Lung Cancer (2007) 57, 175—180

avai lab le at www.sc iencedi rec t .com

journa l homepage: www.e lsev ier .com/ locate / lungcan

Post-operative respiratory rehabilitation afterlung resection for non-small cell lung cancer

Alfredo Cesarioa,b, Luigi Ferri c, Domenico Galettad,∗, Franco Pasquac,Stefano Bonassie, Enrico Clini f, Gianluca Biscionec, Vittorio Cardaci c,Stefania di Toroc, Alessia Zarzanac, Stefano Margaritoraa, Alessio Pirainoa,Patrizia Russog, Silvia Sterzih, Pierluigi Granonea

a Department of Surgical Sciences, Division of Thoracic Surgery, Catholic University, Romeb Clinical Respiratory Pathology Translational Laboratory, IRCCS San Raffaele, Romec Department of Internal Medecine, Rehabilitative Pneumology, IRCCS San Raffaele, Romed Division of Thoracic Surgery, European Institute of Oncology, Via Ripamonti, 435, 20141 Milan, Italye Unit of Molecular Epidemiology, National Cancer Research Institute, Genoa, Italyf Pulmonary Rehabilitation, University of Modena, Ospedale Villa Pineta, Pavullo, Modena, Italyg Translational Research Laboratory B ‘‘Lung’’, National Cancer Research Institute, Genoa, Italyh Rehabilitation Service, Campus Biomedico University, Rome, Italy

Received 17 August 2006; received in revised form 9 February 2007; accepted 26 February 2007

KEYWORDSNon-small cell lungcancer;Pulmonaryrehabilitation;Surgery

SummaryBackground: To investigate the efficacy of an inpatient Pulmonary Rehabilitation program (i-PR)after lung resection (LR) for Non-Small Cell Lung Cancer (NSCLC).Patients and methods: From January 2001 to December 2004, 211 out of 618 patients whounderwent LR were considered eligible for i-PR. Twenty-five patients accepted the i-PR andwere included in the case group. The remaining 186 who refused i-PR were taken as controls.Results: The two study groups were comparable for demographic and surgical characteristics,as well as for the peri-operative morbidity (4% in the controls and 3% among patients undergoingi-PR). Most functional parameters among treated patients were improved when baseline versus1-month figures were compared, despite the strong correction for multiple comparison limitedstatistical significance to Borg scale dyspnoea on exertion — median — (2 versus 0; p < 0.01);

pH (7.45 versus 7.42; p < 0.05); timed walk-6MWD (297.8 m versus 393.4 m; p < 0.01) and Hbsaturation during 6 MWD (95.4% versus 93.9%; p < 0.05). On the contrary, global function in thegroup of controls was homogeneously decreased (FEV1 and PEF p < 0.01) after operation. Thecomparison of treated and untreated patients 1 month after the operation did not show any

∗ Corresponding author. Tel.: +39 02 57489801; fax: +39 02 57489698.E-mail address: mimgaletta@yahoo.com (D. Galetta).

0169-5002/$ — see front matter © 2007 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.lungcan.2007.02.017

176 A. Cesario et al.

significant difference in terms of FEV1, FVC, PEF, distance, Hb saturation, and KCO that insteadwere homogeneously and significantly worse at baseline (before the surgical operation) in thecase group.Conclusions: Respiratory Function and exercise capacity significantly improve following a post-operative 4-week i-PR in lung resected patients. i-PR could be regarded as a component of the

have undergone LR for cancer.All rights reserved.

1

TPap

eiednplptridCceatN

2

2

Ir6lweimabpwbedgegrs

2

Eami(i

3

3

Agtrom

Vrt

3

BpiyRa

3

Tugtbmgby the physical therapist. Every 2 min the patients wereencouraged using standardised phrases. Once familiar with

management of patients who© 2007 Elsevier Ireland Ltd.

. Introduction

here is body of evidence regarding the efficacy ofulmonary Rehabilitation (PR) in the comprehensive man-gement of patients with respiratory disease and PRrograms are practiced worldwide [1].

