Reviews

Delirium has been described as a poten­tial complication of concurrent lithium and electroconvulsive therapy (ECT) for depression, in association with a range of serum lithium levels. Although debate persists about the safety of continuing pre­viously established lithium therapy during a course of ECT for mood symptoms, with­holding lithium for 24 hours before adminis­tering ECT and measuring the serum lithium level before ECT were found to decrease the risk of post-ECT neurocognitive effects.¹

We have found that the conven­tional practice of holding lithium for 24 hours before ECT might need to be re-evaluated in geriatric patients, as the fol­lowing case demonstrates. Only 24 hours of holding lithium therapy might result in a lithium level sufficient to contribute to delir­ium after ECT. 

CASE REPORT
An older woman with recurrent unipolar psychotic depression
Mrs. A, age 81, was admitted to the hospital with a 1-week history of depressed mood, anhedonia, insomnia, anergia, anorexia, and nihilistic somatic delusions that her organs were “rotting and shutting down.” Treatment included nortriptyline, 40 mg/d; lithium, 150 mg/d; and haloperidol, 0.5 mg/d. Her serum lithium level was 0.3 mEq/L (reference range, 0.6 to 1.2 mEq/L); the serum nortrip­tyline level was 68 ng/mL (reference range, 50 to 150 ng/mL). CT of the head and an electrocardiogram were unremarkable.

A twice-weekly course of ECT was initiated.

The day before Treatment 1 of ECT, the serum lithium level (drawn 12 hours after the last dose) was 0.4 mEq/L. Lithium was withheld 24 hours before ECT; nortriptyline and haloperidol were continued at prescribed dosages.

Right unilateral stimulation was used at 50%/mC energy (Thymatron DG, with metho­hexital anesthesia, and succinylcholine for mus­cle relaxation). Seizure duration, measured by EEG, was 57 seconds.

Mrs. A developed postictal delirium after the first 2 ECT sessions. The serum lithium level was unchanged. Subsequently, lithium treat­ment was discontinued and ECT was continued; once lithium was stopped, delirium resolved. ECT sessions 3 and 4 were uneventful, with no post-treatment delirium. Seizure duration for Treatment 4 was 58 seconds. She started breath­ing easily after all ECT sessions.

After Treatment 4, Mrs. A experienced full remission of depressive and psychotic symp­toms. Repeat CT of head, after Treatment 4, was unchanged from baseline.

What is the role of lithium?
Mrs. A did not exhibit typical signs of lithium intoxication (diarrhea, vomiting, tremor). Notably, lithium has an intrinsic anticholinergic activity²; concurrent nor­triptyline, a secondary amine tricyclic anti­depressant with fewer anticholinergic side effects than other tricyclics,² could pre­cipitate delirium in a vulnerable patient secondary to excessive cumulative anti­cholinergic exposure.

No prolonged time-to-respiration or time-to-awakening occurred during treat­ments in which concurrent lithium and ECT were used; seizure duration with and without concurrent lithium was rela­tively similar.

There are potential complications of con­current use of lithium and ECT:
    • prolongation of the duration of muscle paralysis and apnea induced by commonly used neuromuscular-blocking agents (eg, succinylcholine)
    • post-ECT cognitive disturbance.^1,3,4

There is debate about the safety of con­tinuing lithium during, or in close proximity to, ECT. In a case series of 12 patients who underwent combined lithium therapy and ECT, the authors concluded that this combi­nation can be safe, regardless of age, as long as appropriate clinical monitoring is pro­vided.⁴ In Mrs. A’s case, once post-ECT delir­ium was noted, lithium was discontinued for subsequent ECT sessions.

Because further ECT was uneventful with­out lithium, and no other clear acute cause of delirium could be identified, we concluded that lithium likely played a role in Mrs. A’s delirium. Notably, nortriptyline had been continued, suggesting that the degree of anticholinergic blockade provided by nortriptyline was insufficient to provoke delirium post-ECT in the absence of potentia­tion of this effect, as it had been when lithium also was used initially.

Guidelines for dosing and serum lithium concentrations in geriatric patients are not well-established; the current traditional range of 0.6 to 1.2 mEq/L, is too high for geriatric patients and can result in epi­sodes of lithium toxicity, including delirium.⁵ Although our patient’s lithium level was below the reference range for all patients, a level of 0.3 mEq/L can be considered at the low end of the reference range for geriatric patients.5 Inasmuch as the lithium-assisted post-ECT delirium could represent a clinical sign of lithium toxicity, perhaps even a sub­therapeutic level in a certain patient could be paradoxically “toxic.”

Although the serum lithium level in our patient remained below the toxic level for the general population (>1.5 mEq/L), delirium in a geriatric patient could result from:  
   • age-related changes in the pharmacokinetics of lithium, a water-soluble drug; these changes reduce renal clearance of the drug and extend plasma elimination half-life of a single dose to 36 hours, with the result that lithium remains in the body longer and necessitating a lower dosage (ie, a dosage that yields a serum level of approximately 0.5 mEq/L)  
    • the CNS tissue concentration of lith­ium, which can be high even though the serum level is not toxic  
    • an age-related increase in blood-brain barrier permeability, making the barrier more porous for drugs  
    • changes in blood-brain barrier perme­ability by post-ECT biochemical induction, with subsequent increased drug availability in the CNS.^5,6 

What we recommend
Possible interactions between lithium and ECT that lead to ECT-associated delirium need further elucidation, but discontinu­ing lithium during the course of ECT in a geriatric patient warrants your consider­ation. Following a safe interval after the last ECT session, lithium likely can be safely re-introduced 1) if there is clinical need and 2) as long as clinical surveillance for cognitive side effects is provided— especially if ECT will need to be reconsidered in the future.

Two additional considerations:
   • Actively reassess lithium dosing in all geriatric psychiatric patients, especially those with renal insufficiency and other systemic metabolic considerations.
   • Actively examine the use of all other anticholinergic agents in the course of evaluating a patient’s candidacy for ECT.


Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Delirium has been described as a poten­tial complication of concurrent lithium and electroconvulsive therapy (ECT) for depression, in association with a range of serum lithium levels. Although debate persists about the safety of continuing pre­viously established lithium therapy during a course of ECT for mood symptoms, with­holding lithium for 24 hours before adminis­tering ECT and measuring the serum lithium level before ECT were found to decrease the risk of post-ECT neurocognitive effects.¹

We have found that the conven­tional practice of holding lithium for 24 hours before ECT might need to be re-evaluated in geriatric patients, as the fol­lowing case demonstrates. Only 24 hours of holding lithium therapy might result in a lithium level sufficient to contribute to delir­ium after ECT. 

CASE REPORT
An older woman with recurrent unipolar psychotic depression
Mrs. A, age 81, was admitted to the hospital with a 1-week history of depressed mood, anhedonia, insomnia, anergia, anorexia, and nihilistic somatic delusions that her organs were “rotting and shutting down.” Treatment included nortriptyline, 40 mg/d; lithium, 150 mg/d; and haloperidol, 0.5 mg/d. Her serum lithium level was 0.3 mEq/L (reference range, 0.6 to 1.2 mEq/L); the serum nortrip­tyline level was 68 ng/mL (reference range, 50 to 150 ng/mL). CT of the head and an electrocardiogram were unremarkable.

A twice-weekly course of ECT was initiated.

The day before Treatment 1 of ECT, the serum lithium level (drawn 12 hours after the last dose) was 0.4 mEq/L. Lithium was withheld 24 hours before ECT; nortriptyline and haloperidol were continued at prescribed dosages.

Right unilateral stimulation was used at 50%/mC energy (Thymatron DG, with metho­hexital anesthesia, and succinylcholine for mus­cle relaxation). Seizure duration, measured by EEG, was 57 seconds.

Mrs. A developed postictal delirium after the first 2 ECT sessions. The serum lithium level was unchanged. Subsequently, lithium treat­ment was discontinued and ECT was continued; once lithium was stopped, delirium resolved. ECT sessions 3 and 4 were uneventful, with no post-treatment delirium. Seizure duration for Treatment 4 was 58 seconds. She started breath­ing easily after all ECT sessions.

After Treatment 4, Mrs. A experienced full remission of depressive and psychotic symp­toms. Repeat CT of head, after Treatment 4, was unchanged from baseline.

What is the role of lithium?
Mrs. A did not exhibit typical signs of lithium intoxication (diarrhea, vomiting, tremor). Notably, lithium has an intrinsic anticholinergic activity²; concurrent nor­triptyline, a secondary amine tricyclic anti­depressant with fewer anticholinergic side effects than other tricyclics,² could pre­cipitate delirium in a vulnerable patient secondary to excessive cumulative anti­cholinergic exposure.

No prolonged time-to-respiration or time-to-awakening occurred during treat­ments in which concurrent lithium and ECT were used; seizure duration with and without concurrent lithium was rela­tively similar.

There are potential complications of con­current use of lithium and ECT:
    • prolongation of the duration of muscle paralysis and apnea induced by commonly used neuromuscular-blocking agents (eg, succinylcholine)
    • post-ECT cognitive disturbance.^1,3,4

There is debate about the safety of con­tinuing lithium during, or in close proximity to, ECT. In a case series of 12 patients who underwent combined lithium therapy and ECT, the authors concluded that this combi­nation can be safe, regardless of age, as long as appropriate clinical monitoring is pro­vided.⁴ In Mrs. A’s case, once post-ECT delir­ium was noted, lithium was discontinued for subsequent ECT sessions.

Because further ECT was uneventful with­out lithium, and no other clear acute cause of delirium could be identified, we concluded that lithium likely played a role in Mrs. A’s delirium. Notably, nortriptyline had been continued, suggesting that the degree of anticholinergic blockade provided by nortriptyline was insufficient to provoke delirium post-ECT in the absence of potentia­tion of this effect, as it had been when lithium also was used initially.

