Dr. R. Sansone is a professor in the Departments of Psychiatry and Internal Medicine at Wright State University School of Medicine in Dayton, Ohio, and Director of Psychiatry Education at Kettering Medical Center in Kettering, Ohio; Dr. L. Sansone is a family medicine physician (government service) and Medical Director of the Primary Care Clinic at Wright-Patterson Air Force Base. The views and opinions expressed in this column are those of the authors and do not reflect the official policy or the position of the United States Air Force, Department of Defense, or US government.

Psychiatry (Edgemont) 2009;6(9):13–16

Dr. R. Sansone is a professor in the Departments of Psychiatry and Internal Medicine at Wright State University School of Medicine in Dayton, Ohio, and Director of Psychiatry Education at Kettering Medical Center in Kettering, Ohio; Dr. L. Sansone is a family medicine physician (government service) and Medical Director of the Primary Care Clinic at Wright-Patterson Air Force Base. The views and opinions expressed in this column are those of the authors and do not reflect the official policy or the position of the United States Air Force, Department of Defense, or US government.

Psychiatry (Edgemont) 2008;5(12):16–19

ABSTRACT

Pain, including neuropathic pain, is a relatively common complaint in various clinical settings. Several antidepressants have been efficacious in the management of chronic neuropathic pain, including the tricyclic antidepressants (particularly the tertiary-amine subtypes, such as amitriptyline, doxepin, and imipramine) as well as venlafaxine, bupropion, and duloxetine. The selective serotonin reuptake inhibitors have either been less robust (paroxetine, citalopram) or lacked any efficacy at all (fluoxetine). In this article, we review these various medications and offer an interpretive comparison, as there are few head-to-head comparison studies.

INTRODUCTION

Pain complaints, which emerge in both psychiatric and primary care settings, are a costly medical phenomenon. According to Jackson and St. Onge, the management of pain disorders in the US exceeds $100 billion annually, which includes medical care, workers compensation, and lost work productivity.[1] This estimate is echoed by the American Pain Society.[2] While lower back pain appears to be the most prevalent pain disorder,[2] pain encompasses a number of different syndromes, including neuropathic pain. In this overview article, we have elected to focus specifically on neuropathic pain syndromes and will summarize the recent review articles in this area with regard to the use of antidepressants.

The Working Definition of Neuropathic Pain

Neuropathic pain is a subcategory of the larger group of pain syndromes and is characterized by (a) chronic pain, (b) various heterogeneous diagnoses (e.g., diabetic peripheral neuropathy, post-herpetic neuralgia, central post-stroke pain, phantom limb pain), and (c) differing etiologies.[3] In addition, within a specific diagnosis, there is notable variability in the presentation as well as the characteristics of the experienced pain.

Epidemiological Aspects of Neuropathic Pain

According to Wolfe and Trivedi, about 1.5 percent of the US population is affected by neuropathic pain.[4] This prevalence rate is reflected in healthcare costs. For example, in a 2007 study, Barrett et al[5] found that the average annual cost of pain medication per diabetic patient with peripheral neuropathy was just over $1,000.

How Antidepressants Work in Pain Management

The explicit way in which antidepressants are effective in pain management remains unknown, but multiple mechanisms are likely to be involved.[6] Perhaps the most popular theory is that antidepressants exert their effects on serotonin and norepinephrine, particularly along the descending spinal pain pathways. Antidepressants may also exert adjunctive therapeutic influences through histamine receptors as well as modulation of sodium channels.[6]

Neuropathic pain and depression

One common comorbid psychiatric diagnosis encountered in patients with neuropathic pain is depression, which affects the majority of individuals (i.e., 57%).[1] Studies indicate that patients with pain have a substantially increased risk for depression, anywhere from 2 to 5 times that of the general population.[7] Given that these syndromes are commonly comorbid, the assessment of depression in the presence of pain is typically complicated by the presence of shared features between the two syndromes (e.g., fatigue, sleep disturbance). In addition, the format for assessment may affect findings. Specifically, prevalence rates of depression among pain patients are seemingly higher when assessed through self-report measures compared with standardized diagnostic approaches.[2]

Investigators have also found that the prevalence rate of depression among pain patients may vary as a function of the clinical setting. In a review of the literature, Bair et al[8] found that the prevalence of depression among pain patients in psychiatry clinics was 35 percent, in pain clinics 38 percent, in rheumatology clinics 52 percent, and in dental clinics 78 percent. Surprisingly, despite the frequent comorbidity of pain and depression, Jackson and St. Onge indicate that both remain under-recognized and under-treated in this clinical population.[1]

While various relationships between pain and depression might be postulated, the most common hypotheses for their comorbidity are that (a) depression precedes pain; (b) pain precedes depression; (c) prior depression heightens the risk of subsequent depression in the presence of new-onset pain; and (d) the two phenomena occur independently of each other. Regardless of the relationships between the two syndromes, according to Sullivan and Robinson, the level of comorbidity between pain and depression is one of the basic rationales for considering antidepressant therapy as a treatment.[9] However, this is not an exclusive relationship as the doses of TCAs used to treat pain are often much lower than those used to treat depression and depression is not always present in those who experience pain relief.

Antidepressants and Neuropathic Pain

Many antidepressants are effective in the treatment of pain, but not all and not to the same degree.

