Electronic Letters to:

Upper Limb: A. E. Price, P. DiTaranto, I. Yaylali, M. A. Tidwell, and J. A. I. Grossman Botulinum toxin type A as an adjunct to the surgical treatment of the medial rotation deformity of the shoulder in birth injuries of the brachial plexus J Bone Joint Surg Br 2007; 89-B: 327-329 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

Electronic letters published:

Concerns regarding statistical analyis of results

Timothy E J Hems, David A Sherlock   (9 May 2007)

Concerns regarding statistical analyis of results 9 May 2007
Timothy E J Hems, Consultant Hand and Orthopaedic Surgeon The Victoria Infirmary and The Royal Hospital for Sick Children, Glasgow, David A Sherlock Send letter to journal: Re: Concerns regarding statistical analyis of results t.e.j.hems{at}doctors.org.uk Timothy E J Hems, et al.	Sir, We read this paper with interest. The effects of obstetric brachial plexus injury on the shoulder are very complicated and varied. The only assessment given is the Gilbert shoulder score, which is based on active abduction and external rotation only. It is not clear whether external rotation or abduction was most limited. The passive range of movement is not stated. Nor is the extent of skeletal deformity at the shoulder clear. There was a difference in age between the two patient groups, A and B. We would like to point out two problems with analysis of the results. Although there was apparently no statistically significant difference between Gilbert shoulder scores for the two groups before operation, there was a difference in the mean scores (Group A = 3.38, Group B = 2.85). It would therefore be best to assess the improvement in score after operation. The improvement in the mean scores was 1.93 for group A and 2.0 for group B, which would seem unlikely to be significantly different. Furthermore, Student’s t-test was used to compare the results. This test should not be used as the results are a score rather than a continuous variable (eg. measurement of range of abduction in degrees). A non-parametric statistical test should have been used. We could not re-analyse the data since it is not yet available on the website. As they stand, the results are unlikely to provide evidence of a significant improvement in outcome following use of botulinum toxin. We would fully support the prospective randomised study which the authors are carrying out. However, greater detail of the passive and active range of movement should be recorded since even the Mallet score, which includes internal rotation, has its deficiencies. An important issue, which has been previously under-reported is reduction of the range of internal rotation following surgery, which may have functional consequences. The range of internal rotation both before and after surgery should therefore be assessed. Our experience has been that Pectoralis major is rarely a cause of medial rotation contracture. We would entirely agree with the authors that there is a major influence of the central nervous system on the functional outcome after obstetric brachial plexus palsy and this deserves further investigation. T.E.J. Hems, Consultant Hand and Orthopaedic Surgeon, The Victoria Infirmary, Glasgow, UK. D.A. Sherlock, Consultant Orthopaedic Surgeon, The Royal Hospital for Sick Children, Glasgow, UK.