Proin Eye Risks

Hello Summarians!

Sometimes, especially for older drugs or therapies, it can be difficult to remember that all medications have positive and negative effects. Sometimes, we find that the balance tips in favor of using a particular drug to enhance a patient’s quality of life.

There can be other instances where we are not aware of all of the effects that it may have on their body. So is the case with Proin…

If you enjoy VetSummary, please consider sending it on to a friend 😄 

Generative AI For Research

This study evaluated whether generative AI can accurately and efficiently automate case identification for retrospective veterinary research, a task that is traditionally labor-intensive and difficult due to inconsistent medical recordkeeping and lack of standardized coding in veterinary medicine. Using three distinct retrospective study scenarios—trauma-related mortality, benign ureteral obstruction treated with stents, and gallbladder mucocele cases undergoing cholecystectomy—the authors compared AI-based approaches, basic Python scripts, and expert human reviewers applying the same predefined inclusion and exclusion criteria. 

Across all applications, large language models, particularly Google Gemini 2.5 Pro and NotebookLM, demonstrated high agreement with expert evaluators, often matching or exceeding human inter-rater reliability while completing the task in a fraction of the time. In the trauma and ureteral stent datasets, Gemini closely mirrored expert inclusion decisions and showed near-perfect reproducibility across repeated runs. In the more complex gallbladder mucocele dataset, which required longitudinal data interpretation across multiple documents, AI performance remained comparable to expert consensus, though variability increased and performance differed between models. Overall, AI systems consistently outperformed traditional rule-based Python approaches, which showed strong biases toward false negatives or false positives depending on script complexity. 

Time efficiency was a major advantage of AI-based methods. While expert reviewers required roughly one to one-and-a-half hours to assess around 100 cases, AI models processed similar datasets in seconds to minutes once files were prepared, suggesting dramatic scalability for large retrospective studies. The AI systems also provided transparent justifications for inclusion or exclusion decisions, enabling easier auditing and refinement of study criteria, although output formatting varied between runs, highlighting challenges in standardization. 

The study identified important limitations and risks, including occasional missed files, variability in output structure, and the potential for hallucinations in some AI platforms, which led the authors to exclude certain models from evaluation. Performance was assessed using data from a single academic institution, limiting generalizability across diverse practice settings. Additionally, AI-assisted selection does not eliminate the need for final human verification before analysis or publication. 

Overall, the findings support the use of generative AI as a powerful tool to accelerate and scale retrospective veterinary research by automating case identification with accuracy comparable to expert reviewers. While not a replacement for human oversight, AI offers a practical solution to longstanding barriers in veterinary data analysis and holds promise for enabling larger, more efficient studies that can more rapidly translate clinical data into actionable knowledge. 

Brus, A. M., Edwards, T., Atiee, G., Dickerson, V., Ortiz, R., Mosely, S., Hernandez Torres, S. I., & Snider, E. J. (2025). Generative artificial intelligence provides accurate case selection in veterinary retrospective studies. American Journal of Veterinary Research https://doi.org/10.2460/ajvr.25.08.0295 

Bottom line — a powerful tool if you used correctly

Proin Ocular Risks

Phenylpropanolamine (PPA) is a sympathomimetic drug withdrawn from human use because of serious adverse effects, but remains widely prescribed in dogs for urinary incontinence due to its ability to increase urethral sphincter tone. Although generally considered effective and safe in canine patients, PPA has been associated with cardiovascular, neurologic, and hypertensive complications, raising concerns about its potential ocular effects. Sympathomimetic drugs can influence intraocular pressure (IOP) and tear production, but prior studies in other species have yielded conflicting results, and the effects of PPA on these parameters in dogs had not previously been evaluated. 

This study investigated the impact of PPA administration on IOP and Schirmer tear test (STT-I) values in healthy dogs. Overall, PPA induced a statistically significant increase in IOP and a small decrease in STT-I, although these changes were not considered clinically significant in the study population. Importantly, dogs with higher baseline IOPs experienced more pronounced IOP elevations and greater fluctuations during treatment, suggesting that such individuals may be more susceptible to PPA-induced ocular effects. While no clinical signs of glaucoma were observed, these IOP changes could be clinically relevant in dogs predisposed to glaucoma or those with marginal IOP control. 

The mechanism underlying the observed IOP increases remains unclear. PPA predominantly stimulates α1-adrenergic receptors, which can cause mydriasis and potentially impair aqueous humor outflow, a mechanism implicated in angle-closure glaucoma in humans. Although mydriasis was not observed clinically in this study and no anatomical abnormalities were detected on gonioscopy or ultrasound biomicroscopy, subtle or subclinical changes in aqueous outflow pathways cannot be excluded. Notably, in two dogs, increased IOP fluctuations persisted for one to two months after discontinuation of PPA, raising the possibility of longer-lasting functional effects with sympathomimetic exposure. 

Tear production showed a small, statistically significant decrease in one eye, but this change was clinically negligible and inconsistent with prior reports of increased tear production following sympathomimetic drug use. A transient episode of conjunctival hyperemia in one dog resolved quickly and could not be definitively linked to PPA administration. 

