Anticholinergic drugs are known to produce or enhance cognitive deficits. Tolterodine tartrate is marketed as a bladder-selective anticholinergic drug that is reported to be free of significant cognitive adverse effects.
To describe a 46-year-old woman who had memory loss and abnormal memory test results that improved when she discontinued tolterodine therapy.
While taking tolterodine, the patient's score on the delayed free recall portion of the Hopkins Verbal Learning Test–Revised was at the first percentile. One month after discontinuing tolterodine therapy, this test was administered a second time using an alternative form and she showed marked improvement scoring above the 75th percentile.
Tolterodine therapy caused cognitive dysfunction in our patient. It is possible that cognitive dysfunction is a common result of tolterodine treatment, but in the absence of testing, remains undiagnosed. Alternatively, our patient may have had aberrant metabolism of this drug or an increased sensitivity as a result of incipient Alzheimer disease.
MEDICATIONS THAT block muscarinic cholinergic transmission are known to interfere with learning and memory.1 Patients with Alzheimer disease (AD) have cholinergic deficits and have an increased sensitivity to cognitive adverse effects from these agents.2 We examined a woman with memory impairment who was being treated with tolterodine tartrate, a muscarinic antagonist. The online Physician's Desk Reference3 listed no true cognitive adverse effects and no contraindications or precautions that pertain to the use of tolterodine in cognitively impaired or elderly populations. Furthermore, a review of the literature on tolterodine argues against significant cognitive adverse effects (other than a 2% incidence of dizziness and fatigue)4- 10 and even promotes its use in older patients.11 When we discontinued the use of this medication, however, her memory improved.
A 46-year-old, right-handed woman reported decreased memory, which had become worse over the preceding 2 years. Specifically, she reported forgetting events such as meetings at work as well as some outings with her family. She began having to write down everything at work, and reported particular difficulty remembering numbers. She also reported syntactic errors in which word order was disturbed. Her medical history was notable for anxiety, essential tremor, and urinary stress incontinence. Her medications at the time of evaluation included a combination of conjugated estrogens, 0.625 mg/d, and medroxyprogesterone acetate, 5 mg/d; venlafaxine hydrochloride, 150 mg/d; fexofenadine hydrochloride, 180 mg/d; propranolol hydrochloride, 10 mg, three times daily; tolterodine tartrate, 4 mg/d (which had been started 3 months prior to this visit); and vitamin E, 800 IU/d. Her family history was positive for dementia in her mother (mixed, AD and vascular dementia), her maternal grandmother (type unknown), and several maternal aunts and uncles (AD). During testing, the patient appeared calm and put forth good effort. Mental status testing revealed a Mini-Mental State Examination score of 30. Testing of attention revealed no distractibility during visual field examination, a digit span of 7 forward, and normal performance on a letter cancellation task. Intention was also intact with no evidence of impersistance on a sustained motor task, no evidence of bradykinesia, normal performance on a contrasting programs paradigm, and a normal score (39) on the controlled word association test12 for letters /F/, /A/, and /S/. The Boston Naming Test13 was given and she scored 56 of 60, which is normal. Her spontaneous speech was fluent with normal pace, word quantity, grammar, and articulation. Repetition and comprehension were intact. There was no ideomotor apraxia. Finger naming was intact and there was no right-left confusion. Because of her report of impaired memory, we administered the Hopkins Verbal Learning Test–Revised (form 3)14 where she was presented a list of 12 words on 3 sequential trials. After each trial her immediate recall of these words was assessed. She performed well on the learning and immediate free recall test with a combined score of 30 of 36 (50th-75th percentile), which is in the normal range. On delayed free recall, however, she remembered only 5 of 12 words (first percentile), which is abnormal, and had one semantically related intrusion. In contrast, her delayed recognition was good, recognizing 12 of 12 with no false-positive results. Findings from the remainder of her neurological examination showed no abnormalities. Values from laboratory studies including a magentic resonance image of her head, free thyroxine, thyrotropin, treponemal antibody test, vitamin B12, methylmalonic acid, folate, and erythrocyte sedimentation rate were normal. We recommended that she increase her dose of vitamin E up to 2000 IU/d and discontinue tolterodine therapy. We saw her in follow-up 1 month later and she reported improvement in her memory and no further syntactic errors. A second testing with the Hopkins Verbal Learning Test–Revised (form 2), which is equivalent to form 3 for recall and has good test-retest reliability,14 revealed an immediate free recall score of 33 of 36 and a perfect delayed free recall of 12 of 12 (≥75th percentile).
