Wednesday, June 3, 2009
Pain recently published a small cohort study that attempted to provide insight into the question, "Does cachexia decrease the absorption of transdermal fentanyl?" (Not to be confused with the question, "Is transdermal fentanyl less effective in cachectic patients? Read on.)
An urban legend exists (perpetuated by well-meaning hospice and palliative care providers) that fentanyl TD isn't as effective in patients with cachexia (and maybe should be avoided in that population). The absorption of fentanyl TD depends on skin permeability and local blood flow, and perhaps one or both are reduced in cachexia. The "state of the science," as far as I know, says that while this hypothesis may make sense physiologically, the question of reduced effectiveness remains unstudied.
This Finnish study recruited twenty patients with cancer pain, ten who had a normal weight (BMI average BMI 23 kg/m2) and ten who had cachexia (average BMI 16 kg/m2). Eight of the patients (five with normal weight and three with cachexia) were already using fentanyl TD. The other patients were converted from their previous opioid (morphine or oxycodone) to an equianalgesic dose of fentanyl TD. The two groups were similar in age, baseline skin temperature, sweating, and blood flow (all factors that may influence fentanyl pharmacokinetics). The cachectic group had a longer median duration of disease than the normal weight group (5.5 months vs. 2.5 months, respectively), skin fold thickness, and a higher fentanyl TD dose based on conversion from the other opioid or previous fentanyl needs (96 +/- 29 mcg per hour vs. 42 +/-10 mcg per hour).
The authors attribute the higher opioid dose in the cachectic group to greater pain associated with more advanced disease, but do also mention the possibility that of those who were on fentanyl beforehand, participants with cachexia may have required more because of absorption issues. Although the authors presumably had data that may have confirmed or denied this hypothesis (i.e. - the fentanyl TD doses of those already on it before the study) they do not present this data...I'm unsure why they didn't just exclude patients already on fentanyl.
Subjects were followed over the subsequent 72 hours. Plasma fentanyl concentrations were determined at baseline, 4 hours, 24 hours, 48 hours, and 72 hours and then were adjusted for dose. The authors found that at 48 and 72 hours, cachectic subjects had a significantly (based on SEM) lower dose adjusted plasma concentration of fentanyl but at earlier time intervals there we no statistically significant differences. The mean plasma concentration per dose (mcg/l/dose) was 0.014 vs. 0.023 at 48 hours and 0.012 vs. 0.024 at 72 hours in cachectic and normal weight subjects, respectively. When looking at the 95% confidence interval for these time intervals, though, there was some overlap between the plasma concentration per dose (mcg/l/dose) of normal weight patients and cachectic patients, so the difference really isn't significant although it is a very small sample of patients.
The authors report that the median time to maximum fentanyl concentration in subjects naive to fentanyl was 48 hours in cachectic patients compared to 24 hours in normal weight patients. Of course, this is a median, and one would need to do continuous sampling to really determine whether there is a difference between the two groups. Pain intensity as measured by visual analogue scale was comparable between groups and didn't change significantly from baseline to 72 hours. Fentanyl dose changes were not permitted over the 3 days, but patients could use short acting opioids for breakthrough pain. The amount of breakthrough pain medication needed was not reported.
The authors discuss the unpredictable relationship between opioid concentration and analgesia:
There seems to be no concentration-effect relationship for fentanyl or other strong opioids and lack of correlation between plasma fentanyl concentration and analgesia is evident in the present study and that of Solassol et al., where transdermal fentanyl administration resulted in a wide range of individual plasma fentanyl concentrations but provided adequate analgesia. The plasma opioid concentration is an unreliable reflector of opioid analgesia since the opioid analgesic effect is also influenced by factors such as the intensity of pain, opioid tolerance and the concentration of the opioid in the central nervous system after crossing the blood–brain barrier.But then go on to conclude:
The clinical implication of our study is that transdermal fentanyl is not the opioid of choice for cachectic cancer patients with pain. Poor transdermal absorption of fentanyl in these patients may result in inadequate analgesia despite an apparently large dose of opioid. Further research is needed to study the factors that lead to impaired skin permeability and poor fentanyl absorption in patients with cancer cachexia.These two quotes appear contradictory. If there is apparently no concentration-effect relationship for opioids and pain relief, then you can't conclude from this study (which looked at fentanyl concentrations closely, but not very closely at more appropriate outcomes such as VAS scores and breakthrough opioid needs) that fentanyl TD is a second-class opioid in cachectic patients. I'm okay with a conclusion that cachexia may impair fentanyl absorption, but not okay with concluding that fentanyl is not the opioid of choice for cachectic patients, at least on the basis of this study.
Now, before you conclude that I have a financial relationship with Janssen Pharmaceuticals, makers of Duragesic (I don't) or am just one of those that thinks fentanyl patches are a "slice of heaven," I'm firmly planted in the camp of those who think fentanyl patches should generally be reserved for patients who can't swallow or who have some other compelling reason. This is based more on issues related to interindividual variability of absorption (unrelated to weight), cost, and wanting to use a single opioid if possible (e.g. MSIR plus MSSR).
We'll wait for a study with larger cohorts of patients where the primary endpoints are related to effectiveness (e.g. VAS and breakthrough opioid needs) over a longer period of time. In the meantime, this study won't change my practice - in cachectic patients for whom I initiate fentanyl TD, I'll continue to use recommended conversions, make sure they have a good breakthrough opioid, and adjust the dose of the patch as necessary.