Testicular Torsion
- Testicular torsion is a plumbing emergency: the testicle twists on its cord and chokes off its own blood supply. Salvage falls the longer it stays twisted, so nothing about the workup is allowed to be slow.
- It's a clinical diagnosis first. If the story screams torsion, the patient goes to the operating room — imaging must never delay surgery.
- The ultrasound findings that matter most are absent or reduced intratesticular flow on color Doppler compared with the normal side, and a twisted cord (the whirlpool sign) — the whirlpool runs about 92% sensitive and 99% specific.
- Color Doppler ultrasound in adults is good, not perfect: pooled sensitivity 0.86 and specificity 0.95 across all studies, rising to 0.95 and 0.98 in the more recent ones. A "normal" scan does not cancel a convincing story.
- Don't be fooled by gray-scale that looks normal early on, or by the partial or intermittent twist that still has some flow.
Imagine a hanging plant pot suspended by a single cord, and you spin the pot so the cord wrings itself into a tight twist. Water can't get up the kinked cord anymore, and the plant starts to wilt. That, in painfully literal terms, is testicular torsion: the testicle rotates on its spermatic cord, the cord wrings shut, and the blood that's supposed to keep the testicle alive can't get in (or, depending on the timing, can't get back out).
It is one of the few times in radiology where the words "drop everything" are not an exaggeration.
Why everyone panics about it
The testicle is a demanding little organ — it wants a constant blood supply and it does not forgive a long interruption. Once the cord twists, the testicle is on a timer. Salvage is best when it's untwisted early and falls steadily the longer it sits strangled. You will hear people quote a "safe window" in hours; I'm deliberately not going to give you one, because the honest version is a slope, not a cliff: every hour spent waiting costs testicle, and there is no minute at which it becomes fine to dawdle.
That's why classic teaching is blunt about it — time is testicle. The whole job of imaging here is to be fast and right, or to get out of the way entirely.
The anatomy behind the twist is worth one sentence. Normally the testis is anchored to the back wall of its sac so it can't spin; in some people the covering (the tunica vaginalis) wraps too high around the cord, leaving the testis dangling free like the clapper inside a bell. That "bell-clapper" arrangement is the setup; a sudden movement, cold, or sometimes sleep is the trigger.
If the clinical picture is classic — sudden severe scrotal pain, often with nausea, a high-riding testicle, and an absent cremasteric reflex — the patient may go straight to surgery WITHOUT waiting for ultrasound. Imaging is for when the diagnosis is genuinely unclear, not for confirming the obvious. A normal-looking scan should never overrule a screaming clinical story.
The finding: is blood getting in?
Ultrasound with color Doppler is the workhorse. Color Doppler is just ultrasound painting moving blood in color — flow toward the probe one color, away the other. (If that color-equals-motion idea feels shaky, take two minutes on Doppler in plain English.) The single most important thing you're hunting for is simple:
Is there blood flow in this testicle, and how does it compare to the other one?
You always interrogate both sides with identical settings, because the comparison is the whole game. Practically, that means a high-frequency linear probe, a transverse image with both testes in the same frame, and the machine dialed for slow flow: low velocity scale, low wall filter, color gain pushed up until the normal side just starts to speckle. In torsion, the affected testicle shows absent or markedly reduced color flow while the normal side lights up reassuringly. That side-by-side asymmetry is the money finding.
Then confirm with spectral (pulsed) Doppler. Put the gate inside the testis and look for an arterial waveform. In a complete torsion there is nothing to find. In an early or partial torsion, the veins get squeezed shut before the arteries do, so the first thing to change is the diastolic part of the arterial trace: it flattens, then disappears, then reverses — the artery pushing blood in during systole and having it bounce straight back out because there's nowhere for it to go. A testis with reversed diastolic flow is a testis in trouble even if a few pixels of color are still visible.
The cord and the gray-scale clues
Gray-scale (the plain black-and-white image) can be sneaky. Early on, the torsed testicle can look almost normal in gray-scale — the damage hasn't shown up yet. A normal adult testis is roughly 5 × 3 × 2 cm and uniformly echogenic, and in the first hours a twisted one can still meet that description. That's exactly why you can't relax just because the picture looks fine; the Doppler is what tells the truth.
