Fentanyl

Observational studies

In a small open study, in which patients using transdermal fentanyl were switched to an intravenous infusion of fentanyl for acute cancer-related pain relief, there was no excessive sedation or opioid adverse effects [ ].

The safety of intrathecal fentanyl was demonstrated In 31 patients undergoing transurethral resection of the prostate fentanyl (n = 15) or saline (n = 16) was added to ropivacaine for subarachnoid block; the only significant adverse reaction was pruritus [ ].

Transdermal fentanyl 25 micrograms/hour has been assessed in 22 patients with painful oral mucositis induced by high-dose chemotherapy after stem cell transplantation [ ]. Three had severe dizziness, severe vomiting, and extensive rashes. These patients were subsequently withdrawn from the analysis. Mild nausea (in 32%) and dizziness (in 11%) were the two most common adverse events. There was no respiratory depression, constipation, or withdrawal symptoms.

The incidence of adverse events in 14 386 patients receiving fentanyl has been studied after implementation of the Joint Commission on Accreditation of Healthcare Organization guidelines for procedural sedation and analgesia in an urban tertiary-care children's hospital [ ]. The combination of fentanyl + midazolam was the regimen with the highest incidence (9.7%) of adverse events compared with other regimens involving opioids. The most common adverse event was hypoxemia (7.1%), followed by hypotension (1.0%), prolonged sedation (0.7%), vomiting (0.4%), airway obstruction (0.4%), bradycardia (0.1%), pain-related agitation (0.1%), and agitation (not pain-related) (0.04%).

The addition of transdermal fentanyl in 226 patients with severe pain from rheumatoid arthritis improved pain relief, function, sleep, well-being, and patient satisfaction [ ]. Adverse events were experienced by 39 patients (17%) and 10% stopped taking the treatment. The most frequent adverse events were nausea and vomiting; others included dizziness, sedation, and constipation.

Combinations of varying doses of fentanyl with bupivacaine have been compared with bupivacaine alone in 58 infants undergoing lower abdominal and urological surgery [ ]. The treatments were 0.5% hyperbaric bupivacaine alone and bupivacaine + fentanyl 0.25, 0.5, or 1 microgram/kg. There was no significant difference in adverse reactions across the groups. There was pruritus in three infants who received fentanyl.

The safety and efficacy of patient-controlled analgesia (PCA) using fentanyl 1 microgram/kg plus continuous intravenous infusion of midazolam 2 micrograms/kg/minute has been explored in 16 children with moderate to severe postoperative pain [ ]. Minor adverse reactions occurred in 25%. Two had pruritus, one a maculopapular rash, and one vomiting. There were no problems related to PCA. In a similar study PCA fentanyl 1 microgram/kg/hour + booster doses of intravenous fentanyl 1 microgram/kg were given to 18 children with moderate to severe pain [ ]. Adverse reactions were reported by 39% and included pruritus (17%), vomiting (11%), and rashes (11%); there were no major adverse reactions.

In a prospective open trial in 81 patients with moderate to severe osteoarthritis, transdermal fentanyl 25 micrograms/hour was associated with mild to moderate adverse reactions [ ]. Treatment was withdrawn in 32 patients because of nausea, vomiting, or dizziness.

Comparative studies

Bupivacaine + clonidine as an alternative to bupivacaine + fentanyl has similar analgesic efficacy and is associated with fewer adverse reactions, as reported in a study of 47 children undergoing the Nuss procedure [ ]. The children were randomized to bupivacaine + clonidine, bupivacaine + fentanyl, or bupivacaine + fentanyl + clonidine. Vomiting and pruritus were significantly more common in those who received fentanyl. Vomiting occurred in 27% of those who received bupivacaine + clonidine, compared with 69% of those who received bupivacaine + fentanyl and 55% of those who received bupivacaine + fentanyl + clonidine. Pruritus occurred in 85% of those who received bupivacaine + fentanyl and 54% of those who received bupivacaine + fentanyl + clonidine, but in none of those who received bupivacaine + clonidine.

Placebo-controlled studies

In a double-blind, randomized, placebo-controlled study 29 boys undergoing day-case penile surgery were allocated to either intravenous fentanyl 1 microgram/kg intraoperatively or intravenous saline [ ]. Pain and opioid-related adverse reactions were monitored for the first 24 hours after surgery. The authors concluded that intraoperative intravenous fentanyl is associated with an increased incidence of nausea and vomiting, without any significant contribution to postoperative pain relief.

