Patients With Gastrointestinal Disease

Xerostomia

Xerostomia is defined by dry mouth, with or without decreased salivary production. Psychotropic medications are a contributing cause in 10–50% of patients. In a recent analysis that examined the relationship between xerostomia and psychotropic medications (first- and second-generation antipsychotics and anxiolytics) in patients with schizophrenia, Okamoto and colleagues found a negative correlation between the number of antipsychotics and, especially, anxiolytics, and the degree of oral moisture. Mean oral moisture was decreased in patients as opposed to controls and there was no significant correlation with dose, whereas the number of medications used appeared to be significantly correlated with dry mouth. Benzodiazepines, lithium, carbamazepine, anticholinergics, and typical and atypical antipsychotics are all associated with xerostomia. Antihypertensives, diuretics, and opioids are also associated with xerostomia. Other causes include connective tissue disorders, such as Sjogren's syndrome, radiation therapy, anxiety, and depression. Non-pharmacologic treatment includes frequent sips of water, using sugarless gums and candies, avoiding caffeine and alcohol, and using saliva substitutes. Pilocarpine is an effective treatment, though its use is contraindicated in patients with closed-angle glaucoma. Masters reported its successful use in psychiatric inpatients in doses of 10–30 mg/day, divided into dosing of two or three times a day. Sweating and increased urination were the most common side effects. Another medication treatment option for xerostomia is sublingual bethanechol.

Dysphagia

Dysphagia has been reported to occur secondary to antipsychotic medications (first- and second-generation) as a result of drug-induced parkinsonism, dystonia, or tardive dyskinesia. Case reports and case series have implicated various agents, including: haloperidol, loxapine, trifluoperazine, olanzapine, risperidone, quetiapine, clozapine, and aripiprazole. Treatment should address the suspected etiology. Thus, dysphagia due to drug-induced dystonia should be treated with IV or IM benztropine or diphenhydramine. Dysphagia due to drug-induced parkinsonism is typically not responsive to anticholinergic medication and is best handled with reducing or discontinuing the suspected agent. For patients with dysphagia, regardless of primary etiology, care should be taken when prescribing sedating medications, as over-sedation in this population may result in aspiration and other dreaded events.

Globus Hystericus

The feeling that something is lodged in one's throat is an oft-reported symptom that requires medical attention. When pertinent diseases, including gastroesophageal reflux disease, airway masses, or esophageal masses have been ruled out, psychogenic causes are often considered. Functional dysphagia, classically known as globus hystericus , has a wide differential diagnosis as well as a high rate of co-morbidity with other medical conditions. Management includes treating co-morbid conditions, and when applicable, psychotherapy and possibly medications for co-morbid anxiety. If globus is thought to be due to a functional (or conversion) disorder, reassurance, psychoeducation, and psychotherapy is preferred.

Gastroesophageal Reflux Disease

Gastroesophageal reflux disease (GERD) is associated with increased rates of many psychiatric disorders. In a retrospective Taiwanese study, You and colleagues found that GERD was associated with increased rates of depressive (HR = 2.91, 95% CI = 2.34–3.61, p <0.001), anxiety (HR = 2.75, 95% CI = 2.15–3.50, p <0.001), and sleep disorders (HR = 2.65, 95% CI = 2.02–3.47, p <0.001) versus controls without GERD. There did not appear to be a significant increase in rates of bipolar disorder or schizophrenia; however, in another Taiwanese study, Lin and colleagues noted an increased incidence of bipolar disorder (incidence rate ratio, IRR 2.29, 95% confidence interval, CI = 1.58–3.36, p <0.001) among GERD patients than among controls. In another study that examined rates of GERD in patients with major depression, Chou and colleagues found that patients with major depressive disorder had a significantly increased rate of GERD (odds ratio, OR = 3.16; 95% CI = 2.71–3.68; p <0.001). Various presumed mechanisms may account for the correlations between psychiatric illness and GERD, including increased production of pro-inflammatory cytokines from the esophageal mucosa and increases in autonomic (sympathetic) nervous system activation during coughing and arousal from sleep. Most psychotropic medications can be used in patients with GERD without worsening symptoms; an exception is the use of anticholinergic medications that should be avoided if possible. Benzodiazepines and SNRIs have been associated with improvements in sleep, a sense of well-being, and even reduction in core GERD symptoms.

