F

ML

tmBorGsw(itiwpta7sahopndMrpBcfiracc

KS

CLIN

ICA

LA

T

GASTROENTEROLOGY 2011;141:872–880

Faster Recovery of Gastrointestinal Transit After Laparoscopy andFast-Track Care in Patients Undergoing Colonic SurgerySJOERD VAN BREE,* MALAIKA VLUG,‡ WILLEM BEMELMAN,‡ MARKUS HOLLMANN,§ DIRK UBBINK,‡,�

KOOS ZWINDERMAN,¶ WOUTER DE JONGE,* SUSANNE SNOEK,* KAREN BOLHUIS,* ESMERIJ VAN DER ZANDEN,*RANS THE,* ROEL BENNINK,# and GUY BOECKXSTAENS*,**

Departments of *Gastroenterology, ‡Surgery, §Anesthesiology, �Quality Assurance & Process Innovation, ¶Clinical Epidemiology and Biostatistics, and #Nuclear

edicine, Academic Medical Center, Amsterdam, The Netherlands; and **Department of Gastroenterology, University Hospital Leuven, Catholic University of Leuven,

euven, Belgium

gapvts

snag

ca

gss

BACKGROUND & AIMS: Postoperative ileus is charac-erized by delayed gastrointestinal (GI) transit and is a

ajor determinant of recovery after colorectal surgery.oth laparoscopic surgery and fast-track multimodal peri-perative care have been reported to improve clinicalecovery. However, objective measures supporting fasterI recovery are lacking. Therefore, GI transit was mea-

ured following open and laparoscopic colorectal surgeryith or without fast-track care. METHODS: Patients

n � 93) requiring elective colonic surgery were random-zed to laparoscopic or conventional surgery with fast-rack multimodal management or standard care, resultingn 4 treatment arms. Gastric emptying and colonic transitere scintigraphically assessed from days 1 to 3 in 78atients and compared with clinical parameters such asime to tolerance of solid food and/or bowel movementnd time until (ready for) discharge. RESULTS: A total of1 patients without mechanical bowel obstructions orurgical complications requiring intervention were avail-ble for analysis. No differences in gastric emptying 24ours after surgery between the different groups werebserved (P � .61). However, the median colonic transit ofatients undergoing laparoscopic/fast-track care was sig-ificantly faster compared with the laparoscopic/stan-ard, open/fast-track, and open/standard care groups.ultiple linear regression analysis showed that both lapa-

oscopic surgery and fast-track care were significant inde-endent predictive factors of improved colonic transit.oth were associated with significantly faster clinical re-overy and shorter time until tolerance of solid food andrst bowel movement. CONCLUSIONS: Colonic transitecovers significantly faster after laparoscopic surgerynd the fast-track program; laparoscopy and fast-trackare lead to faster recovery of GI motility and improvelinical recovery.

eywords: Postoperative Ileus; Fast Track; Laparoscopicurgery; Gastrointestinal Transit.

Each patient undergoing an abdominal surgical pro-cedure will develop a transient episode of impaired

gastrointestinal (GI) motility or postoperative ileus. Im-portantly, postoperative ileus is the single most important

determinant of hospital stay after abdominal surgery and

consequently significantly contributes to postoperativemorbidity and hospitalization costs.1,2 Laparoscopic sur-

ery and the implementation of an enhanced recoveryfter surgery program, also referred to as fast-track (FT)erioperative care, are the 2 most important recent ad-ances in modern surgical care. Both have been reportedo be safe and effective and to result in a shorter hospitaltay with earlier recovery of GI function3– 6 and less mor-

bidity compared with open colorectal surgery and stan-dard care.7–10

FT programs in colonic surgery were introduced toreduce surgical stress response, organ dysfunction, andmorbidity, thereby promoting a faster recovery after sur-gery.11,12 This multimodal perioperative care strategy con-titutes a multidisciplinary approach involving dieticians,urses, surgeons, and anesthesiologists. It mostly focuses ondequate perioperative fluid management, optimized anal-esia, early oral nutrition, and early mobilization.6,13,14

It should be emphasized that most studies assessingpostoperative ileus have used rough clinical parameterssuch as return of bowel sounds and time to first flatus ordefecation as primary outcomes. These parameters are,however, not only difficult to assess accurately, but it isalso questionable to what extent they reflect restoration ofGI motility. For example, the occurrence of first flatusmay reflect emptying of rectal gas rather than recovery ofcolonic transit. To date, only 2 clinical studies comparinglaparoscopic and open colectomies objectively measuredpostoperative GI motility and at best showed only a mod-est increase in the recovery of GI motility.15,16 Studiesomparing the effect of FT care with standard postoper-tive care on GI transit, however, are lacking.

Although several methods have been used, scinti-raphic recording of GI transit is considered the goldtandard. This technique is not only widely used to mea-ure gastric emptying, it also allows accurate evaluation of

Abbreviations used in this paper: CI, confidence interval; FT, fasttrack; GC, geometrical center; IQR, interquartile range; GI, gastrointes-tinal; IQR, inter-quartile ranges; Lap/FT, laparoscopic colectomy withfast-track care; Lap/Standard, laparoscopic colectomy with standardcare; Open/FT, open colectomy with fast-track care; Open/Standard,open colectomy with standard care; POD, postoperative day.

© 2011 by the AGA Institute0016-5085/$36.00

doi:10.1053/j.gastro.2011.05.034

uj

aoihonapEspP

faGtrussdpd

wpdtlattrp

swpuc

t

(mhHTidpii

tMPifithstic

CLI

NIC

AL

AT

September 2011 GI TRANSIT AFTER LAPAROSCOPY AND FAST TRACK 873

motility throughout the entire GI tract in a safe andnoninvasive way.17–19 In addition, it has been successfully

sed to evaluate therapeutic interventions in healthy sub-ects and patients with colonic motility disorders.19 –21

Therefore, in the present study, scintigraphy was se-lected to objectively evaluate whether laparoscopy, FTcare, or the combination of both result in faster recov-ery of GI transit after colonic surgery.

Patients and MethodsPatients and EligibilityPatients enrolled in the Academic Medical Center,

Amsterdam, as part of the LAFA (LAparoscopy and/or FAsttrack multimodal management versus standard care) multi-center trial were eligible for this study.22 Patients between 40

nd 80 years of age with an American Society of Anesthesi-logists (ASA) status less than IV were invited to participate

f they were to undergo elective segmental colectomy foristologically confirmed adenocarcinoma or adenoma with-ut evidence of metastatic disease. Exclusion criteria wereeoadjuvant radiotherapy, prior midline laparotomy, unavail-bility of a laparoscopic surgeon, emergency surgery, or alanned stoma. The protocol was approved by the Medicalthics Review Board of the Academic Medical Center, Am-terdam. The study was conducted in accordance with therinciples of the Declaration of Helsinki and Good Clinicalractice Guidelines.

