General Practice

Prevalence of arm movements in patients with coronary heart disease: case-control study

BMJ 1997; 314 doi: (Published 11 January 1997) Cite this as: BMJ 1997;314:122
  1. Alan N Rennie, part time medical adviser to Benefits Agencya
  1. a 20 Waterside Avenue, Newton Mearns, Glasgow G77 6JT
  • Accepted 30 August 1996


I had previously noticed that patients with coronary heart disease tended to gesticulate, or sit with their arms up and hands clasped behind their heads, more than patients who did not have ischaemic heart disease.1 I therefore undertook a pilot study using a goniometer2 to measure movements at the elbow joint during a 10 minute interview in a group of patients with proved coronary heart disease and a group of controls. A significant difference, using an analysis of variance, was shown between the patients with coronary heart disease and the controls (P=0.01). As the pilot study had only 10 patients in each group, I therefore undertook a larger study with 25 patients in each group.

Methods and results

The coronary heart disease group were 25 consecutive attenders with established coronary heart disease at a chest pain clinic. They were all white and had electrocardiographic changes of ischaemia or previous myocardial infarction. Ages ranged from 33-75 (mean 59.68) years, and 21 were men. The control group were 25 consecutive attenders at a medical clinic free of cardiac complaints, hypertension, hyperlipidaemia, cerebrovascular disease, diabetes, or a strong family history of cardiac disease. Ages ranged from 15-82 (mean 50.24) years, and 19 were men. Each patient sat in an upright, armless chair. A goniometer was placed over the elbow of the dominant arm to record the amount of arm movement over 10 minutes. During this time the patient was asked standard questions on lifestyle, family, and health. The goniometer measures each movement of the joint of 45° in either direction on a two dimensional plane. At the end of 10 minutes a reading was taken; only then was the function of the goniometer explained to the patient.

Patients with coronary heart disease moved their arms during the 10 minute interview significantly more than those in the control group (fig 1). A one way analysis of variance yielded a significance level of P=0.003. With a non-parametric Mann-Whitney test there was still a highly significant difference between the two groups.

Fig 1
Fig 1

Number of arm movements in 10 minutes in 25 patients with coronary heart disease and 25 controls


The most obvious explanation of these findings is that type A personalities are prone both to gesticulation and to coronary heart disease. It is possible that people with coronary heart disease move their arms more because they are otherwise physically inactive or their disease causes them to become agitated. However, my own suspicion is that arm movements over a lifetime may be a factor–combined with other known factors–in the development of coronary heart disease. I hope that further studies will show a contributing causative role of arm movements in coronary heart disease through haemodynamics.


I thank Professor Tony Unsworth, University of Durham, for supplying goniometers; Professor Ross Lorimer and Dr Iain Hutton, Glasgow Royal Infirmary, for providing access to patients; Helen Young, nursing department, Glasgow Royal Infirmary, for checking for researcher bias; Dr Gilbert Mackay, University of Strathclyde, for statistical analysis; and Dr C M Traini, for help with the pilot study.

Funding: None.

Conflict of interest: None.


  1. 1.
  2. 2.
View Abstract