| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;79:1845-1850
© 2005 The Society of Thoracic Surgeons
a Division of Thoracic and Foregut Surgery, Cancer Institute, Biostatistics Facility, Pittsburgh, Pennsylvania
b Department of Anesthesiology, University of Pittsburgh, Pittsburgh, Pennsylvania
c Department of Cardiothoracic Surgery, Boston Medical Center, Boston, Massachusetts
Accepted for publication October 28, 2004.
* Address reprint requests to Dr Luketich, UPMC Presbyterian, Suite C-800, 200 Lothrop Street, Pittsburgh, PA15213 (E-mail: luketichjd{at}upmc.edu).
Presented at the Fortieth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 26–28, 2004.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
METHODS: The study included 124 randomized patients; 91 had sufficient data for analysis (44 EPI, 47 ICN-PCA). The primary endpoint was pain measurement using a composite of a visual analogue scale, numerical rating, and categorical rating. A second endpoint was the success rate of each method. Pulmonary function tests, antibiotics, intensive care unit (ICU), and hospital days, and use of nonprotocol pain medications were also compared.
RESULTS: There were 12 pain observations per patient (90% completed on days 1 to 5). The pain composite revealed an average postoperative pain score of 2.4 on a scale from 0 (no pain) to 10 (worst pain). There was no difference between the groups. Failures of the planned method of analgesia included 9 in the EPI group and 4 in the ICN group (p = 0.23). Another 20 patients were excluded (no difference between groups) due to unsuspected mediastinal metastases precluding thoracotomy (n = 13), and other miscellaneous factors precluding follow-up (n = 7). The EPI group had an increased number of urinary catheter days (2.5 days vs 1.7, p = 0.002) and increased narcotic supplements (p = 0.03) compared with ICN. Mean ICU days (0.9) and hospital days (6.2) were similar for both groups, and there were no differences in arrhythmias, pneumonias, transfusions, and antibiotic use. Significant differences were seen (p = 0.001) between preoperative and postoperative pulmonary function tests in both groups. However, there were no differences in pulmonary function when the groups were compared with each other.
CONCLUSIONS: Satisfactory pain control was achieved after thoracotomy using either EPI or ICN-PCA. The ICN-PCA achieved equivalent pain control compared with EPI, and was placed by the surgeon with no delays in surgery, and demonstrated a decreased requirement for Foley catheter duration.
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
In many centers epidural anesthesia has emerged as the gold standard for pain control. However, this method is not suitable for all patients and may be associated with potential risks such as dural perforation, bleeding, infection, hypotension, and urinary retention [2]. There are also other potential problematic issues with epidural pain control, such as delaying the start of an operative procedure, technical failures, and the costs of postoperative pain management by a separate pain team. Another method of pain control, which has gained popularity in some centers, is the use of intercostal nerve blockade [3]. As with epidural anesthesia, this method allows local administration of drugs to the pain causing anatomic region, but potentially with lower risks and discomfort to the patient. There may also be fewer delays in surgery and the technical failure rate should be lower since it is placed under direct vision by the surgeon.
Our group previously performed a case-controlled retrospective study comparing 20 patients who underwent epidural anesthesia (EPI) with 20 patients who underwent intercostal nerve catheter with supplemental patient-controlled analgesia (ICN-PCA) [4]. No significant differences were seen in pain control, although Foley catheter days and hospital charges were greater in the EPI group. The higher hospital charges were related primarily to pain-service costs. This prospective randomized study was performed as a follow-up to our earlier study. The goals of this study were to compare the efficacy of pain control and also postoperative morbidity with each technique.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Analgesia Techniques
Patients were randomized at least one day before operation. Those patients randomized to the EPI received the thoracic epidural before the induction of general anesthesia. This was placed at the T3-T6 level. After administering a test dose of 3 mL of 1.5% lidocaine with 1:200,000 epinephrine, the production of a sensory band of analgesia to pin-prick confirmed successful EPI placement. A mixture of 0.125% bupivacaine and morphine 0.05 mg/mL were administered through the epidural catheter at a rate of 4 to 8 mL/h. The amounts given and changes made were recorded throughout their hospital stay.