Nausea and Vomiting

Nausea is a common side effect of serotonergic antidepressants due to agonism of 5HT ₃ receptors in the gut. Medications that antagonize 5HT ₃ receptors (especially mirtazapine, olanzapine, and ondansetron) are less likely to cause GI distress and may be useful in patients with nausea and co-morbid depression or anxiety. Blockage of D ₂ receptors, particularly in the chemoreceptor trigger zone, is also helpful in alleviating nausea and vomiting. Metoclopramide and prochlorperazine (a phenothiazine antipsychotic) are D ₂ receptor antagonists marketed as antiemetics. While they are often effective over the short-term, their use may be limited by extrapyramidal symptoms (EPS), such as akathisia, and long-term effects, such as tardive dyskinesia (TD). For patients with depression or anxiety and chemotherapy-induced nausea and vomiting (CINV), hyperemesis gravidarum (HG), and other conditions, the antiemetic properties of mirtazapine and certain antipsychotics (e.g., olanzapine, perphenazine, chlorpromazine, prochlorperazine, and other antipsychotics), and benzodiazepines can be considered if no other contraindications are present.

In a randomized, double-blind, placebo-controlled trial of patients scheduled to receive emetogenic chemotherapy, Mizukami and colleagues demonstrated that the addition of 5 mg of olanzapine daily was associated with significant reductions in nausea and vomiting as well as with increases in quality of life over the control group. Lohr reviewed treatment options for chemotherapy-induced nausea and vomiting and discussed potential treatment options. The psychotropic medications that have been shown to be efficacious in CINV include lorazepam (also found to be helpful, including systematic desensitization, hypnosis, biofeedback, imagery, and relaxation ), butyrophenone antipsychotics (including haloperidol and droperidol), phenothiazine antipsychotics (e.g., prochlorperazine and promethazine), olanzapine, and gabapentin. Mirtazapine has also been found to be effective in treating nausea and depression in cancer patients. Anticipatory nausea and vomiting may cause a considerable amount of distress in cancer patients who are receiving chemotherapy. Psychological models for chemotherapy-related anticipatory nausea and vomiting include classical conditioning (e.g., associating chemotherapy with environmental cues or other physical sensations), demographic factors (including younger age, female gender, and the propensity to experience certain physical symptoms such as dizziness), and beliefs and negative expectations related to treatment and symptom formation. Adjunctive lorazepam has been shown to be helpful in anticipatory nausea and vomiting in this population.

Although the belief that hyperemesis gravidarum (HG) is psychogenic has been discredited, there is evidence to support the notion that this condition is associated with high degrees of psychological stress and with elevated rates of depression and anxiety disorders, even after pregnancy. In a study of 47 patients with HG, 25.5% had a mood disorder and the prevalence of any mood disorder was 14.9% in women in the first trimester. Psychiatric disorders continued throughout the pregnancy in two-thirds of the women with HG and a psychiatric diagnosis. In another study of 52 women with HG, Uguz and colleagues found that the prevalence of mood disorders was 15.4% and that of anxiety disorders was 36.5%. A significant number of patients (36.5%) with HG had at least one personality disorder diagnosis (particularly an avoidant or obsessive–compulsive personality disorder). Furthermore, most of the mood or anxiety disorders occurred before the pregnancy in women with HG. Treatment of HG typically involves vitamin supplementation (especially thiamine to avoid Wernicke's encephalopathy), pyridoxine (vitamin B ₆ ) with or without doxylamine, and for more severe cases, use of metoclopramide or steroids. Mirtazapine has been used successfully as has chlorpromazine.

