Name: Carpal Tunnel Syndrome: Not Just Typing, and Surgery Actually Works
Creator: Realised
Published: 2026-06-01

Summary

The office-folklore model of carpal tunnel syndrome — "it's from typing, so buy an ergonomic keyboard, a wrist rest and some vitamin B6" — gets the cause, the gadgets and the supplement wrong at once: keyboard and mouse use is a weak driver next to obesity, type-2 diabetes, pregnancy, hypothyroidism and the size of your wrist tunnel, and the ergonomic-device market and B6 have little-to-no evidence (high-dose B6 is itself neurotoxic) — while the genuinely effective moves are cheap and, unusually for this platform, include surgery: a nocturnal neutral-position wrist splint and a time-boxed cort

Why Moderate

Moderate Evidence because the entry's load-bearing therapeutic claims rest on Cochrane RCT syntheses (injection, splinting, surgery-versus-splinting) and a GRADE-based specialty guideline (AAOS surgical indication), but each carries a real limit: the injection benefit is short-window, the splint's long-term data is thin, and the surgical advantage over splinting is sub-threshold for mild-to-moderate disease. This is solid, gold-standard-for-the-question evidence with genuine caveats — not a single undisputed axiom, and not a handful of suggestive studies.

NOT Strong for the headline, because the central surgical claim comes with the honest Cochrane caveat (sub-clinically-important advantage in mild-to-moderate cases), and the conservative levers are short-term and best-model-unknown. The severe/axonal-loss surgical indication alone leans Strong (AAOS graded strong, independently corroborated), and the entry marks that rather than inheriting the milder-case uncertainty across the board.

NOT Foundational because the entry carries live clinical and commercial judgement — a both-ways tension between over- and under-treatment — rather than a settled principle.

The per-lever split (read this, not just the headline):
• Corticosteroid injection (temporary benefit): Moderate (Cochrane RCTs).
• Nocturnal neutral-position splint (short-term benefit): Moderate (Cochrane; weak long-term/duration data).
• Surgery vs splinting, mild-to-moderate: Moderate, with the sub-threshold caveat that anchors the honesty of the title.
• Surgical release for severe/failed-conservative/axonal-loss: Moderate leaning Strong (AAOS graded strong).
• Risk factors dominated by obesity/diabetes/anatomy; keyboard link weak: Moderate.
• Ergonomic gadgets ineffective: weak-to-unsupported debunk.
• Vitamin B6 ineffective and neurotoxic at dose: weak-to-unsupported debunk with a harm flag.

Practical takeaway

The framing to hold: carpal tunnel is mostly a pressure-in-a-fixed-tunnel problem driven by your metabolism, hormones and anatomy — not a verdict on how much you type. Conservative levers genuinely help mild-to-moderate cases; surgery is the well-evidenced, often definitive fix once it is moderate-to-severe, persistent, or damaging the nerve. The gadgets and the B6 tablet are where the money goes and the evidence does not.

Start conservative for mild-to-moderate (in rough order of evidence).
• Nocturnal neutral-position wrist splint. Wear a splint that holds the wrist near neutral at night — the highest-yield first move for mild-to-moderate symptoms. No specific model is proven best; neutral position is the active ingredient.
• Activity modification. Reduce sustained forceful, repetitive or vibrating wrist tasks where you can — this is where genuine occupational load (not office keyboarding) lives.
• A time-boxed corticosteroid injection. One injection gives real but temporary relief (up to about six months in milder disease) and can reduce the need for surgery over the following year. Treat it as a window to reassess, not a cure; a second injection adds little.

Address the actual drivers.
• Weight, blood sugar, and thyroid. Obesity, type-2 diabetes and hypothyroidism raise risk through the tunnel's contents; these are the levers the typing story ignores. (Pregnancy-related carpal tunnel often resolves after delivery.)