Study DesignA 2 � 2 balanced factorial block design was used to

randomize patients to laparoscopic or open colectomy and tothe FT program or standard care by means of an Internetrandomization module. This resulted in 4 treatment groups: (1)laparoscopic colectomy with FT care (Lap/FT), (2) open colec-tomy with FT care (Open/FT), (3) laparoscopic colectomy withstandard care (Lap/Standard), and (4) open colectomy withstandard care (Open/Standard).

Twenty-four hours after surgery, patients underwent a solidgastric emptying test (99mTc labeled pancake, 115 kcal) directlyollowed by the ingestion of 60 mL of indium-111–labeled water tossess colonic transit on postoperative days (POD) 2 and 3 (seeI Transit Studies for details). During hospital admission, pa-

ients were visited at least once daily by a trial nurse and/or aesearch physician for clinical evaluation. Clinical symptoms ofpper and lower GI motility were evaluated daily via self-de-igned questionnaires. Patients were assisted with filling in thiself-assessment sheet daily at the time of scintigraphy untilischarge. Until discharge, nursing staff reported daily on theatient’s progress, that is, intake, passage of stool, and pre-efined discharge criteria were checked.

To avoid crossover treatment by the nursing staff, patientsere admitted either to a ward providing standard care or a wardroviding FT care. The treatment protocols are described inetail in Supplementary Table 1.22 Open surgery was performedhrough a midline laparotomy. The laparoscopic approach was aaparoscopic-assisted technique comprising a medial-to-lateralpproach with intracorporeal devascularization and mobiliza-ion. In right colectomies, the anastomosis was performed ex-racorporeal; in left colectomies, intracorporeal after extracorpo-eal insertion of the anvil in the afferent colon loop was

erformed. All anastomoses were performed using mechanical

taplers. At the end of the surgery, the abdomen was coveredith a large dressing to hide abdominal incisions to blind theatients and nurses on the ward. This dressing was kept in placentil the day of discharge unless wound complaints or compli-ations were suspected.

OutcomesThe primary outcome of efficacy was colonic transit on

day 3, depicted as geometric center (GC) of intracolonic mass 48hours postprandially of 111In-DTPA labeled water.

The secondary end points of this study were (1) gastric reten-ion at 24 hours after surgery, formulated as the percentage of

99mTc-labeled pancake present in the stomach 2 hours afteringestion; (2) time until first bowel movement; (3) time untilfirst tolerance of solid food; (4) the composite outcome of timeto tolerate solid food and bowel movement; (5) length of hos-pital stay; and (6) time until ready for hospital discharge.

GI Transit StudiesOn POD 1, patients underwent a solid gastric empty-

ing test. Two hours after ingestion of a 99mTc-labeled pancake115 kcal), a 5-minute acquisition was performed in a 128

atrix with the patient in a supine position using a single-ead gamma camera (Siemens Orbiter or Diacam; Siemens,offman Estates, IL) fitted with a medium energy collimator.he relative gastric content of the pancake 2 hours after

ngestion was calculated as previously described.23 Briefly, toepict the percentage of activity present in the stomach com-ared with the total activity in the abdominal region of

nterest, the counts in the stomach were divided by the countsn the complete abdominal region, corrected for background.

Directly after completion of the gastric emptying test, pa-ients were asked to drink 60 mL of tap water labeled with 4

Bq 111In-DTPA (diethylenetriaminepentaacetate; Covidien,etten, The Netherlands). A cobalt marker was placed on the

liac crest for anatomic reference, and a baseline scan was per-ormed to determine the total amount of indium activity presentn the abdomen 5 minutes after ingestion. To determine colonicransit, two 5-minute acquisitions were performed 24 and 48ours after ingestion of the radiolabeled water using the sameingle-head gamma camera. To enable calculation of colonicransit, the gut was subdivided into 3 segments (ie, 0 � smallntestine; 1 � proximal colon; 2 � distal colon; 3 � stool). Theenter of mass model24 was applied, expressing colonic transit as

24- and 48-hour postprandial GC of activity. The primary vari-able of interest in overall colonic transit was the GC at 48 hours.The GC is the weighted average of counts in the different colonicregions: proximal colon, distal colon, and stool. Quantificationof the counts in each region was performed using a HermesGold software program (Hermes Medical Solutions, Stockholm,Sweden). At any time, the portion of colonic counts in eachcolonic region (corrected for background activity and isotopedecay) was multiplied by the corresponding weighting factors asfollows: GC � (% Proximal Colon � 1 � % Distal Colon � 2 �% Stool � 3)/100. Thus, a high GC implies faster colonic transit.A GC of 0 implies that none of the isotope has reached thecolon, and a GC of 3 implies that all of the isotope is in thestool. The amount of 111In-DTPA tracer defecated before day 4was computed by subtraction of decay-corrected abdominalcounts on day 3 from total abdominal counts on day 1. Inter-pretation and calculation of gastric retention and colonic transitwas performed in a blinded manner by 2 researchers indepen-

dently (S.v.B. and R.B.) on a Hermes workstation.

tdsot(ew7ssotowKecdpLaB0

b

tlbsa(

D

ana

CLIN

ICA

LA

T

874 VAN BREE ET AL GASTROENTEROLOGY Vol. 141, No. 3

Sample Size Calculation and StatisticalAnalysisBecause resumption of colonic motility occurs on PODs

3 through 5 and typically is the rate-limiting factor for theresolution of ileus (reviewed in Benson and Wingate25), thecolonic transit on day 3 was used as the primary efficacy param-eter. Sample size calculation was based on earlier studies on GItransit after abdominal surgery15,23 and indicated that 18 pa-ients per group were required to detect a significant (P � .05)ifference in the median colonic transit between groups, corre-ponding to a probability of 0.22 (or less) that the colonic transitf a patient in the open surgery group is slower than the colonicransit of a patient in the laparoscopy group using a WilcoxonMann–Whitney) rank-sum test, providing 80% power. A similarffect on colonic transit day 3 was expected for FT care comparedith standard care, and a total sample size of 2 � 2 � 18 �2 was expected to provide sufficient power for both compari-ons. Per-protocol analysis was applied on all data because thistudy was designed and powered to detect a difference in post-perative GI transit between patients who actually received thereatment of interest and had no major surgical complicationsf the intestinal tract. Data were nonnormally distributed andere expressed as median values and interquartile range (IQR). Aruskal–Wallis test was performed to identify statistical differ-

nces among the 4 treatment groups. A 2-sided P value �.05 wasonsidered statistically significant. If a significant difference wasetected, the Mann–Whitney U test was used to conduct the 5airwise comparisons among the 4 treatment groups, that is,ap/Standard vs Open/Standard, Open/FT vs Open/Standard,nd Lap/FT to each of the other 3 groups. Holm’s sequentialonferroni method was used to control for type I error at the.05 level across the comparisons.

