FUNCTION AFTER OPEN ABDOMINAL AORTIC POSOPERATIVE PULMONARY ANEURYSM REPAIR IN PATIENTS WITH CHRONIC OBSTRUCTIVE PULMONARY DISEASE: EPIDURAL VERSUS INTRAVENOUS ANALGESIA
Clinical Research Postoperative Pulmonary Function After Open Abdominal Aortic Aneurysm Repair in Patients With Chronic Obstructive Pulmonary Disease: Epidural Versus Intravenous Analgesia Venetiana Panaretou,1 Levon Toufektzian,2 Ioanna Siafaka,3 Irene Kouroukli,1 Fragiska Sigala,2 Charalambos Vlachopoulos,4 Stilianos Katsaragakis,2 George Zografos,2 and Konstantinos Filis,2 Athens, Greece Background: We reviewed our experience to determine the effect of epidural versus intrave- nous analgesia on postoperative pulmonary function and pain control in patients with chronic obstructive pulmonary disease (COPD) undergoing open surgery for abdominal aortic aneurysm. Methods: A retrospective study with prospective collection of data of 30 COPD patients under- going open abdominal aortic aneurysm repair, during a 5-year period. Group I (n ¼ 16) was oper- ated under combined general and epidural anesthesia and epidural analgesia; group II (n ¼ 14), under general anesthesia and intravenous analgesia. All patients performed pulmonary function tests (PFTs) preoperatively and during postoperative days 1 and 4. Pain assessment was per- formed on all patients during rest and activity on postoperative days 1, 2, and 4 by using the visual analog scale. Data were recorded for PFTs, postoperative pain, length of hospital stay, length of ICU stay, and postoperative pulmonary morbidity, including atelectasis and pulmonary infections. Results: There was no in-hospital mortality. Hospital stay was similar between the two groups (group I: 7.1 ± 1.0, group II: 7.5 ± 1.1). Group I patients showed signiﬁcantly increased postoperative PFT values compared with group II patients at all time points (post- operative day 1: FEV1(%): 32.3 ± 4.4 vs. 27.1 ± 1.6, p ¼ 0.007, FVC(%): 35.4 ± 8,5 vs. 28.3 ± 2.3, p ¼ 0.035; postoperative day 4: FEV1(%): 50.4 ± 6.8 vs. 41.9 ± 6.8, p ¼ 0.017, FVC(%): 51.3 ± 8.3 vs. 43.0 ± 7.9, p ¼ 0.046). However, postoperative clinical pulmonary morbidity was not different between groups. Group I patients showed signiﬁ- cantly reduced postoperative pain at all time points compared with group II patients. These differences were more pronounced during postoperative days 1 and 2, both at rest (visual analog score: 1.1 ± 0.9 vs. 2.6 ± 1.6, p ¼ 0.02 and 0.7 ± 0.8 vs. 1.9 ± 1.1, p ¼ 0.021, respectively) and during activity (2.3 ± 0.8 vs. 4.0 ± 1.7, p ¼ 0.013 and 1.6 ± 0.7 vs. 2.8 ± 1.2, p ¼ 0.019, respectively). 1 Department of Anesthesiology, Hippokration Hospital, Athens, Greece. Correspondence to: Konstantinos Filis, MD, PhD, Division of 2 Division of Vascular Surgery, First Department of Propeudetic Vascular Surgery, First Department of Propeudetic Surgery, Hippokra-Surgery, Hippokration Hospital, Medical School, University of Athens, tion Hospital, Medical School, University of Athens, 34B FaneromenisAthens, Greece. Street, 15562 Athens, Greece; E-mail: kﬁlis@hotmail.com 3 Department of Anesthesiology, Aretaieio Hospital, Medical School, Ann Vasc Surg 2012; 26: 149–155University of Athens, Athens, Greece. DOI: 10.1016/j.avsg.2011.04.009 4 Ó Annals of Vascular Surgery Inc. First Department of Cardiology, Hippokration Hospital, Medical Published online: October 24, 2011School, University of Athens, Athens, Greece. 149
