Psychopharmacologic interventions in the refractory psychotic

Vernon M Neppe MD, PhD

Educational Objectives

  1. To educate in the area of refractory, atypical and nonresponsive psychosis, particularly the clinical psychopharmacologic management.
  2. To create a workable outline with regard to management options and problems.

Non-responsive psychosis refers to a chronic non-neuroleptic responsive psychosis. "Non-responsive psychosis" is the author's term for patients with a history of a hallucinatory-delusional condition who have not returned to their pre-morbid level of functioning despite adequate neuroleptization for a minimum period of at least one year. The term is thus non-specific and descriptive including as a subgroup chronic schizophrenics. This encompasses far more than typical deteriorating schizophrenia as many non-responsive psychotics are not schizophrenic. We should seek out and manage "non-responsive" psychotics who do not exhibit the typical chronic schizophrenia. The patient with refractory psychosis is particularly difficult to treat. These patients have invariably failed trials of several neuroleptics. Their treatments of necessity go beyond FDA approved indications and are still being researched.

A specific treatment approach is outlined in my book Innovative Psychopharmacotherapy.

Three pharmacological principles to an approach to the non-responsive psychotic, namely toleration, responsiveness and pharmacologic tracing are useful. Non-toleration of neuroleptics implies that a "functional psychosis" cannot be present - organic pathology invariably is. Thus, review of diagnosis and exclusion of organicity is critical. Toleration without response may imply non-compliance. If the patient is complying, certain fundamental symptoms should be sought and adjunctive treatment to neuroleptics considered. Pharmacologic tracing is reflected by responsiveness to target symptoms.

This leads to the following plan:

  1. First ensure compliance, if necessary with short-acting intramuscular injection or with depot neuroleptics.
  2. Second, evaluate toleration of antipsychotic doses on the basis of extrapyramidal, hypnogenic, and autonomic side-effects. Non-toleration implies organic disease, often hyperthyroidism.
  3. Third, test the limits of appropriate neuroleptic: low doses in some instances are suitable, high doses in others particularly chronic hallucinosis. Dosage of neuroleptic is critical: while the continued presence of florid symptoms without side-effects implies increasing doses, neuroleptic overdosage is a common error.
  4. Fourth, recognize that not all neuroleptics are equal. Choice of neuroleptic is very important: their different receptor profiles allow specific guidelines for management. Each drug has special effects at several different receptors. Clozapine is especially topical and difficult to use. Two special options are the use of sulpiride in the presence of refractory positive psychotic features, and of pipothiazine palmitate when deficit features predominate, are often worth considering.

Risperidone with its profound serotonin 2 blockade and potent dopamine 2 blocking effects is already proving an exciting, major breakthrough. Management of target symptomatology such as florid psychosis and overmedication is in the context of neuroleptic manipulation.

The major approach, however, is to seek out target features. Add medication usually as adjunct to neuroleptic to treat these target elements.

Management of target symptomatology such as anxiety, extra-pyramidal features, affective components, hostility, tardive syndromes involves appropriate non-neuroleptic adjuncts. Frequently such innovative therapies are given as adjuncts to neuroleptics because they control specific symptoms not psychotic features. Propranolol, Buspirone and Carbamazepine are neglected forms of such innovative therapy and also useful are Lithium, Antidepressants, Dopamine agonists and Dietary / Pharmacologic environmental manipulations.

In the evaluation of non-responsive psychosis, there are several fundamental key symptoms that should be sought namely anxiety, hostility, manic irritability, labile mood, extrapyramidal features, tardive symptoms, depression and nutritional status, diet, use of alcohol, caffeine and smoking. These lead to different approaches. The following are highlighted:

The first fundamental symptom to be sought is anxiety. Tachycardia is an important feature in the non-responsive psychotic, many of whom still seem to have a substantial degree of anxiety, even after eliminating such obvious causes as thyrotoxicosis.

Most of the literature on benzodiazepines indicate that they do not apparently relieve anxiety in the psychotic, except in rare instances in very high doses. Benzodiazepines can, however, be used in mania, catatonia, for nonspecific sedation and in epileptics.

On the other hand, there is a substantial body of research claiming that d1-propranolol, a beta-adrenergic receptor blocking agent without intrinsic sympathomimetic activity and a substantial degree of lipid solubility, may in fact have antipsychotic-like actions. This anti-psychotic activity of propranolol is not established either empirically or on theoretical grounds. On theoretical grounds, it does not have any dopamine antagonism effect and does not therefore cause any kind of hyperprolactinaemia or growth hormone alteration.