For categorical variables, treatment groups were comparedy means of the �2 test or Fisher exact test. For continuous

normally distributed data, the analysis of variance test wasused and data are presented as means � SD.

Multiple linear regression analysis was used to assess theeffect of laparoscopy, FT care, and the combination of both onthe primary and secondary end points. Because the measure-ments of the end points were skewed, these data were logtransformed. First, the 2 main effects (laparoscopy and FT care)were incorporated in the multiple linear regression model, andsubsequently the interaction term (laparoscopy * FT care) wasadded. The interaction term was not incorporated in the final

Table 1. Baseline Characteristics and Surgical Aspects of the

Lap/FT(n � 18)

Lap(n

Age (y)a 65 � 9.4 6Male sex (%) 50Body mass index (kg/m2)a 26.5 � 4.4 24.ASA grade I or II (%) 89Comorbidity (%) 72Type of colectomy (%)

Right-sided 28Left-sided 6Sigmoid 67

uration of surgery, median (IQR) 191 (176–205) 179

ASA, American Society of Anesthesiologists.aValues are mean � SD.bAnalysis of variance, c�2 test, and dKruskal–Wallis test. Per-protocol

model if no significant contribution occurred. Because we log

ransformed the end points, the antilog of the parameters of theinear regression model can be interpreted as the fold increaseetween the geometric means of the laparoscopic versus openurgery groups and FT versus standard care groups. Statisticalnalysis was performed using SPSS for Windows version 17SPSS Inc, Chicago, IL).

ResultsPatientsBetween October 2005 and August 2009, a total of

93 patients were randomly assigned to one of the 4 treat-ment groups. Fifteen patients (n � 1 Lap/FT; n � 3Lap/Standard; n � 6 Open/FT; n � 5 Open/Standard)were withdrawn for various reasons: resection of the smallintestine (n � 2), delirium (n � 1), fascial dehiscence (n �1), metastasis at time of admission (n � 4), protocolviolation (n � 5), and withdrawal of informed consent(n � 2). GI transit was assessed in 78 patients. Seven ofthese had mechanical bowel obstructions or surgical com-plications requiring intervention within the first 5 days: 3Lap/FT patients (internal herniation, n � 2; adhesion,n � 1); 1 Lap/Standard patient (anastomotic leak); 1Open/FT patient (anastomotic leak); and 2 Open/Standard patients (intraperitoneal bleed and adhesion).Therefore, data of 71 patients without major surgical com-plications were available for per-protocol analysis (Supple-mentary Figure 1). In 1 patient, laparoscopy had to be con-verted to open colectomy because of the extent of tumorinvasion. Baseline characteristics and surgical aspects didnot differ significantly among the 4 treatment groupsexcept duration of surgery (P � .001) (Table 1).

Gastric Retention Day 1The residual gastric content 2 hours after inges-

tion of the pancake varied from almost complete empty-ing to gastric stasis with more than 99% of the radiola-beled pancake still present in the stomach. Multiple linearregression analysis showed no effect of laparoscopy or FTcare on gastric retention (Table 2). The median gastricretention did not differ between groups (P � .61, Kruskal–

cluded Patients per Group

andard17)

Open/FT(n � 18)

Open/Standard(n � 18) P

10.2 64 � 10.1 66 � 6.9 .920b

3 61 61 .872c

3.9 24.6 � 3.5 25.0 � 3.6 .246b

8 89 89 1.000c

3 56 50 .535c

.337c

5 28 394 17 222 56 397–202) 146 (122–180) 139 (122–165) �.001d

lysis of 71 patients.

In

/St�

6 �5

0 �85

513

(15

Wallis test) (Figure 1A). Gastric retention was 70% (IQR,

g

tdhPpmm2SPP

tr1c1

Am

bsfl2amvTptf

P

S

NTrmwv

CLI

NIC

AL

AT

September 2011 GI TRANSIT AFTER LAPAROSCOPY AND FAST TRACK 875

36%–94%), 81% (34%–95%), 58% (26%–71%), and 58%(31%–100%) in patients treated with Lap/FT, Lap/Stan-dard, Open/FT, and Open/Standard, respectively.

Colonic TransitThe median calculated GCs of activity at day 2 are

shown in Figure 1B. Colonic transit on day 2 was signif-icantly faster in the Lap/FT group (2.2 [1.8 –2.6]) com-pared with the Lap/Standard (1.3 [1.2–2.0], P � .015),Open/FT (1.5 [1.2–1.9], P � .008), and Open/Standard

roup (1.1 [1.0 –1.4], P � .001).These data indicate that most of the radiolabeled ma-

erial was still located in the proximal part of the colon atay 2 in the latter 3 groups, while in the Lap/FT group itad already moved to the distal part of the colon. OnOD 3, most of the radiolabel was still present in theroximal colon in the Open/Standard group but hadoved into the distal colon in the other 3 groups. Theedian GC for patients in the Lap/FT group (2.6 [2.0 –

.9]) was significantly higher compared with the Lap/tandard (2.2 [1.6 –2.5], P � .044), Open/FT (2.0 [1.6 –2.4],� .020), and Open/Standard care groups (1.3 [1.0 –1.5],� .001) (Figure 1B).Multiple linear regression analysis indicates that colonic

ransit on day 3 was significantly influenced by both lapa-oscopy (fold increase of 1.48; confidence interval [CI], 1.24–.76; P � .001; ie, the geometric mean colonic transit in-reased by 48% [CI, 24%–76%]) and FT care (fold increase of.38; CI, 1.16–1.65; P � .001; ie, geometric mean colonic

transit increased by 38% [CI, 16%–65%] compared with pa-tients receiving standard care). However, the combination ofboth (ie, the interaction of laparoscopy with FT care) was notidentified to add any benefit (Table 2).

Defecation of Radiolabeled FecesIn addition to assessment of colonic transit using

Table 2. Results of Multiple Linear Regression Analysis

Laparos

P valueFac

(fold inc

rimary end pointColonic transit day 3 �.001 1.4

econdary end pointsGastric retention day 1 .407 1.2Time until first defecation .102 0.6Time until tolerance of solid food .012 0.6Time until tolerance of food and defecation �.001 0.6Days until discharge .005 0.7Days until ready for discharge .001 0.6

OTE. The multiple linear regression model incorporated the 2 main ehe interaction term did not show a significant additional effect on anyegression model. Per-protocol analysis of 71 patients. Linear regresation of the dependent variable. The presented factors (fold increashich may be interpreted as the fold increase of the geometric means os standard care group, respectively.