150 Panaretou et al. Annals of Vascular Surgery Conclusions: Epidural anesthesia and postoperative epidural analgesia improve the postoper- ative respiratory function, compared with general anesthesia and systemic analgesia, and reduce postoperative pain as well, in COPD patients undergoing elective infrarenal abdominal aortic aneurysm repair.INTRODUCTION opted to assess the effect on pain management and postoperative pulmonary function in this high-riskAlthough various developments and reﬁnements in group of patients.surgical techniques and anesthetic approaches havereduced perioperative risk, abdominal aortic aneu-rysm (AAA) open repair in patients with severechronic obstructive pulmonary disease (COPD) METHODScontinues to be a high-risk procedure associated Study Groupwith life-threatening complications and increasedperioperative morbidity.1 In addition, COPD has Among 110 patients electively operated for anbeen associated with an increased prevalence of infrarenal AAA (max diameter> 55 mm), duringAAA2,3 and is an independent factor for AAA a 5-year period (2004-2009), we reviewed 30rupture.4 Obviously, endovascular AAA repair has patients with severe COPD. Operations were per-been now established as the treatment of choice formed by the standard transperitoneal exposure,for this high-risk group of patients. However, endo- by the same vascular surgical and anesthesiavascular repair is feasible in less than 60% of teams. None of these patients fulﬁlled criteria forpatients5; therefore, a number of AAA patients endovascular aneurysm repair (in 18 cases due towith COPD will still remain under medical observa- short aortic neck, in 12 due to iliac access prob-tion, waiting to undergo open AAA repair until it lems). We excluded emergency cases, patientsbecomes feasible in their case. Recent studies have with severely impaired left ventricular systolicshown that perioperative morbidity and mortality function with ejection fraction <30%, patientsas well as long-term survival of these patients justify with a history of cerebrovascular accident or ofopen AAA repair,6 although different clinical renal and hepatic insufﬁciency, and patientspredictors (mainly cardiac and renal disease) of allergic to levobupivacaine.severe morbidity and an unfavorable outcome Pulmonary disease was diagnosed by clinicalexist.7 history and PFTs. Smoking history was present in A vast amount of literature underscores the bene- all patients. PFTs, including forced vital capacityﬁcial effects of epidural anesthesia and analgesia in (FVC) and forced expiratory volume in 1 secondimproving pain management and enhancing post- (FEV1), were measured preoperatively in alloperative recovery in comparison with general patients by using a portable spirometer unit (Spiro-anesthesia alone.8e10 Epidural anesthesia and post- lab II, SDI Diagnostics, Rome, Italy). With a goodoperative epidural analgesia attenuate reductions in ﬂow-volume loop, each measurement was per-respiratory parameters during the postoperative formed three times, and the best trial was recorded.period, as compared with systemic analgesia in The diagnosis of moderate to severe COPD waspatients undergoing abdominal vascular proce- deﬁned by the presence of one or more of thedures.11 Although epidural anesthesia has been following criteria: PaCO2 >45 mmHg, FEV1 <70%the common practice for open AAA repair, this prac- of predicted value, and FEV1/FVC <0.7, as thetice comes largely from experience in abdominal same limits have been used by a network of interna-and thoracic surgery. Thus, it has not been tional experts.12 Patients were assigned to receiveadequately studied in COPD patients undergoing either combined general and epidural anesthesiaopen AAA repair in terms of postoperative respira- and epidural postoperative analgesia (group I, n ¼tory function, as evidenced by alterations in pulmo- 16), or general anesthesia alone with postoperativenary function tests (PFTs). patient-controlled intravenous systemic analgesia The objective of this study was to compare respi- (group II, n ¼ 14). Preoperatively, all patientsratory effects of combined general and epidural received bronchodilator treatment with salbutamolanesthesia and postoperative epidural analgesia for 1 week and had preconditioning regularwith those of only general anesthesia and postoper- breathing exercises. All previous pulmonary medi-ative systemic analgesia in COPD patients under- cation had been discontinued at least 1 week beforegoing open AAA surgical repair. Additionally, we hospital admission.
