Cerebral Venous Thrombosis: Finding the Needle Without Burning Down the Haystack

May 4, 2025May 4, 2025 jneuro Neurology, Radiology Bayes, Cerebral venous thrombosis, CT, Neurology, Neuroradiology

1 Headline Case: 37‑year‑old woman, epic headache

Q: How likely is a cerebral venous thrombosis (CVT)?
A: About as likely as your phone staying silent on a Friday night—but let’s run the math anyway.

1.1 Everyday headaches vs. unicorns

Global background noise: On any random day 15.8 % of humanity is nursing a headache (thanks, life).
– 7 % migraine, 8.7 % tension, 2.5 % “I‑live‑on‑ibuprofen”.
CVT incidence: a majestic 1–3 / 100 000 per year.
Back‑of‑napkin: that’s ~10‑30 CVT per million souls vs 158 000 migraineurs per million → 0.006–0.019 ‰.
(Translation: six to nineteen CVT among one million head‑banging humans.)

1.2 Will she hit the ED?

Only 1–4 % of headaches make it to the ED door.
That’s ~3 000 visits per million residents per year.
Blend the two stats and voilà: 0.6 % pre‑test chance of CVT in our migraine‑magnet ED.

(Told you it’s a unicorn.)

1.3 Will she score a CT?

Reality check: In most German EDs, any thunderclap, “worst ever,” or “neuro can’t‑sleep‑tonight” headache triggers a non‑contrast head CT faster than you can say Strahlenschutz.
Surveys show ≈ 35 % of headache patients end up in the scanner—even though only 4 % have secondary causes.
That means our 0.6 % pre‑test CVT probability slides into the gantry with a 0.6 % × 35 % ≈ 0.2 % CT cohortprobability.
Translation for the radiology tech: “One in 500 of today’s head CTs might hide a venous clot. The other 499 just want your warm dark room.”
Fun paradox: Around 50–70 % of ED headaches raise at least one red flag, yet only ≈ 35 % earn a CT ride—make of that what you will.

2 RISK-HEAD: Trying to avoid CT-angiography

Q: Our patient got her CT and it looks clean as a monk’s browser history. How nervous should we be?
A: The probability was negligible before, it is vanishing now – let’s do the math.

2.1 CT test properties

Bayes rule rules: CT might not be as expensive and great as the all-powerful magnet, but with a sensitivity of 79% and specificity of 95% it reduces the probability to about 0,2%.
Apart from the obvious bleed and brain edema you have to look out for the dense‑vessel sign. Literature pegs it at ≈ 25 – 30 % of acute CVT cases (my personal experience is more like 80%), but requires thin-sliced or maximum-intensity-projection reconstructions – just as in the arterial case. And a lot of vigilance.

2.2 CT venography anyone?

It is unclear who and why gets a venogram next. If you look at the clinical signs, a CVT can show up with any kind of symptom, even a thunderclap headache in about 10 % of CVT cases. So we construct a simple score to capture the few very specific and prevalent signs:

RISK‑HEAD item	Adjusted OR(CVT)	Prevalence in CVT	False‑positive rate in non‑CVT headaches
R – Oral contraceptives	7.6 (3.8–15.1)	45 – 70 % of female CVT	~ 10 – 15 %
I – Any thrombophilia	3 – 6	22 – 34 %	< 1 %
S – ≤ 6 wk postpartum	18.7 (8.3–41.9)	10 – 20 %	0.3 – 1 %
K – Active malignancy	4 – 6	6 – 7 %	4 – 6 %
H – Positional / Valsalva pain	3.2 (1.2–8.4)	20 – 35 %	3 – 5 %
E – Emesis (nausea / vomiting)	1.4 (vs primary HA)	30 – 45 %	30 – 35 %
A – Isolated headache (no neuro deficit)	–	14 – 25 % of CVT present only with pain	80 – 90 %
D – Papilledema	3.3 (LR⁺)	28 – 60 %	0.5 – 3 %

Here is how you unleash RISK-HEAD:

Zero hits? Multiply by LR⁻ 0.4 → 0.08 % (1 : 1 250).

One or more hits? We don’t divide—we multiply (LR⁺ ≈ 1.3) → ≥ 0.25 %.
(That’s one CVT hiding in every 400 Red‑Flagged noggins—worthy of contrast.)

So, if Mrs Headache scores nothing on RISK‑HEAD, your odds have already nosedived to the point where her biggest danger is the hospital vending machine.

3 D‑Dimers: the lab test everyone loves to hate

Perfect tool for very low prevalence diseases—just ask pulmonary embolism.

Meta‑analysis	Sens	Spec	LR⁻
Dentali 2012 (mixed crowd)	93.9 %	89.7 %	0.07
Alons 2015 (isolated headache)	97.8 %	84.9 %	0.03

Negative D‑dimer + RISK‑HEAD‑neg → CVT risk ≈ 1 : 17 000 – 1 : 77 000.
That’s safer than most hospital coffee.

4 CT Venography: the Costly Unicorn Detector

Metric	Value
Price tag	€450 – 1 200 per CTV (Germany 2025)
Radiation	3 mSv (aka 13 months of Bavarian sunshine)
Contrast reactions	0.04 – 0.28 % (death: 1 / 150 000)
False positives	6–10 % (cue unnecessary heparin & sad hematomas)
NND after D‑dimer	≈ 20 000 CTV per extra CVT found
NNH (major bleed via FP + heparin)	1 : 500

TL;DR — You’d hurt more patients than you help if you CTV every RISK‑HEAD negative, D‑dimer negative noggin.

5 Bottom Line (and dad‑joke takeaway)

“If you go looking for trouble in every sinus, you’ll mostly find debt, radiation, and paperwork.”

CT negative → run RISK‑HEAD.
All clear? Draw a D‑dimer.
D‑dimer negative? High‑five, treat the hangover.
D‑dimer or RISK‑HEAD positive? Break out the contrast.

Follow this flow and you’ll spare wallets, retinas, and night shifts—while still netting the rare but deadly clots.

Footnotes (a.k.a. bedtime reading)

Nobody actually clicks them but they look smart:

1. Stovner, L. J., Hagen, K., Linde, M. & Steiner, T. J. The global prevalence of headache: an update, with analysis of the influences of methodological factors on prevalence estimates. J. Headache Pain 23, 34 (2022).

2. Devasagayam, S., Wyatt, B., Leyden, J. & Kleinig, T. Cerebral Venous Sinus Thrombosis Incidence Is Higher Than Previously Thought: A Retrospective Population-Based Study. Stroke 47, 2180–2182 (2016).

3. Doretti, A. et al. Headaches in the emergency department –a survey of patients’ characteristics, facts and needs. J. Headache Pain 20, 100 (2019).

4. 12,4 Millionen Behandlungen in Notfallambulanzen im Jahr 2023. Statistisches Bundesamt https://www.destatis.de/DE/Presse/Pressemitteilungen/2024/12/PD24_N061_23.html.

5. Viganò, A. et al. A Study of Clinical Features and Risk Factors of Self-Referring Emergency Department Headache Patients: A Comparison with Headache Center Outpatients. Eur. Neurol. 83, 34–40 (2020).

