Effects of a Clinical Pathway for video-assisted thoracoscopic surgery (VATS) on

quality and cost of care

Matthias HM Schwarzbach1*, Ulrich Ronellenfitsch1*, Qian Wang1, Eric D Rössner1,

Christof Denz2, Stefan Post3, Peter Hohenberger1

*these authors contributed equally to the paper

1Department of Surgery, Division of Surgical Oncology and Thoracic Surgery, University

Medicine Mannheim, Germany

2Department of Anesthesiology, University Medicine Mannheim, Germany

3Department of Surgery, University Medicine Mannheim, Germany

Corresponding author:

Matthias HM Schwarzbach

Department of Surgery, Division of Surgical Oncology and Thoracic Surgery

University Medicine Mannheim

Theodor-Kutzer-Ufer 1-3

D-68135 Mannheim, Germany

Phone: +49-621-383-2227

Fax: +49-6221-805790

E-mail: matthias.schwarzbach@chir.ma.uni-heidelberg.de

Abstract and keywords

Purpose: To evaluate effects of a Clinical Pathway (CP) for video-assisted thoraco-

scopic surgery (VATS) on process quality, outcome quality, and hospital costs.

Methods: We implemented a CP for VATS and compared 34 patients treated with CP to

77 patients treated without CP. Indicators for process quality were: duration of catheter

placement, pain intensity, respiratory exercising and mobilization. Outcome quality was

measured through morbidity, mortality, re-operations, and re-admissions. Cost of hospi-

tal stay was calculated using an imputed daily rate.

Results: Foley catheters were removed significantly earlier after CP implementation. All

patients on CP were mobilised and received pulmonary exercising on the operation day.

Pain levels were low after CP implementation. Median hospital stay significantly reduced

by five days. Perioperative outcome quality remained unchanged. Costs significantly

diminished by 1510 Euro per stay.

Conclusions: CP implementation had positive effects on process quality. Specifically,

catheter management was improved and a good pain control achieved. Patients bene-

fited from shortened stay and were treated at lower cost. A clear effect on outcome qual-

ity was not found. CPs are a promising tool for quality improvement and cost contain-

ment in thoracic surgery.

Keywords: Clinical Pathway; quality of care; thoracoscopy/VATS; outcomes; health

economics

Introduction

Process and outcome quality determine patients’ benefits from a health care inter-

vention [1]. This is especially valid in surgery, where perioperative morbidity and mortal-

ity still constitute a non-negligible challenge for many procedures [2;3]. At the same time,

cost pressure has steadily increased over the last decades [4]. So called Clinical Path-

ways (CPs) are increasingly used with the aim of improving quality of care, effectiveness

of perioperative treatment, and cost management. They are structured longitudinal

treatment plans which include a multidisciplinary and non-hierarchical approach to

streamline patient care at an institutional level [5].

In thoracic surgery, only few studies are available evaluating CP usage in clinical

practice. Lobectomy, video-assisted thoracoscopic surgery (VATS), pneumothorax

treatment and lung transplantation have been studied [6-10]. However, of these studies

one has no comparison group [8] and three do not assess process quality with clinical

treatment parameters such as time of drainage removal [6-8]. Moreover, data are sparse

and stem mostly from the USA and Japan. The single available European study reports

exclusively on pneumothorax treatment with chest drainage and no surgery [9]. Since

CP effects must be assumed to be strongly setting-specific [11], there is a definite need

for more ample evidence from Europe covering all aspects of CP usage including effects

on process quality measured through treatment parameters.

For this reason, we prospectively evaluated the effects of an institutionally devel-

oped CP for patients undergoing video-assisted thoracoscopy (VATS) on process and

outcome quality and hospital costs. In the present analysis we focus on the implementa-

tion phase during the first six months of CP use.

2

Patients and methods

Inclusion criteria and patient populations

The intervention group (CP group) comprised all consecutive patients undergoing

VATS at our institution from April 2007 to October 2007, who were treated with a CP.