Guidelines for dosing and serum lithium concentrations in geriatric patients are not well-established; the current traditional range of 0.6 to 1.2 mEq/L, is too high for geriatric patients and can result in epi­sodes of lithium toxicity, including delirium.⁵ Although our patient’s lithium level was below the reference range for all patients, a level of 0.3 mEq/L can be considered at the low end of the reference range for geriatric patients.5 Inasmuch as the lithium-assisted post-ECT delirium could represent a clinical sign of lithium toxicity, perhaps even a sub­therapeutic level in a certain patient could be paradoxically “toxic.”

Although the serum lithium level in our patient remained below the toxic level for the general population (>1.5 mEq/L), delirium in a geriatric patient could result from:  
   • age-related changes in the pharmacokinetics of lithium, a water-soluble drug; these changes reduce renal clearance of the drug and extend plasma elimination half-life of a single dose to 36 hours, with the result that lithium remains in the body longer and necessitating a lower dosage (ie, a dosage that yields a serum level of approximately 0.5 mEq/L)  
    • the CNS tissue concentration of lith­ium, which can be high even though the serum level is not toxic  
    • an age-related increase in blood-brain barrier permeability, making the barrier more porous for drugs  
    • changes in blood-brain barrier perme­ability by post-ECT biochemical induction, with subsequent increased drug availability in the CNS.^5,6 

What we recommend
Possible interactions between lithium and ECT that lead to ECT-associated delirium need further elucidation, but discontinu­ing lithium during the course of ECT in a geriatric patient warrants your consider­ation. Following a safe interval after the last ECT session, lithium likely can be safely re-introduced 1) if there is clinical need and 2) as long as clinical surveillance for cognitive side effects is provided— especially if ECT will need to be reconsidered in the future.

Two additional considerations:
   • Actively reassess lithium dosing in all geriatric psychiatric patients, especially those with renal insufficiency and other systemic metabolic considerations.
   • Actively examine the use of all other anticholinergic agents in the course of evaluating a patient’s candidacy for ECT.


Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Delirium has been described as a poten­tial complication of concurrent lithium and electroconvulsive therapy (ECT) for depression, in association with a range of serum lithium levels. Although debate persists about the safety of continuing pre­viously established lithium therapy during a course of ECT for mood symptoms, with­holding lithium for 24 hours before adminis­tering ECT and measuring the serum lithium level before ECT were found to decrease the risk of post-ECT neurocognitive effects.¹

We have found that the conven­tional practice of holding lithium for 24 hours before ECT might need to be re-evaluated in geriatric patients, as the fol­lowing case demonstrates. Only 24 hours of holding lithium therapy might result in a lithium level sufficient to contribute to delir­ium after ECT. 

CASE REPORT
An older woman with recurrent unipolar psychotic depression
Mrs. A, age 81, was admitted to the hospital with a 1-week history of depressed mood, anhedonia, insomnia, anergia, anorexia, and nihilistic somatic delusions that her organs were “rotting and shutting down.” Treatment included nortriptyline, 40 mg/d; lithium, 150 mg/d; and haloperidol, 0.5 mg/d. Her serum lithium level was 0.3 mEq/L (reference range, 0.6 to 1.2 mEq/L); the serum nortrip­tyline level was 68 ng/mL (reference range, 50 to 150 ng/mL). CT of the head and an electrocardiogram were unremarkable.

A twice-weekly course of ECT was initiated.

The day before Treatment 1 of ECT, the serum lithium level (drawn 12 hours after the last dose) was 0.4 mEq/L. Lithium was withheld 24 hours before ECT; nortriptyline and haloperidol were continued at prescribed dosages.

Right unilateral stimulation was used at 50%/mC energy (Thymatron DG, with metho­hexital anesthesia, and succinylcholine for mus­cle relaxation). Seizure duration, measured by EEG, was 57 seconds.

Mrs. A developed postictal delirium after the first 2 ECT sessions. The serum lithium level was unchanged. Subsequently, lithium treat­ment was discontinued and ECT was continued; once lithium was stopped, delirium resolved. ECT sessions 3 and 4 were uneventful, with no post-treatment delirium. Seizure duration for Treatment 4 was 58 seconds. She started breath­ing easily after all ECT sessions.

After Treatment 4, Mrs. A experienced full remission of depressive and psychotic symp­toms. Repeat CT of head, after Treatment 4, was unchanged from baseline.

What is the role of lithium?
Mrs. A did not exhibit typical signs of lithium intoxication (diarrhea, vomiting, tremor). Notably, lithium has an intrinsic anticholinergic activity²; concurrent nor­triptyline, a secondary amine tricyclic anti­depressant with fewer anticholinergic side effects than other tricyclics,² could pre­cipitate delirium in a vulnerable patient secondary to excessive cumulative anti­cholinergic exposure.

No prolonged time-to-respiration or time-to-awakening occurred during treat­ments in which concurrent lithium and ECT were used; seizure duration with and without concurrent lithium was rela­tively similar.

There are potential complications of con­current use of lithium and ECT:
    • prolongation of the duration of muscle paralysis and apnea induced by commonly used neuromuscular-blocking agents (eg, succinylcholine)
    • post-ECT cognitive disturbance.^1,3,4

There is debate about the safety of con­tinuing lithium during, or in close proximity to, ECT. In a case series of 12 patients who underwent combined lithium therapy and ECT, the authors concluded that this combi­nation can be safe, regardless of age, as long as appropriate clinical monitoring is pro­vided.⁴ In Mrs. A’s case, once post-ECT delir­ium was noted, lithium was discontinued for subsequent ECT sessions.

Because further ECT was uneventful with­out lithium, and no other clear acute cause of delirium could be identified, we concluded that lithium likely played a role in Mrs. A’s delirium. Notably, nortriptyline had been continued, suggesting that the degree of anticholinergic blockade provided by nortriptyline was insufficient to provoke delirium post-ECT in the absence of potentia­tion of this effect, as it had been when lithium also was used initially.

Guidelines for dosing and serum lithium concentrations in geriatric patients are not well-established; the current traditional range of 0.6 to 1.2 mEq/L, is too high for geriatric patients and can result in epi­sodes of lithium toxicity, including delirium.⁵ Although our patient’s lithium level was below the reference range for all patients, a level of 0.3 mEq/L can be considered at the low end of the reference range for geriatric patients.5 Inasmuch as the lithium-assisted post-ECT delirium could represent a clinical sign of lithium toxicity, perhaps even a sub­therapeutic level in a certain patient could be paradoxically “toxic.”

Although the serum lithium level in our patient remained below the toxic level for the general population (>1.5 mEq/L), delirium in a geriatric patient could result from:  
   • age-related changes in the pharmacokinetics of lithium, a water-soluble drug; these changes reduce renal clearance of the drug and extend plasma elimination half-life of a single dose to 36 hours, with the result that lithium remains in the body longer and necessitating a lower dosage (ie, a dosage that yields a serum level of approximately 0.5 mEq/L)  
    • the CNS tissue concentration of lith­ium, which can be high even though the serum level is not toxic  
    • an age-related increase in blood-brain barrier permeability, making the barrier more porous for drugs  
    • changes in blood-brain barrier perme­ability by post-ECT biochemical induction, with subsequent increased drug availability in the CNS.^5,6 

What we recommend
Possible interactions between lithium and ECT that lead to ECT-associated delirium need further elucidation, but discontinu­ing lithium during the course of ECT in a geriatric patient warrants your consider­ation. Following a safe interval after the last ECT session, lithium likely can be safely re-introduced 1) if there is clinical need and 2) as long as clinical surveillance for cognitive side effects is provided— especially if ECT will need to be reconsidered in the future.

Two additional considerations:
   • Actively reassess lithium dosing in all geriatric psychiatric patients, especially those with renal insufficiency and other systemic metabolic considerations.
   • Actively examine the use of all other anticholinergic agents in the course of evaluating a patient’s candidacy for ECT.


Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Myelodysplastic syndromes (MDS) are a heterogeneous group of hematologic neoplasms with an annual incidence of 4.1 cases per 100,000 Americans. Patients with MDS suffer from chronic cytopenias that may lead to recurrent transfusions, infections, and increased risk for bleeding. They are also at risk for progression to acute myeloid leukemia. Allogeneic hematopoietic cell transplantation is the only potentially curative treatment for MDS, although 3 drugs have been approved by the US Food and Drug Administration for its treatment: lenalidomide, 5-azacitidine, and decitabine.
 

Click on the PDF icon at the top of this introduction to read the full article.

Myelodysplastic syndromes (MDS) are a heterogeneous group of hematologic neoplasms with an annual incidence of 4.1 cases per 100,000 Americans. Patients with MDS suffer from chronic cytopenias that may lead to recurrent transfusions, infections, and increased risk for bleeding. They are also at risk for progression to acute myeloid leukemia. Allogeneic hematopoietic cell transplantation is the only potentially curative treatment for MDS, although 3 drugs have been approved by the US Food and Drug Administration for its treatment: lenalidomide, 5-azacitidine, and decitabine.
 

Click on the PDF icon at the top of this introduction to read the full article.

Myelodysplastic syndromes (MDS) are a heterogeneous group of hematologic neoplasms with an annual incidence of 4.1 cases per 100,000 Americans. Patients with MDS suffer from chronic cytopenias that may lead to recurrent transfusions, infections, and increased risk for bleeding. They are also at risk for progression to acute myeloid leukemia. Allogeneic hematopoietic cell transplantation is the only potentially curative treatment for MDS, although 3 drugs have been approved by the US Food and Drug Administration for its treatment: lenalidomide, 5-azacitidine, and decitabine.
 

Click on the PDF icon at the top of this introduction to read the full article.

Although delirium has many descriptive terms (Table 1), a common unifying term is “acute global cognitive dysfunction,” now recognized as delirium; a consensus supported by DSM-5¹ and ICD-10² (Table 2). According to DSM-5, the essential feature is a disturbance of attention or awareness that is accompanied by a change in baseline cognition that cannot be explained by another preexisting, established, or evolving neurocognitive disorder (the newly named DSM-5 entity for dementia syndromes).¹ Because delirium affects the cortex diffusely, psychiatric symp­toms can include cognitive, mood, anxiety, or psychotic symp­toms. Because many systemic illnesses can induce delirium, the differential diagnosis spans all organ systems.

