Ten Myths About HIV Infection and Aids

Since the first reports of the acquired immuno-deficiency syndrome (AIDS) emerged in 1981, the numbers of reported cases of HIV infection and AIDS have risen. The cumulative total of AIDS cases AIDS cases in the United States through 2003 was 929,985. Of these, 920,566 have been in adults or adolescents and 9,419 in children under age 13. Among the adult and adolescent cases, 749,887 have been in males and 170,679 in females. Over half of these individuals have died, with the number of deaths totaling 524,060 (1).

Transmission of HIV can occur by contact with the virus through unprotected sexual activity of any type. HIV, a retrovirus, can enter the body through the lining of the vagina, vulva, penis, rectum, or mouth during sex with an infected partner. HIV is also spread through use of injection drugs, introduction of infected blood, blood products, or other bodily fluids into the bloodstream, by intrapartum or perinatal transmission, and by passive transmission through breast-feeding (1). Scientific evidence excludes other methods of transmission, including casual transmission by hugging or kissing.

Two strains of HIV have been described. HIV type 1 is the most virulent strain (2). The most common subtype in North America is HIV-1 subtype B. HIV type 2 is a less virulent strain that is found in certain subgroups, mostly contained in western Africa (3). Although HIV-2 is not considered clinically relevant in the United States, individuals who have signs or symptoms suggestive of HIV infection who come from West Africa or who report sexual contact with someone from West Africa should be evaluated for the presence for this strain of the virus. If appropriate, the “Combi-test” should be used—the same antibody test used by U.S. blood banks—which reliably detects antibodies to both HIV-1 and HIV-2 (4). Laboratory tests can measure HIV antibodies in the blood, usually 3 to 6 weeks after exposure to the virus.

Antibody testing is the primary method of making the diagnosis of HIV/AIDS, and patients undergoing such testing should receive standardized pre- and posttest counseling. The two most common antibody detection tests are the enzyme-linked immunosorbent assay (ELISA) and the confirmatory Western blot (WB) test. An oral fluids HIV test is now available (5); one type is used to test for HIV-1, and another for both HIV-1 and HIV-2. Although the oral test does not have approval for home use, it has been approved for use in nonhospital settings, which should greatly facilitate HIV testing in rural areas and through outreach and community programs. A major advantage of the test is that results are available in 20 minutes. Availability of the oral test is expected to have a positive impact on early detection and prevention of HIV infection in the United States (6). As with the serological antibody test, the rapid oral HIV-antibody test should be considered preliminary, and a positive result must be confirmed with more specific tests, such as the WB. Bear in mind that a negative HIV antibody test of any type does not guarantee that the person is not currently infected.

HIV directly infects T cell lymphocytes, with the CD4 (T4) surface marker serving as the primary target of infection. Because the number of CD4 cells correlates with the progression of HIV infection, the CD4 cell count is an important tool in staging the immunological status of patients with HIV infection. CD4 counts from 500 to 1600 cells/mm³ are considered normal. CD4 counts below 200/mm³ are considered diagnostic of AIDS. Some other conditions are also considered diagnostic of AIDS, including HIV-related CNS involvement (see Table 1).

The current method used to predict stage of disease, to monitor disease progression, and to formulate treatment strategies is to determine viral load, which is a measurement of the actual number of viral particles found in a cubic millimeter of blood. Viral load is the preferred measure for monitoring disease progression, treatment responsiveness, and treatment resistance. The relationship between peripheral viral load and the development of HIV-related neurological disease is not yet known. Whether cerebrospinal fluid (CSF) viral load is an important predictor of CNS involvement, progression, or effectiveness of therapy has yet to be determined (7, 8).

Treatment guidelines for HIV/AIDS evolve constantly and are updated frequently by the U.S. Department of Health and Human Services, which issues treatment guidelines for adult and adolescent, pediatric, and perinatal populations (see http://www.AIDSinfo.nih.gov). Other online services are available through local jurisdictions that provide additional resources to supplement the basic guidelines. Among the better sites are those of the New York State Department of Health (http://www.hivguidelines.org/public_html/center/clinical-guidelines/clinical-guidelines.shtml) and any of the various AIDS Education Training Centers (AETC), such as the Florida/Caribbean AETC at the University of South Florida (http://www.faetc.org/Guide/). These sites include important links to services such as the postexposure prophylaxis hot line (http://www.ucsf.edu/hivcntr/Hotlines/PEPline.html), which offers, among other things, around-the-clock advice on the management of occupational exposures. The New York State Department of Health (http://www.hivguidelines.org/public_html/center/clinicalguidelines/mental_health_guidelines/mental_health_intro.htm) and the APA AIDS Program Office (http://www.psych.org/psych_pract/treatg/pg/hivaids_revisebook_index.cfm) provide treatment guidelines for the mental health and psychiatric complications of HIV/AIDS. All of these resources are invaluable for the information they provide on start-up therapies, maintenance recommendations, management of side effects, and complications from antiretroviral therapies.

