Multimodal analgesia using intrathecal diamorphine, and paravertebral and rectus sheath catheters are as effective as thoracic epidural for analgesia post-open two-phase esophagectomy within an enhanced recovery program

Multimodal analgesia using intrathecal diamorphine, and paravertebral and rectus sheath catheters... Summary Thoracic epidural (TE) analgesia has been the standard of care for transthoracic esophagectomy patients since the 1990s. Multimodal anesthesia using intrathecal diamorphine, local anesthetic infusion catheters (LAC) into the paravertebral space and rectus sheaths and intravenous opioid postoperatively represent an alternative option for postoperative analgesia. While TE can provide excellent pain control, it may inhibit early postoperative recovery by causing hypotension and reducing mobilization. The aim of this study is to determine whether multimodal analgesia with LAC was effective with respect to adequate pain management, and compare its impact on hypotension and mobility. Patients receiving multimodal LAC analgesia were matched using propensity score matching to patients undergoing two-phase trans-thoracic esophagectomy with a TE over a two-year period (from January 2015 to December 2016). Postoperative endpoints that had been evaluated prospectively, including pain scores on movement and at rest, inotrope or vasoconstrictor requirements, and hypotension (systolic BP < 90 mmHg), were compared between cohorts. Out of 14 patients (13 male) that received LAC were matched to a cohort of 14 patients on age, sex, and comorbidity. Mean and maximum pain scores at rest and movement on postoperative days 0 to 3 were equivalent between the groups. In both cohorts, 50% of patients had a pain score of more than 7 on at least one occasion. Fewer patients in the LAC group required vasoconstrictor infusion (LAC: 36% vs. TE: 57%, P = 0.256) to maintain blood pressure or had episodes of hypotension (LAC: 43% vs. TE: 79%, P = 0.05). The LAC group was more able to ambulate on the first postoperative day (LAC: 64% vs. TE: 43%, P = 0.14) but these differences were not statistically significant. Within the epidural cohort, three patients had interruption of epidural due to dislodgement or failure of block compared to no disruption in the multimodal local anesthesia catheters group (P = 0.05). Therefore, multimodal anesthesia using spinal diamorphine with combined paravertebral and rectus sheath local anesthetic catheters appears to provide comparable pain relief post two-phase esophagectomy and may provide more reliable and safe analgesia than the current standard of care. INTRODUCTION Thoracic epidural (TE) has been the standard of care for trans-thoracic esophagectomy since the 1990s. It can provide reliable analgesia postoperatively with minimal imposition on mobilization, compared to systemic opioid analgesia.1 Poorly managed pain postthoracotomy can lead to reduced pulmonary function2 and increased postoperative pulmonary complications. Furthermore, well-managed pain in the immediate postoperative period seems to be associated with a reduced incidence of long-term postthoracotomy pain.3 Adequate analgesia is an important component in the reduction of pulmonary complications following esophagectomy. Increasingly respiratory failure and acute respiratory distress syndrome (ARDS) are the most important causes of postoperative early mortality following esophagectomy.4 In a large modern cohort study of 858 patients undergoing esophagectomy from 1998 to 2008 in Australia, a TE was associated with a significantly reduced risk of postoperative respiratory failure (OR 0.12, P = 0.003).5 For this reason, the use of a TE is increasingly seen as standard of care with a TE performed in 93.2% of the Australian cohort. Although TE reduces pulmonary complications, the baseline risk for developing pneumonia has reduced over time6 and with the advent of enhanced recovery programs, other factors may also play a role in reducing complications and its relative importance may be declining. However, the placement of an epidural catheter is not always possible or acceptable to all patients. There are risks associated with epidurals ranging from serious long-term consequences following an epidural hematoma or abscess, to commonly encountered premature failure of analgesia,7 hypotension,8 and an increased need for intravenous fluid administration. Concerns have been raised that the presence of postoperative hypotension may impact on recovery and perfusion of the conduit,9,10 perhaps impacting on the presence of postoperative anastomotic leaks. Although other studies report improved microcirculation perfusion following epidural anesthesia.11 A key focus of enhanced recovery protocols following esophagectomy includes early mobilization,12 which may be prevented by postoperative hypotension as well as inadequate analgesia. An alternative to epidural anesthesia is infusion of local anesthetic using catheters placed intraoperatively into the rectus sheath bilaterally and into the paravertebral space. Evidence from trials, albeit limited in number and design, following thoracotomy indicate equivalent pain outcomes1 and perhaps a favorable side effect and complication profile.13 Since 2015, a combination of a paraverterbral and two rectus sheath catheters has been used for some patients undergoing open two-phase subtotal esophagectomy. Patients receiving local anesthesia catheters (LAC) also receive intrathecal diamorphine at induction of anesthesia and utilize an intravenous fentanyl patient controlled analgesia (PCA) pump in the postoperative period using a multimodal approach to analgesia. This paper aims to report initial outcomes associated with the use of LAC analgesic method compared to TE. METHODS The Northern Oesophagogastric Unit is a high volume center performing specialized upper gastrointestinal tract surgery. As a service evaluation of current practice, ethical approval was not formally sought. Patients undergoing open two-phase transthoracic esophagectomy from January 2015 to December 2016 were analyzed from a single institution. This coincided with the implementation of a formal enhanced recovery after surgery (ERAS) pathway. All patients underwent a similar staging and preoperative assessment in the multidisciplinary team setting. Patients underwent endoscopy with biopsy, endoscopic ultrasound, computerized tomographic scan of the chest, and abdomen and positron emission tomography for all patients with locally advanced tumors. Staging laparoscopy was used where appropriate. Operative fitness was assessed using spirometry and cardiopulmonary exercise testing.14 Patients with locally advanced tumors (T3 or greater, any node-positive disease) were considered for perioperative chemotherapy using ECX15 or neoadjuvant chemoradiation therapy using the CROSS protocol.16 Resections were carried out using a standardized two-phase approach (Ivor–Lewis) with a radical en-bloc abdominal and mediastinal lymphadenectomy as previously described.17 Out of 12 patients who underwent a minimally invasive thoracoscopic and open abdominal esophagectomy and 10 patients who had three-phase esophagectomies during the time period were excluded from the study. Thoracic epidural analgesia was considered for all patients. Contra-indications for epidural included patient choice, technical difficulty with epidural placement, previous back surgery or anatomical malformation precluding safe use of epidural, platelet disorder, or other bleeding diathesis. All patients were counseled at preoperative assessment clinic regarding options for analgesia postoperatively and provided with a copy of the Royal College of Anaesthetists epidural patient information leaflet.18 A final decision was made with the patient at the preoperative anesthetic assessment at time of admission. Selection of patients for multimodal LAC-based analgesia was initially due to either technical difficulties placing the epidural (n = 3) or patient choice (refusal of epidural, n = 3). After increasing experience and confidence with use of this form of analgesia, increasingly selection of LAC was based on patient choice after discussing analgesic options with the surgical and anesthetic team during the consent process on the day preceding surgery (Supplementary Table 1). All patients received either spinal or epidural analgesia before induction of general anesthesia. Epidurals were placed at a mid-thoracic level. Spinal analgesia comprised subarachnoid injection of 1 mg diamorphine and 5–7.5 mg hyperbaric bupivacaine. General anesthesia was maintained with either intravenous 2% propofol (TCI) and remifentanil (TCI) or inhaled desflurane and remifentanil. Single lung ventilation was achieved with a Mallinckrodt double lumen endotracheal tube. All patients had standard monitoring, radial artery catheters, and urinary catheters placed. Central venous cannulation, cardiac output monitoring (LiDCO), and BIS monitoring were used as required on an individual patient basis. Fluid status was continuously monitored intraoperatively, using arterial blood gas analysis, urine output, CVP where available, and stroke volume variation derived from pulse contour analysis; with the aim of limiting intravenous fluid administration while maintaining euvolemia in a goal-directed manner. All patients received standard perioperative antibiotic and thromboembolic prophylaxis, as well as intra-operative warming. A nasogastric tube, as well as anastomotic (24Ch) and basal (24Ch) thoracic drains were used routinely. The rectus sheath and paravertebral catheters were placed prior to closure of incisions. The dual-rectus sheath catheters were placed at the superior aspect of the incision using the On-Q® Painbuster kit® (B Braun, Melsungen, Germany). With the blunt