Correspondence  |   September 2004
Solid (Concrete) Evidence Needs Both Cement and Sand
Author Affiliations & Notes
  • Malcolm Daniel, M.B. Ch.B., F.R.C.P. (EDIN), F.R.C.A.
  • * Glasgow Royal Infirmary, Glasgow, United Kingdom.
Article Information
Correspondence   |   September 2004
Solid (Concrete) Evidence Needs Both Cement and Sand
Anesthesiology 9 2004, Vol.101, 802-803. doi:
Anesthesiology 9 2004, Vol.101, 802-803. doi:
To the Editor:—
London et al.  1 are to be congratulated on highlighting the physiologic foundations and current clinical controversies regarding the use of β-blockade in the perioperative period; however, we cannot agree that there is justification for the editorial statement by Kertai et al.  2 that “β blockers should be prescribed to all patients with one or more risk factors correlated with higher risk of cardiac complications.” Kertai et al.  assume the case for this intervention is made and consider some possible reasons why this intervention has not been more widely implemented. Many of the comments in the editorial are valid but fail to address other important issues important in the integration of evidence into practice.
There are two influential randomized controlled trials showing that perioperative β blockade improves outcome in defined small samples of patients. These trials, carried out in single centers, provide evidence of the efficacy of perioperative β blockade. We need evidence not only of efficacy but also of effectiveness. The efficacy of an intervention is the degree to which the desired health outcomes are achieved in the best possible circumstances. The effectiveness of an intervention is the degree to which the desired health outcomes are achieved in clinical practice.3 There is a history in anesthesia practice that a powerful effect observed in a study with a small sample size carried out in a single center in which efficacy has been demonstrated4 does not necessarily translate into similar effects in a larger multicenter trial examining the effectiveness of an intervention.5 
We acknowledge that the trials of perioperative β blockade use a simple protocol to guide indications for and the administration of these drugs. These trials have not shown an adverse effect associated with the intervention. The nonoccurrence of an adverse event in a series of patients does not necessarily mean that it cannot happen.6 On the basis of the data contained within the two major trials of perioperative β blockade in the perioperative period, in which a total of 158 patients have received the active intervention, the upper 95% confidence interval for adverse effects may be as high as 2% (3 of 158). It is noticeable that four of the six studies Kertai et al.  cite as showing no adverse effect are in fact two studies that have reported separate short-term and long term outcomes in the same groups of patients.
Integrating evidence into clinical practice takes time; this is a challenge in all areas of medical practice. When we compare the number of studies, number of patients recruited, and variety of settings these patients have been recruited from for studies on the use of β blockade after myocardial infarction the difference is stark. In this scenario there is not just evidence of efficacy but also of effectiveness. The current guidelines for the use of β-blockers after myocardial infarction to reducing all cause mortality, cardiac mortality and nonfatal myocardial infarction are based on multicenter studies enrolling more than 16,000 patients7 and systematic reviews including 82 randomized controlled trials enrolling a total of more than 54,000 patients.8 The practitioner is more confident about the magnitude of effect that may be seen in his or her own practice and the standard of care in this area of medical practice is clearly defined.
It is odd that members of an active research group, such as Kertai et al.,  fail to identify the lack of effectiveness studies as a stumbling block to the introduction of perioperative β-blockade into clinical practice. The challenge for us as anesthesiologists is to provide this level of evidence. London et al.  in the concluding paragraph of their article highlight that there are now several such large-scale trials underway. At this time although much has been done to establish the efficacy of perioperative β blockade, only when the effectiveness of perioperative β blockade has been demonstrated in large-scale trials will we be able to state clearly which patients benefit from perioperative β blockade, the size of the effect we might see in our own practice, and for whom it is a standard of care. At that time we will need to reflect on those additional steps that may promote the implementation of evidence into practice.9 
* Glasgow Royal Infirmary, Glasgow, United Kingdom.
London MJ, Zaugg M, Schaub MC, Spahn DR: Perioperative β-adrenoceptor blockade: Physiological foundations and clinical controversies. Anesthesiology 2004; 100:170–5London, MJ Zaugg, M Schaub, MC Spahn, DR
Kertai MD, Bax JJ, Klein J, Poldermans D: Is there any reason to withhold β blockers from high-risk patients with coronary artery disease during surgery? Anesthesiology 2004; 100:4–7Kertai, MD Bax, JJ Klein, J Poldermans, D
Gray JAM: Evidence Based Healthcare, 2nd edition. Edinburgh, Churchill Livingstone, 2001, pp 169–242Gray, JAM Edinburgh Churchill Livingstone
Yeager MP, Glass DD, Neff RK, Brinck-Johnsen T: Epidural anesthesia and analgesia in high risk surgical patients. Anesthesiology 1987; 66:729–36Yeager, MP Glass, DD Neff, RK Brinck-Johnsen, T
Rigg JRA, Jamrozik K, Myles PS, Silbert BS, Peyton PJ, Parsons RW, Collins KS: Epidural anaesthesia and analgesia and outcome of major surgery: A randomised trial. Lancet 2002; 359:1276–82Rigg, JRA Jamrozik, K Myles, PS Silbert, BS Peyton, PJ Parsons, RW Collins, KS
Eypasch E, Lefering R, Kum CK, Troidl H: Probability of adverse events that have not yet occurred: A statistical reminder. BMJ 1995; 311:619–20Eypasch, E Lefering, R Kum, CK Troidl, H
First International Study of Infarct Survival (ISIS-1) Collaborative Group: Randomised trial of intravenous atenolol among 16,027 cases of suspected acute myocardial infarction. Lancet 1986; 2:57–66First International Study of Infarct Survival (ISIS-1) Collaborative Group,
Freemantle N, Cleland J, Young P, Mason J, Harrison J: Beta blockade after myocardial infarction: Systematic review and meta regression analysis. BMJ 1999; 318:1730–7Freemantle, N Cleland, J Young, P Mason, J Harrison, J
Davis D, Evans M, Jadad A, Perrier L, Rath D, Ryan D, Sibbald G, Straus S, Rappolt S, Wowk M, Zwarenstein M: The case for knowledge translation: Shortening the journey from evidence to effect. BMJ 2003; 327:33–5Davis, D Evans, M Jadad, A Perrier, L Rath, D Ryan, D Sibbald, G Straus, S Rappolt, S Wowk, M Zwarenstein, M