Positive results in terms of improvement in dyspnoea,xercise capacity and Quality of Life (QoL) are recognisedn Chronic Obstructive and Non-Obstructive Pulmonary Dis-ase, including COPD, cystic fibrosis, and restrictive thoracicisease [2—5]. Moreover, PR is becoming a crucial compo-ent of the overall treating strategy in high risk surgicalatients (i.e. lung volume reduction surgery — LVRS — andung transplantation) [6]. In particular, PR was part of thereoperative phase in those candidates who met criteria forhe large multicenter NETT trial [7]. There are few dataegarding the effectiveness of inpatient PR (i-PR) programsn the pre and post-operative period of patients who are can-idates or have undergone lung resection (LR) for Non-Smallell Lung Cancer (NSCLC) [8—11]. On the other hand, thora-otomy and parenchymal resection have shown an importantffect on pulmonary function and QoL [12—15]. Our study isimed to evaluate the effect of i-PR on respiratory func-ional parameters in a group of patients subject to LR forSCLC.

. Methods

.1. Population

nstitutional Review Board approval was obtained for theeported study. From January 2001 until December 2004,18 consecutive patients underwent LR for NSCLC. Given theack of established evidence recommending i-PR after LR,e gave all the resected patients who fulfilled the generalligibility criteria for i-PR (no severe peri-operative morbid-ty, no previous i-PR, no unstable angina or joint limitingobility or peripheral vascular disease), the opportunity to

ttend the program. Of the 211 eligible patients who haveeen offered upon informed consent to enter the i-PR, 25atients accepted and were included in the study. Patientsho did not accept entering the i-PR program (N =186)ut who still fulfilled the inclusion criteria were consid-red, upon additional informed consent regarding personalata, as the control group. Demographic, clinical, and sur-

ical data of all patients were prospectically collected andvaluated comparing the two groups. Population and sur-ical characteristics are described in Table 1. All patientseceived a standardised surgical approach (lateral muscleparing thoracotomy) and pharmacological treatment.

ttbvm

.2. Acute pain management

ach patient received thoracic para-vertebral bupivacaines bolus followed by continuous infusion. Ketorolac andorphine ‘‘patient controlled analgesia’’ (PCA) was admin-

stered to all patients in the early post-operative period48 h). Reliever therapy with oral NSAIDs was permitted dur-ng i-PR.

. Measurements

.1. Pulmonary function evaluation

ll patients underwent a baseline spirometry and bloodases analysis as part of the preoperative evaluation. Theseests were repeated 1 month after discharge. For patientseceiving the i-PR, these exams were performed at the endf the rehabilitation program and for control patients 1onth after the operation.Dynamic and static lung volumes were measured using a

MAX® Spirometer (SensorMedics; Yorba Linda, CA) and cor-ected for temperature and barometric pressure, accordingo the American Thoracic Society recommendations [16].

.2. Blood gas analysis

lood gases were sampled at the radial artery whileatients, breathing room air for at least one hour, were lyingn bed in a supine position for 15 min. Samples were anal-sed using an AVL-OMNI® blood gas analyser (Modula System;oche Diagnostic Corporation, Basel, Switzerland) to obtainrterial PO2, PCO2 and pH.

.3. Six-minutes walking test (6MWD)

he exercise endurance was evaluated with a 6MWD testsing the modified 6MWD protocol [17]. The ATS referenceuideline has been followed [18]. Patients were instructedo walk at their fastest pace to cover the longest possi-le distance over 6 min. Oxygen saturation was continuouslyonitored, supplementary oxygen was supplied to keep oxy-

en saturation >88%, and the equipment was carried out

he instructions and after two practice sessions, the twoests were completed in a 50-m corridor with a 30-min restetween tests. The longest walk was taken to represent thealue for that visit. The following variables were recorded:eters walked, heart and respiratory rates, blood pressure

Pulmonary rehabilitation in NSCLC patients 177

Table 1 Baseline and 1-month physiological parameters for the two study groups

Parameter Treated (N = 25) Controls (N = 186)