6. Goldstein, J. N., Camargo, C. A., Pelletier, A. J. & Edlow, J. A. Headache in United States emergency departments: demographics, work-up and frequency of pathological diagnoses. Cephalalgia Int. J. Headache 26, 684–690 (2006).

7. Yang, S. et al. Trends in the Management of Headache Disorders in US Emergency Departments: Analysis of 2007-2018 National Hospital Ambulatory Medical Care Survey Data. J. Clin. Med. 11, 1401 (2022).

8. Chu, K. et al. Predictive performance of the common red flags in emergency department headache patients: a HEAD and HEAD-Colombia study. Emerg. Med. J. EMJ 41, 368–375 (2024).

9. García-Azorín, D., Monje, M. H. G., González-García, N., Guerrero, Á. L. & Porta-Etessam, J. Presence of red flags in patients with cerebral venous sinus thrombosis admitted to the emergency department because of headache: A STROBE compliant cohort-study. Medicine (Baltimore) 99, e20900 (2020).

10. Ulivi, L., Squitieri, M., Cohen, H., Cowley, P. & Werring, D. J. Cerebral venous thrombosis: a practical guide. Pract. Neurol. 20, 356–367 (2020).

11. Botta, R. et al. Headache Patterns in Cerebral Venous Sinus Thrombosis. J. Neurosci. Rural Pract. 8, S72–S77 (2017).

12. Coutinho, J. M., Gerritsma, J. J., Zuurbier, S. M. & Stam, J. Isolated cortical vein thrombosis: systematic review of case reports and case series. Stroke 45, 1836–1838 (2014).

13. Sheth, S. A. et al. Venous collateral drainage patterns predict clinical worsening in dural venous sinus thrombosis. J. Neurointerventional Surg. 10, 171–175 (2018).

14. Johnson, L. N., Hepler, R. S. & Bartholomew, M. J. Accuracy of papilledema and pseudopapilledema detection: a multispecialty study. J. Fam. Pract. 33, 381–386 (1991).

15. Dalay, S., Umar, F. & Saeed, S. Fundoscopy: a reflection upon medical training? Clin. Teach. 10, 103–106 (2013).

16. Chaudhary, S. R. et al. Diagnostic Sensitivity of Unenhanced CT for Cerebral Venous Thrombosis: Can Clot Density Measurement Replace CT Venogram? Indian J. Radiol. Imaging 33, 187–194 (2023).

17. Amoozegar, F., Ronksley, P. E., Sauve, R. & Menon, B. K. Hormonal Contraceptives and Cerebral Venous Thrombosis Risk: A Systematic Review and Meta-Analysis. Front. Neurol. 6, 7 (2015).

18. Lauw, M. N., Barco, S., Coutinho, J. M. & Middeldorp, S. Cerebral venous thrombosis and thrombophilia: a systematic review and meta-analysis. Semin. Thromb. Hemost. 39, 913–927 (2013).

19. Dinc, Y. et al. Evaluation of risk factors for postpartum cerebral venous sinus thrombosis, a multicenter retrospective observational study. Medicine (Baltimore) 103, e40772 (2024).

20. Silvis, S. M. et al. Cancer and risk of cerebral venous thrombosis: a case-control study. J. Thromb. Haemost. JTH 16, 90–95 (2018).

21. Dentali, F. et al. D‐Dimer Testing in the Diagnosis of Cerebral Vein Thrombosis: a Systematic Review and a Meta‐Analysis of the Literature. J. Thromb. Haemost. JTH 10, 582–9 (2012).

Science Without Objectivity – is it still Science?

January 31, 2025 jneuro General medicine Epistemology, Medicine, Objectivity, Philosophy, science, Science communication, Truth, Writing scientific articles

We doctors like things neat and tidy. Clear diagnoses, precise pathomechanisms, hard evidence. And if possible, let’s have it all in numbers. Numbers sound objective – and objectivity is the foundation of science. Right?

The only problem is… it isn’t.

Because if we’re honest, we all know: The kind of science that delivers truth in clean decimal points exists only in textbooks and statistical fantasies. In reality, we trip over contradictory data, unclear causalities, and the unsettling realization that even the most significant p-values aren’t immune to human factors.

Yet we cling to the idea that science is a kind of analytical machine – emotionless, infallible, a pure logic engine that spits out knowledge at the push of a button. A comforting fiction we hold onto, even though we’ve all experienced first-hand that science thrives on doubt, that it doesn’t produce eternal truths but rather hypotheses, and that the truly interesting questions rarely come with simple answers.

The End of Fake Objectivity

Maybe it’s time to admit: Science is not a glass-walled lab; it’s a workshop full of wood shavings. A constant process of thinking, experimenting, and (if we’re lucky) understanding. It’s less a sterile truth factory and more an intellectual adventure—complete with detours, dead ends, and wild ideas.

But that doesn’t mean science has no rules. On the contrary: Precisely because it isn’t objective, it needs principles to keep it from slipping into pure opinion (or worse: politics). But those principles aren’t what you might expect.

If Not Objectivity, Then What?

Helen Longino, one of the most influential philosophers of science today, makes it clear: Science isn’t special because it’s free from values or interests. It’s special because it imposes rules on itself—rules that protect it from becoming arbitrary. Four of these are crucial:

Everything must be open to criticism.
If a scientific claim cannot be challenged, it isn’t science—it’s dogma. Every good study should be written with the secret hope that it will soon be outdated.
Science needs diverse perspectives.
A single viewpoint is a dangerous thing. The greatest scientific advances often came from outsiders, dissenters, and those who refused to accept the status quo.
Not all claims are created equal.
A hypothesis isn’t valuable just because it’s “interesting” or “plausible”—it must be methodologically sound, logically coherent, and empirically testable.
Science corrects itself.
Peer review, replication studies, open data—these are not bureaucratic annoyances. They are the mechanisms that keep science from getting lost in its own enthusiasm.

These principles ensure that, despite all its uncertainties, science remains reliable. It doesn’t hold a monopoly on truth, but it’s the best system we have for minimizing error.

Why This Matters in Medicine

Now, one might ask: Fine, but what does this have to do with everyday neuroradiology?

Well—everything.

Because in medicine, we often mistake science for statistics, forgetting that numbers don’t speak for themselves. A meta-analysis might be methodologically flawless—but if it ignores clinical reality and pathophysiological mechanisms, it produces nothing but numerical noise. An AI model might deliver impressive diagnoses—but without transparency in its algorithms, can we really call it science?

In short: If we reduce science to mere objectivity, we risk losing what makes it truly valuable.

So What’s Left?

Maybe this: Science isn’t infallible. It’s not a simple equation. It’s full of uncertainties, dead ends, and provisional truths. But that’s exactly what makes it strong.

Not because it’s always right—but because it allows itself to be corrected.

And that’s worth far more than any perfectly calibrated p < 0.05.

But if we acknowledge that, we have to ask ourselves: How should we write about science? If objectivity is a useful fiction rather than an absolute, what should an abstract look like—one that doesn’t hide behind passive voice and numerical mystique, but instead embraces transparency and intellectual honesty?

