present study is that we utilized a comprehensive battery of assessments to better assess each neurocognitive domain and the findings are so strikingly large that it’s unlikely that a larger sample size would alter the current findings. Furthermore, to our knowledge this is the first study to examine sleep, abstract thinking, and developmental level of intellectual functioning to measure estimated early intellectual functioning and quantify the impact of HIV illness on intellectual functioning.

Our results suggest there are no significant differences between groups on Attention/Vigilance, Reasoning and Problem Solving, Visual Perceptual Delayed Recall, or Psychomotor Functioning. Nonetheless, our sample size is small and previous studies have reported differences in some of these domains. Cohen et al. (2014), found children (8 to 18-years-old) with PHIV performed worse on all neurocognitive domains assessed (i.e., Verbal IQ, Performance IQ, Processing Speed, Working Memory, Motor Functioning, Memory, and Executive Functioning) than controls. Furthermore, we also previously found a deficit in Attention / Processing Speed in 13 to 25-year-old adolescent PHIV-infected-youth². Some researchers (Laughton et al., 2013; Nichols et al., 2015; Willen et al., 2017) also found a significant difference between PHIV-infected-youth and matched healthy controls on measures of Executive Functioning; however, others (Laughton et al., 2013; Willen et al., 2017;) including our study, did not find significant differences on measures of Psychomotor Speed. Recently, Van den Hof et al. (2019), reported a neurocognitive longitudinal study of the NOVICE cohort PHIV as they progressed into adolescence over 4.6 years and demonstrated worsening executive function over time, while Robbins et al. (2019), evaluated children as they progressed into adulthood longitudinally and noted better performance on both Executive Functioning and Working Memory over time. Their results suggest the need for further studies to assess neurocognitive issues as PHIV-youth mature.

The verbal learning deficit was striking for our PHIV young adults who received prolonged and early treatment for HIV. With a p value of 0.000, the difference was still significant after correction for multiple comparisons. The Verbal Learning domain was one of only two domains for which the standard deviation was smaller for PHIV-infected-youth compared to controls, which supports how common the deficit in verbal learning is in our PHIV patients. For most domains, the standard deviation was higher for the patients than controls suggesting much more variation in the PHIV-infected-youth group. A confounding variable for our study could be the effect on language development of living within a bilingual environment or children born in other countries. We did not track this information since all of the participants were fluent in English. Others have also reported verbal impairments including Cohen et al. (2014), who studied PHIV in early adolescence (median age 13.8 years, p=0.043). Sirois et al. (2016), looked at associations of memory and executive functioning with academic and adaptive functioning. Verbal learning was statistically associated with measures of academic achievement at p<0.001 in PHIV. Adults who acquired HIV have also been noted to have verbal memory impairments (Wright et al., 2011). PHIV-youth performed significantly worse than healthy controls on developmental level of intellectual functioning, memory which were likely present prior to long-term early intervention ART intervention WTAR (t(10) = 4.357, p < .001). Impairments in intellectual functioning have also been noted in the literature for individuals with PHIV prior to ART intervention (Laughton et al., 2013). Therefore, we believe that these impairments are part of the condition of HIV and thus chose not to treat developmental level of intellectual functioning (WTAR) as a covariate in our analysis.

Similar to our study, Ireland (2011), compared a small sample of individuals diagnosed with HIV-Associated Neurocognitive Disorders (HAND) to controls and found early impairments in Social Cognition regardless of functioning on other cognitive domains. Louthrenoo, Oberdorfer, and Sirisanthana (2014), also found that adolescents with PHIV had higher social withdrawal compared to controls. In furtherance of these results, we examined difference in ethnicity (p-value = 0.086) and education (p-value = 0.476) with the results of the Fisher’s exact test, and found no significant difference between groups; thus, we concluded language development was equivalent across both groups. Deficiency in Verbal Learning and Social Cognition might have a significant impact on physician/patient interactions and enhanced awareness of this deficiency would be important to improve medical communication between patients and physicians. Morgan et al. (2019), also noted poorer healthcare interactions between those diagnosed with HAND compared to controls.