Stop buying the things that do not work.
• Skip the ergonomic-gadget cure. Vertical mice, engineered wrist rests and split keyboards change posture without reliably lowering tunnel pressure, and no prevention device is proven to work. General desk-ergonomics and posture questions are owned by physical_counter_modern_postures; for treating carpal tunnel, they are not the answer.
• Skip vitamin B6 for carpal tunnel. No good evidence it treats the condition, and high doses can cause their own nerve damage.

Know when conservative is not enough — and act.
• If symptoms are moderate-to-severe, persist despite a 6-to-12-week conservative trial, or the numbness is now constant — see a clinician about surgical release. For this stage the operation is well-evidenced and often definitive. This is a case where surgery is the right answer, not over-treatment.
• If the thumb-base muscle is visibly wasting or the numbness never lets up, do not keep trialling products. That is nerve damage; get assessed promptly (see Risks).

Evidence detail

Why This Entry Exists

Carpal tunnel syndrome is misadvised in two opposite directions at once, and both errors cost people function. The first error is the cause story and the shopping list it generates: "you typed too much, so fix your desk." That sends people to buy vertical mice, split keyboards, wrist rests and pyridoxine tablets, none of which has good evidence, while leaving the actual drivers — body weight, blood sugar, thyroid, pregnancy-related fluid, and the plain anatomy of a narrow carpal tunnel — untouched. The second error is more dangerous and runs the other way: a reflexive "avoid surgery, keep trying conservative things" that, applied to the wrong patient, lets the median nerve keep dying while the thumb-base muscle quietly wastes. For moderate-to-severe or persistent carpal tunnel, surgical decompression is one of the genuinely definitive operations in musculoskeletal medicine, and under-stating that to keep someone in an indefinite gadget-and-splint loop is the clinically harmful failure mode here.

So this entry holds both truths. The conservative levers are real and worth doing first for mild-to-moderate disease: a night splint that holds the wrist neutral, activity modification, and a single corticosteroid injection that buys a real but temporary window. And the surgical reality is also real: when the nerve is under sustained pressure — moderate-to-severe symptoms, failure of a 6-to-12-week conservative trial, thenar wasting, or electrodiagnostic evidence of axonal loss — releasing the ligament is well-supported and often curative. The typing story is not the villain the folk model makes it, and the gadgets and B6 that story sells do not work; the splint, the injection window, and (when indicated) the operation do.

What bad advice this protects against, in all directions:
• "Carpal tunnel is from typing / computer use" → office keyboard and mouse use is a weak, mixed driver; obesity, type-2 diabetes, pregnancy, hypothyroidism and a congenitally narrow tunnel are bigger. (Occupational force, repetition and vibration — meat and fish processing, chainsaw work, electronics assembly — do carry real risk; the desk-keyboard link specifically is the weak one.)
• "Buy an ergonomic keyboard, vertical mouse and wrist rest to fix it" → these alter wrist posture but do not reliably lower carpal-tunnel pressure, and no primary-prevention device has been shown to work in a systematic review.
• "Take vitamin B6 for carpal tunnel" → the positive studies were small, non-randomised and subjective; a cohort found no link between B6 status and symptoms or nerve conduction, and high-dose B6 can itself cause a sensory neuropathy.
• "Surgery is over-treatment / a last resort to avoid at all costs" → for moderate-to-severe, persistent, or nerve-damage cases, release is the well-evidenced definitive fix; delaying it to keep trialling gadgets risks permanent motor loss.
• "Surgery is decisively better, so just operate on everyone" → for mild-to-moderate disease the mean advantage of surgery over splinting on the standard symptom and function scales falls below the clinically-important threshold; conservative-first is legitimate there.
• "Numbness that comes and goes is nothing to worry about" → persistent numbness with thumb-base (thenar) muscle wasting is axonal nerve damage; it needs prompt assessment, not another product.

This entry owns the cause correction (typing is weak; metabolic/anatomic/hormonal drivers dominate), the conservative levers (night splint, activity modification, a time-boxed injection), the surgery-actually-works message with its severity boundary, and the debunk of the ergonomic-gadget and B6 market for carpal tunnel specifically. It defers general modern-posture and desk-ergonomics questions to physical_counter_modern_postures, and generic conservative joint-pain care to joint_pain_conservative_management. It states those boundaries and routes there rather than re-arguing them.