Carbamazepine may be useful in hostility, lability, temporal lobe pathology and previous hallucinogen usage. The iminostilbene, carbamazepine, (CBZ) (Tegretol), is most commonly used as an anticonvulsant and is equally efficacious to phenytoin in the control of tonic-clonic seizures and is frequently used in complex partial seizures and atypical absences, and for trigeminal neuralgia. Side effects with regard to mental status are far less than with anticonvulsants such as phenytoin and phenobarbital.

CBZ is being increasingly used in the treatment of non-epileptic psychiatric patients although not marketed as such. The improvements in CBZ have consequently been interpreted as psychotropic actions. These may be a property of its special anti-convulsant profile or distinct from it. CBZ is also the prime limbic antikindling agent in experimental animals. The efficacy of CBZ in psychiatric illness is attested by increasing literature arising from Japanese, American and European research, inter alia, of its use in affective illness. There is a growing but still limited evidence for its usefulness in both the prophylaxis and treatment of affective illness and in non-responsive psychosis. Some uncontrolled evidence suggests carbamazepine's efficacy may be particularly marked in those patients who have had affective illness for periods of many years as opposed to new patients. Carbamazepine is today probably most often used in the rapid cycling patient in which fluctuations of mood have at times been noted over a matter of hours and days. In contrast, it has also been used only occasionally in schizo-affective illness and there are supporting double-blind s tudies However, preliminarily responses may have related to the affective features. A broad use of the drug is suggested by this, and it is probable that its mode of action is different from lithium. Lithium-responsive and carbamazepine-responsive affective illness may ultimately represent several separate conditions.

The use of carbamazepine has been extended beyond mood disorder to the non-responsive psychotic. This has been a consequence of the hypothesized control of limbic instability by carbamazepine, implying that it may have effects on schizophreniform illness associated with temporal lobe dysfunction, or effects on dyscontrol, or on temporo-limbic instability linked to the non-responsive psychotic. The only double blind study is Neppe's crossover study in South Africa of carbamazepine in the non-epileptic, non-responsive psychotic admitted on the basis of an electroencephalographic temporal lobe abnormality (the target feature) and chronic non-responsive psychosis using 200 mg three times a day. Statistically significant improvements occurred on all three measuring instruments. Most interesting was the specific improvement noted at an interpersonal level. The patients reported being far more in control of themselves. They felt far less hostile or aggressive. Serum levels of carbamazepine suggest a low therapeutic range for seizures, namely, four to ten mg/l, and this still is the only such study with information on serum levels. No other double-blind studies of CBZ in hostility exist and anger ultimately appeared to be an important consistent factor which improved. Subsequent to the original Neppe study, numerous other investigators have found that hostility diminishes considerably. Consequently, carbamazepine's relevance to non-responsive psychosis may be particularly important in controlling hostility.

No benefit of carbamazepine on a relatively classical chronic schizophrenic population occurs. Indeed, possibly due to the pharmacokinetic decrease of neuroleptic levels, the schizophrenic patients, in fact, at times deteriorated. Carbamazepine possibly may be valuable in a specific subgroup of violent patients.

Carbamazepine adjunctive innovative therapy should be considered in non-responsive psychotics with difficulties in self-control, previous histories of drug use of hallucinogens and in patients in whom a kindling-like process could have occurred.

Other anticonvulsants such as phenytoin may at times have roles.

Another group of accessory symptoms to analyze relates to the presence of an affective component, often labelled "schizo-affective illness": patients in this category do not have the affective blunting of schizophrenia nor do they have the exaggeration of mood that would suggest affective illness, per se. In these patients, lithium carbonate or other lithium preparations are logical attempts at innovative therapy.

Lithium may aggravate schizo-affective illness when used with neuroleptics, increasing the risk of side effects, but in other instances, improvements have been reported. Psychomotor acceleration, episodes of acuity, and pathological excitement have been hypothesized to be more responsive to lithium. There appears no adequate way, at present, to differentiate these two opposing results. The mechanism of effect of lithium manic psychosis is as yet unknown and until this is clarified its use in non-affective illness becomes a shot in the dark. Establishing an underlying biochemical framework for a drug is necessary.

In the non-responsive psychotic with affective features, alternatives exist. Co-existent depression can be handled with an appropriate antidepressant, particularly when patients exhibit symptoms of clinical depression other than anergia. The heterocyclic and monoamine oxidase inhibitor groups of antidepressants can induce improvement, deterioration or no real change in condition. At this point only clinical judgment can be exercised as to specific drug choice as there are no special target features.

