NSAIDs Can Cause Bacterial Resistance?

Hello, Summarians!

As science advances, we gain a deeper understanding of how the world interacts. Every action has some type of reaction, even if we are not able to fully understand how it works.

One of these studies demonstrates how this might occur in relation to combining common drugs with a typical antibiotic to treat a urinary tract infection.

Thank you for your continuing support. If you have found VetSummary informative, please share it with a friend.

Cheers!

AI To Detect Sudden Cardiac Death

Cardiac arrhythmias are a major concern in athletic species because they can impair performance and sometimes cause collapse or sudden death. Atrial fibrillation (AF) is the most common clinically important arrhythmia in both horses and humans, occurring in up to 4.9% of racehorses. Exercise-associated sudden death is much more common in horses than in human athletes, with more than half of these fatalities suspected to be cardiac in origin. Necropsy findings often fail to reveal structural abnormalities, pointing toward arrhythmogenic causes. Electrocardiography (ECG) remains the diagnostic gold standard, but it is time-consuming, subject to interpretation errors, and difficult to use during exercise, where many arrhythmias emerge. 

Artificial intelligence (AI) has shown great promise in human cardiology by automating ECG interpretation, reducing inter-observer variability, and detecting subtle or transient arrhythmias. Convolutional neural networks (CNNs) and other deep learning (DL) methods are particularly effective, functioning in ways similar to expert cardiologists by identifying both simple and complex ECG features. These tools can enhance classification, predict arrhythmic risk, and support continuous monitoring. However, their application in equine cardiology remains limited due to species-specific differences in cardiac conduction and ECG morphology, as well as smaller, less diverse datasets. 

Most of the reviewed literature has focused on humans, with veterinary studies representing a small fraction. Human studies typically benefit from massive datasets, often exceeding 100,000 ECGs, while horse studies are constrained to far fewer recordings, sometimes fewer than 30,000 heartbeats. Despite this, promising results in horses include the prediction of paroxysmal AF with good accuracy and successful classification of ECG complexes, especially when transfer learning from human datasets was applied. These approaches improved performance, addressing data scarcity and the challenges posed by equine-specific ECG features such as dominant S waves and overlapping P and T waves during exercise. 

Performance metrics reported across studies varied, with sensitivities and specificities generally ranging from about 79% to 100%. However, inconsistencies in reporting and the absence of standardized guidelines make comparisons difficult. Most AI models appeared geared toward diagnostic use, prioritizing specificity over sensitivity, although higher sensitivity would be preferable in screening athletic populations where missing a potentially life-threatening arrhythmia carries high risk. Few studies validated models in real-world or exercise conditions, even though arrhythmias in athletes frequently occur during or immediately after exertion. This gap significantly limits the current applicability of AI tools in sports cardiology. 

The review highlights that future work should focus on developing species-specific AI models tailored to horses and dogs, collecting ECG data during exercise, and using transfer learning to overcome limited veterinary datasets. Standardized reporting, rigorous methodological designs, and larger studies are needed to enhance comparability and reliability. Integration of AI-enhanced ECG systems with wearable technologies offers exciting opportunities for real-time monitoring and more accurate risk assessment in both human and veterinary sports medicine. While still underdeveloped in equine cardiology, AI-assisted ECG analysis holds considerable potential to improve arrhythmia detection, risk stratification, and overall athlete safety. 

A. Kapusniak, N. M. Lara, P. L. Hitchens, S. Bailey, L. Nath, and S. Franklin, “ Use of Artificial Intelligence to Detect Cardiac Rhythm Disturbances in Athletes: A Scoping Review,” Journal of Veterinary Internal Medicine 39, no. 6 (2025): e70257, https://doi.org/10.1111/jvim.70257. 

Bottom line — Need to focus on species-specific AI, but it has potential.

Antimicrobial Resistance and NSAIDs

The development and spread of antimicrobial resistance (AMR) is a major public health threat, with nearly 5 million deaths associated with it in 2019. While the overuse of antibiotics in medicine and agriculture is the primary driver, recent work has shown that non-antibiotic medications (NAMs) such as statins, NSAIDs, diuretics, and others can also contribute to AMR. Over 200 widely used drugs have demonstrated antibacterial-like activity on gut bacteria, and since NAMs make up the majority of the global pharmaceutical market, their role in resistance development is concerning, particularly in older populations that are heavily medicated and frequently prescribed antibiotics in residential aged care facilities (RACFs). 