Patient data were gathered prospectively. In an intention-to-treat approach, all patients

were included in the quality analysis irrespective of deviations from the CP or complica-

tions occurring during the course of treatment. As there were no defined exclusion or

abortion criteria and thus treatment with the CP was continued even in case of devia-

tions from the pre-specified plan of treatment.

The comparison group (pre-pathway group) consisted of all consecutive patients

who had been treated before CP implementation in 2006/7. For this group, data was

compiled retrospectively from records. In the pre-pathway group, patients were treated

according to the judgment and decisions taken by the treating physicians. Although sev-

eral semi-formal standards had been existent for single elements of care such as e.g.

analgetic therapy, there was no formulated “guideline” covering the whole treatment

course which was to be followed.

In order to obtain unbiased cost data, only patients who had been directly admitted

to and discharged from the thoracic surgery ward and no patients admitted from or trans-

ferred to other departments were included in the cost analysis.

Patient characteristics of the two study groups are displayed in table 1. There are

no statistically significant differences in sex, age and American Society of Anesthesio-

logy (ASA) classification (www.asahq.org). The majority of patients in both groups un-

derwent VATS for diagnostic or therapeutical pulmonary wedge resection (pulmo-

nary/pleural mass or bullae resection). Fewer patients underwent VATS to receive tal-

cum pleurodesis for pleural effusion. Pleurodesis was significantly more common in the

CP group. The most common underlying diagnoses were malignancies in both groups.

The relative number of operations done for pneumothorax / bullous emphysema was

significantly higher in the pre-pathway group.

CP design, implementation and content

Based on various national and international treatment and nursing recommen-

dations as well as on available evidence from the literature and internal treatment regi-

mens (e.g. standardized pain medication) a detailed CP for VATS was designed. The

approach was primarily interdisciplinary involving medical staff, i.e. physicians and

3

nurses of all hierarchy levels and all participating disciplines (surgery, anesthesiology,

radiology, physiotherapy), and administrative units. After a literature research performed

by a designated physician in order to identify pertinent evidence on perioperative treat-

ment of patients undergoing VATS, a CP structure was proposed. Subsequently, it was

integrated with existing institutional nursing standards by the head nurse of the thoracic

surgery ward. The resulting draft was jointly modified and a final version of the CP was

agreed on in a meeting of representatives of all different groups involved in patient care.

After CP implementation, meetings were held to decide on pertinent issues and possible

reframing of the CP when a respective need was conceived by the project group. The

process of CP development and implementation is shown in figure 2.

The CP was designed for intersectoral application (outpatient and inpatient

management). The diagnostic workup (e.g. spirometry, CT scans or laboratory tests)

was scheduled for the preoperative and follow-up treatment algorithms were included for

the postoperative outpatient phase. Hospital admission was to take place one day before

surgery. For the perioperative inpatient phase, the CP implemented key elements of en-

hanced recovery schemes (standardized analgesia, early ambulation, early removal of

invasive catheters etc.) [12;13]. The CP foresaw extubation in the operation theatre and,

except for cases judged to be at high risk of postoperative complications by the treating

surgeon and anesthesiologist, no routine postoperative monitoring of patients at the ICU

/ intermediate care unit. Patients routinely received one chest suction drainage. The

drainage was to be removed upon draining less than 100 ml per 24 hrs. Foley catheters

were placed for the operation and foreseen to be removed as soon as the patient was

awake. Central venous lines and epidural analgesia catheters were reserved for selected

patients at high perioperative risk or where conversion thoracotomy was likely a priori,

and were scheduled for removal on postoperative day 3. Patients without epidural pain

catheter received a standardized pain medication of ibuprofen 400 mg BID, metamizol 1

g TID and oxycodon/naloxon 10/5 mg BID. Patients with epidural catheter received only

ibuprofen 400 mg BID in addition to the epidural medication (sufentanyl/ropivacain in

variable dosage adjusted to the pain level expressed by the patient). Normal food intake

was restituted immediately after surgery and iv lines were removed as soon as the pa-

tient was able to drink sufficiently. Discharge was planned for the fourth postoperative

day based on the minimum stay stipulated in the German DRG catalogue (http://www.g-

drg.de). CPs were designed as four-page paper-based documents containing all stipu-

lated treatment steps for the single pre- and postoperative days. They were kept with

patient’s daily protocols and thus always available at the bedside. The CP is depicted in

the electronic supplementary material.

