Introduction

It has been known from early on in the current pandemic that human immunodeficiency virus (HIV) infects cells associated with the central nervous system (CNS) and that the CNS serves as a viral reservoir, even with effective combination anti-retroviral therapy (ART) [1, 2]. While the incidence of HIV-associated dementia (HAD) has decreased, HIV-associated neurocognitive disorders (HAND) are, and will continue to be, a significant public health problem, particularly where treatment is harder to access. It is thought that 15 % of those with HIV have a related dementia, and anywhere from 30 to 60 % have less severe cognitive impairment [3, 4]. From onset of infection, HIV stows away in immature monocytes to cross the blood–brain barrier, and it is thought that its replicating presence in the brain triggers a cascade of inflammatory reactions that include toxins damaging to the micro-vascular endothelium that can further disrupt the blood–brain barrier [5–7]. This inflammation causes HIV encephalitis, the primary infection of HIV that continues to chronically fester, even in those with undetectable viral loads and adequate immune markers [1, 8•]. HIV protein gp120 not only inhibits stem cells in the brain from producing new nerve cells [7], but in neuronal cells, gp120 induces caspases, mitochondrial-death proteins that may influence the up-regulation of the death receptor Fas, leading to apoptosis [9].

HIV proteins gp120 and Tat can bind to and interfere with the dopamine transporter (DAT) system and increase both dopamine levels and oxidative stress in the brain, which leads to further dysfunction and impairment [10••, 11•]. Substances of abuse, like cocaine, methamphetamines, and opioids, commonly used by HIV-positive individuals, can also increase oxidative stress, increase blood–brain barrier permeability, and worsen the neurotoxic effects of gp120 [11•]. Dopamine levels are increased in the brain from most drugs of abuse [12–15]. This exposure increases dopamine well beyond the levels derived from more natural satisfactions like food and sex [16–19]. Dopamine has been clearly shown to induce HIV replication, and researchers have found that dopamine-mediated mechanisms involved in HIV replication may depend on the stages of HIV life-cycle, starting with viral attachment to dendritic cells, monocytes, and lymphocytes through HIV reverse transcription to HIV transcription [10••], providing researchers with another moving target.

HIV-Associated Neurocognitive Disorders

HIV alone can induce neurocognitive and neuropsychiatric disorders at almost any point in the infection, whether or not otherwise symptomatic. Even those with good viral suppression and well functioning immune systems can have cognitive problems, from a mild cognitive impairment (MCI) to severe disabling dementia and everything in between [8•]. Individuals with HAND can have motor dysfunction, speech disorders, and behavioral changes, with slow mental processing and poor memory and concentration. Motor problems range from loss of fine motor control to tremors and poor balance. Behaviorally, patients have apathy, lethargy, and reduced emotional responses, with a lack of spontaneity. The brain histopathology of HIV reveals many changes, including neuronal loss and frank structural atrophy [1, 5, 20].

Cognitive Effects of Drug Use

Recreational and dependent drug use has long been associated with adverse cognitive effects, both in the short and the long term [21]. Commonly abused substances like alcohol, the opioids, other sedatives, the stimulants, marijuana, or more exotic substances like hallucinogens, combinations, or so-called ‘designer drugs,’ synthetic cannabinoids, etc., have almost all been associated with either cognitive or neuronal dysfunction or laboratory-derived brain damage on mammals. [22–31].

Confounding and Complicating Factors

Multiple variables can have adverse effects on cognition, and it is known that the many co-morbid risk factors in HIV-positive individuals have differential effects relative to age, HIV disease severity, and other significant indicators. Researchers have recently found that cognitive risk differs by age, and cognitive reserve capacity (educational attainment or pre-morbid intelligence) is the best predictor of neuro-cognition [32••]. About 20–25 % of those living with HIV acquired the infection either directly or indirectly because of drug use, whether the virus was blood borne, or sexually or vertically transmitted [33]. Links are clear between substance use and high-risk behaviors in non-injecting drug-using men and women, especially for stimulant users [34, 35•, 36]. Impaired states influence sexual behavior and lead to risky practices that increase risk of HIV exposure. Injection of drugs can spread HIV, and use of alcohol, stimulants, and other drugs has been associated with disinhibition, risky sexual behaviors, and non-adherence to medical recommendations [37].

