ittpsho

aobitdr

tecc(wqmtct

Laaivct“b

(F

Journal of Surgical Research 81, 15–20 (1999)Article ID jsre.1998.5493, available online at http://www.idealibrary.com on

Patterns of Morbidity and Resource Consumption Associated withLaparoscopic Cholecystectomy in a VA Medical Center

Mark Molloy, M.D.,*,1 Matthew J. Sorrell, M.D.,† Robert H. Bower, M.D.,*,†Per-Olof Hasselgren, M.D.,*,† and Barbara J. Dalton, R.N.*

*Veterans Affairs Medical Center, Cincinnati, Ohio 45220; and †Department of Surgery,University of Cincinnati College of Medicine, Cincinnati, Ohio 45220

Presented at the 22nd Annual Symposium of the Association of Veterans Administration Surgeons,Baltimore, Maryland, April 26–28, 1998

c

caitdfstlptei

e[aDmS[oc

ptCmoti

Background. The generally low incidence of morbid-ty and reduced rate of health care resource consump-ion commonly associated with laparoscopic cholecys-ectomy (LC) have been established from studies ofatient populations which are distinct from thaterved by the Department of Veterans Affairs (VA)ealth care system. We sought to assess the outcomesf this procedure when performed on VA beneficiaries.Materials and methods. Demographic and perioper-

tive data for all patients undergoing attempted LC inur facility were recorded in a prospective databaseeginning 1 January 1993. The information in this reg-stry was analyzed to determine the demographics ofhe treated population, the spectrum of biliary tractisease encountered, and patterns of morbidity andesource consumption.Results. LC was attempted in 141 cases. Median pa-

ient age was 62 years. The indication for surgery wasither acute cholecystitis or biliary pancreatitis in 63ases (45%). Thirteen patients (9%) developed majoromplications. These patients were significantly oldermean age 68 vs 59 years) than patients whose courseas uncomplicated. Twenty-seven cases (19%) re-uired conversion to an open procedure, most com-only for acute cholecystitis. Progressive cholecysti-

is was associated with a conversion rate of 64%. Bothonversion and the development of a major complica-ion produced significant increases in length of stay.

Conclusions. The population undergoing attemptedC in the VA system is characterized by relativelydvanced age and high incidences of comorbid illnessnd complicated biliary tract disease. These attributesncrease the frequency of major morbidity and of con-ersion to open cholecystectomy, which in turn in-rease resource consumption. Comparisons betweenhe outcomes of attempted LC in VA centers andbenchmark” results obtained in other settings shoulde controlled for these factors. © 1999 Academic Press

1 To whom correspondence should be addressed at Surgical Service112), VA Medical Center, 3200 Vine Street, Cincinnati, OH 45220.

tax: (513)-475-6470.

15

Key Words: cholecystectomy; cholecystitis; laparos-opy; complications; elderly.

INTRODUCTION

Laparoscopic cholecystectomy (LC) replaced openholecystectomy as the therapy of choice for symptom-tic gallstones in this country within a few years of itsntroduction to American surgeons in 1989 [1–4]. Theransition to this procedure was accelerated by patientemand for a less painful form of therapy and pressurerom third-party payers to reduce the length of hospitaltay and duration of postoperative disability in pa-ients undergoing this common operation. Severalarge series [5–9] have demonstrated that LC can beerformed with very low morbidity and mortality inhe general American population, despite an appar-ntly sustained small increase in the rate of bile ductnjuries associated with this technique.

The largest population-based studies which havevaluated the morbidity associated with this operation5, 7–10] were performed in patient populations thatre demographically distinct from that served by theepartment of Veterans Affairs (VA). The estimatededian age of the veteran population in the Unitedtates is 56.7 years; 95% of these individuals are males

11]. We sought to assess the impact of this proceduren VA beneficiaries in terms of morbidity and healthare resource consumption.