4

Quality of care

Perioperative process quality was defined as the fulfillment of medical procedures

delivered to the patient as detailed in the CP. Specifically, it was measured by (a) the

number of days invasive catheters (Foley catheter, central venous line, epidural analge-

sia catheter and chest suction drain) remained in situ; (b) early postoperative mobiliza-

tion; and (c) pain levels on day 3 and the day of discharge as expressed by the patient

on a visual analogue scale with possible values from 0 (no pain) to 10 (maximum pain).

Perioperative outcome quality was defined as morbidity, 30-day mortality, readmission

and re-operation rate.

Cost calculation

Since at the time of analysis detailed cost data were not available for the CP

group, cost of hospital stay was computed based on the average daily cost for the 21

patients treated in 2006 meeting the above mentioned inclusion criteria. For these indi-

viduals, expenses from cost centers depending directly on length of stay (treatment on

the regular ward, optionally the ICU / intermediate care unit, radiological investigations

and laboratory testing) were extracted from the internal accounting system and divided

by the mean number of days these selected patients stayed in the facility, thus yielding

an average daily cost. Subsequently, this figure was multiplied with the number of days a

patient stayed in the hospital to obtain an estimate of the hospital cost per treated case

both in the CP and the prepathway group.

Statistical analyses

Between study groups, ordinal variables were compared with the Wilcoxon rank-

sum test. Dichotomous variables were compared with the chi-square test. P-values were

computed and an effect was considered statistically significant at p ≤ 0.05. SPSS 15.0

for Windows (SPSS Inc., Chicago, IL) was used for all statistical analyses.

Results

Standardization and quality of perioperative care

The time of removal of invasive catheters is shown in figure 1. Removal of Foley

catheters took place significantly earlier in the CP group (p<0.01). Central venous lines

in the CP group were removed in all cases on the specified day or earlier whereas the

range of when the line was removed was much wider in the pre-pathway group. In pa-

tients who received an epidural analgesia catheter, it was removed in the majority of

cases in the CP group on day three, as stipulated in the CP, whereas the range of the

removal day was much wider in the pre-pathway group. Regarding chest drain removal,

the median did not differ significantly between groups, but there was considerably less

variation in the CP compared to the pre-pathway group. All patients in the CP group

were mobilized on the day of surgery and received a standardized pulmonary exercise

program including incentive deep breathing exerciser and mechanical chest treatment.

(not shown in figure).

Perioperative outcome quality

30 day mortality was nil in the CP group whereas two patients died in the pre-

pathway group: one case of respiratory failure after pleurodesis for mesothelioma with

malignant pleural effusion and one case of cardiogenic shock in a patient with NYHA IV

congestive heart failure and bullous emphysema operated for recurring pneumothorax

(table 2). The difference was statistically not significant. Morbidity in the CP group oc-

curred in 8.8% of patients in the CP and 27.3% in the pre-pathway group. This difference

did not reach statistical significance (p=0.08). Most of the morbidity in the pre-pathway

group was due to postoperative pneumothorax and/or skin emphysema. evision surgery

occurred in one patient (re-pleurodesis after postoperatively persisting malignant effu-

sion) in the CP group and no patient in the pre-pathway group. None of the patients

treated with the CP required readmission to the hospital in comparison to one patient

with wound infection in the pre-pathway group. Median pain levels on the visual analo-

gue scale (potential range 0-10) were 2 on postoperative day three and 0 on the day of

discharge. Since standardized pain documentation was initiated only upon CP imple-

mentation, no data for comparison are available from the pre-pathway group.