Patients randomized to receive ICN-PCA received a 10 mL bolus of 0.25% bupivacaine injected by percutaneous nerve block before thoracotomy from the third to the eighth intercostal space. This was performed to achieve a block at least two interspaces above and below the thoracotomy incision (which was performed through the fifth intercostal space. This nerve block was administered by the anesthesiologist and was considered preemptive anesthesia, corresponding to the placement of the epidural and test dose given in the EPI patients. At the conclusion of the lung resection, an intercostal nerve catheter was placed through a small 3 mm-stab incision posteriorly at the eighth intercostal space and then tunneled up vertically, below the parietal pleura to the third intercostal space to allow delivery of the local anesthetic to at least two intercostal spaces above and below the incision. The tunnel is created by passing a Stern clamp (Scanlon, St Paul, MN) through the stab incision and into the extrapleural space. The clamp is then tunneled posterior to the thoracotomy incision to the third intercostal space where this breaks through the pleura. The proximal end of the catheter is then grasped and pulled through the extrapleural tunnel and out through the skin. It is important that the parietal pleura is intact in the area of the extrapleural tunnel created for the intercostal nerve catheter, so care is taken to avoid extending the posterior extension of the rent in the parietal pleura when entering the chest. The catheter used was a 5F, 40-cm-length radio-opaque polyurethane catheter with multiple side holes to assure uniform delivery along the nerve roots. After suturing the catheter, 10 mL of 0.5% bupivacaine was routinely injected in the operating room and then on arrival to the postanesthesia care unit, an infusion of 0.25% bupivacaine at 1 mL/10 kg per hour (eg, 7 mL per hour for a 70 kg patient) was started and continued for a minimum of 72 hours. We selected bupivacaine for the continuous ICN infusion as this has been successfully reported previously [3]. Additionally, we wanted to use the same drug that is routinely used in our epidurals to make the patient groups as comparable as possible. If the chest tubes were still in place the infusion continued up to a maximum of 96 hours after operation. A PCA was also connected immediately after surgery in the recovery room. The PCA was set to deliver 1 mg of morphine on demand up to every 8 minutes with a 4-hour maximum of 30 mg. For patients more than 80 kg, the PCA delivered a 2 mg morphine bolus on demand (40 mg, 4 hour limit). The PCA was discontinued within 4 to 6 hours of the intercostal infusion and oral pain medications were initiated. The anesthesia pain service was not involved in the postoperative management of the ICN-PCA patients, whereas the pain service closely followed all patients in the EPI group.
Antibiotic prophylaxis consisted of a single dose of intravenous (IV) antibiotics before the skin incision was made. Antibiotics were not used to cover the chest tubes or extrapleural catheter. Additional antibiotic use was recorded as a secondary endpoint.
Outcome Measurements
The primary endpoint measured in this study was pain. Three self-reported instruments were used. These were (1) a visual analogue scale, (2) a box-score, and (3) a categorical scale. In order to create a composite score the categorical code was first converted to the average of the box score among people who completed both. The three scores were then averaged to produce a composite score (from 0-no pain to 10-severe pain). If the patient did not complete all three metrics, those that were completed were averaged. The three pain scores represent three ways of assessing the same construct. Averaging the three scores was believed to provide several advantages. First, the average of the pain scores is more reliable and less subject to random measurement errors than any simple measure. Second, the average of the three scores has greater precision than any simple measure (than, for example, the categorical item which is a Likert scale with 6 response levels) and can discriminate between fine degrees of pain. Third, some questions had missing values on one or two of the items, in which case we constructed the composite score from the available items, allowing more questionnaires to be included in the analysis. Pain scores were measured preoperatively and daily after surgery to postoperative day 5. The surveys were administered by a clinical research nurse while in the hospital and handed to the patients after discharge. We did not control for the time of day for the administration of the surveys.
Secondary endpoints included the success rate (completion) of the analgesia delivery method, antibiotic use, intensive care unit (ICU) and hospital days, Foley catheter days, and the use of nonnarcotic and narcotic pain medications. Pulmonary function tests were also obtained preoperatively and also in the early postoperative period (days 2 to 6).
