t”ROPEAN ,cJ”KNAL OF CARDIO-THORACIC

European Journal of Cardio-thoracic Surgery 12 (1997) 82-87

Thoracotomy and : postoperative pulmonary function, pain and chest wall complaints’ Downloaded from https://academic.oup.com/ejcts/article/12/1/82/417840 by guest on 25 September 2021

Markus Furrer a,*, R. Rechsteiner a, V. Eigenmann a, Ch. Signer b, U. Althaus a, H.B. Ris a

aDepartment of’ Thoracic and Cardiovascular Surgery, Inselspital, University of Bern, Bern, Switzerland b Institute of Anaesthesiology and Intensive Care, Inselspital, University of Bern, Bern, Switzerland

Received 21 October 1996; received in revised form 22 January 1997; accepted 11 March 1997

Abstract

Objective: Two different surgical accesses combined with standard pain management procedures are compared regarding early and intermediate pulmonary function and pain relief. Methods: In a prospective study, 15 consecutive patients undergoing video-thoracoscopy for pulmonary (group 1) were matched to 15 patients undergoing standard postero-lateral for lobectomy (group 2) according to age, gender and preoperative pulmonary function. Postoperative pain control consisted of patient controlled analgesia in group 1 and epidural analgesia in group 2. Pain intensity was scored from O-4. The predicted postoperative pulmonary function (FVC and FEV 1) after lobectomies was calculated from the preoperative value according to the extent of resection. A clinical measurement was obtained after a mean follow-up time of 4.2 months. Results: The ratios of postoperative measured to predicted values of FVC and FEW for group 1 compared with group 2 were 0.64 f 0.15 and 0.65 f 0.14 compared with 0.60 + 0.19 and 0.59 f 0.13, resp. (both n.s.) at the first day postoperative; 0.92 f 0.18 and 0.95 i 0.17 compared with 0.76 f 0.20 (P < 0.05) and 0.83 f 0.23 (n.s.), resp. at hospital discharge; 0.98 _+ 0.10 and 0.94 f 0.14 compared with 1.01 k 0.17 (n.s.) and 1.10 f 0.17 (P < 0.0.5), resp. at follow-up. Pain intensity score one day after surgery ranged from 0.4 (resting position) to 1.6 (coughing) for group 1, and from 0.3 to 1.2 for group 2. Thirty-six percent of the thoracoscopy patients and 33% of the thoracotomy group complained of persistent pain or discomfort on the site of the operation after 3- 18 months. Conclusion: Post-thoracotomy pain can be effectively controlled with epidural analgesia and pain intensity is no higher than in patients after thoracoscopy who are managed with patient controlled analgesia. FVC is slightly more decreased after thoracotomy during the early postoperative period. FVC and FEV 1 approach the predicted values after four months in both groups. The rate of persistent pain is similar after thoracoscopy and thoracotomy. 0 1997 Elsevier Science B.V.

Keywords: VATS; Postoperative pain; Pulmonary function; Prevalence of chronic pain

