Intended for healthcare professionals


Thermographic changes in keyboard operators with chronic forearm pain

BMJ 1997; 314 doi: (Published 11 January 1997) Cite this as: BMJ 1997;314:118
  1. S D Sharma, research fellowa,
  2. E M Smith, Research nursea,
  3. B L Hazleman, consultant rheumatologista,
  4. J R Jenner, consultant rheumatologistb
  1. a Rheumatology Research Unit, Box 194, Cambridge CB2 2QQ
  2. b Addenbrookes NHS Trust, Cambridge CB2 2QQ
  1. Correspondence to: Dr Sharma
  • Accepted 20 August 1996


Chronic incapacitating forearm pain and disability in the context of repetitive action has caused much debate. Lack of objective measurements in a condition with diverse symptoms, few physical signs, and uncertain pathology is a major problem.1 2 Clinical observations have suggested the presence of vasomotor changes in repetitive strain injury, so we used computer assisted thermography to assess this.

Patients, methods, and results

Ten consecutive keyboard operators (six women) with chronic forearm pain exacerbated by keyboard work presenting for rheumatological assessment and 21 (12 women) asymptomatic controls matched for sex and typing speed (30-50 words/min) were recruited from rheumatology outpatient clinics. All the patients had had diffuse forearm pain for at least three months. We excluded patients with Raynaud's syndrome, neurological causes, compartment syndromes, inflammatory conditions, epicondylitis, algodystrophy, trauma, and diabetes mellitus.

We used a Talytherm infrared camera unit with thermal emission measurement software (Rank Taylor Hobson UK Ltd). Ambient temperature was controlled at 22-24° C. Baseline images were taken after acclimatisation for five minutes; the subjects then typed at their usual speed from a standard text for five minutes, and another thermogram was taken immediately afterwards. Thermograms were taken over 200-350 pixels on the 2nd, 3rd, and 4th proximal phalanges of both hands, avoiding large muscle masses which might interfere with the readings. Mean readings were recorded. All the patients were asked to return after a mean of nine months to assess reproducibility, though one refused further evaluation because of pain, two had moved away, and two did not respond. Differences between means in patients before and after typing were assessed with Student's paired t test and the unpaired test for intergroup means.

After typing all the patients had symptoms. In each patient the mean temperature readings after typing were significantly reduced (fig 1) (mean 2.11°C, range 0.45-3.44°C, 95% confidence interval 1.35 to 2.26°C; P< 0.001). Means before typing in the two groups were similar (P>0.05), though significantly different afterwards (P<0.001). Only four controls showed cooling (mean 0.55°C, range 0.35-1); the 95% confidence interval of the differences between the means before and after typing in the two groups was 0.93 to 2.59°C. Of the five patients who were reassessed, four again showed cooling, while the fifth, who had clinically recovered, did not.

Fig 1
Fig 1

Temperature changes in 10 patients with repetitive strain injury and 21 controls before and after typing


The changes we have described are reproducible and the method is non-invasive. It was notable that the temperature readings in the one patient who became asymptomatic changed significantly at his reassessment. Cooling in symptomatic patients may be secondary to sympathetic overactivity as a result of nociceptor and mechanoreceptor stimulation leading to a reflex neuropathic state; however, cooling after challenge suggests that it is a result rather than a cause.

Many attempts have been made to explain the relation between pain and sympathetic overactivity, mostly in algodystrophy. These suggest a central sensitisation of wide dynamic neurones within the spinal cord.345 This sensitisation is thought to arise via afferents arising in peripheral nociceptors, resulting in increased sympathetic efferent activity leading to painful response rates to subsequent afferents. This results in further sensitisation, setting up a vicious circle of sympathetically maintained pain. This is unlikely to be the sole explanation, however, as sympathetic blockade has not been successful in repetitive strain injury.1

We conclude that thermography needs further evaluation as a diagnostic tool in evaluating repetitive strain injury. It may prove more useful in follow up, particularly in measuring response to treatment, some of which has been prescribed at enormous cost and with little evidence of benefit.


Funding: Arthritis and Rheumatism Council, England. SDS is supported by a travelling fellowship of Roche and Rose Hellaby foundations of New Zealand.

Conflict of interest: None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.