Vol. 26, No. 2, February 2012 Pulmonary function after AAA open repair: epidural vs intravenous analgesia 151 The study was approved by the institutional and data for drugs and material were obtainedethics committee, and all patients had given from the hospital administration and from theinformed consent regarding the procedural risks hospital pharmacy. Postoperative PFTs were per-and outcomes. formed on all patients during postoperative days 1 and 4. For each patient, postoperative pain was assessed at rest and during movement. Pain scoresAnesthetic, Surgical, and Postoperative were assessed during postoperative days 1, 2, andProcedures 4 by using the visual analog scale rating fromFor the standard preoperative evaluation (including 0 (no pain at all) to 10 (worst possible pain). Compli-cardiac echo stress test, PFTs, bowel preparation, cations related to the epidurally administeredand clinical consultation with the anesthesiologist), opioids, including pruritus, nausea, vomiting, motorall patients were admitted to the hospital 3-4 days block, and respiratory depression, were recorded.before surgery. The afternoon before surgery, an Length of ICU and hospital stay and sequelae ofepidural catheter was inserted at L1-L2 or L2-L3 postoperative pulmonary morbidity, including atel-level in group I patients. The evening before ectasis and pulmonary infections, were alsosurgery, all patients received 1.5 mg of oral broma- recorded.zepam for premedication. After entering the oper-ating room, all patients received 2.5 mg of Statistical Analysisintravenous midazolam. Induction of general anes-thesia was achieved with 1 mg/kg intravenous remi- Results for continuous variables were expressed asfentanyl, 2 mg/kg propofol, and 0.6 mg/kg mean and standard deviation (SD). Categoric vari-rocuronium. Group I patients received 10-15 mL ables were presented as numbers. Continuous andof levobupivacaine 0.125% and 100 mg of fentanyl categoric data were compared by the Student t testvia the epidural catheter. Maintenance of anesthesia and 1-way analysis of variance. Probability ( p)in group I patients was achieved with a mixture of values less than 0.05 were considered statisticallysevoﬂurane, oxygen, and cisatracurium and a basal important.infusion of epidural levobupivacaine 0.125% and5 mg/mL fentanyl at a rate of 4-6 mL/hr. Mainte- RESULTSnance of anesthesia in group II patients wasachieved with a mixture of sevoﬂurane, oxygen, The two groups exhibited no signiﬁcant differencesremifentanyl, and cisatracurium. All patients were in terms of demographic data, patient characteris-managed with the same standardized protocols aim- tics, perioperative variables, and length of postoper-ing for early extubation immediately after the ative hospital stay, which are presented in Table I.surgical procedure and were transferred to the ICU Transfusion requirements were comparablefor overnight hemodynamic monitoring. Sixteen between the two groups. However, patients ofpatients were operated using a straight prosthesis; combined general and epidural anesthesia groupand 14, using a bifurcated knitted Dacron prosthesis. were administered additional ﬂuid infusions and During the ﬁrst 5 postoperative days, group I vasopressors (phenlyephrine) to counterbalancepatients received epidural analgesia with a mixture peripheral vasodilation and arterial pressure dropof levobupivacaine 0.125% and 5 mg/mL fentanyl due to epidural anesthesia and to maintain a stablewith a basal rate of 3-5 mL/hr, bolus doses of 3-5 hemodynamic proﬁle (mean arterial pressure abovemL, and a lockout interval of 20 minutes. Postoper- 60 mm Hg or systolic blood pressure above 90 mmative analgesia in group II patients, during the same Hg) throughout the entire surgical procedure.period, was achieved with intravenous patient- Preoperatively, values of FEV1 and FVC were notcontrolled mode of application of 10 mg/mL fentanyl different between groups. During postoperativewith a basal rate of 1.5-2.5 mL/hr, bolus doses