Why is any of this important for psychiatrists in practice today? First, in all categories of mental disorders, there may be behavioral issues—related, for example, to appetitive drives, substance use, impulsivity, judgment, reasoned action, and sexuality—that constitute risk factors for acquiring and transmitting HIV. Indeed, a growing number of studies show high rates of HIV risk behaviors in patients with psychiatric illness. Second, as experts in the field of behavior, we have an essential role in providing effective assessment and medical treatment to patients in all medically ill populations. Third, with the development of increasingly successful antiretroviral therapies, persons with HIV infection and AIDS are living longer. With increasing life expectancy, the rate of psychiatric complications is increasing, and clinicians must be knowledgeable about such complications and their treatment. Finally, given the vast array of neuropsychiatric complications associated with primary CNS infection, secondary CNS disease, and the psychiatric side effects of antiretroviral therapies and related therapies for medical conditions associated with HIV infection, psychiatrists need to be prepared for collaborative work with other health care providers to improve therapeutic outcomes for patients.

The most common neuropsychiatric disorders among individuals with HIV infection and AIDS are HIV-associated minor cognitive/motor disorder (HMCMD), HIV-associated dementia (HAD), and delirium (9). The AIDS Task Force of the American Academy of Neurology has established diagnostic criteria (10) to recognize early subtle cognitive changes associated with HIV infection and to differentiate them from dementia (Tables 2 and 3).

Cognitive/motor disorders in patients with HIV infection are common and usually progressive (9). Some evidence suggests that these disorders lie along on a continuum of CNS involvement rather than constituting different clinical disorders. Although it is true that with the advent of highly active antiretroviral therapy (HAART), the incidence of HAD has dropped to the neighborhood of 5%–10%, the incidence of HAD as a percentage of fully developed AIDS has risen from 4.4% to 6.5% (11). Rates of HMCMD, by contrast, have remained fairly constant at about 20%–25% (12). Thus, psychiatrists need to maintain a high index of suspicion in the evaluation of patients with HIV infection who present with new-onset cognitive or behavioral symptoms at any time in the course of their disease.

In patients with HMCMD, cognitive impairment will manifest as changes in attention and in memory registration, storage, and retrieval as well as in psychomotor speed, information processing rate, and fine motor function. At first, significant changes may be detected only by neuropsychological testing. As the disease progresses, these impairments become more pronounced and may be accompanied by changes in mood and personality. Such changes have minimal effects on activities of daily living and on functional performance status. Changes consistent with more fully developed HAD are commonly associated with a more severe decline in functional status.

Forgetfulness, inattention, difficulty with concentration, mental slowing, loss of interest, depression, and anhedonia should always be evaluated at any stage of HIV disease for possible cognitive disorders masquerading as mood disorders. Even the most subtle cognitive or motor symptoms in previously asymptomatic persons warrant a full neurodiagnostic investigation. At a minimum, a complete neurodiagnostic evaluation includes a metabolic screen, a toxicology screen, lumbar puncture, an electroencephalogram (EEG), and either computed tomography (CT) or magnetic resonance imaging (MRI) of the brain.

Imaging by both CT and MRI is helpful in demonstrating HIV-related brain injury as well as other pathological processes. MRI has revealed volumetric reductions in cerebral gray and white matter in neurologically asymptomatic HIV-positive patients (13–14). Positron emission tomography (PET) (15) has shown an increase in basal ganglia and right parietal lobe metabolism early in nondemented patients with HIV-1 infection, followed by general hypometabolism as CNS involvement of HIV-1 progresses; these findings are clinically associated with dementia. Some studies have noted reversal of imaging abnormalities along with functional neurological improvement of varying degrees after initiation of antiretroviral treatment (16). Magnetic resonance spectroscopy (MRS) has shown sensitivity in detecting early changes in patients with HMCMD, and it differentiates these patients from those with more advanced HAD (17–18). MRS is the most useful imaging study in evaluating brain injury in early HMCMD (19–21), and it may also have a role in monitoring response to treatment of patients with HAD (22–23).

