Ultrasound First for Kidney Stones: What the STONE Trial Showed

Acute one-sided flank pain, nausea, restlessness, and blood in the urine can suggest a ureteral stone. They do not prove one. Appendicitis, ovarian or testicular emergencies, and infection can produce overlapping symptoms. So can aortic disease, bowel disease, and other conditions.

Noncontrast CT is excellent at finding stones and alternative diagnoses. It also uses ionizing radiation, and stone disease often recurs. Repeated CT can create a substantial cumulative dose. The STONE trial asked a practical question: can emergency care safely begin with ultrasound and reserve CT for people who still need it?

The clinical tradeoff#

CT can locate a stone, measure it, show obstruction, and identify many alternative causes of pain. Its sensitivity is higher than ultrasound, especially for small ureteral stones. Noncontrast technique avoids iodinated contrast for the usual stone question.

Ultrasound uses sound waves. It can show hydronephrosis, which is dilation upstream from obstruction, and may see stones in the kidney, renal pelvis, ureteropelvic junction, or near the bladder, though a ureteral stone between those points can be hard to see, particularly with bowel gas or larger body habitus.

The choice is not “accurate CT” versus “inaccurate ultrasound.” It is whether the first test provides enough information for a safe plan, and whether the added CT information is likely to change what you do next.

How STONE was designed#

The trial enrolled adults with suspected nephrolithiasis at 15 academic emergency departments in the United States, and it randomly assigned 2,776 participants to one of three initial strategies: point-of-care ultrasound performed by an emergency clinician, ultrasound performed in radiology, or abdominal and pelvic CT. Baseline data were available for 2,759.

The study was pragmatic and unblinded. Treating clinicians could order another imaging test after the assigned initial study. That freedom is the whole point. It made the comparison resemble real diagnostic strategies rather than a contest between isolated machines; the trial excluded pregnancy and several higher-risk circumstances, including a solitary kidney, kidney transplant, dialysis, and clinical concern for a serious alternative diagnosis. Its safety finding should not be extended automatically to excluded groups.

The primary safety outcome#

The primary outcome was a high-risk diagnosis with complications that could plausibly be related to a missed or delayed diagnosis during the first 30 days. Such diagnoses included dangerous alternative conditions rather than merely failure to visualize a small stone.

The outcome was rare: 0.4 percent in the point-of-care ultrasound group, 0.7 percent in the radiology ultrasound group, and 0.3 percent in the CT group. Differences were not statistically significant.

Rare outcomes create uncertainty. The result does not prove the strategies have exactly equal risk or that no dangerous diagnosis was ever missed. It indicates that the trial did not detect a clinically important safety disadvantage for the ultrasound-first strategies in the selected population. Serious adverse events, related serious adverse events, and pain scores also did not differ significantly over follow-up. Neither did emergency-department returns or hospital admissions.

Radiation was meaningfully lower#

Mean cumulative radiation dose over six months was about 10.1 mSv in the point-of-care ultrasound group, 9.3 mSv in the radiology ultrasound group, and 17.2 mSv in the CT-first group.

The ultrasound groups did not reach zero because some participants received later CT or radiographs, and the trial therefore measured the real effect of an imaging pathway, including crossovers, rather than quoting the dose of one scan.

Radiation risk from an individual CT is small and uncertain at low doses. It becomes more relevant across a population and for younger people who may have repeated episodes. Avoiding a scan that will not change your care can reduce cumulative risk without costing you useful information.

Additional imaging was part of the strategy#

During the initial visit, 40.7 percent of participants assigned to point-of-care ultrasound and 27.0 percent assigned to radiology ultrasound also underwent CT. In the CT group, 5.1 percent underwent ultrasound.

That means most participants in the ultrasound groups avoided immediate CT, but not all. A clinician could escalate when the image was nondiagnostic, the story was atypical, pain persisted, or an alternative diagnosis remained concerning.

Calling the study “ultrasound versus CT” can hide this feature. The policy tested was “ultrasound first, then more imaging if needed” versus “CT first.” Selective CT was not a protocol failure. It was the mechanism that allowed a lower-dose strategy to remain safe.

Single-test accuracy favored CT#

When the investigators examined the first imaging test by itself, ultrasound had lower sensitivity for nephrolithiasis than CT. That is expected from the physics and anatomy. Ultrasound can infer ureteral obstruction from hydronephrosis without showing the stone itself.

The accuracy analysis also used later stone passage or procedure as part of the reference standard, which can miss stones that never pass visibly and need no intervention. Specificity estimates therefore require caution.

For many emergency decisions, exact stone visualization is not the only goal. The team wants to know whether there is dangerous obstruction, infection, or kidney injury. It also wants to know about another diagnosis, or a stone likely to need urgent intervention. A less sensitive test can support a safe pathway if clinical follow-up and escalation work.