tipped introducer tunneled posterior to the rectus muscle on either side of the incision. The paravertebral catheter was tunneled under direct vision using a Portex epidural insertion kit (16G epidural catheter and Touhy needle) into the paravertebral space immediately above the thoracotomy.19 A bolus of 20 mL 0.25% levobupivacaine (50 mg) was split between the two rectus sheath catheters, and a further 15–20 mL of 0.25% levobupivacaine (37.5–50 mg) was given via the paravertebral catheter to ensure both systems were loaded with local anesthetic at the commencement of infusion. A 270 mL elastomeric reservoir of 0.25% levobupivacaine, running at 2 + 2 mL/hour was attached to the dual rectus sheath catheters; and a 400 mL elastomeric reservoir of 0.25% levobupivacaine was attached to the paravertebral catheter, and infused at 6–10 mL/hour. LAC catheters were removed routinely on postoperative day 3 once the reservoirs were empty (total infusion: up to 670 mL). Patients undergoing multimodal analgesia and placement of LAC catheters were provided with a patient controlled intravenous fentanyl infusion in the recovery room (PCA) (10–20 mcg fentanyl bolus with 5 minute lockout and a maximum dose of 480 mcg in 4 hours). In patients receiving an epidural, a catheter was placed at about the T7/8 level prior to induction of GA. An initial test bolus of 6 mL 0.25% levobupivacaine was provided and an infusion of 0.125% levobupivacaine with 4 mcg/mL fentanyl infused per and postoperatively at a mean 5 mL/h (range: 3–8 mL/hour), with a 5 mL PCEA bolus facility provided in addition, maximally every 20 minutes. Removal was planned for the fourth postoperative day if pain was well controlled and no epidural related complications encountered prior to this time point. Patients were extubated at the end of the case and monitored on a high dependency unit (HDU) for at least 16 hours before being discharged to a specialist esophagogastric ward. Pain was monitored and analgesia titrated in order to try and obtain a visual analogue pain score ≤3. Discharge from HDU was dependent on hemodynamic and respiratory stability. Patients were reviewed daily by the acute pain service (specialist pain nurse and/or anesthesia specialist registrar/consultant) and additionally, in instances where the pain score was >4. Pain scores and level of block for patients with epidural were recorded by trained nurses on the ward and analgesia was titrated with the aim of maintaining a pain score of less than 4 on movement. Observations and scores were recorded 4 hourly or sooner as clinically dictated. The pain score was reported on a scale from 0 to 10 at both rest and movement. In cases of hypotension (systolic blood pressure <90 mmHg), patients were assessed for contributing factors and the epidural infusion titrated or suspended until hemodynamic stability was re-established. In most cases, fluid boluses of 20 mL/kg were utilized to increase blood pressure and in cases of persistent hypotension with no pulse pressure variability, vasoconstrictor medications (noradrenaline) were commenced and titrated to keep mean arterial pressure >70 mmHg. An ERAS program was established on the unit utilizing an integrated care pathway (ICP) at the beginning of the time period in question. Briefly, the ERAS program was designed to progress attainment of daily goals using the integrated care pathway. The aim is that patients ambulate for a target distance of 300 m at least three times on postoperative days 1 and 2 and increasing the distance as able on postoperative day 3. Data were prospectively recorded using a standardized proforma. Further details on pain scores and fluid administration were obtained by review of the patient notes. Pain scores were recorded by trained nursing staff at rest and movement at least every 4 hours while LAC or TE were in situ or more frequently in case of poorly controlled pain (score ≥4). Ambulation goals were recorded in the ICP document and medication use on the electronic patient record. Complications were assessed twice daily and recorded using the Accordion classification.20 Statistical analysis Continuous variables were compared using unpaired t- or Mann–Whitney U tests as appropriate. Association of categorical variables (differences for dichotomous variables between groups) was assessed using chi-square (Χ2) test. Propensity score matching was used in order to minimize confounding effects due to nonrandomized treatment decisions to compare outcomes for patients treated with neoadjuvant chemoradiation treatment followed by surgery or surgery only versus definitive chemoradiation therapy, both with curative intent. A propensity score was calculated using logistic regression analysis matched on the following variables: age, sex, and American Society of Anesthesiology (ASA) grade. ASA grade was chosen as a measure of comorbidity and it is predictive of mortality and morbidity following esophagectomy.4 Patients who received LAC catheters were matched on a 1:1 ratio to those who received a TE, using the nearest neighbor matching algorithm without replacement on the logit of the propensity score. A caliper of width equal 0.2 SDs of the logit of propensity score was used.21 After propensity score matching, balance of covariates was tested through matching procedures using standardized differences (an absolute standardized difference of <10% was considered a negligible imbalance) and assessment of global imbalance (P = 0.98 after matching).22 RESULTS Demographics Over the two-year study time period, 184 patients underwent esophagectomy with curative intent with 159 undergoing open two-phase transthoracic esophagectomy. Fourteen patients received multimodal analgesia with spinal diamorphine, paravertebral and rectus sheath catheter local anesthesia infusions, and PCA: the remainder (n = 145) received a TE. The demographics of the patients receiving local anesthetic catheters were broadly similar to the overall cohort but in order to analyze outcomes in more details, propensity score matching was used to match the LAC cohort to similar patients receiving a TE (Table 1). The mean ARISCAT risk score for postoperative pulmonary complications for patients in both the LAC and matched cohorts was 50 compared to a mean score of 51.9 for the overall cohort.23 Table 1 Patient demographics and tumor details Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) *P-value for comparison of multimodal LAC versus matched epidural cohort. ASA, American Society of Anesthesiology; BMI, body mass index; LAC, local anesthesia catheter. View Large Table 1 Patient demographics and tumor details Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) *P-value for comparison of multimodal LAC versus matched epidural cohort. ASA, American Society of Anesthesiology; BMI, body mass index; LAC, local anesthesia catheter. View Large Pain Mean and maximum pain scores at rest and movement on postoperative days 0 to 3 were equivalent between the groups (Table 2). Three patients in the matched TE group experienced malfunctioning of their TE (leakage, high pain scores, and no sensory block) leading to earlier than planned removal of the epidural and provision of a morphine PCA. Table 2 Pain scores Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 2 Pain scores Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 LAC, local anesthesia catheter; TE, thoracic epidural. View Large The mean cumulative dose received via peripheral venous fentanyl PCA was 3369 mcg (+/−2200) and the mean cumulative dose of fentanyl delivered by patient controlled epidural anesthesia to the matched group was 659 mcg (±326.9). Half of patients in both cohorts experienced a pain score of more than 7 on at least one occasion. The mean number of pain episodes scored greater than 7 was 1.5 (range: 0–5) in the LAC group and 2 (range: 0–11) in the TE group (P = 0.371). Functional recovery There were no statistically significant differences between the proportion of patients requiring vasoconstrictor infusion postoperatively: (LAC: 36% vs. TE: 57%, P = 0.256) to maintain blood pressure; and differences in the frequency of episodes of hypotension (LAC: 43% v TE: 79%, P = 0.053) were not different (Table 3). Fluid administered intraoperatively did not differ between the groups (LAC: 3.7 L ± 1.5 vs. TE: 4.5 L ± 0.96, P = 0.107). Intraoperative blood loss did not differ (LAC: 449 mL ± 233 vs. TE: 475 mL ± 346, P = 0.79). There was an improvement in the ability to ambulate on the first postoperative day (LAC: 64% vs. TE: 43%, P = 0.138) as well as the proportion of patients achieving ambulation goals on POD 1 and 2 (Table 4). Table 3 Hemodynamic outcomes LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 3 Hemodynamic outcomes LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 4 Proportion achieving ambulation goals LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 4 Proportion achieving ambulation goals LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Complications are recorded in Table 5. Two patients in the LAC cohort suffered Accordion grade 4 or more complications: one with idiopathic ARDS after neoadjuvant chemoradiotherapy and the other who developed a stroke and respiratory failure on postoperative day 7. Although more grade 4 or higher complications were experienced in the LAC cohort compared to the matched TE cohort, (LAC: 2/14 vs. TE: 0/14, P = 0.203), the difference was not statistically significant. Table 5 Complications LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 5 Complications LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large The median total length of stay for the overall cohort of 184 patients was 15.53 days (SD: ±15.61). The length of stay for the LAC cohort was 10.77 days (±3.66) versus 11.57 days (=/-2.95) for the matched TE cohort (P = 0.535). The length of stay in critical care was 5.0 days ±7.8 in the overall cohort versus LAC: 3.3 days ± 1.0 and TE: 3.0 days ± 2.1(P = 0.912). DISCUSSION Thoracic epidural currently represents the standard of care for patients undergoing esophagectomy due to its reliable reduction in pain and associated reduction in pulmonary complications. Even following minimally invasive esophagectomy, there is a reduction by almost 50% in postoperative mortality associated with TE24 compared to management with systemic intravenous opioid only. TE is not suitable for all patients and the failure rate may be as high as 25%8,25 and was 21% in this study. There is a risk of epidural hematoma or abscess resulting in permanent spinal cord injury and the risk of nerve damage is not insubstantial and possibly underrecognized.26 In a systematic review of analgesic methods postthoracotomy, seven studies were identified comparing epidural to paravertebral infusion and most (five) reported superior pain outcomes with paravertebral infusion—although there was a large degree of heterogeneity in the pain scores reported between studies.1 The incidence of hypotension seemed to be reduced in the three studies reporting this endpoint. In a meta-analysis of 10 RCTs (n = 520 patients) of patients undergoing thoracotomy, comparing paravertebral (PV) catheters versus TE, PV resulted in comparable analgesia with a better side effect profile, specifically reduced pulmonary complications and less hypotension.27 Based on these data, one-third of UK centers now employ PV catheters postpneumonectomy and observational data support an association with fewer postoperative complications with PV catheters compared to TE.28 The difficulty in extrapolating these results to the postesophagectomy patient includes a greater proportion of patients experiencing complications following esophagectomy compared to other types of major surgery29 and the presence of dual abdominal and thoracic incisions—which may result in more pain and associated functional consequences than thoracotomy alone. Subcostal transversus abdominis plane block (single injection) for open gastrectomy provides improved analgesia compared to IV opioid alone but pain scores with epidural anesthesia are superior at randomized controlled trial.30 The evidence base for the use of LAC catheters following esophagectomy is limited to a single center series using paravertebral catheters in conjunction with epidural anesthesia at T7/8 level to cover the laparotomy component (n = 52). This group was compared to patients who received two epidurals (one at T3/4 and another at T7/8) for analgesia. The frequency of hypotension was significantly reduced in the PVB catheter group versus the two epidural group (9.6% vs. 30.2%, P = 0.008).31 A further single center study of 16 patients from Taiwan utilized single-shot paravertebral and transversus abdominis plane blocks in patients undergoing minimally invasive esophagectomy and reported reduced pain scores and reduced opioid use, compared to patients receiving only intravenous systemic opioid postoperatively.32 Patients without blocks were more likely to report moderate to severe pain in the immediate 2–6 hours postoperatively (for example at 2 hours: 65% vs. 30%, P = 0.049). Thus an alternative to TE anesthesia utilizing multimodal analgesia and local anesthestic delivered via combined paravertebral and rectus sheath catheters may be an attractive proposition for the postesophagectomy patient and worthy of further study. The validity of pain scores as an endpoint is an issue since it is the patient's overall recovery, which is of greatest interest to clinicians and it is not known whether pain scores correlate well with ability to mobilize. Inadequate pain control results in increased complications, prolonged length of stay, and reduced mobility33 but the threshold defining inadequate pain control is not well established. Although large meta-analyses have demonstrated improved pain outcomes with epidural compared to intravenous opioids, the mean difference in visual analogue pain scores varies from 2.2 to 19.9 mm, the clinical relevance of which is difficult to determine.34 Within enhanced recovery programs following open abdominal surgery, at meta-analysis, the use of epidural analgesia is associated with reduced pain scores but there does not seem to be a difference in length of stay and systemic analgesia was associated with fewer complications35 and thus, epidural analgesia is no longer a central component of care. Assessment of functional outcomes such as mobility may be of more relevance to the assessment of analgesia postoperatively. If pain scores remain manageable and do not hinder mobility, we deem this to be clinically equivalent. This study reports for the first time the use of spinal diamorphine, LAC (combined PV and rectus sheath) and PCA for patients following open two-phase transthoracic esophagectomy. In this matched cohort, there are no significant differences in pain scores between the two techniques. The requirement for fluids and pressor support is also an important outcome following esophagectomy as hemodynamic instability limits the ability of patients to ambulate and in the postesophagectomy patient is associated with the development of acute lung injury,36 an important complication of surgery and important contributor to mortality. Current practice aims to judiciously limit fluid administration, aiming to balance organ perfusion, vasopressor requirement, and the avoidance of fluid overload. This could increase the incidence of hypotension and pressor requirement compared to where a liberal fluid policy is employed.37,38 Importantly in this study, there were trends to a reduction in hypotension as well as a reduction in need for vasopressor infusion. This translated in more patients ambulating on the first postoperative day as well as the achieving the desired ambulation goals of the enhanced recovery program. These functional endpoints, rather than absolute pain scores alone would be important for future randomized trials of analgesic modalities in esophagectomy patients. This retrospective study has a number of significant limitations. The overall proportion of patients receiving multimodal LAC infusion technique rather than TE is small—owing mainly to the lack of previous experience and the dearth of an evidence base supporting this practice. There is an unintentional selection bias in the cohort given that the practice began as an alternative to failed epidural. The numbers included are too small to provide absolute proof of any benefit of LAC catheters but provide some evidence on which to power future studies. This study indicates that outcomes such as use of vasoconstrictors or ability to ambulate may be important endpoints for future study. An important issue will be any possible impact of the analgesia techniques on complications, specifically pulmonary complications postesophagectomy. The risk of developing pneumonia after major surgery may have decreased in recent decades. One metaanalysis of epidural analgesia to prevent postoperative pneumonia, reports the number needed to treat (NNT) of 25 from studies in the past 20 years versus 4 for studies from the 1970s.6 Therefore, proxy measures of outcomes may be required for randomized studies rather than overall complication rates. Complications and length of stay were not obviously different but the study underpowered to detect any clear difference. It is important to note that the length of stay is not a complete assessment of recovery given LOS is multifactorial.39 Theoretically, epidural neural modulation may attenuate the postoperative neuroendocrine stress response following esophagectomy,40 although this effect has not been observed in any studies.41 The use of local anesthesia catheters may not have any such effects and are placed at incision closure rather than prior to incision, although the use of intrathecal diamorphine prior to induction, may play a role in neuromodulation. Although epidural from C3 to L1 may reduce the postoperative inflammatory cytokine response following esophagectomy,41 the consequences of a lower TE, which is applicable to clinical practice, and any impact of any immunomodulation are not readily clinically apparent and will require much larger, likely population based study to determine impact on morbidity following surgery. Studies of local anesthesia infusion in hepatic surgery did not show any obvious differences in the inflammatory response compared to epidural.42 This study provides evidence of the feasibility of combining spinal diamorphine with local anesthesia catheters in the paravertebral space and rectus sheaths as an alternative to TE for patients after open transthoracic esophagectomy. It indicates that clinically relevant reductions in analgesia failure, hypotension and requirement of vasopressors may reflect a more favorable side effect profile using a multimodal LAC approach compared to TE and future randomized studies assessing functional recovery are warranted. Notes Disclosures: The authors declare no commercial or other interests. There was no funding sought for this study. Specific author contributions: Design of work: Claire L. Donohoe, Emma Flynn, Claire L. Taylor, Catherine Donnison, Arul Immanuel, Alexander W. Phillips; Data acquisition: Claire L. Donohoe, Emma Flynn, Claire L. Taylor, Catherine Donnison, Arul Immanuel, Alexander W. Phillips; Data analysis and interpretation: Claire L. Donohoe, Alexander W. Phillips; Drafting article: Claire L. Donohoe; Critical revision of article: Alexander W. Phillips, Rhona C. F. Sinclair, David Saunders, S. Michael Griffin. References 1 Joshi G P , Bonnet F , Shah R et al. A systematic review of