Evidence

Organised by lever, with the tier signal inline. The headline is Moderate, and it splits sharply: the conservative and surgical levers are Cochrane- and guideline-backed, while the gadget and B6 claims are debunks. Read the tiers, not just the thesis.

Corticosteroid injection helps — temporarily (Moderate Evidence).

1. A local corticosteroid injection beats placebo for symptom relief, with benefit lasting months and reduced surgery need out to a year — but it is a window, not a cure. A 2023 Cochrane review found local corticosteroid injection improved carpal tunnel symptoms versus placebo, with benefit in mild-to-moderate disease persisting up to six months and a reduced need for surgery up to twelve months (relative risk about 0.84; 14 trials, 994 participants). An earlier Cochrane review found injection also outperformed oral corticosteroid and that a second injection added no significant benefit over one, with serious adverse events rare. The load-bearing nuance is temporariness: this buys time and can defer or avert surgery in milder disease, but it is not a durable fix. (Ashworth NL, et al. "Local corticosteroid injection versus placebo for carpal tunnel syndrome." Cochrane Database Syst Rev, CD015148, 2023 — 14 trials, 994 participants; benefit vs placebo to six months, reduced surgery need at twelve months (RR ~0.84). The injection-vs-oral-corticosteroid and one-vs-two-injection comparisons come from the earlier Marshall et al. review, Cochrane CD001554. Moderate Evidence — Cochrane-level RCT synthesis, but a short, temporary effect window. Cochrane reviews are independent of injection-product manufacturers; low commercial conflict.)

Nocturnal neutral-position splinting helps mild-to-moderate cases (Moderate Evidence).

2. A night splint holding the wrist neutral is effective short-term and beats no treatment — best model unknown, long-term data thin. Cochrane and the wider orthosis literature find nocturnal wrist splinting effective in the short term for mild-to-moderate carpal tunnel and more effective than placebo or no treatment, plausibly because holding the wrist near neutral (roughly 0 to 15 degrees) at night lowers pressure inside the carpal tunnel during the hours symptoms usually strike. No single orthosis model is proven superior, and guidance on how long to wear it is limited. (Cochrane orthosis review plus orthosis literature review, PMC9847107, 2023; neutral splint 0–15 degrees. Moderate Evidence — consistent short-term benefit, weak long-term and duration data. Splints are cheap commodity items with low industry incentive to inflate the signal.)

Surgery works — and the honest caveat that keeps the title from overclaiming (Moderate Evidence).

**3. Surgery gives a higher rate of clinical improvement than splinting, but for mild-to-moderate disease the mean advantage sits below the clinically-important threshold.** A 2024 Cochrane update (14 trials, n=1,231, 84% female) found surgery produced a higher rate of clinical improvement than splinting at 6 to 12 months, yet the mean between-group difference on the Boston symptom-severity scale (about 0.26) and functional-status scale (about 0.36) fell below accepted clinically-important thresholds (moderate certainty), and quality of life was not clearly better (low certainty). This is the finding that keeps the entry honest: it means surgery is not decisively superior for mild-to-moderate cases, so conservative-first is defensible there — the surgical case rests on severity, not on beating splints across the board. (Cochrane. "Surgical versus non-surgical treatment for carpal tunnel syndrome." CD001552, 2024 update; 14 trials, n=1,231. Moderate Evidence — this caveat is the counterweight to over-recommending surgery. Independent Cochrane synthesis, a counterweight to fee-for-service surgical incentive.)