METHODS

Demographic, perioperative, and long-term survival data for allatients undergoing attempted LC at our facility have been main-ained in an 83-field, prospective database (Filemaker Pro, Clarisorp., Santa Clara, CA) since the procedure was introduced in ouredical center in 1993. The information for each field is entered by

ne nurse assigned to the surgical service after it is obtained fromhe patient’s record, the hospital information system, or patientnterviews. Longitudinal follow-up is maintained by reviewing re-

urn patient activity to the medical center, patient deaths reported to

0022-4804/99 $30.00Copyright © 1999 by Academic Press

All rights of reproduction in any form reserved.

ts

cipBwckdswllhdddi

smuro

cda0m

O

awTi3tnsli8

ohmuqdtcwpa

nvcdnmdsrf

cw3e

p

PPS

11111

T

1C

16 JOURNAL OF SURGICAL RESEARCH: VOL. 81, NO. 1, JANUARY 1999

he VA, and phone calls to patients. All cases were performed byurgical residents directly supervised by attending surgeons.For purposes of the study, patients were given a diagnosis of acute

holecystitis if they had an acute presentation and any of the follow-ng findings: temperature .100°F, WBC $10,000/mm3, or right up-er quadrant guarding (Murphy’s sign) on physical examination.iliary pancreatitis was defined as a clinical syndrome consistentith pancreatitis associated with hyperamylasemia and findings of

holelithiasis by ultrasound examination. A diagnosis of biliary dys-inesia required a normal right upper quadrant ultrasound andemonstration of a gallbladder ejection fraction of ,35% by HIDAcan. Comorbid conditions were recorded as present only if a patientas being treated for them at the time of the procedure or if histo-

ogic evidence was available for patients with histories of prior ma-ignancy or cirrhosis. The day of the procedure was counted as a fullospital day. Thus, patients both admitted and discharged on theay of surgery were recorded as having a length of stay (LOS) of 1ay. The length of operating room time required (OR time) wasetermined by subtracting the time the patient entered the operat-ng room from the time he left it using a 24-h clock.

The outcomes of attempted LC were then compared to those ob-erved in patients who underwent nonincidental open cholecystecto-ies at our facility during the study period. Study data for individ-als treated with open cholecystectomy were collected in aetrospective fashion after these patients were identified by a reviewf the medical center operative log.Statistical analysis was performed using StatView (Abacus Con-

epts, Inc., Berkeley, CA). Continuous variables were analyzed usingescriptive statistics and analysis of variance (ANOVA) as appropri-te. Discrete data were analyzed using x2 techniques. A P value of.05 was considered significant throughout. Results are reported asean values 6 the standard deviation unless otherwise indicated.

RESULTS

verall Results of Attempted LC

LC was attempted in 141 patients between 1 Janu-ry 1993 and 1 December 1997. Median patient ageas 62 years, and 129 of the subjects (91%) were male.he number of LCs attempted increased from 13 dur-

ng the first year of the study to an incidence between1 and 39 for each of the last 3 years (Table 1). Withhe exception of the first year, the total number ofonincidental cholecystectomies performed in the in-titution remained stable. The percentage of total cho-ecystectomies that were attempted laparoscopicallyncreased from 62% in 1993 to a level between 81 and

TABLE 1

Incidence of Laparoscopic Cholecystectomy as aercentage of All Nonincidental Cholecystectomieserformed in the Institution during Each Year of thetudy

Year Total cholecystectomies Open (%) Laparoscopic (%)

993 21 8 (38) 13 (62)994 37 13 (35) 24 (65)995 45 6 (13) 39 (87)996 42 8 (19) 34 (81)997a 38 7 (18) 31 (82)

otal 183 42 (23) 141 (77)

Note. Cases converted to open are counted as “laparoscopic.”a Includes only first 11 months of 1997.

7% for each of the last 3 years. t

At least one comorbid condition was present in 94%f the population. The most common of these wereypertension, coronary artery disease, and diabetesellitus (Table 2). Twenty-seven patients (19%) hadndergone previous abdominal surgery, most fre-uently appendectomy (n 5 13), or gynecologic proce-ures (n 5 5). The indication for surgery was symp-omatic cholelithiasis in 63 cases (45%), but moreomplex forms of biliary tract disease were as commonhen taken in aggregate (Table 3). Forty-six of theatients in the study (33%) were being treated forcute cholecystitis.All patients were discharged from the hospital alive;

ine (6%) have died during follow-up. The median sur-ival for those patients who died was 15.9 months. Theauses of death were cardiac disease (n 5 3), intraab-ominal carcinoma (n 5 2), and bronchogenic carci-oma, pulmonary fibrosis, cystic fibrosis, and undeter-ined (n 5 1 each). A single death (0.7%) was

etermined to be related to the procedure. This patientuffered a perioperative myocardial infarction and waseadmitted 4 weeks after discharge with a second in-arction that proved fatal.