When CT provides important added value#

CT is more useful when the diagnosis remains uncertain, the presentation is atypical, a serious alternative is plausible, pain or vomiting cannot be controlled, kidney function is impaired, fever or sepsis is present, or ultrasound findings do not fit the clinical picture.

Precise stone size and location influence likelihood of passage and procedural planning. CT can assess stone density and skin-to-stone distance when an intervention is being considered. Contrast-enhanced imaging may be needed when another abdominal or vascular diagnosis is the real concern, even though routine stone CT is noncontrast. Reduced-dose noncontrast CT can preserve high diagnostic performance in suitable patients and settings. “Low dose” should be an actual optimized protocol, not a label assumed from scanner type.

Guidelines do not all give the same first answer#

The 2023 ACR Appropriateness Criteria rate noncontrast CT as usually appropriate for a person with acute flank pain and no or only remote stone history. Ultrasound is rated may be appropriate, with disagreement among panelists.

The multispecialty renal-colic consensus led by emergency medicine, radiology, and urology experts concluded that ultrasound or even no immediate imaging is appropriate in selected scenarios, with reduced-dose CT when CT is needed.

The European Association of Urology recommends ultrasound as the primary diagnostic imaging tool, followed by noncontrast CT to confirm the diagnosis when required. These differences reflect population, setting, and the balance between diagnostic certainty and dose. A guideline statement should be applied to its scenario. First-time severe pain in an older adult is not the same decision as recurrent typical pain in a young adult with prior documented stones.

Red flags change the pathway#

Fever, chills, low blood pressure, confusion, or systemic illness can indicate infection. Obstruction plus infection can deteriorate rapidly because antibiotics may not adequately control infection behind a closed system. Urgent drainage and antibiotics may be necessary.

A solitary functioning kidney, transplant, severe kidney impairment, anuria, or bilateral obstruction raises the stakes of delay. Persistent uncontrolled pain or vomiting can require admission or intervention even without infection. Sudden pain with fainting, pulse abnormality, pregnancy, peritoneal signs, or a presentation unlike prior stones broadens the differential. Ultrasound-first does not mean anchoring on a stone while ignoring discordant findings.

Pregnancy and children#

Ultrasound is the preferred first test during pregnancy and in children because it avoids ionizing radiation. Physiological dilation during pregnancy can make obstruction harder to interpret, and ultrasound may not locate the stone.

The EAU guideline describes MRI as a second-line option during pregnancy and low-dose CT as a last-line option when necessary. This is an individualized specialist decision because missing an obstructed infected kidney also carries serious risk. Children have smaller body size and a long future in which radiation effects could appear. Pediatric protocols use ultrasound first and reserve optimized CT or radiography for unresolved questions.

Imaging is only part of the evaluation#

Urinalysis can show blood, white cells, and nitrites. It can show bacteria, pH, and crystals, though none is decisive alone. Urine culture matters when infection is possible. Blood tests can assess kidney function, blood count, electrolytes, and selected metabolic abnormalities.

Absence of blood does not rule out a stone. White cells may result from irritation without bacterial infection. Conversely, a positive culture plus obstruction is not routine uncomplicated colic. Pregnancy testing, examination, and vital signs all shape how you should read the image. So do pain response, prior imaging, medication history, and comorbidities.

After the emergency visit#

The immediate imaging strategy does not replace follow-up. Before you go home, you should know what fever, worsening pain, or persistent vomiting would mean. You should know what inability to urinate, weakness, or failure to improve would mean, and who to call. A stone may not pass, and hydronephrosis can persist without dramatic symptoms.

First-time and recurrent stone formers may benefit from stone analysis and a prevention assessment. The AUA guideline recommends additional metabolic testing for high-risk or interested first-time stone formers and recurrent stone formers, using one or two 24-hour urine collections with specified analytes. Prevention depends on stone type, urine chemistry, and diet. It depends on fluid intake, medications, anatomy, and associated disease. The ultrasound-versus-CT choice does not tell you why your stone formed.

The deeper comparative-effectiveness lesson#

STONE showed why diagnostic tests should be evaluated as care strategies. CT was more sensitive as one test, yet that did not translate into better measured patient outcomes in the randomized population. Ultrasound first reduced cumulative dose because clinicians could add CT selectively.

This does not make diagnostic accuracy unimportant. It places accuracy inside a chain that includes selection, downstream tests, and treatment. The chain also includes follow-up, harms, and patient outcomes.

The durable message is not “ultrasound is always enough.” It is “start with the lowest-burden test that can safely answer the clinical question, and escalate when the remaining uncertainty matters.”

References#

1. STONE randomized trial
2. ACR acute flank-pain appropriateness criteria
3. Multispecialty consensus on imaging in suspected renal colic
4. EAU urolithiasis guideline
5. NIDDK kidney-stone diagnosis guide
6. AUA medical management of kidney stones guideline

Severe flank pain, fever, systemic illness, reduced urine, pregnancy, or a solitary kidney require prompt clinical assessment.*