Eye Injections for Vision Loss Linked to Sudden Psychosis in One Elderly Patient

A routine injection into the eye, one of the most commonly performed procedures in modern ophthalmology, has been implicated in a startling psychiatric emergency. Clinicians in Istanbul report the case of a 75-year-old woman with no history of mental illness who developed persecutory delusions, auditory hallucinations, and suicidal thoughts within a single day of receiving her third intravitreal injection of bevacizumab, a drug widely used to treat age-related macular degeneration. The report, published in BMC Psychiatry by a team at Basaksehir Cam and Sakura City Hospital, does not claim that the drug caused the psychosis, but it documents a temporal association so close and so well investigated that it raises an uncomfortable question about a therapy millions of older patients receive every year.

Bevacizumab is a monoclonal antibody that blocks vascular endothelial growth factor, or VEGF, a signalling protein that drives the growth of the abnormal, leaky blood vessels that destroy central vision in wet age-related macular degeneration. Although the drug was originally developed and approved as an intravenous cancer therapy, ophthalmologists have used it off-label inside the eye for nearly two decades because it is inexpensive and effective. When injected directly into the vitreous cavity, the dose is tiny, typically 1.25 milligrams in 0.05 millilitres, and the conventional assumption has been that systemic exposure is negligible. Yet pharmacokinetic studies suggest that intravitreal bevacizumab may produce higher blood levels than some other anti-VEGF agents, and the full systemic consequences of repeated exposure in elderly, medically complex patients remain incompletely mapped.

The patient in this report had already received two bevacizumab injections without any psychiatric symptoms. The third injection, delivered to her right eye, was designated Day 0 by the clinical team. On Day +1, before her scheduled left-eye injection on Day +2, she developed persecutory delusions, heard voices, could not sleep, and expressed suicidal ideation. Her behaviour changed markedly. This abrupt onset, arriving within hours of the procedure, is precisely the kind of temporal pattern that prompts clinicians to look for an external trigger rather than a slowly evolving primary psychiatric illness.

What makes the case compelling is the thoroughness of the exclusion workup. The woman had hypertension, type 2 diabetes, and atrial fibrillation, and she was taking multiple medications, a combination that in geriatric patients frequently hides drug interactions, toxicity, or withdrawal syndromes behind a psychiatric presentation. Careful medication reconciliation revealed no recent changes of any kind. Urine toxicology was negative, ruling out illicit substances. There were no major psychosocial stressors and no family history of psychiatric disease. No corticosteroids, which are sometimes co-administered around eye injections and are well known to provoke psychosis in older adults, were documented peri-injection or topically.

The neurological evaluation was equally revealing. On examination the patient was oriented, her memory was intact, and her Mini-Mental State Examination score was 27 out of 30, essentially normal for her age. The Confusion Assessment Method, a bedside screen for delirium, was negative, arguing against the fluctuating inattention that characterises acute confusional states. Brain magnetic resonance imaging showed only mild age-related atrophy and chronic small-vessel ischaemic changes, with no acute lesions, no diffusion restriction, and no evidence of stroke. Laboratory findings were unremarkable. In other words, the usual structural, metabolic, and toxicological culprits of late-life psychosis were systematically absent, leaving the timing of the injection as the most striking anomaly in the clinical picture.

The treating psychiatrists initiated aripiprazole, a second-generation antipsychotic, at 2.5 milligrams per day and gradually titrated the dose to 7.5 milligrams per day. The psychotic symptoms resolved during treatment, and by Day +33 the patient was discharged in stable condition. She then entered a six-month follow-up period during which she experienced no recurrence, including a subsequent re-exposure to bevacizumab while continuing aripiprazole. That re-challenge is scientifically double-edged. It suggests that the antipsychotic may have protected her against a drug-triggered reaction, but it also weakens the causal argument, because a true drug-induced psychosis would typically be expected to recur on re-exposure unless pharmacologically suppressed.

The authors are careful about this ambiguity, and their caution is warranted. Causal attribution in an elderly patient with vascular comorbidities is notoriously difficult. Cerebral small-vessel disease, which her MRI showed, is itself associated with late-onset psychotic symptoms, and the blood-brain barrier in older hypertensive patients may be more permeable than in younger brains. One plausible biological hypothesis is that systemically circulating bevacizumab, or the downstream reduction in VEGF signalling, could interact with vulnerable cerebral vasculature or with neurovascular regulation in a brain already strained by chronic ischaemia and polypharmacy. VEGF plays roles in vascular integrity and possibly in neuronal support, and interfering with it in a compromised brain is a hypothesis that has never been rigorously tested in this population.

The case also highlights a broader blind spot in drug safety monitoring. Intravitreal anti-VEGF injections are among the most frequent procedures in medicine, delivered in repeated monthly or bimonthly cycles, often for years. Because the treated population is overwhelmingly elderly and carries heavy burdens of cardiovascular disease, diabetes, and concurrent medication, new psychiatric symptoms in these patients are almost reflexively attributed to dementia, delirium, or psychosocial factors. A rare signal, occurring in perhaps one patient among thousands, could easily be swallowed by that background noise. Case reports such as this one function as an early-warning system, prompting pharmacovigilance databases and regulators to look for converging reports that would otherwise go unconnected.

For clinicians, the practical message is not to abandon bevacizumab, which remains a mainstay of retinal care with a strong efficacy and safety record, but to maintain diagnostic vigilance. The authors suggest that when new-onset psychosis emerges in a polymedicated geriatric patient shortly after an intravitreal anti-VEGF injection, a medication-related adverse effect deserves a place on the differential diagnosis, alongside the more familiar neuropsychiatric explanations. Structured medication reconciliation, toxicology screening, neuroimaging, and delirium assessment, all of which were performed in this case, provide the framework for distinguishing a possible drug trigger from multifactorial decompensation.

Whether the association is causal or coincidental cannot be settled by a single patient, and the authors state this plainly. What the case offers instead is a precisely documented temporal sequence, a negative workup for competing causes, a documented treatment response, and a six-month outcome including re-exposure, the raw material from which larger pharmacovigilance studies can build. As the population ages and anti-VEGF therapy expands, the incident reminds medicine that even a needle aimed at the retina can, on rare occasions, reach somewhere far more unexpected.

Subject of Research: Possible psychiatric adverse effects of intravitreal bevacizumab in an elderly patient

Article Title: Possible triggering of psychotic symptoms after repeated intravitreal bevacizumab exposure in a polymedicated geriatric patient: a case report

Article References: Gulec, B. B., Bisgin, E., Akbas, I., & Guclu, O. (2026). Possible triggering of psychotic symptoms after repeated intravitreal bevacizumab exposure in a polymedicated geriatric patient: a case report. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08567-z

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08567-z

Keywords: bevacizumab, intravitreal injection, psychosis, age-related macular degeneration, anti-VEGF therapy, adverse drug reaction, geriatric psychiatry, aripiprazole, case report, pharmacovigilance, delusions, small-vessel disease