The Effects of Three Different Analgesia Techniques on Long-Term Postthoracotomy Pain

Abstract

In this clinical, randomized, prospective study, we compared the effects of three different analgesia techniques (thoracic epidural analgesia [TEA] with and without preoperative initiation and IV patient-controlled analgesia [IV-PCA]) on postthoracotomy pain in 69 patients. In two groups, a thoracic epidural catheter was inserted preoperatively. Group Pre-TEA had bupivacaine and morphine solution preoperatively and intraoperatively. Postoperative analgesia was maintained with epidural PCA with a similar solution. Group Post-TEA, with no intraoperative medication, had the same postoperative analgesia as Group Pre-TEA plus the bolus dose. Group IV-PCA received only IV-PCA with morphine for postoperative analgesia. Pain was evaluated every 4 h during the first 48 h at rest, cough, and movement. Pre-TEA was associated with decreased pain compared with the other groups. Six months later, the patients were asked about their pain. The incidence and the intensity of pain were most frequent in Group IV-PCA (78%) and were the least in Group Pre-TEA (45%) (Group Pre-TEA versus Group IV-PCA, P = 0.0233; Group Pre-TEA versus Group IV-PCA, P = 0.014). Patients having pain on the second postoperative day had 83% chronic pain. TEA with preoperative initiation is a preferable method in preventing acute and long-term thoracotomy pain.

Chronic postthoracotomy pain syndrome (pain that recurs or persists along the thoracotomy scar more than 2 mo after surgery) ⁽¹⁾ is reported to have an incidence between 50%^(2,3) and 75%–80%⁽⁴⁾. However, there are only a few studies examining the effects of different medications and techniques used perioperatively ⁽⁵⁾. In this prospective, randomized study, the effects of three different analgesia techniques (thoracic epidural analgesia [TEA] with [pre-TEA] and without [post-TEA] preoperative initiation and IV patient-controlled analgesia [IV-PCA]) have been compared, and the relationship between the immediate postoperative pain and chronic postthoracotomy pain was evaluated.

Methods

The study was approved by the Ethics Committee of the faculty. One-hundred-twelve consenting ASA physical status II–III patients undergoing thoracotomy were randomly divided into three groups to evaluate the effects of three different analgesia techniques. Exclusion criteria during the preoperative period were general contraindications for epidural anesthesia; a history of chronic pain; psychiatric disease; documented cardiac or vascular disease; history of myocardial infarction; arrhythmias; preoperative respiratory function tests showing a forced vital capacity <60% predicted, forced expiratory volume 1 s <60%, or both; renal insufficiency (creatinine >15 mg/dL); or liver dysfunction (aspartate aminotransferase, alanine aminotransferase, or both >40 U/L). A decision of inoperability was the intraoperative exclusion criterion. During the postoperative period, patients who were reoperated, who reported pain related to recurrence, metastases, or infections, or who died were excluded. Twenty-seven patients were excluded according the preoperative exclusion criteria.

The preanesthesia procedure was the same in all patients: the patients were not premedicated before arrival in the operating room. After arrival, they were prehydrated (crystalloid 8 mL · kg⁻¹ · h⁻¹), monitored (electrocardiography, noninvasive blood pressure, and Spo₂) (Horizon XL; Mennen Medicals, Rehovot, Israel), and premedicated (midazolam 2 mg IV).

In the first two groups, an 18-gauge epidural catheter (B. Braun, Melsungen, Germany) was inserted through the T7-8 intervertebral space preoperatively by a midline approach with the loss-of-resistance technique and placed under fluoroscopic control 3–4 cm in the cephalad direction. The placement of the catheter was verified both by administering contrast (iohexol, Omnipaque 300, Opakim, Istanbul, Turkey) under fluoroscopy and by 2 mL of 2% lidocaine.