Tricyclic antidepressants. TCAs are the most studied antidepressants for the treatment of neuropathic pain and are a mainstay in the treatment armamentarium.[1] These antidepressants inhibit the reuptake of serotonin and norepinephrine at the synapse, but do so differentially according to chemical structure. The tertiary amines (e.g., amitriptyline, doxepin, imipramine) inhibit serotonin to a greater degree than norepinephrine. In contrast, the secondary amines (e.g., desipramine, nortriptyline) have more pronounced effects on norepinephrine. While some authors report approximately equal efficacy between these two subclasses of TCAs,[4] others advise that the tertiary amines are somewhat more effective than the secondary amines.[1,2,6] Interestingly, pain relief appears to be independent of the antidepressant effects of these drugs and may be achieved at doses lower than those used in the treatment of depression. While TCAs are reliable and effective, their use is potentially complicated by a host of beleaguering side effects (e.g., weight gain, anticholinergic effects, orthostatic hypotension, cardiovascular effects, lethality in overdose).

Selective serotonin reuptake inhibitors. Selective serotonin reuptake inhibitors (SSRIs) exert their efficacy predominantly through the reuptake inhibition of serotonin. Compared with TCAs in the management of neuropathic pain, the data with regard to SSRIs is more inconsistent[3,6] and studies are considerably fewer in number. To date, paroxetine and citalopram have demonstrated modest efficacy in the management of neuropathic pain, whereas fluoxetine has not demonstrated any efficacy at all.[1] The overall impression is that SSRIs are less effective than other antidepressant options in the treatment of neuropathic pain.[2,4,9–11] Indeed, from a conservative perspective, it appears that SSRIs are not superior to the other types of antidepressants used in the treatment of neuropathic pain, such as the TCAs.[12] While SSRI side effects are generally mild, there is the risk of weight gain with the long-term use of some (e.g., paroxetine) as well as sexual dysfunction and discontinuation symptoms with abrupt cessation.

Venlafaxine. Venlafaxine is an unusual, mixed-action antidepressant that predominantly inhibits serotonin reuptake at low doses and norepinephrine reuptake at higher doses. Therefore, unlike SSRIs and like TCAs, venlafaxine affects both of the key neurotransmitters that are hypothesized to be involved in the modulation of neuropathic pain. In support of this theoretical observation, case reports[1,13] and empirical studies[9,14] indicate that venlafaxine is effective for the management of neuropathic pain at doses of 150mg per day or higher (i.e., typical antidepressant doses). Venlafaxine has also demonstrated efficacy with pain management in a study in which it was added to gabapentin.[4] In the treatment of neuropathic pain, venlafaxine is comparable to imipramine,[2,10,13] suggesting that it may be comparable to other TCAs as well. Despite a milder side effect profile than the TCAs, venlafaxine may elevate blood pressure and has a discontinuation syndrome with abrupt cessation.

Bupropion. Bupropion inhibits the reuptake of norepinephrine and dopamine. In a double-blind crossover study of patients with various forms of neuropathic pain, at doses of 300mg per day, bupropion SR was similar in efficacy to TCAs.[4] According to the proposed theory of how antidepressants work in pain syndromes (i.e., dual effects on serotonin and norepinephrine), bupropion clearly “breaks the rules.” This observation indicates that other types of antidepressants also warrant investigation as potential medications in the treatment of pain. Bupropion has two absolute clinical contraindications (i.e., current or past history of seizures or eating disorder) and may be excessively activating for some patients.[15]

Duloxetine. Duloxetine is the only antidepressant approved by the US Food and Drug Administration for the treatment of neuropathic pain. It is purportedly a dual-action drug (i.e., it inhibits both serotonin and norepinephrine). Duloxetine has been confirmed in several studies as an effective agent in the treatment of neuropathic pain.[13] Doses for the treatment of neuropathic pain as well as depression are between 60mg and 120mg per day. Interestingly, in a study of healthy volunteers who were taking doses of 60mg per day of duloxetine, the drug exhibited a notable effect on serotonin reuptake inhibition, but not on norepinephrine reuptake inhibition.[16] If these data can be generalized to clinical populations, they indicate that the neurotransmitter effects of duloxetine may be dose-dependent. If so, then like bupropion, duloxetine may be an unexpected exception to the hypothesized requirement for dual serotonin/norepinephrine reuptake inhibition in the treatment of neuropathic pain. We are not aware of any comparative studies between duloxetine and TCAs. As for side effects, duloxetine may cause nausea, somnolence, dizziness, and fatigue.

Comparisons of antidepressants. Based upon the presented data, we have developed a comparison table of these various antidepressants (Table 1). As a caveat, please note that there have been few head-to-head comparisons between TCAs and other types of antidepressants; hence, we have had to rely on some interpretation of these data. The most efficacious antidepressants for the treatment of neuropathic pain appear to be the tertiary-amine TCAs (amitriptyline, doxepin, imipramine), venlafaxine, bupropion, and duloxetine. These appear to be closely followed in efficacy by the secondary-amine TCAs (desipramine, nortriptyline). Modestly effective antidepressants may include the SSRIs paroxetine and citalopram. Ineffective antidepressants include fluoxetine.

Conclusions

Neuropathic pain is a relatively common clinical entity. Given the current diversity of available antidepressant options, many different types appear to be efficacious in the treatment of pain, with the exception of SSRIs. Hopefully, future studies will clarify the genuine role of neurotransmitters in pain modulation as well as the explicit comparative efficacies of the available antidepressants. In addition to their effects on pain, future studies may also take into consideration the effect of antidepressants on the overall quality of life and functioning. It seems clear that antidepressants have an important role in the treatment of neuropathic pain—an observation that is highly relevant for both psychiatrists and primary care clinicians.