Something to think about next time we submit a paper.

Measurement improvement measures – the size of intracranial hemorrhage

January 23, 2025January 23, 2025 jneuro Neurology, Radiology ai, CT, health, Intracranial hemorrhage, Neurointensive Care, Radiology, science

In radiology, we measure. Oh boy, do we measure. Lymph nodes, intracerebral hemorrhages, lung nodules, tumors – you name it. If you believe in McCoubrie’s Rule of Radiology #11 (“never measure anything”) – which I do – you should doubt all clinical and radiological classifications that rely on measurements rather than qualitative biomarkers because they are inherently unreliable. But here is the catch: like many of McCoubrie’s truisms, this is wise philosophy and nearly impossible to implement in practice. Medicine loves to put on its objective poker face. It shuns uncertainty, subjectivity, and emotional or gut feelings. And what could feel more objective – more definitive – than a measurement of length, area, or volume to gauge disease severity?

Let’s dive into the bread and butter of radiology: intracranial hemorrhage (ICH). It’s like the low-hanging fruit of CT findings – white, bright, and screaming for attention. ICH is so easy that medical students can find them. Or artificial intelligence. Well, in around 95% of cases, so every twentieth patient falls through AI’s hands, at least two every day. Braive new world.

Reporting ICH on NCCT

Here’s what we need to comment on in the case of a an ICH on a non-enhanced cranial CT:

Location: Where? Typical or atypical?
Measurements: Length, width, height, volume.
Form: Unifocal/multifocal, patchy, ellipsoid, irregular?
Borders: Sharp? Regular? Satellites?
Homogeneity: Including islands, swirl, blend, black hole signs.
Perilesional vasogenic edema: Amount and distribution.
Intraventricular or subarachnoid components?
Mass effects: Compression of ventricles/aqueducts, herniation?
Signs of trauma?

Let’s tackle the measurements in reverse order.

Volume

It seems obvious that volume correlate best with badness, but actually, that has not been studied, let alone established. This is because volume is hard to measure.

Estimation. There are quite a few estimation formulas that use simpler measures, the most common being abc/2 (length × width × height / 2), which has good mathematical motivation: for a pure rotational ellipsoid with diameters a, b, c, the radii are a/2, b/2, c/2, and the volume is computed using the formula:

4/3 π a/2 b/2 c/2 = π/3 abc/2

Approximating π as 3 simplifies this to abc/2. This was popularized by Kothari in one of the most entertaining papers in stroke medicine – an article that managed to make doctors calculate. We love objectivity, precision, and numbers but hate computing them. Other approximations have been proposed, such as 2.5/6 abc (ugh, too many decimals), abc/3 (who wants to divide by 3?), or even π/6 abc. These offer improved precision at the cost of memorability and ease of use.

If your PACS viewer has an area delineation tool, you could calculate volume as Ah/2 or Ah/3, where A is the hemorrhage area and h is height. But let’s be honest: most of us either use abc/2 or rely on research tools with automated volume measurements.

Measuring Volume. There are two approaches:

Counting Voxels: Using a specific Hounsfield unit range (e.g., > 50-100 HU) for blood and summing the bloody voxel volumes.
Sequential Area Measurement: Manually delineating hemorrhage areas on all slices and performing simple math. This is more commonly available but less precise.

Defining the Area

How do you decide whether a voxel belongs to the hemorrhage? It’s easy if it’s bright (say > 50 HU) and surrounded by similar density. This is the thresholding method – patches of blood-like HUs are considered hemorrhage and artifacts are eliminated by requiring hemorrhages to be of a certain size (say 3 mm). Starting from a known hemorrhage region, you can extend its border until you reach a stark difference, a method called region growing. However, you’ll encounter islands of lower density and artifacts, especially at the borders. This is why older approaches used edge detection or morphological algorithms to refine boundaries. Modern methods, like Viz.AI or Rapid.AI, invoke neural networks trained on thousands of cases. These systems vastly improve accuracy but remain limited by the quality and diversity of training data. Atypical cases will still fall through.

Demarcation Lines. So, where do algorithms draw the line between parenchyma and ICH? Wherever we radiologists tell them to. And where do we draw the line? Good question. Some radiologists draw it conservatively (inside hyperdense areas), while others place it in transition zones, balancing over- and underestimation or even outside. This is known as edge ambiguity in image processing, complicating consistent measurements. MRI makes it even trickier for early hemorrhages with their peculiar intensity variations.
Should we over- or underestimate the volume? It depends on the clinical context. Personally, I prefer underestimation – the big, bright hemorrhage is already alarming enough. Also, I like to think surgeons will intervene more readily if the volume seems manageable.

Measuring Length

Linear measurements are like a bad breakup – where do you start, and where do you stop? In healthy parenchyma or the “hell of hemorrhage”? On top of this, you must decide the direction of measurement. Three dimensions are necessary: x, y, and z. But should you:

Use the largest diameter in 3D?
Measure in the axial slice with the largest hemorrhage?
Ensure orthogonal directions for the second and third dimensions?

And what happens with non-isotropic datasets (e.g., sequential CT slices)? Partial volume effects add another layer of complexity.

Guidelines don’t specify. Some tumor templates detail 2D slice-based orthogonal measurements but leave open whether the axes must parallel the coordinate system or not. The z-axis is typically defined by slice thickness, ignoring partial volume effects and different departmental conventions for axial angulation (e.g., bicommissural line vs. Reid’s baseline vs. canthomeatal plane).

A Practical Recipe to maybe improve precision

Verify axial reconstructions follow your department’s standard angulation (e.g., Reid’s baseline).
Choose the slice where the ICH looks largest.
For all measurements use the bulk of the ICH and ignore small spotty satellite lesions.
Measure the length a along the sagittal line.
Measure width b orthogonally to the length – this should be right-left.
Use a sagittal reformatted slice centered at the intersection of your length and width to measure height c orthogonal to your length.
Place endpoints on the last pixel resembling hemorrhage.
Report dimensions in cm with one decimal place.
Calculate volume as abc/2(1 cm³ = 1 ml).
Write “a-p 6,9 cm x r-l 4,5 cm x c-c 6,0 cm =~ 93 ml”
Add comments if irregular form, borders, or inhomogeneity suggest imprecision.

“If You Don’t Measure Volumes, You Can’t Find a Fever”

As noted, beware of scores like ICH or FUNC that rely on quantitative cutoffs (e.g., 20 or 30 ml). Given the imprecision of our measurements, small variations should not dictate treatment. Deadly ICHs are big – think egg-sized. For reference, an egg yolk is around 20 ml. Regardless of your chicken’s breed, eyeball your bleeds (an actual eyeball holds about 6–7 ml), trust your gut, and leave the obsession with decimals to engineers. Oh, and never mix metaphors. Unless it’s eggs and eyeballs – then it’s fair game.

