To the Editor: It is with interest that we read the recent article published in Diabetologia by Holman et al [1]. This piece of work is very similar to that published by our group in the British Journal of Surgery in 2010 [2] and reaches the same unfortunate conclusion: there is significant variation in the rates of amputation around England. Like us, Holman et al have used Hospital Episode Statistics for their analysis and have broadly similar figures. Whereas we investigated the variation in incidence and outcome (in-hospital mortality and the above knee:below knee ratio) of major and minor amputation, the Holman group have focused on the incidence of major and minor amputations within the general population and the ‘at risk’ population with diabetes mellitus by combining the HES data with figures from the Quality Outcome Framework to determine the prevalence of diabetes in the general population.

Our work was novel and the data were presented to a House of Commons Select Committee concerned with the causes of variation not only in incidence but also in outcomes, which, it could be argued, is more important to patients. The data went on to assist the development of the Vascular Society of Great Britain and Ireland’s ‘Quality Improvement Framework for Amputation’, which aims to reduce mortality rates after major amputation and standardise care [3]; this is not mentioned by Holman et al.

Identifying the causes of variation in incidence of and outcome after amputation is of huge importance in tackling the burden of limb loss for patients, and Holman et al should be commended for adding new data to the discussion. However, for completeness, we would recommend our work to readers and also encourage them to read the Quality Improvement Framework on Amputation.