As time passes, the gray-scale catches up: the testicle swells and becomes heterogeneous and darker as it loses its blood supply, often with a reactive hydrocele (more than the 2–3 mL of fluid that normally sits between the layers of the tunica vaginalis), and thickened scrotal skin. A heterogeneous testicle is a worrying sign — it suggests the tissue is already in trouble, and it also lowers the odds the surgeon will be able to save it. The epididymis often swells too, beyond its normal 10–12 mm head, because the cord it hangs from is engorged — which is one reason a torsion can be misread as epididymitis if you only look at the epididymis.
There's also a direct sign worth knowing: the whirlpool sign, where the twisted spermatic cord coils on itself like water spiraling down a drain. On gray-scale and color Doppler, if you follow the cord down from the groin toward the testis and rock the probe, the vessels of a twisted cord wind around a central point instead of running in parallel. Seeing that swirl is a specific tip-off that the cord has wound up — and it is the one finding that doesn't depend on how much flow is left in the testis.
Always sweep up to the spermatic cord, not just the testicle. The twist lives in the cord, and the whirlpool sign there can clinch the diagnosis even when the testicle's flow is ambiguous. Find the cord at the external ring, follow it down, and look for the knot.
How not to miss it
This is a "don't-miss," and it earns the title because there are a few honest ways to get burned. Here are the ones that actually lose testicles.
Partial or intermittent torsion can preserve some flow. A testicle that isn't fully wrung shut may still show some color Doppler signal, which is dangerously reassuring. Reduced or asymmetric flow — not just zero flow — should still raise alarm, and so should an arterial waveform with absent or reversed diastolic flow. When the story fits, trust the story over a "there's a little flow" reading.
The testis that untwisted on its own. A torsion that spontaneously detorses before the scan can show increased flow from reactive hyperemia and look exactly like epididymo-orchitis. The clue is the history — pain that was severe and is now easing — and sometimes a cord that still looks thickened. This patient is not out of the woods; the cord that twisted once will twist again.
The tiny testis. In a small child the testis has very little flow to begin with, so "I can't see flow on either side" doesn't help you. Push the settings, use power Doppler if you have it, and compare with the other side. If you can't confidently demonstrate normal symmetric flow and the clinical suspicion is high, that uncertainty favors surgical exploration — not a comforting "probably fine."
Stopping at the testis. If you never look at the cord you will never see a whirlpool, and the whirlpool is your most specific sign.
Settings that lie. A velocity scale set for the aorta will show nothing in a normal testis either. Before you call flow absent, prove the machine can see slow flow on the normal side at the same settings.
The big mimic is epididymo-orchitis, the infectious inflammation that also causes an acutely painful, swollen scrotum. The tell is the opposite of torsion: infection brings increased blood flow (a hyperemic, angry, lit-up testicle and epididymis), whereas torsion brings decreased flow. So the Doppler doesn't just say "abnormal" — the direction of the flow change points you toward the right diagnosis. The full side-by-side of all the acute-scrotum contenders lives on the approach to the acute scrotum page.
| Feature | Testicular torsion | Epididymo-orchitis |
|---|---|---|
| Onset | Sudden, severe | More gradual, often with fever or urinary symptoms |
| Color Doppler flow | Decreased or absent in the testis | Increased (hyperemia) |
| Cord | Whirlpool sign | Thickened but straight |
| The move | Surgery, now | Antibiotics |
The numbers
There is one number everyone wants from this page and I'm going to withhold it: the number of hours you have before the testicle is lost. It is genuinely time-dependent, salvage falls as the hours pass, and the practical instruction is the same whatever the cutoff — go now. What I can give you are the normal values that let you recognize an abnormal scrotum, and the accuracy figures that tell you how much to trust the scan.