Baseline One-month p Baseline One-month p

Lung functionFVC (Lt) 2.27 (0.55) 2.55 (0.69) NS 3.71 (5.22) 2.61 (0.75) NSFEV1 (Lt) 1.57 (0.35) 1.64 (0.34) NS 2.32 (0.76) 1.79 (0.61) <0.01FEF25—75 (Lt) 1.41 (0.86) 1.25 (0.62) NS 1.63 (0.99) 1.55 (4.30) NSPEF (Lt) 4.25 (1.26) 4.44 (1.05) NS 5.71 (1.94) 4.46 (1.61) <0.01KCO (ml/min/mmHg) 2.88 (1.17) 3.29 (0.95) NS 3.96 (1.21) 3.70 (1.28) NSKCO (%) 56.36 (20.11) 61.61 (15.42) NS 74.61 (17.18) 70.27 (21.29) NSRAW (Lt) 2.62 (1.26) 2.37 (1.18) NSMIP (%) 58.75 (24.49) 68.41 (24.14) NSMIP (CmH2O) 60.61 (26.80) 64.75 (22.10) NSBorg-Dyspa 1 (0—2) 0 (0—0) <0.01b

Borg-Exa 2 (2—3) 0 (0—2) <0.01b

Lung volumesTLC (Lt) 4.77 (1.52) 4.93 (1.78) NSTLC (%) 79.20 (18.85) 82.20 (23.45) NSRV (Lt) 2.24 (1.05) 2.09 (1.14) NSRV (%) 92.00 (35.53) 86.30 (42.74) NSVC (Lt) 2.53 (0.69) 2.84 (0.89) NSVC (%) 74.70 (11.84) 84.10 (17.12) NS

ABG analysisPO2 (mmHg) 70.26 (7.59) 72.33 (17.05) NS 76.81 (21.93) 73.68 (19.84) <0.01PCO2 (mmHg) 38.63 (4.19) 40.63 (3.14) NS 47.43 (20.66) 46.34 (19.26) NSpH 7.45 (0.02) 7.42 (0.02) <0.05 7.39 (0.51) 7.42 (0.04) NS

6-MWTDistance (m) 297.8 (70.7) 393.4 (110.6) <0.01 498.9 (78.2) 466.5 (92.7) <0.01Hb saturation (%) 95.4 (1.5) 93.9 (2.4) <0.05 99.2 (1.0) 94.9 (3.3) <0.01

NS: not significant; FEV1: forced expiratory volume in 1 s; FEF25—75: forced expiratory fraction; PEF: peak expiratory force; KCO: Krogindex (transfer factor of carbon monoxide equal to DLCO/alveolar volume); RAW: airflow resistance; MIP: maximal inspiration pressure;

exert

omitnlwlm(sfsimab

Borg Dysp: Borg scale dyspnoea; Borg Ex: Borg scale dyspnoea ona Median (25—75◦).b Wilcoxon signed rank test.

(BP), oxygen flow, and lowest oxygen saturation (measured)using a pulse oximeter (Pulsox-5® portable pulse oximeter;Minolta; Tokyo, Japan) attached to the left index finger.

4. Rehabilitation program

The Rehabilitation team consisted of a chest physiciandirector, physical therapists, nurses, a psychologist and adietician. The program was performed with patients admit-ted to a 28-bed ward. Subjects participated in five dailysessions each week (3-h supervised sessions) up to a max-imum of 20 (mean, 18.2 sessions) during a hospital stay of26 ± 3 days. The program included: (a) supervised incremen-tal exercise until the patient achieved 30 min of continuouscycling at 70—80% of the maximal work load achieved onan incremental cyclo-ergometer test carried out at admis-sion. This test was a symptom-limited incremental exercise

test and was performed on an electrically braked cyclo-ergometer using a standard incremental cycle exerciseprotocol [19]. EKG activity was monitored continuously andsystemic arterial blood pressure (BP) was registered everyminute using a sphygmomanometer. After a 2-min period

5

Tr

ion.

f unloaded pedalling, the load was increased by 10 W eachinute and patients were encouraged to cycle to the point of

ntolerable breathlessness, discomfort or exhaustion, untilhe maximal heart rate was achieved, an abnormal EKG wasoted, or whenever the patient wanted to stop (symptom-imited exercise test). At rest and every minute patientsere asked for their perceived breathlessness/dyspnoea and

eg fatigue by pointing to a number or phrase on a 10-pointodified Borg scale [20] set on a sheet in front of them.

b) Abdominal muscle activities, inspiratory resistive ses-ions, treadmill, upper and lower extremities training andull arm circling were applied as well [21]. (c) Educationalessions were conducted twice weekly and covered such top-cs as pulmonary physiopathology, pharmacology of patients’edications, dietary counselling, relaxation and stress man-

gement techniques, energy conservation principles, andreathing retraining.

