RightRep EN · PT Open the app
BLOG

Can You Train With Carpal Tunnel Syndrome? The Wrist-Nerve Audit: Where Atroshi Mailed 3,000 Random Swedes and Found the Tingling in 14.4% but the Confirmed File in 2.7% — and 18.4% of 'Normal' Hands Failing the Nerve Test Anyway; Where the Same Researcher Inverted the Folk Cause — Intensive Keyboard Typists Carrying HALF the Carpal Tunnel of Non-Typists, 2.6% Versus 5.2%; Where Padua Followed 80 Untreated Wrists and Watched Worsening Run INVERSE to Severity, and Ortiz-Corredor's Two-Year File Saw 47.6% Recover on Nothing at All; Where the Cochrane Stack Reads Like a Refund Desk — Splinting's 0.37-Point Null, the Steroid's Real but Expiring −0.77, Surgery's 1.27 Edge With Complications MORE Common at RR 1.38; Where AANEM's 2026 Consensus Crowned the Bland Scale Over 12 Rival Grading Systems; Where the Yoga Trial Raised Grip 162 to 187 mmHg in Median-Age-52 Wrists, the Three-Arm Elastography Trial Improved Symptoms Without Moving the Nerve at All, and Even the Severe Door — Constant Numbness, Thenar Atrophy — Resolved Two-Thirds and Regrew the Thumb Pad 1.2 of 1.4 mm After Release; Where Pregnancy Carries 34% of Third-Trimester Wrists and the Hand Clinic's Fingerstick Found 58.3% of Patients in the Prediabetes Range; and Brazil's 174-Surgeon Congress Census Ran the Same Playbook the World Does: The Nerve Is Measured, Not Guessed, and Training Goes On

By RightRep AI Assistant · · 26 min read

There is a fence at the top of this post and it outranks everything below it: thenar muscle wasting at the base of the thumb, constant — not intermittent — numbness in the thumb, index and middle fingers, or weakness abducting the thumb against resistance is the severe door: nerve studies and a hand surgeon's opinion before any program. The second fence: numbness that respects a neck pattern rather than the thumb-index-middle distribution, or that arrives with neck pain radiating down the arm, is a cervical-spine question wearing a wrist question's clothes — clinician, different file. The third fence: a hand that goes cold, pale or discolored, or numbness spreading to the whole hand, is a vascular door — same-week clinician, not a loading question. The fourth fence: new wrist tingling in the third trimester of pregnancy is common and usually resolves after delivery — but severe or persistent symptoms get assessed, not just waited out, and every treatment decision in this file belongs to a clinician who has examined the wrist. And one honesty clause that outranks both internet camps: this file's most important number is a downgrade — 14.4% of adults report the tingling, but only 2.7% carry the confirmed syndrome [1], and the difference between the two is a measurement, not an opinion [2][3]. With the fences respected, the question underneath is the one every gym already knows: the rest script ("it's pinched, stop gripping, wait") versus the ignore script ("it's just tingling, push through"). The evidence convicts the first and fences the second — this is a measure-first, splint-at-night, keep-training file.

The short answer: train — everything, with a neutral-wrist audit and a night splint, while the nerve gets measured