In Group Pre-TEA (n = 28), a 10-mL bolus of a solution of bupivacaine 0.1% plus 0.1 mg/mL morphine in saline was administered at least 30 min before anesthetic induction, followed by a 7 mL/h infusion of the same solution during the operation. Postoperative analgesia was obtained with epidural PCA (Abbott Pain Management Provider; Abbott Laboratories, Istanbul, Turkey) with a concentration of bupivacaine of 0.1% plus 0.05 mg/mL morphine according to the following program: no initial dose, basal infusion rate 5 mL/h, bolus dose 3 mL, and a 30-min lock-out time.

In Group Post-TEA (n = 29), no epidural medication was applied until the postoperative period. Postoperative analgesic treatment was similar to Group Pre-TEA except that the initial dose of 10 mL of the same solution was administered after extubation.

In Group IV-PCA (n = 28), patients received IV-PCA with morphine with a 5-mg initial dose, no basal infusion, and a 2-mg bolus dose with a 15-min lock-out time.

Patients were not blinded to group.

Before the operation, postoperative pain methodology during the first 48 h was explained to all patients. Additionally, before the hospital discharge they were also informed at the interview about long-term pain after 6 mo.

Except for the smaller concentration of isoflurane in Group Pre-TEA, the anesthesia procedure for the operation was similar in all groups: electrocardiogram, invasive blood pressure, and central venous pressure were monitored. Fentanyl 2 μg/kg, propofol 2.5 mg/kg, and vecuronium 0.08 mg/kg were used for anesthesia induction. One-lung ventilation with a double-lumen tube was applied. For the maintenance of anesthesia, isoflurane (0.3%–0.7% in Group pre-TEA and 0.9%–1.5% in the other groups) was used with incremental doses of vecuronium and fentanyl. In Group Pre-TEA, no further administration of fentanyl was needed after the induction of anesthesia; in the other two groups, 5–10 μg/kg boluses of fentanyl were given intermittently until 1 h before the end of surgery. Patients’ tracheas were extubated immediately after the operation.

During the first 48 h after the operation, patients used the IV-PCA or epidural PCA as described in the protocol, and they were questioned about their pain at rest, cough, and movement by using a numeric rating scale (NRS) ⁽⁶⁾, where 0 = no pain and 10 = maximum pain experienced. Pain was evaluated at 4-h intervals by investigators who were blinded to the analgesic method used. The acute postoperative pain data of the patients and the side effects and unwanted effects were recorded.

Six months after surgery, an investigator who was blinded to the acute pain treatment and to the results of the acute pain treatment interviewed patients by telephone, using a standard questionnaire. In the first two questions, patients were asked whether they were having pain at the moment (sixth month) or had it at least 2 mo after surgery. Then they were asked to evaluate their pain with the same NRS as in the acute period. Finally, they were asked whether the pain (if any) was affecting their daily life. Afterward, the results were also compared with the NRS of the acute postoperative period.

Statistical analysis was performed with appropriate χ² tests (with or without Yates’ correction and Fisher’s exact test) to compare the incidences and with the Kruskal-Wallis test to compare the nonparametric data, using Dunn’s test in cases in which there was significance with the Kruskal-Wallis test. Demographic data and the intraoperative fentanyl administration were compared with one-way analysis of variance. P < 0.05 was regarded as significant.

Results

Six of 85 patients were excluded because of the intraoperative exclusion criteria. At the end of 6 mo, 10 more patients were excluded from the study because of postoperative exclusion criteria. In the remaining 69 patients (22 in Group Pre-TEA, 24 in Group Post-TEA, and 23 in Group IV-PCA), there was no statistical difference among the groups in terms of demographic variables and operative data (Table 1).

Table 1: Demographic Data and Durations and Types of Operations

During the operation, patients in Group Pre-TEA received no fentanyl except 2 μg/kg for the induction of anesthesia. This dose was significantly smaller than in Group Post-TEA and Group IV-PCA (148 ± 19 μg versus 310 ± 103 μg and 270 ± 119 μg, respectively, P < 0.001; there was no difference between the last two groups).