Despite the small sample size, repeated measurements allowed detection of meaningful IOP changes in individual dogs. These findings suggest that PPA can influence IOP dynamics, particularly in dogs with higher baseline pressures. Given that urinary incontinence often requires long-term therapy and that glaucoma risk increases with age, cautious use of PPA is advised in dogs with elevated IOP or known risk factors for glaucoma. Baseline and ongoing ophthalmic examinations, including IOP monitoring and periodic gonioscopy or ultrasound biomicroscopy, are recommended when prescribing PPA. Larger and longer-term studies are needed to better define the clinical significance and mechanisms of PPA-associated ocular effects. 

Jung, S., Harris, K., Ra, H., Li, J. Z., Hoard, L., Oladele, O., Janowski, B., Smith, L. H., Weng, H., & Park, S. A. (2025). Phenylpropanolamine (Proin) increases intraocular pressure and alters tear production in dogs. American Journal of Veterinary Research https://doi.org/10.2460/ajvr.25.08.0284 

Bottom line — This may change case selection for this drug.

Nerve Growth Factor and Cognitive Decline

Nerve growth factor (NGF) is a critical neurotrophin required for the development, maintenance, and survival of basal forebrain cholinergic neurons, which supply acetylcholine to key brain regions involved in learning and memory, including the hippocampus and cerebral cortex. Loss of cholinergic function is a defining pathological feature of Alzheimer’s disease (AD), and growing evidence shows that disruption of normal NGF metabolism plays a central role in this process. NGF is synthesized and released as a precursor protein, proNGF, which is normally converted into mature NGF through a tightly regulated enzymatic pathway and subsequently degraded in a controlled manner. Under healthy conditions, this system maintains a predominance of mature NGF, supporting neuronal survival and synaptic plasticity. 

In Alzheimer’s disease, this balance is progressively disrupted in a process termed NGF dysmetabolism. Two key abnormalities characterize this state: impaired conversion of proNGF to mature NGF and excessive degradation of NGF once it is produced. As a result, proNGF accumulates while NGF levels decline. This biochemical shift is compounded by changes in NGF receptor expression. NGF normally binds with high affinity to the TrkA receptor, which mediates neuroprotective and trophic effects, whereas proNGF preferentially binds to the p75 neurotrophin receptor (p75NTR), activation of which promotes neuronal apoptosis and beta-amyloid production. In AD brains, TrkA receptor expression is reduced while p75NTR expression is preserved or increased, creating a receptor environment that favors neurodegeneration. Together, reduced NGF availability and an unfavorable TrkA-to-p75NTR ratio drive progressive neuronal damage, synaptic loss, and cognitive decline. 

Importantly, NGF dysmetabolism appears early in the disease process, preceding overt cognitive impairment by years. Human postmortem studies demonstrate that alterations in NGF processing and degradation occur in individuals with increasing amyloid burden even before clinical symptoms emerge. Early stages are dominated by defective proNGF-to-NGF conversion, while accelerated NGF degradation becomes more prominent later in the disease course. These findings position NGF dysmetabolism as a potential upstream contributor to Alzheimer’s pathology rather than a late consequence of neurodegeneration. 

Strong causal evidence for the role of NGF disruption in cognitive decline comes from transgenic mouse models. Mice engineered to produce antibodies against NGF develop Alzheimer-like brain pathology, including beta-amyloid accumulation, tau pathology, loss of basal forebrain cholinergic neurons, and measurable cognitive deficits. Additional models show that blocking TrkA signaling or enhancing proNGF-p75NTR interactions is sufficient to trigger similar neurodegenerative changes. Notably, even peripheral blockade of NGF can indirectly lead to central neurodegeneration by damaging the sympathetic nervous system, disrupting the blood–brain barrier, and allowing anti-NGF antibodies to access the brain. These models collectively demonstrate that interference with NGF signaling, even outside the central nervous system, can initiate an Alzheimer-like cascade. 

Canine cognitive dysfunction represents a naturally occurring analog of human Alzheimer’s disease and shares many clinical and pathological features, including age-related cognitive decline and cholinergic neuron loss. Although NGF dysmetabolism has not yet been comprehensively studied in dogs with cognitive dysfunction, evidence of basal forebrain cholinergic degeneration in aging dogs and the strong evolutionary conservation of NGF pathways across species suggest that similar mechanisms are likely at play. Given the high prevalence of cognitive dysfunction in senior dogs, this population offers a valuable opportunity to study NGF dysmetabolism in a spontaneous disease model. Such research could improve understanding of dementia progression while opening new avenues for biomarker development and therapeutic intervention relevant to both veterinary and human medicine. 

Dewey, C. W., & Brunke, M. W. (2025). Dysmetabolism of the nerve growth factor pathway in the aging brain plays a pivotal role in cognitive decline. Journal of the American Veterinary Medical Association https://doi.org/10.2460/javma.25.09.0578 

Bottom line — Potentially could aid in early cognitive decline diagnosis.

Just putting things in perspective …