Three subtypes
Delirium can be classified, based on symptoms,^3,4 into 3 sub­types: hyperactive-hyperalert, hypoactive-hypoalert, and mixed delirium. Hyperactive patients present with rest­lessness and agitation. Hypoactive patients are lethargic, confused, slow to respond to questions, and often appear depressed. The differential prognostic significance of these subtypes has been examined in the literature, with conflicting results. Rabinowitz⁵ reported that hypoactive delirium has the worst prognosis, while Marcantonio et al⁶ indicated that the hyperactive subtype is associated with the highest mortality rate. Mixed delirium, with periods of both hyperactivity and hypoactivity, is the most common type of delirium.⁷

A prodromal phase, characterized by anxiety, frequent requests for nursing and medical assistance, decreased attention, rest­lessness, vivid dreams, disorientation imme­diately after awakening, and hallucinations, can occur before an episode of full-spectrum delirium; this prodromal state often is iden­tified retrospectively —after the patient is in an episode of delirium.^8,9

Evidence-based guidelines aim to improve recognition and clinical management.^10-13 Disruptive behavior is the main reason for psychiatric referral in delirium.^14,15 Delayed psychiatric consultation because of non-recognition of delirium is related to variables such as older age; history of a pre-existing, comorbid neurocognitive disorder; and the clinical appearance of hypoactive delirium.¹⁴

The case of Mr. D (Box),¹⁶ illus­trates how the emergence of antipsychotic-associated neuroleptic malignant syndrome (NMS) can complicate antipsychotic treat­ment of delirium in a geriatric medical patient, although delirium also is a common presentation in NMS.¹⁷ Delirium developed after an increase in carbidopa/levodopa, which has central dopaminergic effects that can precipitate delirium, particularly in a geriatric patient with preexisting comorbid neurocognitive disorder. Further complicat­ing Mr. D’s delirium presentation was the development of NMS, which had a multifac­torial causation, such as the use of dopamine antagonists (ie, quetiapine, metoclopramide), and an abrupt decrease of a dopaminergic agent (ie, carbidopa/levodopa), all inducing a central dopamine relative hypoactivity.

Epidemiology
Delirium is more common in older patients,¹⁵ and is seen in 30% to 40% of hospitalized geri­atric patients.¹⁸ Delirium in older patients, compared with other adults, is associated with more severe cognitive impairment.¹⁹ It is com­mon among geriatric surgical patients (15% to 62%)²⁰ with a peak 2 to 5 days postoperatively for hip fracture,²¹ and often is seen in ICU patients (70% to 87%).²⁰ However, Spronk et al²² found that delirium is significantly under-recognized in the ICU. Nearly 90% of terminally ill patients become delirious before death.²³ Terminal delirium often is unrecognized and can interfere with assessment of other clinical problems.²⁴ A preexisting history of comor­bid neurocognitive disorder was evident in as many as two-thirds of delirium cases.²⁵

Pathophysiology and risk factors
The pathophysiology of delirium has been characterized as an imbalance of CNS metab­olism, including decreased blood flow in vari­ous regions of the brain that may normalize once delirium resolves.²⁶ Studies describe the simultaneous decrease of cholinergic transmission and dopaminergic excess.^27,28 Predisposing and precipitating factors for delirium that are of particular importance in geriatric patients include:
   • advanced age
   • CNS disease
   • infection
   • cognitive impairment
   • male sex
   • poor nutrition
   • dehydration and other metabolic abnormalities
   • cardiovascular events
   • substance use
   • medication
   • sensory deprivation (eg, impaired vision or hearing)
   • sleep deprivation
   • low level of physical activity.^27,29,30

Table 3 lists the most common delirium-provocative medications.²⁷

Evaluation and psychometric scales
The EEG can be useful in evaluating delir­ium, especially in clinically ambiguous cases. EEG findings may indicate generalized slowing or dropout of the posterior domi­nant rhythm, and generalized slow theta and delta waves, findings that are more common in delirium than in other neurocognitive dis­orders and other psychiatric illnesses. The EEG must be interpreted in the context of the delirium diagnostic workup, because abnor­malities seen in other neurocognitive disor­ders can overlap with those of delirium.³¹

The EEG referral should specify the clini­cal suspicion of delirium to help interpret the results. Delirium cases in which the patient’s previous cognitive status is unknown may benefit from EEG evaluation, such as:
   • in possible status epilepticus
   • when delirium improvement has reached a plateau at a lower level of cognitive function than before onset of delirium
   • when the patient is unable or unwilling to complete a psychiatric interview.²⁷

Assessment instruments are available to diagnose and monitor delirium (Table 4). Typically, delirium assessment includes examining levels of arousal, psychomotor activity, cognition (ie, orienta­tion, attention, and memory), and percep­tual disturbances.

Psychometrically, a review of Table 4 suggests that validity appeared stable with adequate specificity (64% to 99%) but more variable sensitivity (36% to 100%). These reliability parameters also will be affected by the classification sys­tem (ie, DSM vs ICD) and the cut-off score employed.³² Most measures (eg, Confusion Assessment Method [CAM], CAM-ICU) provide an adequate sample of behavioral (ie, level of alertness), motor (ie, psychomo­tor activity), and cognitive (ie, orientation, attention, memory, and receptive language) function, with the exception of the Global Attentiveness Rating, which is a 2-minute open conversation protocol between physi­cian and patient.

Some measures are stand-alone instru­ments, such as the Memorial Delirium Assessment Scale, whereas the CAM requires administration of separate cogni­tive screens, including the Mini-Mental State Examination (MMSE) and Digit Span.³³ Instruments to detect delirium in critically ill patients are a more recent development. Wong et al³⁴ reported that the most widely studied tool was the CAM. Obtaining collat­eral information from family, caregivers, and hospital staff is essential, particularly given the fluctuating nature of delirium.

Management
Prevention. Identify patients at high risk of delirium so that preventive strategies can be employed. Multi-component, nonphar­macotherapeutic interventions are used in clinical settings but few randomized trials have been conducted. The contributing effectiveness of individual components is not well-studied, but most include staff education to increase awareness of delir­ium. Of 3 multi-component intervention randomized trials, 2 reported a signifi­cantly lower incidence of delirium in the intervention group.^35-37 Implementation of a multi-component protocol in medical/ surgical units was associated with a sig­nificant reduction in use of restraints.³⁸

As in Mr. D’s case, complex drug regimens, particularly for CNS illness, can increase the risk of delirium. Considering the medication profile for patients with complex systemic illness—in particular, minimizing the use anticholinergics and dopamine agonists— may be crucial in preventing delirium.

Prophylactic administration of antipsychotics may reduce the risk of devel­oping postoperative delirium.³⁹ Studies of the use of these agents were characterized by small sample sizes and selected groups of patient populations. Of the 4 random­ized studies evaluating prophylactic anti­psychotics (vs placebo), 3 found a lower incidence of delirium in the intervention groups.^39-41

A study of haloperidol in post-GI sur­gery patients showed a reduced occurrence of delirium,⁴⁰ whereas its prophylactic use in patients undergoing hip surgery⁴² did not reduce the incidence of delirium com­pared with placebo, but did decrease sever­ity when delirium occurred.⁴²

Risperidone³⁹ in post-cardiac sur­gery and olanzapine⁴¹ perioperatively in patients undergoing total knee or hip replacement have been shown to decrease delirium severity and duration. Targeted prophylaxis with risperidone⁴³ in post-cardiac surgery patients who showed disturbed cognition but did not meet cri­teria for delirium reduced the number of patients requiring medication, compared with placebo.⁴³

Dexmedetomidine, an α-2 adrenergic receptor agonist, compared with propofol or midazolam in post-cardiac valve surgery patients, resulted in a decreased incidence of delirium but no difference in delirium duration, hospital length of stay, or use of other medications.⁴⁴ However, other studies have shown that dexmedetomidine reduces ICU length of stay and duration of mechani­cal ventilation.⁴⁵

Treatment. Management of hospitalized medically ill geriatric patients with delirium is challenging and requires a comprehensive approach. The first step in delirium man­agement is prompt identification and man­agement of systemic medical disturbances associated with the delirium episode. First-line, nonpharmacotherapeutic strategies for patients with delirium include:
   • reorientation
   • behavioral interventions (eg, use of clear instructions and frequent eye con­tact with patients)
   • environmental interventions (eg, mini­mal noise, adequate lighting, and lim­ited room and staff changes)
   • avoidance of physical restraints.⁴⁶

Consider employing family members or hospital staff sitters to stay with the patient and to reassure, reorient, and watch for agitation and other unsafe behaviors (eg, attempted elopement). Psychoeducation for the patient and family on the phenomenol­ogy of delirium can be helpful.

The use of drug treatment strategies should be integrated into a comprehensive approach that includes the routine use of nondrug measures.⁴⁶ Using medications for treating hypoactive delirium, formerly con­troversial, now has wider acceptance.^47,48 A few high-quality randomized trials have been performed.^25,49,50

Pharmacotherapy, especially in frail patients, should be initiated at the lowest start­ing dosage and titrated cautiously to clinical effect and for the shortest period of time nec­essary. Antipsychotics are preferred agents for treating all subtypes of delirium; haloperidol is widely used.^46,51,52 However, antipsychotics, including haloperidol, can be associated with adverse neurologic effects such as extrapyra­midal symptoms (EPS) and NMS.

