A non-contrast CT scan is the most accurate imaging test for kidney stones, but it is not perfect. Reported sensitivity generally falls between 86% and 98%, meaning a small percentage of stones go undetected even under ideal conditions. Whether a particular stone slips through depends on a combination of factors: how large the stone is, what it is made of, how the scanner was set up, and the body size of the person being scanned. Understanding these gaps matters, because a missed stone can mean ongoing pain, a delayed procedure, or a wrong diagnosis entirely.
How Often CT Scans Miss Kidney Stones
Non-contrast CT is widely considered the reference standard for detecting urinary stones. A systematic review comparing CT, X-ray, and ultrasound found that non-contrast CT achieves sensitivity between 86% and 98%, with specificity approaching 100%.1The Indonesian Journal of General Medicine. Comparison of CT Scan, X-Ray and Ultrasound in Sensitivity and Specificity for Identifying Kidney Stones in Clinical Settings? : A Comprehensive Systematic Review That sounds close to flawless, but the gap between 86% and 100% is clinically meaningful. In a busy emergency department seeing dozens of patients with flank pain each week, even a few percent of missed stones adds up.
Older reports using CT findings alone often quoted sensitivity at 95% to 100%. But when researchers compared CT results against what surgeons actually found during endoscopy, the numbers were less rosy.2PubMed. Sensitivity of Noncontrast Computed Tomography for Small Renal Calculi With Endoscopy as the Gold Standard The discrepancy makes sense: CT images are only as good as the conditions under which they were obtained and the stones they are trying to show.
Small Stones Are the Ones Most Likely to Be Missed
Stone size is the single biggest predictor of whether a CT will catch a kidney stone. Stones larger than about 3 to 4 millimeters are reliably detected in nearly every study, but detection falls off sharply below 2 to 3 millimeters. One study examining how scanner slice width affected detection found that the rate of picking up 2 mm stones improved from roughly 53% on thick (5 mm) slices to about 79% on the thinnest (1.25 mm) slices.3PubMed. Renal stone detection using unenhanced multidetector row computerized tomography–does section width matter? For stones bigger than 2 mm, the detection rate was roughly the same regardless of slice thickness.
This has a practical consequence. If you have a tiny stone that is causing symptoms but sitting in a part of the urinary tract where it is hard to distinguish from surrounding tissue, a thick-slice scan might not show it at all. A follow-up scan with thinner slices, or simply the passage of time as the stone grows or moves, can reveal what was invisible the first time around.
How Slice Thickness Changes the Picture
When a CT scanner takes images, it captures the body in a series of cross-sectional slices. Thicker slices cover more anatomy per image but blur small objects. Thinner slices show finer detail at the cost of more image noise and slightly longer reconstruction time. A study evaluating 1 mm, 3 mm, and 5 mm slices found that stone detection was highest at 1 mm (about 82%), dropping to roughly 75% at 3 mm and 63% at 5 mm.4Journal of Health, Wellness and Community Research. Effect of Slice Thickness on the Diagnostic Accuracy of Non-Contrast CT KUB for Renal Stone Detection Good stone visibility fell even more steeply: about 73% of stones were clearly seen on 1 mm slices versus 32% on 5 mm slices, and radiologists reported high diagnostic confidence on only about 12% of 5 mm scans compared with 57% on 1 mm scans.
Not every hospital defaults to 1 mm slices. Emergency departments sometimes use thicker slices to speed up scanning and lower radiation, especially when the clinical question is broad (ruling out appendicitis, an aneurysm, and stones all at once). If your CT was done at 5 mm slice thickness and nothing was found but your symptoms strongly suggest a stone, a repeat scan reconstructed at thinner slices could make a difference.