References
1. Jackson KC, St. Onge EL. Antidepressant pharmacotherapy: considerations for the pain clinician. Pain Pract. 2003;3:135–143.
2. Sharp J, Keefe B. Psychiatry in chronic pain: a review and update. Curr Psychiatry Rep. 2005;7:213–219.
3. Colombo B, Annovazzi POL, Comi G. Medications for neuropathic pain: current trends. Neurol Sci. 2006;27:S183–189.
4. Wolfe GI, Trivedi JR. Painful peripheral neuropathy and its nonsurgical treatment. Muscle Nerve. 2004;30:3–19.
5. Barrett AM, Lucero MA, Le T, et al. Epidemiology, public health burden, and treatment of diabetic peripheral neuropathic pain: a review. Pain Med. 2007;8:S50–S62.
6. Gallagher RM. Management of neuropathic pain: translating mechanistic advances and evidence-based research into clinical practice. Clin J Pain. 2006;22:S2–S8.
7. Gureje O, Von Korff M, Simon GE, et al. Persistent pain and well-being: a World Health Organization study in primary care. JAMA. 1998;280:147–151.
8. Bair MJ, Robinson RL, Katon W, et al. Depression and pain comorbidity: a literature review. Arch Int Med. 2003;163:2433–2445.
9. Sullivan MD, Robinson JP. Antidepressant and anticonvulsant medication for chronic pain. Phys Med Rehabil Clin N Am. 2006;17:381–400.
10. Irving GA. Contemporary assessment and management of neuropathic pain. Neurol. 2005;64:S21-27.
11. Vu T-N H. Current pharmacologic approaches to treating neuropathic pain. Curr Pain Headache Rep. 2004;8:15–18.
12. Reisner L. Antidepressants for chronic neuropathic pain. Curr Pain Headache Rep. 2003;7:24–33.
13. Jann MW, Slade JH. Antidepressant agents for the treatment of chronic pain and depression. Pharmacotherapy. 2007;27:1571–1587.
14. Chong MS, Brandner B. Neuropathic agents and pain: new strategies. Biomed Pharmacother. 2006;60:318–322.
15. Clinical Pharmacology [database online]. Tampa, FL: Gold Standard, Inc.; 2008. http://cp.gsm.com. Revised July 2008. Access date November 20, 2008.
16. Turcotte JE, Debonnel G, de Montigny C, et al. Assessment of the serotonin and norepinephrine reuptake blocking properties of duloxetine in healthy subjects. Neuropsychopharmacol. 2001;24:511–521.

Psychiatry (Edgemont) 2008;5(7):24-25

VIOLENCE AND MENTAL ILLNESS

Dear Editor:

The article on violence and mental illness by Rueve and Welton[1] in the May issue of Psychiatry 2008 was quite interesting and informative. One cannot overemphasize the role played by environmental factors, especially staff-patient interactions, as a trigger for violent incidents. There have been several published studies that have highlighted these factors.

A study by Lowe, et al.,[2] suggests that issues associated with limit setting and autonomy are perceived as most important by nurses, and these issues are most likely to lead to disagreements in judgment between nurses of different status. Nurses of higher grades appear to show a greater preference for respectful and autonomy-confirming interventions than their more junior colleagues.

The other factor related to staff-nurse interactions is also highlighted in a study by Jansen, et al.,[3] in which nurses’ attitudes toward inpatient aggression indicate that personal characteristics (gender and work experience), occupational characteristics (contractual status and type of ward), and national sociocultural values and beliefs all affect the formation of attitudes of nurses.

Another earlier study by Whitington and Sykes,[4] investigated the frequency in which violence in inpatient psychiatric units was preceded by an aversive interpersonal situation. The investigators found that a significant proportion of assaults in their study were immediately preceded by the assaulted nurse having delivered an aversive stimulus to the patient (i.e., activity demand or physical contact).

Several studies have highlighted the need for a comprehensive training for staff members to address these factors. This training would go a long way to further implement the Joint Commission of Accreditation of Healthcare Organization’s (JCAHO) standards, which have placed an emphasis on limitation of the use of restraints and seclusion on inpatient units. There are several recognized effective training modules available, including Mandt training, Therapeutic Options training, and the Professional Assault Response Training 2000 (PART 2000). The latter training is mainly used internationally.
Rueve and Welton[1] had a useful table in their article (Table 3), which talked about environmental modifications to control aggression, and in addition to those given in their list, a study by Nijman, et al.,[5] discusses certain effective approaches. That study mentions that discussing treatment goals with patients shortly after admission and discussing certain protocol for the facility (e.g., explaining why the doors are locked, explaining exit rules, providing patients with a schedule for staff meetings to explain absence of staff members from the psychiatric inpatient unit, and clarifying the procedure to make appointments with the psychiatrists) all showed a marginally significant difference in severity of aggressive incidents on wards with these interventions compared to control wards.

The aspect of violence in mental illness will continue to be a topical issue, and continued research in this area should be encouraged.

References
1. Rueve M, Welton R. Violence and mental illness. Psychiatry (Edgemont) 2008;5(5):34-48.
2. Lowe T, Wellman N, Taylor R. Limit-setting and decision making in the management of aggression. J Adv Nurs. 2003;41(2):154–161.
3. Jansen G, Middel B, Dassen WN, Reijneveld MSA. Cross-cultural differences in psychiatric nurse’s attitudes to inpatient aggression. Arch Psychiatr Nurs 2006;20(2):82–93.
4. Whittington R, Wykes T. Aversive stimulation by staff and violence by psychiatric patients. Br J Clin Psychol. 1996;35:11–20.
5. Nijman HL, Merckelbach HL, Allertz WF, Campo JM. Prevention of aggressive incidents on a closed psychiatric ward. Psychiatr Serv 1997;48:694–698.

