Clearing up confusion

“Mr. Smith seems somewhat confused today” is one of the most serious and concerning pre-visit reports you can receive from your staff or the patient’s family. Such a descriptor can be confusing—pardon the pun—not only for the patient, but to even seasoned mental health providers.

The term confusion can be code for diagnoses ranging from delirium^a to a progressive neurocognitive disorder (NCD) such as major NCD due to Alzheimer’s disease (AD), or even a more challenging prob­lem such as beclouded dementia (delirium superimposed on demen­tia/NCD). It is essential for all mental health professionals to have an evidence-based approach when encountering signs or symptoms of confusion.

^aICD-10 code R41.0 encompasses Confusion, Other Specified Delirium, or Unspecified Delirium.

CASE REPORT
Ms. T, age 62, has hypothyroidism and bipolar I disorder, most recently depressed, with comorbid generalized anxiety disorder. She has been tak­ing lithium, 600 mg/d, to control her mood symptoms. Her daughter-in-law reports that Ms. T has been exhibiting increasing signs of confusion. During the office evaluation, Ms. T minimizes her symptoms, only describ­ing mild issues with forgetfulness while cooking and concern over increas­ing anxiety. Her daughter-in-law plays a voicemail message from earlier in the week, in which Ms. T’s speech is halting, disorganized, and in a word, confused. I decide to use the mnemonic decision chart MR. MIND (Table 1) to get to the bottom of her recent confusion.

Measure cognition
It is nice to receive advanced warning about a cognitive change or a change in activities of daily living; however, many patients present with subtle, sub-acute changes that are more difficult to assess. When encountering a broad symptom such as “confusion”—which has an equally broad differential diagnosis—systematic assess­ment of the current cognitive state com­pared with the patient’s baseline becomes the first order of business. However, this requires that the patient has had a baseline cognitive assessment.

In my practice, I often administer one of the validated neurocognitive screening instruments when a patient first begins care—even a brief test such as the Mini- Cog (3-item recall plus clock drawing test), which is comparable to longer screening tests at least for NCD/dementia.¹ During a presentation for confusion, a more detailed neurocognitive assessment instrument would be recommended, allowing one to marry the clinical impression with a validated, objective measure. Formal neu­ropsychological testing by a clinical neuro­psychologist is the gold standard, but such testing is time-consuming and expensive and often not readily available. The screen­ing instrument I use for a more thorough evaluation depends on the clinical scenario.

The Six-Item Screener is used in some emergency settings because it is short but boasts a higher sensitivity than the Mini- Cog (94% vs 75%) with similar specificity when screening for cognitive impairment.² The Mini-Mental State Examination (MMSE) is a valuable instrument, although, recently, the Saint Louis University Mental Status Examination has been thought to be better at detecting mild NCD than the MMSE; more data are needed to substan­tiate this claim.³ The Montreal Cognitive Assessment is another validated screening tool that has been shown to be superior to the MMSE in terms of screening for mild cognitive impairment.⁴ The best delirium-specific assessment tool is the Confusion Assessment Method (Table 2).⁵

Ms. T’s MMSE score was 26/30, down from 29/30 at baseline. Her score fell below the cutoff score of 27 for mild cogni­tive impairment for someone with at least 8 years of completed education. Her results were abnormal mainly in the memory domain (3-item recall), raising the ques­tion of a possible prodromal state of AD although the acute nature of the change made delirium or mild NCD high in the differential.

Review medications
A review of the medication list is not just a Joint Commission mandate (medication reconciliation during each encounter) but is important whenever confusion is noted. Polypharmacy can be a concern, but is not as concerning as the class of medication prescribed, particularly anticholinergic and sedative medications in patients age >65. The Drug Burden Index can be helpful in assessing this risk.⁶ Medications such as the benzodiazepine-receptor agonists, tri­cyclic antidepressants, and antipsychotics should be discontinued if possible, keeping in mind that the addition or subtraction of medications must be done prudently and only after reviewing the evidence and in consultation with the patient. A detailed medication review is as important for confused outpatients as it is for an inpatient case (steps 2 and 3 of the inpatient algo­rithm outlined in Table 3).⁷

In Ms. T’s case, the primary concern on her medication list was that her medical team was prescribing levothyroxine, 112 mcg/d, and desiccated thyroid (combination thy­roxine and triiodothyronine in the form of 20 mg Armour Thyroid), despite a lack of data for such combination therapy. Earlier, I had discontinued lorazepam, leaving lithium, 600 mg/d, quetiapine, 400 mg/d, and escitalopram, 10 mg/d, as her remain­ing psychotropics. Her other medications included atorvastatin, 40 mg/d, for hyper-lipidemia and metformin, 750 mg/d, for type 2 diabetes mellitus.