of days 1 and 4, all patients (both groups) demonstrated1-2.5 mL, and a lockout interval of 6 minutes. After reductions in both parameters. However, thesepostoperative day 5, all patients received oral anal- reductions were signiﬁcantly less pronounced ingesics as required. The same attending and the group I patients (combined general and epiduralsame resident anesthesiologist were responsible for anesthesia) compared with group II patients (generalepidural catheter insertion, pain management, anesthesia only). Group I patients showed signiﬁ-epidural doses and complications, as well as catheter cantly increased postoperative PFT values comparedremoval. The actual cost for staff and pharmacy with group II patients at all time points (postopera-charges, as well as the cost for material and drugs, tive day 1: FEV1(%): 32.3 ± 4.4 vs. 27.1 ± 1.6, p ¼was calculated. Data on cost per hour for the staff 0.007, FVC(%): 35.4 ± 8.5 vs. 28.3 ± 2.3, p ¼ 0.035;
152 Panaretou et al. Annals of Vascular SurgeryTable I. Demographics, patient comorbidities, intraoperative variables, and length of hospital stay;continuous variables are expressed as mean ± standard deviation (SD)Variables Group I (n ¼ 16) Group II (n ¼ 14) p valueAge (years) 68.7 ± 5.9 70.0 ± 6.2 n/sSex (male/female) 16/0 14/0 n/sWeight (kg) 79.3 ± 9.4 76.1 ± 6.3 n/sHeight (cm) 172.9 ± 5.3 173.1 ± 5.6 n/sHypertension 13/16 11/14 n/sSmoking history 16/16 12/14 n/sDiabetes mellitus 4/16 2/14 n/sHyperlipidemia 7/16 6/14 n/sPrevious MI 4/16 3/14 n/sCoronary artery disease 3/16 4/14 n/sAortic cross-clamping duration (minutes) 65.7 ± 5.8 64.6 ± 7.9 n/sDuration of surgery (minutes) 223.2 ± 20.8 238.9 ± 19.5 n/sPerioperative blood transfusion (units) 2.8 ± 1.0 3.3 ± 0.7 n/sLength of hospital stay (days) 7.1 ± 1.0 7.5 ± 1.1 n/sn/s, not signiﬁcant.Table II. Pulmonary function test values as Table III. Postoperative pain assessed at rest andpercentage of predicted values before and after during activity by visual analog scale (VAS);the surgical procedure; values are expressed as values are expressed as mean ± standard deviationmean ± standard deviation Group I Group II Group I Group II Postoperative day (n ¼ 16) (n ¼ 14) p valueVariable (n ¼ 16) (n ¼ 14) p value VAS at restFVC (%) Day 1 1.1 ± 0.9 2.6 ± 1.6 0.020 Preoperative 66.2 ± 12 65.7 ± 4.6 0.922 Day 2 0.7 ± 0.8 1.9 ± 1.1 0.021 Postoperative day 1 35.4 ± 8.5 28.3 ± 2.3 0.035 Day 4 0.5 ± 0.5 1.4 ± 1.1 0.036 Postoperative day 4 51.3 ± 8.3 43.0 ± 7.9 0.046 VAS on activityFEV1 (%) Day 1 2.3 ± 0.8 4.0 ± 1.7 0.013 Preoperative 60.1 ± 10.4 61.3 ± 5.7 0.784 Day 2 1.6 ± 0.7 2.8 ± 1.2 0.019 Postoperative day 1 32.3 ± 4.4 27.1 ± 1.6 0.007 Day 4 0.8 ± 1.1 2.0 ± 1.1 0.036 Postoperative day 4 50.4 ± 6.8 41.9 ± 6.8 0.017postoperative day 4: FEV1(%): 50.4 ± 6.8 vs. 41.9 ± In group I, complications related to epidural or6.8, p ¼ 0.017, FVC(%): 51.3 ± 8.3 vs. 43.0 ± 7.9, intravenous analgesia included pruritus (2 patients)p ¼ 0.046) (Table II). None of the patients in both and peripheral motor block (1 patient); and in groupgroups developed toxic symptoms related to II, pruritus (1 patient), nausea (1 patient), andlevobupivacaine. persistent vomiting (1 patient). Regarding return Group I patients (epidural analgesia) showed of bowel function, bowel movements were evidentsigniﬁcantly decreased postoperative pain at all from postoperative day 2 or 3 in both groups. Alltime points either while being at rest ( p < 0.05) or patients were put on liquids on postoperative daywhile doing some activity ( p < 0.05), compared 3. Of 16 patients from the epidural group, ﬁvewith group II patients (systemic analgesia). Addi- suffered from nausea and vomiting (after liquidtionally, the differences in postoperative pain levels diet), delaying the start of regular diet from postop-between the two groups were more pronounced erative day 5 to postoperative day 6. In the intrave-during activity and less pronounced by postopera- nous analgesia group, liquid diet initiation on day 3tive day 4. Table III presents postoperative pain did not result in nausea or vomiting in any patient,levels at rest and during activity according to visual although one patient had nausea and one experi-analog scale at days 1, 2, and 4 after the surgical enced vomiting unrelated to liquid or foodprocedure. Epidural catheters were removed on consumption in this group. The actual mean timepostoperative day 5. required for the patient to be on regular diet was
