If you wish to respond to a paper or other item already published in the BJA, please go to the abstract/full text version of that item and click on the link "E-Letters: Submit a response to the article".
Electronic Letters to:
|
|
E-letters: Electronic letters published:
-
![[Read E-letter]](/icons/misc/sectionDOWN.gif)
Re: the problem of sample size imported from other authors
- Luc E De Baerdemaeker (5 February 2009)
-
the problem of sample size imported from other authors
- Yasir Al-Tamimi (15 January 2009)
-
Re: Opioid TCI for bariatric surgery and morphine loading dose
- Luc E C De Baerdemaeker, Stefan Jacobs (11 December 2008)
-
Opioid TCI for bariatric surgery
- Yasir A Al-Tamimi (18 November 2008)
|
|
|||
|
Luc E De Baerdemaeker
Send letter to journal:
|
Dear Editor, We would like to thank Dr. Yasir Al-Tamimi Joondalup for his constructive comments on the statistical methods. To calculate sample size, four factors are needed: the values chosen for alpha and beta, the effect size and the variance in the underlying populations. It is common practice to use pilot studies or other published data to estimate variance. We have chosen to derive effect size and variance from the published data of an almost identical protocol in non-obese performed by Derrode et al [2]. Our protocol differed from this work in such a way that we included morbidly obese patients and in the fact that we used paracetamol and diclofenac as adjuvant analgesia to morphine. We have to agree with Dr Yasir that this practice for calculating sample size can import the risk of underestimating the variance and consequently diminish the statistical power of the study. We have to thank Dr. Yasir Al-Tamimi Joondalup to draw the attention of the readers to this statistical pitfall. When calculating a posteriori the power of this study using cumulative morphine consumption as primary endpoint, we found that the statistical power started at 97.6% and was 80.1% with an alpha error of 5% for the samples taken at 120 min. Beyond this time point, statistical power decreased to 30% after 12 hours of registration. Luc De Baerdemaeker, MD, DEAA, Ph.D Department of anaesthesia Ghent university hospital Gent, Belgium Conflict of Interest:None declared |
|||
|
|
|||
|
Yasir Al-Tamimi Joondalup Hospital
Send letter to journal:
|
Dear sir Many thanks for Dr De Baerdemaeker et al for a comprehensive reply to my concerns regarding their trail 'Influence of intraoperative opioid on postoperative pain and pulmonary function after laparoscopic gastric banding:remifentanil TCI vs sufentanil TCI in morbid obesity'[1]. I have to say though that the differences noted in VAS for pain during the first 120 minutes postoperatively may all be due to the way the study protocol is designed and implemented. I still say that the lack of morphine intraoperatively had contributed largely to the significanct differences in pain scores and cumulative morphine consumption in the first 2 hours postoperatively. I have to say that the author may have fallen into importing sample size from another investigator without much scrutiny. Though, most journals and thier editorial boards would accept such a practice, it carries the inherent risk of spreading the effects of an under powered design/methodology to other studies even with a superior methodology and sound research background. I have certainly read the cited reference 'Derrode' that the authors have based their sample size calculation upon [2]. I have to admit that Derrode et al have not actually declared in their published work in this journal in 2003 the whole truth involved in their sample size calculation. It does not take a great deal of expertise to realise that a sample size of 20 in eahc study arm would have required a substantial effect size gone into the calculation process. I have to stress the fact that a substantial effect size may be easy to work with as the number of subjected to be recruited is achievable, but mostly unrealistic to achieve.It consequently may jeorpadise the external validity of the study results. I have to thank the BJA editors and the authors to lend me the space to reply and highlight an area in clinical research that is misunderstood and possible misused. References: 1-L. E. C. De Baerdemaeker, S. Jacobs, P. Pattyn, E. P. Mortier, and M. M. R. F. Struys Influence of intraoperative opioid on postoperative pain and pulmonary function after laparoscopic gastric banding: remifentanil TCI vs sufentanil TCI in morbid obesity Br. J. Anaesth. 2007; 99: 404- 2-Derrode N, Lebrun F, Levron JC, et al. Influence of peroperative opioid on postoperative pain after major abdominal surgery:sufentanil TCI versus remifentanil TCI. A randomized, controlled study. Br J Anaesth 2003; 91: 842–9 Conflict of Interest:None declared |
|||
|
|
|||
|
Luc E C De Baerdemaeker , Stefan Jacobs
Send letter to journal:
|