2

Length of hospital stay

The median length of stay in the hospital was five days shorter in the CP group as

compared to the prepathway group (p<0.05, table 3). Within both groups, patients re-

ceiving pleurodesis stayed significantly longer in the hospital compared to patients un-

dergoing atypical pulmonary resection.

Cost of hospital stay

Based on a calculated rate of €302 per day of hospital stay the mean length-of-

stay dependent cost per patient was €4228 in the pre-pathway group and €2718 in the

CP group (p<0.05, table 4). Consequently, the mean cost saving after introduction of the

CP was €1510 per patient treated. The relative effect of cost saving was stronger in pa-

tients undergoing lung resections (€1510 from a total of €3926; 39%) as compared to

patients undergoing thoracoscopic pleurodesis (€1510 from a total of €5738; 26%).

3

Discussion

Our study assessed the effects of a CP for patients undergoing VATS on process

and outcome quality as well as on hospital costs. We were able to measure several fea-

tures of perioperative care which had not been evaluated before when looking at CP

usage in thoracic surgery [6-10]. Specifically, our study is the first to demonstrate that

CPs are able to thoroughly standardize invasive catheter management. Central and epi-

dural lines were reserved for high-risk patients and removed in the vast majority of cases

at the point of time stipulated in the CP. Likewise, in the vast majority of cases foley ca-

theters were removed at the day of the operation or the day after. Although not formally

demonstrated in our relatively small study population, one can expect that the incidence

of catheter infections will be lowered in consequence of such standardization and de-

crease of overall catheter times [14]. Time of chest drainage removal did not show a

decline in variation in CP patients in the same pronounced manner. The reason is that

our CP foresees removal not on a specific day but rather when daily drainage is below a

defined volume. This specification was made to avoid complications caused by serotho-

rax. In contrast, a CP for VATS in Japan stipulated chest drain removal on postoperative

day four regardless of drainage volume. Unfortunately, it was not reported if this led to a

higher incidence of serothorax and concomitant complications [10].

Pain management is an element of treatment quality with particularly high relev-

ance for patients. A recently conducted study showed that enhanced recovery schemes

allow for timely discharge with no significant pain [15]. We had a similar experience in

our patients where pain levels were low throughout the hospital stay. This facilitated in-

tensified mobilization and respiratory exercising, which in turn can be assumed to pre-

vent pneumonia and thrombotic events [16]. There was no relevant pain observed upon

discharge. This underpins that CPs with enhanced recovery schemes do not lead to has-

tened discharge of unfit patients if sound perioperative standards are established and

adhered to.

As in the other available comparative studies [6;7;9;10], CP introduction led to a

significant shortening of hospital stay. In our case the reduction was particularly pro-

nounced, probably due to the meticulous integration of an enhanced recovery (“fast

track”) scheme into the CP. Frequently, such schemes are viewed with skepticism due to

fears of a higher incidence of postoperative complications. In line with findings from a

large observational series [12] a recently published RCT comparing “fast track” and con-

ventional perioperative care for lung resections demonstrated the opposite by showing

significantly reduced morbidity in the “fast track” arm [17]. In our patients, we also ob-

4

served a reduction in morbidity, which did however not reach statistical significance be-

cause of low case numbers. There seems to be a relevantly higher incidence of post-

operative pneumothorax and/or skin emphysema in the pre-pathway group. From our

data alone, it is not clear if this finding is merely due to chance or if the changes in chest

suction drainage management brought along with CP introduction played a role and led

to decreased morbidity. Likewise, there were no wound infections in the CP group, which

might be a result of improved management of invasive catheters and faster discharge.

These possible associations need to be assessed in studies with more statistical power.