Thirteen patients (9%) experienced a major compli-ation following surgery of which the most commonere bile leaks (n 5 4) and myocardial infarctions (n 5). The frequency and spectrum of major complicationsncountered are shown in Table 4.The mean age of patients experiencing a major com-

lication was 68 years vs 59 years for patients whose

TABLE 3

Distribution of Indications for Surgeryin the Study Population

Indication n (%)

Symptomatic cholelithiasis 63 (45)Acute cholecystitis 46 (33)Biliary pancreatitis 17 (12)Biliary dyskinesia 7 (5)Prior choledocholithiasis 6 (4)Othera 2 (1)

a Other indications were (one each) gallbladder polyps and asymp-

TABLE 2

Spectrum and Incidence of Comorbid Conditions in41 Patients Undergoing Attempted Laparoscopicholecystectomy

Condition n (%)

Hypertension 45 (32)Coronary artery disease 42 (30)Diabetes mellitus 31 (22)Peptic ulcer disease 17 (12)Chronic obstructive pulmonary disease 14 (10)Cirrhosis 11 (8)Others 39 (28)None 9 (6)

omatic cholelithiasis associated with diabetes.

ctctoptast

ati

crtcswtcc(r

qtsDmd

aFrtr2s

tptof4mw

R

fputawipt

BM

P

P

R

E

PP

T

Ut

L

C

M

A

mdicp

17MOLLOY ET AL.: LAPAROSCOPIC CHOLECYSTECTOMY IN A VA MEDICAL CENTER

ourse was uncomplicated (P 5 0.02, ANOVA). Al-hough there were no bile duct injuries, biliary compli-ations occurred in six patients (4%). These includedwo bile leaks from the cystic duct stump and twothers from the gallbladder fossa. One patient who hadreviously been treated with a percutaneous cholecys-ostomy for severe cholecystitis developed cholangitisfter undergoing an interval LC, and one patient pre-ented a week after discharge with a retained stonehat required endoscopic extraction.

Minor complications manifested themselves in andditional 15 patients (11%). The most frequent ofhese were urinary retention (n 5 6, 4%) and woundnfections (n 5 4, 3%) (Table 4).

Conversion to an open procedure was required in 27ases (19%). In 17 of these cases (63%), conversion wasequired because of inflammation or anatomic distor-ion related to acute cholecystitis. The relative frequen-ies of the indications for conversion for all cases arehown in Table 5. Converted cases were associatedith a lower incidence of major morbidity than cases

hat were completed laparoscopically. A major compli-ation occurred in 12 of 114 cases (11%) that wereompleted laparoscopically and in 1 of the 27 cases4%) that required conversion. This difference did noteach statistical significance (P 5 0.27, x2 analysis).

TABLE 5

Indications for Conversion

Indication for conversion n (%)

Inflammation 10 (7)Unclear anatomy 7 (5)Choledocholithiasis 4 (3)Hemorrhage 2 (1)Cirrhosis 1Other 3 (2)

Total conversions 27 (19)

TABLE 4

Incidence and Types of Morbidity Encountered in141 Attempted Laparoscopic Cholecystectomies

Major complications n (%) Minor complications n (%)

ile leak 4 (3) Urinary retention 6 (4)yocardialinfarction

3 (2) Wound infection 4 (3)

ulmonaryembolism

1 Return forabdominal pain

2 (1)

ostoperativecholangitis

1 Mild pulmonaryedema

1

etained stone 1 Exacerbation ofgout

1

nterocutaneousfistula

1 Urinary tractinfection

1

ort site hernia 1rolonged ileus 1

otal majorcomplications

13 (9) Total minorcomplications

15 (11)

LOS data are depicted in Table 6. Patients who re-uired conversion to open cholecystectomy and pa-ients who experienced a major complication had aignificantly longer mean LOS than those who did not.evelopment of a major complication increased theean LOS by 10 days, and conversion increased it by 7

ays (both P , 0.0001, ANOVA).Seventeen patients (12%) underwent some type of

djunct procedure at the time of their attempted LC.or the remaining 124 cases, the mean total operatingoom time (OR time) required was 196 6 62 min. ORime ranged from 80 to 455 min. Converted proceduresequired significantly more time to perform (mean60 6 84 min) than those that were completed laparo-copically (mean 180 6 42 min, P , 0.0001, ANOVA).Intraoperative cholangiography (IOC) was at-

empted in 99 of the 114 cases (87%) which were com-leted laparoscopically and was successful in 83 ofhese (84%). When 15 cases which involved some typef adjunct procedure were excluded, the mean OR timeor cases in which an IOC was either performed2 (185 60 min) or attempted (181 6 34 min) was approxi-ately 30 min longer than for cases in which no IOCas attempted2 (153 6 50 min).

