Wrist & Hand: Carpal Tunnel, Thumb Base Arthritis and De Quervain's

Three different conditions with three different answers — and one of them has a time limit.

Start Here: Which Condition Do You Have?

These share a body part and almost nothing else. Treating them as one problem with one answer is the commonest mistake made about hands.

⚠️ This page covers these three plus post-traumatic wrist arthritis. It does not cover Dupuytren's contracture, trigger finger, ganglion cysts or TFCC injuries — those need separate advice, and we would rather say so than generalise.

⛔ Carpal Tunnel Syndrome: The Part With a Deadline

Most of what is written about hand conditions can safely be read at leisure. This section cannot.

Carpal tunnel syndrome is a compression neuropathy — sustained pressure on the median nerve. In mild and moderate cases the nerve is irritated but intact, and symptoms are recoverable. In severe or long-standing cases, motor nerve fibres to the thumb muscles die, the muscle at the base of the thumb wastes (thenar atrophy), and that loss may be permanent even after a technically perfect decompression.

Get assessed promptly if you have any of these

• Numbness that is now constant rather than intermittent or night-time only
• Visible flattening or hollowing of the muscle pad at the base of the thumb, compared with your other hand
• Weakness of thumb opposition — difficulty touching your thumb to your little finger, dropping things
• Symptoms that have been getting steadily worse over months

None of these is a reason to try an injection first. They are a reason to see a hand surgeon and have nerve conduction studies. The operation for carpal tunnel syndrome is quick, is done under local anaesthetic, and probably reduces the risk of needing further surgery to almost nothing³. What it cannot do is bring back muscle that has already wasted.

We say this plainly because it is the one place on this page where waiting has a cost that cannot be undone — and because a patient with thenar wasting is not a candidate for anything we offer.

Diagnosis, and an important caveat about the tests

Nerve conduction studies are the standard objective test, alongside Phalen's and Tinel's signs.

⚠️ But a positive test does not by itself prove that carpal tunnel syndrome is causing your symptoms. In a general-population study of 2,466 people, 18.4% of asymptomatic control subjects (95% CI 12.0-26.3) had electrophysiological median neuropathy at the carpal tunnel¹. Nearly one person in five with no complaints tests positive.

A Lancet Neurology review puts it directly: electrodiagnostic testing and nerve imaging "provide additional information about the extent of axonal involvement and structural change, but their exact benefit to patients is unknown", and there remains "a paucity of evidence about the best approaches for assessment of carpal tunnel syndrome and to guide treatment decisions"⁹.

How common is it, really?

The answer depends entirely on the definition used. In the same population study¹:

The authors' conclusion is worth carrying: "1 in 5 symptomatic subjects would be expected to have CTS based on clinical examination and electrophysiologic testing." Most hand tingling is not carpal tunnel syndrome.

What the treatment evidence shows

Surgery versus splinting — Cochrane 2024, 14 randomised trials, 1,231 participants, 84% women³:

The Cochrane authors' summary: surgery "probably results in a higher rate of clinical improvement" but "probably does not provide clinically important benefit in symptoms or hand function compared with splinting."

Surgery versus corticosteroid injection³: hand function probably does not differ (SMD −0.12, 95% CI −0.80 to 0.56, moderate certainty). Clinical improvement and symptom relief are uncertain (very low certainty).

So the honest sequence for mild-to-moderate carpal tunnel syndrome is: splint first. It is cheap, it is well evidenced, and roughly half of people will not need anything further. If it fails, injection and surgery are both reasonable, and surgery's clearest advantage is that it is usually definitive.

PRP for carpal tunnel syndrome: a genuinely disputed evidence base

We offer this treatment. Here is the whole picture rather than the favourable part of it.

In favour — Jiang 2022⁶, systematic review and meta-analysis of 8 RCTs, 220 patients, all judged high quality:

• Short-term symptom severity: MD −2.00 (95% CI −3.15 to −0.85)
• Mid-term: VAS −0.63 (−1.22 to −0.04); symptom severity −3.56 (−4.93 to −2.18); function −2.29 (−3.03 to −1.56); sensory peak latency −0.39 (−0.58 to −0.19); median nerve cross-sectional area −0.20 (−0.31 to −0.10)

The single most-cited trial — Wu 2017²: 60 patients with unilateral mild-to-moderate CTS, randomised single-blind to one ultrasound-guided PRP injection or night splinting. At 6 months the PRP group had significantly better VAS, BCTQ and median nerve cross-sectional area (p < 0.05). ⚠️ Note the comparator is an active, effective treatment, not a placebo — and the trial was single-blind.

Against, and more recent — Du 2025⁷, meta-analysis of 7 studies, 365 patients:

• Symptom severity improved at 1, 3 and 6 months (P < .05) ✅
• ⛔ Median nerve cross-sectional area: differences NOT statistically significant (P > .05)
• ⚠️ Sensory nerve conduction velocity was lower in the PRP group at 1 month (P < .0001), not significant at 3 or 6 months
• ⚠️ In these pooled trials, PRP was given on top of conventional conservative treatment, not instead of it

And in context — Hong 2022⁸, network meta-analysis of 19 studies, 1,066 patients, ranking all injectable options:

⚠️ A 5% dextrose injection — among the cheapest interventions available — outranked PRP for pain. We are including that because it is true and because you would not otherwise hear it from us.