Acute pain data at rest, cough, and movement are described in Table 2. During movement and cough, Group Pre-TEA had significantly less pain compared with the other two groups during the entire period. At rest, patients in Group Pre-TEA reported having significantly lower pain scores during the first 12 h compared with those in Group Post-TEA and during the first 48 h compared with those in Group IV-PCA.

Table 2: Pain Levels During the First 48 Hours After the Operation at Rest, Coughing, and Movement

There were statistically significant differences between Group Post-TEA and Group IV-PCA during rest from 8 h after surgery until the end of 48 h, but no difference during cough or movement was recorded.

Regarding the side effects and unwanted effects of the acute pain treatment, there was no statistical difference between the groups. In no patient in the TEA groups was any motor blockade observed because of the technique and the drug used. There was no technical complication of TEA either in Group Pre-TEA or Group Post-TEA. Two (9%) patients in Group Pre-TEA, two (8%) in Group Post-TEA, and four (17%) in Group IV-PCA reported pruritus. In one patient each in Group Pre-TEA and Group Post-TEA and in three patients in Group IV-PCA, postoperative vomiting was observed (5%, 4%, and 13%, respectively).

The results of the interview concerning the long-term data are shown in Table 3.

Table 3: Results of the Questionnaire at the Sixth Month

Generally, 62% (43 of 69) of the patients reported having pain after 6 mo. According to the χ² test, the incidence of pain in Group IV-PCA was significantly more frequent than in Group Pre-TEA (18 of 23 vs 10 of 22; χ² = 5.148;P = 0.0233), but the difference between Group Post-TEA (15 of 24) versus Group Pre-TEA and Group IV-PCA was not significant.

The values of pain lasting more than 2 mo after surgery were also statistically significant between Group Pre-TEA and Group IV-PCA (11 of 22 vs 20 of 23;P = 0.0106). Group Post-TEA (16 of 24) was not significantly different compared with the other two groups.

The intensity of pain was significantly higher in Group IV-PCA (NRS, 1.4 ± 1.2; maximum, 4) than in Group Pre-TEA (NRS, 0.6 ± 0.8; maximum, 3), without a significance versus Group Post-TEA (NRS, 0.9 ± 0.9; maximum, 3) (between Group Pre-TEA and Group IV-PCA, u = 15;P = 0.014). No patients reported that their lives were affected by pain.

Of the 29 patients who had an NRS ≥2 on the second postoperative day during coughing or movement, 24 (83%) reported having chronic pain, whereas only 19 (48%) of 40 patients had pain at the sixth month, despite their reporting pain of NRS <2 after the acute pain management at the 48th hour (χ² = 0.746, P = 0.0063). In other words, only five patients had acute, but not chronic, pain.

At the end of the 48-h period, 6, 11, and 12 patients had pain during cough or movement in Groups Pre-TEA, Post-TEA, and IV-PCA, respectively. Of these patients, 5, 9, and 10, respectively, reported chronic pain.

Discussion

A frequent incidence of chronic postthoracotomy pain has been reported in previous studies ^(2–4). The results of our study (62% after 6 months and 68% for more than 2 months) are consistent with those of others. However, there are few studies examining this syndrome, and most report the incidence and prediction of chronic postthoracotomy pain without any comparisons of different methods. There is still no standard method to treat or prevent chronic postthoracotomy pain effectively. This study shows that pre-TEA is associated with a decreased incidence and intensity of pain compared with post-TEA or IV analgesia during the acute and the following 6-month period; therefore, it is a method of choice to prevent chronic postthoracotomy pain.

The clinical effectiveness of preemptive analgesia is controversial ⁽⁷⁾. The results of acute pain levels in our study have shown beneficial effects of preoperative initiation of TEA. Pre-TEA was associated with significantly less pain than post-TEA and IV-PCA during rest, coughing, and movement, whereas the difference between post-TEA and IV-PCA could be observed only during rest.