Physical Examination of Chronic Back Pain Patients – A Method

January 20, 2025January 20, 2025 jneuro General medicine, Neurology, Radiology Back pain, health, Interventional pain medicine, Neurology, Neuroradiology, Physical examination, sciatica

In our neuroradiological department, we run a spine clinic that often feels like the final chapter of a long diagnostic novel. Patients arrive after consulting 2-5 other specialists, enduring countless needles, and collecting at least one MRI along the way. By the time they reach us, they’re not just seeking relief – they’re seeking closure.

To ensure we don’t add unnecessary or counterproductive procedures to their saga, we dedicate up to 45 minutes per patient, before even contemplating sticking a needle in them. Yes, that’s a rare luxury in modern medicine, but it’s still a race against the clock. Every second counts, and the big question is: what parts of the physical examination truly matter? Goniometers, anyone?

Functions of the physical examination

The physical examination is an art we rarely practice in full after medical school. Each patient’s needs dictate a bespoke menu of tests, chosen from a buffet so vast it could make a Michelin-starred chef jealous. Why do we examine patients at all? Let’s break it down:

Hypothesis Testing: For example, if we suspect a spine fracture, we might perform percussion tenderness tests and compute post-test probabilities. (Okay, maybe not compute. Let’s be honest, we’re often winging it with intuition and the occasional clinical decision-making app.)
Hypothesis Generation: This is the “add-to-cart” phase of diagnostics, where new ideas and incidental findings (like an S3 gallop on auscultation) join the list of suspects.
Triggered Routine: Admit it: a lot of what we do is automatic. It’s in the SOP, or a mentor once told us to do it.
Relationship Building: Let’s face it: the physical exam also reassures patients that we’re more than just interpreters of high-tech scans. It builds trust, helps us remember them (“The one with the hump…”), and occasionally provides a face-saving “no procedure needed” exit strategy.

Hypotheses for Chronic Back Pain

When examining chronic back pain patients, consider these questions:

Is it really back pain? Could it be the hip, or something more exotic like a kidney or … Lupus? Think of frequent differentials/mimics/chameleons:
- Piriformis syndrome
- The tunnels: carpal, cubital, tarsal, peroneal (ok, that is not a tunnel)
- Lyme
- Peripheral musculoskeletal problems: hip, greater trochanter, plantar fascia, ilitiotibial band, hamstring, SIJ
Which pain-sensitive structures are involved?
- Discs? Nope, they’re not pain-sensitive. Bony endplates, however, love to complain.
- Muscles? Usually secondary, rarely the primary culprits.
- Joints? Always eager participants in degenerative drama.
- Bones? Possible inflammation, which hurts. Or fracture, which really hurts.
- Nerves? Compression mostly, but might be inflamed – greetings from Lyme.
- Brain? Chronification takes pain perception to new heights (or depths).
What’s the pathophysiological mechanism? Mechanical inflammation, primary inflammation (rheumatologic or infectious), stenoses, Baastrup…pick your poison.
Which functions are impaired? Walking, sitting, dressing…living?
Any distant causes? Weight, sedentary habits, smoking, or overzealous sports.
Signs of chronification? Spread of tenderness? Antalgic postures perpetuating the original problem?
Therapy challenges? Will the patient tolerate physiotherapy, prone positioning for interventions, or needles?

Example: If chronification has led to widespread tenderness, the patient may not even tolerate lying prone or being touched. In such cases, reducing overall pain must precede local measures.

What?

Here’s the buffet in logical order:

Spinal mobility
Posture
Radiculopathy: sensory, motor, pain (Lasegue, inverted Lasegue), reflexes
Long tract signs (Babinski, Romberg’s)
Other red flags: percussion tenderness, signs of ivdu, heart murmur, calor/rubor
Local: paraspinal muscles, spinous processes, facet joint pain, ISJ pain
Provocation tests: Mennell’s 3-phases-test, reclination, spinous processes (Baastrup), foraminal occlusion test

When?

The actual exam follows the logic of convenience rather than the logical one. It depends on whether cervical or lumbar complaints predominate, maybe combined. The following thorough examination takes at least 8 minutes and may extend depending on patient complexity or teaching needs.

Observation

Observe the patient when he enters the room and when he rises from his chair and disrobes (e.g. while washing your hands): range of motion, signs of pain?

Standing and walking (as much of the examination as possible should happen here, unless the pain prevents it)

Walking with and without aid: Observe gait, step width and length, symmetry, and Trendelenburg’s. Learn to distinguish an algophobic limp from issues in the lumbar spine, hip, knee, or ankle.
Toe standing and walking; heel standing and walking: Interestingly, some patients show reduced toe/foot extension when lying down but not when standing.
Unipedal standing: Trendelenburg’s sign? Asymmetry?
Unipedal hopping: Look for asymmetry, weakness, or postural instability – an ultrashort strength test for most lumbar segments
Squat and rise test: Check for asymmetry or weakness (lumbar segments and S1).
Mobility tests: Finger-floor distance, reclination, and rotation (feet fixed on the ground, measuring the combined mobility of the knee, hip, and lumbar complex in degrees).
Spinal tenderness on percussion: Fracture? Spondylodiscitis?
For cervical pain – I do as much as possible standing so as not to strain my back
- Range of motion: Anteflexion, reclination, lateral flexion, head turning
- Pain sensitive structures: myogeloses, spinous processes, facet joints, nuchal
- Ultrashort strength test: Have the patient grab your fingers (2-3) bilaterally and try to push you away. This tests multiple nerves and segments efficiently.
- Short strength test: Covering all major nerves and segments, including: shoulder raise (accessory nerve), rhomboids (dorsalis scapulae, C4-C5), shoulder abduction (axillary, C5), elbow flexion (musculocutaneous, C5-C6), wrist dorsiflexion (radial, C6), elbow extension (radial, C7), finger flexion (median/ulnar, C8), finger abduction (ulnar, C8-T1), and thumb abduction (median, C6-T1)
- Foraminal occlusion test
For lumbar pain: Palpate for scoliosis and atrophy; assess the rest in the prone position.

Sitting

For lumbar pain, not much can be gained sitting, while for cervical pain I do most of the exam here:

Complete tests left out during standing, especially foraminal occlusion and strength testing.
Sensory testing: Check superficial touch and pain for C5 (anterior shoulder), C6 (thumb/digit 3), C7 (digit 5), and C8.
Hoffmann-Tinel sign: Check for carpal tunnel syndrome (C6-C7) and ulnar sulcus syndrome (C8).
Reflexes: Test pectoral (C5), biceps (C5-C6), brachioradialis (C6-C7), triceps (C7), finger flexors (C8), and finger abductors (T1).

Lying on the back

This is mostly for lumbar back pain. Not all patients tolerate lying straight – you might need a knee roll.

