- Top of page
- What is Already Known about This Subject
- What This Study Adds
- Conflict of Interest
Postoperative pain is a frequent problem in cardiac surgery . It is characterized by a large interindividual variability, even when the patient is at rest . In contrast, the numerous unavoidable painful healthcare procedures in the intensive care unit (ICU) produce pain levels that may differ greatly between patients. Turning of the patient and drain removal have been identified as the most painful procedures [3, 4]. This interpatient variability in pain sensitivity may partly be explained by environmental factors, such as age, sex or anxiety [5-7]. Furthermore, some candidate genes have been associated with differential pain sensitivity [2, 8].
The cathechol-O-methyltransferase (COMT) enzyme may contribute to the variability in pain sensitivity because it has a role in pain processing. Cathechol-O-methyltransferase metabolizes dopamine, adrenaline and noradrenaline, and is a key modulator of dopaminergic and adrenergic neurotransmission . There are two major forms of the COMT enzyme that differ by 50 amino acids at the N-terminus, namely membrane bound (MB-COMT) and soluble (S-COMT) . One single nucleotide polymorphism (SNP; rs2097903) is located at position 21217 in the estrogen-sensitive portion of the MB-COMT promoter region [11, 12]. The second SNP (rs6269) is located in the promoter region of S-COMT [12, 13]. Three SNPs (rs4633, rs4818 and rs4680) are situated within the coding region for both S- and MB-COMT (National Center for Biotechnology Information genome database; www.ncbi.nlm.nih.gov/gene). The SNP rs4680 is nonsynonymous and codes for a substitution of valine (Val) to methionine (Met) at codon 158 (Val158Met) . This substitution has been associated with a three- to fourfold decrease in COMT activity . Zubieta et al.  have suggested that lower COMT activity leads to an enhanced activation of dopaminergic neurotransmission, with lower endogenous levels of enkephalins and thus exaggerated pain sensitivity as a result . More specifically, in an experimental study these authors showed that healthy volunteers with the Met/Met genotype reported higher pain ratings than did those with the Val/Val genotype. In addition, subjects with the Met/Met genotype showed weaker activation of the endogenous opioid system on experimental pain stimuli than did subjects with the Val/Val genotype. These results were confirmed in two other experimental studies with healthy volunteers [16, 17] and in one study evaluating morphine consumption in patients with chronic cancer pain . There are, however, no reports about the relevance of this polymorphism for acute postoperative pain, such as pain after cardiac surgery, experienced either at rest or upon an unavoidable painful healthcare procedure, such as drain removal or turning of the patient, when patients are treated with intravenous morphine infusions.
In a recent clinical trial by our group in postoperative cardiac patients , pain levels were studied around an unavoidable routine painful procedure, i.e. turning of the patients and/or drain removal. Despite continuous morphine infusions and a bolus dose of morphine before the procedure, 25% of the patients experienced unacceptable pain, rated on the numerical rating scale (NRS ≥ 4, range 0–10 ) . In the present study, we tested the hypothesis that the COMT Val158Met polymorphism may explain these high pain levels upon the unavoidable painful procedure in these postoperative cardiac patients. Therefore, the aim of this study was to investigate the influence of the COMT Val158Met polymorphism on pain sensitivity during an unavoidable painful routine healthcare procedure in morphine-treated adult patients after cardiac surgery in the ICU.
- Top of page
- What is Already Known about This Subject
- What This Study Adds
- Conflict of Interest
We evaluated the influence of the COMT Val158Met polymorphism in postoperative cardiac patients treated with continuous morphine infusions to manage pain at rest and with an additional bolus dose of morphine before an unavoidable postoperative painful procedure (drain removal and/or turning of the patient) in the ICU. The main finding was that patients carrying the Met-variant allele experienced both significantly increased overall pain and significantly increased pain during the painful procedure, in comparison to patients with the Val/Val genotype. More specifically, patients with the Met/Met genotype showed a statistically significant and clinically relevant increase in pain scores during the painful procedure, unlike patients with the Val/Val genotype. To our knowledge, this is the first study to evaluate the impact of the COMT Val158Met polymorphism on pain sensitivity within a clinical research design in morphine-treated postoperative ICU patients undergoing an unavoidable painful stimulus.