. Data analysis

he distribution of treated and untreated patients withespect to demographic and surgical parameters was tested

178 A. Cesario et al.

wee

wocivttawgitTiaapps

6

Tsmgeptp(a

pufpv(6hcsF4(agtipv34pdipec

Fig. 1 Comparison of functional parameters at baseline bet

ith the Chi-square test or the Fisher’s exact test. The effectf iRP on pulmonary functional parameters was assessedomparing each parameter before and after the treatmentn the group of treated patients. A mixed model of multi-ariate analysis of variance was performed, which includedhe overall differences in the treated and control groups,he time-effect between the pre-test and the post-test,nd the interaction term. The interaction effect indicatedhether the effect at 30 days is different in the treatedroup as opposed to the control group. The non homogene-ty of variances, mostly due to the unequal sizes of thewo groups, was accounted for comparing groups with theamhane’s test, which is a conservative pairwise compar-sons test based on a t test [22]. Categorical covariates werenalysed using the non-parametric Wilcoxon signed rank testnd applying the classical Bonferroni correction for multi-le comparisons (95% significance level was reached with a< 0.0011). All analyses were done using the SPSS statistical

oftware, version 11.5.

. Results

he two study groups were comparable for demographic andurgical characteristics, as well as for peri-operative minororbidity (4% in the controls and 3% among patients under-

oing i-PR). The comparability of the two groups has beenvaluated testing the distribution of treated and untreated

atients according to those variables potentially affectinghe evaluation of functional parameters. None of thesearameters significantly differed in the two study groupsTable 1). The progression of functional parameters 1 monthfter the operation is compared in treated and untreated

7

Tt

n subjects included in the treatment group and the controls.

atients. The almost general improvement in patientsndergoing i-PR is evident. Significant increases were foundor Borg scale dyspnoea at rest — median — (1 versus 0;< 0.01); Borg scale Dyspnoea on exertion — median — (2ersus 0; p < 0.01); pH (7.45 versus 7.42; p < 0.05); 6 MWD297.8 m versus 393.4 m; p < 0.01) and Hb saturation duringMWD (95.4% versus 93.9%; p < 0.05) (Table 1). On the otherand, a reverse trend has been observed in the group ofontrols, with many parameters decreasing 1 month afterurgery. Among these significant differences were found inEV1 (2.32 Lt versus 1.79 Lt; p < 0.01); PEF (5.71 Lt versus.46 Lt; p < 0.01); and PO2 (76.81 versus 73.68; p < 0.01)Table 1). The direct comparison of functional parameterst baseline between subjects included in the treatmentroup and the controls, clearly reveals as subjects acceptinghe i-PR had much lower performances (Fig. 1). Signif-cant differences were found in FEV1 (1.57 versus 2.32;< 0.01); FVC (2.27 Lt versus 3.71 Lt; p < 0.05); PEF (4.25 Ltersus 5.71 Lt; p < 0.05), KCO (2.88 ml/min/mmHg versus.96 ml/min/mmHg; p < 0.05); distance (m) (297.8 versus98.9; p < 0.01); and Hb saturation (%) (95.4 versus 99.2;< 0.01). Fig. 1 shows that after the i-PR the general con-ition of treated patients remarkably improved and thismprovement, combined with a physiological reduction oferformances in untreated patients, cancels baseline differ-nces between groups, which are after 1 month of surgeryomparable in all parameters.

. Discussion

his observational study indicates that an early rehabili-ative intervention prevents deterioration and speeds up