Although reported less frequently than with haloperidol, other agents have been implicated in development of EPS and NMS, including atypical antipsychotics and anti­emetic dopamine antagonists, particularly in parkinsonism-prone patients.⁵³ Strategies that can minimize such risks in geriatric inpatients with delirium include oral, rather than par­enteral, use of antipsychotics—preferential use of atypical over typical antipsychotics— and lowest effective dosages.⁵⁴

In controlled trials, atypical antipsychot­ics for delirium showed efficacy compared with haloperidol.^52,55 However, there is no research that demonstrates any advantage of one atypical over another.²⁵

In Mr. D’s case, the most important inter­vention for managing delirium caused by NMS is to discontinue all dopamine antag­onists and treat agitation with judicious doses of a benzodiazepine, with supportive care.¹⁷ In cases of sudden discontinuation or a dosage decrease of dopamine agonists, these medications should be resumed or optimized to minimize the risk of NMS-associated rhabdomyolysis and subse­quent renal failure.¹⁷ Antipsychotics carry an increased risk of stroke and mortality in older patients with established or evolving neurocognitive disorders^56,57 and can cause prolongation of the QTc interval.⁵⁷

Other medications that could be used for delirium include cholinesterase inhibitors^58,59 (although larger trials and a systematic review did not support this use⁶⁰), and 5-HT receptor antagonists,⁶¹ such as trazodone. Benzodiazepines, such as lorazepam, are first-line treatment for delirium associated with seizures or withdrawal from alcohol, sedatives, hypnotics, and anxiolytics and for delirium caused by NMS. Be cautious about using benzodiazepines in geriatric patients because of a risk of respiratory depression, falls, sedation, and amnesia.

Geriatric patients with alcoholism and those with malnutrition are prone to thia­mine and vitamin B₁₂ deficiencies, which can induce delirium. Laboratory assessment and consideration of supplementation is recom­mended. Despite high occurrence of delirium in hospitalized older adults with preexisting comorbid neurocognitive disorders, there is no standard care for delirium comorbid with another neurocognitive disorder.⁶² Clinical practice guidelines for older patients receiv­ing palliative care have been developed⁶³; the goal is to minimize suffering and discomfort in patients in palliative care.⁶⁴

Post-delirium prophylaxis. Medications for delirium usually can be tapered and discontinued once the episode has resolved and the patient is stable; it is common to discontinue medications when the patient has been symptom-free for 1 week.⁶⁵ Some patients (eg, with end-stage liver disease, disseminated cancer) are prone to recur­rent or to prolonged or chronic delirium. A period of post-recovery treatment with antipsychotics—even indefinite treatment in some cases—should be considered.

Post-delirium debriefing and aftercare. The psychological complications of delirium are distressing for the patient and his (her) caregivers. Psychiatric complications asso­ciated with delirium, including acute stress disorder—which might predict posttraumatic stress disorder—have been explored; early recognition and treatment may improve long-term outcomes.⁶⁶ After recovery from acute delirium, cognitive assessment (eg, MMSE⁶⁷ or Montreal Cognitive Assessment⁶⁸) is recommended to validate current cognitive status because patients may have persistent decrement in cognitive func­tion compared with pre-delirium condition, even after recovery from the acute episode.

Post-delirium debriefing may help patients who have recovered from a delirium episode. Patients may fear that their brief period of hallucinations might represent the onset of a chronic-relapsing psychotic dis­order. Allow patients to communicate their distress about the delirium episode and give them the opportunity to talk through the experience. Brief them on the possibility that delirium will recur and advise them to seek emergency medical care in case of recur­rence. Advise patients to monitor and main­tain a normal sleep-wake cycle.

Family members can watch for syndro­mal recurrence of delirium. They should be encouraged to discuss their reaction to hav­ing seen their relative in a delirious state.

Health care systems with integrated electronic medical records should list “delirium, resolved” on the patient’s illness profile or problem list and alert the patient’s primary care provider to the delirium his­tory to avoid future exposure to delirium-provocative medications, and to prompt the provider to assume an active role in post-delirium care, including delirium recur­rence surveillance, medication adjustment, risk factor management, and post-recovery cognitive assessment.

Bottom Line
Evaluation of delirium in geriatric patients includes clinical vigilance and screening, differentiating delirium from other neurocognitive disorders, and identifying and treating underlying causes. Perioperative use of antipsychotics may reduce the incidence of delirium, although hospital length of stay generally has not been reduced with prophylaxis. Management interventions include staff education, systematic screening, use of multicomponent interventions, and pharmacologic interventions.

Related Resources
• Downing LJ, Caprio TV, Lyness JM. Geriatric psychiatry review: differential diagnosis and treatment of the 3 D’s - delirium, dementia, and depression. Curr Psychiatry Rep. 2013;15(6):365.
• Brooks PB. Postoperative delirium in elderly patients. Am J Nurs. 2012;112(9):38-49.

Drug Brand Names
Carbidopa/levodopa • Sinemet                       Midazolam •  Versed
Dexmedetomidine • Precedex                        Olanzapine •  Zyprexa
Haloperidol • Haldol                                      Propofol  •  Diprivan
Lithium • Eskalith, Lithobid                            Quetiapine  •  Seroquel
Lorazepam • Ativan                                      Risperidone  •  Risperdal
Metoclopramide •  Reglan                              Trazodone  •  Desyrel

Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Although delirium has many descriptive terms (Table 1), a common unifying term is “acute global cognitive dysfunction,” now recognized as delirium; a consensus supported by DSM-5¹ and ICD-10² (Table 2). According to DSM-5, the essential feature is a disturbance of attention or awareness that is accompanied by a change in baseline cognition that cannot be explained by another preexisting, established, or evolving neurocognitive disorder (the newly named DSM-5 entity for dementia syndromes).¹ Because delirium affects the cortex diffusely, psychiatric symp­toms can include cognitive, mood, anxiety, or psychotic symp­toms. Because many systemic illnesses can induce delirium, the differential diagnosis spans all organ systems.

Three subtypes
Delirium can be classified, based on symptoms,^3,4 into 3 sub­types: hyperactive-hyperalert, hypoactive-hypoalert, and mixed delirium. Hyperactive patients present with rest­lessness and agitation. Hypoactive patients are lethargic, confused, slow to respond to questions, and often appear depressed. The differential prognostic significance of these subtypes has been examined in the literature, with conflicting results. Rabinowitz⁵ reported that hypoactive delirium has the worst prognosis, while Marcantonio et al⁶ indicated that the hyperactive subtype is associated with the highest mortality rate. Mixed delirium, with periods of both hyperactivity and hypoactivity, is the most common type of delirium.⁷

A prodromal phase, characterized by anxiety, frequent requests for nursing and medical assistance, decreased attention, rest­lessness, vivid dreams, disorientation imme­diately after awakening, and hallucinations, can occur before an episode of full-spectrum delirium; this prodromal state often is iden­tified retrospectively —after the patient is in an episode of delirium.^8,9

Evidence-based guidelines aim to improve recognition and clinical management.^10-13 Disruptive behavior is the main reason for psychiatric referral in delirium.^14,15 Delayed psychiatric consultation because of non-recognition of delirium is related to variables such as older age; history of a pre-existing, comorbid neurocognitive disorder; and the clinical appearance of hypoactive delirium.¹⁴

The case of Mr. D (Box),¹⁶ illus­trates how the emergence of antipsychotic-associated neuroleptic malignant syndrome (NMS) can complicate antipsychotic treat­ment of delirium in a geriatric medical patient, although delirium also is a common presentation in NMS.¹⁷ Delirium developed after an increase in carbidopa/levodopa, which has central dopaminergic effects that can precipitate delirium, particularly in a geriatric patient with preexisting comorbid neurocognitive disorder. Further complicat­ing Mr. D’s delirium presentation was the development of NMS, which had a multifac­torial causation, such as the use of dopamine antagonists (ie, quetiapine, metoclopramide), and an abrupt decrease of a dopaminergic agent (ie, carbidopa/levodopa), all inducing a central dopamine relative hypoactivity.

Epidemiology
Delirium is more common in older patients,¹⁵ and is seen in 30% to 40% of hospitalized geri­atric patients.¹⁸ Delirium in older patients, compared with other adults, is associated with more severe cognitive impairment.¹⁹ It is com­mon among geriatric surgical patients (15% to 62%)²⁰ with a peak 2 to 5 days postoperatively for hip fracture,²¹ and often is seen in ICU patients (70% to 87%).²⁰ However, Spronk et al²² found that delirium is significantly under-recognized in the ICU. Nearly 90% of terminally ill patients become delirious before death.²³ Terminal delirium often is unrecognized and can interfere with assessment of other clinical problems.²⁴ A preexisting history of comor­bid neurocognitive disorder was evident in as many as two-thirds of delirium cases.²⁵

Pathophysiology and risk factors
The pathophysiology of delirium has been characterized as an imbalance of CNS metab­olism, including decreased blood flow in vari­ous regions of the brain that may normalize once delirium resolves.²⁶ Studies describe the simultaneous decrease of cholinergic transmission and dopaminergic excess.^27,28 Predisposing and precipitating factors for delirium that are of particular importance in geriatric patients include:
   • advanced age
   • CNS disease
   • infection
   • cognitive impairment
   • male sex
   • poor nutrition
   • dehydration and other metabolic abnormalities
   • cardiovascular events
   • substance use
   • medication
   • sensory deprivation (eg, impaired vision or hearing)
   • sleep deprivation
   • low level of physical activity.^27,29,30

Table 3 lists the most common delirium-provocative medications.²⁷

Evaluation and psychometric scales
The EEG can be useful in evaluating delir­ium, especially in clinically ambiguous cases. EEG findings may indicate generalized slowing or dropout of the posterior domi­nant rhythm, and generalized slow theta and delta waves, findings that are more common in delirium than in other neurocognitive dis­orders and other psychiatric illnesses. The EEG must be interpreted in the context of the delirium diagnostic workup, because abnor­malities seen in other neurocognitive disor­ders can overlap with those of delirium.³¹

The EEG referral should specify the clini­cal suspicion of delirium to help interpret the results. Delirium cases in which the patient’s previous cognitive status is unknown may benefit from EEG evaluation, such as:
   • in possible status epilepticus
   • when delirium improvement has reached a plateau at a lower level of cognitive function than before onset of delirium
   • when the patient is unable or unwilling to complete a psychiatric interview.²⁷

Assessment instruments are available to diagnose and monitor delirium (Table 4). Typically, delirium assessment includes examining levels of arousal, psychomotor activity, cognition (ie, orienta­tion, attention, and memory), and percep­tual disturbances.