In a placebo-controlled study in 399 patients with moderate to severe osteoarthritis transdermal fentanyl produced better pain management [ ]. The most common adverse reactions were nausea, vomiting, and somnolence.

Systematic reviews

Transdermal iontophoretic fentanyl was as effective as morphine PCA in the management of acute postoperative pain, but was associated with fewer treatment failures related to adverse events, less pruritus, and less somnolence [ ].

Cardiovascular

A hypertensive crisis occurred in a patient with a previously unknown pheochromocytoma [ ].

Bolus intravenous fentanyl 1 micrograms/kg in stereotactic brain biopsy for intracranial mass lesions in 135 patients was well tolerated but provided less hemodynamic stability than alfentanil and remifentanil [ ].

Fentanyl failed to produce adequate protection of the myocardium from ischemic injury following cardiopulmonary bypass in a comparative study with morphine in 46 patients [ ]. Global cardiac function was assessed by the myocardial performance index. Fentanyl significantly improved left ventricular function, but morphine improved global ventricular function.

Midazolam + fentanyl has been compared with dexmedetomidine during carotid endarterectomy in 56 patients [ ]. Those who received fentanyl required more interventions to control hypertension and/or tachycardia (72% versus 40%), and in the post-anesthesia care unit fewer interventions were required for hypotension (11% versus 28%). Many more of the patients who were given dexmedetomidine required additional pain relief (72% versus 38%).

Midazolam (median dose 0.11 mg/kg) and fentanyl (median dose 2.77 micrograms/kg) were given intravenously to 1226 children, mean age 9 years, undergoing endoscopy; there were mild to moderate cardiovascular events, mainly hypertension and hypotension, in 10% [ ].

Patient-controlled epidural analgesia consisting of 0.16% ropivacaine + 3.5 micrograms/ml fentanyl (n = 25) has been compared with epidural boluses of 0.1% methadone 4–6 mg 8-hourly (n = 24), in controlling acute post-thoracotomy pain [ ]. Fentanyl + ropivacaine caused more hypotension during the first postoperative hour.

Heart rate variability during anesthesia has been studied in patients who received sevoflurane + fentanyl; in two cases there was reduced heart rate variability associated with junctional rhythm [ ]. In a 9-year-old girl with hereditary sensory autonomic neuropathy type 2 there was altered heart rate variability with improved hemodynamic stability after the administration of propofol + fentanyl [ ].

A 68-year-old man with Brugada syndrome developed ventricular tachycardia during general anesthesia and thoracic paravertebral block using anesthetic agents + fentanyl [ ]. Is not clear how fentanyl contributed to the cardiac rhythm changes in this case.

Respiratory

Even small doses of fentanyl can cause respiratory depression. Delayed respiratory depression can be a particular problem in the elderly, in whom the half-life is approximately three times longer than in younger patients [ ]. Respiratory depression has been reversed with nalbuphine; doxapram could only antagonize this effect for 2–5 minutes [ ]. However, the need for prolonged treatment of respiratory depression with naloxone, because of pharmacokinetic variability and/or transdermal drug reservoir, has been emphasized by several authors [ , ].

The respiratory effects of fentanyl have been demonstrated in 21 volunteers who were given a high-dose infusion [ ]. Four of them developed apnea shortly after the infusion, thought to be due to rapid crossing of the blood–brain barrier and rapid depression of respiratory neurons.

Fentanyl can evoke the pulmonary chemoreflex, as evidenced by 50% of patients in one study and 28% in another, who coughed after the administration of fentanyl through a central line [ ]. The coughing caused by fentanyl is inhibited by terbutaline [ ].

Intravenous fentanyl is associated with coughing in 28–45% of patients. Coughing due to fentanyl may not always be benign and brief; it can sometimes be explosive, requiring immediate intervention on the operating table. Coughing occurs because fentanyl constricts the tracheal smooth muscle, stimulating the irritant receptors. Other possible factors are release of histamine, leukotrienes, interleukins, and other inflammatory mediators from mast cells. In 200 patients scheduled for elective laparoscopic cholecystectomy under general anesthesia and given intravenous fentanyl 2 micrograms/kg, pretreatment with salbutamol, beclomethasone, or sodium cromoglicate by aerosol before the fentanyl bolus reduced the incidence of cough [ ].

