Our recent editorial has generated widespread debate about the National Institute for Health and Clinical Excellence (NICE) Technology Guidance on the CardioQ™ Doppler, and in particular the evidence on which this guidance was based [1]. Our greatest concern was that this guidance, which was intended to make the case for adoption of the product, seemed to be based on a small number of randomised trials each containing only a small number of patients, a point reiterated by Minto et al. [2]. Nesbitt et al. have asked ‘how much evidence is enough’ and this is indeed a fair question [3], but it is our view that the evidence regarding the CardioQ is not yet sufficient, amounting to eight trials and about 700 patients spread over more than a ten-year period, to warrant the robust commendation given to it by NICE. Analysis of data pooled from a small number of patients over a period of a decade or more will always be questionable because of changes in standards of care and practice over time, and thus we would answer the question ‘how much evidence is enough?’ by saying it depends on how old the evidence is. This is particularly pertinent when considering time to discharge from hospital, which has changed radically for nearly all forms of surgery over the past fifteen years. It is interesting to note that the most recent randomised trial of the CardioQ in major colorectal surgery has failed to demonstrate any benefit from use of the device [4]. As the authors of that trial point out, this may be simply because their standard fluid regimen resulted in higher cardiac outputs in their control group than in previous studies that did demonstrate benefit. We noted in our editorial that the difference between controls and study groups in the trials considered by NICE amounted to a small bag of fluid per patient [1], and this latest trial bears out the fact that a less conservative approach to fluid management during major surgery may yield the same beneficial results with or without the CardioQ.

The process by which NICE develops such guidance has been well described by Campbell et al. [5], however they fail to make the point that every NICE guidance is given a great deal of weight by hospital managers, to the extent that pressure is placed on clinicians to justify why they choose not to adopt such guidance. Adoption or failure to adopt is audited within trusts, and clinicians thus fear claims of negligence if they fail to act within the guidelines. In addition, financial pressure may also be applied, as the NICE guidelines have been adopted into the Commissioning for Quality and Innovation (CQUIN) payment framework [6]. This places the onus on NICE to restrict issuing guidance on technologies to those for which the evidence is deemed to be of high quality, as well as with sufficient depth and breadth of supporting opinion. Our contention is that this is not the case for the CardioQ, especially as use of the device and the subsequent intravascular volume loading it may trigger is contentious in itself, as has also been pointed out by Woodcock [7]. In addition, we agree with Ghuman et al. [8], that advocating a single device for measuring haemodynamic data over other devices, which is inherent in the publishing of such technology guidance, is not merited when there isn’t conclusive evidence of superior benefit. In this final point, we agree with Singer, who also points out that the evidence base for comparing technologies is ‘scanty and inconsistent’ [9].

We hope that NICE will take these views into account, and consider issuing guidance on technology only when a strong body of independent evidence is available.


  1. Top of page
  2. References