Vol. 26, No. 2, February 2012 Pulmonary function after AAA open repair: epidural vs intravenous analgesia 153Table IV. Postoperative pulmonary Table V. The estimation of cost in terms of costcomplications and length of ICU stay; prolonged for staff, pharmacy charges, and costs for drugs,intubation deﬁned as inability of extubation calculated for each patient and for a 4-dayimmediately after the surgical procedure; treatmenthypoxemia deﬁned as PaO2 <60 mm Hg Group I Group II Group I Group II Costs (V) (V)Variable (n ¼ 16) (n ¼ 14) p value Personnel cost for applying the 290 180Prolonged intubation 0/10 1/14 n/s methodICU stay >24 hours 1/16 1/14 n/s Personnel cost for intervention for 250 250Atelectasis 4/16 4/14 n/s pain relief and pulmonaryPulmonary infection 0/16 2/14 n/s function tests Personnel cost for minor side 95 65 effects/technical problems/ changing treatment strategy4.7 days for the epidural group compared with 3.5 Pharmacy charges and cost for 169 72days for the intravenous analgesia group. material In group I, all patients were transferred extubated Personnel cost for catheter removal 30 ein the ICU. One patient in group II could not be Total postoperative care 824 567extubated immediately after surgery and requiredadditional mechanical ventilatory support untilpostoperative day 1, resulting in a 2-day stay in of AAAs have also demonstrated improvement inthe ICU. This patient, however, was not excluded respiratory status with the use of epidural anesthesiafrom pain assessment and PFTs during postoperative in similar settings. However, these studies empha-day 1. In group I, four patients developed atelectasis sized the beneﬁts of epidural anesthesia withoutwith no evidence of infection. In group II, four general anesthesia or in combination with mini-patients developed postoperative atelectasis, which mally invasive surgical approaches. Besides,in two of them advanced to pulmonary infection although there was a signiﬁcant beneﬁt in terms of(Table IV). postoperative pulmonary morbidity, improvement Regarding the economic analysis, the most rele- of pulmonary function was not demonstrated withvant medical costs were estimated: the cost of PFTs.14,15human resources as well as pharmacy charges and In our study, we showed a beneﬁt of postopera-the cost of drugs and material (Table V). Epidural tive PFTs, (FVC and FEV1) on postoperative days 1analgesia resulted in a cost of V824 for each patient and 4, in COPD patients operated under combinedfor the 4-day care, compared with V567 for intrave- general and epidural anesthesia and postoperativenous analgesia (50% increased cost). epidural analgesia compared with COPD patients All patients had an uneventful surgical recovery, with general anesthesia and intravenous analgesia.with no reoperations or postoperative blood transfu- In addition, the assessment of postoperative painsions. There was no in-hospital mortality. during rest and activity revealed reduced levels of pain at all time points in patients who receivedDISCUSSION epidural analgesia compared with those who received systemic analgesia. The difference in painDecision of elective open AAA repair depends on the levels between the two groups during activityestimate of the risks and beneﬁts involved. The difﬁ- underscores the important contribution of epiduralculty in making a decision stems from the necessity analgesia in patient mobilization.to determine whether a high-risk COPD patient Although the statistical differences lack powerwould beneﬁt from the open operation, when the because of