Length of stay significantly differed between patients undergoing lung resection

and pleurodesis due to prolonged pleural fluid production in the latter, a phenomenon

which is well known [18]. Therefore, the relative reduction of hospital stay was smaller in

this group. Compared to regional and national standards, length of stay in the pre-

pathway group was above average whereas in the CP group it was roughly comparable

to that in similar facilities [18;19]. Compared to other countries like the US [20;21], length

of stay in both study groups appears exceedingly long. On the other hand side, in coun-

tries like Japan length of stay following VATS before and after CP implementation was

reported to be much longer than in our study [10]. This shows that there are structural

disparities in health systems and care mentalities which make direct cross-country com-

parisons of length of stay extremely difficult.

In addition to the evaluation of medical effects, we estimated the CP’s impact on

length-of stay dependent cost. We found a significant decrease of hospital costs as a

direct consequence of shortened length of stay. The magnitude of saved costs is roughly

in line with that reported in other studies [6;7;9;10]. Relative stay reduction and cost sav-

ing depended on the type of VATS. A much more pronounced effect was observed in

patients undergoing VATS wedge resection. Patients requiring pleurodesis had a longer

overall length of stay and thus the effect was smaller. This fact should incite a critical

consideration if two distinct CPs, but also two distinct DRGs, are needed for pleurodesis

and non-pleurodesis VATS. As an institutional consequence, we instituted a separate

CP for pleurodesis which foresees discharge one day later than in patients with non-

pleurodesis VATS.

The main limitation of our study is its non-randomized design with one retrospec-

tive group. Albeit desirable, the evaluation of CPs is virtually not feasible through rando-

mization within institutions. Once staff is trained to use specific elements of care as sti-

pulated in a CP, single patients can hardly be treated without applying these elements

even though they are formally not on a CP. In fact, to our knowledge there is only one

RCT assessing CPs in surgery [22]. Therefore, as in all other studies, the only viable

5

option was a “before-after” design, even though it bears the risk of structural inequality

between study groups. We tested for such inequality by comparing age, sex, type of sur-

gery and ASA status between groups and could not find significant differences. Howev-

er, there might have been residual differences in perioperative risk not addressed by

these variables. In addition, the frequency of the single types of VATS and the underly-

ing diagnoses differed significantly between the two groups. A drawback of the retros-

pective design is that secular changes in treatment quality occurring independently from

CP implementation might have biased our results. Data in our study are available only

for the first six months after CP implementation. Thus, case numbers and the power to

detect significant differences in perioperative and outcome quality are relatively low. In

analogy to technical procedures in surgery such as laparoscopy [23], there might be a

“learning curve” also for newly established care standards like CPs. If this was the case,

it would mean that implementation of our CP shows its full effects only after years and

the effects measured in this analysis might underestimate the real impact. Lastly, our

cost analysis is merely an approximation. Per diem length-of-stay-dependent costs were

estimated based on data of patients from one single year. Real costs for specific ele-

ments of care might have undergone a secular change unrelated to CP implementation,

which would have not been reflected in our analysis. In addition, we assumed that on

every day of hospital stay, on average the same length-of-stay-dependent cost was ac-

crued. This disregards a potential condensation of costs in the early postoperative phase

e.g. due to a more intensive need of nursing. The strength of our study is that we ana-

lyzed patients on an intention-to-treat basis. Even if certain goals of the CP such as re-

moval of a catheter at a given day were not met, patients were not taken “off the path-

way”. CP treatment was planned for all consecutive patients undergoing VATS, and all

patients were included in the analysis. Thus, selection bias in the CP group can be ruled

out. We believe that this approach is the best one to evaluate the true clinical value of

CPs because it is the closest representation of clinical reality, where CPs are meant to

be a tool for the treatment of all patients with a given condition or intervention and not

only one used for a highly selected group of patients who meet certain criteria.