Strip the file to its spine. The diagnosis is a measurement, not a vibe: the symptoms are common enough to be noise — 14.4% of a random general population reported median-distribution tingling, but only 3.8% were clinically certain and 2.7% clinically and electrophysiologically confirmed — and 18.4% of symptom-free controls failed the nerve test too, which is why the field measures before it programs [1][3]. The folk cause is inverted: in the same population registry, intensive keyboard users carried half the carpal tunnel of non-typists — 2.6% at four-plus hours a day against 5.2% with no keyboard work (prevalence ratio 0.55) — the "typing did this" story runs backwards [4]. Untreated is not doom: 80 untreated wrists followed prospectively worsened inversely to their severity — the mild ones shifted, the moderate ones improved by a third — and in a two-year file of 132 untreated patients, 47.6% recovered clinically while only 23.4% deteriorated [5][6]. The conservative stack is modest and honest: splinting's Cochrane verdict is an insufficient-evidence shrug with a cheap-and-harmless clause and one strong night-wear trial (RR 3.86, NNTB 2) [7]; the steroid injection is real but expiring — SMD −0.77 at three months, benefits to six, surgery deferred not cancelled [8]; against surgery the injection data collapse into uncertainty (I²=95%) [9]. Surgery works, and costs: RR 1.27 better symptom improvement at one year, re-operation 0.04 — but complications significantly more common in the surgical arm (RR 1.38), and the mild-symptoms question honestly open [10]. Exercise moves symptoms without visibly moving the nerve: yoga's 1998 trial raised grip 162→187 mmHg in median-age-52 wrists [12], and a 2026 three-arm randomized trial improved symptoms while leaving nerve elastography exactly where it was — "improvement may occur through mechanisms other than changes in tissue stiffness" [13]. Train legs, press, pull, carry — audit the wrist positions, splint at night, and let the nerve study, not the tingling, set the ceiling [2][3][7].

Who is asking: the scale file, and the measurement that downsizes it

This is the most common entrapment neuropathy of the upper limb, and the census that sized it is the file's first door. Atroshi's team drew a sex- and age-stratified random sample of 3,000 subjects aged 25–74 from a southern Swedish population register, mailed the survey, and got 83% back (n = 2,466, 46% men) — then examined and nerve-tested the symptomatic [1]. The layered result: 354 responders (14.4%, 95% CI 13.0–15.8) reported pain, numbness or tingling in the median-nerve distribution; 94 (3.8%) were clinically certain CTS on examination; nerve conduction found median neuropathy in 120 (4.9%); and only 66 (2.7%, 95% CI 2.1–3.4) carried the confirmed file — both the clinical picture and the electrical signature [1]. Two honesty spines run through those numbers. First, the downsizing: about 1 in 5 symptomatic subjects actually had carpal tunnel syndrome — the tingling alone is not the diagnosis [1]. Second, the false-positive floor: of 125 symptom-free controls who let themselves be tested, 18.4% (12.0–26.3) had electrophysiological median neuropathy anyway — the test sees mild slowing that symptoms never noticed, which is why a number without a clinical picture programs nothing [1]. The gym translation: the 48-year-old whose hands fall asleep at night is standing in a very common crowd, and the crowd is five times bigger than the confirmed diagnosis inside it — the wrist's first training decision is a measurement, not a rest week [1][3].

What the tunnel actually is: the anatomy door, and the 2026 consensus on grading it

The tunnel is a real place with a lid: a rigid arch of carpal bones floored by the transverse carpal ligament, through which the median nerve runs with nine flexor tendons — a fixed-volume corridor where swelling, tendon-sheath thickening or fluid retention raises pressure on the nerve, and the nerve's insulation (myelin) complains first, its axons only later. That geography explains the file's signature pattern — night-time numbness in the thumb, index and middle fingers (the wrist curls while you sleep), shaking-the-hand relief, symptoms that respect the median distribution and stop at the ring finger's halfway line. It also explains why the field's central test is electrical, not pictorial: nerve conduction studies measure demyelination and axonal loss directly — as Bland's review puts it, they are "not a test that determines the diagnosis but a measure of impaired nerve function," the best available indicator of overall severity, with prognostic value for surgical outcome and sensitivity to change under treatment [3]. And the grading of that measurement just got its own consensus: AANEM's 2026 guideline counted twelve published CTS grading systems, surveyed 1,253 practitioners, and found only 43% used nerve-conduction-only grading while 27% used no system at all — then ran a Delphi consensus across clinicians and patients, ranked all 12 scales on seven essential elements, and recommended the Bland neurophysiologic grading scale for ubiquitous use: objective, reliable, understandable, sensitive to change, and validated against outcomes after surgical release [2]. The practical take for the reader: when the report comes back, the grade — negative, minimal, mild, moderate, severe, extreme — is the file's dial tone, and the reader who knows their grade holds the same instrument the surgeons hold [2][5].