• A 7-year-old boy with trisomy 21 (Down syndrome) had explosive coughing, 30 seconds after fentanyl 50 micrograms (2 micrograms/kg) had been injected and flushed through an intravenous cannula [ ]. The cough was unproductive and persisted in spasmodic bursts for a further 2–3 minutes until anesthesia was induced with propofol 60 mg and atracurium 15 mg intravenously. The coughing immediately ceased. A petechial rash in the conjunctivae and periorbital regions was subsequently noted and disappeared by the end of the first postoperative day.

The incidence of fentanyl-induced cough has been studied in patients undergoing surgery [ ]. Fentanyl (100 micrograms for those weighing 40–69 kg; 150 micrograms for those weighing 70–90 kg) was given intravenously at three different rates, over 2, 15, or 30 seconds. A longer injection time resulted in a reduced incidence of cough. Light smoking was associated with a reduced incidence of cough.

The addition of intrathecal fentanyl 20 micrograms to hyperbaric bupivacaine in 40 women undergoing cesarean section improved the quality of subarachnoid blockade but did not result in worsening of respiratory function [ ].

A method of estimating analgesic fentanyl requirements after surgery, while avoiding respiratory depression, has been evaluated prospectively [ ]. The method was based on a fentanyl challenge before surgery. The patients required only very small adjustments to the settings that were based on estimates from the challenge. There was no evidence of respiratory depression.

The combination of midazolam + fentanyl in 100 children caused adverse events in 25%, with respiratory events in 9% [ ]. Most of the respiratory events were mild to moderate, and apnea occurred in two patients.

• Severe respiratory depression simulating a stroke occurred in a 78-year-old woman on patient-controlled analgesia with fentanyl after routine surgery for endoluminal graft repair for abdominal aortic aneurysm [ ]. The fentanyl analgesia was programmed at 20 microgram doses with 5 minutes lockout, resulting in administration of 600 micrograms of fentanyl 16 hours after the operation. She had slightly impaired renal function (creatinine 167 μmol/l) and a history of obstructive sleep apnea, which together with the fentanyl contributed to the development of respiratory depression.

Intravenous fentanyl is often associated with a reflex cough. In 300 adults undergoing elective surgery, intravenous fentanyl 2 micrograms/kg in 2 seconds was followed by cough in 39%. Perioperative administration of clonidine 2 micrograms/kg reduced the incidence of cough to 17% [ ].

Low-dose intravenous ketamine (0.15 mg/kg) reduced fentanyl-induced cough and the time of onset of cough in a placebo-controlled study in 360 patients undergoing elective surgery who received fentanyl 1.5 micrograms/kg in 5 seconds [ ].

Nervous system

Movement disorders after withdrawal of continuous infusion, without the characteristic autonomic signs of opioid withdrawal, have been reported in children [ ].

Fentanyl-induced seizures have been reported [ ].

Life-threatening complications have included raised intracranial pressure and critically reducing cerebral perfusion [ ].

• A 55-year-old man was given fentanyl 0.05 mg for treatment of left chest pain and immediately developed an acute confusional state and fluctuating tetraparesis [ ]. The symptoms abated 12 hours after withdrawal. A provocation test confirmed that fentanyl 0.1 mg was enough to cause myoclonic and dystonic reactions with increased agitation. Administration of intravenous naloxone 0.8 mg improved the condition.

• A 14-year-old girl developed a dystonic reaction to fentanyl 50 micrograms given as a general anesthesia for dental extraction; her abnormal movements stopped completely after 3 days [ ].

Fentanyl can displace bilirubin from albumin in neonates, with a risk of kernicterus; other drugs should be used [ ].

Serotonin syndrome occurred in a 25-year-old man who was given intrathecal fentanyl [ ]. The authors suggested that the combination of fentanyl with dihydroergotamine, which the patient was taking for migraine, precipitated the serotonin syndrome and that his illicit use of 3,4-methylenedioxymetamfetamine and ephedrine exaggerated the clinical symptoms.

In a comparison of propofol + ketamine and propofol + fentanyl for analgesia in children aged 5–60 months, both regimens produced sedation and analgesia, but there was more restlessness (in 7 patients) among those who received propofol + fentanyl [ ].