randomized trials evaluating regional techniques for postthoracotomy analgesia . Anesth Analg 2008 ; 107 : 1026 – 40 . Google Scholar CrossRef Search ADS PubMed 2 Richardson J , Sabanathan S , Shah R . Postthoracotomy spirometric lung function: the effect of analgesia: a review . J Cardiovasc Surg (Torino) 1999 ; 40 : 445 . Google Scholar PubMed 3 Katz J , Jackson M , Kavanagh B P et al. Acute pain after thoracic surgery predicts long-term postthoracotomy pain . Clin J Pain 1996 ; 12 : 50 – 55 . Google Scholar CrossRef Search ADS PubMed 4 Sauvanet A , Mariette C , Thomas P et al. Mortality and morbidity after resection for adenocarcinoma of the gastroesophageal junction: predictive factors . J Am Coll Surg 2005 ; 201 : 253 – 62 . Google Scholar CrossRef Search ADS PubMed 5 Zingg U , Smithers B M , Gotley D C et al. Factors associated with postoperative pulmonary morbidity after esophagectomy for cancer . Ann Surg Oncol 2011 ; 18 : 1460 – 8 . Google Scholar CrossRef Search ADS PubMed 6 Pöpping D M , Elia N , Marret E et al. Protective effects of epidural analgesia on pulmonary complications after abdominal and thoracic surgery . Arch Surg 2008 ; 143 : 990 – 9 . Google Scholar CrossRef Search ADS PubMed 7 Hermanides J , Hollmann M , Stevens M et al. Failed epidural: causes and management . Br J Anaesth 2012 ; 109 : 144 – 54 . Google Scholar CrossRef Search ADS PubMed 8 Viscusi E R . Emerging techniques in the management of acute pain: epidural analgesia . Anesth Analg 2005 ; 101 ( Supplement ): S23 – 9 . Google Scholar CrossRef Search ADS PubMed 9 Al-Rawi O Y , Pennefather S H , Page R D et al. The effect of thoracic epidural bupivacaine and an intravenous adrenaline infusion on gastric tube blood flow during esophagectomy . Anesth Analg 2008 ; 106 : 884 – 7 . Google Scholar CrossRef Search ADS PubMed 10 Pathak D , Pennefather S H , Russell G N et al. Phenylephrine infusion improves blood flow to the stomach during oesophagectomy in the presence of a thoracic epidural analgesia . Eur J Cardiothorac Surg 2013 ; 44 : 130 – 3 . Google Scholar CrossRef Search ADS PubMed 11 Michelet P , Roch A , D’Journo X B et al. Effect of thoracic epidural analgesia on gastric blood flow after oesophagectomy . Acta Anaesthesiol Scand 2007 ; 51 : 587 – 94 . Google Scholar CrossRef Search ADS PubMed 12 Yip V , Dunne D , Samuels S et al. Adherence to early mobilisation: key for successful enhanced recovery after liver resection . Eur J Surg Oncol 2016 ; 42 : 1561 – 7 . Google Scholar CrossRef Search ADS PubMed 13 Scarci M , Joshi A , Attia R . In patients undergoing thoracic surgery is paravertebral block as effective as epidural analgesia for pain management ? Interact Cardiovas Thorac Surg 2010 ; 10 : 92 – 6 . Google Scholar CrossRef Search ADS 14 Sinclair R , Navidi M , Griffin S et al. The impact of neoadjuvant chemotherapy on cardiopulmonary physical fitness in gastro-oesophageal adenocarcinoma . Ann R Coll Surg Engl 2016 ; 98 : 396 – 400 . Google Scholar CrossRef Search ADS PubMed 15 Cunningham D , Allum W H , Stenning S P et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer . N Engl J Med 2006 ; 355 : 11 – 20 . Google Scholar CrossRef Search ADS PubMed 16 van Hagen P , Hulshof M C C M , van Lanschot J J B et al. Preoperative chemoradiotherapy for esophageal or junctional cancer . N Engl J Med 2012 ; 366 : 2074 – 84 . Google Scholar CrossRef Search ADS PubMed 17 Phillips A W , Dent B , Navidi M et al. Trainee involvement in Ivor–Lewis esophagectomy does not negatively impact outcomes . Ann Surg 2018 ; 267 : 94 – 98 . Google Scholar CrossRef Search ADS PubMed 18 Royal College of Anaesthetists . Epidural pain relief after surgery: patient information leaflet . 4TH edition , 2014 London . Accessed at www.rcoa.ac.uk/patientinfo . 19 Tighe S , Greene M D , Rajadurai N . Paravertebral block . Contin Educ Anaesth Crit Care Pain 2010 ; 10 : 133 – 7 : mkq029 . Google Scholar CrossRef Search ADS 20 Strasberg S M , Linehan D C , Hawkins W G . The accordion severity grading system of surgical complications . Ann Surg 2009 ; 250 : 177 – 86 . Google Scholar CrossRef Search ADS PubMed 21 Austin P C . Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies . Pharm Stat 2011 ; 10 : 150 – 61 . Google Scholar CrossRef Search ADS PubMed 22 Bowers J , Fredrickson M , Hansen B . RItools: Randomization inference tools . R Package, Version 01-11 , Vienna, Austria : R. Foundation for Statistical Computing , 2010 . 23 Mazo V , Sabaté S , Canet J et al. Prospective external validation of a predictive score for postoperative pulmonary complications . Anesthesiology 2014 ; 121 : 219 – 31 . Google Scholar CrossRef Search ADS PubMed 24 Zingg U , McQuinn A , DiValentino D et al. Minimally invasive versus open esophagectomy for patients with esophageal cancer . Ann Thorac Surg 2009 ; 87 : 911 – 9 . Google Scholar CrossRef Search ADS PubMed 25 Ready L B . Acute pain: lessons learned from 25,000 patients . Reg Anesth Pain Med 1999 ; 24 : 499 – 505 . Google Scholar PubMed 26 Moen V , Dahlgren N , Irestedt L . Severe neurological complications after central neuraxial blockades in Sweden 1990–1999 . J Am Soc Anesthesiol 2004 ; 101 : 950 – 9 . Google Scholar CrossRef Search ADS 27 Davies R , Myles P , Graham J . A comparison of the analgesic efficacy and side-effects of paravertebral vs epidural blockade for thoracotomy—a systematic review and meta-analysis of randomized trials . Br J Anaesth 2006 ; 96 : 418 – 26 . Google Scholar CrossRef Search ADS PubMed 28 Powell E , Cook D , Pearce A et al. A prospective, multicentre, observational cohort study of analgesia and outcome after pneumonectomy . Br J Anaesth 2011 ; 106 : 364 – 70 . Google Scholar CrossRef Search ADS PubMed 29 Merkow R P , Bilimoria K Y , McCarter M D et al. Short-term outcomes after esophagectomy at 164 American College of Surgeons National Surgical Quality Improvement Program Hospitals . Arch Surg 2012 ; 147 : 1009 – 16 . Google Scholar CrossRef Search ADS PubMed 30 Wu Y , Liu F , Tang H et al. The analgesic efficacy of subcostal transversus abdominis plane block compared with thoracic epidural analgesia and intravenous opioid analgesia after radical gastrectomy . Anesth Analg 2013 ; 117 : 507 – 13 . Google Scholar CrossRef Search ADS PubMed 31 Niwa Y , Koike M , Torii K et al. Combination of continuous paravertebral block and epidural anesthesia in postoperative pain control after esophagectomy . Esophagus 2016 ; 13 : 42 – 47 . Google Scholar CrossRef Search ADS 32 Li N-L , Liu C-C , Cheng S H-C et al. Feasibility of combined paravertebral block and subcostal transversus abdominis plane block in postoperative pain control after minimally invasive esophagectomy . Acta Anaesthesiol Taiwan 2013 ; 51 : 103 – 7 . Google Scholar CrossRef Search ADS PubMed 33 Wu C L , Fleisher L A . Outcomes research in regional anesthesia and analgesia . Anesth Analg 2000 ; 91 : 1232 – 42 . Google Scholar PubMed 34 Block B M , Liu S S , Rowlingson A J et al. Efficacy of postoperative epidural analgesia . JAMA 2003 ; 290 : 2455 – 63 . Google Scholar CrossRef Search ADS PubMed 35 Hughes M J , Ventham N T , McNally S et al. Analgesia after open abdominal surgery in the setting of enhanced recovery surgery . JAMA Surg 2014 ; 149 : 1224 – 30 . Google Scholar CrossRef Search ADS PubMed 36 Tandon S , Batchelor A , Bullock R et al. Peri-operative risk factors for acute lung injury after elective oesophagectomy . Br J Anaesth 2001 ; 86 : 633 – 8 . Google Scholar CrossRef Search ADS PubMed 37 Chau E H L , Slinger P . Perioperative fluid management for pulmonary resection surgery and esophagectomy . Seminars in cardiothoracic and vascular anesthesia , Vol. 18 , Los Angeles, CA : SAGE Publications , 2014 ; pp. 36 – 44 . Google Scholar CrossRef Search ADS 38 Neal J M , Wilcox R T , Allen H W et al. Near-total esophagectomy: the influence of standardized multimodal management and intraoperative fluid restriction 1, 2 . Reg Anesth Pain Med 2003 ; 28 : 328 – 34 . Google Scholar PubMed 39 Maessen J , Dejong C , Hausel J et al. A protocol is not enough to implement an enhanced recovery programme for colorectal resection . Br J Surg 2007 ; 94 : 224 – 31 . Google Scholar CrossRef Search ADS PubMed 40 Ahlers O , Nachtigall I , Lenze J et al. Intraoperative thoracic epidural anaesthesia attenuates stress-induced immunosuppression in patients undergoing major abdominal surgery† . Br J Anaesth 2008 ; 101 : 781 – 7 . Google Scholar CrossRef Search ADS PubMed 41 Yokoyama M , Itano Y , Katayama H et al. The effects of continuous epidural anesthesia and analgesia on stress response and immune function in patients undergoing radical esophagectomy . Anesth Analg 2005 ; 101 : 1521 – 7 . Google Scholar CrossRef Search ADS PubMed 42 Hughes M , Harrison E , Peel N et al. Randomized clinical trial of perioperative nerve block and continuous local anaesthetic infiltration via wound catheter versus epidural analgesia in open liver resection (LIVER 2 trial) . Br J Surg 2015 ; 102 : 1619 – 28 . Google Scholar CrossRef Search ADS PubMed SUPPLEMENTARY DATA Supplementary data are available at DOTESO online. Supplementary Table 1: Patient selection factors for paravertebral/rectus sheath catheters © The Author(s) 2018. Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Diseases of the Esophagus Oxford University Press

Multimodal analgesia using intrathecal diamorphine, and paravertebral and rectus sheath catheters are as effective as thoracic epidural for analgesia post-open two-phase esophagectomy within an enhanced recovery program

Loading next page...
 
/lp/ou_press/multimodal-analgesia-using-intrathecal-diamorphine-and-paravertebral-HVjSMxNzeE
Publisher
The International Society for Diseases of the Esophagus
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus.
ISSN
1120-8694
eISSN
1442-2050
D.O.I.