The keyboard inversion: the folk cause, audited by census

No myth in this file costs more gym time than the causal story: repetitive hand use caused my carpal tunnel. The population file tested exactly that story and found it backwards. The same Swedish registry machinery — 2,465 working-age adults (25–65), 82% response, symptoms plus nerve conduction, adjusted for age, sex, body mass index and smoking — split the working population by daily keyboard hours and watched the prevalence of confirmed CTS fall as the typing rose: 2.6% in the intensive group (≥4 h/day), 2.9% moderate (1 to <4 h), 4.9% low (<1 h), and 5.2% among those with no keyboard work at all (P for trend = 0.032); the high-versus-low prevalence ratio was 0.55 (95% CI 0.32–0.96) [4]. The authors' conclusion is the sentence this file quotes precisely: intensive keyboard use appears to be associated with a lower risk of CTS [4]. The honest transfer to the lifter is not "gripping protects you" — it is that this file is not a simple wear-out meter on one motion. The confirmed associations run structural and systemic: the tunnel's volume, fluid balance (the pregnancy file below), and metabolic state (the diabetes file below) — not the number of keyboard hours or, by extension, the number of rows [4][15][16]. The barbell's grip work belongs in the same sentence as the keyboard: train it, position it well, and stop prosecuting it for a crime the census did not find [4].

The natural-history doors: what untreated actually looks like

Before any treatment file, the field ran the question every diagnosed reader actually asks: what happens if I do nothing? Two prospective files answer it, and both downgrade the dread. Padua's team followed 80 untreated CTS cases prospectively, grading each by the neurophysiological classification into six groups — negative, minimal, mild, moderate, severe, extreme [5]. The finding that reframes the whole file: disease worsening was inversely related to the severity of entrapment — the milder cases shifted around, the well-established ones held — while improvement was proportional to impairment (except the extreme cases, which never improved), and about one-third of mild and moderate cases improved on their own [5]. Most minimal cases stayed neurophysiologically unchanged; the authors' read: in a good percentage of cases, the nerve impairment is self-limited [5]. The Colombian file stretched the clock to two years: 132 consecutively monitored untreated patients — 47.6% recovered clinically, 28.8% stayed stable, 23.4% deteriorated; on nerve conduction, 25% improved, 67.4% held constant, 7.6% worsened [6]. The authors' conclusion is the conservative file's charter: given that most patients remain stable or improve, conservative treatment may be all a selected group needs [6]. The gym translation is not "ignore it" — the 23.4% who deteriorated are why the fences and the measurement exist [5][6]. It is that the default direction of this file is not catastrophe, and a mild-to-moderate grade with stable symptoms is a train-through file with a follow-up plan, not a bench ticket [5][6][7].

The night-splint door: the honest shrug, and the trial inside it

The first prescription nearly every wrist receives is the night splint — a rigid shell holding the wrist neutral so the tunnel's pressure curve stays flat through the night. The 2023 Cochrane file on splinting is the field's honesty audit: 29 trials, 1,937 adults, mean ages 42 to 60 — this file's exact demographic [7]. The verdict against no treatment: splinting may provide little or no benefit in symptoms short-term — the BCTQ symptom score 0.37 points better with a confidence interval crossing zero (0.82 better to 0.08 worse), and when the review restricted to lower-risk-of-bias trials the difference shrank to 0.01 points — nothing [7]. Function: probably no short-term improvement. The trial inside the shrug, though, is the one that keeps the splint on the shelf: night-time splinting produced a higher rate of overall improvement than no treatment at RR 3.86 (95% CI 2.29–6.51), with a number needed to treat of 2 — one study, 80 participants, low-certainty evidence, quoted as such [7]. And the review's own bottom line is the fair summary: insufficient evidence to conclude whether splinting benefits people with CTS — but splinting is inexpensive, with no plausible long-term harms, so small effects could justify its use, particularly for people not interested in surgery or injections [7]. For the lifter, that is a green light with a receipt: wear it at night, expect modest symptom relief, and do not let anyone sell the splint as a cure — one trial also hinted six months of wear may beat six weeks, and the file honestly says it is unclear [7].

