Case Presentation: A 74-year-old man with Parkinson’s disease on carbidopa-levodopa had started pramipexole for tremor control about 3 months prior to presentation. He initially started with 3 (0.125 mg) tablets three times a day and was titrated to 8 tablets three times a day, which he had maintained for 3 weeks. He presented to the emergency department with increased thirst and urination, fatigue, hypotension, nausea, chills, and confusion without frank hallucinations. His labs were significant for thrombocytopenia of 38k and lymphopenia. His symptoms were attributed to pramipexole due to similarities in presentation, and this medication was discontinued. However, thrombocytopenia is not a common effect of pramipexole overdose. The following day, his platelets continued to drop to 32k. The family was adamant that pramipexole was to blame and gave no history of tick exposure, no travel outside the house, and no dog exposure. TTP was considered, so a blood smear was ordered, but it revealed no schistocytes. Neurology considered the possibility that symptoms were due to too rapid of pramipexole discontinuation and reinstituted the medication at a lower dose. Pramipexole-induced symptomatology was questioned since the top dosage was not exceptionally high. The following day, platelets dropped to 24k. Labs for tickborne illnesses were ordered, and doxycycline was initiated. He later developed progressive shock requiring multiple pressors and was pronounced dead. Post-discharge testing returned Ehrlichia DNA positive.
Discussion: The patient’s medication history and lack of risk factors for tickborne illnesses led clinicians to misattribute symptoms to the medication. Anchoring and confirmation biases likely contributed to the initial diagnostic conclusion. Additionally, the patient’s symptom presentation (excluding thrombocytopenia) led clinicians to suspect dopamine agonist adverse effects. Subsequently, the family’s adamant history of the patient’s lack of tick exposure contributed to anchoring on pramipexole toxicity. Confirmation bias diverted attention from the tickborne illness despite the severe thrombocytopenia. Although fulminant ehrlichiosis may progress despite the doxycycline therapy, the delayed administration represents a modifiable cognitive error. Early administration of doxycycline when tickborne illness is suspected, without waiting for confirmatory testing, is critical in improving patient outcomes. Despite the patient’s social history of no prior risk factors for tickborne illnesses, infectious etiology should have been considered earlier.
Conclusions: This case emphasizes the importance of caution in diagnostic decision-making. Clinicians should work to maintain a broad differential diagnosis to avoid anchoring bias and weigh confirmatory data with discordant data in an effort to avoid confirmatory bias. This openness and humility can prevent missed, time-sensitive diagnoses like ehrlichiosis.