Some Stones Are Nearly Invisible on CT
Most kidney stones contain calcium, which lights up brightly on CT. But certain stone types have little or no calcium, making them hard or even impossible to see. The best-known example involves indinavir, a protease inhibitor used to treat HIV. Stones made of indinavir are radiolucent, meaning they do not absorb X-rays the way calcium does. In one series of patients with indinavir-related stones, the stones were either radiolucent or outright invisible on CT in every case. Doctors diagnosed the stones based on indirect signs like ureteral obstruction and surrounding tissue inflammation rather than seeing the stone itself.5PubMed. Urolithiasis associated with protease inhibitors Earlier research had already established that these calculi do not show up on plain radiographs either.6PubMed. Imaging characteristics of indinavir calculi
Matrix stones are another rare type that can fool a CT scan. These stones consist almost entirely of protein rather than mineral, giving them a density close to that of soft tissue. A case report documented a patient with type 1 diabetes and chronic kidney disease whose pure matrix stones were completely invisible on CT.7PubMed Central. Radiolucent Pure Matrix Stones on Computed Tomography Scan, Arising in Patient with Type I Diabetes and Chronic Kidney Disease: A Case Report Pure uric acid stones can also be relatively low-density, although they usually still appear on a well-done non-contrast CT. The point is that not all stones behave the same way under the scanner, and stone composition is one of the less obvious reasons a CT can come back negative in someone who truly has a stone.
Low-Dose Protocols and the Trade-Off With Detection
Radiation exposure is a legitimate concern, particularly for younger patients and people who need repeated scans for recurrent stones. This has driven the development of low-dose and reduced-dose CT protocols that cut the radiation to a fraction of a standard scan. In one study, a reduced-dose protocol had a sensitivity of about 90% for any ureteral stone and 100% for stones larger than 5 mm, while reducing the radiation dose from roughly 22 milligray to about 3 milligray.8PubMed Central. Accuracy of Reduced-Dose Computed Tomography for Ureteral Stones in Emergency Department Patients That is an enormous radiation savings, with the trade-off mostly concentrated among the smallest stones.
Another study found that in people with a body mass index under 30, low-dose CT was about 95% sensitive for ureteral stones overall and 100% sensitive for stones larger than 3 mm. Below 3 mm, sensitivity dipped to 86%.9PubMed. Low-dose versus standard-dose CT protocol in patients with clinically suspected renal colic For most clinical decisions, a stone under 3 mm is one that will likely pass on its own, so missing it may not change management. But it does mean that a negative low-dose scan in someone with ongoing pain does not completely rule out a tiny stone.
Body Size Affects Image Quality
In larger patients, X-ray photons have more tissue to travel through, which degrades the image and increases noise. This is especially relevant for low-dose protocols, which already use fewer photons. A study of obese patients (average BMI around 29) undergoing low-dose CT found that all scans were diagnostic when processed with advanced image reconstruction algorithms, but about 41% were rated non-diagnostic when reconstructed with the older, standard method.10PubMed Central. Low-dose computed tomography of urolithiasis in obese patients: a feasibility study to evaluate image reconstruction algorithms In other words, the scanner hardware matters, but so does the software used to build the images. A facility using modern iterative reconstruction can achieve diagnostic-quality images in larger patients at lower doses, while one still relying on older algorithms might produce noisier scans that obscure small stones.
What Happens When the CT Was Done With Contrast
The standard CT for kidney stones is a non-contrast scan, because the bright white appearance of calcium-containing stones stands out best against the dark background of the kidney and ureter when no contrast dye is present. But sometimes a patient gets a contrast-enhanced CT for a different reason, such as evaluating abdominal pain that could be anything from pancreatitis to a tumor, and the radiologist is then asked whether there are also any stones.
Contrast dye collects in the kidneys and ureters, raising the density of the surrounding urine. This can camouflage stones whose density is similar to the opacified urine. One study looking at contrast-enhanced portal venous phase scans found that overall sensitivity for stones was about 81%, considerably lower than the 95%-plus expected from a non-contrast study. For stones 3 mm or larger, sensitivity rose to about 95%, and for stones 4 mm and up it was essentially 99%.11PubMed Central. Renal stones on portal venous phase contrast-enhanced CT: does intravenous contrast interfere with detection? So a contrast CT can still catch most clinically significant stones, but it is not the right tool if small stones are the specific concern.