With regards,
Adegboyega Oyemade, MD
Addiction Psychiatrist, Heritage Behavioral Health Center, Inc., Decatur, Illinois

REGARDING ANTIDEPRESSANTS AND EFFICACY IN SHORT-TERM CLINICAL TRIALS

Dear Editor:

I would like to add two more factors besides the ones Dr. Feifel discusses in his article, “More depressing news on antidepressants: Should we panic?” which appeared in the April issue of Psychiatry 2008.[1]

One: There is a strong incentive for the clinical investigators to push marginally depressed patients into drug trials. Since the reimbursement is based upon number of patients enrolled, there is an unconscious bias to rate the depressive symptoms higher than they are to help patients meet the inclusion criteria.

The patients with milder and atypical forms of depression often show high therapeutic response to placebo and psychosocial interventions thus diluting the evidence of an antidepressant’s efficacy with the more severely depressed. Furthermore, the mildly depressed are more likely to show decline of their symptoms spontaneously irrespective of the treatment. This also works against showing differences between the antidepressant and the placebo.

Two: Once the patient is enrolled in the study, there is a strong incentive to rate the treatment response liberally. The investigator has a powerful motivation to see that the drug works. If the drug is effective, it means publication and the honor of bringing a new drug to the market. Thus, when rating the treatment response, there is an unconscious bias to show greater improvement than what is actually occurring. This once again dilutes the differences between the antidepressant and placebo response.

The above two factors were borne upon me while doing my first clinical trial—a four-week, double-blind, placebo–controlled study in depression. I observed how eager we were to enroll anyone who declared himself or herself to be depressed during the screening rounds that we conducted with all the new admissions throughout the state hospital where the research unit was situated. To enable patients to meet the study’s inclusion criteria, there was a strong inner compulsion to downplay comorbid conditions and to exaggerate the intensity of the depression. It did not take me long to realize that all the research subjects were improving quite rapidly, and, therefore, the study would run into the problem of falsely showing that the antidepressant was no better than placebo.

References
1. Feifel D. More depressing news on antidepressants: Should we panic? Psychiatry (Edgemont) 2008;5(4)–35–36.

With regards,
Surendra Kelwala MD
Livonia, Michigan

Dr. Yanofski is a fellow in child psychiatry at the University of Texas, Southwestern, Dallas, Texas, and Dr. Croarkin is a child psychiatry consult attending at the University of Texas, Southwestern, Dallas, Texas.

ABSTRACT

Human immunodeficiency virus (HIV) has a high comorbidity with major depression. Symptoms of depression may be attributed to ongoing HIV infection, thereby reducing the likelihood of timely treatment with antidepressants. This may contribute to the morbidity of both illnesses. This review focuses on an evidence-based approach to selecting antidepressants for first-line treatment of major depressive disorder in patients with HIV and acquired immune deficiency syndrome (AIDS). Some antidepressant medications have side effect profiles that may exacerbate the symptoms commonly seen in patients with HIV and AIDS. Others have side effects that, while normally problematic in the general population, may be helpful in counteracting the difficulties seen in HIV and AIDS patients. Other challenges in treatment include an array of possible drug-drug interactions between antidepressants and HIV medications. Clinicians should focus more on capitalizing on the side effects of psychotropic medications in this patient population than on trying to avoid drug-drug interactions.

Key Words

major depressive disorder, HIV, AIDS, antidepressants

INTRODUCTION

Diagnosing and treating major depressive disorder in patients with human immunodeficiency virus (HIV) and acquired immune deficiency syndrome (AIDS) is a critical issue for clinicians as it is common and its effects are significant. Symptoms of both illnesses often overlap and obscure diagnostic and treatment interventions. Depression in HIV is also a significant contributor to the morbidity of HIV itself. It can decrease quality of life for these patients, decrease adherence with HIV medications, and it has shown to decrease positive outcomes overall.[1] A study of over 3,000 patients with HIV started on nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) found the proportion of individuals with adherence of 80 percent or better was 26 percent, and concluded that nonadherence is therefore the biggest cause of failure of these medicines.[2] Furthermore, depression and HIV both negatively impact the course of each other concurrently. It is imperative that clinicians treat both conditions simultaneously with well-coordinated treatment plans.

This review will examine the efficacy, side effects, and drug interactions of antidepressants in this patient population. Clinicians must consider all three concerns simultaneously when selecting a first-line antidepressant medication for a patient with HIV or AIDS.

EFFICACY AND ADVANTAGES

At first glance, there is no reason to think that antidepressants that successfully treat depression in the general population would not be effective in treating depression in people with HIV. However, in the world of evidence-based medicine, inductive reasoning needs to be backed up by data. Hence, there are studies that have shown that similarly to major depression in the general population, antidepressants have efficacy in HIV patients. Selective serotonin reuptake inhibitors (SSRIs) continue to cause less side effects than tricyclic antidepressants (TCAs), but no single SSRI has emerged as clearly more effective than others.[11]

Data like this often shape the practice of treating major depression by persuading psychiatrists to use multiple trials of SSRIs before trying a TCA or other older classes of medicine. However, studies have not been able to show that interclass switching as opposed to switching to a new class has a clear advantage in treating symptoms.[12,29] Often combination treatment and augmentation treatment can be helpful in nonresponders or partial responders.

There are many studies that have looked at the efficacy of particular antidepressants in HIV and AIDS patients specifically. A 20-person, prospective study of sustained-release bupropion suggested it is effective for the treatment of depression in HIV-positive patients, regardless of HIV clinical staging.[19] A 15-person, prospective study of nefazodone showed that depressed HIV outpatients responded as well to nefazodone as other outpatient populations did.[20]

Medications that have shown efficacy in treating depression in patients with HIV include imipramine, desipramine, nortriptyline, amitriptyline, fluoxetine, sertraline, paroxetine, citalopram, escitalopram, fluvoxamine, venlafaxaine, nefazodone, trazodone, bupropion, and mirtazapine.21 However, only imipramine, fluoxetine, sertraline, and paroxetine have evidence from double-blind trials.[27]

Despite all these studies, no type of antidepressant has emerged as clearly superior. However, just because they all seem to work equally well, does not mean they are equal in every way. Antidepressants have different likelihoods of side effects and causing interactions with other medications. Side effects, the subject of the next section, are often the most helpful factors in making decisions about which antidepressants to use. Negative side effects will impact patient adherence, strongly impacting efficacy of treatment.