esults of Attempted LC in Acute Cholecystitis

Forty-six patients (33%) underwent attempted LCor the treatment of acute cholecystitis. Two dominantatterns of illness were observed among these individ-als. Twenty-four patients responded favorably to ini-ial medical treatment with bowel rest and intravenousntibiotics. These patients underwent attempted LCithin 72 h of presentation if they were identified early

n the course of their illness. More commonly, theyresented with a history of several days of pain andherefore underwent a staged attempted LC 3 to 6

2

TABLE 6

Length of Stay (LOS) Data for 141 Patientsndergoing Attempted Laparoscopic Cholecystec-

omy

Case type (n) Pre-op LOS Post-op LOS Total LOS

aparoscopic (102) 2.0 6 3.5(1.0)

2.2 6 2.3(1.5)

4.3 6 4.7(3.0)

onverted (26) 5.0* 6 7.3(3.0)

6.0 6 2.7(5.0)

11.0** 6 8.2(8.5)

ajor morbidity(13)

2.7 6 4.0(1.0)

11.3 6 1.04(10.0)

14.0** 6 10.5(13.0)

ll cases (141) 2.6 6 4.6 3.8 6 4.7 6.4 6 7.1(1.0) (2.0) (4.0)

Note. Results are reported as mean values 6 standard deviation;edian values are shown in parentheses. Both conversion and the

evelopment of a major complication produced significant increasesn mean total LOS. The longer pre-op LOS observed for convertedases reflects the high incidence of progressive acute cholecystitis foratients in this category.

*P 5 0.004, ANOVA.**P # 0.0001, ANOVA.

P 5 0.008, ANOVA.

wr

sdidst0f

doaattrtdi

R

dtybt

mvuc1dpb

lutprsppaws

ttcoogg

tt0tdf0td

cwsa(

cfttosaaalampsvt

d(ptrarmfrtth

toTpfst

18 JOURNAL OF SURGICAL RESEARCH: VOL. 81, NO. 1, JANUARY 1999

eeks after their initial presentation. Conversion wasequired in 3 of these 24 cases (13%).

The other 22 patients with acute cholecystitis pre-ented with a history of several days of symptoms andid not improve with a trial of medical therapy. Thesendividuals underwent attempted LC during their in-ex hospitalization because of persistent or progressiveymptoms. Conversion was required in 14 (64%) ofhese patients with “progressive” cholecystitis (P 5.0003, x2 analysis). This clinical scenario accountedor 52% of the total conversions that were performed.

Five of the 46 patients treated for acute cholecystitiseveloped major complications (11%). These consistedf two myocardial infarctions, an episode of postoper-tive cholangitis, a bile leak from the gallbladder bed,nd an enterocutaneous fistula from a port site. All ofhese adverse events occurred in the group of patientsreated after their acute episode of cholecystitis hadesolved. No major complications were encountered inhe 22 patients with progressive cholecystitis who un-erwent attempted LC during the acute phase of theirllness (P 5 0.02, x2 analysis).

esults of Open Cholecystectomy

Forty-two patients underwent a primary, noninci-ental open cholecystectomy in the institution duringhe study period. The incidence of these procedures byear is shown in Table 1. Complete records were able toe retrieved for 35 of these patients (83%), who formhe basis for the analysis that follows.