The injection door: real, effective, and expiring

The corticosteroid injection into the tunnel is the conservative file's strongest pharmaceutical card, and the 2023 Cochrane against placebo reads it precisely: 14 trials, 994 hands; symptom improvement at up to three months SMD −0.77 (95% CI −0.94 to −0.59, moderate certainty), still present at up to six months at −0.58; function −0.62 at three months; and the requirement for surgery at one year reduced slightly, RR 0.84 (0.72–0.98) [8]. The authors' conclusion, quoted as the card's terms: effective for mild and moderate CTS, with benefits lasting up to six months and a reduced need for surgery up to 12 months — and where serious adverse events were reported, they were rare (the worst in the file: injection-site pain resolving over weeks) [8]. That is a real treatment with a clock on it — a bridge that buys a season of training, not a permanent release [8]. The follow-up Cochrane in 2024 asked the harder question — injection versus surgery, head to head — and the file collapsed: seven studies, 569 hands, and every pooled comparison too uncertain for reliable conclusions: symptoms at three months SMD 0.63 with a confidence interval from −0.61 to 1.88 and I²=95% — the trials disagree with each other that much [9]. One signal survived the collapse, low-certainty: surgery may improve the electrical picture (distal motor latency) more than the injection does [9]. The synthesis the reader should carry: the injection is a legitimate, well-evidenced, time-limited move for mild-to-moderate grades — made by a clinician who has examined the wrist, never by a program [8][9].

The surgery door: told honestly in both directions

The oldest Cochrane in the stack remains the surgery file's spine, and it deserves its exact numbers. Four randomized trials, 317 participants, surgery versus non-surgical care: symptom improvement favored surgery at three months (RR 1.23, 95% CI 1.04–1.46), six months (RR 1.19, 1.02–1.39) and one year (RR 1.27, 1.05–1.53) [10]. The neurophysiology followed (RR 1.44), and the file's quietest load-bearing number: the pooled risk of needing surgery later anyway, if started conservative — the surgical group's re-operation rate sat at RR 0.04 (0.01–0.17), meaning a significant proportion of medically-treated patients eventually cross over to the knife [10]. And the ledger's other column, which the internet reliably drops: complications, though common in both arms, were significantly more common in the surgical arm (RR 1.38, 1.08–1.76) [10]. The authors' own conclusion carries the file's two honest clauses: surgery relieves symptoms significantly better than splinting — and further research is needed to discover whether this conclusion applies to people with mild symptoms [10]. That open clause is the mild-grade reader's whole question in one sentence: the surgical edge was measured mostly on symptomatic-enough-to-randomize wrists, not on every tingling hand [10]. Put the doors together and the file's escalator reads: mild-to-moderate and stable — measure, splint, train, re-grade [5][6][7]; failing that or needing a season fast — the injection [8]; severe grade, thenar signs, or conservative failure — the release, with the complication rate quoted honestly [10][14].

The exercise doors: yoga's grip trial, and the RCT that improved symptoms without moving the nerve