Our position: PRP is a defensible option for mild-to-moderate carpal tunnel syndrome after splinting has failed and before considering surgery. It is not established, the syntheses disagree, and the claim that it measurably restores the nerve is not supported by the most recent pooled data.

Thumb Base (CMC) Arthritis

Osteoarthritis of the trapeziometacarpal joint is the commonest arthritis of the hand and a frequent cause of real disability — difficulty pinching, opening jars, turning keys.

On prevalence, we want to be careful, because the most-quoted figures come from a study of a particular group. Sodha and colleagues¹¹ reviewed the radiographs of 615 consecutive patients presenting with an isolated distal radius fracture — people who had broken their wrist, not a general population sample. In that group, radiographic trapeziometacarpal arthrosis rose steadily from age 41, reaching 91% in those over 80 (94% of women, 85% of men), with a totally destroyed joint in 66% of women over 80 versus 23% of men. The condition is common and affects women more severely; precise general-population percentages are not what that study measured.

What works

First: hand therapy and splinting. A 2026 systematic review of 5 RCTs (n=260, PEDro 6-8/10) found proprioceptive training added to conventional rehabilitation produced moderate-to-large effects on pain and function¹⁴. This is the best-evidenced non-operative option and it is routinely under-delivered.

Injections. Corticosteroid gives better pain relief than hyaluronic acid at 1 month (MD 0.73, 95% CI 0.02-1.43, p = 0.043), but across 4 RCTs and 222 patients there was no significant difference at 3 months (p = 0.670) or 6 months (p = 0.654)¹². Steroid buys a few weeks, not a few years.

Surgery, if it comes to that. A network meta-analysis of 26 RCTs, 1,193 hands and 19 surgical techniques¹³ found simple trapeziectomy ranked first for long-term pain relief and was the most favourable for avoiding complications. More elaborate reconstructions were associated with greater complication severity. If a surgeon offers you a complex procedure, it is reasonable to ask why not the simple one.

The regenerative evidence for small joints

One study, thirteen patients, and it was an operation.

Murphy et al. (ASCOT, 2017)⁴ treated osteoarthritis of the thumb CMC joint by microfracturing the joint surface and applying autologous bone marrow stem cells directly — a surgical procedure under anaesthetic, not an injection. Fifteen patients met criteria; two were excluded, one because of postoperative cellulitis. In the remaining 13, at 12 months:

• VAS pain at rest 3.23 → 1.69 (p = 0.029); on activity 7.92 → 4.23 (p = 0.002)
• DASH 51.67 → 23.08 (p = 0.007); thumb opposition score 7.62 → 9.23 (p = 0.015)
• All patients converted from a positive to a negative Grind test
• ⚠️ Grip strength did not significantly improve — 4.7 kg to 5.53 kg (p = 0.126)

⚠️ There was no control group, and the authors describe it as "an innovative pilot study… a new approach". The strength result is worth noting on a page about hand function: the pain and disability scores moved, the measured strength did not.

⚠️ What we are not going to do is borrow knee data. Trials of mesenchymal stem cells in knee osteoarthritis exist and are themselves disputed — a meta-analysis of 13 Level-I RCTs found intra-articular MSC injection no different from placebo in the knee. Extrapolating from a large weight-bearing joint to the thumb is not evidence about the thumb.

De Quervain's Tenosynovitis

Pain on the thumb side of the wrist, worse on gripping and thumb movement, with a positive Finkelstein's test. Common in the postpartum period and in occupations involving repetitive thumb use.

Corticosteroid injection works, and we are not going to talk you out of it

The Cochrane review of corticosteroid injection for de Quervain's¹⁰ identified one controlled trial of 18 participants (all pregnant or lactating women), comparing a single steroid injection with thumb spica splinting:

All 9 patients in the steroid injection group achieved complete relief of pain. None of the 9 in the splinting group did (number needed to treat = 1, 95% CI 0.8 to 1.2). No side effects or local complications of steroid injection were noted.

⚠️ Read both halves of that. A number needed to treat of 1 is about as favourable a result as exists in medicine. It also comes from 18 patients in a single small trial, and the review's authors say so explicitly. The evidence base for the standard treatment is strong in direction and thin in volume.

Where PRP stands

Kumar et al. (2023)⁵ randomised 60 patients (30 per group) to a single PRP or corticosteroid injection, following them for a year. Level of evidence: 2, prospective comparative study.

• Both groups improved significantly on Finkelstein's test, VAS, DASH and Modified Mayo Wrist Score at every timepoint
• Comparison between the two groups showed no statistically significant difference — the treatments performed equivalently
• Complications occurred in 8 of 30 corticosteroid patients (26.7%, p = 0.026) — subcutaneous fat atrophy, depigmentation, temporary increase in pain — and in none of the PRP group

⚠️ The authors' own conclusion, in full: "PRP may have a lower complication profile, however, this benefit should be weighed against the slight increase in cost and time of PRP preparation and injection."