Thus, even with the advent of HAART, it remains vital to detect cognitive changes early rather than late. HIV-related cognitive disorders are treatable when properly diagnosed (see Myth 3). It is a significant issue for our patients’ quality of life and their ability to function in everyday life. Cognitive impairment can have a negative impact on employment, medical compliance, sense of well-being, productivity, and the ability to express oneself meaningfully.

Before the advent of HAART, zidovudine was reported to ameliorate cognitive impairment and delay progression to HAD. The optimal dosing of zidovudine to obtain this effect was not precisely determined, although Sidtis and colleagues (24) reported statistically significant gains in neuropsychological testing with doses of 2000 mg/day. Other agents, such as didanosine (ddI) and other nucleosides, zalcitabine (ddC), stavudine (d4T) and lamivudine (3TC), and the nonnucleoside reverse transcriptase inhibitors, are being investigated for use in monotherapy or in multiagent therapy, but they have not been demonstrated to be especially effective in the treatment of neurocognitive dysfunction associated with HIV infection (9). Encouraging results have been reported in various studies that support the use of HAART in patients with HAD and HMCMD (9, 25). In cross-sectional as well as longitudinal studies, patients on HAART regimens have shown significant improvements in performance on neuropsychological testing. Although some specialists advocate the use of agents that have good CSF penetrance and can reduce CSF viral load, this characteristic does not appear to be required for achieving positive results.

Beyond the standard antiretroviral therapies that target systemic infection, adjunctive pharmacotherapy with agents that specifically target CNS infection may be used. Attempts to prevent the neurotoxic chain of events secondary to HIV involvement of the CNS have focused on blocking the toxicity generated from the binding of HIV to CD4 receptors in the CNS (26). Reducing stimulation of N-methyl-d-aspartate (NMDA) receptors by removing or antagonizing excitatory amino acids with NMDA receptor antagonists such as memantine or by preventing calcium influx with calcium channel blockers such as nimodipine, dantrolene, or BAPTA (bis-o-aminophenoxyethane-N,N,N’,N’-tetra-acetic acid) has been proposed, but as yet there is no clinical indication for any of these drugs (9).

Palliative use of psychostimulants, such as methylphenidate (27) and dextroamphetamine (28), has been shown to improve symptoms of fatigue and cognitive dysfunction in HIV-related cognitive impairment (29). The general clinical experience has been that methylphenidate is better tolerated than dextroamphetamine. Clinically useful doses range from 10 to 30 mg/day in divided doses, with adjustments made as needed. The usual dosage schedule would be 5–10 mg of methylphenidate taken at 7 a.m., 10 a.m., and 1 p.m. The side effects of stimulants are mild and may include anxiety, insomnia, tachycardia, and, rarely, anorexia. The use of stimulants must be carefully assessed for patients with a history of substance use disorders, and the benefit of its use must clearly outweigh the risk.

The use of modafinil in patients with severe fatigue (30) or as augmentation to standard antidepressant treatment is growing (31). It is unclear whether it will have the same cognitive-enhancing benefit as dextroamphetamine and methylphenidate (32). Although the abuse potential of modafinil in patients with HIV infection is unknown, it is worth considering as an alternative to conventional stimulants because of its low abuse potential and its schedule IV status. Low doses (e.g., 100 mg) have been found to be effective, with limited side effects.

Other agents, such as amantadine (33) and atomoxetine (34), have been used, but no data are available on their cognitive-enhancing properties specifically in HIV-related cognitive impairment. Controlled trials with selegiline, a selective monoamine oxidase (MAO) B inhibitor, are under way, and the results are promising. In one randomized, double-blind, placebo-controlled trial of selegiline in HIV cognitive impairment, patients receiving the drug showed statistically significant improvement on neuropsychological testing (35). No adverse drug-drug interactions were noted between selegiline and antiretroviral therapies.

With the use of today’s more intensive and aggressive treatments for HIV infection and its complications, delirium can present as psychosis in ambulatory patients with HIV/AIDS. In our setting, we have seen hypoactive delirious processes in ambulatory patients with oral infections masquerade as “depression.” Hence it is important not to disregard the possibility—and the importance—of delirium, even in ambulatory care settings.

