The study has demonstrated that in RP patients, age and depression are found to be predictors of postoperative pain. A pain level > VAS 30 mm at one occasion during the three postoperative days was predicted by age, and a pain level > VAS 70 mm was predicted by depression. From a treatment perspective we wanted to predict whether the patient needs treatment in the next future, so that we, in the best case, can give the treatment before the pain has increased above 30 mm on the VAS scale. For that purpose we did not need to make a prediction at baseline or by use of the baseline values. We needed to make a prediction with a time horizon of a few hours only to get the opportunity to treat the patient. By use of logistic regression analysis we found that the only factor that could predict pain was the previous VAS score, except for day two, when we found that patients with EDA reported significantly higher pain scores than the patients with other pain treatments. A surprisingly finding was that as much as one out of four patients (23%) of the total sample and 30% of the EDA patients experienced severe pain.
Patients undergoing the same type of surgery are often given the same type of postoperative pain treatment, but age-related pharmacokinetic and pharmacodynamic factors may influence the variance in analgesic needs [26
]. With a univariate analysis, age was found to be predictor of pain > 30 mm, with younger patients at a higher risk of experiencing pain. Others have also found younger patients to report higher pain scores than older ones [12
], possibly reflecting that young patients with a cancer diagnosis may experience greater distress than older patients because of the effect of serious illness on their life and therefore report higher pain levels [4
]. In patients using postoperative patient-controlled analgesia, age is found to be the best predictor of postoperative morphine requirements [26
]. We found no correlation between age and the opioid consumption in the present study.
Preoperative emotional variables such as anxiety and depression have been found to influence pain experience. In this study we found depression to be of importance for a pain level > VAS 70 mm. In a previous study, we found depression to be a predictor of pain in RP patients, and that preoperative depression also affected pain and depression after discharge from hospital [28
]. Even in multivariate studies, depression has been shown to be a strong predictor of postoperative pain [15
]. Pre-operative state anxiety has repeatedly been shown to correlate with post-operative pain severity [15
]. In this study we only found a tendency that preoperative anxiety would influence the postoperative pain. High expected pain severity has been found to predict severe pain [11
]. This was not confirmed in the present study when more patients expected moderate/severe pain to a higher degree than was actually experienced.
Regarding MHLC, it has previously been shown that patients who are more internal, i.e. who believe that they can influence and are responsible for their own health [13
], have lower pain scores and use less postoperative morphine [14
]. This was however not confirmed in the present study, where we found no significant correlation between any of the different dimensions of the MHLC instrument and the pain intensity. The low predictive power of the MHLC variables might result from a relatively low sensitivity of the general MHLC scales to various problems of post-surgery patients [30
In a multivariate analysis model we found the only predictor of the severity of postoperative pain to be the previous VAS value. Seventy percent of the patients with a pain score > VAS 30 mm at four hours after surgery continued to be in pain. Pre-emtive analgesia is a frequently discussed matter but studies comparing pre-incisional with post-incisional treatment have failed to provide convincing evidence for the value of preemptive analgesia [31
]. Kissin [31
] discusses the definition of preemptive analgesia and defines it as "treatment that prevent establishment of central sensitization caused by incisional and inflammatory injuries; it starts before incision and covers both the period of surgery and the initial postoperative period". This means that effective blockade of noxious stimuli during the initial postoperative period reduces subsequent postoperative pain [32
]. Patients who wake up after surgery with insufficient pain relief should be treated immediately to avoid further pain.
Several studies have reported EDA with local anaesthetics combined with opioids as a safe and effective method [33
]. Some studies though, report a fair amount of different complications to the epidural treatment, not only with hypotension, parestesis and motor blockade, but of technical complications and premature removal [35
] which also may result in insufficient pain relief. In the present study, patients with EDA reported higher pain score day two than patients with ITA or systemic opioids. We have previously reported on EDA as an insufficient method for pain management in RP patients with the average patient not experiencing a pain score that was sufficient until day three [9
]. The findings in our previous study were related to problems/barriers associated with the individual epidural pain treatment regime and an inadequate service response in general to patients with moderate or severe pain. Others have reported insufficient pain relief in patients with postoperative EDA, with one third of the patients suffering from significant pain [37
]. In contrast to these findings, Caumo et al [15
] found EDA to protect against moderate to intense postoperative pain. The method for treatment of postoperative pain at our hospital is today changed to ITA [10
Nurses play an important role in the pain management. They assess and document pain, decide whether to administer analgesics, and they monitor the effect of medication which is prescribed and administered in a variety of ways. In the present study supplemental systemic opioids were supposed to be given on a PRN basis until pain relief was sufficient. There is evidence that nurses are conservative when making decisions about opioid dosing and frequency of administration [38
]. Consistent with our findings, it has previously been reported that patients with mild pain receive significantly lower doses of opioids and even if higher doses of opioids are given to those with severe pain, these doses are not titrated to optimal reduction of pain severity [39
]. Nurses do not always titrate opioids appropriately and do not increase subsequent doses of opioids when the previous dose has been safe but ineffective [40
]. Under-medication of pain is often the result of the nurses' failure to involve patients in pain decisions and also of a lack of trust regarding the patients' reports concerning the quality of their pain [41
]. Nurses have to be aware of the fact that in general, younger patients need more opioids than older patients. For their perception of patients' pain, doctors generally rely on nurses to report pain in their patients. If nurses underestimate and/or do not document pain, this is likely to result in under-medication in many patients [42