Stone Composition from Hounsfield Units Interpreter
Stone Composition from Hounsfield Units Interpreter
What a CT attenuation value is consistent with, and why it cannot be more than that. The published ranges for uric acid, cystine, struvite and the calcium stones overlap so widely that one series found a 700–1,000 HU threshold misclassified 68 per cent of calcium oxalate monohydrate stones.
Which compositions this attenuation is consistent with
Attenuation + size + pH → a set, never one answerA 12 mm renal stone with an unenhanced attenuation of 1,200 HU and a urine pH of 6.0
What the number can support, and what it cannot
No non-calcium stone exceeded 76 HU/mm in the series that proposed it
Uric acid, combined rule: 500 HU or less and urine pH 5.5 or less → 87 per cent sensitivity, 83 per cent specificity
- this page classifies nothing
- This page classifies nothing. It reports which published compositions are consistent with the attenuation entered, and the published ranges overlap so heavily that more than one always is. Dual-energy CT, not a single attenuation number, is what actually separates a uric acid stone from a calcium one. A single attenuation number from a single-energy scan cannot identify a composition, and the rules above are written to report a set for that reason
- the ranges disagree between series
- uric acid is published as 100 to 200 HU, as 258 to 679, as 67 to 769, as 566 to 632, as 347 to 512 and as 367 to 556 in six readings of the same composition. Calcium oxalate monohydrate is published as 1,707 to 1,925, as 496 to 1,865, as 507 to 1,639 and as 783 to 1,010. Those are not measurement noise around a true value; they are different studies measuring different things on different scanners
- size changes the answer
- attenuation varied significantly with stone size for both calcium oxalate hydrates in a series of 91 patients, and a 700 to 1,000 HU threshold misclassified every stone under 5 mm. Partial volume averaging is the mechanism, and it always makes a small stone read softer
- mixed stones
- the series with the clearest separations studied PURE or near-pure stones — 100 of them in one case, 180 in another. The authors of the size-dependence study attributed the poor prediction for calcium stones largely to mixed composition, and most clinical stones are mixed
- a different use of the same unit
- the adrenal incidentaloma interpreter reads unenhanced attenuation at a 10 HU threshold as a statement about intracytoplasmic LIPID. That is the same unit doing a different job three orders of magnitude lower down the scale, and the two must not be confused. This page reports a figure and what the published sources attach to it. It does not make a clinical decision and cannot.
Worked example
A 12 mm renal stone with an unenhanced attenuation of 1,200 HU and a urine pH of 6.0
The stone is 12 mm, so it clears the 5 mm floor below which attenuation carried no information in the series of 91 patients
Hounsfield unit density = 1,200 ÷ 12 = 100 HU/mm
100 HU/mm is above the 76 HU/mm that no non-calcium stone exceeded in Motley's series of 100 pure stones, so the calcium group is the consistent one
It is not a calcium oxalate monohydrate stone, a calcium oxalate dihydrate stone or a calcium phosphate stone on this evidence — the same series reported that the calcium stones could not be distinguished from one another, and Plata's values put apatite at 835 to 1,034 HU and monohydrate at 783 to 1,010 HU, which overlap almost completely
Change one number: at 1,200 HU in a 30 mm stone the density falls to 40 HU/mm, below the ceiling, and the answer widens to include struvite and uric acid. The attenuation did not change
The published attenuation ranges, as the sources give them
| Composition | Published values (HU) |
|---|---|
| Uric acid | 100 to 200 (review introduction); 67 to 769, 566 to 632 and 347 to 512 in three further series; 367 to 556 (Plata) |
| Cystine | mostly below 700 to 800; grouped WITH uric acid below Gallioli’s 825 HU cut-off |
| Struvite | 549 to 869, 862 to 944 and 790 to 2,143 in three series; 540 to 693 (Plata) |
| Calcium oxalate dihydrate | 1,853 to 2,536, 1,416 to 1,938 and 324 to 1,015 in three series; 873 to 1,218 (Plata) |
| Calcium oxalate monohydrate | 1,707 to 1,925, 496 to 1,865 and 507 to 1,639 in three series; 783 to 1,010 (Plata); 1,548 and above (Silva) |
| Carbonate apatite | 835 to 1,034 (Plata) |
What happens when a threshold is actually applied
| Stone size | Calcium oxalate monohydrate stones misclassified by a 700 to 1,000 HU threshold |
|---|---|
| Under 5 mm | 10 of 10 — every one |
| 5 to 10 mm | 13 of 22 (59 per cent) |
| Over 10 mm | 5 of 9 (55 per cent) |
| All sizes | 28 of 41 (68 per cent) |
The overlap is the finding
Uric acid stones are characteristically low in attenuation and calcium oxalate monohydrate and brushite high, and that much is true. What follows from it clinically is much less than it appears, because the published ranges for the compositions overlap across most of the usable scale. Taking the widest range reported for each, a stone at 600 Hounsfield units lies simultaneously inside the uric acid range, the struvite range, the calcium oxalate dihydrate range and the calcium oxalate monohydrate range. This page classifies nothing. It reports which published compositions are consistent with the attenuation entered, and the published ranges overlap so heavily that more than one always is.