1. Introduction are major goals during the early postoperative period following major thoracic surgery. Thoracotomy is considered to be among the most During the past decade, thoracic surgeons have pur- painful operative procedures [9,15,16]. Perioperative sued innovative techniques to minimize the surgical pain control and maintenance of pulmonary function insult. Modifications of conventional techniques such as a muscle sparing postero-lateral thoracotomy have been shown to be associated with a marked reduction in * Corresponding author. Present address: Thoracic and Vascular postoperative pain [6]. Video-assisted-thoracic-surgical Surgery, Kantonsspital Chur, CH-7000 Chur, Switzerland. Tel: + 41 (VATS) techniques seem not only to improve patient 81 2566207; fax: f41 81 2566279. ’ Prcscnted at the 10th Annual Meeting of the European Associa- comfort, but also to reduce early postoperative depres- tion for Cardio-thoracic Surgery, Prague, Czech Republic, 6-9 Octo- sion of pulmonary function [20]. Most of the reports on ber 1996. VATS techniques, however, have only adressed feasibil- lOlO-7940/97/$17.00 0 1997 Elsevier Science B.V. All rights reserved. P11s1010-7940(97)00105-x hf. Furrer et al. /European Journal of Cardio-thoracic Surgery 12 (1997) 82-87 83 ity and technical outcome [1,2,8,10,11,14,17,18]. The ing procedure: Two patients from group 1 younger impact on functional outcome still needs to be clearly than 20 years of age were excluded in the consecutive defined by randomized studies comparing these new series of 32 patients for a correct matching of the two minimally invasive techniques to accepted conventional groups). procedures. The second approach to minimizing postoperative 2.4. Endpoints of the study pain focuses on techniques which improve pain control such as regional and epidural anaesthesia (EDA, by opioids, local anaesthesia, or both), or patient controlled 2.4.1. Pulmonary function test In all patients the same method of bedside intravenous analgesia (PCA, by opioids). While patients was used (Microlab 33, Micro Medical Ltd. Rochester, may benefit from EDA after conventional thoracotomy, regional techniques seem unnecessary in patients under- Kent ME12 AZ England). Forced vital capacity (FVC) Downloaded from https://academic.oup.com/ejcts/article/12/1/82/417840 by guest on 25 September 2021 and forced expiratory volume in 1 s (FEV 1) were going simple thoracoscopic procedures [7]. measured the day before surgery, the day after surgery, In cancer patients performing major pulmonary resec- at hospital discharge and after an intermediate follow-up tions through a thoracotomy is still considered the gold period (mean follow-up time 4.2 months). The best values standard for such proceclures, as VATS techniques might after three attempts were recorded. compromize oncologic principles in thoracic surgery [3,5]. The main effort, therefore, should be to optimize For lobectomy patients postoperative predicted opti- mal pulmonary function was calculated according to the perioperative pain management in those patients who recommendations of Loddenkemper [13]. In the case of still have to undergo thoracotomy. Nevertheless, com- small pulmonary wedge resections postoperative pre- paring the functional postoperative outcomes after dicted optimal values were approximated as being iden- VATS and conventional thoracotomy procedures as well tical to preoperative values. as different pain control techniques has important clini- cal relevances. The degree of pulmonary function depression was In accordance with our routine clinical practice we expressed as a ratio of measured postoperative to pre- dicted values. therefore compared in a prospective study two different standard procedures for pulmonary resections and their respective and perioperative pain control regimens 2.4.2. Pain intensity scoring (Video-thoracoscopic and PCA vs. postero-lateral thora- Pain was evaluated at rest and during coughing at the cotomy and EDA). The study compared differences in same time as pulmonary function was measured. Postop- early and intermediate postoperative pulmonary func- erative pain intensity was ranked by a nurse using a visual tion, pain control and prevalence of chest wall com- analog pain scale which ranged from 0 to 4. plaints. 2.4.3. Intermediate follow-up All patients were well managed clinically in terms of 2. Methods and material persistent postoperative pain and shoulder girdle func- tion. 2.1. Study design 2.5. Patients In a prospective study 15 consecutive patients under- going video-thoracoscopy for a pulmonary wedge resec- tion (group 1) were matched to 15 patients undergoing 2.5.1. Age, gender and preoperative pulmonary function a standard postero-lateral thoracotomy for a lobectomy The two groups did not differ in age (55.7 f 18.3 vs. (group 2) according to age, gender and preoperative 60.7 + 9.2 years), gender (9 male and 6 female vs. 11 male pulmonary function. and 4 female) and preoperative pulmonary function (FVC3.56f1.04vs.3.41f0.551andFEV12.71+0.96 2.2. Inclusion criteria vs. 2.25 f. 0.55 1).

For group 1 (video-thoracoscopies): Peripheral pul- 2.5.2. Postoperative diagnosis monary wedge resections and for group 2 (postero-lateral In group 1, six patients had benign nodules (four thoracotomy): Lobectomy for stage I or II cancer. hamartomas, two granulomas), five had malignant pulmonary nodules (two from metastatic lung cancers, 2.3. Exclusion criteria three from metastatic disease or malignant lymphoma), and one patient each suffered from pulmonary infarc- Preoperative chest lube, retained pulmonary secre- tion, pneumonia, apical bullous emphysema and histio- tions, atelectasis or pneumonia, age < 20 years (match- cytosis. 84 M. Furrer et al. /European Journal of Cardio-thoracic Surgery 12 (1997) 82-87