Median patient age was 69 years; 33 subjects wereale (94%). These population characteristics did not

ary significantly from those observed in the groupndergoing attempted LC. The frequencies of the indi-ations for surgery were as follows: acute cholecystitis,5 (43%); symptomatic cholelithiasis, 10 (29%); chole-ocholithiasis, 4 (11%); biliary pancreatitis, 3 (9%);rior choledocholithiasis, 2 (6%); and one case of gall-ladder polyps.The most common reasons given for not using a

aparoscopic approach in these patients were previouspper abdominal surgery [12 (34%)], acute cholecysti-is [8 (23%)], choledocholithiasis [4 (11%)], and staffreference [4 (11%)]. The contraindications to a lapa-oscopic approach in the remaining cases were cirrho-is (n 5 2), the need for an unrelated intraabdominalrocedure (n 5 2), and possible gallbladder carcinoma,resence of a thrombosed hepatic artery aneurysm,nd a porcelain gallbladder (1 each). A cholangiogramas attempted in 23 of the 35 cases (66%) and was

uccessfully obtained in all attempts.Four major complications occurred in the group

reated by open cholecystectomy. These consisted ofwo episodes of prolonged respiratory failure, one myo-ardial infarction, and one wound dehiscence. Theverall incidence of major morbidity associated withpen cholecystectomy was 11%, not significantlyreater than the incidence observed in patients under-

2

oing attempted LC (9%, P 5 0.52, x analysis). p

The mean total LOS associated with open cholecys-ectomy (11.7 6 6.4 days) was nearly twice as long ashat associated with attempted LC (6.4 6 7.1 days, P ,.0001, ANOVA). The mean postoperative LOS for pa-ients undergoing open cholecystectomy was 7.2 6 4.3ays. This was significantly longer than that observedor patients who had successful LC (3.2 6 4.9 days, P ,.0001, ANOVA), but not significantly different thanhat for patients who required conversion (6.2 6 2.8ays, P 5 0.53, ANOVA).The mean operating room time required for an open

holecystectomy with intraoperative cholangiographyas 179 6 45 min. The mean OR time for a laparo-

copically completed procedure with a successful chol-ngiogram was essentially the same: 185 6 40 minP 5 0.65, ANOVA).

Eleven of the 15 patients (73%) undergoing openholecystectomy for acute cholecystitis met the criteriaor progressive cholecystitis described above (symp-oms for .72 h at presentation and failure to respondo medical treatment). Three of these patients devel-ped major complications (27%). This incidence wasignificantly higher than the rate of zero observedmong the 22 patients with this condition who had anttempt at laparoscopic management (P 5 0.04, x2

nalysis). This difference is likely to be related to se-ection bias. The mean WBC of patients undergoingttempted LC for this condition was 9.9 6 3.4 3 103/m3 vs 13.7 6 4.2 3 103/mm3 for patients undergoing

rimary open cholecystectomy under the same circum-tances (P 5 0.009, ANOVA). Patients with more se-ere cholecystitis appear to have been preferentiallyreated using an open procedure.

DISCUSSION

The dominant clinical finding of this study is theemonstration of the extremely high conversion rate64%) associated with attempted LC in the setting ofrogressive cholecystitis. This finding does not meanhat a laparoscopic approach to this situation is unwar-anted. Approximately one-third of these patients wereble to be managed laparoscopically, and those thatequired conversion did not incur any increased risk oforbidity. We would advocate a treatment algorithm

or this condition that begins with an attempt at lapa-oscopic management during which an extremely lowhreshold for conversion is maintained. Patients withhis condition should be advised preoperatively of theigh risk of conversion under these circumstances.The overall risk of major morbidity observed for at-

empted LC (9%) was equivalent to that observed forpen cholecystectomy (11%) during the study period.he length of OR time required was similar for bothrocedures when converted patients were excludedrom consideration. LC resulted in a significantlyhorter mean LOS than open cholecystectomy by in-ent to treat analysis. Thus, a liberal laparoscopic ap-

roach to biliary tract disease in this population does

of

b(tlaaabamf[pwsmiWsLm(

pcreb“dctmpnsorts

it

aVbwbhdhpeapTmio5tt

wbVsms

tslL1cfuspm

SBAFWC

C

19MOLLOY ET AL.: LAPAROSCOPIC CHOLECYSTECTOMY IN A VA MEDICAL CENTER

ffer considerable resource savings compared to a uni-orm open approach.