That is a trade-off, not a verdict. Steroid is cheaper, faster and immediately available; PRP appears to avoid the skin and fat changes that steroid can cause around a superficial tendon. Which matters more depends on you.

Treatment Options Compared

Carpal tunnel syndrome

Thumb base and wrist arthritis

Is Regenerative Therapy Right for You?

May be appropriate

• Mild-to-moderate carpal tunnel syndrome that has not responded to a genuine trial of night splinting, where you would rather try an injection before considering surgery
• Thumb CMC arthritis (Eaton II-III) after hand therapy and splinting have been tried
• De Quervain's that has recurred after corticosteroid, or where the skin and fat changes steroid can cause around a superficial tendon are a particular concern
• Anyone who has understood that the small-joint evidence is 13 uncontrolled patients and the carpal tunnel evidence is disputed

Not appropriate

• ⛔ Severe carpal tunnel syndrome with thenar muscle wasting, constant numbness or thumb weakness. See a hand surgeon promptly — this is the time-critical situation described above, and delay risks permanent loss
• ⛔ Anyone who has not yet tried night splinting for carpal tunnel syndrome, or hand therapy for thumb base arthritis
• ⛔ Advanced wrist arthritis with complete cartilage loss and deformity
• ⛔ Active infection at the injection site
• ⛔ Unstable wrist fractures requiring surgical fixation

What to Expect

Our Assessment and Treatment Protocol

Assessment

• Hand and wrist examination, including specific assessment of thenar bulk and thumb opposition
• Nerve conduction studies where carpal tunnel syndrome is suspected
• High-resolution ultrasound of tendons, joints and the median nerve
• Grip and pinch dynamometry
• Review of what has already been tried, and for how long

Treatment

• Ultrasound-guided injection to the specific target: perineural for carpal tunnel, intra-articular for thumb base arthritis, peritendinous for De Quervain's
• Custom splinting for post-injection protection

Rehabilitation

• Session with a certified hand therapist
• Nerve gliding exercises where indicated
• Ergonomic and activity assessment
• Home programme

⚠️ On timelines. We are not publishing a week-by-week recovery schedule. The published studies assessed patients at fixed research timepoints — Wu at 6 months, Murphy at 12, Malahias at 12 weeks — and none of them supports a milestone calendar. Your hand therapist will set progression by what your hand is doing, not by a chart.

Frequently Asked Questions

Q: Can PRP cure carpal tunnel syndrome permanently?

A: No published trial has followed patients long enough to answer that. The longest follow-up in the studies cited here is 6 months²; the placebo-controlled study ran 12 weeks¹⁵. What can be said is that PRP improved symptoms over that horizon in several randomised trials, that the most recent pooled analysis found no significant effect on nerve size⁷, and that severe carpal tunnel syndrome with muscle wasting needs surgical release rather than any injection.

Q: Should I just have the surgery?

A: If you have red-flag features, yes — promptly. If your symptoms are mild to moderate, splinting first is well supported, and surgery's advantage over splinting on symptom and function scores is below the threshold for a clinically important difference³. Its clearest benefit is that it usually settles the matter: 44% of splinted patients came to surgery anyway³.

Q: Isn't a steroid injection bad for the tendon?

A: Repeated steroid injections around tendons are used cautiously for good reason, and in de Quervain's a quarter of patients in one trial had local skin or fat changes⁵. But a single, well-placed steroid injection is genuinely effective for de Quervain's — the Cochrane trial reported complete pain relief in 9 of 9 patients with no side effects recorded¹⁰. We are not going to tell you it does not work.

Q: How does hand arthritis treatment differ from knee treatment?

A: More than the mechanism suggests. Injection technique differs, but so does the evidence: the knee has large randomised trials that the thumb does not, and those knee trials are themselves disputed. Results from one joint are not evidence about another, which is why this page uses the thumb study rather than the knee ones.

Q: I'm a surgeon, a pianist, a craftsman. How quickly can I get back to fine work?

A: After an injection, most people resume light fine-motor activity within a few weeks, but we set this individually and there is no published return-to-work data for these treatments in these conditions. If your livelihood depends on your hands, that uncertainty should weigh in your decision.

Q: What about trigger finger, Dupuytren's, or a ganglion?

A: This page does not cover them, and we are not going to make claims about conditions we have not reviewed here. Trigger finger responds well to steroid injection and to surgery. Dupuytren's is a fibrotic contracture treated with needle fasciotomy, collagenase or surgery — it is not a condition where a regenerative injection makes mechanistic sense, and we do not offer one for it.

Take the Next Step

⛔ If you have constant numbness, visible wasting at the base of your thumb, or weakness of thumb opposition, do not book an assessment with us first. See a hand surgeon. That is the advice we would give a member of our own family.

Otherwise:

• Take our 2-minute Joint Health Assessment
• Book a Discovery Consultation with our hand and wrist specialists

If you have not yet completed a proper trial of night splinting for carpal tunnel syndrome, or hand therapy for thumb base arthritis, we will tell you to do that first.