Because the etiology of delirium in patients with HIV infection is multifactorial, a thorough diagnostic workup for psychotic symptoms is required when delirium is detected. Important possibilities to rule out include hypoxemia, hypoglycemia, opportunistic infections (such as toxoplasmosis and cryptococcosis), viral infections, cerebrovascular accident, neurotoxicity from substances of abuse or from antiretroviral therapies, and metabolic and electrolyte disturbances.

Delirium can be effectively treated pharmacologically with high-potency neuroleptics (given orally or intramuscularly) without serious adverse effects. In a controlled study, Breitbart and associates (36) found that low doses of haloperidol (2.8±2.4 mg/day) and chlorpromazine (50±23 mg/day) may be beneficial in treating delirious patients without causing significant extrapyramidal side effects; lorazepam given as monotherapy was neither effective nor well tolerated, and in some patients it exacerbated the delirium. Atypical antipsychotic agents, with their low risk of adverse effects, can be used safely to control delirium (9). In our setting, risperidone, olanzapine, quetiapine, aripiprazole, and ziprasidone are noted to be more effective in treating hypoactive delirium with psychotic symptomatology than the high- or low-potency neuroleptics. Now that ziprasidone and olanzapine are available for parenteral use, they will likely be used in favor of haloperidol.

In a critical care or inpatient setting, intravenous administration of haloperidol remains the preferred treatment for delirium. Although this approach is clinically effective (37), it remains investigational because haloperidol has not been formally approved by the Food and Drug Administration for intravenous use. When symptoms of delirium or agitation do not respond adequately to oral or intramuscular haloperidol or other neuroleptics, intravenous haloperidol may be used. Repeated boluses ranging from 0.5 mg to 100 mg may be administered every 30 to 60 minutes until the patient is either calm or sedated. Administering the same total dose to achieve sedation may be repeated for another 24 hours, and then the dose should be reduced by 20%–30% per day until treatment can safely be discontinued. Treatment-related side effects are negligible but should be monitored. If agitation is refractory to a schedule of discrete doses of intravenous haloperidol, a continuous infusion may be used (38).

Almost 10% of patients with HIV infection have a diagnosed psychosis of one type or another (9). Some psychoses are preexisting conditions unrelated to the HIV infection, whereas others are thought to result from CNS involvement of the HIV infection (39, 40). Many patients with HIV-associated psychosis also manifest symptoms of cognitive decline, such as HMCMD or HAD. There is a high association between psychosis and HIV-related mood disorders. The initial evaluation for psychosis includes the same medical workup as that for delirium and dementia.

Psychosis in the context of CNS involvement of HIV requires timely intervention, which is often accomplished with neuroleptics. However, patients with HIV infection are more sensitive to side effects such as extrapyramidal reactions and neuroleptic malignant syndrome with high-potency neuroleptics and confusion and seizures with low-potency neuroleptics. The atypical antipsychotics—risperidone, olanzapine, quetiapine, aripiprazole, and ziprasidone—have been shown to control HIV-related psychosis without causing sedation or notable cognitive impairment (9). Quetiapine has been shown to have the least antiparkinsonian effects (41) and, as with patients with Parkinson’s disease, should be considered for patients with symptomatic HIV-related CNS disease. Risperidone is an effective alternative and has been reported to be well tolerated in low doses (1–3 mg/day) (42). Low doses should be in used in any regimen involving protease inhibitors because extrapyramidal reactions (43) and alterations in consciousness (44) have been reported.

Mania and hypomania are known to accompany HIV infection and may be due to bipolar illness, use of illicit drugs or prescribed medications (including antiretroviral agents), neurological conditions, systemic infections, endocrinopathies, or metabolic disturbances. HIV infection itself has been associated with psychotic symptoms with manic features (9, 45, 46). In some cases, especially in patients with advanced disease, neuropsychiatric impairment is noted on resolution of the manic symptoms, with progress to dementia within months (47, 48). A comprehensive medical evaluation should accompany the assessment of mania in patients with HIV infection to identify potential organic components of the illness and treat them accordingly.

Patients who have used lithium carbonate for their manic episodes before HIV was diagnosed can continue their regimen with careful monitoring (9). Patients taking lithium carbonate who develop diarrhea or significant fluid loss from any source may be prone to neurotoxicity or nephrotoxicity. The anticonvulsant agent valproic acid and the divalproex sodium preparation control mania better than lithium and neuroleptics in patients with MRI abnormalities of the brain (49). Although increased viral replication has been reported in vitro with valproic acid or valproate (50), no clinical correlates have reported. Valproic acid has been reported to increase zidovudine levels by decreasing its glucuronidation (51).