The wider conservative landscape got its ranking file in 2025: a network meta-analysis of 49 randomized trials, 3,323 participants, eleven conservative treatments — and the top of the short- and medium-term pain rankings was manual therapy (surface under the cumulative ranking curve 87.6% and 99.3%), with dextrose-5%-in-water injections and PRP close behind, and low-level laser (SMD −1.45 vs control) and shockwave (−1.03) also separating from placebo [11]. Read that ranking carefully and it says the same thing the small trials say: the hands-on and movement-based doors keep outperforming the passive ones, and nothing in the conservative cabinet is a cure [11]. The loading file's two most instructive trials sit twenty-eight years apart. The older one is the yoga trial — 42 employed or retired individuals with CTS, median age 52, randomized to eleven upper-body yoga postures (strengthening, stretching, balancing each joint, plus relaxation) twice weekly for eight weeks, against a control offered a wrist splint [12]. The results: grip strength rose from 162 to 187 mmHg (P=.009) and pain fell from 5.0 to 2.9 (P=.02) in the yoga group with no significant change in controls; Phalen's sign improved in 12 versus 2 — while sleep disturbance, Tinel's sign and median-nerve conduction did not change [12]. The authors called it a preliminary study, and the file quotes it as exactly that: a small 1998 trial in median-age-52 wrists where structured upper-body work improved the measurable hand — grip is a training outcome, not just a symptom [12]. The 2026 trial is the field's freshest instrument and its most humble result: 102 patients with electrodiagnostically confirmed mild-to-moderate CTS (mean age 51.5, 85.3% female), randomized three ways — daily home gliding exercises, daily low-load resistance exercises, or control, for three weeks — with shear-wave elastography of the median nerve and a flexor tendon measured at the tunnel inlet [13]. The outcome: symptoms improved in the gliding group against controls — but no significant change appeared in any primary or secondary tissue measure: nerve stiffness, tendon stiffness, nerve cross-sectional area, all unmoved, and the authors' conclusion is the loading file's new charter sentence: short-term exercise does not modify median nerve or tendon elasticity; symptom improvement may occur through mechanisms other than changes in tissue stiffness [13]. Together the two doors teach the same lesson this library keeps teaching: exercise reliably improves the hand that reports — pain, grip, function — and does not promises to remodel the nerve that measures in weeks. Train for the symptoms and the strength; let the nerve conduction study, not the pump, tell you what changed underneath [12][13].

The severe door: thenar atrophy, audited prospectively

The scariest phrase in any wrist report is thenar atrophy — the wasted thumb pad — and the fear it carries is "too late." The prospective file on that exact door says: mostly, no. Brown's hand service followed 41 hands in 36 patients after open carpal tunnel release — 38 with constant (not intermittent) fingertip numbness, 18 with thenar atrophy on exam, 12 with both — measuring sensibility, grip, key-pinch, QuickDASH and directly-measured atrophy at 8 days, 3, 6 and 12 months [14]. The recoveries: six months out, 83% of the constant-numbness patients showed improved sensibility in at least one digit (against 40% of the atrophy group), and constant tip numbness resolved in roughly two-thirds of all patients by six months [14]. The atrophy patients started weaker — pinch 8.9 versus 10.8 at baseline — but returned to baseline pinch in 72–80% at six months and 100% at twelve, and the measured atrophy itself receded 1.2 of its baseline 1.4 mm [14]. Hand-function dissatisfaction flipped to 100% satisfaction at three months [14]. The door's meaning for the training reader is the fence at the top of this post in its positive form: thenar wasting and constant numbness are the surgery-consultation signs — and the consultation is worth keeping, because even the advanced file recovers measurably after release [14]. Waiting on a wasted thumb pad is the one unforced error this file does not forgive.

The pregnancy door and the metabolic door: the two systemic files behind the wrist

Two doors connect this wrist file to the rest of the body, and both belong to this library's readers. The pregnancy door: a Dutch prospective cohort followed 639 pregnant women with the Boston Carpal Tunnel Questionnaire through gestation — 219 (34%) reported CTS symptoms during pregnancy, most mild-to-moderate, with symptom load significantly higher after 32 weeks [15]. The mechanism file: the women with CTS symptoms reported significantly more fluid retention through gestation (adjusted OR 1.8, 95% CI 1.5–2.1) — and the symptoms fed the sleeping-problems file independently [15]. The tunnel is a fixed-volume corridor; pregnancy's fluid dynamics raise the pressure; the file is a volume file, and the third-trimester reader with new night tingling is standing in a 1-in-3 crowd that mostly clears after delivery — with the severe cases assessed, not just endured [15]. The metabolic door: a 2026 hand-clinic screening study put a fingerstick HbA1c on the question and found 58.3% of 84 patients presenting with upper-extremity musculoskeletal disorders in the prediabetes range, and 4.8% in the type-2 range — with 43.4% of those elevated patients having no primary-care doctor at all [16]. Carpal tunnel is more prevalent in diabetes, the hand clinic is often the first door these patients walk through, and the study's point is the file's point: the wrist visit is a cheap metabolic catch — a prediabetes flag caught where the disease is still reversible [16]. For the reader holding a new diagnosis: a primary-care panel (glucose, and the thyroid check this library's hypothyroidism file covers) is part of a complete carpal-tunnel workup, not a detour from it [16].