Phleboliths and Other Look-Alikes
A CT scan can also get the answer wrong in the other direction: calling something a stone when it is not one. Phleboliths are tiny calcified blood clots that sit in pelvic veins. They show up on CT as bright white dots in roughly the same neighborhood as the lower ureter, which is exactly where a stone would lodge on its way to the bladder. Studies have found that phleboliths appear on CT at least as often as actual ureteral stones.12PubMed. Is This Your Stone? Distinguishing Phleboliths and Nephroliths on Imaging in the Emergency Department Setting
Radiologists use several clues to tell them apart. A true ureteral stone tends to have a “soft-tissue rim sign,” a thin ring of swollen ureteral wall around it, which phleboliths lack. Phleboliths, on the other hand, often have a “comet sign,” an eccentric tail of soft tissue representing the rest of the vein, and sometimes a characteristic central lucency or bifid density peak. In one study, 76% of ureteral stones showed the rim sign, while no phleboliths did, and 21% of phleboliths had the comet sign, while no true stones did.13PubMed. Unenhanced helical CT criteria to differentiate distal ureteral calculi from pelvic phleboliths Despite these distinguishing features, the overlap can still cause confusion, particularly on quick emergency reads.
When a CT Shows No Stone but You Still Have Pain
A negative CT in someone with classic flank pain does not always mean “no stone.” It might mean the stone already passed before the scan, or that it was too small or too low-density to see. But it can also mean the pain is coming from something else entirely. In one large review of CT scans ordered for suspected stones, a ureteral stone explained the symptoms in about 48% of cases, no cause was found in roughly 43%, and a non-stone diagnosis accounted for about 9%. Among those non-stone diagnoses, about 6% were considered acutely important.14PubMed. Prevalence and clinical importance of alternative causes of symptoms using a renal colic computed tomography protocol in patients with flank or back pain and absence of pyuria
A separate study found that in about 14% of patients presenting with acute flank pain, the actual cause was something other than a ureteral stone. CT identified most of these alternatives but missed a handful, including a complicated ovarian cyst, pyelonephritis, and pleuritis.15PubMed. Acute flank pain: comparison of unenhanced helical CT and ultrasonography in detecting causes other than ureterolithiasis This is worth keeping in mind: a non-contrast CT protocol that is optimized for stone detection can sometimes underperform when the real diagnosis is an infection, a vascular problem, or a gynecological condition, because those often benefit from contrast-enhanced imaging.
Motion Artifacts and Size Distortion
Even when a stone is detected, the CT image may not represent it faithfully. Patient movement during the scan, including involuntary movement from breathing or pain, can create motion artifacts that warp the appearance of a stone. A case report described a stone that appeared significantly larger on CT than it actually was, because motion artifact stretched its image.16PubMed Central. Deceived by a CT Scan: The Case of the Misrepresented Stone Size This kind of distortion can influence treatment decisions, since stone size is one of the main factors in choosing between watchful waiting and surgical intervention. A stone that looks like 7 mm on a motion-blurred image but is actually 4 mm might lead to an unnecessary procedure.
Stones That Come and Go on Imaging
Kidney stones are not always static. A stone sitting quietly in the kidney may cause no symptoms at all, and a stone that begins to move into the ureter can trigger severe pain that sends someone to the emergency department. Timing matters: if the stone passes into the bladder or out of the body between the onset of symptoms and the scan, the CT will be negative even though the stone was real. A prospective cohort study tracking patients with known asymptomatic stones found that about half showed radiographic evidence of stone passage within five years, and symptoms accompanied that passage only about half the time.17PubMed Central. Symptomatic and Radiographic Manifestations of Kidney Stone Recurrence and Their Prediction by Risk Factors: A Prospective Cohort Study Stones can move quickly, and a scan taken hours after peak pain may simply arrive too late to catch one.
Unusual Anatomy Can Hide Stones
Calyceal diverticula are small pouches that branch off the kidney’s collecting system. They are uncommon, occurring in roughly 2 to 5 per 1,000 imaging exams, but stones form in them at a disproportionately high rate.18PubMed Central. The imaging appearances of calyceal diverticula complicated by uroliathasis Because these pouches sit in unusual locations within the kidney, a stone tucked inside one can be misidentified as a vascular calcification or a parenchymal abnormality, especially if the radiologist is not looking for a diverticulum. Knowing you have this anatomy ahead of time can help guide the radiologist’s search.