Anecdotal reports support the use of carbamazepine and oxcarbamazepine in the treatment of mania in patients with HIV-related mania; however, toxicity has been noted when these agents are combined with antiretroviral therapy (9). A report of carbamazepine-induced failure of antiviral therapy is cause for concern (52) and raises the question of significant drug-drug interactions with antiretroviral therapies. Oxcarbamazepine is theoretically safe to use, but no clinical reports on it have been issued yet.

There are no specific reports on the use of lamotrigine in HIV-related mania. However, lamotrigine has been used effectively in the treatment of HIV-related neuropathy without added burden or side effects (53).

One important consideration in the use of anticonvulsants for patients with HIV infection is the potential interaction with the protease inhibitors. Because these medications are metabolized through the cytochrome P450 system, plasma levels may become elevated as a result of competition for the enzymatic system. Since no definitive guidelines have been formulated, caution should be exercised when adding an antiepileptic agent to a regimen of medications that includes a protease inhibitor. Clinical and, when possible, laboratory monitoring is recommended.

The use of risperidone, olanzapine, quetiapine, aripiprazole, and ziprasidone in the treatment of acute mania has not yet been studied in this population. Clozapine can be used, but with caution because of the potential for agranulocytosis. One report indicated possible benefits from clozapine in low doses (below 30 mg/day) in patients with HIV infection who had documented drug-induced movement disorders with conventional neuroleptics (54).

Nearly 85% of individuals with HIV infection exhibit some symptoms of depression during the course of their illness (55). As many as 50% of patients experience a major depressive disorder. Persons with HIV infection are twice as likely as those who are HIV seronegative to experience depression (56). The diagnosis of depression in patients with HIV disease may be complicated by factors such as appropriate sadness related to having a life-threatening illness, acute grief and multiple bereavement, or other psychological reactions. Depression due to a general medical condition, substance abuse, HIV-related medications, sepsis secondary to opportunistic infection, systemic malignancies, and CNS complications must also be considered.

If depression is not accurately diagnosed and treated, there is a risk of suicidality becoming manifest. Suicidal ideation in the context of HIV disease should not be considered a normal or understandable reaction to having a stigmatizing and potentially fatal disease. Clinically, in the context of HIV infection, there is a high risk of suicide (9, 57). Risk factors for suicide in patients with HIV infection include current or past episodes of major depression, previous suicide attempts, substance abuse, social isolation, perceived lack of social support, adjustment disorder with depressed mood, personality disorder, HIV-related interpersonal or occupational problems, and a complicated or multiple bereavement (9, 58).

Although the tricyclic antidepressants, such as amitriptyline, nortriptyline, imipramine, desipramine, and doxepin, are effective in HIV-related depression (59), they have troublesome anticholinergic effects that limit their use. If a tricyclic is being considered, it is well worth selecting one that has a low level of anticholinergic effects, such as desipramine or nortriptyline. Tricyclics may be of benefit in the treatment of neuropathic pain whether or not depression is present.

Trazodone, which has minimal anticholinergic effects, seems to be well tolerated and can be initiated at a dose of 25–50 mg at bedtime, with the dose increased by 25–50 mg every 3–5 days until a clinical response is evident (9, 59). Trazodone’s sedating effect can be used with benefit for patients with severe insomnia. Orthostatic hypotension as a side effect of trazodone has also been reported and can be of significance in patients with HIV-related dysautonomia.

Selective serotonin reuptake inhibitors (SSRIs), such as fluoxetine, paroxetine, sertraline, citalopram, and escitalopram, which also have minimal anticholinergic effects, have been reported to be useful in treating HIV-related depression (9). Fluvoxamine was shown in one trial to be poorly tolerated among depressed patients with HIV infection (60).

Bupropion’s stimulatory effect can be particularly useful in the withdrawn or apathetic patient (9, 59). Unlike stimulants, no cognitive enhancement has been associated with its use. However, it should be used with caution in patients with CNS lesions, because seizures have been induced with doses greater than 200 mg/day. In patients receiving protease inhibitors, specifically ritonavir, the lowest possible dose of bupropion should be used (61).