Statistical Methods
Repeated measures analysis (normal linear mixed effects modeling) was used to determine whether the groups had different patterns of pain control through time (group x time interaction) and if there was no interaction, whether there was a group effect on pain control. The baseline (presurgery) average pain scores, the time from surgery, and an interaction between time from surgery and treatment group were included as covariates. A quadratic function of time from surgery was also included. Serial correlation was modeled. Primary analyses were performed on an intent-to-treat basis (as randomized). Secondary analyses performed with treatments as given showed no differences in the results. The pulmonary function measures were also compared across treatment arm with normal mixed effects linear modeling, with a linear predictor for the number of days after surgery [1–6] and a factor for the treatment arm. Other measures of perioperative morbidity were compared using Wilcoxon signed-rank tests.
|
Results |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
|
Supplemental nonprotocol narcotic and nonnarcotic pain medication usage was documented up to day 23. The nonnarcotic medications included Tylenol, toradol, Motrin, and other nonsteroidal antiinflammatory agents. Nonnarcotic medication usage days was 76 in total for the ICN-PCA group compared to 82 in total for the EPI group (p = 0.1709). Supplemental (beyond the planned randomized treatments) narcotic medication days were significantly lower (p = 0.0316) in the ICN-PCA patients (248 days total) compared with the EPI patients (286 days total). The other positive finding in this study was the lower number of Foley catheter days (p = 0.002) in the ICN-PCA group at 81 (mean, 1.7) days versus 108 (mean, 2.5) days for the EPI patients.
The forced expiratory volume in 1 second (FEV1%, forced vital capacity (FVC%), and peak expiratory flow rate (PEFR%) were measured at three assessment time frames; this was before surgery, and postoperatively at days 2 to 3, and at days 4 to 6. Each factor decreased significantly (p = 0.0001) after surgery by about 38% (Fig 2). However, there were no significant differences between the treatment arms.
|
|
Comment |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
Systemic narcotics, usually in combination with antiinflammatory drugs, are often used after thoracotomy [8]. However, high doses of narcotics may lead to respiratory depression, nausea, and bowel dysfunction. Nonsteroidal antiinflammatory drugs may also be associated with complications such as gastrointestinal problems, acute renal failure, and platelet dysfunction. For this reason our group reserves the use of systemic opioid delivered by PCA, in combination with local anesthesia injection, for the less painful VATS procedures rather than after thoracotomy. We believe that a regional delivery method of analgesia is preferential after thoracotomy.
Epidural anesthesia has proven to be an effective method of pain control and has become the gold standard in many centers [9, 10]. Epidural anesthesia, however, is not effective in all patients and carries some risks such as dural perforation, bleeding, infection, hypotension, bradycardia, and urinary retention [9–11]. A particularly devastating complication is the development of an epidural abscess. Fortunately the incidence of this complication is rare and is related to the duration of catheter placement [12]. A 1-year multicenter study from Denmark demonstrated an epidural abscess incidence of 1:1,930 catheter placements. Apart from the low risk of epidural complications additional considerations are that catheter placement may be time consuming and often requires considerable experience for optimal placement. Furthermore, anatomic factors such as previous spinal operations and the significant degrees of scoliosis and osteophytes present in elderly patients with lung cancer may preclude successful epidural placement. In our study there were no significant complications associated with epidural placement. However in 9 of 61 (14.8%) patients the epidural catheter placement was not successful.
Another method of regional analgesia, which has previously been used by thoracic surgeons, is cryoanalgesia [13]. Although relatively simple to perform, this technique has been associated with long-term intercostal neuralgia and has now fallen out of favor in most centers [13].
Sabanathan and colleagues [3] published one of the earlier reports on the use of continuous intercostal nerve block using an extrapleural catheter. Although embraced by some surgical groups this technique has not been widely used, possibly because of concerns of leakage into the pleural space [14], and the potentially high serum concentrations of bupivacaine from absorption from the extrapleural space [15]. As a result of concerns of possible serum toxicity with the relatively longer acting bupivacaine, some centers have used lidocaine [16]. Potential side effects related to high serum levels of bupivacaine include dry mouth, central nervous toxicity, and myocardial suppression. Despite these concerns, bupivacaine toxicity has not been a problem in most series. Usually the bupivacaine concentrations reported range from 0.25% to 0.5% [17, 3]. In a study using a continuous infusion of 0.5% bupivacaine, mean serum levels were 4.2 µmol/L and 4.7 µmol/L on the first and third days postoperatively [18]. This was significantly below 14 µmol/L, a level below which toxicity is rarely seen [18]. In our series, we used the lower concentration of 0.25% bupivacaine. No toxicity was observed and patients still received effective analgesia.