Other sources of CNS-mediated signs and symptoms that adversely affect cognition can confound the picture. Many HIV-infected people have psychiatric reactions or exacerbations or psychological responses to being infected (depression, anxiety, agitation, suicidal ideation); others have co-morbid conditions, either acquired or pre-existing, including substance use disorders (SUDs), co-infection with hepatitis C virus (HCV) or other infections or malignancies [38••]. Individuals around the world with HIV are an aging population and are confronting the chronic illnesses of the older adult, like cardiovascular disease, diabetes, arthritis, neoplasms, dementia, and an even further decline in the immune system that often accompanies old age [32••, 39].

Ellis and colleagues [4, 40] found that the historically lowest CD4 count predicts cognitive impairment and is associated with the diagnosis of HAND [41]. ART is considered a positive advancement since it supports and assists the immune system and reduces peripheral viral load; however, there is evidence that the ART used to treat HIV may be responsible for exacerbating common illnesses, especially heart, hepatic, and renal disease [42]. The comorbidities themselves can cause neuro-inflammation that can further worsen cognitive functioning [43]. Aging itself may also contribute to worsening of HIV disease [44, 45]. Some ARTs have greater brain penetration than others and have been associated with hallucinations and other symptoms associated with increased dopamine [46, 47•]. While there has been a reduction in incidence of HAD since ART, prevalence remains high and studies show that, despite adequate immune reconstitution and virologic control, cognitive disorders persist in many patients [48, 49]. It is estimated that over 1 million people in the USA have HAND [50•], and there is mounting evidence that HIV infection will be the leading cause of early age dementia worldwide [8•, 51, 52]. Genetics of the host and the infectious agent also present active variables that affect cognition and may impact whether someone would have dementia without HIV [53•, 54, 55]. There is also evidence that different clades of HIV can have a differential effect on cognition due to differing virulence and pathogenesis [56••].

Understanding the intersection of the effects of SUD on cognition in those with HIV infection can be complicated by the frequent use of multiple substances and variable and inconsistent durations of use. Brain polymorphisms can change with drug use, and it has been shown that studies planning to examine genetic risk for HAND in drug users should carefully assess substance use patterns because the neurobiological substrates related to cognition in HIV-positive individuals vary with tonic changes of chronic substance exposure [57].

Neuro-psychological measures for the study of cognition have been standardized over many years and are valid and reliable; however, the study of cognition in drug users with HIV has numerous methodological issues. Quantifying the measurement of actual drug use can be problematic related to poor recall, uneven patterns of use, and, sometimes, lack of full disclosure in self-reported drug use. Multiple drug use is also common, and sorting out what effect is from what substance in poly-drug users presents its own problems. Other complicating co-factors that frequently accompany drug use include lower socioeconomic status, nutritional abnormalities, lower education levels, histories of head injury or other neurological compromise, and higher rates of psychiatric disorders, all of which can complicate interpretation of the effects of drug use [58].

Given this context, herein we review recent relevant literature that has evaluated cognition in those with HIV infection and the effects of specific substances and present the most salient findings.

Drugs of Abuse

Alcohol

It is well known that alcohol adversely affects many brain functions, whether when an individual is intoxicated, in withdrawal, or experiencing sustained damage from chronic, heavy use. The behaviorally disinhibiting and judgment-impairing effects of alcohol can lead to acquisition of HIV or other sexually transmitted infections. Specific cognitive effects are well documented and Wernicke-Korsakoff’s dementia was described over 100 years ago in 1897 [59]. Alcohol has some specific preferences for damage to cerebellar and prefrontal structures and the limbic system [60, 61]. HIV itself shows an affinity for structures in the basal ganglia and the associated frontal-striatal pathway [62, 63]. A few studies looked at whether the cognitive effects of both HIV and alcoholism are synergistic and found that those with both disorders are more compromised than those with either HIV or alcoholism alone and found independent adverse effects on cognition from both HIV and alcohol. Patients with both conditions often show greater deficits in visual-spatial perception, verbal reasoning, episodic memory, motor, and reaction times than individuals with either HIV infection or alcoholism alone, and this has suggested a synergistic effect [62, 64–66]. Alcohol-use disorders are common in those living with HIV and in injection drug users (IDUs) and, as is well documented, heavy alcohol use has many adverse effects on people living with HIV, including increased risk of HIV transmission to others by increased risk behaviors, poor retention in care, poor adherence to treatment, impaired ability to suppress HIV, and acceleration of hepatic fibrosis, especially if there is co-infection with HCV and in the setting of prescribed ART [67–70].