Psychometrically, a review of Table 4 suggests that validity appeared stable with adequate specificity (64% to 99%) but more variable sensitivity (36% to 100%). These reliability parameters also will be affected by the classification sys­tem (ie, DSM vs ICD) and the cut-off score employed.³² Most measures (eg, Confusion Assessment Method [CAM], CAM-ICU) provide an adequate sample of behavioral (ie, level of alertness), motor (ie, psychomo­tor activity), and cognitive (ie, orientation, attention, memory, and receptive language) function, with the exception of the Global Attentiveness Rating, which is a 2-minute open conversation protocol between physi­cian and patient.

Some measures are stand-alone instru­ments, such as the Memorial Delirium Assessment Scale, whereas the CAM requires administration of separate cogni­tive screens, including the Mini-Mental State Examination (MMSE) and Digit Span.³³ Instruments to detect delirium in critically ill patients are a more recent development. Wong et al³⁴ reported that the most widely studied tool was the CAM. Obtaining collat­eral information from family, caregivers, and hospital staff is essential, particularly given the fluctuating nature of delirium.

Management
Prevention. Identify patients at high risk of delirium so that preventive strategies can be employed. Multi-component, nonphar­macotherapeutic interventions are used in clinical settings but few randomized trials have been conducted. The contributing effectiveness of individual components is not well-studied, but most include staff education to increase awareness of delir­ium. Of 3 multi-component intervention randomized trials, 2 reported a signifi­cantly lower incidence of delirium in the intervention group.^35-37 Implementation of a multi-component protocol in medical/ surgical units was associated with a sig­nificant reduction in use of restraints.³⁸

As in Mr. D’s case, complex drug regimens, particularly for CNS illness, can increase the risk of delirium. Considering the medication profile for patients with complex systemic illness—in particular, minimizing the use anticholinergics and dopamine agonists— may be crucial in preventing delirium.

Prophylactic administration of antipsychotics may reduce the risk of devel­oping postoperative delirium.³⁹ Studies of the use of these agents were characterized by small sample sizes and selected groups of patient populations. Of the 4 random­ized studies evaluating prophylactic anti­psychotics (vs placebo), 3 found a lower incidence of delirium in the intervention groups.^39-41

A study of haloperidol in post-GI sur­gery patients showed a reduced occurrence of delirium,⁴⁰ whereas its prophylactic use in patients undergoing hip surgery⁴² did not reduce the incidence of delirium com­pared with placebo, but did decrease sever­ity when delirium occurred.⁴²

Risperidone³⁹ in post-cardiac sur­gery and olanzapine⁴¹ perioperatively in patients undergoing total knee or hip replacement have been shown to decrease delirium severity and duration. Targeted prophylaxis with risperidone⁴³ in post-cardiac surgery patients who showed disturbed cognition but did not meet cri­teria for delirium reduced the number of patients requiring medication, compared with placebo.⁴³

Dexmedetomidine, an α-2 adrenergic receptor agonist, compared with propofol or midazolam in post-cardiac valve surgery patients, resulted in a decreased incidence of delirium but no difference in delirium duration, hospital length of stay, or use of other medications.⁴⁴ However, other studies have shown that dexmedetomidine reduces ICU length of stay and duration of mechani­cal ventilation.⁴⁵

Treatment. Management of hospitalized medically ill geriatric patients with delirium is challenging and requires a comprehensive approach. The first step in delirium man­agement is prompt identification and man­agement of systemic medical disturbances associated with the delirium episode. First-line, nonpharmacotherapeutic strategies for patients with delirium include:
   • reorientation
   • behavioral interventions (eg, use of clear instructions and frequent eye con­tact with patients)
   • environmental interventions (eg, mini­mal noise, adequate lighting, and lim­ited room and staff changes)
   • avoidance of physical restraints.⁴⁶

Consider employing family members or hospital staff sitters to stay with the patient and to reassure, reorient, and watch for agitation and other unsafe behaviors (eg, attempted elopement). Psychoeducation for the patient and family on the phenomenol­ogy of delirium can be helpful.

The use of drug treatment strategies should be integrated into a comprehensive approach that includes the routine use of nondrug measures.⁴⁶ Using medications for treating hypoactive delirium, formerly con­troversial, now has wider acceptance.^47,48 A few high-quality randomized trials have been performed.^25,49,50

Pharmacotherapy, especially in frail patients, should be initiated at the lowest start­ing dosage and titrated cautiously to clinical effect and for the shortest period of time nec­essary. Antipsychotics are preferred agents for treating all subtypes of delirium; haloperidol is widely used.^46,51,52 However, antipsychotics, including haloperidol, can be associated with adverse neurologic effects such as extrapyra­midal symptoms (EPS) and NMS.

Although reported less frequently than with haloperidol, other agents have been implicated in development of EPS and NMS, including atypical antipsychotics and anti­emetic dopamine antagonists, particularly in parkinsonism-prone patients.⁵³ Strategies that can minimize such risks in geriatric inpatients with delirium include oral, rather than par­enteral, use of antipsychotics—preferential use of atypical over typical antipsychotics— and lowest effective dosages.⁵⁴

In controlled trials, atypical antipsychot­ics for delirium showed efficacy compared with haloperidol.^52,55 However, there is no research that demonstrates any advantage of one atypical over another.²⁵

In Mr. D’s case, the most important inter­vention for managing delirium caused by NMS is to discontinue all dopamine antag­onists and treat agitation with judicious doses of a benzodiazepine, with supportive care.¹⁷ In cases of sudden discontinuation or a dosage decrease of dopamine agonists, these medications should be resumed or optimized to minimize the risk of NMS-associated rhabdomyolysis and subse­quent renal failure.¹⁷ Antipsychotics carry an increased risk of stroke and mortality in older patients with established or evolving neurocognitive disorders^56,57 and can cause prolongation of the QTc interval.⁵⁷

Other medications that could be used for delirium include cholinesterase inhibitors^58,59 (although larger trials and a systematic review did not support this use⁶⁰), and 5-HT receptor antagonists,⁶¹ such as trazodone. Benzodiazepines, such as lorazepam, are first-line treatment for delirium associated with seizures or withdrawal from alcohol, sedatives, hypnotics, and anxiolytics and for delirium caused by NMS. Be cautious about using benzodiazepines in geriatric patients because of a risk of respiratory depression, falls, sedation, and amnesia.

Geriatric patients with alcoholism and those with malnutrition are prone to thia­mine and vitamin B₁₂ deficiencies, which can induce delirium. Laboratory assessment and consideration of supplementation is recom­mended. Despite high occurrence of delirium in hospitalized older adults with preexisting comorbid neurocognitive disorders, there is no standard care for delirium comorbid with another neurocognitive disorder.⁶² Clinical practice guidelines for older patients receiv­ing palliative care have been developed⁶³; the goal is to minimize suffering and discomfort in patients in palliative care.⁶⁴

Post-delirium prophylaxis. Medications for delirium usually can be tapered and discontinued once the episode has resolved and the patient is stable; it is common to discontinue medications when the patient has been symptom-free for 1 week.⁶⁵ Some patients (eg, with end-stage liver disease, disseminated cancer) are prone to recur­rent or to prolonged or chronic delirium. A period of post-recovery treatment with antipsychotics—even indefinite treatment in some cases—should be considered.

Post-delirium debriefing and aftercare. The psychological complications of delirium are distressing for the patient and his (her) caregivers. Psychiatric complications asso­ciated with delirium, including acute stress disorder—which might predict posttraumatic stress disorder—have been explored; early recognition and treatment may improve long-term outcomes.⁶⁶ After recovery from acute delirium, cognitive assessment (eg, MMSE⁶⁷ or Montreal Cognitive Assessment⁶⁸) is recommended to validate current cognitive status because patients may have persistent decrement in cognitive func­tion compared with pre-delirium condition, even after recovery from the acute episode.

Post-delirium debriefing may help patients who have recovered from a delirium episode. Patients may fear that their brief period of hallucinations might represent the onset of a chronic-relapsing psychotic dis­order. Allow patients to communicate their distress about the delirium episode and give them the opportunity to talk through the experience. Brief them on the possibility that delirium will recur and advise them to seek emergency medical care in case of recur­rence. Advise patients to monitor and main­tain a normal sleep-wake cycle.

Family members can watch for syndro­mal recurrence of delirium. They should be encouraged to discuss their reaction to hav­ing seen their relative in a delirious state.

Health care systems with integrated electronic medical records should list “delirium, resolved” on the patient’s illness profile or problem list and alert the patient’s primary care provider to the delirium his­tory to avoid future exposure to delirium-provocative medications, and to prompt the provider to assume an active role in post-delirium care, including delirium recur­rence surveillance, medication adjustment, risk factor management, and post-recovery cognitive assessment.

Bottom Line
Evaluation of delirium in geriatric patients includes clinical vigilance and screening, differentiating delirium from other neurocognitive disorders, and identifying and treating underlying causes. Perioperative use of antipsychotics may reduce the incidence of delirium, although hospital length of stay generally has not been reduced with prophylaxis. Management interventions include staff education, systematic screening, use of multicomponent interventions, and pharmacologic interventions.

Related Resources
• Downing LJ, Caprio TV, Lyness JM. Geriatric psychiatry review: differential diagnosis and treatment of the 3 D’s - delirium, dementia, and depression. Curr Psychiatry Rep. 2013;15(6):365.
• Brooks PB. Postoperative delirium in elderly patients. Am J Nurs. 2012;112(9):38-49.