10.1093/dote/doy006
Publisher site
See Article on Publisher Site

Abstract

Summary Thoracic epidural (TE) analgesia has been the standard of care for transthoracic esophagectomy patients since the 1990s. Multimodal anesthesia using intrathecal diamorphine, local anesthetic infusion catheters (LAC) into the paravertebral space and rectus sheaths and intravenous opioid postoperatively represent an alternative option for postoperative analgesia. While TE can provide excellent pain control, it may inhibit early postoperative recovery by causing hypotension and reducing mobilization. The aim of this study is to determine whether multimodal analgesia with LAC was effective with respect to adequate pain management, and compare its impact on hypotension and mobility. Patients receiving multimodal LAC analgesia were matched using propensity score matching to patients undergoing two-phase trans-thoracic esophagectomy with a TE over a two-year period (from January 2015 to December 2016). Postoperative endpoints that had been evaluated prospectively, including pain scores on movement and at rest, inotrope or vasoconstrictor requirements, and hypotension (systolic BP < 90 mmHg), were compared between cohorts. Out of 14 patients (13 male) that received LAC were matched to a cohort of 14 patients on age, sex, and comorbidity. Mean and maximum pain scores at rest and movement on postoperative days 0 to 3 were equivalent between the groups. In both cohorts, 50% of patients had a pain score of more than 7 on at least one occasion. Fewer patients in the LAC group required vasoconstrictor infusion (LAC: 36% vs. TE: 57%, P = 0.256) to maintain blood pressure or had episodes of hypotension (LAC: 43% vs. TE: 79%, P = 0.05). The LAC group was more able to ambulate on the first postoperative day (LAC: 64% vs. TE: 43%, P = 0.14) but these differences were not statistically significant. Within the epidural cohort, three patients had interruption of epidural due to dislodgement or failure of block compared to no disruption in the multimodal local anesthesia catheters group (P = 0.05). Therefore, multimodal anesthesia using spinal diamorphine with combined paravertebral and rectus sheath local anesthetic catheters appears to provide comparable pain relief post two-phase esophagectomy and may provide more reliable and safe analgesia than the current standard of care. INTRODUCTION Thoracic epidural (TE) has been the standard of care for trans-thoracic esophagectomy since the 1990s. It can provide reliable analgesia postoperatively with minimal imposition on mobilization, compared to systemic opioid analgesia.1 Poorly managed pain postthoracotomy can lead to reduced pulmonary function2 and increased postoperative pulmonary complications. Furthermore, well-managed pain in the immediate postoperative period seems to be associated with a reduced incidence of long-term postthoracotomy pain.3 Adequate analgesia is an important component in the reduction of pulmonary complications following esophagectomy. Increasingly respiratory failure and acute respiratory distress syndrome (ARDS) are the most important causes of postoperative early mortality following esophagectomy.4 In a large modern cohort study of 858 patients undergoing esophagectomy from 1998 to 2008 in Australia, a TE was associated with a significantly reduced risk of postoperative respiratory failure (OR 0.12, P = 0.003).5 For this reason, the use of a TE is increasingly seen as standard of care with a TE performed in 93.2% of the Australian cohort. Although TE reduces pulmonary complications, the baseline risk for developing pneumonia has reduced over time6 and with the advent of enhanced recovery programs, other factors may also play a role in reducing complications and its relative importance may be declining. However, the placement of an epidural catheter is not always possible or acceptable to all patients. There are risks associated with epidurals ranging from serious long-term consequences following an epidural hematoma or abscess, to commonly encountered premature failure of analgesia,7 hypotension,8 and an increased need for intravenous fluid administration. Concerns have been raised that the presence of postoperative hypotension may impact on recovery and perfusion of the conduit,9,10 perhaps impacting on the presence of postoperative anastomotic leaks. Although other studies report improved microcirculation perfusion following epidural anesthesia.11 A key focus of enhanced recovery protocols following esophagectomy includes early mobilization,12 which may be prevented by postoperative hypotension as well as inadequate analgesia. An alternative to epidural anesthesia is infusion of local anesthetic using catheters placed intraoperatively into the rectus sheath bilaterally and into the paravertebral space. Evidence from trials, albeit limited in number and design, following thoracotomy indicate equivalent pain outcomes1 and perhaps a favorable side effect and complication profile.13 Since 2015, a combination of a paraverterbral and two rectus sheath catheters has been used for some patients undergoing open two-phase subtotal esophagectomy. Patients receiving local anesthesia catheters (LAC) also receive intrathecal diamorphine at induction of anesthesia and utilize an intravenous fentanyl patient controlled analgesia (PCA) pump in the postoperative period using a multimodal approach to analgesia. This paper aims to report initial outcomes associated with the use of LAC analgesic method compared to TE. METHODS The Northern Oesophagogastric Unit is a high volume center performing specialized upper gastrointestinal tract surgery. As a service evaluation of current practice, ethical approval was not formally sought. Patients undergoing open two-phase transthoracic esophagectomy from January 2015 to December 2016 were analyzed from a single institution. This coincided with the implementation of a formal enhanced recovery after surgery (ERAS) pathway. All patients underwent a similar staging and preoperative assessment in the multidisciplinary team setting. Patients underwent endoscopy with biopsy, endoscopic ultrasound, computerized tomographic scan of the chest, and abdomen and positron emission tomography for all patients with locally advanced tumors. Staging laparoscopy was used where appropriate. Operative fitness was assessed using spirometry and cardiopulmonary exercise testing.14 Patients with locally advanced tumors (T3 or greater, any node-positive disease) were considered for perioperative chemotherapy using ECX15 or neoadjuvant chemoradiation therapy using the CROSS protocol.16 Resections were carried out using a standardized two-phase approach (Ivor–Lewis) with a radical en-bloc abdominal and mediastinal lymphadenectomy as previously described.17 Out of 12 patients who underwent a minimally invasive thoracoscopic and open abdominal esophagectomy and 10 patients who had three-phase esophagectomies during the time period were excluded from the study. Thoracic epidural analgesia was considered for all patients. Contra-indications for epidural included patient choice, technical difficulty with epidural placement, previous back surgery or anatomical malformation precluding safe use of epidural, platelet disorder, or other bleeding diathesis. All patients were counseled at preoperative assessment clinic regarding options for analgesia postoperatively and provided with a copy of the Royal College of Anaesthetists epidural patient information leaflet.18 A final decision was made with the patient at the preoperative anesthetic assessment at time of admission. Selection of patients for multimodal LAC-based analgesia was initially due to either technical difficulties placing the epidural (n = 3) or patient choice (refusal of epidural, n = 3). After increasing experience and confidence with use of this form of analgesia, increasingly selection of LAC was based on patient choice after discussing analgesic options with the surgical and anesthetic team during the consent process on the day preceding surgery (Supplementary Table 1). All patients received either spinal or epidural analgesia before induction of general anesthesia. Epidurals were placed at a mid-thoracic level. Spinal analgesia comprised subarachnoid injection of 1 mg diamorphine and 5–7.5 mg hyperbaric bupivacaine. General anesthesia was maintained with either intravenous 2% propofol (TCI) and remifentanil (TCI) or inhaled desflurane and remifentanil. Single lung ventilation was achieved with a Mallinckrodt double lumen endotracheal tube. All patients had standard monitoring, radial artery catheters, and urinary catheters placed. Central venous cannulation, cardiac output monitoring (LiDCO), and BIS monitoring were used as required on an individual patient basis. Fluid status was continuously monitored intraoperatively, using arterial blood gas analysis, urine output, CVP where available, and stroke volume variation derived from pulse contour analysis; with the aim of limiting intravenous fluid administration while maintaining euvolemia in a goal-directed manner. All patients received standard perioperative antibiotic and thromboembolic prophylaxis, as well as intra-operative warming. A nasogastric tube, as well as anastomotic (24Ch) and basal (24Ch) thoracic drains were used routinely. The rectus sheath and paravertebral catheters were placed prior to closure of incisions. The dual-rectus sheath catheters were placed at the superior aspect of the incision using the On-Q® Painbuster kit® (B Braun, Melsungen, Germany). With the blunt tipped introducer tunneled posterior to the rectus muscle on either side of the incision. The paravertebral catheter was tunneled under direct vision using a Portex epidural insertion kit (16G epidural catheter and Touhy needle) into the paravertebral space immediately above the thoracotomy.19 A bolus of 20 mL 0.25% levobupivacaine (50 mg) was split between the two rectus sheath catheters, and a further 15–20 mL of 0.25% levobupivacaine (37.5–50 mg) was given via the paravertebral catheter to ensure both systems were loaded with local anesthetic at the