The ranges do not merely overlap; they disagree. Uric acid is published as 100 to 200 HU in the introduction of one review, as 258 to 679 in a systematic review, and as 67 to 769, 566 to 632, 347 to 512 and 367 to 556 in four separate series. Calcium oxalate monohydrate appears as 1,707 to 1,925 in one series and 496 to 1,865 in another. These are not error bars around a single truth. Attenuation depends on the scanner, the tube voltage, the slice thickness and whether the report quotes a mean, a peak or a region of interest — in one dual-energy series of 270 stones the same non-uric-acid stones averaged 1,099 HU at 80 kVp and 706 HU at 140 kVp.
Stone size is the other variable, and it is the one most often forgotten. In 91 patients with surgical stone analysis and a non-contrast CT, attenuation varied significantly with size for both calcium oxalate hydrates, and a 700 to 1,000 HU threshold for monohydrate misclassified 28 of 41 such stones — including every single stone under 5 mm. Partial volume averaging is the mechanism: a small stone’s measured attenuation is pulled towards the tissue around it, so small stones read softer than they are. That is why this page asks for the diameter, and why the first thing it reports for a stone under 5 mm is that the attenuation has not narrowed anything.
Two derived figures do carry real information and both are used above. Attenuation divided by maximum diameter separated uric acid (50 ± 24 HU/mm) from calcium stones (105 ± 43 HU/mm) in 100 pure stones, with no non-calcium stone exceeding 76 HU/mm — a one-way statement that is useful above the ceiling and silent below it. And attenuation of 500 HU or less combined with a urine pH of 5.5 or less predicted a uric acid stone at 87 per cent sensitivity and 83 per cent specificity. Dual-energy CT, not a single attenuation number, is what actually separates a uric acid stone from a calcium one. The clinically useful reading of a high value is not chemical at all: the EAU notes that stones above 1,000 HU with high homogeneity are less likely to be disintegrated by shockwave lithotripsy, while a separate series found fragility did not correlate with attenuation and that morphological inhomogeneity correlated better. The stone’s size, which matters more than its density for most decisions, is on the stone volume page.
Frequently asked questions
Can Hounsfield units tell me what my stone is made of?
No. This page classifies nothing. It reports which published compositions are consistent with the attenuation entered, and the published ranges overlap so heavily that more than one always is. In the series that tested a threshold directly, a 700 to 1,000 HU cut-off for calcium oxalate monohydrate misclassified 68 per cent of such stones. Dual-energy CT, not a single attenuation number, is what actually separates a uric acid stone from a calcium one.
What attenuation suggests a uric acid stone?
The best-performing published rule combines two things: attenuation of 500 HU or less AND a urine pH of 5.5 or less, which predicted uric acid at 87 per cent sensitivity and 83 per cent specificity. Even then, cystine occupies the same attenuation range as uric acid, and one series found cystine, uric acid and struvite could not be accurately distinguished from one another.
Why does my stone’s attenuation differ between two scans?
Because attenuation is a property of the measurement as much as of the stone. Tube voltage alone accounts for a large part of it: in a dual-energy series the same non-uric-acid stones averaged 1,099 HU at 80 kVp and 706 HU at 140 kVp. Slice thickness, the scanner, the reconstruction and whether a mean or a peak was reported all contribute.
Why does the stone’s size matter for its attenuation?
Partial volume averaging. A small stone occupies only part of each voxel, so its measured attenuation is pulled towards the surrounding tissue and it reads softer than it is. In 91 patients, every stone under 5 mm fell below the attenuation threshold for its own composition.
Is this the same as the 10 HU threshold used for adrenal lesions?
No — same unit, different question, different scale. The adrenal incidentaloma interpreter uses unenhanced attenuation at or below 10 HU as evidence of intracytoplasmic lipid in an adenoma. Stone attenuation is measured in the hundreds and thousands and is about mineral density. The two thresholds have nothing to do with each other.
What this page will not tell me
Every threshold, range and performance figure on this page is the published figure from the source named beside it, and each one depends on the population, the method and the equipment it was derived in. Where your own report, laboratory or local guideline gives a different figure, that figure governs. A score, an index, a measured volume or an attenuation value is not a diagnosis, and a proportion measured in a cohort is not a probability for one patient. This page reports a figure and what the published sources attach to it. It does not make a clinical decision and cannot.
Related calculators
References
- Merticariu M, Rascu S, Anghelescu DV, Merticariu CI. Hounsfield measurements for detection of stone composition, density, and overall hardness — a brief report. Surg Gastroenterol Oncol. 2022;27(2):152–156.
- Stewart G, Johnson L, Ganesh H, Davenport D, Smelser W, Crispen P, Venkatesh R. Stone size limits the use of Hounsfield units for prediction of calcium oxalate stone composition. Urology. 2015;85(2):292–295.
- The role of Hounsfield units in predicting urinary stone composition: a systematic review. MAGNA MEDIKA. 2026;13(2):147–155 — abstract read; the full text is PDF-only and was not read, so only its abstract’s figures are quoted.
- Predicting the composition of urinary stones by non-enhanced spiral and dual-energy CT: mean attenuation at 80 and 140 kVp in 270 cases. Korean J Urol, via KoreaMed Synapse.
- European Association of Urology. Guidelines on Urolithiasis, 2026 edition — Guidelines chapter, including the cumulative-diameter size strata, the spontaneous passage proportions in section 3.4.9.a and the shockwave-lithotripsy attenuation statement. uroweb.org.
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/