Venlafaxine and mirtazapine have been reported to be effective and well tolerated (9, 59), and they have a low potential for drug interactions. One report cautions that venlafaxine may decrease indinavir levels (62). Because mirtazapine may cause weight gain and has antiemetic properties, it is often very useful for patients with HIV disease (59) and may have specific therapeutic benefits for patients with wasting syndrome. As yet, there are no reports on the use of duloxetine in HIV-related depression.

Psychostimulants, such as methylphenidate, have been found to be especially effective for patients who have coexisting depression and cognitive impairments and those for whom other antidepressants have proved to be ineffective as monotherapy or intolerable because of severe or untoward side effects (63). A rapid beneficial effect may be observed with psychostimulants, often within hours after the first dose, producing psychomotor activation, appetite stimulation, and overall improvement of cognition. Similarly, the use of modafinil to augment antidepressant treatment, especially in patients with residual fatigue, is effective and free of serious side effects (30, 31).

MAO inhibitors are generally avoided in treating depression in patients with HIV infection (9, 59). Patients who are already being treated effectively with an MAO inhibitor should continue their regimen but should be cautious especially with zidovudine, an inhibitor of catechol O-methyltransferase. Concomittant use may contribute to a hypertensive crisis. Selegiline, a selective MAO-B inhibitor, has been reported to have a significant antidepressant effect. A transdermal delivery system for selegiline is under development that would bypass both intestinal and liver inhibition and thus make it a safe alternative to oral preparations (64). It may be used for both depression and cognitive impairment.

Anxiety disorders affect some 17% to 36% of patients with HIV infection (9, 59, 65). Persistent and chronic anxiety following notification of seropositive status affects about 20% of patients. This anxiety may evolve into a fully developed posttraumatic stress disorder. Denial or avoidant behaviors secondary to untreated HIV-related anxiety may diminish treatment compliance and interfere with medical management of HIV disease.

Among HIV patients with diagnosed anxiety disorders, adjustment disorder with anxious mood seems to be the most prevalent. Anxiety is commonly associated with depression, and a careful assessment to rule out depressive illness is critical. Anxiety may also arise from medical conditions, especially in patients with pain, respiratory compromise due to pneumonia, neurological compromise due to opportunistic infections or malignancies, and substance-induced anxiety. As with other psychiatric disorders, a comprehensive medical evaluation is essential to rule out medical problems that could be causally related to anxiety symptoms. Antiretroviral medications have been associated with anxiety, irritability, restlessness, agitation, and insomnia. Although efavirenz (66) is the most common cause of antiretroviral-induced anxiety disorder, all medications are suspect, and, whenever possible, consideration should be given to discontinuing all possible offending agents.

Psychotherapeutic interventions, including self-hypnosis and biofeedback, are very effective in the treatment of anxiety disorders (59, 67). While adjunctive pharmacotherapy with a benzodiazepine may be necessary for acute reduction of symptoms, all psychotherapies are useful for the development of adaptive behaviors that allow the individual to reduce anxiety while improving coping style.

In the long-term management of anxiety, especially in individuals who have a concomitant substance use disorder, antidepressants are most effective (59, 67). All SSRIs and venlafaxine have been reported to be effective. In some cases, intermediate-acting benzodiazepines without active metabolites (oxazepam and lorazepam) are considered a good choice for the patient with severe anxiety because of their ease of use and low neurotoxicity (9, 67). Long-acting benzodiazepines are not recommended for patients with cognitive compromise, including disinhibition, frontal lobe dysfunction, delirium, and confusion.

The combination of a benzodiazepine with an antidepressant or with buspirone can be helpful. Buspirone (59, 67), with its low level of side effects, can be increased by 5–10 mg every 3 days until a therapeutic dose (45–90 mg/day) is reached over a 2-week course of concomitant benzodiazepine therapy. It is advisable then to taper the benzodiazepine.

While the use of antihistamines is sometimes recommended in managing anxiety, our clinical experience suggests otherwise. Anxious patients, especially those with cognitive compromise, do not do well with antihistamines. Using trazodone in low doses (25–50 mg t.i.d.) is preferable (9).

Sleep-related EEG findings indicate early CNS involvement by HIV. Sleep is often disturbed in patients with HIV disease (68). Polysomnography has been used to investigate patients’ complaints of dyssomnia, uncovering gross disturbances of sleep architecture. Although the viral infection may play a central role in such disorders, medication effects, such as those of zidovudine and illicit drugs, must be included in the analysis of sleep disturbances related to HIV infection.

