Reply to: anaesthetic considerations in measuring Aldrete score and long-term outcomes in craniotomy patients
We read with interest the letter by Bhavsar and Dang1 confirming the interest of the neuroanaesthesia community in our work.2 We shall try to respond to their comments point by point.
- Rationale for the use of midazolam and particularly the relatively high dose (5 mg)?
- To define drug dosage in the three groups, we surveyed the 14 participating neuroanaesthesia units and agreed on the dosing details with all the investigators during a preenrolment meeting. We tried to replicate, as much as possible, everyday practice. We, therefore, designed a pragmatic equivalence trial. Pragmatic trials measure effectiveness, that is the benefit the treatment produces in ‘routine clinical practice’, as opposed to explanatory trials that measure efficacy, that is the benefit a treatment produces under ‘ideal’ conditions.3 Premedication with midazolam was standard in most of the units and it is a short-acting benzodiazepine in adults with an elimination half-life of 1 to 4 h. Average durations of surgery were similar; 312 min [standard deviation (SD) 88], 294 min (SD 88) and 318 min (SD 95) for the sevoflurane-fentanyl, sevoflurane-remifentanil and propofol-remifentanil groups, respectively. If a negative effect on time to achieve an Aldrete postanaesthesia score of 9 could be hypothesised (and we are not sure of it), this effect is equally distributed among all the groups.
- Rationale for both urine and serum biomarkers.
- No extensive data are available on the use of stress biomarkers in anaesthesia. According to the predefined protocol of the NeuroMorfeo trial, blood and urine samples were collected in more than 400 patients to assay two stress markers, cortisol (in plasma and urine) and catecholamines (in urine only). Blood samples were drawn immediately before the induction of anaesthesia and again 1 h after postsurgery awakening. The first urine sample was collected over the 24 h preceding anaesthesia induction; the second urine sample was collected through a urinary catheter and drain bag from induction of anaesthesia to 1 h after patient's awakening. To account for different volumes of urine collected and period of collection, the concentration of all urinary markers was normalised to creatinine concentration. In the context of the present multicentre trial, this protocol seemed to us simple enough to grant its implementation in the already busy wards, while being ethically acceptable to the patients. Whether urine samples accurately reflect biological stress at a given time is suggested by the observation that changes in cortisol levels in the three experimental groups were qualitatively similar when measured in plasma or in urine. The point concerning the chronobiology of stress markers is well taken. However, as comparisons of changes over time in these markers were made across the three experimental groups (with no systematic bias in the daytime of sample collection), we feel that this should not be a major issue in the interpretation and validity of the data. Finally, Bhavsar and Dang are certainly right in putting a word of caution on the impact of dexamethasone on stress markers. Dexamethasone presurgical usage was equally distributed between the groups (sevoflurane-fentanyl: 58 of 137; sevoflurane-remifentanil: 56 of 136; propofol-remifentanil: 58 of 136). Its effect is homogenously distributed among the study groups.
- Why was not a fourth study arm included?
- When we initially planned the study, we designed it with two groups. We decided to opt for such a design because proven strategies already existed (sevoflurane-fentanyl) and the most recent treatment (propofol-remifentanil) was unlikely to have superior efficacy, but may have had other positive effects such as being safer, easier to use or less expensive. During the discussions we had with the centres, it was requested that a sevoflurane-remifentanil group be added because this strategy was frequently used. The fourth group Bhavsar and Dang are suggest, even if theoretically interesting, was not added because it was not frequently used in the Italian Neuroanaesthesia departments that participated in the trial, and even with three groups, Neuromorfeo was already the largest neuroanaesthesia trial ever conducted. The burden of adding yet another group of more than 100 patients was regarded as not easy feasible and would not have been covered by the funding that we obtained from the Italian Ministry of Health.
- We believe that we have to challenge our beliefs and practices and stop arguing that small differences, mostly clinically insignificant and probably irrelevant, are crucial. Neuromorfeo, even if it leaves room for improvements and future studies are awaited, was a pragmatic trial designed to provide more solid elements to our clinical practice, to globally evaluate our anaesthesiological performance, and to improve patient outcome. We are grateful to Dr Bhavsar and Dang for their interest in our work and their appreciation.
Assistance with the letter: none declared.
Financial support and sponsorship: none declared.
Conflicts of interest: none declared.
1. Bhavsar S, Dang AQ. Anaesthetic considerations in measuring Aldrete score and long-term outcomes in craniotomy patients. Eur J Anaesthesiol
2. Citerio G, Pesenti A, Latini R, et al NeuroMorfeo Study Group. A multicentre, randomised, open-label, controlled trial evaluating equivalence of inhalational and intravenous anaesthesia during elective craniotomy. Eur J Anaesthesiol
© 2013 European Society of Anaesthesiology
3. Roland M, Torgerson DJ. What are pragmatic trials? BMJ