In group 2, 14 patients had stage I or II lung cancers. Additionally in all patients of both groups, 1 g of In one patient with preoperative suspected Paracetamol (PanadoP) was administered every 8 h. histological analyses was consistent with a hamartoma. 2.6. Surgical techniques 2.8. Statistical analyses

All interventions were performed under general Data of the two groups were compared using un- anaesthesia and single-lung ventilation (double-lumen paired t-test analyses. tube) with the patient in the lateral decubitus position. Two Charriere 28 G chest tubes were inserted at the end of the surgery. 3. Results

In group l-patients three trocar sleeves ranging in Downloaded from https://academic.oup.com/ejcts/article/12/1/82/417840 by guest on 25 September 2021 diameter from seven to 11 mm were placed in the seventh and fourth intercostal spaces thus forming a 3.1. Pulmonary function test triangle. A straight, 0 degree 7 mm telescope (Richard Wolf GmbH, Knittlingen, Germany) was used. Pul- Mean FVC and FEV 1 depressions measured the first monary resection was performed using a multiple fire postoperative day ranged from 59% to 64% of the Endo-GIA-30 mm stapler (Autosuture Company Divi- predicted postoperative values with no significant dif- sion, United States Surgical Corporation, Norwalk, ferences between the two groups. At hospital discharge Conn.). One incision was enlarged if necessary for the both FVC and FEV 1 nearly reached preoperative removal of an undamaged specimen, but rib spreading values in group 1 (92 and 95%, respectively), while in was always avoided. group 2 (thoracotomies) FVC reached only 76% (P < In group 2 a postero-lateral thoracotomy was per- 0.05) and FEV 1 reached only 83% (ns.) of the pre- formed by dividing the latissimus dorsi muscle. The dicted postoperative values (Table 1). After a mean follow-up of 4.2 months pulmonary function was nor- serratus anterior muscle was mobilized and partially divided at its distal insertion. The fourth intercostal malized in both groups, with FEV 1 significantly higher than expected in group 2-patients (FVC: 98% for group space was entered at the upper border of the fifth rib, which was divided by a dorsal resection osteotomy of 1 and 101% for group 2, n.s.; FEV 1: 94% and 110% one centimeter. Rib spreading was limited to 15 cm and respectively, P < 0.05). intercostal nerves were not resected. 3.2. Pain scoring 2.7. Perioperative pain control Postoperative pain control was very effective in both In group 1 intravenous PCA opioid infusions were groups (mean pain score < 0.5 at rest and < 2.0 during used exclusively in the early postoperative period (24- cough) and there was no difference in pain scores the 48 h). PCA pump settings and administered opioid: first day after surgery (0.4 (range O-l) and 0.3 (range SIMS Deltec Inc., Mod. 5800 R, St. Paul Minnesota, O-2) at rest; 1.6 (range O-4) and 1.2 (range O-2.5) USA. Morphine: initial load lo-15 mg, single dose: 2 during cough for group 1 and 2, Fig. 1). mg, shortest interval: 8 min, maximal dose 4 single doses/h. Table 1 In all patients in group 2 EDA was used. Before Mean preoperative values of FVC and FEV 1 and mean ratios of induction of an epidural catheter measured/predicted postoperative values at the first postoperative was introduced at a level between the fourth and fifth day, at hospital discharge and at follow-up (mean follow-up 4.2 thoracic vertebra. 90 min before the estimated end of months) the operation the continous epidural Bupivacain (0.25- Group Mean Mean ratio of measured postoperative/ 0.5%) infusion was changed to ‘Breivik’s solution’ preoperative predicted value (Bupivacain O.l%, Fentanyl 2 mcg/ml, Adrenalin 2 value (L) Pg/ml) administered in increasing doses of up to 6- 15 I day post- Hospital Follow-up ml/h which corresponded to the usual dose required operative discharge during the early postoperative period. EDA effects were FVC checked every 4 h by a nurse using a standard protocol Group 1 3.56 k 1.04 0.64*0.15 0.92 + 0.18 0.98 k 0.10 and modifications in epidural drug application were Group 2 3.41 k 0.55 0.60+0.19 0.76kO.20 1.01 kO.17 then managed by a ‘pain service anaesthesiologist’. The t-test n.s. n.s. <0.05 ns. epidural catheter was removed after a short period of FEV 1 Group 1 2.71 10.96 0.65 k 0.14 0.95 * 0.17 0.94 + 0.14 dose reduction typically 3-7 days postoperatively after Group 2 2.25 i 0.55 0.59+0.13 0.83kO.23 l.lOkO.17 removal and a reduction in the patient’s pain t-test n.s. ns. ns. <0.05 intensity score. M. Furrer et al. /European Journal of Cardio-thoracic Surgery 12 (1997) 82-87 85