commencement of infusion. A 270 mL elastomeric reservoir of 0.25% levobupivacaine, running at 2 + 2 mL/hour was attached to the dual rectus sheath catheters; and a 400 mL elastomeric reservoir of 0.25% levobupivacaine was attached to the paravertebral catheter, and infused at 6–10 mL/hour. LAC catheters were removed routinely on postoperative day 3 once the reservoirs were empty (total infusion: up to 670 mL). Patients undergoing multimodal analgesia and placement of LAC catheters were provided with a patient controlled intravenous fentanyl infusion in the recovery room (PCA) (10–20 mcg fentanyl bolus with 5 minute lockout and a maximum dose of 480 mcg in 4 hours). In patients receiving an epidural, a catheter was placed at about the T7/8 level prior to induction of GA. An initial test bolus of 6 mL 0.25% levobupivacaine was provided and an infusion of 0.125% levobupivacaine with 4 mcg/mL fentanyl infused per and postoperatively at a mean 5 mL/h (range: 3–8 mL/hour), with a 5 mL PCEA bolus facility provided in addition, maximally every 20 minutes. Removal was planned for the fourth postoperative day if pain was well controlled and no epidural related complications encountered prior to this time point. Patients were extubated at the end of the case and monitored on a high dependency unit (HDU) for at least 16 hours before being discharged to a specialist esophagogastric ward. Pain was monitored and analgesia titrated in order to try and obtain a visual analogue pain score ≤3. Discharge from HDU was dependent on hemodynamic and respiratory stability. Patients were reviewed daily by the acute pain service (specialist pain nurse and/or anesthesia specialist registrar/consultant) and additionally, in instances where the pain score was >4. Pain scores and level of block for patients with epidural were recorded by trained nurses on the ward and analgesia was titrated with the aim of maintaining a pain score of less than 4 on movement. Observations and scores were recorded 4 hourly or sooner as clinically dictated. The pain score was reported on a scale from 0 to 10 at both rest and movement. In cases of hypotension (systolic blood pressure <90 mmHg), patients were assessed for contributing factors and the epidural infusion titrated or suspended until hemodynamic stability was re-established. In most cases, fluid boluses of 20 mL/kg were utilized to increase blood pressure and in cases of persistent hypotension with no pulse pressure variability, vasoconstrictor medications (noradrenaline) were commenced and titrated to keep mean arterial pressure >70 mmHg. An ERAS program was established on the unit utilizing an integrated care pathway (ICP) at the beginning of the time period in question. Briefly, the ERAS program was designed to progress attainment of daily goals using the integrated care pathway. The aim is that patients ambulate for a target distance of 300 m at least three times on postoperative days 1 and 2 and increasing the distance as able on postoperative day 3. Data were prospectively recorded using a standardized proforma. Further details on pain scores and fluid administration were obtained by review of the patient notes. Pain scores were recorded by trained nursing staff at rest and movement at least every 4 hours while LAC or TE were in situ or more frequently in case of poorly controlled pain (score ≥4). Ambulation goals were recorded in the ICP document and medication use on the electronic patient record. Complications were assessed twice daily and recorded using the Accordion classification.20 Statistical analysis Continuous variables were compared using unpaired t- or Mann–Whitney U tests as appropriate. Association of categorical variables (differences for dichotomous variables between groups) was assessed using chi-square (Χ2) test. Propensity score matching was used in order to minimize confounding effects due to nonrandomized treatment decisions to compare outcomes for patients treated with neoadjuvant chemoradiation treatment followed by surgery or surgery only versus definitive chemoradiation therapy, both with curative intent. A propensity score was calculated using logistic regression analysis matched on the following variables: age, sex, and American Society of Anesthesiology (ASA) grade. ASA grade was chosen as a measure of comorbidity and it is predictive of mortality and morbidity following esophagectomy.4 Patients who received LAC catheters were matched on a 1:1 ratio to those who received a TE, using the nearest neighbor matching algorithm without replacement on the logit of the propensity score. A caliper of width equal 0.2 SDs of the logit of propensity score was used.21 After propensity score matching, balance of covariates was tested through matching procedures using standardized differences (an absolute standardized difference of <10% was considered a negligible imbalance) and assessment of global imbalance (P = 0.98 after matching).22 RESULTS Demographics Over the two-year study time period, 184 patients underwent esophagectomy with curative intent with 159 undergoing open two-phase transthoracic esophagectomy. Fourteen patients received multimodal analgesia with spinal diamorphine, paravertebral and rectus sheath catheter local anesthesia infusions, and PCA: the remainder (n = 145) received a TE. The demographics of the patients receiving local anesthetic catheters were broadly similar to the overall cohort but in order to analyze outcomes in more details, propensity score matching was used to match the LAC cohort to similar patients receiving a TE (Table 1). The mean ARISCAT risk score for postoperative pulmonary complications for patients in both the LAC and matched cohorts was 50 compared to a mean score of 51.9 for the overall cohort.23 Table 1 Patient demographics and tumor details Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) *P-value for comparison of multimodal LAC versus matched epidural cohort. ASA, American Society of Anesthesiology; BMI, body mass index; LAC, local anesthesia catheter. View Large Table 1 Patient demographics and tumor details Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) Multimodal LAC cohort Epidural cohort Matched epidural cohort P-value* Age 64.7 (5.8) 64.33 (8.6) 63.6 (8.1) 0.181 Sex (male:female) 13:1 108:39 12:2 0.518 ASA 1 0 4 (2.7) 0 0.629 2 8 (57.1) 72 (49.0) 8 (57.1) 3 6 (42.9) 71 (48.3) 6 (42.9) BMI (mean ±SD) 25.98 (5.4) 27.04 (4.9) 27.78 (7.0) 0.634 Neoadjuvant 11 (78.6) 123 (83.7) 9 (64.2) 0.678 pT 0 2 (14.3) 19 (12.9) 2 (14.3) 0.748 1 1 (7.1) 41 (27.9) 6 (42.9) 2 1 (7.1) 18 (12.2) 0 3 8 (57.1) 61 (41.5) 5 (35.7) 4 2 (14.3) 8 (5.4) 1 (7.1) pN 0 7 (50.0) 76 (51.7) 9 (64.3) 0.717 1 5 (35.7) 36 (24.5) 3 (21.4) 2 2 (14.3) 21 (14.3) 0 3 0 13 (8.8) 2 (14.3) *P-value for comparison of multimodal LAC versus matched epidural cohort. ASA, American Society of Anesthesiology; BMI, body mass index; LAC, local anesthesia catheter. View Large Pain Mean and maximum pain scores at rest and movement on postoperative days 0 to 3 were equivalent between the groups (Table 2). Three patients in the matched TE group experienced malfunctioning of their TE (leakage, high pain scores, and no sensory block) leading to earlier than planned removal of the epidural and provision of a morphine PCA. Table 2 Pain scores Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 2 Pain scores Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 Mean VAS score (SD) LAC cohort Matched TE cohort P-value POD 0–rest 0.996 (1.382) 1.74 (2.442) 0.353 POD 0–movement 2.033 (2.26) 2.338 (2.534) 0.753 POD 1–rest 1.173 (1.165) 1.212 (1.543) 0.936 POD 1–movement 3.43 (1.876) 2.926 (2.07) 0.529 POD 2–rest 1.259 (1.418) 1.71 (2.201) 0.545 POD 2-movement 2.9 (1.664) 3.6 (2.842) 0.458 POD 3–rest 0.685 (0.741) 1.057 (1.923) 0.54 POD 3-movement 1.779 (1.138) 2.2 (1.82) 0.509 LAC, local anesthesia catheter; TE, thoracic epidural. View Large The mean cumulative dose received via peripheral venous fentanyl PCA was 3369 mcg (+/−2200) and the mean cumulative dose of fentanyl delivered by patient controlled epidural anesthesia to the matched group was 659 mcg (±326.9). Half of patients in both cohorts experienced a pain score of more than 7 on at least one occasion. The mean number of pain episodes scored greater than 7 was 1.5 (range: 0–5) in the LAC group and 2 (range: 0–11) in the TE group (P = 0.371). Functional recovery There were no statistically significant differences between the proportion of patients requiring vasoconstrictor infusion postoperatively: (LAC: 36% vs. TE: 57%, P = 0.256) to maintain blood pressure; and differences in the frequency of episodes of hypotension (LAC: 43% v TE: 79%, P = 0.053) were not different (Table 3). Fluid administered intraoperatively did not differ between the groups (LAC: 3.7 L ± 1.5 vs. TE: 4.5 L ± 0.96, P = 0.107). Intraoperative blood loss did not differ (LAC: 449 mL ± 233 vs. TE: 475 mL ± 346, P = 0.79). There was an improvement in the ability to ambulate on the first postoperative day (LAC: 64% vs. TE: 43%, P = 0.138) as well as the proportion of patients achieving ambulation goals on POD 1 and 2 (Table 4). Table 3 Hemodynamic outcomes LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 3 Hemodynamic outcomes LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC cohort Matched TE cohort P-value Hypotension 6 (43%) 11 (79%) 0.053 Pressors used POD 0 5 (36%) 8 (57%) 0.256 Pressors used POD 1 2 (14%) 6 (43%) 0.094 Fluid bolus given 7 (50%) 11 (79%) 0.115 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 4 Proportion achieving ambulation goals LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 4 Proportion achieving ambulation goals LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC cohort Matched TE cohort P-value POD 1 8 (57%) 3 (21%) 0.053 POD 2 12 (86%) 8 (57%) 0.094 POD 3 12 (86%) 12 (86%) 1.0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Complications are recorded in Table 5. Two patients in the LAC cohort suffered Accordion grade 4 or more complications: one with idiopathic ARDS after neoadjuvant chemoradiotherapy and the other who developed a stroke and respiratory