Psychological

Mood alteration during patient-controlled epidural anesthesia with either morphine or fentanyl was compared in a randomized, double-blind study of 52 patients undergoing elective hip or knee joint arthroplasty under general anesthesia [ ]. Mood was assessed preoperatively and at 24, 48, and 72 hours, using the bipolar version of the Profile of Mood States. Pain intensity postoperatively did not vary with morphine or fentanyl and, as expected, both fentanyl and morphine users had significant somnolence, pruritus, and nausea compared with baseline. With morphine, the mean score for measures of composure/anxiety, elation/depression and clear-headedness/confusion increased, indicating a change toward the more positive pole, but there were negative changes for the fentanyl users’ scores for five of the six components of the Profile of Mood States. The difference in test scores between morphine and fentanyl was significant at 48 hours of patient-controlled anesthesia and 24 hours after withdrawal. There was no correlation between mood scores and pain scores, and mood scores with fentanyl fell with increasing plasma concentrations. Previous investigations have shown transient positive feelings with intravenous fentanyl, followed by more negative feelings in the longer term. The authors suggested that the differences in mood between the two groups may have been explained by differences in the lipid solubility and pharmacokinetics of epidural morphine and fentanyl.

Endocrine

Hypothalamic–pituitary–adrenal (HPA) axis suppression has been attributed to chronic administration of fentanyl [ ].

• A 64-year-old man with chronic sciatic pain had taking transdermal fentanyl 200 micrograms/hour for 2 years. He developed back pain, miosis, somnolence, and a blood pressure of 70/40 mmHg. Adrenocortical insufficiency was diagnosed, but the cause was unclear, and he was given hydrocortisone 25 mg/day. After poor compliance with hydrocortisone he again presented in adrenal crisis. On re-stabilization, opiate-induced suppression of the HPA axis was suspected. On gradual reduction of the dose of fentanyl, HPA axis function improved markedly.

Gastrointestinal

In a retrospective cohort study, 1836 patients using long-term opioids for chronic malignant and non-malignant pain were analysed to compare the incidence of constipation; 601 used transdermal fentanyl, 514 used morphine CR, and 721 used oxycodone CR) [ ]. Crude (unadjusted) rates of constipation were 3.7% for transdermal fentanyl, 5.1% for morphine, and 6.1% for oxycodone CR. Transdermal fentanyl had a lower annual incidence density and risk of constipation than the other two medications. The adjusted risk of constipation was estimated as 78% greater with oxycodone CR and 44% greater with morphine CR than with transdermal fentanyl.

Constipation is a common adverse reaction to opioids. In a 13-month, open, parallel-group comparison of a fentanyl transdermal reservoir and oral modified-release morphine in 677 patients with chronic low back pain who were given transdermal fentanyl or oral modified-release morphine there was less severe constipation with fentanyl [ ]. The Patient Assessment of Constipation Symptoms 12-item questionnaire was a reliable measure of the severity of opioid-induced constipation.

Ileus occurred more commonly in 30 mechanically ventilated patients receiving co-sedation with midazolam + fentanyl (two episodes) than in those who received midazolam alone (no episodes) [ ]. However, the combination provided more reliable sedation.

Fentanyl and remifentanil have been compared in patients undergoing plastic surgery, anesthetized with propofol [ ]. There was a higher incidence of postoperative nausea and vomiting in those who received fentanyl. Despite this, patient satisfaction was the same in the two groups.

In a comparison of transdermal fentanyl (n = 299) and modified-release oral morphine (n = 298) in patients with chronic low back pain, transdermal fentanyl was associated with significantly less constipation (31 versus 48%); a smaller proportion of those who received transdermal fentanyl needed to use laxatives [ ]. In both groups, the most common adverse events leading to treatment withdrawal were nausea, vomiting, and constipation.

Urinary tract

There have been two cases of urinary retention leading to renal pelvocalyceal dilatation as a result of continuous infusion of fentanyl (3 micrograms/kg/hour) in premature neonates [ ]. In both cases the problem was resolved by inserting an indwelling catheter.

Skin

A rash has been attributed to fentanyl [ ].

• A 70-year-old man with metastatic cancer of the colon was given transdermal fentanyl 50 mg for analgesia. After 10 days he developed an itchy pustular eruption on the trunk and limbs. The lesions subsided on withdrawal of fentanyl. When he restarted transdermal fentanyl 2 months later, the skin lesions reappeared and became more generalized. The pustules were scattered, sparse, and superficial, and included the tongue and buccal mucosa, but not the conjunctivae or genitalia. A history of eosinophilia suggested an immunoallergic origin.