nant diseases. In this study, therefore, the hypothesis we tested were whether adequate postoperative pain control would not only be possible by minimizing the surgical insult, but also by combining conventional surgical techniques with effective regional anaesthetic regimens such as epidural analgesia. In this study pa- group 1 tients in the two groups differed in regard to the type of group 2 disease and the extent of pulmonary resection, but were similar in age, gender and preoperative pulmonary function. The disease characteristics of patients, how- ever, should not have an important impact on early postoperative pain intensity or depression of the indi- Downloaded from https://academic.oup.com/ejcts/article/12/1/82/417840 by guest on 25 September 2021 vidually calculated predicted postoperative pulmonary function. Mean pain scores (O-4) one day after surgery at rest during cough which were recorded every 4 h by nurses not involved 1 2 in the study protocol, were remarkably low both in the resting position (0.3-0.4) and during cough (1.2- 1.6) Fig. 1. Mean pain scores during the first day after surgery (no for patients in both groups with no significant differ- differences between groups 1 and 2. ences. This is in contrast to a study by Giudicelli where chest pain intensity on the first postoperative day was 3.3. Infermediate foIIow-up higher even after video-assisted minithoracotomy [4]. One patient in group 1 with metastatic non small cell Our data confirmed the empiric clinical impressions lung cancer died during the follow-up period. Five of leading to the study’s hypotheses that postoperative 14 patients in group 1 (36%) and five of 15 patients in pain after thoracotomy can be efficiently controlled by group 2 (33%) complained of some persistent pain or EDA and its intensity seems to be no greater than after discomfort on the side of the operation after 3- 18 thoracoscopic approached pulmonary resections with months. Pain intensity was always judged to be low and PCA pain management. in most cases was described as burning sensation. One Pain control is thought to be associated with mainte- patient in each group had persistent localized sensory nance of early postoperative pulmonary function. In a deficit of the thoracic wall. Shoulder girdle function comparative study by Hazelrigg, however, patients after was normal in all but one patient after thoracotomy. muscle-sparing thoracotomies, as compared to standard thoracotomies, had less postoperative pain but pul- monary function remained unchanged [6]. The random- ized pleurectomy study by Waller [20] showed a 4. Discussion significant difference in depression of pulmonary func- tion on the third postoperative day (for FVC and FEV While the debate between proponents of thoraco- 1 81 and 75% of preoperative values in the VATS vs. 61 scopic and conventional surgical techniques has focused and 57% in the conventional surgery group). In this in most reports on technical feasibility and operative study pulmonary function was measured on the first complications, there are only a few studies comparing postoperative day. This was due to standardized early data on postoperative pain or postoperative respiratory postoperative nursing care for all patients one day after function [4,6,20]. The thoracoscopic approach seems to surgery and furthermore represents the most painful1 reduce both acute and chronic pain. In a randomized postoperative day for most of the patients. Sixty four study by Waller narcotic requirements were decreased and 60% of FVC and 65 and 59% of FEV 1 was after VATS compared to open pleurectomies during the maintained after thoracoscopy and thoracotomy re- first postoperative day [20]. In lobectomy patients spectively, with no significant differences between the Walker reported a reduced hourly morphine consump- two groups. These favorable postoperative functional tion of 1.36 mg/h in VATS vs. 1.87 mg/h in patients values are also in contrast to Giudicelli’s study where undergoing conventional procedures [ 18,191. FVC and FEV 1 after thoracotomy and VATS were In our study we compared two standard procedures more decreased (38-U”/a of predicted FVC and 32- as well as two perioperative pain control regimens. 34% of predicted FEV 1 on day 2 with no differences Despite our positive experience with the more than 700 after thoracotomy or VATS) [4]. Our data confirm that thoracoscopies that have been performed at our insti- early postoperative function and pain relief after thora- tute since 1990, thoracotomy for major pulmonary coscopy is excellent, but the same favorable results can resection has not been abandoned because adequacy of be obtained by improving the perioperative pain con- resection remaines the main goal when treating malig- trol with modern regional anaesthetic techniques. In 86 M. Furrer et al. /European Journal of Cardio-thoracic Surgery 12 (1997) 82-87 this study hospital stay was not compared because References the two groups were different concerning the patients’ diagnoses. Conclusions regarding surgical access are Ill Coltharp WH, Arnold JH, Alford WC, Burrus GR, Glassford therefore not possible. At hospital discharge FVC and DM, Lea JW, Petracek MR, Starkey TD, Stoney WS, Thomas FEV 1 were almost normalized after thoracoscopy CS, Sadler RN. Video-thoracoscopy: Improved technique and expanded indications. Ann Thorac Surg 1992;53:776-779. (92% and 95%), while after thoracotomy