tprafi(tiripTdcpPtdaprothogoowrrfgoiu

ivltmtdob

8

OfaCrsd

Pulmonary rehabilitation in NSCLC patients

recovery of function in surgical patients resected for NSCLC.Improvement in exercise tolerance following a pulmonaryrehabilitation program is a timely and well-established find-ing in COPD patients [1—5,23,24]. In truth, although thereis general agreement that physical training promotes sub-jective patients’ sense of well being [25] conflicting resultshave been obtained in those studies investigating the realeffect of pulmonary rehabilitation on physiological parame-ters dealing with lung functions. A major contradiction lies inthe fact that while no positive effects of exercise training onlung volumes have been detected [26,27] a somewhat inter-esting effect has been reported on ventilatory, gas exchangeand homodynamic parameters (work of breathing, lung com-pliance, alveolar—arterial difference, P[A − a]O2, maximumoxygen consumption, arterial oxygen tension, PaO2), heartrate, arterial lactate concentration, cardiac index and pul-monary artery pressure [28,29]. To date and to our bestknowledge the evaluation of the possible beneficial effectof an inpatient Pulmonary Rehabilitation program (i-PR) fol-lowing LR on respiratory and global functions has never beeninvestigated in a controlled study. Many reports exist regard-ing the association of i-PR to Lung Volume Reduction Surgery(LVRS) in the treatment of emphysema [30] or tuberculosis[31,32], but the characteristics of patients included in thosestudies do not fit the needs for comparison with regards tothe population we examined in the present study.

Our data have evidenced a substantially beneficial effectof the i-PR program in patients who have undergone LR pro-cedures for NSCLC through a thoracotomic approach. Wetruly believe a warning is necessary to correctly read themessage. In fact it is important to keep in mind that thebaseline functional status of the case group was noticeablyworse than that of the control group patients (see Table 2).This fact somewhat weakens the clinical modulation of databut offers an interesting interpretation of findings. Never-theless the two groups have a clear and statistically provedcomparability.

The large difference between the two groups observedat baseline is strongly reduced 1 month after the operation:thus, whilst it may be assumed that the ‘‘better’’ func-tional status in the early post-operative period may haveinfluenced the decision of the patients to undergo a i-PR(introducing a possible bias), the benefit of the i-PR in thegroup functionally worse, is certainly an outstanding result,even considering that there were no differences in termsof extent of resection (type of operation) among the twogroups. This being stated the interpretation of the signifi-cant difference inside the treated group evidenced in Borgscale dyspnoea; in pH, in 6 MWD, and in the percentage ofHb saturation (but also in the borderline significant param-eters), could be justified by a positive effect of the i-PR onthe patients who have undergone a LR. Moreover, we havesubstantially confirmed that i-PR did not modify static anddynamic lung volumes, according to previously outlined evi-dences in patients with chronic airway obstruction [26,27].On the other hand, outstanding is the fact that whilst in thecontrol group a significant and expected reduction in these

volumes has been detected, this has not been confirmed inthe patients undergoing i-PR.

In addition, this is the first controlled study demonstrat-ing the improvement of exercise tolerance and exertionaldyspnoea in subjects undergoing i-PR after LR. Thanks to

tpLafi

179

he agreement among our surgical and rehabilitation units,atients could be transferred directly at discharge withoutecovery discontinuation. Notwithstanding, some difficultiesnd limitations may arise with this approach. The first dif-culty to face was the presentation of a novel programa typical rehabilitation program with a novel indication)o a cohort of patients without validated evidence regard-ng the true usefulness of the program itself. We had toeformulate the informed consent forms to provide the bestnformation to the patients, since his/her will to enter therogram was the real discriminating criterion for inclusion.hus, since feasibility was given top priority, the study wasesigned in a non-randomised way and data were prospecti-ally observed. Secondly, we had to make a choice regardingarameters to be considered to compare the impact of i-R on the patients’ outcome and make a comparison withhose patients who had not received any i-PR. A typicalataset based on baseline pulmonary function, gas exchangend exercise tolerance tests was used to provide a generalicture of the effect of i-PR. In this setting, pain was dis-egarded as a matching parameter among the two groupsf patients. Pain, in fact, is essentially connected withhe thoracotomic approach [32—35] and since thoracotomyad been performed in all patients its impact on the post-perative outcome was considered homogeneous in the tworoups. The only significant concern was the continuationf hospital stay for a four weeks long period in the groupf patients who had undergone an i-PR. Cost evaluationas beyond the scope of this experience and has not been

ealised. As well we recognise a bias of this study being rep-esented by the lack of a Quality of Life (QoL) analysis. As aurther support to propose i-PR programs to patients under-oing LR it should be noted that no side or adverse effectsf the PR program have been recorded, and the programtself has been well accepted by all the patients who havendergone it.