the limited number of patients includedendovascular approach is not feasible. Previous in the study, the results parallel the documentedprospective studies have shown that epidural anes- effects of epidural anesthesia in larger studies onthesia offers many beneﬁts during major abdominal abdominal vascular operations. Major centers treat-vascular surgery, including inhibition of the surgical ing AAA patients have already adopted, as a part ofstress response, fewer pulmonary complications, their clinical pathways, the strategy of administeringand better postoperative pain control, in comparison combined general and epidural anesthesia to theirwith patient-controlled systemic analgesia.13,14 COPD patients,6,7 owing to previously known posi-Additionally, studies on high-risk pulmonary tive effects of combined general and epidural anes-compromised patients undergoing reconstruction thesia on the overall outcome. By reporting the
154 Panaretou et al. Annals of Vascular Surgeryﬁndings of postoperative lung function tests, our been estimated in our hospital to be V9,000-study reconﬁrms the speciﬁc positive beneﬁt of V12,500 (uncomplicated cases with 9-10-daythis clinical practice for COPDeAAA surgical length of stay and 1-2 days in the ICU). In general,patients. However, postoperative increased PFT the cost of epidural analgesia has been evaluatedvalues did not altered the overall clinical outcome. to be up to two to three times the cost of intravenousPostoperative atelectases were equal between analgesia, although the cost has been deemedgroups; however, pulmonary infections were reasonable in relation to the beneﬁt of pain control.evident only in two of the patients in group II. In our study, we showed, except for the clinicalThis lack of superiority of clinical complications in beneﬁt in pain management, a clearly improvedgroup I, compared with group II, is probably owing pulmonary function in a highly compromisedto the small number of patients included in the patient group. Improvement in pulmonary functionstudy. Unfortunately, the feasibility of a large decreases the possibility of pulmonary complica-single-center study is questionable because endo- tions and the potential cost associated with theirvascular repair is the surgical treatment of choice treatment.for these patients. However, the beneﬁt of improved The use of epidural catheters for 4 days waspulmonary function is mandatory, as major vascular a protocol decision, based on the possible risks andsurgery in COPD patients may be followed by pro- beneﬁts and the experience expressed in previouslonged intubation or reintubation in the ICU. These studies.16,17 In a previous study on postoperativepossible complications are likely to be prevented by pulmonary complications, the authors reportedpulmonary function improvement. their results after 5 days of epidural analgesia in Postoperative lung function after major abdom- patients who had undergone major abdominalinal surgery can be impaired with marked reduc- surgery.18 Although most of the recent studiestions in VC and FEV1 lasting up to 1 month follow a protocol of 3 days of postoperative painpostoperatively. Perhaps, the most profound effect management, it was our decisiondbased on anof major abdominal and thoracic surgery on pulmo- extensive personal experience with postoperativenary function is due to diaphragmatic dysfunction, pulmonary complications after major abdominaldecreased chest wall compliance, and pain-limited vascular surgery19dto evaluate lung functioninspiration. Although factors other than pain have improvement after 1 and 4 days of treatment.been implicated in the development of postopera- However, the analgesic methods used in this studytive respiratory dysfunction, such as reﬂex inhibi- have been thoroughly evaluated and accepted fortion of phrenic nerve on diaphragmatic activation their safety by numerous previous studies.13e15 Pro-and residual effects of general anesthetics, it is longed duration of epidural analgesia