the values I21Fuentes P. Major pulmonary resection by video-assisted were still slightly more decreased (FVC 76%, P < 0.05 minithoracotomy. Eur J Cardiothorac Surg 1994;8:254-258. and FEV 1 83%, ns.). Although functional recovery [31 Ginsberg RJ. Thoracoscopy: a cautionary note. Ann Thorac was favorable in all patients it was significantly de- Surg 1993;56:801-803. layed after thoracotomy as compared to thora- 141Giudicelli R, Thomas P, Ponjon T, Ragni J, Morati N, Ottomani coscopy. R, Fuentes PA, Shennib H, Noirclerc M. Video-assisted minitho- racotomy for performing lobectomy. Ann Thorac Surg Downloaded from https://academic.oup.com/ejcts/article/12/1/82/417840 by guest on 25 September 2021 We also were interested whether long term chest 1994;58:712-718. complaints after thoracotomy or thoracoscopy would 151Goldstraw P. -assisted microthoracotomy (editorial) be different. Landreneau analysed the incidence of 1992;47:489. chronic pain in 165 patients after muscle-sparing PI Hazelrigg SR, Landreneau RJ, Boley TM, Priesmeyer M, lateral thoracotomies and in 178 patients after VATS Schmaltz RA, Nawarawong W, Johnson JA, Walls JT, Curtis JJ. procedures. During the first year the results in re- The effect of muscle-sparing versus standard posterolateral tho- racotomy on pulmonary function, muscle strength, and postop- gard to the presence and intensity of pain as well erative pain. J Thorac Cardiovasc Surg 1991;3(101):394-401. as shoulder function were superior after VATS re- [71 Horswell JL. Anesthetic Techniques for Thoracoscopy. Ann section, but there was no difference after 1 year be- Thorac Surg 1993;56:624-629. tween the two groups [12]. Our results after an inter- PI Inderbitzi R, Furrer M. The surgical treatment of spontaneous mediate follow-up (4.2 months) did not confirm this, by video-thoracoscopy. Thorac Cardiovasc Surg revealing no difference between the two groups with 1992;40:330%333. 191Kehlet H. Post-operative pain, American College of Surgeons, 36 and 33% persistant minor chest wall complaints care of the surgical patient, Committee on pre- and post-opera- and normal shoulder girdle function in all but one tive care, Scientific American Medicine, New York, 1989;1:3-12. patient. It is remarkable but in agreement with the 1101Kirkby TJ, Mack MJ, Landreneau RJ, Rice TW. Initial experi- findings of other authors [12] that even after so called ence with video-assisted thoracoscopic lobectomy. Ann Thorac minimally invasive techniques every third patient Surg 1993;56:1248-1253. seems to have some persistant pain or discomfort at Ull Landreneau RJ, Mack MJ, Hazelrigg SR, Dowling RD, Acuff TE, Magee MJ, Ferson PF. Video-assisted thoracic surgery: the site of the operation. Postoperative pulmonary basic technique concepts and intercostal approach strategies. function was normalized in both groups and FEV 1 Ann Thorac Surg 1992;54:800-807. after thoracotomy was even 10% higher than pre- [I21 Landreneau RJ, Mack MJ, Hazelrigg SR, Naunheim K, Dowl- dicted. We speculate that most patients with lung ing RD, Ritter P, Magee MJ, Nunchuck S, Keenan RJ, Ferson cancer stopped smoking after surgery and also im- PF. Prevalence of chronic pain after pulmonary resection by proved their pulmonary function with drugs and in- thoracotomy or video-assisted thoracic surgery. J Thorac Car- diovasc Surg 1994;107:1079-1086. halation therapy. [I31 Loddenkemper R. Criteria of functional operability in patients with bronchial carcinoma: Preoperative assessment of risk and prediction of postoperative function. J Thorac Cardiovasc Surg 5. Conclusion 1983;31:334. [I41 McKenna RJ. Lobectomy by video-assisted thoracic surgery Post-thoracotomy pain can be effectively controlled with mediastinal node sampling. J Thorac Cardiovasc Surg 1994;107:879-882. with epidural analgesia and early postoperative pain u51 Mulder DS. Pain management principles and tech- intensity is no higher than in patients after thora- niques for thoracoscopy. Ann Thorac Surg 1993;56:630-632. coscopy who are managed with patient controlled [I61 Oden R. Acute post-operative pain: incidence, severity and the analgesia. FVC seems to be slightly more decreased etiology of inadequate treatment. Anesth Clin North Am after thoracotomy than after thoracoscopy during the 1989;7:1-15. early postoperative period. FVC and FEV 1 approach 1171Roviaro GC, Varoh F, Rebuffat C, Vergani C, D’Hoore A. Scalambra SM Maciocco M, Grignani F. Major pulmonary the predicted values after four months in both resections: pneumonectomies and lobectomies. Ann Thorac Surg groups. The rate of persistent pain after 4 months is 1993;56:7799783. similar after thoracoscopy and thoracotomy. 1181Walker WS, Camochan FM, Mattar S. Video-assisted thoraco- Our data suggest that the early postoperative func- scopic . Br J Surg 1994;81:81-82. tional disadvantage of more extensive surgical ap- I191 Walker WS, Craig SR. Video-assisted thoracoscopic pulmonary proaches-governed by underlying disease and surgery-current status and potential evolution. Eur J Cardio- thorac Surg 1996;10:161-167. resectability-can nearly be compensated by the addi- [201 Wailer DA, Forty J, Morritt G. Video-assisted thoracoscopic tional use of effective regional anaesthetic regimens surgery versus thoracotomy for spontaneous pneumothorax. such as epidural analgesia. Ann Thorac Surg 1994;58:372-377. M. Furrer et al. /European Journal of Cardio-thoracic Surgery 12 (1997) 82-87 81