| What | Threshold / value | Why it matters |
|---|---|---|
| Time from twist to surgery | No fixed cutoff stated here; salvage falls as time passes | The only threshold that matters is "as soon as possible" — imaging never delays the surgeon |
| Normal adult testis | About 5 × 3 × 2 cm, homogeneously echogenic | An early torsed testis can still look like this; a swollen heterogeneous one is late |
| Normal epididymal head | 10–12 mm in diameter | Swells beyond this in torsion as well as infection — don't let an enlarged epididymis talk you out of torsion |
| Normal tunica vaginalis fluid | 2–3 mL | More than this is a reactive hydrocele, a supporting (not diagnostic) sign |
| Whirlpool sign | Sensitivity 92%, specificity 99% | The most specific single finding; look for it in the cord |
| Color Doppler US, adults, all studies | Sensitivity 0.86 (95% CI 0.79–0.91); specificity 0.95 (0.92–0.97) | Roughly one in seven torsions had a falsely reassuring scan in the pooled data |
| Color Doppler US, adults, studies after 2010 | Sensitivity 0.95 (0.84–0.99); specificity 0.98 (0.93–0.99) | More recent studies do better, but still not perfectly |
How good is the test
Color Doppler ultrasound is the first-line test, and in an adult meta-analysis it performed well: overall pooled sensitivity 0.86 (95% CI 0.79–0.91) and specificity 0.95 (0.92–0.97). Restricted to studies published after 2010 — plausibly reflecting newer scanners and better low-flow technique — sensitivity rose to 0.95 (0.84–0.99) and specificity to 0.98 (0.93–0.99). Read that as: a positive scan is very believable, and a negative scan on a modern machine is fairly reassuring, but a meaningful number of torsions still slip through when the twist is partial, early, or intermittent. Those figures come from adults; there is no robust separate pooled figure for children to quote here, where small-testis flow makes the exam harder.
The whirlpool sign, as summarized in the scrotal imaging literature, is about 92% sensitive and 99% specific. Its specificity is the point: a swirling cord means a twisted cord.
A sensitivity of 0.86 means that in the pooled data roughly one torsion in seven had a scan that looked reassuring. That's the whole argument for the rule in the red box above: the ultrasound can add confidence, but it cannot subtract a strong clinical suspicion. When the story is classic and the scan is "normal," the answer is a phone call to the surgeon, not a discharge.
What to say in the report
Short and unambiguous. State whether intratesticular flow is present, reduced, or absent compared with the other side at identical settings; whether an arterial waveform was obtained and what diastole looked like; whether the cord shows a whirlpool; and the gray-scale state of the testis (normal echotexture versus heterogeneous). If flow is absent or the cord is twisted, say "findings consistent with testicular torsion" and pick up the phone — this is a critical result, and the report's timestamp should be later than the call.
The one thing to carry out the door
Torsion is the twisted garden hose of the scrotum, and the ultrasound's job is to answer one question fast: is blood getting into this testicle compared to the other side, and is the cord twisted? Decreased or absent flow, or a whirlpool in the cord, plus a convincing story means the operating room, quickly. The mirror-image emergency in women — ovarian torsion — follows the very same logic. When in doubt, remember the timer: the testicle is far more forgiving of an unnecessary trip to the OR than of an hour wasted admiring a normal-looking scan.
References
- Ota K, Fukui K, Oba K, et al. The role of ultrasound imaging in adult patients with testicular torsion: a systematic review and meta-analysis. J Med Ultrason 2019;46(3):325–334. Used for: the pooled color Doppler ultrasound sensitivity 0.86 (95% CI 0.79–0.91) and specificity 0.95 (0.92–0.97) for all studies, and sensitivity 0.95 (0.84–0.99) and specificity 0.98 (0.93–0.99) for studies after 2010, in Key Points, "The numbers," "How good is the test," and the warning callout ("roughly one in seven").
- Lam Shin Cheung J, Bhaduri M. A sound approach to stay on the ball—a review of scrotal pathologies on ultrasound imaging. Br J Radiol 2026;99(1181):847–858. Used for: the whirlpool sign sensitivity 92% and specificity 99% in Key Points, "The cord and the gray-scale clues," "The numbers," and "How good is the test"; the normal adult testis size of about 5 × 3 × 2 cm, the normal epididymal head of 10–12 mm, and the normal 2–3 mL of tunica vaginalis fluid in "The cord and the gray-scale clues" and "The numbers."
Sign in to take the quiz and track your progress