for more thanwidely assumed that when postoperative patients 5 days has been reported to be followed by increasedare relatively pain-free, their pulmonary function probability of complications, whereas shorter dura-is improved as a result of enhanced chest expansion, tion has been inadequate for pain relief.17better breathing pattern, increased cough, Complications in our study regarding the use ofdecreased sputum retention, and cooperation with epidural or intravenous analgesia were mild and tran-physical therapy. The improvement of postoperative sient. A slight prolongation of normal bowel functionlung function can be explained by a direct beneﬁcial return was evident in the epidural group comparedeffect of epidural analgesia on diaphragmatic with the intravenous group. However, major compli-contractility and breathing pattern, as well as cations were absent in both groups. The improvementa signiﬁcant superiority of epidural analgesia on in lung function after 4 days of postoperative treat-postoperative pain relief compared with a patient- ment with epidural analgesia overcomes the possi-controlled mode of application of systemic bility of postoperative complications and outweighsanalgesia.13 the relatively low increase in cost. The same attending and the same resident anes- Our study included open AAA repair only bythesiologistdfrom the division of pain service- transperitoneal aneurysm exposure because of thedwere responsible for epidural analgesia during team’s preference and for reasons of comparabilitythe four postoperative days. Estimated difference between groups. Other researchers have proposedin ﬁnancial cost resulted in an increase of 50% using a retroperitoneal approach for repairingowing to use of epidural analgesia (cost: V824/ AAA, but that approach results in increased postop-patient) instead of intravenous analgesia cost: erative pain and questionable reduction of postoper-V567/patient). The actual difference of V257/ ative pulmonary morbidity.20patient may be considered low compared with the Pulmonary disease has been shown to increasemean total cost of open AAA repair, which has rupture risk of AAA at a given diameter.21,22
Vol. 26, No. 2, February 2012 Pulmonary function after AAA open repair: epidural vs intravenous analgesia 155Difﬁculties arise in treatment of speciﬁc COPD aneurysmectomy in patients with chronic obstructivepatients, as some studies have shown increased pulmonary disease. J Vasc Surg 2003;37:594e9. 8. Liu S, Carpenter RL, Neal JM. Epidural anesthesia and anal-mortality and morbidity after AAA repair.23,24 The gesia. Their role in postoperative outcome. Anesthesiologyperceived increased risk of surgery in patients’ 1995;82:1474e506.consideration may outweigh the increased risk of 9. Ballantyne JC, Carr DB, deFerranti S, et al. The comparativeaneurysm rupture. Ultimately, the risk of interven- effects of postoperative analgesic therapies on pulmonarytion depends mainly on the cardiac and renal outcome: cumulative meta-analyses of randomized, controlled trials. Anesth Analg 1998;86:598e612.comorbidities; so, while making the ﬁnal decision, 10. Benzon HT, Wong HY, Belavic AM Jr, et al. A randomizedone must consider other aspects apart from the double-blind comparison of epidural fentanyl infusionCOPD factor alone. versus patient-controlled analgesia with morphine for post- Our current study and another previous report horacotomy pain. Anesth Analg 1993;76:316e22.show a low mortality and morbidity in COPD 11. Safwat AM. Epidural anesthesia is a valuable adjunct to general anesthesia for abdominal vascular surgery. J Cardi-patients after open AAA repair.24 Although endo- othorac Anesth 1989;3:505e8.vascular repair is the treatment of choice, COPD 12. Rabe KF, Hurd S, Anzueto A, et al. Global strategy for thepatients can successfully undergo open AAA repair; diagnosis, management, and prevention of chronic obstruc-however, careful attention to additional risk factors tive pulmonary disease: GOLD executive summary. Am Jis essential for overall risk modiﬁcation. Epidural Respir Crit Care Med 2007;176:532e55. 