Drug Brand Names
Carbidopa/levodopa • Sinemet                       Midazolam •  Versed
Dexmedetomidine • Precedex                        Olanzapine •  Zyprexa
Haloperidol • Haldol                                      Propofol  •  Diprivan
Lithium • Eskalith, Lithobid                            Quetiapine  •  Seroquel
Lorazepam • Ativan                                      Risperidone  •  Risperdal
Metoclopramide •  Reglan                              Trazodone  •  Desyrel

Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Although delirium has many descriptive terms (Table 1), a common unifying term is “acute global cognitive dysfunction,” now recognized as delirium; a consensus supported by DSM-5¹ and ICD-10² (Table 2). According to DSM-5, the essential feature is a disturbance of attention or awareness that is accompanied by a change in baseline cognition that cannot be explained by another preexisting, established, or evolving neurocognitive disorder (the newly named DSM-5 entity for dementia syndromes).¹ Because delirium affects the cortex diffusely, psychiatric symp­toms can include cognitive, mood, anxiety, or psychotic symp­toms. Because many systemic illnesses can induce delirium, the differential diagnosis spans all organ systems.

Three subtypes
Delirium can be classified, based on symptoms,^3,4 into 3 sub­types: hyperactive-hyperalert, hypoactive-hypoalert, and mixed delirium. Hyperactive patients present with rest­lessness and agitation. Hypoactive patients are lethargic, confused, slow to respond to questions, and often appear depressed. The differential prognostic significance of these subtypes has been examined in the literature, with conflicting results. Rabinowitz⁵ reported that hypoactive delirium has the worst prognosis, while Marcantonio et al⁶ indicated that the hyperactive subtype is associated with the highest mortality rate. Mixed delirium, with periods of both hyperactivity and hypoactivity, is the most common type of delirium.⁷

A prodromal phase, characterized by anxiety, frequent requests for nursing and medical assistance, decreased attention, rest­lessness, vivid dreams, disorientation imme­diately after awakening, and hallucinations, can occur before an episode of full-spectrum delirium; this prodromal state often is iden­tified retrospectively —after the patient is in an episode of delirium.^8,9

Evidence-based guidelines aim to improve recognition and clinical management.^10-13 Disruptive behavior is the main reason for psychiatric referral in delirium.^14,15 Delayed psychiatric consultation because of non-recognition of delirium is related to variables such as older age; history of a pre-existing, comorbid neurocognitive disorder; and the clinical appearance of hypoactive delirium.¹⁴

The case of Mr. D (Box),¹⁶ illus­trates how the emergence of antipsychotic-associated neuroleptic malignant syndrome (NMS) can complicate antipsychotic treat­ment of delirium in a geriatric medical patient, although delirium also is a common presentation in NMS.¹⁷ Delirium developed after an increase in carbidopa/levodopa, which has central dopaminergic effects that can precipitate delirium, particularly in a geriatric patient with preexisting comorbid neurocognitive disorder. Further complicat­ing Mr. D’s delirium presentation was the development of NMS, which had a multifac­torial causation, such as the use of dopamine antagonists (ie, quetiapine, metoclopramide), and an abrupt decrease of a dopaminergic agent (ie, carbidopa/levodopa), all inducing a central dopamine relative hypoactivity.

Epidemiology
Delirium is more common in older patients,¹⁵ and is seen in 30% to 40% of hospitalized geri­atric patients.¹⁸ Delirium in older patients, compared with other adults, is associated with more severe cognitive impairment.¹⁹ It is com­mon among geriatric surgical patients (15% to 62%)²⁰ with a peak 2 to 5 days postoperatively for hip fracture,²¹ and often is seen in ICU patients (70% to 87%).²⁰ However, Spronk et al²² found that delirium is significantly under-recognized in the ICU. Nearly 90% of terminally ill patients become delirious before death.²³ Terminal delirium often is unrecognized and can interfere with assessment of other clinical problems.²⁴ A preexisting history of comor­bid neurocognitive disorder was evident in as many as two-thirds of delirium cases.²⁵

Pathophysiology and risk factors
The pathophysiology of delirium has been characterized as an imbalance of CNS metab­olism, including decreased blood flow in vari­ous regions of the brain that may normalize once delirium resolves.²⁶ Studies describe the simultaneous decrease of cholinergic transmission and dopaminergic excess.^27,28 Predisposing and precipitating factors for delirium that are of particular importance in geriatric patients include:
   • advanced age
   • CNS disease
   • infection
   • cognitive impairment
   • male sex
   • poor nutrition
   • dehydration and other metabolic abnormalities
   • cardiovascular events
   • substance use
   • medication
   • sensory deprivation (eg, impaired vision or hearing)
   • sleep deprivation
   • low level of physical activity.^27,29,30

Table 3 lists the most common delirium-provocative medications.²⁷

Evaluation and psychometric scales
The EEG can be useful in evaluating delir­ium, especially in clinically ambiguous cases. EEG findings may indicate generalized slowing or dropout of the posterior domi­nant rhythm, and generalized slow theta and delta waves, findings that are more common in delirium than in other neurocognitive dis­orders and other psychiatric illnesses. The EEG must be interpreted in the context of the delirium diagnostic workup, because abnor­malities seen in other neurocognitive disor­ders can overlap with those of delirium.³¹

The EEG referral should specify the clini­cal suspicion of delirium to help interpret the results. Delirium cases in which the patient’s previous cognitive status is unknown may benefit from EEG evaluation, such as:
   • in possible status epilepticus
   • when delirium improvement has reached a plateau at a lower level of cognitive function than before onset of delirium
   • when the patient is unable or unwilling to complete a psychiatric interview.²⁷

Assessment instruments are available to diagnose and monitor delirium (Table 4). Typically, delirium assessment includes examining levels of arousal, psychomotor activity, cognition (ie, orienta­tion, attention, and memory), and percep­tual disturbances.

Psychometrically, a review of Table 4 suggests that validity appeared stable with adequate specificity (64% to 99%) but more variable sensitivity (36% to 100%). These reliability parameters also will be affected by the classification sys­tem (ie, DSM vs ICD) and the cut-off score employed.³² Most measures (eg, Confusion Assessment Method [CAM], CAM-ICU) provide an adequate sample of behavioral (ie, level of alertness), motor (ie, psychomo­tor activity), and cognitive (ie, orientation, attention, memory, and receptive language) function, with the exception of the Global Attentiveness Rating, which is a 2-minute open conversation protocol between physi­cian and patient.

Some measures are stand-alone instru­ments, such as the Memorial Delirium Assessment Scale, whereas the CAM requires administration of separate cogni­tive screens, including the Mini-Mental State Examination (MMSE) and Digit Span.³³ Instruments to detect delirium in critically ill patients are a more recent development. Wong et al³⁴ reported that the most widely studied tool was the CAM. Obtaining collat­eral information from family, caregivers, and hospital staff is essential, particularly given the fluctuating nature of delirium.

Management
Prevention. Identify patients at high risk of delirium so that preventive strategies can be employed. Multi-component, nonphar­macotherapeutic interventions are used in clinical settings but few randomized trials have been conducted. The contributing effectiveness of individual components is not well-studied, but most include staff education to increase awareness of delir­ium. Of 3 multi-component intervention randomized trials, 2 reported a signifi­cantly lower incidence of delirium in the intervention group.^35-37 Implementation of a multi-component protocol in medical/ surgical units was associated with a sig­nificant reduction in use of restraints.³⁸

As in Mr. D’s case, complex drug regimens, particularly for CNS illness, can increase the risk of delirium. Considering the medication profile for patients with complex systemic illness—in particular, minimizing the use anticholinergics and dopamine agonists— may be crucial in preventing delirium.

Prophylactic administration of antipsychotics may reduce the risk of devel­oping postoperative delirium.³⁹ Studies of the use of these agents were characterized by small sample sizes and selected groups of patient populations. Of the 4 random­ized studies evaluating prophylactic anti­psychotics (vs placebo), 3 found a lower incidence of delirium in the intervention groups.^39-41

A study of haloperidol in post-GI sur­gery patients showed a reduced occurrence of delirium,⁴⁰ whereas its prophylactic use in patients undergoing hip surgery⁴² did not reduce the incidence of delirium com­pared with placebo, but did decrease sever­ity when delirium occurred.⁴²

Risperidone³⁹ in post-cardiac sur­gery and olanzapine⁴¹ perioperatively in patients undergoing total knee or hip replacement have been shown to decrease delirium severity and duration. Targeted prophylaxis with risperidone⁴³ in post-cardiac surgery patients who showed disturbed cognition but did not meet cri­teria for delirium reduced the number of patients requiring medication, compared with placebo.⁴³

Dexmedetomidine, an α-2 adrenergic receptor agonist, compared with propofol or midazolam in post-cardiac valve surgery patients, resulted in a decreased incidence of delirium but no difference in delirium duration, hospital length of stay, or use of other medications.⁴⁴ However, other studies have shown that dexmedetomidine reduces ICU length of stay and duration of mechani­cal ventilation.⁴⁵

Treatment. Management of hospitalized medically ill geriatric patients with delirium is challenging and requires a comprehensive approach. The first step in delirium man­agement is prompt identification and man­agement of systemic medical disturbances associated with the delirium episode. First-line, nonpharmacotherapeutic strategies for patients with delirium include:
   • reorientation
   • behavioral interventions (eg, use of clear instructions and frequent eye con­tact with patients)
   • environmental interventions (eg, mini­mal noise, adequate lighting, and lim­ited room and staff changes)
   • avoidance of physical restraints.⁴⁶

Consider employing family members or hospital staff sitters to stay with the patient and to reassure, reorient, and watch for agitation and other unsafe behaviors (eg, attempted elopement). Psychoeducation for the patient and family on the phenomenol­ogy of delirium can be helpful.

The use of drug treatment strategies should be integrated into a comprehensive approach that includes the routine use of nondrug measures.⁴⁶ Using medications for treating hypoactive delirium, formerly con­troversial, now has wider acceptance.^47,48 A few high-quality randomized trials have been performed.^25,49,50

Pharmacotherapy, especially in frail patients, should be initiated at the lowest start­ing dosage and titrated cautiously to clinical effect and for the shortest period of time nec­essary. Antipsychotics are preferred agents for treating all subtypes of delirium; haloperidol is widely used.^46,51,52 However, antipsychotics, including haloperidol, can be associated with adverse neurologic effects such as extrapyra­midal symptoms (EPS) and NMS.