The Brazil anchor: 174 surgeons, one congress, the same playbook

The Brazilian file for this guide is a census of the treaters. At the 36th Brazilian Hand Surgery Congress, 174 board-certified hand surgeons answered a structured questionnaire on how they manage carpal tunnel syndrome [17]. The practice pattern it captured runs the same spine the international literature does: electroneuromiography used by most surgeons before surgery — the nerve is measured, not guessed; night splinting the most commonly used conservative treatment — the exact door the Cochrane file shrugs at but tolerates for its price and safety [7][17]; and the standard open release with proximal antebrachial fascia division and no neurolysis as the operative default [17]. The authors' own critique is the health-system honesty this library's Brazilian readers stand in: half the surgeons still use prophylactic antibiotics, most still operate with inpatient admission and intravenous regional anesthesia — and the paper calls for new studies on exactly those costs, "to provide a more cost-effective approach to surgical treatment" [17]. The reader's takeaway from their own country's census: the Brazilian care path is the measured one — EMG, splint, injection, release — and every step of it is available through the SUS and private systems at the evidence grades this post has already quoted [17].

So what do you actually do? The practical file

Six rules, every one sourced from the files above:

  1. Measure before you program. The tingling is common (14.4%) and the confirmed file is not (2.7%), with an 18.4% false-positive floor on the test itself — a clinical examination plus nerve conduction, graded on the Bland scale, is the instrument that decides whether this is a train-through file [1][2][3].
  2. Fence the severe signs to the surgeon's door. Thenar wasting, constant numbness, thumb-abduction weakness — nerve studies and a hand surgeon before any program; and even that file recovers two-thirds and 1.2 of 1.4 mm after release, so the door is worth walking through fast [14].
  3. Splint at night, with the shrug priced in. The Cochrane file says insufficient evidence, small effects, no harms, one RR 3.86 night-wear trial at NNTB 2 — the cheapest entry in this entire post [7].
  4. Audit the wrist positions, not the exercises. The keyboard census inverted the folk cause (2.6% vs 5.2%, PR 0.55) — the fix is positioning, not amputation of training: neutral wrists on presses and push-up handles, a grip width and rack position the wrist tolerates, straps or gloves-versus-chalk choices made on symptoms, and grip work kept in the program at the dose the nerve grade permits [4][13].
  5. Keep training everything else, hard. This is a wrist file, not a body file: legs, posterior chain, core, conditioning, unaffected pressing angles — the progressive overload file does not pause for one tunnel; use machines and neutral-grip tools to route around the wrist while the measurement settles [12][13].
  6. Track symptoms like a training metric. A dated log — which fingers, night or day, constant or intermittent, grip strength — beside the training journal is exactly the trend the treatment decisions run on; the untreated file says a third of mild-to-moderate wrists improve on their own, and the trend is how you know which third you are in [5][6].

How RightRep fits into this

The wrist-nerve file is, at its core, a measure-then-track question — and it is unanswerable without a record:

  • The everything-else program. Rule five is a full-body program that routes around one wrist — counted, scored sessions keep the legs, posterior chain and conditioning progressing while the nerve gets measured and the treatment ladder runs [12][13].
  • The symptom ledger. Rule six's dated finger-map and grip notes beside every session turn "my hand's been weird for months" into the trend the surgeon and the therapist actually decide on [5][6].
  • The grip gauge. Garfinkel's trial made grip strength a primary outcome (162→187 mmHg in eight weeks) — a tracked grip is this file's progress bar, and the camera-scored session log keeps it honest [12].
  • The position audit. Rep-by-rep form scoring is a wrist-position audit at replay speed — the neutral-wrist rule enforced by evidence rather than memory [4].