GC calculation, we also measured the amount of radiola- m

beled feces evacuated 48 hours after intake of the 111In-DTPA tracer. The median percentage of 111In-DTPA tracerdefecated was 70% (19%–94%) in the Lap/FT-treatedgroup compared with 37% (0 –58%), 26% (0 –57%), and 0%(0 – 0) in the Lap/Standard-, Open/FT-, and Open/Stan-dard-treated groups, respectively (Figure 1C).

Clinical Signs of Recovery and ClinicalEvaluationThe median time until the composite outcome

of time until first defecation and tolerance of solid foodwas significantly shorter for patients who underwentLap/FT treatment compared with the other groups(Figure 2). Laparoscopic resection and FT care indepen-dently resulted in a significantly shorter time untiltolerance of solid food and the composite of first def-ecation and tolerance of solid food (Table 2). Thecombination (ie, the interaction) of both was not iden-tified to add any benefit.

There was no significant difference on a per-protocolbasis between the groups in readmission rate (P � .850)and overall morbidity (P � .217) until 30 days after surgery.

list of the complications categorized by grade and treat-ent group is available in Supplementary Table 2.Individualized perioperative fluid management was

ased on dynamic preload optimization and led toimilar (P � .092, Kruskal–Wallis test) intraoperativeuid loading in all 4 treatment groups with a median of.0 L (IQR, 1.3–3.0). Overall compliance of surgical,nesthesiologic, and nursing care personnel with theultimodal perioperative rehabilitation pathway was

ery good. Fifteen FT items were evaluated per patient.he following items were scored if successfully applied:reoperative counseling, omission of bowel prepara-ion, intake of carbohydrate-loaded drinks the day be-ore surgery, intake of carbohydrate-loaded drinks the

Treatment effect

y vs open Fast track vs standard care

se) 95% CI P valueFactor

(fold increase) 95% CI

1.236–1.762 .001 1.380 1.159–1.648

0.764–1.923 .693 0.912 0.575–1.4490.431–1.081 .042 0.619 0.392–0.8320.489–0.912 .003 0.612 0.449–0.8360.455–0.791 �.001 0.601 0.457–0.7930.598–0.908 .017 0.774 0.628–0.9530.500–0.839 .034 0.753 0.582–0.977

cts (laparoscopy and FT) and the interaction term (laparoscopy * FT).he dependent variables and was removed from the final multiple linear

analysis of all end points was performed after logarithmic transfor-nd CI represent the back-transformed (antilog) parameter estimates,e end points of the laparoscopy group vs the open group and FT group

cop

torrea

76

138267003847

ffeof tsione) af th

orning before surgery, no preoperative fasting since

1(

fo

dMsszib

lmwgma

tadswLOui3pM

CLIN

ICA

LA

T

876 VAN BREE ET AL GASTROENTEROLOGY Vol. 141, No. 3

midnight, omission of premedication, thoracic epiduralanalgesia, prevention of hypothermia, adequate periop-erative fluid loading, removal of nasogastric tube beforeextubation, omission of abdominal drains, suprapubiccatheter or no catheter, more than 500 mL of intake atPOD 0 including 200 mL carbohydrate-loaded drink,mobilization for at least 15 minutes at POD 0, andstarting with laxative at POD 1. In the Lap/FT group,12.5 � 1.5 of the 15 items and in the Open/FT group

0.9 � 2.6 items were successfully applied per patientSupplementary Table 3). Some FT items have also been

implemented in the standard care groups: in the Lap/Standard 5.2 � 1.1 items and in the Open/Standard4.8 � 0.7 items per patient (mean � SD). Other items(not mentioned in Supplementary Table 3) applied inall groups were prevention of hypothermia in 100%of patients, removal of the nasogastric tube beforeextubation in 89%, and omission of abdominal drainsin 87%.

Relationship Between Colonic Transit andClinical RecoveryColonic transit is considered to mainly deter-

mine clinical recovery.25 The latter (ie, time until readyor discharge) correlated significantly with colonic GCn day 3 (r � �0.56, P � .001; Spearman’s rank

correlation). The time until patients were ready fordischarge was significantly shorter in patients who un-derwent the Lap/FT treatment compared with theOpen/FT and Open/Standard Care groups. Similarly,length of stay for patients in the Lap/FT group (me-dian, 3.9 days) was significantly shorter compared withthe Open/FT (5.9 days, P � .002) and the Open/Stan-

ard care groups (6.0 days, P � .001) (Figure 2D and E).ultiple linear regression analysis showed that laparo-

copic resection and FT care resulted in a significantlyhorter time until ready for discharge and hospitali-ation (Table 2). However, the combination (ie, thenteraction) of both was not identified to add anyenefit.

4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™Figure 1. Postoperative GI transit. Scintigraphic evaluation of GI transit ofan 111In-labeled noncaloric liquid test meal and gastric emptying of a 99mTc-abeled pancake 24 hours after surgery. All data are expressed as the

edian � IQR in corresponding treatment groups. The Kruskal–Wallis testas performed to identify statistical differences among the 4 treatmentroups. The Mann–Whitney U test was used to compare the different treat-ent groups. P values were calculated using Holm’s sequential Bonferroni

djustment for multiple (n � 5) comparisons. P � .05 was consideredstatistically significant. NS, nonsignificant. Per-protocol analysis of 71 pa-tients: Lap�FT, n � 18; Lap�Stand, n � 17; Open�FT, n � 18;Open�Stand, n � 18. (A) Scintigraphic evaluation of gastric emptying of a99mTc-labeled pancake 24 hours after surgery. Gastric retention was deter-mined 2 hours after ingestion of a radiolabeled pancake and is depicted aspercentage in stomach compared with total abdominal region of interest.Bars represent percentage of gastric retention in the corresponding treat-ment group. Note that the Kruskal–Wallis test identified no statistical differ-ence among the 4 treatment groups. (B) Scintigraphic evaluation of colonicransit on PODs 2 and 3. Colonic transit was determined 24 hours (POD 2)nd 48 hours (POD 3) after ingestion of 60 mL radiolabeled tap water andepicted as GC of colonic contents. Bars represent the GC in the corre-ponding treatment group. Significant differences (Mann–Whitney U test)ere observed for colonic transit day 2 between Lap�FT and Lap�Stand,ap�FT and Open�FT, Lap�FT and Open�Stand, Lap�Stand andpen�Stand, and Open�FT and Open�Stand. P values were calculatedsing Holm’s sequential Bonferroni adjustment for multiple (n � 5) compar-

sons. (C) Amount of 111In-DTPA tracer defecated within 48 hours (on POD) after ingestion of 60 mL radiolabeled tap water. Data are depicted asercentage of the total amount of 111In-DTPA tracer ingested on day 1.ann–Whitney U test was used for statistical analysis.

atga

d, n

CLI

NIC

AL

AT

September 2011 GI TRANSIT AFTER LAPAROSCOPY AND FAST TRACK 877

DiscussionThis study provides objective data indicating that

laparoscopic surgery and FT care improve recovery of GItransit compared with open colectomy and standard care,contributing to faster clinical recovery. In a multiple lin-ear regression analysis, both laparoscopic surgery and FTcare were found to be significant predictors of improvedcolonic transit and reduced time to tolerance of solid foodand bowel movement. These data suggest that laparo-scopic resection and FT care lead to faster recovery of GItransit after colorectal surgery.

