This Week in Urology — May 28, 2026

Welcome to This Week in Urology. This week we're covering 10 notable papers spanning the rise of artificial intelligence in surgical and diagnostic urology, a critical look at safety and complications in BPH management, and innovations in endourologic devices and techniques. Let's dive in.

The AI Revolution in Urology This week features a major focus on artificial intelligence, with four papers exploring its role from the operating room to the clinic and even in assessing medical knowledge. Two papers in particular highlight AI's growing utility in robot-assisted radical prostatectomy.

First, in *BJU International*, investigators developed an AI model to automatically recognize the seven distinct phases of a robotic prostatectomy [2].

The Study Using 72 complete RARP videos, they trained a hierarchical vision transformer model called Hiera. The model was then tested on its ability to correctly identify phases like ‘intrafascial dissection’ and ‘urethral anastomosis’ from surgical video footage.

Results The model performed well, achieving an overall accuracy of 91%. It was particularly good at identifying urethral anastomosis, with a performance score of 0.96. When deployed in a real-world setting, the system successfully processed live video streams and provided continuous, accurate phase predictions without any technical interruptions. This technology holds promise for standardizing surgical training, providing objective feedback, and enabling automated quality review.

Building on the theme of AI in RARP, a study in the *International Journal of Urology* used AI to provide a more objective measure of surgical quality [4]. Investigators aimed to see if an AI's assessment of nerve-sparing, combined with preserved urethral length, could better predict postoperative urinary continence.

Methods They analyzed intraoperative images from 195 patients. A deep learning model evaluated whether the neurovascular bundle was truly preserved, and this was compared to the surgeon’s own assessment. Patients were then grouped based on whether they had both AI-confirmed nerve preservation and a preserved urethral length of at least 26 millimeters.

Conclusions The AI model disagreed with the surgeon’s assessment in nearly 31% of cases, suggesting that subjective evaluation can be unreliable. More importantly, patients who met both the AI nerve-sparing and urethral length criteria had significantly faster recovery of urinary continence. This combined factor was an independent predictor of continence at 60, 90, and 365 days post-surgery, nearly doubling the likelihood of continence recovery at 60 days. The findings suggest that AI can offer a valuable, objective tool for assessing surgical quality and predicting functional outcomes.

Beyond surgery, AI is also being applied to diagnostics. In the *Journal of Endourology*, researchers developed a deep learning system to support cystoscopy [3]. Recognizing that diagnostic accuracy in cystoscopy is highly dependent on operator experience, they trained models on over 9,000 images of bladder tumors, cystitis, and postoperative scars.

Results They developed two models: one for classifying lesions and another for segmenting the boundaries of bladder tumors. The classification model demonstrated high performance with an area under the curve of 0.872, while the segmentation model achieved a Dice similarity coefficient of 0.821, accurately delineating tumor margins. The study concludes that AI-powered analysis can enhance both the accuracy and efficiency of cystoscopy, helping clinicians identify subtle lesions and reducing interobserver variability.

Finally, moving from visual data to text-based reasoning, a study in the *World Journal of Urology* tested how well different versions of ChatGPT could handle novel, high-level medical questions [6]. Researchers used 120 brand-new questions from the 2024 Taiwan Urology Board Examination, all created after the models' training data cutoff. The latest model, o1-preview, achieved 66.7% accuracy, significantly outperforming the more widely known GPT-4o, which scored 55.8%. Interestingly, the newer model performed better on high-complexity questions than low-complexity ones. On challenging surgical anatomy questions, it scored 80%, while previous models scored zero. This suggests that newer, reasoning-oriented AI models may have an advantage in specialized medical domains, though it came at the cost of slightly longer response times.

Re-evaluating Safety in BPH and Prostate Cancer Treatment Next, we turn to patient safety, with three studies examining complications from common procedures for BPH and prostate cancer. While minimally invasive therapies for BPH are popular due to their favorable safety profiles, a report in the *World Journal of Urology* serves as a sobering reminder of potential severe adverse events [1].

The Study This was a retrospective analysis from high-volume international centers, which collected data on 10 cases of life-threatening, Clavien-Dindo grade 4 or higher complications following water vapor thermal therapy, or WVTT. These events occurred despite negative preoperative urine cultures and standard antibiotic prophylaxis.

Findings Most complications were infectious, including septic shock with intraprostatic abscesses, often requiring ICU admission and vasopressors. Some fatal events due to sepsis from multidrug-resistant pathogens were recorded. Other severe events included destructive orchitis requiring orchiectomy and thermal injury to the sigmoid colon. The authors note that many of these patients had significant comorbidities or prior catheterization. They conclude that while WVTT is generally safe, rare but devastating complications can occur. The paper underscores the need for strict infection control, careful patient selection, and comprehensive counseling, particularly in frail patients.

Staying with BPH surgery, a prospective study also in the *World Journal of Urology* highlights a different, often overlooked complication, this time with bipolar transurethral resection of the prostate [8]. While bipolar TURP avoids the hyponatremia of traditional TUR syndrome by using 0.9% saline for irrigation, this introduces the risk of hyperchloremia.

Results In a cohort of 614 patients, hyperchloremia—defined as a postoperative chloride level of 110 mmol/L or higher—occurred in 11.4% of cases. The volume of irrigation fluid absorbed was directly correlated with the rise in plasma chloride. Patients with hyperchloremia had a dramatically higher rate of postoperative acute kidney injury, or AKI: 30% versus just 5% in the non-hyperchloremia group. The severity of AKI was also greater. Hyperchloremia was found to be an independent risk factor for a composite of major adverse outcomes, including ICU admission and renal replacement therapy. The authors strongly advocate for perioperative chloride monitoring and suggest that balanced irrigation fluids should be investigated to improve patient safety during bipolar TURP.