Dual-Energy CT and Stone Composition
Standard CT can tell you whether a stone is there and roughly how big it is, but it gives limited information about what the stone is made of. Dual-energy CT (DECT) uses two different X-ray energy levels simultaneously, allowing the scanner to differentiate materials by how they absorb energy at each level. This is particularly useful for distinguishing uric acid stones from calcium-based stones, because the treatment differs: uric acid stones can sometimes be dissolved with medication, while calcium stones generally cannot.
One study found that dual-energy CT correctly classified 225 of 227 kidney stones, achieving an accuracy of about 99% for distinguishing uric acid from non-uric acid stones, even for stones as small as 3 mm.19PubMed Central. Dual-energy CT kidney stone characterization—can diagnostic accuracy be achieved at low radiation dose? Separating non-uric acid subtypes from each other (calcium oxalate versus calcium phosphate versus struvite, for instance) is harder, since these materials have more similar density profiles. Research into using spectral analysis and patient-size-adjusted thresholds is ongoing, with promising early results for finer classification.20African Journal of Urology. Renal calculus composition analysis using dual-energy CT: a prospective observational study 21PubMed Central. A Generalizable Framework for Kidney Stone Composition Characterization Using Dual-Energy CT
How Ultrasound Compares
Ultrasound is sometimes used as the first imaging test for suspected kidney stones, especially in pregnant patients, children, and situations where radiation exposure is a concern. Its strengths are zero radiation, wide availability, and low cost. Its weakness is accuracy. A literature review found that ultrasound has a pooled sensitivity of about 45% for both renal and ureteral stones, meaning it misses more than half.22PubMed. Limitations to ultrasound in the detection and measurement of urinary tract calculi Specificity is decent (around 88% to 94%), so if ultrasound says a stone is there, it usually is. But a negative ultrasound does little to rule one out. For surveillance of known stone disease, ultrasound performance did not significantly vary by stone location or the time gap between scans, but it remained substantially less sensitive than CT overall.23PubMed Central. Limitations of ultrasound compared with computed tomography for kidney stone surveillance
MRI is occasionally mentioned as an alternative, but conventional MRI sequences are essentially blind to kidney stones. Stones appear as signal voids that blend with other dark structures, making them extremely difficult to identify.24PubMed. Kidney stone imaging with 3D ultra-short echo time (UTE) magnetic resonance imaging. A phantom study. Specialized research sequences are being developed to address this, but MRI is not a practical tool for stone detection in routine clinical settings today.
Artificial Intelligence in Stone Detection
Deep learning algorithms are being trained to automatically flag kidney stones on CT, potentially catching ones that a radiologist in a time-pressed emergency shift might overlook. One system, validated on 100 emergency department CT images, achieved about 94% accuracy. The stones it missed were mostly located at the ureterovesical junction, the spot where the ureter enters the bladder. That is a notoriously tricky location for both human readers and algorithms, partly because of the phlebolith confusion described earlier. False positives, where the system flagged something that was not actually a stone, occurred about 3% of the time, often triggered by artifacts or vascular calcifications.25PubMed Central. A Novel Deep Learning-based Artificial Intelligence System for Interpreting Urolithiasis in Computed Tomography These tools are still being refined, but they represent a second set of eyes that could reduce the small-but-real miss rate of human interpretation alone.
Kidney Stones in Children and Radiation Concerns
Children are more sensitive to ionizing radiation than adults, so there is strong motivation to minimize CT use in pediatric stone evaluation. Ultrasound is often the first-line study, with CT reserved for cases where ultrasound is inconclusive and surgical planning requires precise stone localization. When CT is used, reduced-dose protocols tailored to smaller body size can cut radiation substantially. A scoping review of the literature found that these radiation-reduction approaches did not negatively affect the ability to detect or characterize stones.26PubMed Central. Radiation exposure associated with computed tomography for pediatric urolithiasis evaluation: A scoping review of the literature Even so, the general principle holds: smaller stones in smaller patients on lower-dose scans create more opportunities for a miss, and a clinical picture that does not match a negative scan should prompt further investigation rather than reassurance.