Like other opioids, fentanyl can cause pruritus. Prophylactic intravenous ondansetron 8 mg with hyperbaric bupivacaine 7–10 mg and fentanyl 25 micrograms significantly reduced the incidence of intrathecal fentanyl-induced pruritus in 125 patients undergoing knee arthroscopy or urological surgery in a randomized, double-blind, placebo-controlled trial [ ]. The incidence of pruritus was 39% with ondansetron and 68% with placebo.

Musculoskeletal

There have been three reports and a prospective study of muscle rigidity after fentanyl administration in neonates [ , ].

• Two neonates had transient (0.5–2 minutes) chest wall rigidity after intravenous boluses of fentanyl 2 and 4 micrograms/kg. They were already compromised, one with a respiratory distress syndrome and one with a diaphragmatic hernia.

• A premature male infant of 28 weeks gestation was given an intravenous bolus of fentanyl 3 micrograms/kg before intubation and this was followed by isolated rigidity of the tongue lasting 20 seconds [ ].

In a prospective case series study of 89 preterm and term infants who received fentanyl out of a total of 404 neonatal intensive care patients in one year, eight neonates (9%) had chest wall rigidity [ ]. The spectrum of neuromuscular activity extends from mild muscle rigidity through abnormal muscle movements (chewing) to tonic-clonic movements. In two cases there was laryngospasm with chest wall rigidity. In all cases low-dose fentanyl (3–6 micrograms/kg) had been given for analgesia or sedation.

Chest wall rigidity, sometimes lasting for more than 24 hours and causing hypoxia, can occur postoperatively; it can be attenuated with naloxone or neuromuscular blockers [ ].

Autacoids

Surgical procedures generally cause perioperative stress, with release of cortisol, cytokines (interleukin-6 and tumor necrosis factor alfa) and acute phase proteins (C-reactive protein and leptin). Opioids are believed to have an inhibitory effect on cortisol release, reducing the neuroendocrine perioperative response. This has been studied in 14 patients undergoing hemorrhoidectomy, who were given either general anesthesia with thiopental 5–7 mg/kg (n = 7) or fentanyl 0.5 micrograms/kg (n = 7) [ ]. There were higher leptin concentrations in the general anesthesia group, but this was not clinically significant. There were no other differences between the two groups.

Death

Four deaths have been attributed to intravenous injection of fentanyl extracted from transdermal patches [ ].

• A 35-year old man, with no history of drug abuse was found by his wife on the floor of his workshop. The police recovered a fentanyl patch, needle, and syringe on the scene and toxicological analysis of the aortic blood established fentanyl poisoning.

• A 38-year old man, with a history of polydrug abuse, undergoing a treatment program using “morphine patches”, was found by his brother dead in bed. The police found evidence of recent intravenous injection. Toxicological analysis established fentanyl poisoning.

• A 42-year old man, with a history of polydrug abuse, was found dead in his house with evidence of having injected substantial amounts of fentanyl from several fentanyl patches, and having taken cocaine and oral diazepam. Fentanyl overdose was listed as the main cause of death due to suicide.

• A 39-year old man, with a history of drug and alcohol abuse, died from fentanyl toxicity from illicitly procured fentanyl patches. Hydrocodone and oxycodone were listed as significant contributing factors in his death.

Drug abuse

Abuse of fentanyl-containing analgesics is increasing. In the USA, reports of fentanyl abuse increased to over 6000% (1506 cases) from 1995 to 2002 [ ].

Oral/transmucosal abuse of transdermal fentanyl patches has been reported and can be fatal. In 7 cases of fentanyl-related deaths after oral administration of fentanyl patches, blood fentanyl concentrations were 7–97 micrograms/l [ ]. Fentanyl alone was considered responsible for two of the deaths and the combination of fentanyl, ethanol, other substances, and medical causes were causative in the others.

Drug withdrawal

In a prospective interventional cohort study in 19 neonates who received fentanyl by continuous infusion for a minimum of 24 hours, those who received fentanyl in a total dose of at least 415 micrograms/kg or as an infusion for more than 8 days were at risk of developing opioid withdrawal symptoms on stopping fentanyl [ ]. Having identified patients who are at risk of developing withdrawal symptoms, one needs to be able to monitor withdrawal signs and symptoms adequately and objectively, reduce the dose of fentanyl gradually, or even use methadone.