TOLERABILITY AND SIDE EFFECTS

TCAs have been shown to be equally effective to the newer antidepressant medications. Many psychiatrists have removed them from first-line treatment for depression because they feel their side effect profiles are less tolerable. However, while the side effects are different than those of SSRIs, in certain patients the side effects may actually be more tolerable. Other older antidepressants like monoamine oxidase inhibitors (MAOIs) have generally fallen below both SSRIs and TCAs in the treatment line, mainly because of concerns about strict dietary precautions that must be maintained when using these medications.

TCAs have anticholinergic side effects that include dry mouth, blurred vision, decreased gastrointestinal (GI) motility (promoting constipation and decreasing diarrhea), and urinary retention. Other side effects include drowsiness, confusion, dizziness, weight gain, hypotension, and tachycardia. However, some of these side effects can be taken advantage of when treating HIV and AIDS patients, specifically weight gain, increased sleep, and decreased diarrhea. By starting at low doses and increasing slowly, the clinician may see a favorable ratio of the positive and negative effects of the medicine being reached in the patient.

SSRIs can be activating, sedating, or neutral. These medications also commonly affect patients gastrointestinally and sexually in a negative manner. However, bupropion is one medicine in this group that does not show sexual side effects. Sustained-release bupropion appears to be well tolerated in patients with AIDS-related medical conditions.[22] Other newer medicines have also been studied for tolerability. Depressed HIV outpatients taking nefazodone had few adverse effects.[20]

USING SIDE EFFECTS TO TREAT SYMPTOMS OF HIV

Knowledge of individual side effects associated with the various antidepressants and antidepressant classes can be helpful when deciding which to use with any depressed patient. However, HIV patients present specific challenges because of their medical problems and diverse symptoms. Ideally, a well-informed selection of antidepressant treatment a patient with HIV or AIDS involves broad consideration of all of the patient’s symptoms. Side effect profiles of antidepressants may be exploited in that they can be beneficial in some circumstances.

Fatigue. One study showed that the prevalence of clinical fatigue in men with CD4 counts less than 500 was 14 percent. This was significantly higher than HIV- men and HIV+ men with CD4 counts greater than 500. The study showed that the fatigue was not directly correlated with CD4 count or HIV ribonucleic acid (RNA). It was thought to be associated with depression in HIV patients, but did not seem to be merely a symptom of depression.[22] If a patient complains of fatigue, the clinician would not want to worsen that symptom by prescribing an antidepressant that would cause sedation, and the clinician may want to offer an antidepressant that has been shown to be more stimulating.

Studies comparing antidepressants specifically treating symptoms of fatigue are rare. It is usually expected that the fatigue will decrease as the depression improves. Buproprion is known to be stimulating, but has shown a slight increase in risk of seizures.26 Of the SSRIs, we believe fluoxetine, anecdotally, is the most stimulating.
Insomnia. Many HIV patients may experience insomnia as part of their depression. These patients may benefit from sedating antidepressants and may have worsened insomnia by stimulating antidepressants.

All of the TCAs can be sedating. One study suggests that stimulation of serotonin 5-HT2 receptors as seen in SSRIs can worsen insomnia, but that antidepressant drugs with 5-HT2 blocking properties are a good option for treating depressed patients with insomnia. These include medicines such as trazodone, nefazodone, and mirtazapine.25 Paroxetine is an SSRI that has sedative effects in some people.

Treating weight loss. A two-year, prospective, HIV study showed that 58 (31%) of 187 enrolled HIV+ men had significant body mass depletion at some point during the study, and many of them showed consistent weight loss during the study. The additional weight loss correlated with ‘increased fatigue, global distress and depressive symptomatology, and reduced life satisfaction.’[23]

Many antidepressants are associated with weight gain. All MAOIs and TCAs have been shown to have associated weight gain. Within the SSRI group, paroxetine has shown the most association with weight gain, but sertraline, mirtazapine, and fluvoxamine also have associations. Venlafaxine and nefazadone have no weight changes, and patients on bupropion have been shown to lose weight.[24]

Treating diarrhea. The anticholinergic side effects of TCAs that normally cause constipation may be helpful when patients have diarrhea. Mirtazapine can also be constipating.

Treating chronic pain. Many HIV-infected patients have chronic pain from relating medical issues or comorbidity with opioid dependence. Some antidepressants have a role in decreasing chronic pain.

Trazodone does not appear to be effective for treating chronic pain. Bupropion, venlafaxine, and duloxetine have been shown to relieve different kinds of chronic pain. Other antidepressants do not have clear data on this matter.28

CONCERNS OF DRUG INTERACTIONS

The HIV drug ritonavir is a protease inhibitor that can be given at a small dose as a synergistic enhancer, reducing the liver metabolism of other antiretroviral drugs. It is often given in combination with the protease inhibitor lopinavir (in a form called kaletra) in a 1:5 ratio. It also enhances other protease inhibitors (saquinavir, atazanavir). Ritonavir has been a main focus of study when looking at interactions between antidepressants and HIV medicines.