Although reported less frequently than with haloperidol, other agents have been implicated in development of EPS and NMS, including atypical antipsychotics and anti­emetic dopamine antagonists, particularly in parkinsonism-prone patients.⁵³ Strategies that can minimize such risks in geriatric inpatients with delirium include oral, rather than par­enteral, use of antipsychotics—preferential use of atypical over typical antipsychotics— and lowest effective dosages.⁵⁴

In controlled trials, atypical antipsychot­ics for delirium showed efficacy compared with haloperidol.^52,55 However, there is no research that demonstrates any advantage of one atypical over another.²⁵

In Mr. D’s case, the most important inter­vention for managing delirium caused by NMS is to discontinue all dopamine antag­onists and treat agitation with judicious doses of a benzodiazepine, with supportive care.¹⁷ In cases of sudden discontinuation or a dosage decrease of dopamine agonists, these medications should be resumed or optimized to minimize the risk of NMS-associated rhabdomyolysis and subse­quent renal failure.¹⁷ Antipsychotics carry an increased risk of stroke and mortality in older patients with established or evolving neurocognitive disorders^56,57 and can cause prolongation of the QTc interval.⁵⁷

Other medications that could be used for delirium include cholinesterase inhibitors^58,59 (although larger trials and a systematic review did not support this use⁶⁰), and 5-HT receptor antagonists,⁶¹ such as trazodone. Benzodiazepines, such as lorazepam, are first-line treatment for delirium associated with seizures or withdrawal from alcohol, sedatives, hypnotics, and anxiolytics and for delirium caused by NMS. Be cautious about using benzodiazepines in geriatric patients because of a risk of respiratory depression, falls, sedation, and amnesia.

Geriatric patients with alcoholism and those with malnutrition are prone to thia­mine and vitamin B₁₂ deficiencies, which can induce delirium. Laboratory assessment and consideration of supplementation is recom­mended. Despite high occurrence of delirium in hospitalized older adults with preexisting comorbid neurocognitive disorders, there is no standard care for delirium comorbid with another neurocognitive disorder.⁶² Clinical practice guidelines for older patients receiv­ing palliative care have been developed⁶³; the goal is to minimize suffering and discomfort in patients in palliative care.⁶⁴

Post-delirium prophylaxis. Medications for delirium usually can be tapered and discontinued once the episode has resolved and the patient is stable; it is common to discontinue medications when the patient has been symptom-free for 1 week.⁶⁵ Some patients (eg, with end-stage liver disease, disseminated cancer) are prone to recur­rent or to prolonged or chronic delirium. A period of post-recovery treatment with antipsychotics—even indefinite treatment in some cases—should be considered.

Post-delirium debriefing and aftercare. The psychological complications of delirium are distressing for the patient and his (her) caregivers. Psychiatric complications asso­ciated with delirium, including acute stress disorder—which might predict posttraumatic stress disorder—have been explored; early recognition and treatment may improve long-term outcomes.⁶⁶ After recovery from acute delirium, cognitive assessment (eg, MMSE⁶⁷ or Montreal Cognitive Assessment⁶⁸) is recommended to validate current cognitive status because patients may have persistent decrement in cognitive func­tion compared with pre-delirium condition, even after recovery from the acute episode.

Post-delirium debriefing may help patients who have recovered from a delirium episode. Patients may fear that their brief period of hallucinations might represent the onset of a chronic-relapsing psychotic dis­order. Allow patients to communicate their distress about the delirium episode and give them the opportunity to talk through the experience. Brief them on the possibility that delirium will recur and advise them to seek emergency medical care in case of recur­rence. Advise patients to monitor and main­tain a normal sleep-wake cycle.

Family members can watch for syndro­mal recurrence of delirium. They should be encouraged to discuss their reaction to hav­ing seen their relative in a delirious state.

Health care systems with integrated electronic medical records should list “delirium, resolved” on the patient’s illness profile or problem list and alert the patient’s primary care provider to the delirium his­tory to avoid future exposure to delirium-provocative medications, and to prompt the provider to assume an active role in post-delirium care, including delirium recur­rence surveillance, medication adjustment, risk factor management, and post-recovery cognitive assessment.

Bottom Line
Evaluation of delirium in geriatric patients includes clinical vigilance and screening, differentiating delirium from other neurocognitive disorders, and identifying and treating underlying causes. Perioperative use of antipsychotics may reduce the incidence of delirium, although hospital length of stay generally has not been reduced with prophylaxis. Management interventions include staff education, systematic screening, use of multicomponent interventions, and pharmacologic interventions.

Related Resources
• Downing LJ, Caprio TV, Lyness JM. Geriatric psychiatry review: differential diagnosis and treatment of the 3 D’s - delirium, dementia, and depression. Curr Psychiatry Rep. 2013;15(6):365.
• Brooks PB. Postoperative delirium in elderly patients. Am J Nurs. 2012;112(9):38-49.

Drug Brand Names
Carbidopa/levodopa • Sinemet                       Midazolam •  Versed
Dexmedetomidine • Precedex                        Olanzapine •  Zyprexa
Haloperidol • Haldol                                      Propofol  •  Diprivan
Lithium • Eskalith, Lithobid                            Quetiapine  •  Seroquel
Lorazepam • Ativan                                      Risperidone  •  Risperdal
Metoclopramide •  Reglan                              Trazodone  •  Desyrel

Disclosures
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Mild cognitive impairment (MCI) is a transitional clinical stage between normal aging and demen­tia. Together with aging, it is considered the most significant risk factor for developing dementia, often the Alzheimer’s type.¹

MCI is a challenging neuropsychiat­ric diagnosis to discuss with patients and their family because it is characterized by overlapping features of normal aging and because of its heterogeneity of etiol­ogy, clinical presentation, and outcome.^2,3 The evolution to dementia and the lack of effective treatments for preventing or forestalling this outcome can be difficult to address—particularly when the patient is in good health and has been leading a productive life.

Successful communication is key
You can take steps to communicate in a helpful way, build a strong treatment alli­ance, and reduce the potential for the iat­rogenic effects of disclosing this diagnosis and its prognostic implications.

Clarify that your findings are consistent with the patient’s or family’s report of sustained and concerning change in cog­nition and, depending on the patient, concurrent alterations in affect, behavior, or both. Emphasize that these changes are disproportionately severe relative to expectations for the patient’s age and are not caused by psychiatric or clear-cut medical factors.

Highlight contexts in which the patient’s symptoms are likely to become more dis­ruptive and impaired, and situations in which the patient can be expected to func­tion more effectively.

Provide evidence-based support for the rate of progression of symptoms and func­tional impairment.³

Emphasize that major lifestyle adjust­ments usually are unnecessary in the absence of progression, especially for patients who are retired or not involved in endeavors that involve significant cogni­tive and executive functioning demands.

Discuss the role that cognition-enhancing medications might play in managing symptoms.⁴

Address indications for additional services, including formal psychiatric care for patients who have concomitant affec­tive or behavioral symptoms and who are highly distressed by the diagnosis. Pair these services with longitudinal monitor­ing for possible exacerbation of symptoms.

Identify psychiatric, medical, and life­style factors that can increase the risk of dementia. Depending on the patient’s history, this might include diabetes, hypertension, elevated lipid levels, obe­sity, smoking, head trauma, depres­sion, physical inactivity, and lack of intellectual stimulation.

Review compensatory strategies. In MCI predominantly amnestic type, for example, having the patient make system­atic lists for shopping and other activities of daily living, as well as establishing rou­tines for organizaton, can bolster success­ful coping.

If psychometric testing was not utilized to establish the diagnosis, discussion can include the value of performing such an assessment for a more finely tuned profile of preserved and impaired neurobehavioral functions. Such a profile can include test patterns that 1) have prognostic value with regard to the likelihood of progression to dementia and 2) establish a baseline against which you can assess stability or progression over time.⁵

Disclosure
Dr. Pollak reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Mild cognitive impairment (MCI) is a transitional clinical stage between normal aging and demen­tia. Together with aging, it is considered the most significant risk factor for developing dementia, often the Alzheimer’s type.¹

MCI is a challenging neuropsychiat­ric diagnosis to discuss with patients and their family because it is characterized by overlapping features of normal aging and because of its heterogeneity of etiol­ogy, clinical presentation, and outcome.^2,3 The evolution to dementia and the lack of effective treatments for preventing or forestalling this outcome can be difficult to address—particularly when the patient is in good health and has been leading a productive life.

Successful communication is key
You can take steps to communicate in a helpful way, build a strong treatment alli­ance, and reduce the potential for the iat­rogenic effects of disclosing this diagnosis and its prognostic implications.

Clarify that your findings are consistent with the patient’s or family’s report of sustained and concerning change in cog­nition and, depending on the patient, concurrent alterations in affect, behavior, or both. Emphasize that these changes are disproportionately severe relative to expectations for the patient’s age and are not caused by psychiatric or clear-cut medical factors.

Highlight contexts in which the patient’s symptoms are likely to become more dis­ruptive and impaired, and situations in which the patient can be expected to func­tion more effectively.

Provide evidence-based support for the rate of progression of symptoms and func­tional impairment.³

Emphasize that major lifestyle adjust­ments usually are unnecessary in the absence of progression, especially for patients who are retired or not involved in endeavors that involve significant cogni­tive and executive functioning demands.

Discuss the role that cognition-enhancing medications might play in managing symptoms.⁴

Address indications for additional services, including formal psychiatric care for patients who have concomitant affec­tive or behavioral symptoms and who are highly distressed by the diagnosis. Pair these services with longitudinal monitor­ing for possible exacerbation of symptoms.

Identify psychiatric, medical, and life­style factors that can increase the risk of dementia. Depending on the patient’s history, this might include diabetes, hypertension, elevated lipid levels, obe­sity, smoking, head trauma, depres­sion, physical inactivity, and lack of intellectual stimulation.

Review compensatory strategies. In MCI predominantly amnestic type, for example, having the patient make system­atic lists for shopping and other activities of daily living, as well as establishing rou­tines for organizaton, can bolster success­ful coping.