Recent studies have shown that other risk factors such as stressful life events, socioeconomic status (Malee et al., 2016; Mellins & Malee, 2013; Nichols et al., 2015; Smith & Wilkins, 2015), urban violence (Smith & Wilkins, 2015; Willen et al., 2017), disadvantaged neighborhoods (Abrams et al., 2018; Malee et al., 2016; Mellins & Malee, 2013), objective perception of their neighborhoods (Kang et al., 2019), familial instability (Smith & Wilkins, 2015), family history of mental illness (Mellins & Malee, 2013; Nichols et al., 2015), individual health status (Mellins & Malee, et al., 2013; Smith et al., 2019), low educational enrichment (Nichols et al., 2015), and the stigmatizing nature of HIV (Abrams et al., 2018; Smith et al., 2019; Willen et al., 2017) may also relate to neurocognitive deficits and mental health outcomes. Smith et al. (2019), reported that when PHIV youth were compared to HIV-exposed but uninfected youth, 36% of the youth had a previous or current mental health diagnosis with no significant HIV status group differences. Due to these potential risk factors individuals diagnosed with PHIV may have limited access to educational enrichment opportunities and these differences may also act as a potential confound when reporting neuropsychological outcomes.

Although there was no evidence of sleep difficulties, our PHIV-infected-youth were significantly more depressed than the controls, even though none of the patients were actively receiving psychiatric care for depression. These results are consistent with other studies of this population (Hermetet-Lindsay et al., 2017; Malee et al., 2011; Mellins & Malee, 2013; Smith et al., 2019; Vreeman et al., 2015) which suggest that undiagnosed depression could be an important issue of care for these patients. Le Prevost et al. (2018), showed that there were no differences in anxiety and depression between PHIV and control participants in an adolescent population. Depression and anxiety appear to be more prevalent than is clinically recognized. However, research suggests that co-morbid mental health diagnoses are common, and may not be related to HIV-infection status (Smith et al., 2019). Additionally, parent support decreases depression, as well as increases cognitive development (Laughton et al., 2013). Due to increased risk of mental health diagnoses, Smith et al. (2019) have recommended multi-faceted approaches that incorporate both individual and family participation in treatment programs.

5. Study Limitations

There are a number of limitations that should be considered when interpreting the generalizability of our results including our relatively small sample size (Sarma et al., 2021). Our study did not assess participants’ bilingual status, level of acculturation, or sociodemographic factors, which could confound our findings. Furthermore, we were also unable to match for race; however, our results did show that race was not deemed significantly pertinent to the results of this study. Although our results indicate PHIV-participants differ from our control group on various neurocognitive factors, these results may be less generalizable due to the potential confounding variables listed above.

6. Conclusions

Findings show that despite long-term cART treatment, neurocognitive deficiencies remain for PHIV-infected-youth into adulthood. Although the majority of our PHIV (7/11) were treated before the first birthday, even earlier treatment could enhance the neuropsychiatric and neurocognitive outcomes by aborting development of HIV encephalopathy or subclinical involvement of the central nervous system. The deficiency in the Verbal Learning domain was highly significant and withstood correction for multiple comparisons. Awareness of this defect could improve communication between patients and their medical care team. Furthermore, mental health diagnoses such as depression should be considered and addressed routinely.

Funding

This work was supported by research grant from the National Institute of Neurological Disorders and Stroke at National Institutes of Health Grant (NINDS) awarded to Dr. Sarma [R21NS090956].

Conflict of Interest / Competing Interests

None of the corresponding authors have any conflicts of interest to report.

Ethics & Consent to participate & publish

Informed, signed consent was obtained from each subject using an Institutional Review Board (IRB) consent form approved by the Los Angeles Biomedical Research Institute at Los Angeles County Harbor-UCLA Medical Center (now Lundquist Institute). Both the Los Angeles Biomedical Research Institute and UCLA gave IRB approval which also included MRI/MRS imaging, neurocognitive assessment, and a detailed review of the patient subjects’ medical records (see Sarma et al., 2021).

Data Availability Statement

The authors do not have permission to share raw subject data.

Code Availability

The authors are able to share methods of coding and analyzing data

Authors’ contributions

TW, MK, MKS, JS, JV, and MAT, contributed to the design of design and the conduct of the study. TW, MK, MKS, JH, JV, and MAT, contributed to data collection. TW, MK, MKS, IW, DEM, JH, KN, JD, AK, EO, JAC, MAT, and JV, contributed to patient-related coordination and management of recruitment and enrollment. TW, MK, JS, and JV contributed to data analysis and interpretation. All authors contributed to writing the manuscript.