Stimulants

Cocaine

Many researchers have found that stimulants enhance HIV infection and its progression [71•, 72, 73]. Cocaine increases cell toxicity in astrocytes, accelerates inflammation in microglia, and increases blood–brain barrier permeability [74]. Cocaine-induced neuropsychological impairment is well documented, even though some studies have not supported such deficits. Difference in outcomes has been attributed to varying amounts of cocaine use and methodological differences in testing. Dose-dependent deficits were found in executive functioning, short-term memory, concentration, manual dexterity, psychomotor speed, and visuo-perception, even after weeks of abstinence [75, 76].

In HIV-positive individuals, cocaine has clear pathophysiological effects: it increases HIV replication, enhances HIV-infected monocyte movement into the brain, increases the toxicity of certain HIV viral proteins, and accelerates disease progression [77]. While it is fairly clear that either cocaine use or HIV infection independently lead to neurocognitive impairment, study results have been inconsistent in determining an interactive effect of cocaine and HIV infection [78••]. However, Meyer et al. [77] did find significantly impaired verbal learning and memory in HIV-positive daily/weekly crack cocaine users as compared with cocaine-using HIV-negative subjects. However, long-term neurocognitive effects of cocaine in abstinent HIV-positive individuals has not been well studied.

Methamphetamine

Methamphetamine is highly neurotoxic, resulting in damage to the axon terminals of dopaminergic neurons and depleting both dopamine and norepinephrine, and even leading to a Parkinsonian-like choreatic syndrome in some individuals [79]. HIV infection and methamphetamine use appear to have a synergistic effect, and multiple investigators have documented increased neurocognitive impairment in HIV-positive methamphetamine users as compared with control groups. Rippeth et al. [13] and Rodriquez et al. [80] found significantly worse global neuropsychological impairment in a combined HIV/methamphetamine group, and Carey et. al. [81] noted that immunosuppression appeared to also play a role: the worst neuropsychological performance was found in HIV-positive methamphetamine users with CD4 < 200 (as compared to less immunosuppressed HIV/methamphetamine users and all non-methamphetamine HIV-positive users). Carey et al. [81] postulated that synergistic damage to fronto-striatal circuits, which have already been observed separately in HIV and methamphetamine users, could be an explanation for these cognitive deficits.

Imaging studies have revealed reduced hippocampal volume and greater damage to frontal cortex interneurons in HIV-positive methamphetamine users as compared with control groups [76, 82–84]. Reduced N-acetyl aspartate in the anterior cingulate has also been observed [85]. Recently, there has even been evidence for a biologic marker: a specific dopamine receptor D3 polymorphism has been associated with higher rates of cognitive impairment in men with HIV and methamphetamine dependence [86•]. Among the drugs of abuse, methamphetamine appears to be particularly deleterious for cognition in HIV-infected individuals [71•].

Opioids

HIV-negative opioid users have significant neurocognitive impairment during regular use and in early withdrawal, including deficits in episodic and working memory, attention, and executive function. These deficits seem to recede when re-evaluated in the weeks to months after withdrawal, although a few studies have found persisting impairment in executive function (nonverbal flexible thinking, shifting cognitive sets, planning strategies) or subtle deficits in impulse control [87, 88]. There are less data regarding effects in HIV-positive individuals, but one study of HIV-positive active intravenous heroin and cocaine users did find significant impairment in verbal memory, problem solving, multitasking, visual-motor coordination, and cognitive flexibility [89]. The authors argued that these are important skills for engaging in treatment programs for SUD and HIV, and hence contribute to poorer outcomes for this group. There are also data to suggest that ongoing heroin use is correlated with increased neopterin concentration and hence increased progression to acquired immunodeficiency syndrome (AIDS), which in turn could exacerbate HAND [90]. There is a paucity of research regarding the long-term effects of opioids in subsequently abstinent HIV-positive individuals.