4. The specialty guideline recommends surgical release, on strong evidence, for the severe end. The American Academy of Orthopaedic Surgeons (AAOS) guideline gives a strong recommendation for surgical release of the transverse carpal ligament when conservative measures fail (typically after a roughly 6-to-12-week trial), when thenar atrophy is present, or when electrodiagnostic testing shows severe axonal loss. Thenar atrophy is strongly associated with carpal tunnel syndrome, though its absence does not rule the condition out. Read together with finding 3, the picture is clean: conservative-first for mild-to-moderate; surgery is the well-evidenced, often definitive move once the disease is moderate-to-severe, persistent, or showing nerve damage. (AAOS Clinical Practice Guideline, "Management of Carpal Tunnel Syndrome," 2016, updated 2024. Moderate Evidence leaning Strong for the severe/axonal-loss surgical indication. AAOS is a surgeons' professional body — a mild pro-surgical conflict — but the severe-case recommendation is graded strong and is independently corroborated.)

Vitamin B6 — a debunk with a harm signal (Emerging / Experimental-grade claim, i.e. weak-to-unsupported).

5. Vitamin B6 (pyridoxine) is not supported as a carpal tunnel treatment, and high doses are neurotoxic. A review of the pyridoxine literature found the studies claiming benefit were small, non-randomised and reliant on subjective outcomes, while a University of Michigan occupational cohort found no relationship between B6 status, carpal tunnel symptoms, or nerve-conduction impairment. Separately, high-dose B6 is a recognised cause of sensory neuropathy — so the supplement most heavily marketed for nerve symptoms can, in excess, produce them. This claim sits at the weak-to-unsupported end and carries an active harm signal. (Nutrition Reviews (Oxford). "Pyridoxine Hydrochloride Treatment of Carpal Tunnel Syndrome: A Review." 62(3):96, 2004; University of Michigan occupational cohort. Weak-to-unsupported debunk with a toxicity flag. B6 is a supplement-industry product — a direct cui-bono to sell an ineffective, potentially harmful pill.)

Ergonomic gadgets — fail on the mechanism (weak-to-unsupported debunk).

6. Ergonomic devices change wrist posture but do not reliably lower carpal-tunnel pressure, and no primary-prevention device is proven to work. A study of a vertical mouse and ergonomic mouse pads found they altered wrist posture but did not reduce carpal-tunnel pressure in patients with the condition — the mechanism the products are sold on simply did not fire. Cochrane found insufficient RCT evidence that ergonomic positioning or equipment treats carpal tunnel, and a systematic review of prevention found no primary-prevention intervention (keyboard design, mouse, wrist rest, workstation set-up) demonstrably effective. This is the evidentiary base for treating the ergonomic-gadget and wrist-rest market as unsubstantiated for carpal tunnel. (Applied Ergonomics, vertical-mouse pressure study, S0003687014001598, 2014; Cochrane ergonomic-equipment review, 2012; Applied Ergonomics prevention systematic review, S000368702030301X / PMID 33453588, 2021. Weak-to-unsupported debunk of the ergonomic-gadget/wrist-rest market. Device makers — $100-plus keyboards, vertical mice, wrist rests — profit from unsubstantiated prevention and cure claims.)

Risk factors — metabolic and anatomic, not the keyboard (Moderate Evidence).

7. The dominant drivers are metabolic, anatomic and hormonal; occupational force/vibration matters, but office keyboarding specifically is weak. A phenome-wide and Mendelian-randomization analysis flagged type-2 diabetes, obesity-related factors, and hormone- and psychiatric-related traits among 19 significant associations with carpal tunnel syndrome. A GRADE meta-analysis of work-relatedness found that force, repetition, vibration and sustained wrist-bending are associated with the condition — with high-risk occupations including meat and fish processing, chainsaw forestry, and electronics assembly — while a systematic review of computer mouse and keyboard use found the office-keyboarding link specifically to be weak and mixed. The defensible claim is therefore narrow: not "carpal tunnel is never occupational," but "office keyboard and mouse use is a weak driver relative to obesity, diabetes, pregnancy, hypothyroidism and tunnel anatomy." (Work-relatedness GRADE meta-analysis, PMC9629628, 2022; computer mouse/keyboard systematic review, PMC2569035, 2008. Moderate Evidence — solid on the obesity/diabetes drivers; the keyboard link is genuinely weak. No strong commercial actor here, though the "typing causes it" folk belief indirectly props the ergonomic-gadget market.)