failure on postoperative day 7. Although more grade 4 or higher complications were experienced in the LAC cohort compared to the matched TE cohort, (LAC: 2/14 vs. TE: 0/14, P = 0.203), the difference was not statistically significant. Table 5 Complications LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large Table 5 Complications LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC cohort Epidural cohort Matched TE cohort P-value (LAC v overall TE cohort) Any complication 10 (71.4) 100 (70.0) 9 (64.3) 0.756 Accordion grade 0 3 (21.3) 57 (39.3) 5 (35.7) 0.932 1 1 (7.1) 8 (5.5) 5 (35.7) 2 5 (35.7) 42 (30.0) 4 (28.6) 3 2 (14.2) 23 (15.9) 0 4 1 (7.1) 12 (8.3) 0 5 1 (7.1) 1 (0.7) 0 6 0 4 (2.8) 0 LAC, local anesthesia catheter; TE, thoracic epidural. View Large The median total length of stay for the overall cohort of 184 patients was 15.53 days (SD: ±15.61). The length of stay for the LAC cohort was 10.77 days (±3.66) versus 11.57 days (=/-2.95) for the matched TE cohort (P = 0.535). The length of stay in critical care was 5.0 days ±7.8 in the overall cohort versus LAC: 3.3 days ± 1.0 and TE: 3.0 days ± 2.1(P = 0.912). DISCUSSION Thoracic epidural currently represents the standard of care for patients undergoing esophagectomy due to its reliable reduction in pain and associated reduction in pulmonary complications. Even following minimally invasive esophagectomy, there is a reduction by almost 50% in postoperative mortality associated with TE24 compared to management with systemic intravenous opioid only. TE is not suitable for all patients and the failure rate may be as high as 25%8,25 and was 21% in this study. There is a risk of epidural hematoma or abscess resulting in permanent spinal cord injury and the risk of nerve damage is not insubstantial and possibly underrecognized.26 In a systematic review of analgesic methods postthoracotomy, seven studies were identified comparing epidural to paravertebral infusion and most (five) reported superior pain outcomes with paravertebral infusion—although there was a large degree of heterogeneity in the pain scores reported between studies.1 The incidence of hypotension seemed to be reduced in the three studies reporting this endpoint. In a meta-analysis of 10 RCTs (n = 520 patients) of patients undergoing thoracotomy, comparing paravertebral (PV) catheters versus TE, PV resulted in comparable analgesia with a better side effect profile, specifically reduced pulmonary complications and less hypotension.27 Based on these data, one-third of UK centers now employ PV catheters postpneumonectomy and observational data support an association with fewer postoperative complications with PV catheters compared to TE.28 The difficulty in extrapolating these results to the postesophagectomy patient includes a greater proportion of patients experiencing complications following esophagectomy compared to other types of major surgery29 and the presence of dual abdominal and thoracic incisions—which may result in more pain and associated functional consequences than thoracotomy alone. Subcostal transversus abdominis plane block (single injection) for open gastrectomy provides improved analgesia compared to IV opioid alone but pain scores with epidural anesthesia are superior at randomized controlled trial.30 The evidence base for the use of LAC catheters following esophagectomy is limited to a single center series using paravertebral catheters in conjunction with epidural anesthesia at T7/8 level to cover the laparotomy component (n = 52). This group was compared to patients who received two epidurals (one at T3/4 and another at T7/8) for analgesia. The frequency of hypotension was significantly reduced in the PVB catheter group versus the two epidural group (9.6% vs. 30.2%, P = 0.008).31 A further single center study of 16 patients from Taiwan utilized single-shot paravertebral and transversus abdominis plane blocks in patients undergoing minimally invasive esophagectomy and reported reduced pain scores and reduced opioid use, compared to patients receiving only intravenous systemic opioid postoperatively.32 Patients without blocks were more likely to report moderate to severe pain in the immediate 2–6 hours postoperatively (for example at 2 hours: 65% vs. 30%, P = 0.049). Thus an alternative to TE anesthesia utilizing multimodal analgesia and local anesthestic delivered via combined paravertebral and rectus sheath catheters may be an attractive proposition for the postesophagectomy patient and worthy of further study. The validity of pain scores as an endpoint is an issue since it is the patient's overall recovery, which is of greatest interest to clinicians and it is not known whether pain scores correlate well with ability to mobilize. Inadequate pain control results in increased complications, prolonged length of stay, and reduced mobility33 but the threshold defining inadequate pain control is not well established. Although large meta-analyses have demonstrated improved pain outcomes with epidural compared to intravenous opioids, the mean difference in visual analogue pain scores varies from 2.2 to 19.9 mm, the clinical relevance of which is difficult to determine.34 Within enhanced recovery programs following open abdominal surgery, at meta-analysis, the use of epidural analgesia is associated with reduced pain scores but there does not seem to be a difference in length of stay and systemic analgesia was associated with fewer complications35 and thus, epidural analgesia is no longer a central component of care. Assessment of functional outcomes such as mobility may be of more relevance to the assessment of analgesia postoperatively. If pain scores remain manageable and do not hinder mobility, we deem this to be clinically equivalent. This study reports for the first time the use of spinal diamorphine, LAC (combined PV and rectus sheath) and PCA for patients following open two-phase transthoracic esophagectomy. In this matched cohort, there are no significant differences in pain scores between the two techniques. The requirement for fluids and pressor support is also an important outcome following esophagectomy as hemodynamic instability limits the ability of patients to ambulate and in the postesophagectomy patient is associated with the development of acute lung injury,36 an important complication of surgery and important contributor to mortality. Current practice aims to judiciously limit fluid administration, aiming to balance organ perfusion, vasopressor requirement, and the avoidance of fluid overload. This could increase the incidence of hypotension and pressor requirement compared to where a liberal fluid policy is employed.37,38 Importantly in this study, there were trends to a reduction in hypotension as well as a reduction in need for vasopressor infusion. This translated in more patients ambulating on the first postoperative day as well as the achieving the desired ambulation goals of the enhanced recovery program. These functional endpoints, rather than absolute pain scores alone would be important for future randomized trials of analgesic modalities in esophagectomy patients. This retrospective study has a number of significant limitations. The overall proportion of patients receiving multimodal LAC infusion technique rather than TE is small—owing mainly to the lack of previous experience and the dearth of an evidence base supporting this practice. There is an unintentional selection bias in the cohort given that the practice began as an alternative to failed epidural. The numbers included are too small to provide absolute proof of any benefit of LAC catheters but provide some evidence on which to power future studies. This study indicates that outcomes such as use of vasoconstrictors or ability to ambulate may be important endpoints for future study. An important issue will be any possible impact of the analgesia techniques on complications, specifically pulmonary complications postesophagectomy. The risk of developing pneumonia after major surgery may have decreased in recent decades. One metaanalysis of epidural analgesia to prevent postoperative pneumonia, reports the number needed to treat (NNT) of 25 from studies in the past 20 years versus 4 for studies from the 1970s.6 Therefore, proxy measures of outcomes may be required for randomized studies rather than overall complication rates. Complications and length of stay were not obviously different but the study underpowered to detect any clear difference. It is important to note that the length of stay is not a complete assessment of recovery given LOS is multifactorial.39 Theoretically, epidural neural modulation may attenuate the postoperative neuroendocrine stress response following esophagectomy,40 although this effect has not been observed in any studies.41 The use of local anesthesia catheters may not have any such effects and are placed at incision closure rather than prior to incision, although the use of intrathecal diamorphine prior to induction, may play a role in neuromodulation. Although epidural from C3 to L1 may reduce the postoperative inflammatory cytokine response following esophagectomy,41 the consequences of a lower TE, which is applicable to clinical practice, and any impact of any immunomodulation are not readily clinically apparent and will require much larger, likely population based study to determine impact on morbidity following surgery. Studies of local anesthesia infusion in hepatic surgery did not show any obvious differences in the inflammatory response compared to epidural.42 This study provides evidence of the feasibility of combining spinal diamorphine with local anesthesia catheters in the paravertebral space and rectus sheaths as an alternative to TE for patients after open transthoracic esophagectomy. It indicates that clinically relevant reductions in analgesia failure, hypotension and requirement of vasopressors may reflect a more favorable side effect profile using a multimodal LAC approach compared to TE and future randomized studies assessing functional recovery are warranted. Notes Disclosures: The authors declare no commercial