Rounding out this section, a study in *Prostate Cancer and Prostatic Diseases* provides crucial, practice-changing guidance on the timing of BPH treatment for men also undergoing radiation for localized prostate cancer [5]. Using the large TriNetX database, researchers compared genitourinary toxicity in men who had a bladder outlet obstruction procedure, or BOOP, *after* radiation therapy versus those who had it *before*.

Conclusions After propensity score matching, the results were stark. Men who had their BPH treated after radiation had a more than four-fold increased risk of stress urinary incontinence, a more than two-fold increased risk of urge incontinence and urinary retention, and a roughly three-fold increased risk of urethral strictures. When stratified by procedure type, resective surgeries like TURP after radiation were associated with the highest toxicity. The clinical implication is clear: in men with both localized prostate cancer and symptomatic BPH, urologists should evaluate and consider treating the BPH *before* the patient undergoes radiation to significantly limit long-term genitourinary toxicity.

Innovations in Endourology and Pediatric Reconstruction Our final theme covers new devices and comparative data for endourologic and pediatric procedures. First, from the *Journal of Endourology*, a simple but clever innovation aims to make percutaneous renal access for PCNL faster and safer [9]. Investigators evaluated a novel 'bubble needle', which is a standard nephrostomy needle with an integrated two-dimensional bubble level.

The Study The device is designed to maintain perfect anterior-posterior alignment after just a single fluoroscopic image. In a phantom model, 21 participants ranging from students to attending surgeons performed punctures using the bubble needle and two traditional fluoroscopic techniques. The study measured radiation dose, procedure time, and accuracy.

Results The bubble needle achieved comparable accuracy to traditional techniques but dramatically reduced radiation exposure, with a 74% reduction compared to the bull's-eye technique. It also significantly reduced procedure time. These benefits were consistent across all experience levels, suggesting the device is intuitive and could streamline a challenging part of PCNL while improving radiation safety for both the patient and the surgical team.

Also in the *Journal of Endourology*, a prospective study presents early data on a novel metallic ureteral stent for managing complex ureteral strictures [10]. This new device, a thermally expanding nickel-titanium alloy stent, was implanted in 21 patients with strictures from various causes, including stones, radiation, and iatrogenic injury.

Findings Over a median follow-up of 5 months, the primary goal of effective decompression was achieved in 87% of cases. Patients showed significant improvements in renal function, with decreased serum creatinine and increased eGFR, as well as a marked reduction in hydronephrosis. The main complication was stent migration, which occurred in about 17% of cases but was manageable endoscopically. These initial results suggest the stent is a promising minimally invasive option for a challenging clinical problem.

Finally, a paper in *The Journal of Urology* tackles the classic pediatric urology dilemma of how to surgically manage vesicoureteral reflux [7]. The study combined a two-center retrospective cohort with a systematic review and meta-analysis to compare dextranomer/hyaluronic acid, or Dx/HA, injection with open ureteral reimplantation.

Results The institutional data highlighted a clear trade-off. Dx/HA injection was associated with much shorter operative times (21 vs 140 minutes) and length of stay (0 vs 1 day). However, it also came with a significantly higher rate of postoperative urinary tract infections—nearly 20% versus less than 4% for reimplantation. The meta-analysis of 15 studies confirmed that reimplantation was superior for achieving radiographic resolution and had a much lower likelihood of requiring reoperation. Interestingly, the meta-analysis found no significant difference in the rates of febrile or recurrent UTIs between the two approaches, though heterogeneity was high. The paper concludes that neither approach is definitively superior, and the choice involves a trade-off between the durability of reimplantation and the significant perioperative advantages of Dx/HA injection, supporting an individualized, risk-adapted approach based on shared decision-making.

Editor's Pick If you only have time for one paper this week, make it the study in *Prostate Cancer and Prostatic Diseases* on the timing of bladder outlet procedures relative to radiation therapy [5]. It provides compelling evidence that urologists should evaluate and consider treating significant BPH *before* a patient undergoes radiation for prostate cancer to substantially reduce the risk of long-term genitourinary toxicity.

Clinical Bottom Line Here are the key takeaways from this week in Urology:

First, for patients with localized prostate cancer and symptomatic BPH, consider treating the BPH with a bladder outlet procedure *before* they undergo radiation therapy. This can significantly lower their risk of future incontinence, retention, and strictures.

Second, artificial intelligence is becoming a practical tool. It can now analyze surgical video in real-time to recognize operative steps, objectively assess surgical quality to predict outcomes, and improve the accuracy of diagnostic cystoscopy.

Third, even our safest BPH therapies have underappreciated risks. Minimally invasive treatments like water vapor thermal therapy carry a rare but real risk of severe septic complications, while bipolar TURP is associated with hyperchloremia and subsequent acute kidney injury. Vigilance, careful patient selection, and perioperative monitoring are critical.

Fourth, in managing pediatric VUR, ureteral reimplantation offers better radiographic success and lower reoperation rates, but endoscopic injection with Dx/HA provides major perioperative advantages. This highlights a crucial trade-off that requires shared decision-making with families.

Finally, simple innovations in endourology, like an air-bubble guided needle for PCNL, can lead to significant reductions in radiation exposure and procedure time without sacrificing accuracy.

That's your roundup for This Week in Urology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.