Mechanism

This entry owns the lever hierarchy and the severity boundary, not the full neurophysiology of nerve compression. What follows is only enough mechanism to make the hierarchy, the ergonomic-gadget failure and the surgical logic intelligible.

Why the tunnel, not the typing, is the story. Carpal tunnel syndrome is compression of the median nerve where it passes through a fixed bony-and-ligamentous canal at the wrist. Anything that raises pressure inside that fixed space — extra soft tissue from obesity, fluid retention in pregnancy or hypothyroidism, connective-tissue and vascular changes in diabetes, or simply being born with a narrow tunnel — squeezes the nerve. The volume of keystrokes is a minor contributor next to what is taking up room in, or shrinking, the canal. This is why the metabolic and anatomic risk factors dominate the epidemiology and the keyboard link is weak: the causal action is inside a fixed compartment, and desk typing does little to change its contents.

Why ergonomic gadgets fail on the mechanism. The devices are sold on the premise that a "better" wrist posture lowers the pressure hurting the nerve. But when this was measured, a vertical mouse and ergonomic pads changed wrist posture without reducing carpal-tunnel pressure in affected patients. Posture and intracanal pressure are not the same variable, and the products act on the wrong one — which is exactly why the prevention trials come back null. The gadget targets the folk-model cause, not the actual one.

Why a night splint and an injection help, but only so far. A neutral-position night splint keeps the wrist out of the flexed and extended positions that spike intracanal pressure during sleep, when symptoms cluster — it lowers the pressure episodically rather than removing the cause. A corticosteroid injection reduces local inflammation and swelling in the tunnel, easing compression for weeks-to-months. Both relieve pressure temporarily; neither enlarges the tunnel or removes the tissue crowding it, which is why their benefit is real but time-limited and why milder disease is where they earn their place.

Why surgery is the definitive move once the nerve is genuinely under threat. Releasing the transverse carpal ligament physically enlarges the canal, permanently lowering the pressure on the nerve rather than nudging it. When compression has been sustained enough to cause axonal loss and thenar (thumb-base) muscle wasting, the limiting factor is ongoing mechanical damage to the nerve — and only decompression removes that. This is why AAOS grades release strongly for the severe and axonal-loss end, and why, for that patient, continuing to trial splints and gadgets is not conservative caution but a delay that risks permanent motor loss. For mild-to-moderate disease, where the nerve is not yet losing axons, the pressure-relieving conservative levers can hold, which is what finding 3's sub-threshold surgical advantage reflects.

Risks And Contraindications

• The load-bearing red flag: persistent numbness plus thenar (thumb-base) muscle wasting is ongoing nerve damage. This is not a symptom to manage with another gadget or an indefinite splint trial. Thenar atrophy and constant (rather than intermittent) numbness signal axonal loss; they warrant prompt medical assessment and, usually, surgical decompression. Delaying release at this stage to keep trying conservative measures risks permanent loss of thumb strength and function. This safety point must not be softened — it is the reason the entry treats surgery as a genuine, sometimes urgent option rather than a last resort.
• The dangerous failure mode cuts both ways. Under-treatment (keeping a severe or nerve-damaged patient in a gadget-and-splint loop) is the harm the red flag guards against. But over-treatment is also real: for mild-to-moderate disease, the 2024 Cochrane update shows surgery's mean advantage over splinting falls below the clinically-important threshold, so operating on everyone is not supported. Reserve "definitive / well-evidenced surgery" for moderate-to-severe, persistent, or axonal-loss cases.
• Do not oversell the conservative levers as cures. The night splint and the injection relieve pressure temporarily; the injection's benefit is a window (up to about six months in milder disease), and a second injection adds little. Present them as time-limited help, not a permanent fix.
• High-dose vitamin B6 is neurotoxic. Beyond being ineffective for carpal tunnel, sustained high-dose pyridoxine can cause a sensory peripheral neuropathy — a supplement sold for nerve symptoms that can produce them. Do not recommend it.
• Do not overshoot the typing debunk into "carpal tunnel is never occupational." Force, repetition, vibration and sustained wrist-bending have a graded association with the condition (meat/fish processing, chainsaw forestry, electronics assembly are high-risk). The defensible, narrower claim is that office keyboard and mouse use specifically is a weak driver relative to obesity, diabetes, pregnancy, hypothyroidism and tunnel anatomy.
• Rule out mimics and get a diagnosis. Numbness and tingling in the hand can also come from cervical nerve-root compression, generalised peripheral neuropathy (e.g. diabetic or B6-toxicity), or thoracic outlet issues. Persistent or worsening symptoms deserve a clinical assessment (including nerve-conduction studies where indicated) rather than self-directed product trials.