In a previous study comparing the effects of preoperative and postoperative initiation of TEA, we showed that preoperative initiation of TEA is more effective in controlling acute pain after thoracotomy compared with postoperative initiation ⁽⁸⁾. The results showed that preemptive analgesia was associated with a better control of acute postoperative pain, with less need of supplemental IV and epidural medications and with no difference in undesired side effects. The results of both pre-TEA and post-TEA of acute pain are comparable to the acute data in this study.

It may be suggested that the time interval between the epidural medication and surgical incision (at least 30 minutes in our series) can play an important role in successful preemptive analgesia. Obata et al. ⁽⁹⁾ have shown that an epidural block with mepivacaine before surgery reduces long-term postthoracotomy pain. That study was also associated with a frequent incidence of chronic postthoracotomy pain; it compared the effects of pre- and postoperative initiation of TEA and found a significant clinical efficacy of preemptive analgesia. There are two major differences between that study and ours. First, Obata et al. ⁽⁹⁾ used only a local anesthetic drug (mepivacaine) for epidural block (without opioid supplementation); second, the effects of IV analgesia were not investigated and compared.

Katz et al. ⁽¹⁰⁾ have shown that acute pain after thoracic surgery predicts long-term postthoracotomy pain. This is a pioneer study. However, Katz et al. did not compare different methods, and patients served as their own controls, showing a relationship between acute and chronic pain; moreover, the number of patients was relatively small. Therefore, subjective differences of pain thresholds may have affected the results. In our study, a relationship between the acute and chronic pain states was also shown, similar to the study of Katz et al. Furthermore, the possible effects of individual differences were also minimized by using and comparing three different methods.

Chronic postthoracotomy pain consists of different types of pain. Wallace and Wallace ⁽¹¹⁾ reported in their review of both myofascial and neuropathic characteristics of chronic postthoracotomy pain. Several studies in which patients were also asked about the characteristics and location of their pain have shown the dominance of neuropathic characteristics ^(2,4,11). We have evaluated the incidence of chronic postthoracotomy pain and its correlation with the acute postoperative analgesia technique. Consequently, patients in our study were not asked about the characteristics and location of their pain because it was not the aim of the study to investigate the type of chronic postthoracotomy pain.

In addition, in the six-month interview, patients were not asked whether they had pain at rest, cough, or movement. Obviously, a separate assessment similar to the one in the acute postoperative period could have resulted in a detailed explanation, but we thought this would complicate the telephone interview without adding much to the result. Therefore, the pain scores at six months are to be considered as average (or approximate) scores.

However, the type of chronic postthoracotomy pain and aggravating factors (such as cough, movement) need to be investigated in further studies.

Despite the generally frequent incidence (62%) of long-term pain, the intensity of pain seems to be low. Actually, there was no report of pain affecting daily life (even in Group IV-PCA), and no patient needed analgesic medication. Except for some episodic increases in pain levels, which had not been asked about in the interview, none of the patients had reported pain of >4 on a 10-point scale at the sixth month. However, the generally frequent incidence must be investigated. It seems that the difference in incidence of pain is of more clinical importance than the difference of its intensity (45% vs 78% in Group Pre-TEA and Group IV-PCA, respectively). Even in the group with best results (Group Pre-TEA), almost half of the patients reported a “continuous mild unpleasant sensation” or more, when asked. It must be kept in mind that even a “mild unpleasant sensation” (which would be equal to 1 point of 10) would be clinically important if it persists over a long period or disables the patient either physically or psychologically.

It can be concluded that thoracotomies are associated with a frequent incidence of chronic postoperative pain, which needs investigation. Pre-TEA is an appropriate application for thoracotomy pain because it is more effective in preventing chronic pain. In addition, acute pain seems to be a good predictor of chronic pain; therefore, prevention of acute pain could also be effective in prevention of chronic postthoracotomy pain.