TCAs. TCAs, including nortriptyline, desipramine, imipramine, amitriptyline, clomipramine, and doxepin, have all been shown to have increased levels in plasma by ritonavir and ritonavir combinations. However, these increases have often been shown to be problematic even to the point of needing dose adjustments. Furthermore, TCA blood levels can and should be checked even in patients not on other medications with possible interactions.

Desipramine, in the setting of ritonavir, was shown to have clearance decreased by 59 percent. In the setting of nelfinavir, desipramine was shown to have increased blood levels. Both ritonavir and nelfinavir independently showed potential to increase TCA side effects.[13,14] Despite possible increased desipramine levels in the setting of antiretrovirals, it is not thought that desipramine would require dose adjustments.[6] Amitriptyline in the setting of ritonavir showed increased levels and potential for increased TCA side effects.

There have also been studies that show that some antifungals, such as terbinafine, can inhibit the metabolism of TCAs, specifically nortriptyline and desipramine.[7]
SSRIs. SSRIs that have been shown to have decreased metabolism in the setting of ritonavir include sertraline and citalopram. Paroxetine and escitalopram have no data on this. Fluoxetine and fluvoxamine are both decreased by nevirapine, and they both increase the levels of the following HIV medications: amprenavir, delarvidine, efavirenz, indinavir, lopinavir/ritonavir, nelfinavir, ritonavir, and saquinavir.[4]
Escitalopram. Escitalopram in the setting of ritonavir showed no effect on levels or side effects.[13,16]

Fluoxetine. Fluoxetine in the setting of delavirdine showed that the delavirdine Cmin increased by 50 percent, and there were some increased delavirdine effects. Fluoxetine in the setting of ritonavir had a ritonavir AUC increased by 19 percent, no change to Cmax, and increased ritonavir effects. There was no effect on fluoxetine levels or side effects.[13,15]

Paroxetine. Paroxetine in the setting of darunavir showed a decreased paroxetine AUC by 39 percent, Cmax by 36 percent, Cmin by 37 percent, and showed decreased paroxetine effects. In the setting of fosamprenavir, paroxetine had similar results with a decreased AUC by 58 percent, Cmax by 60 percent, and half-life of 25 percent, with decreased paroxetine effects. The retrovirals did not have affected levels.[13,17]
Sertaline. Sertraline in the setting of darunavir showed decreased AUC by 49 percent, Cmax by 44 percent, and Cmin by 49 percent with decreased sertraline effects. There were no effects on the darunavir levels or efficacy.[13]

Citalopram. Citalopram has been shown to have no interactions with ritonavir.[18]
Trazodone. Trazodone has the most studies (darunavir, indinavir, lopinavir/ritonavir, ritonavir [x2]), and all five of these studies showed potential to increase the trazodone side effects of nausea, dizziness, hypotension, and syncope.[13] Additionally, in the setting of darunavir and lopinavir/ritonavir, there were increased trazodone concentrations, and both ritonavir studies showed specifically that trazodone had an increased AUC by 240 percent, Cmax by 34 percent, and a half-life by 220 percent.[13]
Theoretical interactions. Outside of these measured effects, knowledge about the metabolism of other similar antidepressant medications can lead to theoretical effects of combination with antiretroviral medications. Many SSRIs and new multireceptor targeting medications are known to be weak or potent inhibitors on the CYP isoforms.[9,10]

While most studies show no clinical significance of the interactions of antidepressants and HIV medications, one report alerts of five cases of serotonin syndrome that developed after patients who were taking fluoxetine ingested antiretrovirals that were P450 inhibitors.[8]

However, theoretical possibilities or small case reports should not guide treatment in all cases. Until bigger trends can be shown, clearly defined side effects of drugs and drug classes are better suited to serve as a guide for treatment. This is what will determine patient adherence with treatment plans, and in the real world setting, nothing is more important than getting patients to take their medicines.

See Table 1 for theoretical CYP inhibitions and actual effects of SSRIs and other antidepressants in the setting of retroviruses.

GUIDELINES AND SUMMARY

Patients with HIV and AIDS should be frequently monitored for depressive symptoms, and those that have them should be monitored by one of many available scales of depression symptoms. Education about HIV and mental health is always helpful, and talk therapy of different types can have benefit regardless of psychiatric diagnosis because of the stressors related to having HIV and AIDS.

Major depressive disorder should be diagnosed carefully. However, in many cases, clinicians should have a low threshold for initiating treatment in this patient population. If the syndrome of major depression can be discerned and criteria are met, then treatment is certainly warranted. However, treatment can be justified even without meeting full criteria if certain symptoms are intense or dangerous, such as suicidality, if patients are nonadherent with HIV medications in the setting of low mood or other depressive symptoms, or if there are side effects of antidepressants that can be used for their positive treatment effects.

When it is decided to treat depression with a medication, normal methods of treating depression are applicable. Start slow and taper up. Do not recommend discontinuation of a drug because of side effects during the first week, unless severe. If the medication is tolerable but ineffective, then a full trial should be continued for 4 to 6 weeks at a therapeutic dose. If it is working but causing undesirable side effects, use other medications to fight those side effects to a reasonable degree. If there is partial success, use augmentation or a second agent.

Both TCAs and SSRIs/multireceptor antidepressants are equally effective treatments. Both have theoretical drug interactions with antiretroviral medications, but these effects have not been shown clinically to be strong enough to affect the choice of antiretroviral or antidepressant class of medicine to use first line. SSRIs are considered the safest, particularly in the case of overdose, and they tend to have less side effects. Though there are theoretically different risks of interacting with HIV medicines, it is more practical to choose medicines based on their side effect profiles.
TCAs, while not first line generally, have the advantage in that their levels can be monitored, which helps to prevent toxicity and increase patient adherence. In treating major depression in the general population, it is not unusual to try several antidepressants in the newer classes before initiating a trial of TCAs. In patients with more advanced AIDs or with somatic symptoms that may be amenable to side effects of TCA, such as diarrhea, insomnia, or weight loss, TCAs would be appropriate first-line treatment for depression.