Despite this study is observational, it seems to clearlyndicate a possible role for PR to early treat and to pre-ent function decline in patients who underwent surgicalung resection for NSCLC. Indeed, it has been suggestedhat well planned observational studies may provide infor-ation that closely resemble that provided by randomised

rials [36]. Nonetheless, we have planned a prospective ran-omised controlled trial that is currently under way. In thisngoing trial the positive results described in this study wille checked out along with a proper cost and QoL evaluation.

. Conclusions

verall, our experience demonstrated that there is placeor cooperation between Pulmonary Rehabilitation Programsnd Lung Resection in patients with Non-Small Cell Lungancer: a positive trend on the post-operative functionalecovery, dyspnoea and exercise tolerance has been demon-trated with no adverse side effects; furthermore the i-PRid not interfere with any oncological follow up and adjuvant

herapy planning. Rehabilitation could be regarded as a com-onent of the management of patients who have undergoneR for cancer. Our preliminary data prompted the design ofprospective randomised controlled trial to confirm these

ndings along with costs and QoL evaluation. Additional

1

rt

A

TFlAR

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

80

esearch to better understand the ideal candidate profileo benefit from this treatment should also be encouraged.

cknowledgments

his study was in part granted by the Catholic University D1unding Program, year 2003, the Associazione Italiana pera Ricerca sul Cancro (AIRC), and the Ministry of Health. Theuthors wish to thank Francesco Pezzella MD, from Canceresearch UK, for paper review.

eferences

[1] Troosters T, Casaburi R, Gosselink R, Decramer M. Pulmonaryrehabilitation in chronic obstructive pulmonary disease. Am JRespir Crit Care Med 2005;172:19—38.

[2] Ries AL, Kaplan R, Linbeg T. Effects of pulmonary rehabili-tation on physiologic and psychosocial outcomes in patientswith chronic obstructive pulmonary disease. Ann Intern Med1995;122:823—32.

[3] Goldstein RS, Gort EH, Stubbung D, Avendano MA, GuyattGH. Randomised controlled trial of respiratory rehabilitation.Lancet 1994;344:1394—7.

[4] De Torres JP, Pinto-Plata V, Ingenito E, Bagley P, Gray A, BergerR, et al. Power of outcome measurements to detect clinicallysignificant changes in Pulmonary Rehabilitation of patients withCOPD. Chest 2002;121:1092—8.

[5] Lacasse Y, Wong E, Guyatt GH, King D, Cook DJ, Golstein RS.Meta-analysis of respiratory rehabilitation in chronic obstruc-tive pulmonary disease. Lancet 1996;348:1115—9.

[6] Celli BR. Chronic respiratory failure after lug resection: the roleof pulmonary rehabilitation. Thorac Surg Clin 2004;14:417—28.

[7] National Emphysema Treatment Trial Research Group. Patientsat high risk of death after lung-volume-reduction surgery. NEngl J Med 2001;345:1075—83.

[8] Smetana S. Pre-operative pulmonary evaluation. N Engl J Med1999;340:937—44.

[9] Smith T, Kinasewitz GT, Tucker WY, Spillers WP, George RB.Exercise capacity as a predictor of post-thoracotomy morbidity.Am Rev Resp Dis 1984;129:730—4.

10] Reilly JJ. Evidence based pre-operative evaluation of candi-dates for thoracotomy. Chest 1999;116:474s—6s.

11] Spruit MA, Janssen PP, Willemsen SC, Hochstenbag MM, WoutersEF. Exercise capacity before and after an 8-week multi-disciplinary inpatient rehabilitation program in lung cancerpatients: a pilot study. Lung Cancer 2006;52:257—60.

12] Yushang C, Zhiyong Z, Xiequn X. The analysis of changes andinfluencing factors of early postthoracotomy pulmonary func-tion. Chin Med Sci J 2003;18:105—10.

13] Nomori H, Horio H, Naruke T, Suemasu K. Posterolateral tho-racotomy is behind limited thoracotomy and thoracoscopicsurgery in terms of postoperative pulmonary function and walk-ing capacity. Eur J Cardiothorac Surg 2002;21:155—6.

14] Li WWL, Lee TW, Lam SS, Ng CSH, Sihoe ADL, Wan IYP, et al.Quality of life following lung cancer resection: video-assistedthoracic surgery vs. thoracotomy. Chest 2002;122:584—9.

15] Kesten S. Pulmonary rehabilitation and surgery for end stage

lung disease. Clin Chest Med 1997;18:173—81.