Previously reported clinical trials have shown a benefi-cial effect of laparoscopic procedures and FT care on theduration of postoperative ileus.6,11–13 In our study, clinical

Figure 2. Clinical (secondary) end points. (A) Time until first defecationnd (D) days until discharge and (E) days until ready for discharge for Laphe median � IQR in corresponding treatment group. The Kruskal–Wallisroups. The Mann–Whitney U test was used to compare the different treadjustment for multiple (n � 5) comparisons. P � .05 was considered sta

Lap�FT, n � 18; Lap�Stand, n � 17; Open�FT, n � 18; Open�Stan

hallmarks were not the primary outcome measures for

postoperative ileus because these are less accurate andreliable to objectively evaluate the effect of different treat-ment strategies. Parameters such as nausea, vomiting, andtolerance of solid food strongly depend on patient report-ing, whereas first passage of stool or flatus may simplyreflect rectal emptying and therefore do not necessarilyadequately reflect recovery of effective GI motility. In thepresent study, we therefore assessed GI transit (ie, gastricretention and colonic transit) after surgery using scintig-raphy. We observed major differences in colonic transitamong the 4 groups. Patients in the Lap/FT group inparticular had a significantly higher GC measured at 24and 48 hours after ingestion of Indium-111–labeled water.On the other hand, colonic transit was slowest in patients

) tolerance of solid food, (C) first defecation and tolerance of solid food,T, Lap�Stand, Open�FT, and Open�Stand. All data are expressed ast was performed to identify statistical differences among the 4 treatmentent groups. P values were calculated using Holm’s sequential Bonferronically significant. NS, nonsignificant. Per-protocol analysis of 71 patients:� 18.

, (B�Ftestmtisti

who underwent conventional surgery and postoperative

us

dset

croneer

Gliidttrsapiatm

CLIN

ICA

LA

T

878 VAN BREE ET AL GASTROENTEROLOGY Vol. 141, No. 3

care. In line with these data, 70% of the ingested 111Intracer was defecated 48 hours after ingestion in patientsin the Lap/FT group, whereas no 111In tracer had beenexcreted in patients in the Open/Standard group. Multi-ple linear regression analysis indeed indicated laparo-scopic surgery and FT care as predictive factors of fastercolonic transit. Our scintigraphic data support the out-come of the multicenter LAFA trial, that is, the main trialof which this study was a substudy. This study showed aquicker passage of first stool and tolerance of solid foodafter laparoscopy and FT care and showed that the com-bination of laparoscopic surgery with FT care resulted ina significantly faster recovery than all other treatmentcombinations.26 Moreover, we also observed shorter time

ntil the composite of first defecation and tolerance ofolid food.

Our study is the first providing objective scintigraphicata supporting improved clinical recovery after laparo-copic surgery and FT care. Only one study has previouslyvaluated GI transit after open and laparoscopic colec-omy within an FT care program.15 This study, however,

did not compare FT care with conventional perioperativecare. Similar to our study, Basse et al detected a signifi-cantly higher colonic transit in the laparoscopic group.However, no significant difference in the amount of 111Intracer defecated was reported. It should be emphasized,however, that the time window of 48 hours after surgerymay have been too short to detect differences in fecal111In tracer because colonic motility remains impaireduntil 48 to 72 hours after surgery.25

In contrast to improvement in colonic transit, nosignificant effect of laparoscopy or FT care was detectedon gastric emptying. Gastric emptying of a radiolabeledpancake was determined 24 hours after surgery andvaried considerably; some patients had emptied thestomach almost completely, whereas in others morethan 99% of the pancake was still present in theirstomach. Our data are in line with the study by Gelpi etal, which also failed to show differences in gastricemptying between the open- and laparoscopic-treatedpatients.27 A possible explanation for the discrepancybetween gastric and colonic transit data could be thefact that gastric emptying was assessed earlier aftersurgery. Twenty-four hours after surgery, one may an-ticipate that motility is still largely impaired irrespec-tive of the type of surgery or perioperative care. In otherwords, the effect of postoperative stress, pain, anesthe-sia, and other factors related to the surgical procedurein the early postoperative period most likely have amajor impact on GI function, overriding the potentialbenefit of laparoscopy or FT care. Alternatively, differ-ences in dosage of opioids or antiemetics in the peri-operative period may have obscured the beneficial effectof laparoscopy and FT care on gastric emptying. How-ever, the amount of analgesics and antiemetics did notdiffer between the different groups, making the latter

explanation less likely (Supplementary Table 4).

The pathophysiologic mechanisms involved in post-operative ileus are still not completely understood, butrecent studies have stressed the importance of inflam-mation of the intestinal muscularis resulting from sur-gical handling.28 –31 The faster clinical recovery observedafter laparoscopic surgery compared with open surgerycould be explained by decreased tissue trauma withconcomitant decreased mast cell activation leading toattenuated intestinal inflammation and thus a quickerGI recovery.28,29,31 The mechanisms behind the benefi-ial effect of the FT program remain unclear, but in aat model enteral feeding was shown to improve post-perative GI transit through activation of the vaguserve–mediated anti-inflammatory pathway.32 To whatxtent vagal activation in response to early feeding mayxplain improved transit in the FT group in our studyemains to be studied.