Range of motion: Let the patient raise his leg as far as possible on either side.
Lasegue’s
Strength test: There is no short test for lumbar back pain – you need to do all. Hip flexion, extension, abduction, adduction, knee flexion, extension, plantar flexion, extension, inversion, eversion, toe flexion, extension. See below for a method.
Sensory testing: Assess touch and pain (broken tongue blade) for L3 (anterior thigh), L4 (knee), L5 (dorsal foot), and S1 (sole).
Reflexes: Test patellar (L2-4), Achilles tendon (S1), and adductor (L2-4) reflexes and Babinski’s.
Hip provocation tests: FABER and IROS are sensitive for hip but not specific.
Sacroiliac joint tests: The SIJ is a b… – you need 3 out 5 maneuvers! The following are performed on the back: Gaenslen, thigh thrust, distraction

Prone

Inspect and palpate for atrophy.
Assess tenderness of spinous processes, facet joints, and sacroiliac joints.
Test the medial hamstring reflex (see below)
SIJ tests: sacral thrust

Lateral decubitus

FAIR or other piriformis test
Only if ruling out conus/cauda: Conduct perianal sensory testing or rectal exam if necessary.
SIJ tests: compression test

Final observations

Observe as the patient stands and dresses while washing your hands.

The Medial Hamstring Reflex

Once upon a time, reflex testing L5 was elusive. Sure, there were rumors of the tibialis posterior reflex, but it felt like Bigfoot sightings: lots of hearsay, no evidence. Enter the Medial Hamstring Reflex (MHR), the unsung hero of L5 testing.

It’s simple. Put the patient in the prone position, press down on the tendon just above the knee’s medial side, and use a proper reflex hammer (hint: not the flimsy American ones). If the reflex doesn’t appear, try raising the calf slightly with your knee. Voilà, the elusive MHR springs to life.

I’ve yet to see a patient with an Achilles Tendon Reflex (ATR) and no MHR where L5 wasn’t involved. There’s even data to back this up and Practical Neurology published a neat case report on it recently.

Summary

That is all. It is a lot. What do you think? I would love to hear from you, especially if you are working in the field, or even better: if your are not.

Intimate musings

January 6, 2025January 7, 2025 jneuro Neurology, Radiology Atherosclerosis, cholesterol, health, Medicine, Neurology, nutrition, Radiology, Stroke, Ultrasound

Imagine you’re an experienced cerebrovascular sonographer. Sitting in front of you is a middle-aged healthy patient you’ve somehow managed to coax into biannual ultrasound sessions. Last time, you measured their Intima-Media Thickness (IMT) at 0.7 mm – the insurance company was happy – and this time it’s 0.9 mm. So, what now? Statin? Yoga? Or just a new measuring tape?

The IMT is the Swiss Army knife of vascular medicine: perfect for crafting pharmacological studies, a brilliant surrogate marker for atherosclerosis – just like LDL, obscure lipid ratios, or blood pressure. It correlates with cardiovascular events, at least back in the preprevention era of the 1990s to 2010s, before lifestyle interventions became a serious hobby. Population-based analyses show: IMT correlates with the risk of atherosclerotic events – in the brain, maybe the heart, but also with age, sex, blood pressure, blood sugar, cholesterol, and everything else on the checklist. So, no surprise it became a standard part of cerebrovascular ultrasound.

And that’s where the fun begins: now even those who can’t spell the difference between Communis and Interna can use IMT as an excuse for invasive diagnostics – from screening coronary angiograms to scintigraphies. The worried middle-aged need somewhere to channel their leisure-time anxiety, after all.

Does the literature support this?

Yes and no. Or rather, not in the way you’d need it to. Despite decades and many tens of thousands of patients in studies, there’s (to my knowledge) no research proving that basing treatment decisions (like medication) on IMT, or even tracking its progression, provides a patient benefit in terms of cardiovascular events or mortality. Instead, you get lovely odds ratios for populations – because they look so delightfully significant above 1. Broken down to the individual level? The best meta-analysis claims to analyze individual patient data, but those odds ratios still come from pooled calculations.¹ And in the end, it doesn’t even predict actual events – just plaques, the surrogate for a surrogate.

In this evidence-starved desert, all we’re left with is pathophysiological imagination.

So what is this Intima-Media-Thickness?

IMT is the distance between two lines in an ultrasound scan of the carotid arteries, using a linear probe in a longitudinal view. The two parallel lines represent the leading edges of two anatomical barriers: lumen/intima and media/adventitia.

IMT isn’t supposed to have anything to do with plaques – plaques are focal lumen-narrowing structures in the carotids that, according to the Mannheim (Hennerici)² definition, are at least 0.5 mm thick, occupy 50% of the surrounding Intima-Media complex, or are over 1.5 mm thick. The underlying assumption is that the Intima-Media complex shouldn’t normally be that thick. But of course, that’s a matter of definition.

Physical Limits

Using a typical 7–11 MHz linear probe, and assuming a typical sound speed in tissue of c = 1540 m/s, we calculate

λ = c/f = 1540 m/s / 9 × 10⁶ = 171 μm = 0.171 mm

Axial resolution is typically half the wavelength: 0.171 / 2 = 0.085mm.
The lumen/intima boundary is anatomically only 50–100 µm (0.05–0.1 mm) thick and cannot be resolved ultrasonographically. On the other hand, the media/adventitia boundary is more clearly visible thanks to its bright echo. Both boundaries, however, are constrained by the axial resolution limit of 0.085 mm, resulting in a physical measurement uncertainty of 2 × 0.085 = 0.17 mm.

Of course, you could drastically improve this with automation, higher-frequency probes (e.g., 17 MHz), and so on. But we need to accept clinical practice as a compromise – otherwise, reliability suffers. Some ultrasonographers claim IMT recognition exceeds these physical limits, and I even believe them.³ But ultimately, physics imposes a hard limit of 0.15 mm on measurement precision.

Methodological Questions

In principle, the method is clear. In practice, though, you have to decide where and how to measure IMT:

CCA or ICA? The common carotid artery (CCA) is much easier to image because it runs parallel to the probe, unlike the ICA, which meanders unpredictably. But atherosclerosis tends to develop in the ICA, not the CCA, so it’s a compromise between validity and reliability.
Near or far wall? The near wall looks great in ultrasound, but some esoteric sonographic secrets argue against using it.³ Many studies, unfortunately, ignore this advice.
IMT or plaques? Be careful not to accidentally include an early plaque in your IMT measurement. While plaques have a distinct configuration and brightness, anything can happen if you set an unsupervised tech on the task.
Left or right? Which side to measure? Or both, then average them? In healthy volunteers, it doesn’t seem to matter, but in actual patients, it might. I always start with the left side and measure only there. Is this optimal? Probably not.
Single point or entire segment? Overview articles insist it’s better to measure an entire segment rather than a single point, but there’s little evidence supporting this. If you do measure a segment, should you use the mean or the maximum value? Both approaches have merit.
Where in the CCA? If you’re measuring at just one point, the Mannheim consensus suggests 5 mm below the bifurcation. But what if there’s a plaque there? Do you just switch to the other side?
Measurement point? How precisely do you place your measurement points? Do you go for the middle between the dark and bright region? Or stick to the border below it? And how do you measure the distance – vertically (y-axis) or diagonally (if the vessel’s at an angle)? Do you pick the thickest spot (as long as it’s not a plaque) or the thinnest?
Measurement precision? How exact should the result be? Down to 2–3 decimal places? Or just one, like 1.1 mm?
Cardiac cycle? The Mannheim consensus says to measure at the end of diastole. But who actually knows when that is, especially with an irregular heartbeat?
Blood pressure? Since blood pressure directly correlates with IMT, you should always measure at the same blood pressure level or at least note it on your report and forget that it differed.
Vessel diameter? The Mannheim consensus also recommends recording the vessel diameter itself, because, naturally, it might have something to do with the IMT (cue balloon model, Laplace, and friends). But what if the diameter isn’t the same as last time? When do two vessel diameters count as “equal”?

