Kidney Findings Lurk Undetected in Most Cancer PET Scans, Study Reveals

Every day, thousands of cancer patients undergo positron emission tomography combined with computed tomography, a powerful imaging technique that lights up metabolically active tumors throughout the body. The scan is designed to answer one question: where is the cancer? But in the process of hunting for malignant disease, these machines capture exquisitely detailed images of every organ they pass through, including the kidneys. A new retrospective study from researchers in Türkiye now suggests that a striking proportion of kidney abnormalities visible on these oncologic scans are going unmentioned in the official reports that guide patient care, raising uncomfortable questions about what else radiologists may be missing when they focus their attention elsewhere.

The study, published in BMC Medical Imaging, was led by Seckin Bilgic and colleagues at the Department of Nuclear Medicine of Bursa Uludag University, with additional contributions from Canakkale Onsekiz Mart University and Bursa Yuksek Ihtisas Training and Research Hospital. The team set out to answer a deceptively simple question: when adults undergo fluorine-18 fluorodeoxyglucose positron emission tomography and computed tomography, known as 18F-FDG PET/CT, for cancer evaluation, how often do their kidneys show abnormalities, and how often do those abnormalities actually make it into the written report? The answer, according to their data, is that kidney findings are remarkably common and remarkably under-reported.

The researchers examined 283 consecutive adult patients who underwent oncologic 18F-FDG PET/CT at their institution between March and April 2023. Consecutive enrollment matters here, because it means the cohort was not cherry-picked; these were simply the patients who came through the door during a defined two-month window. The study received ethical approval from the Bursa Uludag University Ethics Research Committee and was conducted in accordance with the 1964 Helsinki Declaration, with informed consent obtained from all subjects. Because the design was retrospective and observational, the investigators could go back and scrutinize the images with a level of systematic attention that busy clinical workflows do not always permit.

What did that scrutiny reveal? The numbers are eye-opening. At least one cortical or pelvicalyceal renal abnormality was identified in 192 of the 283 patients, which works out to 67.8 percent of the cohort, with a 95 percent confidence interval spanning 62.2 to 73.0 percent. In other words, roughly two out of every three cancer patients scanned carried some detectable kidney abnormality on their PET/CT images. The abnormalities fell into two broad morphological categories. Cortical abnormalities, which affect the outer functional tissue of the kidney, were present in 123 patients, or 43.5 percent, and were most commonly large cortical cysts, seen in 22.3 percent of the full cohort. Pelvicalyceal abnormalities, which involve the urine-collecting system deep within the kidney, appeared in 129 patients, or 45.6 percent.

The technical distinction between these two categories is worth unpacking, because it reflects the dual nature of the information contained in a PET/CT scan. The CT component provides high-resolution anatomical detail without contrast enhancement in this setting, allowing reviewers to assess the renal cortex for cysts, masses, and structural changes. The pelvicalyceal system, meanwhile, can be evaluated for dilatation, stones, and other collecting-system pathology. The researchers even developed a stepwise visual grading algorithm, described in their supplementary material, to classify pelvicalyceal dilatation from Grade 0 to Grade 3 based on renal pelvic fullness, calyceal dilatation, and associated thinning of the renal parenchyma. Notably, they used the fused PET/CT images only for anatomical correlation and deliberately ensured that the intense urinary excretion of FDG, which naturally accumulates in the collecting system, did not influence grade assignment. That methodological care matters, because urinary FDG activity is a well-known confounder that can obscure or mimic collecting-system findings.

But prevalence alone was not the heart of the study. The investigators also wanted to know whether the kidney findings they identified on systematic review had ever been documented before. For the 192 patients with at least one renal abnormality, they reviewed available pre-index institutional records, meaning prior imaging and reports already on file at the institution, and classified each patient’s findings as previously documented, mixed, or incidentally identified. The result was stark: only 20 patients, or 10.4 percent, had findings that were entirely previously documented. Another 31 patients, or 16.1 percent, had a mixed picture, with some findings known and some new. The remaining 141 patients, a full 73.4 percent, had kidney abnormalities that were entirely new discoveries, never before captured in the institutional record. Taken together, 172 of the 192 affected patients, or 89.6 percent, had at least one renal abnormality that had never been documented anywhere in their available records.