Controversy

Nature: a condition with a sticky folk cause ("typing did it") and a matching consumer market (ergonomic gadgets, B6) sitting on top of a genuinely effective clinical toolkit — splint, injection, and, unusually, surgery — with error at both poles: over-treating mild disease surgically on one side, and under-treating severe, nerve-damaging disease conservatively on the other.

Position A — "The cheap levers help and, when needed, surgery is the definitive fix." The actionable take.
• Best evidence: a nocturnal neutral splint and a time-boxed corticosteroid injection genuinely help mild-to-moderate carpal tunnel (Cochrane), and surgical release is strongly recommended and often curative for moderate-to-severe, persistent, or axonal-loss disease (AAOS). Meanwhile "it's from typing," ergonomic gadgets and B6 are weak-to-unsupported, and B6 is neurotoxic at dose. The load-bearing safety point — thenar wasting plus constant numbness is nerve damage needing prompt referral — is not negotiable.
• Where it goes wrong if overstated: it tips into "just operate," ignoring that for mild-to-moderate disease surgery's advantage over splinting is sub-threshold.

Position B — the steelman of the counters.
• (1) Surgery is over-recommended for mild disease. The 2024 Cochrane update found surgery's edge over splinting on the Boston symptom (about 0.26) and function (about 0.36) scales falls below clinically-important thresholds (moderate certainty), and quality of life is not clearly better — so for mild-to-moderate carpal tunnel, durable splinting is a legitimate option and reflexive surgery is over-treatment.
• (2) Work-relatedness is not zero. A GRADE meta-analysis found force, repetition, vibration and wrist-bending are associated with the condition; "not occupational at all" overshoots. The narrow, correct claim is that office keyboard and mouse use specifically is the weak link, not that occupation is irrelevant.

The funding/bias dimension — cui bono, both ways. Toward over-treatment: the ergonomic-device and supplement markets profit from the "typing causes it, buy this" narrative (the vertical-mouse pressure study and the prevention systematic review show the devices do not work; B6 is actively harmful at dose), and a fee-for-service surgical incentive could push release for mild cases — where the honest Cochrane caveat is the counterweight, and AAOS carries a mild pro-surgical conflict as a surgeons' body. Toward under-treatment: an anti-surgery, "conservative-forever" reflex is the dangerous one for the severe patient, deferring a genuinely definitive operation while the thenar muscle wastes. Cochrane sits independent of all three.

Realised Position: Both true, held in tension. Conservative-first — night splint plus activity modification plus a time-boxed injection window — for mild-to-moderate carpal tunnel; surgery is the genuinely definitive, evidence-backed fix for moderate-to-severe, persistent, or axonal-loss cases, a real instance where surgery is not over-treatment. Typing and office-computer use is a weak driver relative to obesity, type-2 diabetes, pregnancy, hypothyroidism and wrist anatomy, though occupational force, vibration and repetition are a genuine smaller factor. Ergonomic gadgets, wrist rests, vertical mice and vitamin B6 are unsupported for the condition (and high-dose B6 is neurotoxic). The non-negotiable: numbness with thenar atrophy is nerve damage — refer promptly, do not keep trialling products.