16] American Thoracic Society. Standardization of spirometry:1994 update. Am J Respir Crit Care Med 1995;152:1107—36.

17] Steel B. Timed walking tests of exercise capacity in chroniccardiopulmonary illness. J Cardiopulm Rehabil 1996;16:25—33.

[

A. Cesario et al.

18] ATS statement: guidelines for the six-minute walk test. Am JRespir Crit Care Med 2002;166:11—7.

19] Maltais F, LeBlanc P, Jobin J, Berube C, Bruneau J, Carrier L, etal. Intensity of training and physiologic adaptation in patientswith chronic obstructive pulmonary disease. Am J Respir CritCare Med 1997;155:555—61.

20] Borg GAV. Psychophysical bases of perceived exertion. Med SciSports Exerc 1982;14:377—81.

21] Simpson K, Killian K, McCartney N, Stubbing DG, Jones NL. Ran-domised controlled trial of weightlifting exercise in patientswith chronic airflow limitation. Thorax 1992;47:70—5.

22] Dunnett CW, Tamhane AC. Step-up multiple testing ofparameters with unequally correlated estimates. Biometrics1995;51:217—27.

23] Moser KM, Bokinski GE, Savage RT, Archibald CJ, Hansen PR.Results of a comprehensive rehabilitation program: physiologicand functional effects on patients with chronic obstruc-tive pulmonary disease. Arch Intern Med 1980;140:1596—601.

24] Bradley BL, Garner AE, Billiu D, Mestas JM, Forman J. Oxygenassisted exercise in chronic obstructive lung disease. Am RevRespir Dis 1978;118:239—43.

25] Mohsenifar Z, Horak D, Brown HV, Koerner SK. Sensitive indicesof improvement in a pulmonary rehabilitation program. Chest1983;83:189—92.

26] Alpert JS, Bass H, Szucs MM, Banas JS, Dalen JE, Dexter L.Effects of physical training on haemodynamics and pulmonaryfunction at rest and during exercise in patients with chronicobstructive pulmonary disease. Chest 1974;66:647—51.

27] Degre S, Sergysels R, Messin R, Vandermoten P, Salhadin P,Denolin H, et al. Haemodynamic responses to physical train-ing in patients with chronic lung disease. Am Rev Respir Dis1974;110:395—402.

28] Daly JJ, Duff RS, Jackson E, Turino GM. Effect of exercise onarterial lactate, pyruvate and excess lactate in chronic bron-chitis. Br J Dis Chest 1967;61:193—7.

29] Pierce AK, Taylor HF, Archer RF, Miller WF. Responses to exer-cise training in patients with emphysema. Arch Intern Med1964;113:28—36.

30] Fishman A, Martinez F, Naunheim K, Piantadosi S, Wise R, RiesA, et al. National Emphysema Treatment Trial Research Group.A randomized trial comparing lung-volume-reduction surgerywith medical therapy for severe emphysema. N Engl J Med2003;348:2059—73.

31] Sokolova VS, Strel’tsov VP, Talumbum EA, Ivanova TI, NovikovaLN. Methods of functional rehabilitation of patients after lungsurgery. Probl Tuberk 2002;10:56—63.

32] Ivanova TI, Sokolova VS, Novikova LN, Borukaev AM. Exercisetherapy in rehabilitation of patients with pulmonary tuberculo-sis before and after surgical treatment. Vopr Kurortol FizioterLech Fiz Kult 2002;6:14—7.

33] Akcali Y, Demir H, Tezcan B. The effect of standard pos-terolateral versus muscle-sparing thoracotomy on multipleparameters. Ann Thorac Surg 2003;76:1050—4.

34] Soto RG, Fu ES. Acute pain management for patients undergo-ing thoracotomy. Ann Thorac Surg 2003;75:1349—57.

35] Richardson J, Sabanathan S, Jones J, Shah RD, Cheema S,Mearns AJ. A prospective, randomized comparison of preop-erative and continuous balanced epidural or paravertebral

bupivacaine on post-thoracotomy pain, pulmonary functionand stress responses. Br J Anaesth 1999;83:387—92.

36] Benson K, Hartz AJ. A comparison of observational studies andrandomized, controlled trials. N Engl J Med 2000;342:1878—86.