A limitation of our study is the omission of baselineI transit measurements. This design would have al-

owed comparison before and after surgery in eachndividual patient, possibly reducing the variability andncreasing the power of the study. However, such aesign would imply an even larger burden to our pa-ients and moreover would have required earlier hospi-alization of at least 24 hours to perform the baselineecording and avoid interference with the postoperativecintigraphy. Nevertheless, our study is the first tossess GI transit in a large cohort of postoperativeatients. Moreover, in contrast to previous stud-

es,15,23,27,33 early postoperative gastric emptying wasssessed after intake of a solid meal, which is superioro liquid emptying to demonstrate diminished gastric

otility.18,19,21 Ideally, to compare both gastric empty-ing and colonic transit, gastric emptying should havebeen measured at the same time colonic transit wasdetermined. In practice, performing additional gastricemptying tests on the clinically more relevant days 2and 3 after surgery is not possible because of technicaland ethical reasons. Secondly, compared with previousstudies, our reoperation rate was relatively high. Oneexplanation might be that surgery was performed in anacademic hospital where residents/fellows perform theprocedure assisted by a (laparoscopic) colorectal sur-geon. However, we do not expect that the relatively highreoperation rate has biased the comparisons on intes-tinal transit among the 4 groups because there was nosignificant difference in the reoperation rate betweenthe FT care and standard care groups or the laparo-scopic and open groups. Thirdly, although all measureswere installed to prevent the treating physicians andnurses on the ward from being informed on the type ofsurgery, it remains difficult to obtain a truly blindedstudy design. It should be emphasized, however, thatthe discharge criteria were strictly defined and the peri-operative care protocols were protocolized in great de-tail. The adherence to the treatment protocols washigh, thus limiting the influence of the attending phy-

sicians and nurses on postoperative care.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3CLI

NIC

AL

AT

September 2011 GI TRANSIT AFTER LAPAROSCOPY AND FAST TRACK 879

In conclusion, our study shows that laparoscopic sur-gery and the FT care program lead to a significantly fasterrecovery of GI transit. These data provide objective evi-dence confirming the faster clinical recovery with thisperioperative strategy.

Supplementary Material

Note: To access the supplementary materialaccompanying this article, visit the online version ofGastroenterology at www.gastrojournal.org, and at doi:10.1053/j.gastro.2011.05.034.

References

1. Augestad KM, Delaney CP. Postoperative ileus: impact of pharma-cological treatment, laparoscopic surgery and enhanced recoverypathways. World J Gastroenterol 2010;16:2067–2074.

2. Cali RL, Meade PG, Swanson MS, et al. Effect of morphine andincision length on bowel function after colectomy. Dis Colon Rec-tum 2000;43:163–168.

3. Delaney CP, Marcello PW, Sonoda T, et al. Gastrointestinalrecovery after laparoscopic colectomy: results of a prospective,observational, multicenter study. Surg Endosc 2010;24:653–661.

4. Story SK, Chamberlain RS. A comprehensive review of evidence-based strategies to prevent and treat postoperative ileus. Dig Surg2009;26:265–275.

5. Veldkamp R, Kuhry E, Hop WC, et al. Laparoscopic surgery versusopen surgery for colon cancer: short-term outcomes of a ran-domised trial. Lancet Oncol 2005;6:477–484.

6. Wind J, Polle SW, Fung Kon Jin PH, et al. Systematic review ofenhanced recovery programmes in colonic surgery. Br J Surg2006;93:800–809.

7. Clinical Outcomes of Surgical Therapy Study Group. A comparisonof laparoscopically assisted and open colectomy for colon cancer.N Engl J Med 2004;350:2050–2059.

8. Gouvas N, Tan E, Windsor A, et al. Fast-track vs standard care incolorectal surgery: a meta-analysis update. Int J Colorectal Dis2009;24:1119–1131.

9. Guillou PJ, Quirke P, Thorpe H, et al. Short-term endpoints ofconventional versus laparoscopic-assisted surgery in patients withcolorectal cancer (MRC CLASICC trial): multicentre, randomisedcontrolled trial. Lancet 2005;365:1718–1726.

0. Muller S, Zalunardo MP, Hubner M, et al. A fast-track programreduces complications and length of hospital stay after opencolonic surgery. Gastroenterology 2009;136:842–847.

1. Kehlet H. Fast-track colorectal surgery. Lancet 2008;371:791–793.

2. Basse L, Thorbol JE, Lossl K, et al. Colonic surgery with acceler-ated rehabilitation or conventional care. Dis Colon Rectum 2004;47:271–277.

3. Kehlet H, Wilmore DW. Multimodal strategies to improve surgicaloutcome. Am J Surg 2002;183:630–641.

4. Fearon KC, Ljungqvist O, Von MM, et al. Enhanced recovery aftersurgery: a consensus review of clinical care for patients undergo-ing colonic resection. Clin Nutr 2005;24:466–477.

5. Basse L, Madsen JL, Billesbolle P, et al. Gastrointestinal transitafter laparoscopic versus open colonic resection. Surg Endosc2003;17:1919–1922.

6. Hotokezaka M, Dix J, Mentis EP, et al. Gastrointestinal recoveryfollowing laparoscopic vs open colon surgery. Surg Endosc 1996;10:485–489.

7. Bonapace ES, Maurer AH, Davidoff S, et al. Whole gut transitscintigraphy in the clinical evaluation of patients with upper andlower gastrointestinal symptoms. Am J Gastroenterol 2000;95:

2838–2847.

8. Lin HC, Prather C, Fisher RS, et al. Measurement of gastrointes-tinal transit. Dig Dis Sci 2005;50:989–1004.

9. Maurer AH, Parkman HP. Update on gastrointestinal scintigraphy.Semin Nucl Med 2006;36:110–118.

0. Camilleri M. Scintigraphic biomarkers for colonic dysmotility. ClinPharmacol Ther 2010;87:748–753.

1. Odunsi ST, Camilleri M. Selected interventions in nuclear medi-cine: gastrointestinal motor functions. Semin Nucl Med 2009;39:186–194.

2. Wind J, Hofland J, Preckel B, et al. Perioperative strategy incolonic surgery; LAparoscopy and/or FAst track multimodalmanagement versus standard care (LAFA trial). BMC Surg2006;6:16.

3. The FO, Buist MR, Lei A, et al. The role of mast cell stabilizationin treatment of postoperative ileus: a pilot study. Am J Gastroen-terol 2009;104:2257–2266.

4. Kamm MA. The small intestine and colon: scintigraphic quantita-tion of motility in health and disease. Eur J Nucl Med 1992;19:902–912.

5. Benson MJ, Wingate DL. Ileus and mechanical obstruction. In:Kumar D, Wingate D, eds. An illustrated guide to gastrointes-tinal motility. London, England: Churchill Livingstone, 1993:547–566.

6. Vlug MS, Wind J, Hollmann MW, et al. Laparoscopy in combination withfast track multimodal management is the best perioperative strategy inpatients undergoing colonic surgery: a randomized clinical trial (LAFA-study). Ann Surg 2011 May 18 [Epub ahead of print].

7. Gelpi JR, Dorsey-Tyler K, Luchtefeld MA, et al. Prospective com-parison of gastric emptying after laparoscopic-aided colectomyversus open colectomy. Am Surg 1996;62:594–596.

