John W. (Bill) McEvoy
Administering IV fentanyl during PCI delayed the absorption of ticagrelor by as many as 4 hours, according to a poster presented at the European Society of Cardiology Congress.
It is common to use opiates such as fentanyl during PCI in the United States, but not in other countries, presenter John W. (Bill) McEvoy, MBBCh, MEHP, MHS, assistant professor of medicine and epidemiology at Johns Hopkins Medicine and a Cardiology Today Next Gen Innovator, told Cardiology Today.
Previous studies have shown that other opiates such as morphine delay absorption of P2Y12 inhibitors during PCI, but little was known about fentanyl, McEvoy said.
“There is a stark difference in practice between European and U.S. cath labs in terms of how opiates are administered,” he said in an interview. "Opiates, and fentanyl in particular, are very uncommonly given in the cath lab for regular coronary angiography and PCI in Europe. When I came to the United States from Europe for further medical training, I was surprised to see that almost everybody was getting opiates for a minimally invasive procedure. Given recent data demonstrating that morphine delays the absorption of P2Y12 inhibitors, we thought that maybe there is a similar association for fentanyl, which is routinely given to patients in U.S. cath labs."
He said fentanyl is often given at the same time patients are loaded with P2Y12 inhibitors before coronary angiography or PCI. Also, “routine thinking is that when a stent goes in, you want the platelets inhibited so they don’t stick to the stent,” he said.
For the double-blind PACIFY trial, McEvoy and colleagues randomly assigned 212 patients undergoing coronary angiography at a single center who were not taking chronic P2Y12 inhibition or anticoagulation to receive IV fentanyl or not. The 70 patients requiring PCI were loaded with a 180-mg dose of ticagrelor (Brilinta, AstraZeneca). All patients received a local anesthetic and IV midazolam for pain.
The primary outcome was ticagrelor blood concentration at 24 hours after administration. Secondary outcomes included platelet inhibition assessed by two different methods at 2 hours, and self-reported maximum pain experienced during the procedure.
For the pain and anxiety outcome, there were 105 patients (mean age, 65 years; 38% women) in the no-fentanyl group and 107 patients (mean age, 61 years; 20% women) in the fentanyl group. The platelet function and pharmacokinetics analyses included an intention-to-treat cohort of 35 patients in both the no-fentanyl and fentanyl groups and a per-protocol cohort of 33 patients in the no-fentanyl group and 37 patients in the fentanyl group. The protocol allowed operators to administer fentanyl to patients assigned to no fentanyl if their pain was not being controlled, McEvoy said.
Blood concentration of ticagrelor was lower at 24 hours in the fentanyl group vs. the no-fentanyl group (area under the plasma concentration-time curve, 2,016 ng/mL vs. 3,341 ng/mL; P = .03), according to the researchers.
High on-treatment platelet reactivity at 2 hours was 0% in the no-fentanyl group and 37% in the fentanyl group as assessed by light transmission platelet aggregometry (P = .004). Similar results were observed (3% vs. 22%; P = .02) after assessment by the VerifyNow system (Accriva Diagnostics).
There were no differences between the groups in procedural anxiety, maximum procedural pain, pain score greater than 5 of 10, pain at 2 hours and nurse-recorded maximum pain, McEvoy and colleagues found.
“The results replicated what was seen for morphine,” McEvoy told Cardiology Today. “It was important to determine that fentanyl had a similar effect on P2Y12 inhibitors as morphine does, because fentanyl is much more short-acting than morphine. I would argue the implications are very sizable, because fentanyl is a drug given routinely in the cath lab at a time when patients need to have their platelets inhibited, especially if they are getting a stent. This drug is very rarely used outside the U.S., which begs the question, why are we using it here at all?”
The results could prompt U.S. operators to rethink administering fentanyl routinely, especially if it is for anxiety instead of pain, McEvoy said.
“There’s a strong argument that fentanyl should be restricted to as-needed, rather than given routinely and nonselectively,” he said.
A question, he said, is whether the 4-hour delay in ticagrelor absorption is clinically important.
“Our study is not powered to answer that question, but it certainly generates the motivation to look at clinical outcomes in patients who receive fentanyl,” McEvoy said. “Unfortunately, this is such an off-the-radar topic that I have not been able to find any data on stent thrombosis and opiate use. These results justify the acquisition of those data going forward. Certainly, there is a signal that’s concerning.” – by Erik Swain
McEvoy JW, et al. Poster P3221. Presented at: European Society of Cardiology Congress; Aug. 26-30, 2017; Barcelona, Spain.
Disclosure: McEvoy reports he has no relevant financial disclosures.