References
1. Dew MA, Becker JT, Sanchez J, et al. Prevalence and predictors of depressive, anxiety and substance use disorders in HIV-infected and uninfected men: a longitudinal evaluation. Psychol Med. 1997;27(2):395–409.
2. Becker SL, Dezii CM, Burtcel B, et al. Young HIV-infected adults are at greater risk for medication nonadherence. MedGenMed. 2002;4(3):21.
3. Ouellet D, Hsu A, Qian J, et al. Effect of fluoxetine on pharmacokinetics of ritonavir. Antimicrob Agents Chemother. 1998;42(12):3107–12.
4. Triesman G, Angelino A. The Psychiatry of AIDS: A Guide to Diagnosis and Treatment. Baltimore, MD: Johns Hopkins Press, 2004.
5. Gutierrez MM. An evaluation of the potential for pharmacokinetic interaction between escitalopram and the cytochrome p450 2A4 inhibitor ritonavir. Clin Ther. 2003;25(4):1200–10.
6. Aarnoutse RE. Effect of low-dose ritonavir (100 mg twice daily) on the activity of cytochrome P450 2D6 in healthy volunteers. Clin Pharmacol Ther. 2005;78(6):664-74.
7. Venkatakrishnan K, von Moltke LL, Greenblatt DJ. Effects of the antifungal agents on oxidative drug metabolism: clinical relevance. Clin Pharmacokinet. 2000;38(2):111–80.
8. DeSilva KE, Serotonin syndrome in HIV-infected individuals receiving antiretroviral therapy and fluoxetine. AIDS. 2001;15(10):1281–5.
9. Spina E, Scordo MG, D’Arrigo C. Metabolic drug interactions with new psychotropic agents. Fundam Clin Pharmacol. 2003;17(5):517–38.
10. Hemeryck A, Belpaire FM. Selective serotonin reuptake inhibitors and cytochrome P-450 mediated drug-drug interactions: an update. Curr Drug Metab. 2002;3(1):13–37.
11. Caballero J, Nahata MC. Use of selective serotonin-reuptake inhibitors in the treatment of depression in adults with HIV. Ann Pharmacother. 2005;39(1):141–5. Epub 2004 Nov 23.
12. Ruhe HG. Switching antidepressants after a first selective serotonin reuptake inhibitor in major depressive disorder: a systematic review. J Clin Psychiatry. 2006;67(12):1836–55.
13. McNicholl IR. USCF Center for HIV Information. Database of antiviral interactions. http://hivinsite.ucsf.edu/InSite. Access date: April 23, 2008.
14. von Moltke LL, Greenblatt DJ, Duan SX, et al. Inhibition of desipramine hydroxylation (Cytochrome P450-2D6) in vitro by quinidine and by viral protease inhibitors: relation to drug interactions in vivo. J Pharm Sci 1998;87(10):1184–89
15. Ouellet D, Hsu A, Qian J, et al. Effect of fluoxetine on pharmacokinetics of ritonavir. Antimicrob Agents Chemother 1998;42:3107–12.
16. Gutierrez MM, Rosenberg J, Abramowitz W. An evaluation of the potential for pharmacokinetic interaction between escitalopram and the cytochrome P450 3A4 inhibitor ritonavir. Clin Ther. 2003;25:1200–10.
17. Blenke A, van der Lee M, Varweij-van Wissen C, et al. [abstact #13]. 6th International Workshop on Clinical Pharmacology of HIV Therapy; 2005 April 28-30; Quebec City, Quebec, Canada.
18. Gutierrez MM, Rosenberg J, Abramowitz W. An evaluation of the potential for pharmacokinetic interaction between escitalopram and the cytochrome P450 3A4 inhibitor ritonavir. Clin Ther. 2003;25(4):1200–10.
19. Currier MB, Molina G, Kato M. A prospective trial of sustained-release bupropion for depression in HIV-seropositive and AIDS patients. Psychosomatics. 2003;44(2):120–5.
20. Elliott AJ, Russo J, Bergam K, et al. Antidepressant efficacy in HIV-seropositive outpatients with major depressive disorder: an open trial of nefazodone. J Clin Psychiatry. 1999;60(4):226–31.
21. Elliott AJ, Roy-Byrne PP. Mirtazapine for depression in patients with human immunodeficiency virus. J Clin Psychopharmacol. 2000;20(2):265–7.
22. Ferrando S, Evans S, Goggin K, et al. Fatigue in HIV illness: relationship to depression, physical limitations, and disability. Psychosom Med. 1998 Nov-Dec;60(6):759–64.
23. Wagner GJ, Ferrando SJ, Rabkin JG. Psychological and physical health correlates of body cell mass depletion among HIV+ men. J Psychosom Res. 2000;49(1):55–7.
24. Deshmukh R, Franco K. Managing weight gain as a side effect of antidepressant therapy. Cleve Clin J Med. 2003;70(7):614–618, passim.
25. Thase ME. Antidepressant treatment of the depressed patient with insomnia. J Clin Psychiatry. 1999;60(Suppl 17):28–31; discussion 46–8.
26. Mainie I, McGurk C, McClintock G, Robinson J. Seizures after buproprion overdose. Lancet. 2001;357(9268):1624.
27. Stephen J. Ferrando S. Managing depression in HIV disease, viral hepatitis, and substance abuse. PRN Notebook 2005. http://www.prn.org/index.php/complications/article/managing_depression_in_hiv_disease_viral_hepatitis_and_substance_abuse/. Access date: May 2, 2008.
28. Miller A, Rabe-Jablonska J. The effectiveness of antidepressants in the treatment of chronic non-cancer pain. Psychiatr Pol. 2005;39(1):21–32.
29. Warden D, Rush AJ. The STAR*D Project results: a comprehensive review of findings. Curr Psychiatry Rep. 2007;9(6):449–59.