8. Bauer AJ, Boeckxstaens GE. Mechanisms of postoperative ileus.Neurogastroenterol Motil 2004;16(Suppl 2):54–60.

9. Boeckxstaens GE, de Jonge WJ. Neuroimmune mechanisms inpostoperative ileus. Gut 2009;58:1300–1311.

0. Kalff JC, Turler A, Schwarz NT, et al. Intra-abdominal activation ofa local inflammatory response within the human muscularis ex-terna during laparotomy. Ann Surg 2003;237:301–315.

1. The FO, Bennink RJ, Ankum WM, et al. Intestinal handling-inducedmast cell activation and inflammation in human postoperativeileus. Gut 2008;57:33–40.

2. Lubbers T, Buurman W, Luyer M. Controlling postoperativeileus by vagal activation. World J Gastroenterol 2010;16:1683–1687.

3. Basse L, Madsen JL, Kehlet H. Normal gastrointestinal transitafter colonic resection using epidural analgesia, enforced oralnutrition and laxative. Br J Surg 2001;88:1498–1500.

4. Dindo D, Demartines N, Clavien PA. Classification of surgicalcomplications: a new proposal with evaluation in a cohort of6336 patients and results of a survey. Ann Surg 2004;240:205–213.

5. Foley KM. The treatment of cancer pain. N Engl J Med 1985;313:84–95.

Received December 1, 2010. Accepted May 12, 2011.

Reprint requestsAddress requests for reprints to: Guy Boeckxstaens, MD, PhD,

Department of Gastroenterology, University Hospital Leuven, CatholicUniversity of Leuven, Herestraat 49, 3000 Leuven, Belgium. e-mail:guy.boeckxstaens@med.kuleuven.be; fax: (32) 16-330238.

AcknowledgmentsThe authors thank F. J. Slors†, M. van Berge Henegouwen, D. J.

Gouma, the GI surgery fellows, and the nursing staff for their supportand enthusiasm in realizing this study in the department of surgery;the department of anesthesiology for guarding a standardizedperioperative care and postoperative analgesia protocol; B. Braak, J.Wind, A. Lei, O. Ayubi, and S. El Temna for assistance; N. van

Geloven for advice during the statistical analysis; and J. W. de Jong,

CLIN

ICA

LA

T

880 VAN BREE ET AL GASTROENTEROLOGY Vol. 141, No. 3

M. Spaeth, and the nuclear medicine technicians and staff for thepreparation of the scintigraphically labeled test meals.

Trial registration identifier NTR1884 (http://www.trialregister.nl).

Conflicts of interest

The authors disclose no conflicts.

FundingG. Boeckxstaens, S. van Bree (VICI), and W. de Jonge (VIDI) were

supported by governmental grants from the Netherlands Organizationfor Scientific Research (NWO). G. Boeckxstaens was supported by agovernmental grant (Odysseus program, G.0905.07) of the Flemish

“Fonds Wetenschappelijk Onderzoek” (FWO).

P

D

D

C

September 2011 GI TRANSIT AFTER LAPAROSCOPY AND FAST TRACK 880.e1

Supplementary Table 1. Treatment Protocols

Standard care

reoperative phaseOutpatient department of

surgery- Scheduling of operation- Informed consent

Outpatient department ofanesthesiology

- Preassessment for risk adjustment- Open discussion about different possibimanagement of perioperative analgesia

Preadmission counseling andguided tour on surgical ward

- No

Day of admissionIntake - RoutineBowel preparation - Only enemaPreoperative carbohydrate-

loaded liquids- No

Diet - Last meal until midnightPreanesthetic evening

medication- Lorazepam 1 mg or temazepam 10 or 2

Day of surgeryPreoperative fasting - YesPreanesthetic medication - Yes, lorazepam 1 mg or midazolam 7.5Anesthetic management - Placement thoracic epidural conform FT

level or PCA pump

- Combined with balanced general anesth- Standard preoperative fluid infusion regilactate 20 mL � kg�1 in the first hour fo12 mL � kg�1 � h�1)

- Use of extra fluid challenge as first choimanagement of mean blood pressure drbaseline

- Forced body heating (Bair hugger systemIV fluids)

- Removal of nasogastric tube before extu- Use of ondansetron, dexamethasone, oPONV management according to attendianesthesiologist

Surgical management - Median laparotomy approach/laparosco- Urine catheter according to attending su- No infiltration of surgical wounds with lodrugs

- No standard use of abdominal drainsEarly postoperative management - Epidural or PCA morphine to which para

g � day�1 and/or diclofenac 3 � 50 mgadded

- Small amount of water orally

- IV infusion of Ringer’s lactate 2.5 L � d- No mobilization scheme

ay 1 after surgeryPostoperative management - Diet increased on daily basis

- IV fluid administration (2.5 L � day�1) isadequate oral fluid intake

- Mobilization according to attending surg

Day 2 after surgeryPostoperative management - Epidural removed according to attending

anesthesiologist- Continue as on day 1 until discharge cri

ay 3 after surgery - Continue as on day 2 until discharge cri

FT care

- Scheduling of operation- Informed consent plus information about the FT program- Preassessment for risk adjustment

lities for - Discussion focusing on placement of thoracic epidural catheterfor management of perioperative analgesia plus discussion ofthe essence of the FT program

- Yes

- Additional FT information- Only enema- 4 units (pre-Op)

- Last meal 6 h before operation0 mg - Lorazepam 1 mg the evening before operation, if necessary

- No, 2 units CHL 2 h before surgerymg - Nogroup, or lower - Placement thoracic epidural (T6-T10 depending on the surgical

resection); test dose (bupivacaine 0.25% with adrenaline 1:200,000), top-up dose (bupivacaine 0.25% [� 10 mL] withsufentanil 25 �g, followed by continuous infusion (bupivacaine0.125% with fentanyl 2.5 �g � mL�1) until day 2postoperative

esia - Combined with balanced general anesthesiamen (Ringer’sllowed by 10–

- Restricted per-operative fluid infusion regimen (Ringer’s lactate20 mL � kg�1 in the first hour followed by 6 mL � kg�1 � h�1)

ce forop �20% below

- Use of vasopressor drugs as first choice for management ofmean blood pressure drop �20% of baseline

and warmed - Forced body heating (Bair hugger system and warmed IV fluids)

bation - Removal of nasogastric tube before extubationr droperidol forng

- Prophylactic use of ondansetron (4 mg) to prevent PONV

py - Minimal invasive incisions/laparoscopyrgeon - Suprapubic urine cathetercal anesthetic - Infiltration of surgical wounds with bupivacaine

- No standard use of abdominal drainscetamol 4 � 1� day�1 are

- Use of epidural catheter as mentioned before to whichparacetamol 4 � 1 g.d�1 is added

- First oral drinks at 2 h postsurgery, supplemented with CHLliquids, 2 units (Nutridrink)

ay�1 - IV infusion of Ringer’s lactate 1.5 L � d�1

- Mobilization in the evening (�2 h out of bed)- First semi-solid food intake in the evening