Reproducibility and Comparability

Even with a precise protocol, today’s training and staffing variability make consistent measurements nearly impossible – the bane of ultrasound reliability.

Physical measurement error: As mentioned, at least ±0.17 mm.
Intra-rater variability: Studies report deviations of 0.03–0.1 mm, depending on expertise and caffeine levels.
Inter-rater variability: Differences between observers add another 0.1 mm or so.
Variability in vessel diameter: Temperature, blood pressure, relaxation, or infection can all cause fluctuations.

Combined, errors can easily total 0.3–0.4 mm. Welcome to the world of ultrasound.

Individual Changes in IMT

So, is an increase in IMT from 0.7 to 0.9 mm in your patient cause for alarm?

IMT depends on environmental variables, like current and chronic blood pressure, temperature, hormonal fluctuations, and possibly even medication effects (e.g., statins).
IMT is operator-dependent and shows significant variability due to physical and measurement-related factors. These can be averaged out in epidemiological studies, but they’re still a headache in practice.
IMT correlates with cardiovascular events only at the population level, not on an individual basis for predicting the risk of vascular heart or brain diseases.
IMT can only be tracked over time if it’s measured under identical conditions (e.g., mean vs. max, left vs. right vs. mean, CCA vs. ICA, far wall vs. near wall, etc.).
The more often you measure IMT, the greater the uncertainty. Precision doesn’t improve with frequency – it just gets blurrier.
Threshold effects: If you only focus on IMTs that are “pathological,” you risk falling into the traps of anchoring bias and regression-to-the-mean effects.

If you retreat to the point where you only evaluate data from carotids you’ve personally scanned, taking all these imponderables into account – measuring at exactly the same blood pressure and time of day – what threshold for a change in IMT would you consider acceptable? Based on the arguments above, I’d say 0.3 mm is reasonable.

And that means, from a change of 0.3 mm onward, I’d suggest lifestyle adjustments: more exercise, weight loss, healthier eating, more movement, more joy, and more friends. Which brings us to the key point: I could’ve made all those recommendations without even picking up the probe.

What about a statin? In primary prevention, given the number-needed-to-treat, I’d be pretty cautious about starting a new medication based solely on an ultrasound finding. What do you think?

References

Tschiderer, L. et al. Association of Intima‐Media Thickness Measured at the Common Carotid Artery With Incident Carotid Plaque: Individual Participant Data Meta‐Analysis of 20 Prospective Studies. J. Am. Heart Assoc. Cardiovasc. Cerebrovasc. Dis. 12, e027657 (2023).
Touboul, P. et al. Mannheim Carotid Intima-Media Thickness and Plaque Consensus (2004–2006–2011): An Update on Behalf of the Advisory Board of the 3rd and 4th Watching the Risk Symposium 13th and 15th European Stroke Conferences, Mannheim, Germany, 2004, and Brussels, Belgium, 2006. Cerebrovasc. Dis. Basel Switz. 34, 290–296 (2012).
Wikstrand, J. Methodological considerations of ultrasound measurement of carotid artery intima–media thickness and lumen diameter. Clin. Physiol. Funct. Imaging 27, 341–345 (2007).

Allegorical cavepersons and radiologists

January 5, 2025January 5, 2025 jneuro Professionalism, Radiology Philosophy, plato, Radiology

A comment on Plato’s Republic, Book 7

And now, I said, let me show in a figure how far our nature is enlightened or unenlightened: –Behold! human beings living in a underground den, which has a mouth open towards the light and reaching all along the den; here they have been from their childhood, and have their legs and necks chained so that they cannot move, and can only see before them, being prevented by the chains from turning round their heads. Above and behind them a fire is blazing at a distance, and between the fire and the prisoners there is a raised way; and you will see, if you look, a low wall built along the way, like the screen which marionette players have in front of them, over which they show the puppets.
I see.

What holds the (neuro)radiologist in chains? It is the endless lists of reports and reviews, the steady stream of seemingly vital duties that prevent us from perceiving the fires of the wards and emergency rooms. We see the shadows of the puppets only as grayscale images, measured in Hounsfield units.

And do you see, I said, men passing along the wall carrying all sorts of vessels, and statues and figures of animals made of wood and stone and various materials, which appear over the wall? Some of them are talking, others silent.
You have shown me a strange image, and they are strange prisoners.

The poor clinicians, pitiful as they whisper among themselves or, in trance-like shock, parade patients, fates, and dreadful diseases before us. All we see are illusions, phantasms, allegories. What strange prisoners we are!

Like ourselves, I replied; and they see only their own shadows, or the shadows of one another, which the fire throws on the opposite wall of the cave?
True, he said; how could they see anything but the shadows if they were never allowed to move their heads?

Why does the radiologist remain in their cave? Why not turn their head, take a step into the emergency department, the trauma bay, or at least the CT suite? The chains are self-imposed; the cave, comfortable. Yet the clinic is more than shadows – it is a pulsing flame.

And of the objects which are being carried in like manner they would only see the shadows?
Yes, he said.

Of course, the impression of the clinicians is itself but a pale reflection of true medical reality. Nonetheless, the radiologist’s perspective is unique, if rarely colorful.

And if they were able to converse with one another, would they not suppose that they were naming what was actually before them?
Very true.

So the clinicians talk among themselves, philosophize, speculate, make decisions, and use terms unfamiliar to us.

And suppose further that the prison had an echo which came from the other side, would they not be sure to fancy when one of the passers-by spoke that the voice which they heard came from the passing shadow?
No question, he replied.
To them, I said, the truth would be literally nothing but the shadows of the images.
That is certain.

So, we never truly know whether we interpret the images alone or their relation to the clinic – the clinicopathological correlation. We act as though CTs and MRIs provide objective information, all the while knowing they are but faint impressions of reality.

And now look again, and see what will naturally follow it’ the prisoners are released and disabused of their error. At first, when any of them is liberated and compelled suddenly to stand up and turn his neck round and walk and look towards the light, he will suffer sharp pains; the glare will distress him, and he will be unable to see the realities of which in his former state he had seen the shadows; and then conceive some one saying to him, that what he saw before was an illusion, but that now, when he is approaching nearer to being and his eye is turned towards more real existence, he has a clearer vision, -what will be his reply?

The confrontation with reality. Nowhere do we come closer to the “real world” than in ultrasound. Here, we are forced to engage directly with the patient’s anatomy – hands palpably on the probe, the tangible connections of medical understanding. The pain of sudden realization: A clinically uninformed radiologist is like a prisoner forced into the light. The images alone are not enough.