Another key symptom to be sought in the non-responsive psychotic is extra-pyramidal features. These must be adequately treated. The psychotic patient with even the mildest degree of extra-pyramidal feature may manifest not necessarily as rigidity, tremor or akinesia but as very mild akathisia, some degree of listlessness, or a complaint of mild stiffness. The nonresponsive relevance is a dynamic one: Such symptoms and signs may be interpreted by the psychotic that an external agency is controlling his thinking, affecting his motoric behavior and his impulses. This sequence can exacerbate the psychotic condition. Modification of such features using anticholinergic agents may induce substantial psychodynamic change in the patient: his delusional external agency is no longer controlling him and he may be able to improve. However, anticholinergics have major disadvantages. A reduction in dosage or a shift to a more appropriate neuroleptic may be a more appropriate method of management.

One other aspect of innovative therapy of the withdrawn non-responsive psychotic is controversial: In the absence of response to dopamine antagonists, add dopamine agonists such as the various forms of levodopa (combined for example with bensarimide or carbidopa to prevent the peripheral dopamine agonist effects). There is a literature on the use of levodopa in schizophrenia, and it is interesting that the author has recently described a series of patients with catatonic stupor, who did not respond to neuroleptics, but improved from a catatonic state with levodopa. Following this, they became floridly psychotic and required neuroleptics in a then-uneventful recovery. The use of levodopa is innovative therapy at its most extreme and speculative.

Drug Interactions

The use of combination psychotropic and other drugs is extremely common. Often the potential for pharmacokinetic interactions is neglected. This is particularly so, as almost all are psychotropic agents act other than the beta-adrenergic blocking agents act directly at brain level and are lipid soluble. This in turn implies that they need hepatic drug biotransformation. This generally involves a particular cytochrome, the P-450 cytochrome enzyme system in the liver. The situation is complex indeed. At times, the theoretical hypotheses do not stand up in practice and studies do not demonstrate marked interaction. It is probable that degree of psychotropic interactions varies widely among individuals and is partly genetically determined. Besides hepatic biotransformation, clearly medications may interfere with absorption of the drug. Drugs should be administered at the same time and under similar circumstances each day, for example, with meals. This may have the advantage of diminishing side effects. Alternatively, there may be changes in degree of absorption. Similarly, agents such as antacids and as noted later common beverages such as coffee may substantially effect absorption from the gut.

In addition to the pharmacokinetic interactions, predominantly at the hepatic level, and any limitation pertaining to renal excretion of drugs, a second major consideration when giving drugs in combination is a pharmacodynamic effect. This implies that the new drug is having specific actions in the psychiatric context at the levels of modification of behavior or psychological functions. The drug is therefore demonstrating its psychoactive functions, and these psychoactive functions may be modulated directly at neurotransmitter levels or indirectly by modifying complex or basic biochemical cycles. The addition of one drug to a regimen does not necessarily imply that its effects on neurotransmitters and hormonal or biochemical cycles will be exactly the same as if that drug were given alone. Clearly this is not so, as for example particular receptors may already be blockaded. Consequently, side effects of unexpected kind may occur. A useful principle, therefore, is to prescribe one drug at a time. If the second drug is added, it should be added during a period when the side effects and therapeutic effects of the interaction of the combination regimen can easily be evaluated. Two drugs should not initially be prescribed together, because their individual effects will be more difficult to evaluate. Similarly, not only should one drug change be made at a time, but only single dosage changes should be made at any one time, again allowing better quantification of side effects. In similar fashion, the effects of a drug initially may not be the same as the effects of that drug at a later point, due to changes at kinetic, dynamic and psychological levels.

The role of diet, cigarette smoking, caffeine, and alcohol in affecting these parameters is very significant and should always be considered. Cigarette smoke can induce the P448 and P450 cytochrome systems and alcohol chronically induces and acutely competes at the P450 level. Nicotine in cigarettes has its own added pharmacology as does the stimulant caffeine. Finally, diet may effect absorption, diurnal rhythms, metabolism and have potential pharmacodynamic and neurotransmitter effects - for example, tryptophan as a serotonin precursor induces sleep.


A substantial proportion of patients previously regarded as non-responsive will respond using one or at times a combination of the above psychopharmacological techniques with appropriate psychotherapeutic and environmental interventions.


© Copyright 1997 Pacific Neuropsychiatric Institute.