Dear Editor, We would like to thank Dr. Yasir Al-Tamimi Joondalup for his comments and his interest in our paper and for the opportunity to reply to him. We will address his points in turn. 1. Details on sample size calculation and patient selection can be found in our methods section. Since the primary endpoint of the study was the postoperative morphine consumption, power analysis was based on a similar protocol performed in non-obese patients by Derrode and coworkers1. Group allocation was at random with blinded envelopes. 2. We would like to point out that patients in both study groups received paracetamol 2 g IV and diclofenac 150mg IV at the beginning of surgery. This implies that patients in the remifentanil group received a non-opioid transitional postoperative analgesia. The initial mean (SD) VAS score in the remifentanil group on admission in the post anaesthesia care unit (PACU) was 4.3 (1.7) and this reflects sub optimal analgesia but to our opinion not in the range of unethical clinical practice. In non-obese patients, most clinicians will use an intraoperative intravenous morphine dose of 0.1-0.25 mg.kg–1 administered 30-60 minutes before the end of surgery for the immediate postoperative analgesia after a remifentanil-based anaesthesia. From the moment of starting our study until now, there are no clear guidelines in the literature on the safe use of an intra-operative loading dose of morphine calculated on ideal body weight (IBW) in morbidly obese patients. When studying the influence of diurnal variation and morbid obesity on the morphine requirements using IV patient controlled analgesia (PCIA), Graves and colleagues 2 found that morphine dosing rate (mg.kg-1.hr-1) normalized to IBW was a better predictor of analgesic requirements. More recent studies on the use of PCIA with morphine in morbidly obese patients have demonstrated its safety 3, 4, even in obese patients with obstructive sleep apnoea syndrome (OSAS) 5. All authors used intravenous morphine in the PACU titrated individually in divided doses of 2.5-5mg with 10 minutes intervals to achieve an acceptable pain score before instituting their PCA device. Only Ahmad and co-workers 5 used a prudent intraoperative loading dose of 50µg.kg-1 IBW morphine at the moment of stopping the pneumoperitoneum. In our study, performed on morbidly obese patients without OSAS nor serious cardiopulmonary disease, the PCIA device delivered a morphine loading dose (0.15 mg.kg-1 IBW) at the moment of first analgesic request. Our postoperative data on arterial blood gas analysis and spirometry show the safety of this strategy in this type of obese patients. Nevertheless, it took two hours to register VAS pain scores below three in both study groups. Until we see prospective studies on the accurate timing and safety of an intraoperative loading dose of morphine for morbidly obese patients in the range of 0.15mg.kg-1 IBW, the PCIA recommendations of Levin 6 still apply: no basal infusion of morphine, bolus doses of 0.5-1.0 mg with a ten minutes interval and titration to a desirable effect within the first few hours after surgery. 3. We agree with Dr Yasir that PCIA morphine is not a routinely used postoperative analgesic regimen for laparoscopic gastric banding. For study purposes, morphine consumption was used as an objective and quantitative parameter for measuring postoperative pain. We are convinced that with simple oral analgesics (and infiltration with local anaesthetics), satisfactory postoperative analgesia can be achieved. Luc De Baerdemaeker, MD, DEAA, Ph.D. Stefan Jacobs, MD Department of anaesthesia Ghent university hospital Gent, Belgium References: 1. Derrode N, Lebrun F, Levron JC et al. Influence of peroperative opioid on postoperative pain after major abdominal surgery: sufentanil TCI versus remifentanil TCI. A randomized, controlled study. Br J Anaesth 2003; 91: 842-9 2. Graves DA, Batenhorst RL, Bennett RL et al. Morphine requirements using patient-controlled analgesia: influence of diurnal variation and morbid obesity. Clin Pharm 1983; 2: 49-53 3. Choi YK, Brolin RE, Wagner BK et al. Efficacy and safety of patient- controlled analgesia for morbidly obese patients following gastric bypass surgery. Obes Surg 2000; 10: 154-9 4. Charghi R, Backman S, Christou N et al. Patient controlled i.v. analgesia is an acceptable pain management strategy in morbidly obese patients undergoing gastric bypass surgery. A retrospective comparison with epidural analgesia. Can J Anaesth 2003; 50: 672-8 5. Ahmad S, Nagle A, McCarthy RJ et al. Postoperative hypoxemia in morbidly obese patients with and without obstructive sleep apnea undergoing laparoscopic bariatric surgery. Anesth Analg 2008; 107: 138-43 6. Levin A, Klein SL, Brolin RE et al. Patient-controlled analgesia for morbidly obese patients: an effective modality if used correctly. Anesthesiology 1992; 76: 857-8 Conflict of Interest:None declared |
|||
|
|
|||
|
Yasir A Al-Tamimi, Anaesthetist University of Western Australia
Send letter to journal:
|
Dear sir, I read with interest the paper by De Baerdemaeker et al in 2007, I wish first to thank the authors for this valuable publication. I however, want to raise some concerns about the paper: 1- the authors didn't mention anything about sample size calculation and how these 40 patients were selected. 2- Of concern to me was not to give morphine iv earlier than the recovery time in PACU, as we know morphine peak effect takes a long time to set in and I believe delaying giving these patients (especially the remifentail group) morphine is unethical. 3- in our hospital we don't prescribe PCIA to laparoscopic banding patients post operatively routinely, and most patients do well and leave hospital after thier gastrograhine swallow test next morning on simple oral analgesia. Thanks, Dr Yasir Al-Tamimi Joondalup Hospital, Perth WA, Australia University of Western Australia Conflict of Interest:None declared |
|||