The incidences of major morbidity (9%), minor mor-idity (11%), and conversion to open cholecystectomy19%) as well as mean LOS and OR time are substan-ially higher in this series than reported in previousarge studies of attempted LC performed in generaldult populations [5–10, 12] (Table 7). The populationsnalyzed in these studies consisted of mostly middle-ged females, and some of the differences observedetween this and previous reports may therefore bescribed to the fact that our patients were mostly olderen. Increased patient age has been shown to be a risk

actor for both conversion and morbidity following LC9, 10, 12–14]. This series also has a high incidence ofatients operated on for acute cholecystitis (33%),hich is recognized as a major risk factor for conver-

ion [9, 10, 12–14]. These two factors (conversion andajor morbidity) were the primary determinants of the

ncreased LOS that was seen in the overall group.hen the 39 patients who experienced either conver-

ion or a major complication were excluded from theOS analysis, mean LOS fell from 6.4 to 4.3 days, andean postoperative LOS decreased from 3.8 to 2.2 days

Table 6).The reasons for the mean OR times of 196 min for all

atients and 180 min for nonconverted cases are lesslear. All of the attending surgeons and many of theesidents involved in the study had 1 or 2 years ofxperience with the technique of LC in other facilitiesefore beginning to use it at the VA. This absence of alearning curve” is reflected by the fact that OR timeid not show any trend toward improvement over theourse of the study. We chose to focus on the total ORime required since this is the variable that best esti-ates the consumption of resources. Other studies re-

ort “operative time” or “duration of pneumoperito-eum” instead of total time in the operating room, andome of the difference between these times and thosebserved in our study may be accounted for by the timeequired for perioperative treatment of the patient inhe operating room [5–7, 10, 12]. Still, the cases in this

TAB

Comparison of Rates of Morbidity and Resource CoThose Observed in 141 Consecutive V

Author Year nMorbi

(%

outhern Surgeons[5] 1991 1518 5ailey et al. [6] 1991 375 3.iran et al. [7] 1992 1771 NRried et al. [12] 1994 1677 6herry et al. [9] 1996 9054 6.ollet [10] 1997 4624 4.

urrent series 1998 141 20

a Median value.b Excludes converted patients.

tudy clearly took a long time to perform. This finding t

s probably another reflection of the severity of biliaryract disease encountered.

The National Veterans Affairs Surgical Risk Studynalyzed the introduction of LC into 43 tertiary careA medical centers between October 1991 and Decem-er 1993 [15]. The authors found that attempted LCas performed in 21% of all cholecystectomies at theeginning of the study period and that this percentagead increased to 51% by the end of 1993. They alsoetermined that patients undergoing attempted LCad lower American Society of Anesthesiologists (ASA)reoperative risk classes, were less likely to requireither an emergent or technically complex operation,nd had a lower incidence of acute cholecystitis thanatients undergoing primary open cholecystectomy.hey inferred from these findings that LC was adoptedore slowly and applied more selectively during its

ntroduction into the VA than into other settings. Theverall morbidity rate of 7.6% and conversion rate of.9% reported by these authors are much lower thanhose found in the current study (20 and 19%, respec-ively).

Once LC began being performed in our hospital, itas applied to a wider spectrum of both patients andiliary tract problems than described by the Nationaleterans Affairs Surgical Risk Study report. This ob-ervation probably accounts for the higher rates oforbidity and conversion that were encountered in our

eries.Another examination of the use of attempted LC in

he setting of a VA medical center identified VA patienttatus as an independent risk factor for conversion byogistic regression analysis [13]. In that study, 13 of 72Cs (18%) attempted in a VA facility between July990 and August 1993 were converted to open chole-ystectomy, compared to 32 of 492 cases (6.5%) per-ormed in other adult hospitals served by the sameniversity surgical group. The conversion rate de-cribed by these authors is also higher than that re-orted by the National VA Surgical Risk Study andirrors our experience.The findings of our study indicate that the popula-

7

umption between Previously Published Series andPatients Undergoing Attempted LC

Conversion(%)

Mean LOS(days)

Mean OR time(min)

4.7 1.2 905 1.3 1024.6 NR 1055.4 1* 659.8 NR NR6.9 4.5** 59**

19 6.4 196

LE

nsA

dity)

5

19

ion currently undergoing attempted LC in the VA

shtwtmetsssatt

gnbt

1

1

1

1

1

1

20 JOURNAL OF SURGICAL RESEARCH: VOL. 81, NO. 1, JANUARY 1999

ystem is characterized by relatively advanced age andigh incidences of comorbidity and complicated biliaryract disease. If the experiences of other VA centersith LC parallel our own, the procedure is being at-

empted nationally in veteran patients with more co-orbid conditions and with more advanced biliary dis-

ase than previously reported. These factors contributeo increased incidences of major morbidity and conver-ion to open cholecystectomy, which in turn produce aignificant increase in length of stay. Such issueshould be considered when comparing the outcomes ofttempted laparoscopic cholecystectomy in VA facili-ies with “benchmark” results obtained in other set-ings.