Other studies have demonstrated the efficacy and safety of intercostal nerve catheters for postoperative analgesia after thoracotomy. A study from the United Kingdom [17] randomized patients to extrapleural analgesia with local anesthetic or saline. Patients were still allowed the group’s standard of care analgesia, which consisted of IV morphine and oral analgesics. Pain scores and pulmonary function were significantly better in the patients receiving extrapleural local anesthetic.
A previous randomized study [18] compared EPI with ICA, but this involved a smaller number of patients than ours with only 15 in each group. There were no significant differences in complications in this study, and both methods were found to be similar in controlling pain and preserving pulmonary function [18]. Our own data support these findings. In fact, there may be some advantages to the ICA-PCA because of the reduction in Foley days and total narcotic use that was demonstrated in our study. Although not recorded as an outcome in our study, the use of ICA-PCA may offer some time-saving advantages as placement is usually simply performed at the end of the procedure by the surgeon, whereas EPI placement may be difficult, requiring several minutes before starting the operative procedure. In addition, there may be a cost savings with the ICN-PCA method as there is no requirement for a separate pain service and no daily consulting fees typically associated with an epidural in the postoperative period.
In summary we have demonstrated that intercostal catheters can be placed safely and efficiently by the surgeon. There were no adverse effects from continuous bupivacaine infusion and pain control was as effective as thoracic epidural analgesia. There were some advantages with decreased Foley requirements and supplemental narcotic requirements. The ICN-PCA is recommended after thoracotomy for those patients where epidural placement is not feasible, and perhaps should be considered for all cases where there is no disruption of the parietal pleura.
|
Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
DR FERNANDO: I believe the PCA provided some additional analgesia. I’m not sure that the intrapleural catheter by itself completely took care of all the pain issues. However, one of the interesting findings from our study was that even with using the PCA in combination with the intercostal catheter, we still found that the epidural patients had a higher requirement for additional narcotics, which is difficult to explain.
DR BENNY WEKSLER (Rio de Janeiro, Brazil): Dr Fernando, I just have a question in regards to the supplemental narcotics that were used in the epidural catheter patients. Were those supplemental through the epidural IV?
DR FERNANDO: Some of those were supplemental IV narcotics (in addition to the PCA) and others were supplemental oral medications, such as Percocet. As part of the protocol, the intercostal nerve catheter patients all received supplemental PCA, but anything additional to that was counted as supplemental.
DR WEKSLER: So the difference was the supplemental IV, not through the epidural catheter?
DR FERNANDO: That is correct, not through the epidural. Also, when we looked at our data, there was an increased requirement for supplemental narcotics in the epidural group even after the initial 4 or 5 days when both analgesia methods were terminated. This is difficult to explain.
DR WEKSLER: And the management of the epidural catheter was through the surgeon or through the pain service? I’m asking that because frequently I’ve seen that when the pain management service manages the patients, sometimes you end up with some gaps in analgesia, or too much or too little, and my experience has been that when the surgeon manages it, it’s usually better; the patient becomes more comfortable. So the question is, was the pain management service managing it or were the surgeons managing it?
DR FERNANDO: That is an excellent point, and I think it demonstrates one of the advantages of the intercostal catheter because you don’t have to rely on a pain service. At our institution, we do use a pain service for management of the epidural. In this study most of these epidurals were placed by one of the senior anesthesiologists, who was also a coauthor, so theoretically they got the best postoperative management that our pain service could deliver.
DR TAINE T. PECHET (Philadelphia, PA): Dr Fernando, that was a very enjoyable presentation. Can you clarify the use of COX-2 inhibitors in your patients? What percentage of your patients took them? Did anybody go home on COX-2’s?
DR FERNANDO: I don’t have the data on what the specific use of COX-2 inhibitors was. There are seven other surgeons within our group and we all differ in our use of supplemental medications. For instance, my preference is to use a combination of a PCA and Toradol when an epidural or intercostal catheter is not used, and I will usually send patients home with a combination of Motrin and Percocet.
DR DANIEL L. MILLER (Atlanta, GA): Chris, I just have one question. I think one of the most important issues here is in regard to respiratory complications. Was there a difference between the groups in regard to postoperative pneumonia, requirement for bronchoscopy, and so forth?
DR FERNANDO: That specific information wasn’t tracked; however, in terms of antibiotics, which could be regarded as a surrogate of whether the patients developed atelectasis, pneumonia, and the need for postoperative bronchoscopy, there were no differences between the two groups.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResultsCommentDiscussionAcknowledgmentsReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