And you may further imagine that his instructor is pointing to the objects as they pass and requiring him to name them, -will he not be perplexed? Will he not fancy that the shadows which he formerly saw are truer than the objects which are now shown to him?
Far truer.

The arduous path of radiological training. Education in radiology is not just about mastering the art of imaging but also about understanding the underlying clinical realities. This journey is painful because it demands more than merely analyzing pixels. It requires a deep understanding of diseases.

And if he is compelled to look straight at the light, will he not have a pain in his eyes which will make him turn away to take and take in the objects of vision which he can see, and which he will conceive to be in reality clearer than the things which are now being shown to him?
True, he says.

The confrontation with clinical reality hurts us because we are not masters of it, and because it tarnishes the magical clarity of our cross-sectional images.

And suppose once more, that he is reluctantly dragged up a steep and rugged ascent, and held fast until he ‘s forced into the presence of the sun himself, is he not likely to be pained and irritated? When he approaches the light his eyes will be dazzled, and he will not be able to see anything at all of what are now called realities.
Not all in a moment, he said.

A well-known phenomenon: In comparison with clinical reality, the clear view of an MRI becomes a distorted image, the purity of morphology contrasted with the messiness of contradictory information.

He will require to grow accustomed to the sight of the upper world. And first he will see the shadows best, next the reflections of men and other objects in the water, and then the objects themselves; then he will gaze upon the light of the moon and the stars and the spangled heaven; and he will see the sky and the stars by night better than the sun or the light of the sun by day?
Certainly.

This may be exaggerated and perhaps too daunting – but do you interpret a CT, an angiogram, or a fluoroscopy differently once you are aware of the patient, rather than merely addressing their shadow on a list of cases to be reported?

Last of he will be able to see the sun, and not mere reflections of him in the water, but he will see him in his own proper place, and not in another; and he will contemplate him as he is.
Certainly.
He will then proceed to argue that this is he who gives the season and the years, and is the guardian of all that is in the visible world, and in a certain way the cause of all things which he and his fellows have been accustomed to behold?
Clearly, he said, he would first see the sun and then reason about him.

Thus, you begin to philosophize not about the images themselves, but about their significance in treatment.

And when he remembered his old habitation, and the wisdom of the den and his fellow-prisoners, do you not suppose that he would felicitate himself on the change, and pity them?
Certainly, he would.

That sounds more arrogant than necessary. It is the perspective on the clinic, the relevance, the consequences of radiological work that motivates us, every day anew, to approach the images with uncertainty, provocation, and curiosity.

And if they were in the habit of conferring honours among themselves on those who were quickest to observe the passing shadows and to remark which of them went before, and which followed after, and which were together; and who were therefore best able to draw conclusions as to the future, do you think that he would care for such honours and glories, or envy the possessors of them? Would he not say with Homer, “Better to be the poor servant of a poor master”, and to endure anything, rather than think as they do and live after their manner?
Yes, he said, I think that he would rather suffer anything than entertain these false notions and live in this miserable manner.

This does not mean that radiological discussions, whether daily or in specialized journals, lack their charm and challenge. Yet as long as they take place within their own domain, their cave, they remain shadow plays. Whether a particular neuroradiological sign is characteristic of PSP is not determined by the gold standard of another MRI finding, but by the clinical consequence and the practical benefit for the patient.

Imagine once more, I said, such an one coming suddenly out of the sun to be replaced in his old situation; would he not be certain to have his eyes full of darkness?
To be sure, he said.

This, then, is the true challenge: to walk in both worlds, to take the shadows, their limitations, their regularities, and their dimensions seriously, to live within them while simultaneously respecting their multidimensionality.

And if there were a contest, and he had to compete in measuring the shadows with the prisoners who had never moved out of the den, while his sight was still weak, and before his eyes had become steady (and the time which would be needed to acquire this new habit of sight might be very considerable) would he not be ridiculous? Men would say of him that up he went and down he came without his eyes; and that it was better not even to think of ascending; and if any one tried to loose another and lead him up to the light, let them only catch the offender, and they would put him to death.
No question, he said.

OK, we haven’t gone that far yet.

Emotion is cognition – but does that feel true?

January 4, 2025 jneuro Uncategorized brain, cognition, depression, emotion, Stroke

I was genuinely proud of myself when I not only stumbled upon Lisa Feldman Barrett’s “Theory of Constructed Emotion” but also felt like I might have actually understood it. The gist: our emotions aren’t external institutions (à la Freud) but rather an integral part of our thinking. Cool, I thought – until the next punch landed. Emotions themselves? Nothing but cognition! The difference we feel between the two is merely an evolutionary placebo, a trick that somehow turned out to be useful for gene survival. Which raises a fascinating question: what was our good old evolution thinking—or better yet, feeling—when it came up with the idea that the two concepts are opposed to eachother?

Perhaps I’m hopelessly behind the times, but this doesn’t mesh well with the neuropsychological cognition toolbox I grew up with. My practical neurological prejudices about emotions will need some serious time to adjust to iconoclasm.

Emotional lesions in the brain?

Take lesions in the “classical limbic system,” for instance. Think cingulum, amygdala, hypothalamus, and all the other structures that Paul Broca identified in 1875 (and James Papez 60 years later) as part of the emotional system, before MacLean slapped the label “limbic” on it post-war, and everyone nodded along. Connections between elements of the Papez circuit? Sure, but connections are hardly exclusive to the limbic system—just about every brain region chats with every other, smalltalk.

In my younger days as a neurologist, I learned about pathological laughter and crying, Klüver-Bucy, and the classic fear seen in temporal lobe seizure aura. I’ve even witnessed the emotional rollercoasters brought on by thalamic strokes—not to mention postictal dysphoria or the way deep depression sometimes seems electrically accessible. But let’s be honest: for these “emotional” phenomena, it’s remarkably difficult to separate out the cognitive effects. Case in point: the legendary overeating Klüver-Bucy trauma patient (yes, there are a few documented cases, though they feel like an urban neurological myth). After much searching, I’ve yet to find solid counterexamples. Worse still, many “affective disorders” are closely tied to cognitive impairments, whether acutely (pseudodementia) or in the long term (increased conversion rates from depression to dementia).

Is this relevant?

For the acute treatment of strokes or psychiatric conditions, probably not much, I think/feel. But rehabilitation? That’s where we may need a paradigm shift. Psychologists focus on emotional aftermaths, while occupational and physical therapists work on cognition, movement, and employability. This divide might need rethinking.

And in the follow-up care for brain lesions—be it trauma, subarachnoid hemorrhage, stroke, or a Parkinson’s crisis—the same principle applies: a patient with apparent emotional issues (e.g., post-stroke depression) will likely have cognitive deficits as well, and vice versa. Chronic neurodegenerative diseases? Same story.

What about our neat dichotomy in drug treatments? That might also need revisiting. Emotional problems? SSRIs or stronger antidepressants. Cognitive impairments? Uh … not much, maybe some cholinesterase inhibitors, memantine, amantadine, or modafinil. Is that really the best we can do? Interestingly, meta-analysis show SSRIs do have positive effects on cognition—but only in depressed patients. Who knows if this has been studied systematically enough? A quick glance at the theoretical sections of related studies (short as they are nowadays) reveals that the separability of emotion and cognition is a foundational premise. This is even more pronounced in research on improving cognition in major depression and similar conditions.

