We would like to comment on the paper by Park and colleagues on the use of remifentanil for major abdominal surgery . These days, anaesthetists, like other health care professionals, are being asked to do ‘more for less’. Park and colleagues found themselves in a situation where complex procedures, such as liver transplantation, have become routine but without a concomitant increase in the necessary resources. The work by Park and colleagues is a demonstration of the difference anaesthetists can make to patient care and costs if they improve their technique, and also play a greater role in the management and organization of patient care.
We admire the efforts of Park and colleagues to verify scientifically the benefits of their changes to their anaesthetic technique and to the organization of patient care, but we would like to make a few comments on their statistical methods. On checking the data in Table 1 we found errors in the mean and standard deviations given in the 5th, 6th and 7th columns. For example the mean ‘time for recovery after surgery to be ready for discharge’ is 139 and not 132 min, as published. These errors are minor and probably clinically insignificant, but nonetheless undermine the confidence of the reader when evaluating the data in subsequent tables. Tables 2 and 3 contain summaries of important data such as costs, length of time spent on ICU and period of ventilation – mean and standard deviations are given, but without the raw data. The numbers of patients in the various groups are rather small, and judging by the sizes of the standard deviations relative to the means it is apparent that for some of the measured parameters the data are not normally distributed. For nonparametric data it is more appropriate and informative to present the median and interquartile range.
In the text, the authors identify two groups of patients that they say are referred to in Tables 1 and 2 as the ‘remifentanil routine’ and the ‘nonremifentanil group’. We are unable to find these terms in the tables, but assume that they refer to the ‘After study’ and ‘Before study’ groups in Table 2. In the results section, the authors mention difficulties experienced in matching these groups, and we feel that it would have been more informative to include the demographic data for these groups to enable the reader to judge for himself how well-matched the groups were. Also, it is difficult for the reader to assess the significance of the times to extubation in the various groups without knowing which criteria were used for the timing. In those patients anaesthetized with remifentanil the aim was probably to extubate the trachea as soon as possible, but this may not have been the case for the (retrospective) study of patients anaesthetized without remifentanil.
Finally, we are a little concerned by the statement in the results section that the final dose of remifentanil was accepted as the dose required for adequate anaesthesia. From the description in the methods section we gather that that the ‘final infusion rate’ was the rate at the end of surgery, and that an epidural infusion was started 30 min before the end of surgery. It also appears that if hypotension was present, the management included reductions in the remifentanil infusion rate. For any operation the dose of the analgesic and hypnotic agent required for anaesthesia varies dynamically with the intensity of surgical stimulus, the use of local or regional analgesia techniques, and the presence of other drugs that exert hypnotic and analgesic effects. Thus, the final infusion rate may have borne little relationship to the dose required for adequate intraoperative anaesthesia.
1 Park GR, Evans TN, Hutchins J, Borissov B, Gunning KE, Klinck JR. Reducing the demand for admission to intensive care after major abdominal surgery by a change in anaesthetic practice and the use of remifentanil. Eur J Anaesthesiol
2000; 17: 111–119.