If psychometric testing was not utilized to establish the diagnosis, discussion can include the value of performing such an assessment for a more finely tuned profile of preserved and impaired neurobehavioral functions. Such a profile can include test patterns that 1) have prognostic value with regard to the likelihood of progression to dementia and 2) establish a baseline against which you can assess stability or progression over time.⁵

Disclosure
Dr. Pollak reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Mild cognitive impairment (MCI) is a transitional clinical stage between normal aging and demen­tia. Together with aging, it is considered the most significant risk factor for developing dementia, often the Alzheimer’s type.¹

MCI is a challenging neuropsychiat­ric diagnosis to discuss with patients and their family because it is characterized by overlapping features of normal aging and because of its heterogeneity of etiol­ogy, clinical presentation, and outcome.^2,3 The evolution to dementia and the lack of effective treatments for preventing or forestalling this outcome can be difficult to address—particularly when the patient is in good health and has been leading a productive life.

Successful communication is key
You can take steps to communicate in a helpful way, build a strong treatment alli­ance, and reduce the potential for the iat­rogenic effects of disclosing this diagnosis and its prognostic implications.

Clarify that your findings are consistent with the patient’s or family’s report of sustained and concerning change in cog­nition and, depending on the patient, concurrent alterations in affect, behavior, or both. Emphasize that these changes are disproportionately severe relative to expectations for the patient’s age and are not caused by psychiatric or clear-cut medical factors.

Highlight contexts in which the patient’s symptoms are likely to become more dis­ruptive and impaired, and situations in which the patient can be expected to func­tion more effectively.

Provide evidence-based support for the rate of progression of symptoms and func­tional impairment.³

Emphasize that major lifestyle adjust­ments usually are unnecessary in the absence of progression, especially for patients who are retired or not involved in endeavors that involve significant cogni­tive and executive functioning demands.

Discuss the role that cognition-enhancing medications might play in managing symptoms.⁴

Address indications for additional services, including formal psychiatric care for patients who have concomitant affec­tive or behavioral symptoms and who are highly distressed by the diagnosis. Pair these services with longitudinal monitor­ing for possible exacerbation of symptoms.

Identify psychiatric, medical, and life­style factors that can increase the risk of dementia. Depending on the patient’s history, this might include diabetes, hypertension, elevated lipid levels, obe­sity, smoking, head trauma, depres­sion, physical inactivity, and lack of intellectual stimulation.

Review compensatory strategies. In MCI predominantly amnestic type, for example, having the patient make system­atic lists for shopping and other activities of daily living, as well as establishing rou­tines for organizaton, can bolster success­ful coping.

If psychometric testing was not utilized to establish the diagnosis, discussion can include the value of performing such an assessment for a more finely tuned profile of preserved and impaired neurobehavioral functions. Such a profile can include test patterns that 1) have prognostic value with regard to the likelihood of progression to dementia and 2) establish a baseline against which you can assess stability or progression over time.⁵

Disclosure
Dr. Pollak reports no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Severe and enduring anorexia ner­vosa (SE-AN) is persistent anorexia nervosa (AN) lasting for ≥7 years with or without a history of treat­ment. Evidence points to the effec­tiveness of a patient-tailored plan for treating SE-AN over any universal fix. Proper medication, therapeutic alliance, and strategic discharge planning are the ingre­dients for treating SE-AN that avoids re-hospitalization (Table).

Nutritional support and pharmacotherapy required
Comprehensive metabolic analysis and initiating nutrition should be the first pri­ority for the medical team. Starved-state patients can have electrolyte and metabolic derangements that place them at risk of fatal arrhythmias or multi-system organ failure. Do not hesitate to initiate nasogastric tube feeding under the observation of a certified nutritionist when necessary for survival. A double-blind, randomized controlled trial demonstrated the benefit of olanzap­ine compared with placebo to increase body mass index (BMI) of hospitalized AN patients. Olanzapine was titrated from 2.5 to 10 mg/d over a 13-week period, and was associated with higher patient achieve­ment of a BMI > 18.5 kg/m2.¹

Although the patient is receiving nutri­tional support in conjunction with psycho­tropic medication, the road to BMI recovery can be long. Don’t forget that SE-AN can be incapacitating. In SE-AN, the fear of gaining weight is so severe that the idea of starvation-induced death initially might seem more palatable. Although counterintuitive, as the patient recovers metabolically, self-image deteriorates. Statements praising any new weight gain can derail any therapeutic relationship.

Therapeutic alliance is key
Establishing high-quality therapeutic alliance, as measured by the Helping Relationships Questionnaire, has been shown to have a positive outcome on eating disorder symptoms and comorbid depressed mood in later phases of SE-AN treatment.^2,3 Although therapeutic alliance is individual­ized, maintaining open communication and reiterating how it is the patient’s decision to consume whole food at a level at which the feeding tube can be discontinued are good places to start treatment.

Proper discharge timing and transition to outpatient care for SE-AN patients is paramount. In multicenter studies, treat­ment ends too early in 57.8% of patients; discharge at sub-ideal BMI is linked to rehospitalization.³ Slower weight gain and delayed establishment of therapeutic alli­ance are predictors of patients who exit treatment programs too early.³ Clinicians who remain vigilant for the above metrics are less likely to feed into the unacceptably high rate of treatment failure for SE-AN.

Disclosures
The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Severe and enduring anorexia ner­vosa (SE-AN) is persistent anorexia nervosa (AN) lasting for ≥7 years with or without a history of treat­ment. Evidence points to the effec­tiveness of a patient-tailored plan for treating SE-AN over any universal fix. Proper medication, therapeutic alliance, and strategic discharge planning are the ingre­dients for treating SE-AN that avoids re-hospitalization (Table).

Nutritional support and pharmacotherapy required
Comprehensive metabolic analysis and initiating nutrition should be the first pri­ority for the medical team. Starved-state patients can have electrolyte and metabolic derangements that place them at risk of fatal arrhythmias or multi-system organ failure. Do not hesitate to initiate nasogastric tube feeding under the observation of a certified nutritionist when necessary for survival. A double-blind, randomized controlled trial demonstrated the benefit of olanzap­ine compared with placebo to increase body mass index (BMI) of hospitalized AN patients. Olanzapine was titrated from 2.5 to 10 mg/d over a 13-week period, and was associated with higher patient achieve­ment of a BMI > 18.5 kg/m2.¹

Although the patient is receiving nutri­tional support in conjunction with psycho­tropic medication, the road to BMI recovery can be long. Don’t forget that SE-AN can be incapacitating. In SE-AN, the fear of gaining weight is so severe that the idea of starvation-induced death initially might seem more palatable. Although counterintuitive, as the patient recovers metabolically, self-image deteriorates. Statements praising any new weight gain can derail any therapeutic relationship.

Therapeutic alliance is key
Establishing high-quality therapeutic alliance, as measured by the Helping Relationships Questionnaire, has been shown to have a positive outcome on eating disorder symptoms and comorbid depressed mood in later phases of SE-AN treatment.^2,3 Although therapeutic alliance is individual­ized, maintaining open communication and reiterating how it is the patient’s decision to consume whole food at a level at which the feeding tube can be discontinued are good places to start treatment.

Proper discharge timing and transition to outpatient care for SE-AN patients is paramount. In multicenter studies, treat­ment ends too early in 57.8% of patients; discharge at sub-ideal BMI is linked to rehospitalization.³ Slower weight gain and delayed establishment of therapeutic alli­ance are predictors of patients who exit treatment programs too early.³ Clinicians who remain vigilant for the above metrics are less likely to feed into the unacceptably high rate of treatment failure for SE-AN.

Disclosures
The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Severe and enduring anorexia ner­vosa (SE-AN) is persistent anorexia nervosa (AN) lasting for ≥7 years with or without a history of treat­ment. Evidence points to the effec­tiveness of a patient-tailored plan for treating SE-AN over any universal fix. Proper medication, therapeutic alliance, and strategic discharge planning are the ingre­dients for treating SE-AN that avoids re-hospitalization (Table).

Nutritional support and pharmacotherapy required
Comprehensive metabolic analysis and initiating nutrition should be the first pri­ority for the medical team. Starved-state patients can have electrolyte and metabolic derangements that place them at risk of fatal arrhythmias or multi-system organ failure. Do not hesitate to initiate nasogastric tube feeding under the observation of a certified nutritionist when necessary for survival. A double-blind, randomized controlled trial demonstrated the benefit of olanzap­ine compared with placebo to increase body mass index (BMI) of hospitalized AN patients. Olanzapine was titrated from 2.5 to 10 mg/d over a 13-week period, and was associated with higher patient achieve­ment of a BMI > 18.5 kg/m2.¹

Although the patient is receiving nutri­tional support in conjunction with psycho­tropic medication, the road to BMI recovery can be long. Don’t forget that SE-AN can be incapacitating. In SE-AN, the fear of gaining weight is so severe that the idea of starvation-induced death initially might seem more palatable. Although counterintuitive, as the patient recovers metabolically, self-image deteriorates. Statements praising any new weight gain can derail any therapeutic relationship.

Therapeutic alliance is key
Establishing high-quality therapeutic alliance, as measured by the Helping Relationships Questionnaire, has been shown to have a positive outcome on eating disorder symptoms and comorbid depressed mood in later phases of SE-AN treatment.^2,3 Although therapeutic alliance is individual­ized, maintaining open communication and reiterating how it is the patient’s decision to consume whole food at a level at which the feeding tube can be discontinued are good places to start treatment.

Proper discharge timing and transition to outpatient care for SE-AN patients is paramount. In multicenter studies, treat­ment ends too early in 57.8% of patients; discharge at sub-ideal BMI is linked to rehospitalization.³ Slower weight gain and delayed establishment of therapeutic alli­ance are predictors of patients who exit treatment programs too early.³ Clinicians who remain vigilant for the above metrics are less likely to feed into the unacceptably high rate of treatment failure for SE-AN.

Disclosures
The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.