Skip to Content
Interactive Textbook on Clinical Symptom Research Logo


Home button

Trial Design: Pain Sections
Author Bio
Currently selected section: Introduction
Placebo Effects
Single Dose Trials
Repeated Dose Trials
Explanatory Versus Pragmatic
Dose-Response
Parallel Group Versus Crossover
Conclusion
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Commentary of Eugene Laska, Statistician

Chapter 1: Clinical Trials of Pain Treatment: Introduction
 
  The lumpers' arguments: Animal experiments also show that a variety of disease models share some pain mechanisms. For example, any type of peripheral inflammation, whether from surgery, arthritis, or chemical irritation increases the sensitivity of nociceptors, and the resulting barrage of impulses sensitizes dorsal horn neurons. The sensitization of dorsal horn neurons is largely mediated by the excitatory transmitter glutamate, the same system that sensitizes dorsal horn neurons after nerve injury. For these reasons, anti-inflammatory drugs and NMDA glutamate receptor antagonists have broad activity across animal pain models and pain patients.

The search to distinguish mechanisms of pain syndromes has great intellectual and grant-getting appeal and has dominated basic research. It is not surprising that clinicians have been trying to make similar distinctions in patients. However, some claims that different pain syndromes have differential analgesic sensitivity have been overstated. For example, a widely repeated axiom that neuropathic pain (pain associated with peripheral nerve injury) was insensitive to opioid analgesics, based on a single drug challenge in eight patients (Arner and Meyerson, 1988), has been refuted by further work. Although occasional patients are refractory to opioids, opioid responses to neuropathic cancer pain (Cherny et al., 1994), post-herpetic neuralgia (Watson et al., 1998) and diabetic neuropathy (tramadol paper) are readily shown in clinical trials.

A reasonable middle ground in this controversy is that there are distinct pain mechanisms with different patterns of responses to drugs, but these mechanisms may cross the borders of conventional disease-based diagnoses (Woolf and Decosterd, 1999). A challenge in clinical research is that current tests to determine pain mechanisms in individual patients may require several days of uncomfortable quantitative sensory testing and nerve blocks. Moreover, these tests are persuasive in only the small subset of patients with findings such as light touch-evoked pain in a distal limb (Gracely et al., 1992). For the large majority of patients, simple measures such as questionnaires that evaluate pain quality and evocative factors (Galer and Jensen, 1997), and standard bedside sensory exam must suffice (Woolf and Decosterd, 1999), but these are only indirect assessments of pain mechanisms. Furthermore, if some mechanisms are common to many pain syndromes, differences in drug response may be modest. The standard sample size formulas for clinical trials (below) show that in order to reduce the detection threshold for a treatment difference by 50%, one needs a four-fold increase in sample size, so large multi-center analgesic trials will probably be needed to uncover differential responses to treatment due to differences in pain mechanisms.

Table 1.3 Sample Size Formula for Clinical trial: Capital N equals two times sigma squared times a function of alpha and beta, divided by delta squared, where capital N equals the number of patients needed per treatment group, sigma equals the standard deviation of effect, delta equals the size of change one wishes to detect, alpha equals the Type I error one can tolerate, and beta equals the Type 2 error one can tolerate.

 

Page 4 of 52
 Previous Page