Treatment generally is multimodal, requiring pharmacotherapy along with behavioral interventions. HIV-related insomnia is difficult to treat (59). Use of benzodiazepines can easily lead to tolerance and dose escalation. Antidepressants with significant antihistamine effects are often helpful, but because of anticholinergic side effects, optimal dosing may not be possible. Trazodone is useful in treating insomnia, but orthostatic changes can make optimal dosing difficult. Of all the antidepressants, mirtazapine is the best tolerated. Although low-dose atypical antipsychotics that have sedating effects are sometimes recommended, such as quetiapine and olanzapine, the potential for side effects, including exacerbating metabolic syndrome in patients receiving HAART, places the patient at excessive risk. Zolpidem and zaleplon (69) are effective, but once initiated they are almost always required indefinitely. Anticonvulsants such as valproate, gabapentin, and tiagabine are an alternative to hypnotic agents, antidepressants, and atypical antipsychotics (70). These agents are well tolerated and can be used for chronic insomnia without the development of tolerance or abuse.

Clinically relevant interactions between HAART and psychotropic agents have been noted above, although this is not the place for a full review. The science of pharmacokinetic and dynamic interactions and the pharmacotherapy of neuropsychiatric syndromes in patients with HIV infection are reviewed elsewhere (59, 71). The optimal treatment recommendation in prescribing medications for patients with HIV infection who are taking antiretroviral medications is to avoid kinetic and metabolic drug-drug interactions by selecting psychotropic agents that minimally alter the activity of the enzyme systems involved in hepatic metabolism of HIV therapies. Since few psychotropic agents meet this criterion, one should select a psychotropic agent on the basis of whether it is metabolized by multiple enzymes. Having alternative pathways reduces the likelihood that inhibition of one enzyme will have clinically significant effects.

Large variations in enzyme activities are known to exist among individuals, including among different ethnic groups. The ability to genotype patients, which is becoming widespread, allows identification of patients with known metabolic defects specifically in the cytochrome P450 system. Clinicians can then tailor treatment such that anomalous responses are identified and potentially adverse drug interactions avoided.

In HIV disease, the potential for drug-drug interactions is a concern. Thus, clinicians should have a working knowledge of potential interactions between psychotropic agents and antiretroviral agents as well as other medical pharmacotherapy used in treating HIV/AIDS. Although patients with HIV infection are very knowledgeable about the side effects associated with HIV regimens, they are less knowledgeable about critical interactions that may cause fluctuations in levels of antiretroviral agents; such fluctuations can place them at risk by reducing pharmacotherapeutic control of their disease process. Patients should be made aware that interactions can even occur with recreational drugs, over-the-counter medications, and nutraceuticals.

A collaborative partnership between the clinician, the patient, the medical team, and a knowledgeable pharmacist is necessary to avoid drug interactions that will adversely affect the patient’s disease status. Regular updates for the treatment team are necessary. Exchange of information between all the professionals on the team, without placing the patient in the middle of turf battles, is necessary. This may be accomplished through the form of a journal club for the entire treatment team where drug interactions are reviewed.

One tool our clinical group uses to keep up-to-date on information about drug interactions is the publication HIV CareLink Newsletter (www.faetc.org/Newsletter/index.asp), published by the Florida/Caribbean AIDS Education Training Center. Another resource is the APA AIDS Program Office’s AIDS Network Update (www.psych.org/aids/infoclinc.cfm); their database is updated quarterly specifically on psychotropic drug interactions with HIV therapies. Other helpful Web sites include www.hivpharmacology.com, www.drug-interactions.com, and www.ucsf.edu/ hivcntr/Clinical_Resources/Pharmacy/Drug_Intxn_Web.htm.

Neuropsychiatric complications and disorders may arise at any time during the course of HIV infection. Even the most subtle neurobehavioral complaints from previously asymptomatic patients warrant a full neurodiagnostic and comprehensive medical evaluation. Specific pharmacotherapies can be used, sometimes innovatively, to treat the neuropsychiatric complications associated with HIV infection. Research continues to improve therapeutic options and outcomes for patients with HIV-related neuropsychiatric disorders.

Acknowledgments The author thanks Carol Svoboda, Diane Pennessi, and Candice Peggs from the APA AIDS Program Office for their assistance. The author also thanks Ron Cardoso, who lived courageously with HIV and yet did not die from AIDS. His memory served as an inspiration for this manuscript.