Background: Despite a large number of clinical investigations in posto
perative patients, the pharmacology of epidural fentanyl, alfentanil,
and sufentanil has not been well characterized in a human laboratory s
etting. In this double-blind, placebo-controlled crossover study, we e
valuated analgesia and side effects produced by epidural fentanyl (30
and 100 mu g), alfentanil (300 and 1,000 mu g), and sufentanil (3 and
10 mu g) in volunteers. Methods: Each of 12 volunteers participated in
four separate study sessions. The pain model was cutaneous electrical
stimulation of intensity sufficient to produce a pain report of 5 on
a 0-5 scale, delivered alternately to the finger and toe. Ventilatory
drive, pupil size, and subjective ratings of alertness, nausea, and pr
uritus were measured using visual analog scales. On each study day, af
ter baseline measurements, an epidural catheter was placed at the L2-L
3 or L3-L4 interspace. Subjects received the small dose of study drug,
and the tests were repeated at 2, 55, and 35 min after drug administr
ation. Subjects then received the large dose of the same study drug, a
nd the series of tests was repeated at 2, 55, and 95 min later. Plasma
opioid concentrations were measured using gas chromatography-mass spe
ctrometry. Results: Dose-dependent analgesia was found for all study o
pioids. For both doses of all opioids, toe analgesia was significantIy
greater than finger analgesia. Epidural fentanyl and alfentanil provi
ded greater analgesia than sufentanil at the doses investigated. Sedat
ion, increased end-tidal carbon dioxide, and pupillary constriction oc
curred only after the large epidural doses of all opioids. Overall, th
e incidence of subjective side effects was low, but four subjects expe
rienced pruritus after 100 mu g fentanyl, and four were nauseated afte
r 1,000 mu g alfentanil. Plasma opioid concentrations were near minimu
m effective analgesic plasma concentrations only after larger epidural
doses. Conclusions: Lumbar epidural fentanyl, alfentanil, and sufenta
nil produce selective lower-extremity analgesia. Low plasma opioid con
centrations measured after small epidural opioid doses suggest a spina
l mechanism for analgesia. Larger doses of epidural opioids result in
systemic absorption and are likely to produce supraspinal analgesia an
d other side effects.