So far, studies of healthy volunteers have evaluated this polymorphism in relation to pain sensitivity by applying experimental pain stimuli, with similar results to those in the present study (Table 4). Loggia et al.  demonstrated that individuals with the Met/Met genotype exhibit stronger pain signals in numerous cortical and subcortical structures after repeated noxious stimulation when compared with other genotype groups. This is also consistent with other studies evaluating the impact of the COMT Val158Met polymorphism on pain sensitivity. Zubieta et al.  demonstrated that Met/Met subjects were characterized by higher sensory and affective pain ratings; two other studies [16, 17] found that individuals with the Met/Met genotype were more susceptible to pain after repeated thermal stimuli. Jensen et al.  showed that, after remifentanil bolus injection and repeated heat pain stimuli, Met/Met subjects reported higher pain scores than did Val/Val subjects, while heterozygous subjects reported intermediate scores.
Table 4. Overview of studies that investigated the COMT Val158Met polymorphism in relation to pain sensitivity
|Study design||Study population||Intervention||Main outcome||Reference|
|Experimental||29 healthy volunteers||Binding potential of the μ receptor and muscle pain was measured twice: during intensity-controlled sustained pain induced by infusion of 5% hypertonic saline into the masseter muscle and during the infusion of nonpainful 0.9% isotonic saline.||Subjects with the Met/Met genotype of the COMT Val158Met polymorphism showed diminished regional μ-opioid system responses to pain compared with Met/Val and Val/Val subjects. These effects were accompanied by higher sensory and affective ratings of pain.|||
|Experimental||202 female healthy volunteers||Threshold and tolerance to thermal, ischaemic and mechanical stimuli, as well as temporal summation to heat pain, were determined.||The Val158Met polymorphism was associated with sensitivity to painful heat stimuli, which suggests that the Val158Met polymorphism plays a primary role in variation in temporal summation of pain.|||
|Experimental||43 healthy volunteers||Five applications of thermal heat pain were made to the hand. After each stimulus, subjects rated pain on a VAS. Before the second and the fourth stimulus, respectively, an intravenous injection of remifentanil (0.08 mg kg−1) and placebo was administered.||Met/Met subjects reported significantly more pain compared with Val/Val subjects in the case of repeated pain stimuli, although not during an initial response of the descending pain system. The opioid intervention induced analgesia without a separating effect for genotype.|||
|Experimental||54 healthy volunteers||Subjects received two heat pain stimuli on the right forearm during functional magnetic resonance imaging. After each stimulus, subjects rated their pain intensity using the Gracely Sensory Scale.||Met/Met subjects showed stronger pain-related functional magnetic resonance imaging signals than Val/Val subjects in several brain structures, only for high-intensity pain stimuli after repeated administration.|||
|Experimental||500 healthy volunteers||Subjects received a painful thermal and cold stimulus, measured on the hand up to the wrist, with VAS ratings and temperature dimensions of personality. In total, three single nucleotide polymorphisms were evaluated.||In a classification and regression tree analysis, the Val158Met polymorphism did not provide a strong predictive value for heat or cold pain sensitivity.|||
|Clinical research design||207 patients treated with morphine for chronic cancer pain||Between the genotype groups, morphine doses, serum concentrations of morphine and morphine metabolites were compared.||After a mean treatment period of 3.5 months, patients with the Val/Val genotype needed more morphine when compared with the Val/Met and the Met/Met genotype groups. Pain scores were not reported.|||
|Clinical research design||221 acute postsurgical patients||Clinically induced pain was recorded using the VAS after oral surgery every 20 min until subjects requested analgesic medication (ketorolac), followed by VAS scores at 15 min intervals for 180 min.||Although the COMT enzyme showed significant associations with the maximal postoperative pain, this association was not sustained after correcting for multiple comparisons.|||
|Clinical research design||2294 cancer patients||Among other things, pain intensity, time on opiods and 112 single nucleotide polymorphisms that may predict opioid dose were included as covariates in a regression model.||None of the 112 single nucleotide polymorphisms showed significant associations with opioid dose.|||
In contrast, there are reports in which it was not possible to identify a significant association between COMT Val158Met polymorphism and pain sensitivity, when studying patients with postoperative pain or cancer pain, or healthy subjects with experimental pain . Kim et al.  concluded that in 221 acute postsurgical patients, genetic polymorphisms, including those of the COMT enzyme, show only a weak association with clinically induced acute injury. In a large study by Klepstad et al. , the evaluation of 112 SNPs in 23 candidate genes proposed to influence opioid efficacy did not predict the need for opioids in cancer patients with pain. Finally, a study in which experimental pain was observed in healthy volunteers did not find associations between the genotypes of the COMT enzyme and pain sensitivity, measuring painful thermal and cold stimuli .