or other interests. There was no funding sought for this study. Specific author contributions: Design of work: Claire L. Donohoe, Emma Flynn, Claire L. Taylor, Catherine Donnison, Arul Immanuel, Alexander W. Phillips; Data acquisition: Claire L. Donohoe, Emma Flynn, Claire L. Taylor, Catherine Donnison, Arul Immanuel, Alexander W. Phillips; Data analysis and interpretation: Claire L. Donohoe, Alexander W. Phillips; Drafting article: Claire L. Donohoe; Critical revision of article: Alexander W. Phillips, Rhona C. F. Sinclair, David Saunders, S. Michael Griffin. References 1 Joshi G P , Bonnet F , Shah R et al. A systematic review of randomized trials evaluating regional techniques for postthoracotomy analgesia . Anesth Analg 2008 ; 107 : 1026 – 40 . Google Scholar CrossRef Search ADS PubMed 2 Richardson J , Sabanathan S , Shah R . Postthoracotomy spirometric lung function: the effect of analgesia: a review . J Cardiovasc Surg (Torino) 1999 ; 40 : 445 . Google Scholar PubMed 3 Katz J , Jackson M , Kavanagh B P et al. Acute pain after thoracic surgery predicts long-term postthoracotomy pain . Clin J Pain 1996 ; 12 : 50 – 55 . Google Scholar CrossRef Search ADS PubMed 4 Sauvanet A , Mariette C , Thomas P et al. Mortality and morbidity after resection for adenocarcinoma of the gastroesophageal junction: predictive factors . J Am Coll Surg 2005 ; 201 : 253 – 62 . Google Scholar CrossRef Search ADS PubMed 5 Zingg U , Smithers B M , Gotley D C et al. Factors associated with postoperative pulmonary morbidity after esophagectomy for cancer . Ann Surg Oncol 2011 ; 18 : 1460 – 8 . Google Scholar CrossRef Search ADS PubMed 6 Pöpping D M , Elia N , Marret E et al. Protective effects of epidural analgesia on pulmonary complications after abdominal and thoracic surgery . Arch Surg 2008 ; 143 : 990 – 9 . Google Scholar CrossRef Search ADS PubMed 7 Hermanides J , Hollmann M , Stevens M et al. Failed epidural: causes and management . Br J Anaesth 2012 ; 109 : 144 – 54 . Google Scholar CrossRef Search ADS PubMed 8 Viscusi E R . Emerging techniques in the management of acute pain: epidural analgesia . Anesth Analg 2005 ; 101 ( Supplement ): S23 – 9 . Google Scholar CrossRef Search ADS PubMed 9 Al-Rawi O Y , Pennefather S H , Page R D et al. The effect of thoracic epidural bupivacaine and an intravenous adrenaline infusion on gastric tube blood flow during esophagectomy . Anesth Analg 2008 ; 106 : 884 – 7 . Google Scholar CrossRef Search ADS PubMed 10 Pathak D , Pennefather S H , Russell G N et al. Phenylephrine infusion improves blood flow to the stomach during oesophagectomy in the presence of a thoracic epidural analgesia . Eur J Cardiothorac Surg 2013 ; 44 : 130 – 3 . Google Scholar CrossRef Search ADS PubMed 11 Michelet P , Roch A , D’Journo X B et al. Effect of thoracic epidural analgesia on gastric blood flow after oesophagectomy . Acta Anaesthesiol Scand 2007 ; 51 : 587 – 94 . Google Scholar CrossRef Search ADS PubMed 12 Yip V , Dunne D , Samuels S et al. Adherence to early mobilisation: key for successful enhanced recovery after liver resection . Eur J Surg Oncol 2016 ; 42 : 1561 – 7 . Google Scholar CrossRef Search ADS PubMed 13 Scarci M , Joshi A , Attia R . In patients undergoing thoracic surgery is paravertebral block as effective as epidural analgesia for pain management ? Interact Cardiovas Thorac Surg 2010 ; 10 : 92 – 6 . Google Scholar CrossRef Search ADS 14 Sinclair R , Navidi M , Griffin S et al. The impact of neoadjuvant chemotherapy on cardiopulmonary physical fitness in gastro-oesophageal adenocarcinoma . Ann R Coll Surg Engl 2016 ; 98 : 396 – 400 . Google Scholar CrossRef Search ADS PubMed 15 Cunningham D , Allum W H , Stenning S P et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer . N Engl J Med 2006 ; 355 : 11 – 20 . Google Scholar CrossRef Search ADS PubMed 16 van Hagen P , Hulshof M C C M , van Lanschot J J B et al. Preoperative chemoradiotherapy for esophageal or junctional cancer . N Engl J Med 2012 ; 366 : 2074 – 84 . Google Scholar CrossRef Search ADS PubMed 17 Phillips A W , Dent B , Navidi M et al. Trainee involvement in Ivor–Lewis esophagectomy does not negatively impact outcomes . Ann Surg 2018 ; 267 : 94 – 98 . Google Scholar CrossRef Search ADS PubMed 18 Royal College of Anaesthetists . Epidural pain relief after surgery: patient information leaflet . 4TH edition , 2014 London . Accessed at www.rcoa.ac.uk/patientinfo . 19 Tighe S , Greene M D , Rajadurai N . Paravertebral block . Contin Educ Anaesth Crit Care Pain 2010 ; 10 : 133 – 7 : mkq029 . Google Scholar CrossRef Search ADS 20 Strasberg S M , Linehan D C , Hawkins W G . The accordion severity grading system of surgical complications . Ann Surg 2009 ; 250 : 177 – 86 . Google Scholar CrossRef Search ADS PubMed 21 Austin P C . Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies . Pharm Stat 2011 ; 10 : 150 – 61 . Google Scholar CrossRef Search ADS PubMed 22 Bowers J , Fredrickson M , Hansen B . RItools: Randomization inference tools . R Package, Version 01-11 , Vienna, Austria : R. Foundation for Statistical Computing , 2010 . 23 Mazo V , Sabaté S , Canet J et al. Prospective external validation of a predictive score for postoperative pulmonary complications . Anesthesiology 2014 ; 121 : 219 – 31 . Google Scholar CrossRef Search ADS PubMed 24 Zingg U , McQuinn A , DiValentino D et al. Minimally invasive versus open esophagectomy for patients with esophageal cancer . Ann Thorac Surg 2009 ; 87 : 911 – 9 . Google Scholar CrossRef Search ADS PubMed 25 Ready L B . Acute pain: lessons learned from 25,000 patients . Reg Anesth Pain Med 1999 ; 24 : 499 – 505 . Google Scholar PubMed 26 Moen V , Dahlgren N , Irestedt L . Severe neurological complications after central neuraxial blockades in Sweden 1990–1999 . J Am Soc Anesthesiol 2004 ; 101 : 950 – 9 . Google Scholar CrossRef Search ADS 27 Davies R , Myles P , Graham J . A comparison of the analgesic efficacy and side-effects of paravertebral vs epidural blockade for thoracotomy—a systematic review and meta-analysis of randomized trials . Br J Anaesth 2006 ; 96 : 418 – 26 . Google Scholar CrossRef Search ADS PubMed 28 Powell E , Cook D , Pearce A et al. A prospective, multicentre, observational cohort study of analgesia and outcome after pneumonectomy . Br J Anaesth 2011 ; 106 : 364 – 70 . Google Scholar CrossRef Search ADS PubMed 29 Merkow R P , Bilimoria K Y , McCarter M D et al. Short-term outcomes after esophagectomy at 164 American College of Surgeons National Surgical Quality Improvement Program Hospitals . Arch Surg 2012 ; 147 : 1009 – 16 . Google Scholar CrossRef Search ADS PubMed 30 Wu Y , Liu F , Tang H et al. The analgesic efficacy of subcostal transversus abdominis plane block compared with thoracic epidural analgesia and intravenous opioid analgesia after radical gastrectomy . Anesth Analg 2013 ; 117 : 507 – 13 . Google Scholar CrossRef Search ADS PubMed 31 Niwa Y , Koike M , Torii K et al. Combination of continuous paravertebral block and epidural anesthesia in postoperative pain control after esophagectomy . Esophagus 2016 ; 13 : 42 – 47 . Google Scholar CrossRef Search ADS 32 Li N-L , Liu C-C , Cheng S H-C et al. Feasibility of combined paravertebral block and subcostal transversus abdominis plane block in postoperative pain control after minimally invasive esophagectomy . Acta Anaesthesiol Taiwan 2013 ; 51 : 103 – 7 . Google Scholar CrossRef Search ADS PubMed 33 Wu C L , Fleisher L A . Outcomes research in regional anesthesia and analgesia . Anesth Analg 2000 ; 91 : 1232 – 42 . Google Scholar PubMed 34 Block B M , Liu S S , Rowlingson A J et al. Efficacy of postoperative epidural analgesia . JAMA 2003 ; 290 : 2455 – 63 . Google Scholar CrossRef Search ADS PubMed 35 Hughes M J , Ventham N T , McNally S et al. Analgesia after open abdominal surgery in the setting of enhanced recovery surgery . JAMA Surg 2014 ; 149 : 1224 – 30 . Google Scholar CrossRef Search ADS PubMed 36 Tandon S , Batchelor A , Bullock R et al. Peri-operative risk factors for acute lung injury after elective oesophagectomy . Br J Anaesth 2001 ; 86 : 633 – 8 . Google Scholar CrossRef Search ADS PubMed 37 Chau E H L , Slinger P . Perioperative fluid management for pulmonary resection surgery and esophagectomy . Seminars in cardiothoracic and vascular anesthesia , Vol. 18 , Los Angeles, CA : SAGE Publications , 2014 ; pp. 36 – 44 . Google Scholar CrossRef Search ADS 38 Neal J M , Wilcox R T , Allen H W et al. Near-total esophagectomy: the influence of standardized multimodal management and intraoperative fluid restriction 1, 2 . Reg Anesth Pain Med 2003 ; 28 : 328 – 34 . Google Scholar PubMed 39 Maessen J , Dejong C , Hausel J et al. A protocol is not enough to implement an enhanced recovery programme for colorectal resection . Br J Surg 2007 ; 94 : 224 – 31 . Google Scholar CrossRef Search ADS PubMed 40 Ahlers O , Nachtigall I , Lenze J et al. Intraoperative thoracic epidural anaesthesia attenuates stress-induced immunosuppression in patients undergoing major abdominal surgery† . Br J Anaesth 2008 ; 101 : 781 – 7 . Google Scholar CrossRef Search ADS PubMed 41 Yokoyama M , Itano Y , Katayama H et al. The effects of continuous epidural anesthesia and analgesia on stress response and immune function in patients undergoing radical esophagectomy . Anesth Analg 2005 ; 101 : 1521 – 7 . Google Scholar CrossRef Search ADS PubMed 42 Hughes M , Harrison E , Peel N et al. Randomized clinical trial of perioperative nerve block and continuous local anaesthetic infiltration via wound catheter versus epidural analgesia in open liver resection (LIVER 2 trial) . Br J Surg 2015 ; 102 : 1619 – 28 . Google Scholar CrossRef Search ADS PubMed SUPPLEMENTARY DATA Supplementary data are available at DOTESO online. Supplementary Table 1: Patient selection factors for paravertebral/rectus sheath catheters © The Author(s) 2018. Published by Oxford University Press on behalf of International Society for Diseases of the Esophagus. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Diseases of the EsophagusOxford University Press

Published: May 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off