13. Wu CL, Cohen SR, Richman JM, et al. Efﬁcacy of postoper-anesthesia with postoperative epidural analgesia ative patient-controlled and continuous infusion epiduraldecreases the immediate postoperative pulmonary analgesia versus intravenous patient-controlled analgesiadysfunction compared with systemic analgesia, with opioids: a meta-analysis. Anesthesiology 2005;103:and is justiﬁed to be routinely used in all COPD 1079e88.patients operated for AAA. However, possible clin- 14. P€ pping DM, Elia N, Marret E, Remy C, Tramr MR. Protec- o e tive effects of epidural analgesia on pulmonary complicationsical beneﬁt concerning better pulmonary outcome after abdominal and thoracic surgery: a meta-analysis. Archneeds to be evaluated in larger prospective random- Surg 2008;143:990e9.ized studies. 15. McGregor WE, Koler AJ, Labat GC, et al. Awake aortic aneurysm repair in patients with severe pulmonary disease. Am J Surg 1999;178:121e4. 16. Kindler CH, Seeberger MD, Staender SE. Epidural abscessThis study was funded by a grant from the National and complicating epidural anesthesia and analgesia: an analysisKapodistrian University of Athens, Medical School No. of the literature. Acta Anaesthesiol Scand 1998;42:614e20.70/48112. 17. Sicard GA, Reilly JM, Rubin BG, et al. Transabdominal versus retroperitoneal incision for abdominal aortic surgery: report of a prospective randomized trial. J Vasc SurgREFERENCES 1995;21:174e81. 18. Jayr C, Thomas H, Rey A, Farhat F, Lasser P, Bourgain JL. 1. Kroenke K, Lawrence VA, Theroux JF, Tuley MR, Postoperative pulmonary complications. Epidural analgesia Hilsenbeck S. Postoperative complications after thoracic using bupivacaine and opiods versus parenteral opiods. and major abdominal surgery in patients with and without Anesthesiology 1993;78:666e76. obstructive lung disease. Chest 1993;104:1445e51. 19. Bastounis E, Filis K, Georgopoulos S, Bakoyannis C, 2. Van Laarhoven CJ, Borstlap AC, VanBerge-Hegwegouwen DP, Xeromeritis N, Papalambros E. Selective use of the intensive Palmen FM, Verpalen MC, Schoemaker MC. Chronic obstruc- care unit after elective infrarenal abdominal aortic aneu- tive pulmonary disease and abdominal aortic aneurysms. Eur J rysm repair. Int Angiol 2003;22:308e16. Vasc Surg 1993;7:386e90. 20. Cronenwett JL, Murphy TF, Zelenock GB, et al. Actuarial 3. Bengston H, Bergvist D, Ekberg O, Janzon L. A population analysis of variables associated with rupture of small abdom- based screening of abdominal aortic aneurysms. Eur J Vasc inal aortic aneurysms. Surgery 1985;37:724e32. Surg 1991;7:53e7. 21. Powel JT, Brown LC. The natural history of abdominal aortic 4. Cronenwett JL, Murphy TF, Zelenok GB, et al. Actual anal- aneurysms and their risk of rupture. Acta Chir Belg ysis of variables associated with rupture of small abdominal 2001;101:11e6. aortic aneurysms. Surgery 1985;98:472e83. 22. Feinglass J, Cowper D, Dunlop D, et al. Late survival risk 5. Arko FR, Filis KA, Seidel SA, et al. How many patients with factors for abdominal aortic aneurysm repair: experience infrarenal aneurysms are candidates for endovascular from fourteen Departments of Veterans Affairs hospitals. repair? The Northern California experience. J Endovasc Surgery 1995;118:16e24. Ther 2004;11:33e40. 23. Akersdijk GJ, Van der Graaf Y, Moll FL, et al. Complications 6. Compton CN, Dillavou ED, Sheehan MK, Rhee RY, of standards elective abdominal aortic aneurysm repair. Eur Makaroun MS. Is abdominal aortic aneurysm repair appro- J Vasc Endovas Surg 1998;15:505e10. priate in oxygen-dependent chronic obstructive pulmonary 24. Eskandari MK, Proctor MS, Henke PK, et al. Oxygen- disease patients? J Vasc Surg 2005;42:650e3. depended chronic obstructive pulmonary disease does not 7. Upchurch GR, Proctor MC, Henke PK, et al. Predictors of prohibit aortic aneurysm repair. Am J Surg 1999;178: severe morbidity and death after elective abdominal aortic 125e8.