Conclusion

Our study showed that perioperative quality of care, measured by selected impor-

tant parameters of perioperative VATS treatment, improved after implementation of a

multidisciplinary CP. In addition, we observed a significant cost reduction after CP im-

plementation. In the light of these findings we recommend the use of CPs for selected

interventions in thoracic surgery as an instrument to improve care. However, before

6

drawing definite conclusions, we recommend that further studies with more statistical

power and, ideally, two prospectively assessed groups, are conducted.

Acknowledgments

We thank Bärbel Rothhaar for providing financial data and Brigitte Pagel for valu-

able support in teaching nursery staff and explaining the importance of process im-

provement.

7

Table 1: Characteristics of the two study groups

Patient Characteristics

CP group

(n=34)

April-October 2007

Pre-pathway group

(n=77)

January 2005 - December 2006

Median age (years) 66 60

Gender (percent)

male 26 (76.5) 52 (67.5)

female 8 (23.5) 25 (32.5)

Underlying disorder

chronic pleural effusion 11 (32.4) 9 (11.7)

malignancy 13 (38.2) 32 (41.6)

pneumothorax / bullae 3 (8.8) 22 (28.5)

other 7 (20.6) a 14 (18.2)

a

Type of surgery (percent)

VATS wedge resection 23 (67.6) 68 (88.3)

VATS pleurodesis 11 (32.4)a 9 (11.7)

a

ASA class (percent)

1 1 (2.9 ) 14 (18.2)

2 11 (32.4) 26 (33.8)

3 22 (64.7) 36 (46.7)

4 0 1 (1.3)

ASA: American Society of Anaesthesiology

ap<0.05

Table 2: Parameters of outcome quality in the two study groups

CP Group

(n=34)

April-October 2007

Pre-pathway Group

(n=77)

January 2005 - December 2006

p-value

n % n %

30-day mortality

0

0

2

2.6

1.00

Morbidity

pneumothorax/skin

emphysema

wound infection

cardiopulmonary

other

3

1

0

1

1

8.8

2.9

0

2.9

2.9

21

11

4

2

4

27.3

14.3

5.2

2.6

5.2

0.08

Revision surgery

1 2.9 0 0 1.00

Re-admission 0 0 1 1.3 1.00

CP: clinical pathway

Table 3: Length of hospital stay in days in the two study groups

VATS: video-assisted thoracoscopic surgery

CP group

(n=34)

April-October

2007

Pre-pathway group

(n=77)

January 2005 - Decem-

ber 2006

Reduction

absolute in days

(relative in %)

All 9 14 5 (36)

VATS wedge resection 8 13 5 (39)

VATS pleurodesis 14 19 5 (26)

Table 4: Length-of-stay dependent hospital costs in the two study groups

CP group

(n=34)

April-October 2007

Pre-pathway group

(n=77)

January 2005 - December 2006

Saving per patient

absolute

(relative in %)

All

€2718

€4228

€1510 (36)

VATS wedge resection €2416 €3926 €1510 (39)

VATS pleurodesis €4228 €5738 €1510 (26)

VATS: video-assisted thoracoscopic surgery

Figure 1: Day of removal of invasive catheters. 2005/6: prepathway group; CP: standard

stipulated in the Clinical Pathway; 2007: CP group. Bold horizontal bars indicate median, box

indicates interquartile range, “Whiskers” indicate 5th and 95th percentile and small asterisks

and circles indicate outliers..

Figure 2: Sequential visualization of steps in the development and implementation of the CP.

References

1. Donabedian A. The quality of care. How can it be assessed? JAMA

1988;260(12):1743-8.

2. Maessen JM, Dejong CH, Kessels AG et al. Length of Stay: An Inappropriate Rea-

dout of the Success of Enhanced Recovery Programs. World J Surg 2008;32(6):971-

5.

3. Goodney PP, Siewers AE, Stukel TA et al. Is surgery getting safer? National trends in

operative mortality. J Am Coll Surg 2002;195(2):219-27.