This study investigated nine NAMs commonly used by older adults—acetaminophen, ibuprofen, diclofenac, furosemide, atorvastatin calcium, metformin, pseudoephedrine, temazepam, and tramadol—alongside ciprofloxacin, a fluoroquinolone commonly used to treat urinary tract infections. Results showed that ibuprofen and acetaminophen significantly enhanced antibiotic resistance in E. coli at gut-relevant concentrations, with diclofenac and furosemide also increasing mutation frequency. Other drugs like temazepam, tramadol, and pseudoephedrine had minimal impact. Notably, ibuprofen and acetaminophen in combination did not increase mutation frequency beyond ibuprofen alone, but the resulting mutants exhibited higher levels of ciprofloxacin resistance. 

Exposure to NAMs with ciprofloxacin led to increased resistance to multiple antibiotics, with evidence pointing to the overexpression of the AcrAB-TolC efflux pump, a key mechanism that enables bacterial survival under antibiotic stress. Whole genome sequencing revealed mutations in efflux pump regulators such as marR, acrR, and soxR, as well as in quinolone resistance determinant regions (e.g., gyrA), which together conferred high-level resistance. Some mutants accumulated multiple mutations, further boosting resistance across antibiotic classes. 

Although NAMs alone did not always provide sufficient selective pressure, their combination with ciprofloxacin clearly promoted cross-resistance. This effect mirrors findings with other NAMs like antidepressants, which can synergize with antibiotics to induce resistance. The study emphasizes that polypharmacy, common in RACFs, not only complicates antibiotic prescribing but also creates conditions favorable to resistance development. Furthermore, some NAMs, such as proton pump inhibitors, can disrupt the gut microbiome and increase risks of infections like Clostridium difficile, especially when combined with antibiotics. 

Overall, the findings highlight that certain widely used non-antibiotic medications—particularly acetaminophen, ibuprofen, diclofenac, and furosemide—can promote antibiotic resistance when combined with ciprofloxacin. This underscores the need to reconsider prescribing practices in polypharmacy settings, especially for older populations, and to evaluate how drug combinations may inadvertently accelerate the spread of AMR. 

Chen, H., Sapula, S.A., Turnidge, J. et al. The effect of commonly used non-antibiotic medications on antimicrobial resistance development in Escherichia coli. npj Antimicrob Resist 3, 73 (2025). https://doi.org/10.1038/s44259-025-00144-w 

Bottom line — Wow! Polypharmacy rears its ugly head.

Bravecto for Horses

Ectoparasites such as ticks, mites, lice, and mosquitoes cause irritation and disease in horses, with tick-borne conditions like piroplasmosis and Lyme borreliosis being of particular concern. Current treatments, mainly topical pyrethrins or oral ivermectin, require frequent use and face resistance issues. Fluralaner, an isoxazoline ectoparasiticide approved for dogs and cats, offers a longer treatment interval of 4–12 weeks and has shown efficacy in various species. Its pharmacokinetic profile, however, differs across species, making species-specific studies necessary for horses. 

In this study, horses were given a single oral dose of either 10 or 25 mg/kg fluralaner. All horses remained clinically healthy over a three-month monitoring period, with no adverse effects attributable to the drug, aside from one transient fever in the lower-dose group considered unrelated. Plasma concentrations peaked at 12 hours, shorter than the 24 hours seen in dogs and bears, possibly due to compounding methods that altered absorption. Maximal concentrations and total drug exposure were lower than in dogs or bears, and elimination half-life in horses was shorter (3–6 days) compared to dogs (12–15 days), though similar to bears. This suggests differences in absorption, metabolism, and enterohepatic recirculation between species. 

Despite lower systemic levels, fluralaner remained detectable in plasma for at least 28 days, and skin concentrations were measurable throughout an 80-day period. Plasma concentrations exceeded lethal thresholds for ticks, mosquitoes, and Culicoides for 1–3 weeks, though they were below levels effective against some mites, indicating potential need for higher or repeated dosing for full ectoparasite control. The safety profile was consistent with prior studies in dogs and with limited prior equine reports. 

The study had limitations, including a small sample size, healthy subject selection, fasting administration that may have reduced bioavailability, and a lack of tissue extraction controls. Cost and dosing frequency may also pose challenges for practical use. Nonetheless, these findings support fluralaner as a safe, potentially effective ectoparasiticide in horses, particularly against ticks and mosquitoes, with further studies needed to optimize dosing strategies and confirm efficacy against other parasites. 

Morgan, J. M., Gentille, S. R., Goyette, F. D., Lehman, M. L., Boss, A. L., Cassano, J. M., Knych, H. K., & White, S. D. (2025). Pharmacokinetics and preliminary safety of single-dose oral fluralaner at 10 and 25 mg/kg in healthy horses revealed no adverse reactions. American Journal of Veterinary Research https://doi.org/10.2460/ajvr.25.06.0200 

Bottom line — Could be very effective.

Just putting things in perspective …