Progressive cholecystitis in this population is notenerally able to be treated using laparoscopic tech-iques, and a very low threshold for conversion shoulde maintained when performing an attempted LC forhis condition.

REFERENCES

1. Reddick, E. J., and Olsen, D. O. Laparoscopic laser cholecystec-tomy, a comparison with mini-lap cholecystectomy. Surg. En-dosc. 3: 131, 1989.

2. Legoretta, A. P., Silber, J. H., Constantino, G. N., Kobylinski,R. W., and Zatz, S. L. Increased cholecystectomy rate after theintroduction of laparoscopic cholecystectomy. JAMA 270: 1429,1993.

3. Steiner, C. A., Bass, E. B., Talamini, M. A., Pitt, H. A., andSteinberg, E. P. Surgical rates and operative mortality for openand laparoscopic cholecystectomy in Maryland. New Engl.J. Med. 330: 403, 1994.

4. National Institutes of Health Consensus Development Confer-ence Statement on Gallstones and Laparoscopic Cholecystec-tomy. JAMA 269: 1018, 1993.

5. The Southern Surgeons Club. A prospective analysis of 1518laparoscopic cholecystectomies. New Engl. J. Med. 324: 1073,

1991.

6. Bailey, R. W., Zucker, K. A., Flowers, J. L., Scovill, W. A.,Graham, S. M., and Imbembo, A. L. Laparoscopiccholecystectomy—Experience with 375 consecutive patients.Ann. Surg. 214: 531, 1991.

7. Airan, M., Appel, M., Berci, G., Coburg, A. J., Cohen, M., Cus-chieri, A., Dent, T., Duppler, D., Easter, D., Greene, F., Halevey,A., Hammer, S., Hunter, J., Jenson, M., Ko, S. T., McFaydan,B., Perissat, J., Ponsky, J., Ravindranathan, P., Sackier, J. M.,Soper, N., Van Steigmann, G., Traverso, W., Udwadia, T., Un-ger, S., Wahlstrom, E., and Wolfe, B. Retrospective and prospec-tive multi-institutional laparoscopic cholecystectomy study or-ganized by the Society of American Gastrointestinal EndoscopicSurgeons. Surg. Endosc. 6: 169, 1992.

8. Deizel, D. J., Millikan, K. W., Economou, S. G., Doolas, A., Ko,S. T., and Airan, M. C. Complications of laparoscopic cholecys-tectomy: A national survey of 4,292 hospitals and an analysis of77,604 cases. Am. J. Surg. 165: 9, 1993.

9. Wherry, D. C., Marohn, M. R., Malanoski, M. P., Hetz, S. P., andRich, N. M. An external audit of laparoscopic cholecystectomyin the steady state performed in medical treatment facilities ofthe Department of Defense. Ann. Surg. 224: 145, 1996.

0. Collet, D. Laparoscopic cholecystectomy in 1994—Results of aprospective survey conducted by SFCERO on 4,624 cases. Surg.Endosc. 11: 56, 1997.

1. Department of Veterans Affairs home page: www.va.gov/data.htm.

2. Fried, G. M., Clas, D., and Meakins, J. L. Minimally invasivesurgery in the elderly patient. Surg. Clin. N. Am. 74: 375,1994.

3. Wiebke, E. A., Pruitt, A. L., Howard, T. J., Jacobson, L. E.,Broadie, T. A., Goulet, R. J., and Canal, D. F. Conversion oflaparoscopic to open cholecystectomy—An analysis of risk fac-tors. Surg. Endosc. 10: 742, 1996.

4. Flowers, J. L., Bailey, R. W., Scovill, W. A., and Zucker, K. A.The Baltimore experience with laparoscopic management ofacute cholecystitis. Am. J. Surg. 161: 388, 1991.

5. Chen, A. Y., Daley, J., Pappas, T. N., Henderson, W. G., andKhuri, S. F. Growing use of laparoscopic cholecystectomy in theNational Veterans Affairs Surgical Risk Study—Effect on vol-ume, patient selection, and selected outcomes. Ann. Surg. 227:

12, 1998.