Cyclical vomiting syndrome (CVS) is a condition consisting of recurrent episodes of incapacitating nausea and vomiting that are interspersed with symptom-free intervals lasting anywhere from a few days to several months. While most common in children, it also occurs in adults and may be associated with specific triggers or environmental cues (e.g., migraines, seizures, stress, menstrual cycles). There have been many proposed etiologies; more recent plausible causes include sympathetic and parasympathetic dysfunction, stress/anxiety/depression, and a central mechanism of nausea/vomiting involving corticotropin releasing factor. Treatment of CVS is multi-factorial, complex, and involves both pharmacologic and non-pharmacologic strategies. Interventions include identification and avoidance of any known triggers, use of prophylactic drug therapy to prevent recurrent episodes, use of abortive treatment and/or supportive care to ameliorate acute episodes, and provision of psychological support of the patients and family. Various medications (including prokinetics, antiemetics, erythromycin, sumatriptan, TCAs, benzodiazepines, and anticonvulsants) have been used with varying success. Acute vomiting episodes may require high-dose IV antiemetics and benzodiazepines. Long-term management with TCAs has been recently reviewed and found to be effective in most patients in decreasing the duration and frequency of episodes, emergency visits, and hospitalization in adults with CVS. While side effects with TCAs are common, Hejazi and McCallum reported that they were generally mild and well-tolerated. The role of neuropsychiatric consultation and intervention may be particularly valuable in these patients; the major risk factors for non-response to treatment are co-existing poorly controlled migraine headache, a psychiatric disorder, and chronic narcotic and marijuana use. Additionally, a personal or family history of migraine and the presence of co-morbid psychiatric disorders is thought to be associated with an increased risk (and likelihood if symptoms are present) of CVS.

Cannabis hyperemesis syndrome (CHS) is similar to CVS but it can be distinguished by pathognomonic behavior of frequently bathing in hot water and by chronic cannabis use. Paradoxically, cannabis usually functions as an antiemetic, however, it is thought that with chronic use, toxic levels accumulate and activate CB ₁ receptors in the gut. The peripheral CB ₁ binding over-rides the common antiemetic properties on CB ₁ receptors in the central nervous system and CB ₂ receptors on glial cells. The result is decreased GI motility and emesis. The mechanism related to hot water bathing is unknown but may include CB ₁ activation by THC in the hypothalamus as well as a “cutaneous steal syndrome,” which is thought to occur when blood flow is re-directed from the gut to the skin, resulting in emesis. Management, other than replenishing fluids and nutrients, may involve the judicious and cautious use of benzodiazepines. Antiemetics have not been shown to be effective. The most effective treatment is likely the cessation of cannabis use; indeed, the diagnosis is supported by the fact that patients with CHS tend to have a remission of symptoms when stopping cannabis use and that symptoms tend to resume when cannabis is used again.

Nausea and GI distress are the most common reason for discontinuation of SSRIs by patients. SSRI-induced nausea is generally benign, transient, and if needed, can be managed with over-the-counter antiemetics. In addition to use of SSRIs, nausea is a common early side effect of psychostimulants, mood stabilizers, and anticholinesterase inhibitors. Lithium is associated with nausea and vomiting but longer-acting formulations (e.g., Eskalith CR) or lithium citrate may reduce this effect (though lower GI symptoms are possible). For patients taking sodium valproate, switching to divalproex sodium may help reduce nausea and vomiting. For patients taking carbamazepine, dividing the total daily dose to twice daily may also alleviate nausea. If nausea occurs in the setting of other side effects associated with drug toxicity, such as neurologic or other GI symptoms, consideration of drug toxicity and prompt evaluation should ensue.

Gastroparesis

Gastroparesis is characterized by delayed gastric emptying without an identifiable bowel obstruction. Patients with gastroparesis may experience symptoms including abdominal pain, nausea, vomiting, bloating, and early satiety. Common causes include complications from diabetes mellitus, post-surgical complications, neurologic injury, and medication side effects. Anticholinergic medications worsen gastroparesis and should be avoided. This includes anticholinergic medications used for prophylaxis and/or treatment of extrapyramidal symptoms (EPS) or dystonic reactions, such as benztropine and Benadryl, as well as anticholinergic antipsychotic medications such as clozapine, chlorpromazine, and other low potency first-generation agents. Pharmacologic treatment for gastroparesis involves using prokinetic agents, such as erythromycin, bethanechol, and metoclopramide. Metoclopramide is a dopamine- ₂ receptor antagonist that is used in a variety of GI disorders. Long-term use has been historically associated with an increased risk of tardive dyskinesia (TD) with rates ranging from 1% to 10% in different studies; however, a 2010 review by Rao and Camilleri showed the risk of TD to be <1%. As with other D ₂ -receptor antagonists, metoclopramide may also be associated with other dreaded reactions including akathisia, dystonic reactions, and possibly, neuroleptic malignant syndrome. A recent meta-analysis found that continuous IV administration is associated with a reduced rate of EPS when compared with bolus administration, and this may be an option for patients prone to these conditions.