The most sobering comparison, however, came when the researchers turned to the original clinical PET/CT reports, the documents written by the interpreting physicians at the time of the scan and sent to the treating oncologists. Of the 192 patients whose scans showed renal abnormalities on standardized retrospective review, only 58, or 30.2 percent, had at least one of those abnormalities mentioned in the original report. The other 134 patients, 69.8 percent of those with findings, had no renal abnormality documented at all. The gap between what a systematic review could see and what the clinical report conveyed is the central finding of the paper, and it is difficult to read those numbers without wondering how many clinically meaningful findings slip through the cracks in routine practice.

Why does this happen? The authors point to the inherent structure of oncologic imaging interpretation. A PET/CT study ordered for cancer staging or surveillance generates hundreds of images, and the interpreting physician’s primary obligation is to the clinical question at hand: tumor location, metabolic activity, treatment response, and new metastatic disease. Kidney cysts and mild collecting-system dilatation are, in the vast majority of cases, benign and clinically silent, and flagging every one of them in every report would risk burying the critical oncologic information in a flood of incidental detail. This is the classic tension of incidental findings in modern imaging: as scanners have grown more sensitive and datasets more voluminous, the sheer volume of non-target abnormalities has outpaced any practical reporting strategy. Yet the study’s data suggest the pendulum may have swung too far, with nearly 90 percent of affected patients carrying at least one renal finding absent from all available documentation.

The clinical stakes are not trivial. Some incidental renal abnormalities, particularly complex cysts and solid-appearing cortical lesions, can represent early renal cell carcinoma, a malignancy that is highly curable when caught at a small size but far more dangerous when it grows undetected. Hydronephrosis and collecting-system dilatation can signal obstructing stones or tumors that merit follow-up. Chronic kidney disease, too, often announces itself first through subtle structural changes visible on cross-sectional imaging. Whether the abnormalities catalogued in this study would have changed outcomes for any individual patient is a question the retrospective design cannot answer, and the authors are appropriately measured on this point. They explicitly state that further studies should determine whether more systematic recognition and reporting of these findings influences diagnostic evaluation, clinical decision-making, and patient management.

Still, the study lands at a moment when the imaging community is actively grappling with how to handle the incidentalome, the ever-growing list of unexpected findings generated by modern scans. Structured reporting templates, standardized checklists for organ systems outside the region of interest, and artificial intelligence tools that pre-screen images for non-target abnormalities are all under active investigation as potential solutions. This study provides a concrete baseline for one organ system in one high-volume imaging context, and its message is clear: kidney abnormalities are present in more than two-thirds of oncologic PET/CT patients, yet nearly 70 percent of affected patients leave the encounter with none of those findings on paper. Closing that gap, without drowning the oncologic signal in incidental noise, may be one of the defining reporting challenges of the PET/CT era. The authors received no specific funding for the work and declared no competing interests, and the article is published open access under a Creative Commons license, making the full data and methods available to any research group hoping to replicate or extend the findings.

Subject of Research: Prevalence and under-reporting of incidental renal abnormalities detected on oncologic 18F-FDG PET/CT imaging

Article Title: Incidental renal abnormalities on 18F-FDG PET/CT in oncologic patients: prevalence and reporting practices

Article References: Bilgic, S., Koroglu, I., Akovalı, B., Sahin, H. G., & Alper, E. (2026). Incidental renal abnormalities on 18F-FDG PET/CT in oncologic patients: prevalence and reporting practices. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02859-2

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02859-2

Keywords: 18F-FDG PET/CT, incidental findings, renal abnormalities, oncologic imaging, kidney cysts, pelvicalyceal dilatation, reporting practices, nuclear medicine, radiology, renal cell carcinoma, patient safety, medical imaging