- Oral intake �2 L (including 4 units CHL liquids)continued until - Normal diet

eon - Stop IV fluid administration (leave canula)- Start laxative (MgO, 2 � 1 g � d�1)- Close suprapubic urine catheter and remove when residue�50 mL

- Expand mobilization (�6 h out of bed)

- Remove epidural; add diclofenac 3 � 50 mg � day�1

teria fulfilled - Remove IV canula- Continue Paracetamol 4 � 1000 mg and laxative- Normal diet- Expand mobilization (�8 hours)- Plan discharge

teria are fulfilled - Continue as on day 2 until discharge criteria are fulfilled

HL, carbohydrate loaded; IV, intravenous; MgO, magnesium oxide; PCA, patient-controlled anesthetics; PONV, postoperative nausea and vomiting.

a

880.e2 VAN BREE ET AL GASTROENTEROLOGY Vol. 141, No. 3

Supplementary Table 2. Postoperative Complications and Readmissions Within 30 Days After Surgery

End pointLap/FT

(n � 18)Lap/Standard

(n � 17)Open/FT(n � 18)

Open/Standard(n � 18) P value

Rehospitalization 1 2 1 2 .850Patients with a complication (%)a 2 (11) 5 (29) 7 (39) 7 (39) .217

Grade I 1 3 5 3Grade II 0 1 2 2Grade III 1 0 0 0Grade IV 0 1 0 2Grade V 0 0 0 0Surgical complicationsb 1 4 3 5General complicationsb 2 7 7 3

NOTE. Total complications occurring until 30 days after surgery.aThe severity of complications was graded I to V,34 counting the severest complication per patient. Fisher exact test was used for statisticalnalysis to compare the different treatment arms. Per-protocol analysis of 71 patients.

bMore than 1 complication was counted per patient.

Supplementary Table 3. Protocol Compliance

Lap/FT(n � 18)

Lap/Standard(n � 17)

P value for Lap/FTvs Lap/Standard

Open/FT(n � 18)

Open/Standard(n � 18)

P value for Open/FTvs Open/Standard

Preoperative phase, yes (%)Preoperative counselinga 100 0 �.001b 100 0 �.001b

Omission of bowel preparation 100 94 .486c 94 100 .310c

Intake of CHL day before surgery (L), median(IQR)

0.8 (0.4–0.8) 0.0 (0–0) �.001d 0.8 (0–0.8) 0.0 (0–0) �.001d

Day of surgery, yes (%)Intake of CHL 2 hours before surgery (L),

median (IQR)0.3 (0–0.4) 0.0 (0–0) �.001d 0.4 (0–0.4) 0.0 (0–0) �.001d

No preoperative fasting since midnight 78 13 �.001b 67 31 .084c

Omission of premedication 50 19 .080c 56 12 .012b

Effective thoracic epidural analgesia 89 6 �.001b 83 28 �.002b

First 24-hour intravenous intake (mL), median(IQR)

1600 (1500–1900) 1800 (1150–2260) .901d 1650 (1283–2050) 2000 (1525–2375) .452d

Suprapubic catheter or no catheter 78 71 .627c 67 44 .180c

Intake of CHL after surgery (L), median (IQR) 0.0 (0–0) 0.0 (0–0) .010d 0.0 (0–0) 0.0 (0–0) .008d

Total oral intake after surgery (L), median (IQR) 0.5 (0.1–0.8) 0.1 (0–0.5) �.047d 0.3 (0.1–0.5) 0.0 (0–0.1) �.001d

Mobilization after surgery (L), median (IQR) 0 (0–23) 0 (0–0) .082d 0 (0–10) 0 (0–0) .004d

Start laxative POD 1, no. (%) 72 0 �.001b 50 0 .001b

Intake of CHL (L), median (IQR)POD 1 0.0 (0–0.2) 0.0 (0–0) .005d 0.2 (0–0.6) 0.0 (0–0) �.001d

POD 2 0.0 (0–0.2) 0.0 (0–0) .002d 0.0 (0–0.4) 0.0 (0–0) .002d

POD 3 0.0 (0–0.2) 0.0 (0–0) .087d 0.0 (0–0.3) 0.0 (0–0) .004d

Total oral intake (L), median (IQR)POD 1 1.4 (0.4–1.9) 0.9 (0.3–1.2) .170d 1.1 (0.7–1.8) 0.6 (0.4–0.9) .051d

POD 2 2.0 (1.0–2.1) 1.3 (0.8–1.8) .044d 1.4 (0.7–2.0) 1.0 (0.6–1.6) .426d

POD 3 2.0 (1.5–2.0) 2.0 (1.0–2.0) .603d 1.9 (1.0–2.0) 1.5 (1.0–1.9) .134d

Mobilization (min), median (IQR)POD 1 95 (30–133) 45 (10–75) .035d 105 (60–130) 30 (13–60) �.001d

POD 2 270 (120–364) 60 (60–150) .008d 120 (98–180) 60 (38–110) .005d

POD 3 400 (300–410) 180 (90–270) �.001d 210 (120–300) 120 (80–245) .056d

CHL, carbohydrate-loaded drink.aSeparate consultation before admission with an FT trial nurse to discuss the essence of the FT program.Analyzed by bFisher exact test, c�2 test, and dMann–Whitney U test.

Septem

ber2

01

1G

ITR

AN

SIT

AFTER

LAP

AR

OS

CO

PY

AN

DFA

ST

TRA

CK

88

0.e3

w

880.e4 VAN BREE ET AL GASTROENTEROLOGY Vol. 141, No. 3

Supplementary Table 4. Consumption of Analgesics and Antiemetics

Lap/FT(n � 18)

Lap/Standard(n � 17)

Open/FT(n � 18)

Open/Standard(n � 18) P

Cumulative antiemetic consumption onPOD 0 ondansetron (mg ondansetron)

2.6 � 3.2 2.5 � 3.0 3.1 � 3.8 2.4 � 2.4 .982

Cumulative morphine sulfate consumptionduring intraoperative surgery (mg)

9.8 � 11.2 12.1 � 9.6 9.9 � 9.1 19.9 � 24.2 .174

Cumulative morphine equivalentconsumption until POD 6

108.1 � 32.4 93.7 � 54.0 135.2 � 67.9 119.4 � 46.9 .165

NOTE. Values are expressed as mean � SD unless otherwise noted. The antiemetic doses include the ondansetron equivalent doses ofdroperidol (0.625 mg of droperidol equals 4.0 mg of ondansetron). The morphine doses include the parenteral morphine equivalent doses offentanyl, sufentanil, and tramadol (equianalgesic doses according to Foley35). Per-protocol analysis of 71 patients. The analysis of variance test

as used for statistical analysis.