Ms. Cascade is Vice President, Quintiles Inc./iGuard, Falls Church, Virginia; Dr. Kalali is Vice President, Global Therapeutic Group Leader CNS, Quintiles Inc., San Diego, California, and Professor of Psychiatry, University of California, San Diego; and Dr. Buckley is Professor and Chairman, Department of Psychiatry,
Medical College of Georgia, Augusta, Georgia.

Financial Disclosures

Dr. Buckley has received grant/research support from AstraZeneca, National Institute of Mental Health, Pfizer, Solvay, and Wyeth, is a consultant to AstraZeneca, Bristol Myers Squibb, Eli Lilly, Janssen Pharmaceutica, Lundbeck, Pfizer, Solvay, and Wyeth, and has received honorarium/expenses from Bristol Myers Squibb, Janssen Pharmaceutica, Lundbeck, and Pfizer.

Abstract

We investigated the current practice for treatment of schizophrenia. According to our data, 47 percent of patients with schizophrenia are treated with only antipsychotic medication. An additional 43 percent of patients receive one additional class to supplement their antipsychotic medication, and 10 percent of patients are prescribed two or more classes of drugs in addition to an antipsychotic medication. The most common classes used to supplement antipsychotic medications in the management of schizophrenia include antidepressants (28%), mood stabilizers (18%), agents to treat extrapyramidal symptoms (7%), and sleep aids (5%).

Key words

schizophrenia, monotherapy, combination therapy, augmentation, antipsychotics, antidepressants

Introduction

To better understand current practice for the management of schizophrenia, we investigated the use of antipsychotics alone and the need for supplementation with other classes of agents to control symptoms.

Methods

We obtained data on product treatment regimen from Verispan’s Prescription Drug and Diagnosis Audit (PDDA) database from March, 2007, to February, 2008, for schizophrenia as defined by ICD-9 diagnosis code 295. PDDA captures data on disease state and associated therapy from 3,100 office-based physicians representing 29 specialties across the United States.

Results

As seen in Figure 1, 47 percent of patients with schizophrenia are treated with only antipsychotic medication. An additional 43 percent of patients receive one additional class to supplement their antipsychotic medication, and 10 percent of patients are prescribed two or more classes of drugs in addition to an antipsychotic. According to our data, 28 percent of patients with schizophrenia are prescribed an antidepressant in combination with their antipsychotic (Figure 2). Other therapeutic classes used to supplement antipsychotics in the management of schizophrenia include mood stabilizers (including lithium or antiepileptics, 18%), agents to treat extrapyramidal symptoms (7%), and sleep aids (5%).

Commentary

These data, drawn from clinical practice-prescription profile analyses, “jive well” with what is currently happening in our clinics and in our hospitals. The treatment of schizophrenia and bipolar disorder have (paradoxically) become increasingly complex with the greater availability and choice among antipsychotic medications. At the same time, there is still a substantial unmet need, as confirmed by recent, large, pragmatic trials in schizophrenia and bipolar disorder, that provides the therapeutic context for antipsychotic polypharmacy as well as the use of other psychotropic medications alongside antipsychotics. For patients and clinicians, then, the question of “why and when do I combine medications?” is now very challenging. As these data also bear out, other available evidence suggests that antipsychotic polypharmacy is common in clinical practice. Additionally, it is a topic of enduring interest among clinicians who are always eager to understand the information contributing to key therapeutic strategies. These data also attest to the high rates of concomitant use of other psychotropic agents, particularly antidepressant medications.

At the present time, we are lacking robust research literature to guide this decision-making process that is the clinician’s dilemma: Why, when, and how should I use antipsychotic polypharmacy to its best advantage in my practice? Available scientific evidence for polypharmacy strategies in schizophrenia is scant and does not preferentially endorse any single strategy. This latter point is particularly relevant when proponents of polypharmacy say “Hey, this approach works for my patients” yet cannot articulate a single drug option, and so it again resides in the individualization of care for patients with schizophrenia. Another consideration to bear in mind here is that some polypharmacy is not so much driven by augmentation strategies, but rather by “suspended” switching strategies. That is, you and your patient decide to switch medications but you ended up stopping this midway, and so the patient stays on both medications…and ends up in the so called psychopharmacological purguratory as Dr. Peter Weiden has aptly termed this.

The sizeable rate of concomitant use of antidepressants recorded in these analyses is also worthy of comment. It is well known that patients with schizophrenia get depressed. Depression in schizophrenia worsens the outcome, both in terms of heightening the risk of psychotic relapses and also greatly increasing the risk of suicide. Accordingly, the prescription of antidepressant medications for patients with schizophrenia who become depressed is an important therapeutic strategy. Interestingly, recent work on antidepressant therapy for mood disorders suggests that these drugs may have neuroprotective effects, perhaps through increasing the brain’s neurotrophins, such as brain-derived neurotrophic factor (BDNF). Interestingly, the clinical and translational impact of antidepressant therapy has not yet been studied in the context of second generation antipsychotic medications (beyond clozapine). These data would suggest that clinicians are using these agents regularly in the treatment of patients with schizophrenia.

It would be useful to have some data from pragmatic trials to inform this fundamental practice strategy for clinicians and to examine the extent to which polypharmacy truly impacts both efficacy and tolerability considerations in treating patients with schizophrenia.