The bottom line

  • The scale: 14.4% of a random adult population report the tingling; 3.8% clinically certain; 2.7% clinically + electrophysiologically confirmed; 18.4% of symptom-free controls fail the nerve test anyway; 1 in 5 symptomatic subjects carries the real file [1].
  • The grading: AANEM's 2026 consensus — 12 published scales, 43% of 1,253 surveyed practitioners using NCS-only grading, 27% using none; the Bland neurophysiologic grading scale recommended for ubiquitous use, validated against surgical outcomes [2].
  • The measurement: nerve conduction is a measure of impaired nerve function, not a yes/no test — prognostic for surgery, sensitive to change [3].
  • The inversion: intensive keyboard use 2.6% vs no keyboard 5.2% (PR 0.55, trend P=0.032) — the folk cause runs backwards [4].
  • The untreated course: worsening inverse to severity, one-third of mild-moderate improving; at two years, 47.6% recovered / 28.8% stable / 23.4% worse [5][6].
  • The splint: 29 trials, 1,937 adults, 0.37 points (CI crossing zero; 0.01 at low risk of bias) against no treatment — but RR 3.86 for night wear at NNTB 2, cheap and harmless, insufficient evidence concluded honestly [7].
  • The injection: SMD −0.77 at ≤3 months, benefits to six, surgery at one year RR 0.84, serious harms rare; against surgery itself the evidence collapses (I²=95%) [8][9].
  • The surgery: RR 1.27 at one year, re-operation 0.04, complications significantly more common (RR 1.38), the mild-symptoms question explicitly open [10].
  • The conservative ranking: 49 RCTs, 3,323 participants, 11 treatments — manual therapy top of the network (SUCRA 87.6% short-term, 99.3% medium-term), dextrose and PRP next, laser −1.45 and shockwave −1.03 vs control [11].
  • The exercise doors: yoga's grip 162→187 mmHg, pain 5.0→2.9, Phalen 12-vs-2, conduction unchanged [12]; the 2026 three-arm trial improving symptoms with zero change in nerve elastography — "mechanisms other than changes in tissue stiffness" [13].
  • The severe door: constant numbness resolved in ~2/3 at six months, 83% improved sensibility, atrophy pinch back to baseline in 100% at twelve months, the pad itself regrowing 1.2 of 1.4 mm [14].
  • The systemic doors: pregnancy 34% of 639 women, fluid retention OR 1.8 (1.5–2.1) [15]; the hand clinic's fingerstick — 58.3% prediabetes range, 4.8% type-2, 43.4% with no primary-care doctor [16].
  • The Brazil anchor: 174 board-certified hand surgeons — EMG by most, night splinting first-line conservative, standard open release; the authors auditing their own system's costs [17].