Opioid Substitution Treatment or Medication-Assisted Treatment

In general, individuals receiving methadone maintenance have been found to have impairments in attention, information processing, visual and verbal memory, and problem solving compared with non-opioid-using control groups [79]. HIV infection appears to exacerbate the problem. One study found that even when patients had asymptomatic HIV, the methadone-maintained group was more impaired than the control groups (both opioid-abstinent HIV-positive patients and methadone-maintained HIV-negative patients) [80]. However, a caveat is in order: the methadone-maintained HIV-positive patients were arguably the most burdened group, with fewer years of education, higher viral loads, more significant immunosuppression, and less compliance with anti-retroviral regimens. Although causation cannot be established, the authors note that methadone has been found to have immunosuppressive effects in HIV patients and to activate HIV replication in macrophages and microglial cells of the CNS [23].

Most data indicate that individuals maintained on buprenorphine show less cognitive impairment than those receiving methadone, although crossover studies comparing subjects who switch from methadone to buprenorphine and vice versa are needed to more accurately assess independent cognitive effects [91]. Less is known about the effects of buprenorphine on cognition in HIV-positive individuals.

Cannabis

Studies have not consistently found neuropsychological deficits in HIV-positive long-term users of cannabis, but the varied results may be due to study design [92]. If subjects are included with a history of cannabis use, rather than recent or current use, deficits may have resolved with abstinence. Pope et al. [83] found resolution of neuropsychological impairments after 1 month of abstinence in adult HIV-negative heavy cannabis users. Grant et al. [93] performed a meta-analysis of neurocognitive studies in HIV-negative heavy cannabis users, and found a small negative effect of cannabis in the domains of learning and recall. However, they note that some studies included users who were only abstinent for hours or days, so they may be recording deficits that would have resolved after several more weeks of abstinence. There is a larger concern for persistent deficits in short-term memory and attention in adolescent cannabis users, although these data have not been entirely consistent either [94•, 95, 96].

Cristiani et al. [97] did find poorer performance on memory tasks for daily cannabis users with symptomatic HIV as compared with those with asymptomatic HIV and a sero-negative control group, and the amount of impairment appeared to increase across the stages of HIV disease. The authors concluded that cannabis affected users more as the HIV disease progressed; they postulated that subtle effects may become more significant in an immune-compromised person. However, it is important to note that findings were mixed, as researchers did not require a period of abstinence, cannabis use was determined by self-report, and there was not a significant effect on cognitive performance for cannabis users in all groups.

In a group of asymptomatic individuals with HIV, Chang et al. [98] did not find any neuropsychological deficits in cannabis users, although they also failed to find any abnormalities in either HIV or cannabis user control groups. They did document brain metabolite abnormalities in cannabis users, but found that frontal white matter glutamate actually normalized in cannabis-using HIV-positive subjects. Subjects with HIV in this study were mostly asymptomatic, and therefore did not display the well documented deficits seen in later stages of HIV.

Ecstasy/MDMA and other ‘Club’ drugs

Methamphetamine, discussed above, is often used in the club or party setting and, as noted, presents multiple risks for users, HIV-positive or not. Methylenedioxy-methamphetamine (MDMA), another popular party drug, affects the serotonin system and can interfere with encoding long-term memory, verbal learning, attention, and concentration [99]. A systematic review of club drug use and HIV infection found that MDMA has an adverse effect on cellular immune responses, can increase cytokine levels and increase viral replication; however, these studies are difficult to perform, often due to multiple substances used and other confounders. Other drugs like gamma-aminobutyric acid can affect sleep and memory, and some evidence points to an increase of dopamine levels in the brain when using [100]. The greatest ongoing risk from club drugs is HIV acquisition or transmission.

Tobacco

Though not well researched in this context, nicotine has been shown to adversely affect the immune system [101, 102]. Nicotine has been studied in patients with mental illness, and reviewers have found an enhancement of cognition in schizophrenics [103].

Conclusion

Preventive measures that can ameliorate damage caused by HIV and drugs of abuse are necessary to sustain quality of life and prevent significant morbidity and early mortality in these patients. Prevention strategies must include regular neurocognitive screening during health encounters [38••] and targeted and appropriate techniques accounting for differences in gender, age, disease specificity, psychiatric illness, and baseline cognitive status. Harm-reduction methods, lifestyle management, support of cognitive reserve, cognitive remediation therapy, use of ART that provides optimal CNS penetration in hopes of reducing viral load in the CNS, and a research focus that could investigate new compounds to reduce neuro-inflammation are all necessary to prevent a looming global crisis of neurocognitively impaired individuals with HIV [55, 104, 105••, 106•, 107••, 108••, 109••].