And what about you?

Dear reader, who has bravely endured this blog-length challenge far exceeding any tweet: the cognitive strain of this text cannot have left you emotionally unaffected. Perhaps you feel smarter. Or frustrated. But hey—that’s exactly what makes the interplay of cognition and emotion so fascinating, isn’t it?

From a glance

January 2, 2025 jneuro Neurology, Radiology

For the light-sensitive neuroradiologist, whose eyes are accustomed to the dark and who thrives best in the solitary hush of a reporting room, patient contact is practically corrosive. Thrust into the glaring lights and cacophony of hospital operations, they feel like a mole dragged into midday sunlight. Yet, even this delicate creature requires at least a rudimentary understanding of the patient’s condition to craft their diagnostic masterpiece. No report without understanding.

And so, this timid being occasionally ventures into the CT room, braving the bustling hive of radiologic technologists, nerve whisperers, nurses, and other harried clinicians. These individuals may have spoken to, touched, or even probed the patient, thus gaining valuable insights—but articulating those insights? That’s a different story. They’d rather tell you a score than recall whether the patient was able to stand, frail, diapered, or a hulking powerhouse.

Not wanting to steal anyone’s time or, heaven forbid, spark an actual conversation, our shy radiologist might snatch a brief glance through the console window at the CT table. And what do they see? Feet. And movement. It’s remarkable just how much information can be packed into these fleeting moments.

Movement

A restless patient, as any radiologist knows, is the sworn enemy of any scan. There’s an old radiological adage: never examine them who actively resist examination. And its corollary: an uncooperative patient has no business in the gantry. The supposed time savings of a quick but blurry CT scan rarely translate into actual benefits for the patient.

“I don’t see any major bleeding” may sound reassuring, but it carries the uncomfortable subtext that another radiologist or better examination might indeed find major bleeding. And let’s not forget, even small bleeds are treacherous, both for the patient and for the neurologist, because they tend to invite other departments, or worse, a breathing tube into the fray. A CTA or CTP for someone resisting with head, hands, or feet? Absolutely not—unless, of course, the neuroradiology department has a vested interest (say, it’s one of their own patients).

Quick sedation? Easier said than done. Propofol, the gold standard for swift and effective sedation, has no business in the hands of amateurs or in the poorly monitored CT setting. Midazolam often does more harm than good, amplifying delirium and Cheyne-Stokes respiration while also taking its sweet time to kick in. Ketamine, that trusty old friend, is a safer bet but sadly spends most of its life collecting dust on the shelf. Administering such “pharmaceutical courage” requires an anesthesiologist (or equivalently fearless internist), proper monitoring, and—crucially—time, all of which are scarce commodities in the CT suite.

Of course, not all movement is created equal. Involuntary movements, like convulsions and myoclonus—slight or dramatic, up to full-body jerks—are a sure sign of investigational futility and underdosed anticonvulsants. In contrast, the humble tremor rarely causes leg-centric mischief. Then there’s the symphony of coughs and snores, each a clarion call for an artificial airway, or the pain-driven flinches that demand effective analgesia and all the monitoring and support that entails.

Stillness can be just as telling. A patient who’s too quiet—unmoving, or, slightly worse, pulseless—offers their own grim clues. What pleases the technologist and radiologist isn’t always in the patient’s best interest. If contrast medium for a CTA is taking its time, it’s worth ensuring the ECG signal on the distant monitor still translates into actual blood pressure—and life. Here, an oxygen saturation or, optimally, an arterial blood pressure curve speaks volumes.

Feet

Clothed feet are a diagnostic blessing—a simple rule: those with socks are generally okay. The seriously ill shed their warming layers for the wings of a hospital gown faster than you can say “admission.” This grants curious observers a treasure trove of insights: body habitus, mobility, care level, venous status (handy for assessing IV access readiness), femur fractures (hello, external rotation), hemiparesis, and even Babinski reflexes.

Speaking of Babinski: in the two seconds it takes to scratch for the reflex, a radiologist can substantiate an untrusted neurologic exam and assess the patient’s reaction to unpleasant stimuli, like, incidentally, that from a contrast injection. This critical tidbit belongs in the case notes, where future radiologists might stumble upon it like a hidden gem. After all, cryptic clinical details like “code stroke” or “shock room” are woefully inadequate.

A holistic view on the images

Ultimately, a holistic—not merely radiological—perspective on the images often reveals profound insights into the patient’s condition:

The Eyes: Eyes reflect consciousness (e.g., divergent gaze) and provide clues to focal pathology, from forced deviation (stroke, postictal) to contraversion (ictal, pons) and nystagmus (vestibular, rarely brainstem). While pupil size isn’t discernible in scans, the crystalline lens’s density can hint at anisocoria.
The atlantodental joint: This joint, along with (if present) the cervical spine, offers a wealth of information about the patient’s head and, a fortiori, overall mobility.
Facial 3D Reconstruction: A volume-rendered reconstruction with the right windowing reveals the patient’s face, complete with age (wrinkles), facial symmetry (watch for peripheral facial palsy), hematomas, and more.

Left-sided Bell’s palsy in VRT of CCT (not indicated)

Clinical Reasoning: Uncertainty

October 3, 2023October 3, 2023 jneuro General medicine Clinical reasoning, Medicine, Schema, Uncertainty

As one of my pet projects, I am part of our local Clinical Reasoning Education Team in Augsburg (called CREDA, yeah, I know, hate acronyms, too), where we create educational experiences as part of the curriculum to improve clinical thinking and clinical reasoning skills – we try to separate what goes on in your head (clinical thinking) and what you do as a team (you reason together).

Dealing with uncertainty has a long tradition as a professional skill that obviously needs to be developed, although there is only indirect evidence for it. As the great clinical problem solvers we like to offer our students simple (and usually oversimplifying) schemas, so we started GPT-4 to come up with an acronym (I hate acronyms but what else is there?) for such a schema, based on Gheihmann’s great 2019 article on uncertainty: BIG HERO – follow up! It is an awful acronym and there must be a better and more english solution, that includes the patient earlier (step O) and has the Essential step before the Heuristics. Can I challenge both of my readers to come up with one?

B auchgefühl (gut reaction): What is my intuition?
I dentify: State your uncertainty
G et help: Team up! Don’t worry alone!
H euristics: What are the potential biases and errors?
E ssential: Is it knowable? Is the uncertainty reducable?
R oadmap: Safety netting – What if I’m wrong?
O pen up: Include the patient into your plan
Follow-up

LP, CT, ICP and mass effect

April 6, 2022April 6, 2012 jneuro Uncategorized

We discuss 4 scenarios:

R/o meningitis
Very sick patient with probable bacterial meningitis
R/o SAH
http://www.ncbi.nlm.nih.gov/pubmed/17712055

For each case there are some ways to make sure that your LP does not lead to herniation, and, of course, for each case there are exceptions to the rule.

References: