In this study, the ranges of fentanyl dose among 128 patients were 5.4–17.3 and 12.4–29.9 μg/kg for the first 24 and 48 h after surgery, respectively, which demonstrates that postoperative fentanyl analgesic dose varied significantly among individuals. A number of reasons, including the characteristics and intensity of external pain stimuli, age, gender, weight and pre-operative psychological state, may affect the perception of pain and the requirement for analgesics. We selected patients undergoing open radical gastrectomy, which is a surgical procedure that causes considerable post-operative pain, to identify differences in fentanyl dose. Surgery was performed on all patients by the same two surgeons. There were no differences in the time of surgery, the length of incision or patient age, gender, pre-operative psychological state, height or weight among the different genotype groups. Since the analgesic effect of fentanyl that was administered intra-operatively may outlast the duration of surgery and thus affect post-operative fentanyl use, particularly in patients who received large doses of fentanyl intra-operatively, we discontinued the infusion of fentanyl once the surgeons had opened the abdominal cavity and controlled the total dose of fentanyl so that it was <12 μg/kg. Following surgery, all patients were administered with PCIA so they were in the same pain state (VAS=10–30). This gave all patients a true baseline pain intensity at the start of the observation period and the same endpoint (VAS=10–30). Doses of fentanyl administered post-operatively were also normalized to body weight. As described above, a number of factors affect post-operative fentanyl dose, so our intent was to be as consistent as possible across patients. When we ruled out interference from these confounding factors, we studied the association between the OPRM1 A118G polymorphism and post-operative dose of fentanyl in Han Chinese patients undergoing radical gastrectomy.
We did not find a clear correlation between the OPRM1 A118G SNP and fentanyl analgesic dose at 24 or 48 h after surgery. Results similar to ours have been reported for labor analgesia and cancer pain (15
). Wong et al
) identified no difference in the median duration of intrathecal use of fentanyl in labor analgesia among patients with the OPRM1 A118G polymorphism. Klepstad et al
) reported that OPRM1 A118G demonstrates no significant association with fentanyl dose in a European genetic association study of 394 cancer pain patients. Our findings were, however, contrary to those of a number of other studies (11
). These studies reported that fentanyl is less effective in subjects with the G allele of the OPRM1 A118G SNP than in those with the A allele and subjects with the G allele required more fentanyl for adequate post-operative pain control than those with the A allele. Kasai (18
) considered that these controversial results may be attributable to the different minor allele frequencies between races and ethnicities in the sample populations. In the above mentioned four studies, however, the samples were all from Asian individuals. The frequencies of the mutant G allele were 45 and 43.8% in two Japanese populations (11
) and 31.3 and 37.1% in a Chinese population in a previous study (10
) and in the present study, respectively. Use of the same population with such high mutation frequencies may have enabled us to conduct a more reliable statistical analysis and make a better comparison. These studies and our study were all related to acute post-operative pain.
There are two possible reasons for the conflicting findings. First, the analgesia method and analgesic used for post-operative pain differed. In the present study, we administered only fentanyl by PCIA with no other rescue analgesic. In contrast, in the Hayashida et al
) study, 138 patients that underwent major, open abdominal surgery received primarily continuous epidural analgesia with fentanyl or morphine, with a rescue analgesic, including buprenorphine, pentazocine and pethidine and non-steroidal anti-inflammatory drugs (NSAIDs). Analgesic requirements in the first 24 h post-operative period were determined as the sum of systemic fentanyl equivalent doses of all opioids and NSAIDs used for analgesia during the first 24 h after surgery. We consider that the conversion may not reflect precise fentanyl doses due to the varying properties and various routes of administration for different analgesics. Secondly, in the Fukuda et al
) study, although the analgesic effects of fentanyl evaluated with the cold pressor prior to surgery were reduced in subjects carrying the G allele compared with subjects not carrying this allele, the A118G SNP demonstrated no significant association with 24 h post-operative fentanyl use, peri-operative fentanyl use, total peri-operative analgesic use or VAS at 3 or 24 h in patients undergoing orofacial cosmetic surgery. Thus, this result was consistent with ours with regard to the association between OPRM1 A118G polymorphism and post-operative dose of fentanyl.
