Pain is difficult to measure.
As protagonists of regional analgesia, we were interested to read the article by Stevens et al (1). They have used morphine consumption via patient controlled analgesia (PCA) as their primary outcome measure. The anaesthetic literature is filled with similar studies (2,3). The problem is that this does not show that the intervention improves analgesia. Morphine PCA consumption is an investigator-driven outcome variable and has thus been used repeatedly in acute pain studies. It is easy to measure and relatively easy to show a statistical significance. The principal outcome should surely rather relate to the patients' experience of pain and secondary outcomes should include side-effects from the pain or the analgesia. Unfortunately the variability in PCA requirements is so large that this is not a reliable design substitute(4).
The intervention (modified fascia lata block in this particular study) does not show a significant improvement in the patient pain experience. Nor does the study show a decrease in morphine-related side- effects. The use of clonidine in the trial group clouds the matter somewhat further, however that is not the point of this letter.
If an intervention reduces pain, one needs to have measurable evidence that this is indeed the case. This is often difficult(4). It can be done by independently measuring pain scores numerically or with visual analogue scales (5,6). Farrar et al define "meaningful improvement" as being a 30% reduction in numerical rating of pain intensity. If morphine consumption is used as an objective measure, one should compare patients in the two groups that have similar morphine consumption and then look at their pain scores.
It is also useful to show reduced side-effects in the group that has the intervention, for example, a decrease in morphine-associated nausea and vomiting. However decreasing morphine consumption from, say, 35 mg to 25 mg over 24 hours in a particular patient is very unlikely to decrease the incidence of nausea. One needs larger numbers of patients or a dramatic decrease in opiate consumption to show decreases in nausea and vomiting (6).
We believe it is important to consider this issue in future acute pain intervention studies, whether the intervention is a regional technique or the use of a new or different drug.
(1.) Stevens M, Harrison G, McGrail M., A modified fascia iliaca compartment block has significant morphine sparing effect after total hip arthroplasty. Anaesth Intensive Care 2007; 35:949-952.
(2.) Biboulet P, Morau D, Aubas P, Bringuier-Branchereau S, Capdevila X. Postoperative analgesia after total-hip arthroplasty: Comparison of intravenous patient-controlled analgesia with morphine and single injection of femoral nerve or psoas compartment block. A prospective, randomized, double-blind study. Reg Anesth Pain Med 2004; 29:102-109.
(3.) Stevens RD, Van Gessel E, Flory N, Fournier R, Gamulin Z. Lumbar plexus block reduces pain and blood loss associated with total hip arthroplasty. Anesthesiology 2000; 93:115-121.
(4.) McQuay HJ, Moore RA. Placebo by default. Pain 2000; 86:321- 322.
(5.) Farrar JT, Portenoy RK, Berlin JA, Kinman JL, Strom BL. Defining the clinically important difference in pain outcome measures. Pain 2000; 88:287-294.
(6.) Farrar JT, Berlin JA, Strom BL. Clinically important changes in acute pain outcome measures: a validation study. J Pain Symptom Manage 2003; 25:406-411.
C. W BIRCH
A. J. MCLINTIC
Auckland, New Zealand
Pain is difficult to measure--Reply
Birch and McLintic mention the difficulty of measuring pain and question the validity of morphine consumption as a measure of the utility of a pain intervention.
First, we designed our study on working lists in a regional hospital with no registrar/research assistant. We needed a tool which was objective, easy to measure and free from bias/inadvertant manipulation. We believe that absolute morphine consumption is that tool.
Second, we agree that numerical and visual analogue scales are useful tools for clinically assessing pain in real time. These measures are problematic as a method of assessing research data due to their temporal variability. That is, if you ask the patient about their pain just prior to PCA activation it is invariably higher than if you ask them just after PCA activation. How often is this bias controlled for in studies in the literature?
Our objective was to bring to the literature a modification of an old block so that others may utilise it if they thought it would be of benefit to the patient. It was both our clinical impression (anaesthetists, nurses and patients) and the study's findings that it would benefit patients undergoing total hip replacement by reducing their morphine consumption.