Thus, although there is some inconsistent literature, the results of our clinical study confirm the results of the experimental pain studies in healthy volunteers, where the pain sensitivity of patients carrying the Met-allele was higher than that of patients with the Val/Val genotype. We showed that this genotype-related difference in pain sensitivity is clinically relevant in morphine-treated patients after cardiac surgery undergoing an unavoidable (instead of experimental) pain stimulus in the ICU.
Although we found a clinically relevant increase in NRS score in patients with the Met/Met genotype, the proportion of patients whose pain increased from an acceptable level (NRS < 4) at baseline to an unacceptable level (NRS ≥ 4) during the painful procedure did not differ between the genotype groups. We realize that in our study the mean NRS scores generally were low (NRS < 4), and that in particular the increase to unacceptable pain during the painful procedure should be prevented. However, the knowledge that patients with the Met/Met genotype are at higher risk for a clinically relevant increase in pain during a painful intervention may be of value for procedural-related pain management.
The results of our study suggest that patients carrying the Met-allele may benefit from a more individualized therapy in the event of a second surgery. Higher morphine doses may be anticipated, although only a small increase in efficacy may be expected at an increased incidence of adverse events. In this respect, we hypothesized that non-opioid analgesics may be preferable, because they probably act independently from endogenous enkephalin levels and the μ receptor density. However, agents such as nonsteroidal anti-inflammatory drugs may be disadvantageous in postoperative cardiac patients because they carry the risk of cardiovascular side-effects. Thus, further research should focus on these and other aspects of individualized pain management.
In our study, patients received continuous morphine infusions for pain at rest, with a bolus dose of morphine before a painful procedure. The morphine requirement did not differ between the genotype groups during the ICU stay and during the painful procedure. Interestingly, at first sight our results deviate from the observation made by Rakvåg et al.  that Met/Met homozygous cancer patients required less morphine compared with patients who had the Val/Val variant. The authors suggest that this may be explained by an increase in μ-opioid receptor density in Met/Met patients, which causes morphine to be more effective. However, those patients were treated for prolonged cancer pain for approximately 3.5 months , in contrast to our patients, who were treated for postoperative acute pain (maximum of 48 h). It is possible that an increase in μ-opioid receptor density may occur after a more prolonged period of opioid treatment. Moreover, the study of Rakvåg et al.  concerned a heterogeneous group of patients, with differences in disease severity and progression, and thus several nociceptive stimuli, whereas in our study all patients underwent the same type of painful procedure.
Considering the bolus morphine dose (2.5 vs. 7.5 mg), we observed no significant effect of the genotype on the analgesic response to morphine. These results are supported by the study of Jensen et al. (Table 4, ), in which a single dose of remifentanil before a heat stimulus induced analgesia, without a separating effect for genotype. They also found that Met/Met subjects reported significantly more pain than did Val/Val subjects in the case of repeated pain stimuli, except after the initial pain stimulus. Given that our patients underwent cardiac surgery the day before the painful procedure, it is likely that the pain system was already triggered at the time of the painful procedure. Therefore, we suggest that Met/Met patients receiving the same dose as Val/Val patients are more sensitive to pain as a result of repeated painful stimuli, which is independent of a morphine bolus dose.
Two limitations of our pilot study should be addressed. Firstly, our study was in fact underpowered, because this analysis was conducted within a randomized controlled trial and was thus not designed primarily as a pharmacogenetic study. However, even with these low numbers of patients, we found a significant effect of genotype on both overall pain levels and pain during the painful procedure. Secondly, pain sensitivity is influenced by many factors, probably including genetic contributions of more than one gene. Therefore, this study cannot determine the relative importance of the Val158Met polymorphism in comparison to other genes involved in pain.
In conclusion, the results of the present clinical study suggest that the COMT Val158Met polymorphism is correlated to pain sensitivity in morphine-treated patients undergoing a painful healthcare procedure after cardiac surgery, showing that Met-allele carriers are more sensitive to overall pain and procedural pain.