4. Canales MG, Macario A, Krummel T. The surgical suite meets the new health econ-

omy. J Am Coll Surg 2001;192(6):768-76.

5. Ronellenfitsch U, Rössner E, Jakob J et al. Clinical Pathways in surgery-should we

introduce them into clinical routine? A review article. Langenbecks Arch Surg

2008;393(4):449-57.

6. Zehr KJ, Dawson PB, Yang SC et al. Standardized clinical care pathways for major

thoracic cases reduce hospital costs. Ann Thorac Surg 1998;66(3):914-9.

7. Wright CD, Wain JC, Grillo HC et al. Pulmonary lobectomy patient care pathway: a

model to control cost and maintain quality. Ann Thorac Surg 1997;64(2):299-302.

8. Vigneswaran WT, Bhorade S, Wolfe M et al. Clinical pathway after lung transplanta-

tion shortens hospital length of stay without affecting outcome. Int Surg

2007;92(2):93-8.

9. Soria V, Coll A, Flores B et al. Clinical pathway for the treatment of primary sponta-

neous pneumothorax in a general surgery department. Am J Med Qual

2005;20(5):268-76.

2

10. Maruyama R, Miyake T, Kojo M et al. Establishment of a clinical pathway as an effec-

tive tool to reduce hospitalization and charges after video-assisted thoracoscopic

pulmonary resection. Jpn J Thorac Cardiovasc Surg 2006;54(9):387-90.

11. Campbell H, Hotchkiss R, Bradshaw N et al. Integrated care pathways. BMJ

1998;316(7125):133-7.

12. McKenna RJ, Jr., Mahtabifard A, Pickens A et al. Fast-tracking after video-assisted

thoracoscopic surgery lobectomy, segmentectomy, and pneumonectomy. Ann Thorac

Surg 2007;84(5):1663-7.

13. Kehlet H. Fast-track colorectal surgery. Lancet 2008;371(9615):791-3.

14. Curtis LT. Prevention of hospital-acquired infections: review of non-pharmacological

interventions. J Hosp Infect 2008;69(3):204-19.

15. Gregor JI, Schwenk W, Mall J et al. ["Fast-track" rehabilitation in thoracic surgery.

First experiences with a multimodal, interdisciplinary, and proven perioperative treat-

ment course]. Chirurg 2008;79(7):657-64.

16. Stiller KR, Munday RM. Chest physiotherapy for the surgical patient. Br J Surg

1992;79(8):745-9.

17. Muehling BM, Halter GL, Schelzig H et al. Reduction of postoperative pulmonary

complications after lung surgery using a fast track clinical pathway. Eur J Cardiotho-

rac Surg 2008;34(1):174-80.

18. Schwarz RE, Posner MC, Ferson PF et al. Thoracoscopic techniques for the man-

agement of intrathoracic metastases. Results. Surg Endosc 1998;12(6):842-5.

19. Winter H, Meimarakis G, Pirker M et al. Predictors of general complications after vid-

eo-assisted thoracoscopic surgical procedures. Surg Endosc 2008;22(3):640-5.

3

20. Tovar EA, Roethe RA, Weissig MD et al. One-day admission for lung lobectomy: an

incidental result of a clinical pathway. Ann Thorac Surg 1998;65(3):803-6.

21. Ueda K, Kaneda Y, Sakano H et al. Obstacles for shortening hospitalization after vid-

eo-assisted pulmonary resection for lung cancer. Ann Thorac Surg 2003;76(6):1816-

20.

22. Delaney CP, Zutshi M, Senagore AJ et al. Prospective, randomized, controlled trial

between a pathway of controlled rehabilitation with early ambulation and diet and tra-

ditional postoperative care after laparotomy and intestinal resection. Dis Colon Rec-

tum 2003;46(7):851-9.

23. Avital S, Hermon H, Greenberg R et al. Learning curve in laparoscopic colorectal sur-

gery: our first 100 patients. Isr Med Assoc J 2006;8(10):683-6.