Appendix A. Conference discussion you even can improve the results after open surgery with muscle-spar- ing thoracotomy procedures. Dr J.-M. WihIm (Strasbourg, France): The prospective study is Dr P. Van S&B (Edegem, Belgium): Your results are very interesting always interesting, but we could think that you should have four and thoracoscopy is sometimes more painful than we would like it to groups for purpose of your mix, changing the PCA and EDA for each be. On the other hand you are comparing two kinds of operation: a group. lobectomy versus wedge excision. Don’t you think you should perform Dr M. Furrer: No, we had only two groups. a randomized study comparing patients with the same type of opera- Dr J.-M. Wihlm: Yes, I know. But you could have four groups, tion, a wedge excision, and then performing an axillary thoracotomy prospective groups, looking at the PCA on the thoracoscopy and EDA or a muscle-sparing thoracotomy versus a VATS procedure? on the other one. Dr M. Furrer: I agree. But if you consider a lobectomy to be a more Dr P. Keszler (Buaizpest, Hungary): You compared here, if I invasive procedure than a wedge resection, functional results should be understood well, the posterolateral thoracotomy with the thora- even worse, but they aren’t. coscopy. But if you could compare the axillary thoracotomy, which is Dr F.C. Wells (Cambridge, England, UK): One can’t resist asking much more muscle sparing, with the thoracoscopy, in my opinion the what you’re going to do with your VATS equipment now that the Downloaded from https://academic.oup.com/ejcts/article/12/1/82/417840 by guest on 25 September 2021 differences would have been less significant. advantage seems to be so small? Dr M. Furrer: You’re absolutely right. Our hypothesis was to have Dr M. Furrer: We have performed about 700 VATS procedures since nearly the same functional results after thoracotomy with effective 1990, and we still believe that VATS is a very very good technique, but epidural analgesia as after thoracoscopy. But I absolutely agree that not for lobectomy in cancer patients. That’s the message.