Cross-Pillar Connections

• Physical (physical_counter_modern_postures): owns general desk-ergonomics, modern-posture and workstation questions; this entry holds only that ergonomic gadgets do not treat carpal tunnel and that office keyboarding is a weak driver, and defers the broader posture discussion there.
• Physical / conditions (joint_pain_conservative_management): owns the generic conservative-care approach to musculoskeletal pain; this entry holds only the carpal-tunnel-specific conservative levers (neutral night splint, activity modification, time-boxed injection) and the severity boundary for surgery.
• Evidence & method (publication_bias_and_evidence_distortion): why the small, subjective, non-randomised positive studies (the B6 and some ergonomic-device claims) over-represent benefit — the mechanism behind keeping those on the debunk list.
• Evidence & method (minimum_effective_dose): the framing for the injection and splint as time-boxed, sufficient-not-maximal interventions, and for not escalating to surgery in mild disease where the marginal benefit is sub-threshold.

What would change our mind

Falsifiability: explicit upgrade/downgrade criteria from source

• We'd take the ergonomic gadgets off the debunk list if a large, low-risk-of-bias RCT showed a device (vertical mouse, engineered wrist rest, split keyboard) actually reduces incident carpal tunnel or lowers carpal-tunnel pressure in affected wrists.
• We'd rehabilitate vitamin B6 on a well-powered RCT with objective electrodiagnostic endpoints — not just subjective pain — showing pyridoxine improves nerve conduction in carpal tunnel.
• We'd broaden the surgical recommendation earlier if a future Cochrane update found surgery's Boston-scale advantage over splinting crossed the clinically-important threshold even for mild-to-moderate disease.
• We'd push to a more conservative default if evidence emerged that early surgery in mild carpal tunnel causes net harm — complication rates or persistent post-surgical pain — exceeding what splinting costs.
• We'd revive the occupational-typing claim we currently downweight if a stronger causal signal (e.g. Mendelian-randomization on cumulative keyboard exposure) linked office keyboarding to incident carpal tunnel.
• What would NOT move us: that a nocturnal neutral splint and a time-boxed injection help mild-to-moderate disease, that surgery is definitive for the severe/axonal-loss end, or that thenar wasting plus constant numbness needs prompt referral. Across all of it, independent (non-seller) funding is the decisive variable — the load-bearing anchors (Cochrane, AAOS) already have it.

Industry bias note

Structural incentives the evidence base may reflect

Cui bono runs in three directions here, and the strongest evidence is conflict-clean.
• The ergonomic-device and supplement markets profit from the "typing causes it, buy this" story. Vertical mice, engineered wrist rests, $100-plus split keyboards and pyridoxine tablets all sell on a cause the evidence downweights and a mechanism (posture, "nerve vitamin") the data does not support — the vertical-mouse pressure study and the prevention systematic review show the devices fail, and high-dose B6 is actively neurotoxic. Weight the independent syntheses over device-maker and supplement-seller claims.
• Fee-for-service surgical incentive could over-recommend release for mild cases. The honest counterweight is built into the entry: the 2024 Cochrane update showing no clinically-important advantage over splinting in mild-to-moderate disease. AAOS, as a surgeons' professional body, carries a mild pro-surgical conflict — but its severe-case recommendation is graded strong and independently corroborated, so it survives the discount.
• The opposite bias — "avoid surgery, conservative forever" — is the dangerous one for the severe patient. It can defer a genuinely definitive, evidence-backed decompression while the thenar muscle wastes and axons are lost. This is not commercially driven so much as culturally driven (a general aversion to surgery), which makes it easy to miss.
• Cochrane sits independent of all three. It has no stake in the gadgets, the supplements, the surgical fee, or the anti-surgery reflex — which is exactly why the Cochrane injection, splint and surgery-versus-splint reviews are the anchors. Net: the load-bearing claims are independent syntheses and a graded guideline, and they cut against sellers and reflexes in every direction.

Carpal Tunnel Syndrome: Not Just Typing, and Surgery Actually Works