Scopolamine, a prototypic antimuscarinic drug, is known to impair verbal learning, to severely impair delayed free recall, and to have only minor or no effect on recognition recall for items learned under its influence. Additionally, it has little or no effect on immediate recall as tested by digit span.15- 17
While our patient had normal verbal learning, her severely impaired delayed free recall with normal recognition and a normal digit span is consistent with the expected deficits from an anticholinergic medication. The resolution of her free recall deficit when tolterodine therapy was discontinued further supports the causal role of this medication. The patient's memory impairment began prior to the initiation of tolterodine therapy suggesting that her memory deficit could not be solely due to this medication. Nevertheless, in the absence of tolterodine therapy, results from the patient's memory testing were normal and any underlying diagnosis remained obscure in the absence of objective deficits. The relationship of her reported language disturbance to tolterodine was unclear since we had no objective documentation of this report.
Despite what is known about the effect of antimuscarinic drugs on memory and cognition, tolterodine does not carry specific warnings about true cognitive adverse effects or use in at-risk populations. The clinical trials leading to Food and Drug Administration approval of tolterodine relied primarily on spontaneous reports of adverse events for dry mouth, and detailed cognitive testing was not done.4,7 Another muscarinic antagonist, oxybutynin chloride, likewise carried no warnings of cognitive adverse effects (except for dizziness and somnolence)18 but was later shown to cause a decline in 7 of 15 cognitive measurements in healthy elderly subjects.19
It has been argued that tolterodine would be less likely to cause cognitive deficits than oxybutynin because oxybutynin caused significant changes in the quantitative electroencephalographic power spectrum while tolterodine produced only a slight power reduction within the theta frequency band.20 Such measurements, however, may not fully predict if an agent can induce amnesia. Another argument cites the relatively low penetrance across the blood-brain barrier of tolterodine and its metabolites as measured in mice21 and as suggested by the lipophilicity of the active compounds. Tolterodine is more than 30 times less lipophilic than oxybutynin. The active 5-hydroxymethyl metabolite, which is the predominant species in extensive metabolizers, is more than 350 times less lipophilic than oxybutynin.22
Despite those arguments, there are at least 3 reasons that, alone or in combination, may explain why our patient suffered memory problems secondary to tolterodine therapy. First, since the effect on memory in humans has not been directly studied, perhaps memory impairment is a common but heretofore undocumented adverse effect of tolterodine. Second, our patient may have been one of the 7% of the population, who lack the CYP2D6 enzyme that converts tolterodine to the 5-hydroxymethyl metabolite.3 In these poor metabolizers, the more lipophilic parent compound predominates, resulting in the likelihood of greater central nervous system penetration. Third, she may have had an unusual sensitivity to antimuscarinic medications, which is interesting in light of the known increased sensitivity seen in AD.2
Ever since Sunderland et al2 demonstrated this increased sensitivity to scopolamine in patients with AD, there has been speculation about the possibility of detecting a preclinical phase of AD by evaluating asymptomatic individuals for increased anticholinergic sensitivity.23 Although one small study looked at this prospectively with negative results24 and another study25 has argued that the cholinergic degeneration may occur later in the course of AD than previously thought, the question remains open. Our patient's strong family history of AD makes this possibility particularly salient. In addition to impaired free recall, patients with AD typically have deficits of recognition recall,26,27 which our patient did not have. If preclinical changes, however, merely increase the susceptibility to anticholinergic drugs, then the pattern of deficits would be expected to be consistent with these drugs rather than with AD. Additionally, one study that looked at healthy, very elderly subjects with no dementia found that those who later developed AD had performed worse on recall and no differently on recognition than the members of their cohort who remained free of dementia.28
Studies are needed to directly examine the cognitive effect of tolterodine therapy on young as well as elderly populations using neuropsychological instruments. Correlation with metabolic type, age, and risk factors for AD (apolipoprotein E genotype, family history) could identify groups at particular risk for the cognitive adverse effects of tolterodine to note whether a general risk exists. Additionally, we could also learn if increased sensitivity to drugs like tolterodine might be indicative of incipient AD.
Given the aging population and the direct marketing of this drug through television advertisements, more information is needed to prevent avoidable adverse effects and to recognize them when they occur. In addition to raising suspicions about the safety profile of tolterodine, this case also illustrates the use of detailed mental status testing. A screening test such as the Mini-Mental State Examination, which has limitations because of ceiling effects,29,30 would have missed an identifiable problem that, in this case, was treatable.
Corresponding author: Kyle B. Womack, MD, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Blvd, Dallas, TX 75390 (e-mail: firstname.lastname@example.org).
Accepted for publication August 23, 2002.
Author contributions: Study concept and design (Drs Womack and Heilman); acquisition of data (Dr Womack); analysis and interpretation of data (Drs Womack and Heilman); drafting of the manuscript (Dr Womack); critical revision of the manuscript for important intellectual content (Dr Heilman); administrative, technical, and material support (Dr Womack); study supervision (Dr Heilman).
This study was supported in part by a grant from the State of Florida, Department of Elder Affairs, Tallahassee.
Womack KB, Heilman KM. Tolterodine and MemoryDry but Forgetful. Arch Neurol. 2003;60(5):771-773. doi:10.1001/archneur.60.5.771