As with clinical studies, not all data from in vitro
studies of the OPRM1 A118G mutation and opioid interaction supported positive results. Bond et al
) identified that the mutant μ
-opioid receptor did not show altered binding affinities for the majority of opioid peptides and alkaloids, including fentanyl, although the variant receptor expressed in AV-12 cells had a three-fold higher binding affinity for β
-endorphin than the wild-type opioid receptor. Two other studies also reported no difference in the binding characteristics of ligands to the mutant μ
-opioid receptor (22
). These results suggest that the A118G polymorphism does not change the overall binding properties of the μ
-opioid receptor. In a study by Mahmoud et al
) in sensory neurons isolated from a humanized mouse model, the potency and efficacy of fentanyl were similar for 118 AA and 118 GG sensory neurons and the fentanyl pharmacological profile for the two groups of neurons was also similar, which was consistent with our results. Unlike the effects of fentanyl, morphine was less potent and efficacious in 118 AA neurons and morphine-mediated analgesia in 118 GG mice was significantly reduced compared with the 118 AA mice, also supporting the previous clinical association studies on morphine and the OPRM1 A118G polymorphism. It is unclear why fentanyl and morphine exert a different pharmacological profile in the mouse model. It may be due to opioid-specific responses, so the results from one drug may not easily extrapolate to other drugs.
Post-operative nausea and vomiting are common side-effects of fentanyl. In the present study, the incidences of post-operative nausea and vomiting were 36.7% (47/128) and 21.1% (27/128), respectively, for all the patients, which were much higher than the previously reported range of 5.6–28.5% (6
). Patients undergoing radical gastrectomy are particularly prone to post-operative nausea and vomiting due to the site of surgery and stimulation of the gastric tube. However, we identified that there was no significant difference in nausea and vomiting among the different genotype groups. Our results were similar to previous reports (6
). A number of factors are known to affect post-operative nausea and vomiting, including age, gender, type of surgery and anesthesia, pre-operative psychological state and the particular drugs used following surgery. In this study, there were no significant differences in the above factors, including opioid dose, among the genotype groups. In contrast, Tan et al
), Klepstad et al
) and Sia et al
) all reported that the AA group had the highest nausea score and incidence compared with the other two groups. However, the above two studies (10
) also showed that the nature of pain was malignant pain and post-cesarean pain. These patients may have different physiological or pathological conditions, including increased hormone levels in pregnant women. Another difference was the use of morphine, which has different pharmacological features to fentanyl, despite belonging to the same class of drug. Fentanyl has inhibitory effects on gastrointestinal motility, which is thought to be the main reason for frequent nausea and vomiting. Walldén et al
) identified a large variation in gastric myoelectrical activity following intravenous fentanyl administration; however, this was not explained by the OPRM1 A118G polymorphism. Therefore, we consider that there may be no correlation between the nausea and vomiting induced by fentanyl and the OPRM1 A118G genotypes.
Post-operative dizziness is often observed during PCIA by fentanyl. In this study, the incidence of dizziness was 33.6% (43/128), similar to the report by Lin et al
) and it mainly occurred in the first 24 h after surgery. Similarly, we did not observe an association between the OPRM1 A118G polymorphism and the incidence of dizziness. However, the adverse effects were assessed as events, not on a scale, which may not reflect the real changes and may have led to differences in the results.
In the present study, we controlled a number of confounding factors, including biological and psychological variation, and analgesia management had the same start- and endpoint with intravenous fentanyl, which reflects the real gene-dose association. However, our sample size may be inadequate to make comments on negative genetic effects on fentanyl dose and side-effects. Additionally, we studied only one polymorphism of the OPRM1 gene and other genetic variations in OPRM1 and other candidate genes may also play a role in the response to opioids. Moreover, the mixed-gender study population may have increased variability in post-operative fentanyl requirement, although there was no statistical difference in gender between the different genotypes.
Multiple, complex biological systems are involved in opioid pharmacology and pain processing. Age, gender, body weight, psychological state, disease conditions, tolerance to side-effects, genetic variation, race and ethnicity cause different opioid requirements for patients. For different drugs, there are various routes of administration, potential interference from drug interactions and different pharmacokinetics and pharmacodynamics. The differences in the nature and intensity of nociceptive stimuli, including labor pain, acute post-operative pain and chronic pain, as well as the modalities used to describe pain perception and analgesic responses, may also contribute to inconsistent findings. In addition, although a number of functional analyses of the A118G SNP have been performed, the results of these studies are also controversial in animal and clinical studies. Therefore, further studies are required to clarify the association between the A118G SNP and treatment efficacy or side-effects of fentanyl with sufficient samples for each effect size in different races and ethnicities.
In conclusion, our study suggests that the OPRM1 A118G polymorphism does not play a significant role in either post-operative fentanyl analgesic dose or the incidence of postoperative side-effects in Han Chinese patients undergoing radical gastrectomy.