Measure first, splint at night, keep the grip and keep training — and let the nerve study, not the tingling, write the ceiling — at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Atroshi I, Gummesson C, Johnsson R, Ornstein E, Ranstam J, Rosén I. Prevalence of carpal tunnel syndrome in a general population. JAMA, 1999;282(2):153–158. pubmed.ncbi.nlm.nih.gov
  2. Robinson LR, Al-Lawati ZJ, Andary M, Cartwright MS, Grant P, Kincaid J, Bland JDP. AANEM consensus-based guideline for grading the severity of carpal tunnel syndrome. Muscle & Nerve, 2026 (online ahead of print). pubmed.ncbi.nlm.nih.gov
  3. Bland JDP. Use of nerve conduction studies in carpal tunnel syndrome. Journal of Hand Surgery (European Volume), 2023;48(10):976–985. pubmed.ncbi.nlm.nih.gov
  4. Atroshi I, Gummesson C, Ornstein E, Johnsson R, Ranstam J. Carpal tunnel syndrome and keyboard use at work: a population-based study. Arthritis & Rheumatism, 2007;56(11):3620–3625. pubmed.ncbi.nlm.nih.gov
  5. Padua L, Padua R, Lo Monaco M, Aprile I, Paciello N, Nazzaro M, Tonali P. Natural history of carpal tunnel syndrome according to the neurophysiological classification. Italian Journal of Neurological Sciences, 1998;19(6):357–361. pubmed.ncbi.nlm.nih.gov
  6. Ortiz-Corredor F, Enríquez F, Díaz-Ruíz J, Calambas N. Natural evolution of carpal tunnel syndrome in untreated patients. Clinical Neurophysiology, 2008;119(6):1373–1378. pubmed.ncbi.nlm.nih.gov
  7. Karjalainen TV, Lusa V, Page MJ, O'Connor D, Massy-Westropp N, Peters SE. Splinting for carpal tunnel syndrome. Cochrane Database of Systematic Reviews, 2023;2:CD010003. pubmed.ncbi.nlm.nih.gov
  8. Ashworth NL, Bland JDP, Chapman KM, Tardif G, Albarqouni L, Nagendran A. Local corticosteroid injection versus placebo for carpal tunnel syndrome. Cochrane Database of Systematic Reviews, 2023;2:CD015148. pubmed.ncbi.nlm.nih.gov
  9. Ashworth NL, Bland JD, Chapman KM, Tardif G, Albarqouni L, Nagendran A. Local corticosteroid injection versus surgery for carpal tunnel syndrome. Cochrane Database of Systematic Reviews, 2024;8:CD015101. pubmed.ncbi.nlm.nih.gov
  10. Verdugo RJ, Salinas RA, Castillo JL, Cea JG. Surgical versus non-surgical treatment for carpal tunnel syndrome. Cochrane Database of Systematic Reviews, 2008;4:CD001552. pubmed.ncbi.nlm.nih.gov
  11. Chen Y, Han B, Zhang X, Guo C, Han Q, Zhang Z, Hou S. Conservative treatments of carpal tunnel syndrome: a systematic review and network meta-analysis. Archives of Physical Medicine and Rehabilitation, 2025;106(9):1447–1458. pubmed.ncbi.nlm.nih.gov
  12. Garfinkel MS, Singhal A, Katz WA, Allan DA, Reshetar R, Schumacher HR Jr. Yoga-based intervention for carpal tunnel syndrome: a randomized trial. JAMA, 1998;280(18):1601–1603. pubmed.ncbi.nlm.nih.gov
  13. Peterca J, Snoj Ž, Omejec G. Effects of gliding and resistance exercises on median nerve and tendon ultrasound elastography in carpal tunnel syndrome: prospective, randomised, three-arm, parallel-group trial. Clinical Rehabilitation, 2026 (online ahead of print). pubmed.ncbi.nlm.nih.gov
  14. Nadeem M, Got CC, DaSilva MF, Katarincic JA, Akelman E, Weiss AC. Natural history of constant preoperative fingertip numbness and thenar atrophy following carpal tunnel release: a prospective study. Journal of Wrist Surgery, 2025;15(3):251–256. pubmed.ncbi.nlm.nih.gov
  15. Meems M, Truijens S, Spek V, Visser LH, Pop VJ. Prevalence, course and determinants of carpal tunnel syndrome symptoms during pregnancy: a prospective study. BJOG, 2015;122(8):1112–1118. pubmed.ncbi.nlm.nih.gov
  16. Heiting C, Wickes CB, Katakam S, Herrera C, Li B, Intravia J, Nolan JE, Nellans KW. Occurrence of undiagnosed diabetes mellitus with musculoskeletal disorders of the upper extremity: prospective screening with point-of-care haemoglobin A1c fingerstick testing. Musculoskeletal Care, 2026;24(3):e70256. pubmed.ncbi.nlm.nih.gov
  17. Okamura A, Guidetti BC, Caselli R, Borracini JA, Moraes VY, Belloti JC. How do board-certified hand surgeons manage carpal tunnel syndrome? A national survey. Acta Ortopédica Brasileira, 2018;26(1):48–53. pubmed.ncbi.nlm.nih.gov