Accepte

Reprintgery, Acsterdam

0039-60

� 2014

http://d

384 S

Insufficient safety measures reportedin operation notes of complicatedlaparoscopic cholecystectomiesKlaske A. C. Booij, MD, Philip R. de Reuver, MD, Bram Nijsse, MD, Olivier R. C. Busch, MD,Thomas M. van Gulik, MD, and Dirk J. Gouma, MD, Amsterdam, The Netherlands

Background. In 2007, a Dutch guideline for laparoscopic cholecystectomy (LC) was composed thatadvocates the critical view of safety (CVS) to prevent bile duct injury (BDI). Conversion to opencholecystectomy is recommended in complicated cholecystectomy, but young surgeons are hardly trained inthis procedure. The aim of this study was to analyze the accuracy of dictated operation notes, the use ofCVS before and after guideline implementation, and the severity of injury after conversion.Methods. Between 1990 and 2012, 800 patients were referred for treatment of BDI. All availableoperation notes (n = 528; 66%) were scored for procedural conditions, reasons for conversion, the use ofsafety measures, and postoperative care in BDI patients.Results. Patient demographics, indication for cholecystectomy, conversion rates, and injury type werecomparable to the total cohort of BDI patients. LC (n = 479; 91%) was converted in 180 patients(34%). The CVS technique or dissection of Calot’s triangle were reported in 33 patients (6.3%) and 87patients (16.5%), respectively. Guideline implementation increased the reporting of CVS from 4% (16/425) to 17% (17/103; P < .001), and the consultation of a hepatic-pancreatic-biliary (HPB) colleaguefrom 3% (14/425) to 8% (8/103; P < .01). Conversion to open surgery leads to more complex injury(Bismuth III–V injury rate of 34% [24/64] vs 65% [46/116]; P = .013).Conclusion. The insufficient use of safety measures to prevent BDI during LC in this selected patientgroup is of concern. Although guideline implementation significantly improved the use of CVS duringLC, further improvement is necessary. Conversion cannot simply be used as an ‘‘escape’’ procedure,because this may lead to more complex injury. (Surgery 2014;155:384-9.)

From the Department of Surgery, Academic Medical Centre, Amsterdam, The Netherlands

SAFETY MEASURES to prevent bile duct injury (BDI)caused during laparoscopic cholecystectomy (LC)are extensively described and scrutinized.1-6

Several groups advocate the use of intraoperativecholangiography (IOC) as large cohort and popu-lation based studies showed this reduces BDI by25–40%.7-10 However, the interpretation of IOCmight be difficult during LC.7-11

Complete dissection of Calot’s triangle andreaching the critical view of safety (CVS) is themost practical way to be sure the common bileduct is safe before transecting the cystic duct andartery.12,13 The Dutch Society of Surgery imple-mented a guideline for LC in 2007, in which

d for publication October 8, 2013.

requests: Philip R. de Reuver, MD, Department of Sur-ademic Medical Center, Meibergdreef 9, 1105 AZ Am-, The Netherlands. E-mail: p.r.dereuver@amc.uva.nl.

60/$ - see front matter

Mosby, Inc. All rights reserved.

x.doi.org/10.1016/j.surg.2013.10.010

URGERY

documentation of CVS is required and conversionto open surgery is advised if CVS is not achieved.However, young residents are scarcely experiencedin this open ‘‘rescue’’ procedure.14,15 Conversionseems, in that sense, to be an arbitrary safety mea-sure in difficult LC patients.

The conventional way to document the opera-tional procedure and measures taken to preventBDI have been the postoperative dictated opera-tion report. Despite an adequate description of theprocedure in this report, several studies showedpoor or limited assessment of CVS compared witha photograph or video made during the sameprocedure.4,16-18 In medical litigation after BDI,experts are therefore generally confronted withthe operation report as the only documenteddescription of the procedure.

A national survey in 2011, 4 years after theimplementation of the guideline, reported the useof CVS by 98% of Dutch surgeons.3 Although CVSis advocated by the Dutch Society of Surgeons andseems to be embraced by their members, the num-ber of BDI patients referred for to a tertiary center

SurgeryVolume 155, Number 3

Booij et al 385

did not decline in the last decade.19 It is of interestto analyze whether the implementation of thenational guideline changed the use of CVS andmeasures taken to prevent BDI.

The purpose of this study was to analyze theaccuracy of dictated operation notes and the use ofCVS during LC in BDI patients before and afterguideline implementation, and to describe theoutcome in terms of injury classification afterconversion.

METHODS

Patients and operative report. Between January1990 and January 2012, 800 consecutive patientsreferred to the Academic Medical Centre in Amster-dam for management of BDI were included in aprospective database. Clinical data obtained fromthe referring centers included indication for chole-cystectomy, type of operation, and type of injury. Thetypeof injurywas classifiedaccording theAmsterdamClassification, Strasberg classification, and Bismuthclassification.20-22 For the present study, all availablecopies of operation reports of the cholecystectomywere analyzed in detail. According to a list of 30 items2 authors (PR and BN) independently scored the re-ports. The presence or absence of conditions werescored (eg, inflammation, adhesions, bleeding, bileleakage, or bad equipment), identification ofanatomic structures (eg, cystic duct, cystic artery,common bile duct, right hepatic duct and artery),safety measurements (CVS, IOC, consultation of acolleague), and perioperative proceedings (detec-tion of BDI, conversion, treatment of bile leakage,etc). Suspicion of BDI was noted if injury was diag-nosed during surgery or the operation report de-scribes doubt about possible injury. The outcomein terms of injury classification was compared be-tween BDI patients in whom BDI occurred beforeconversion to open surgery and patients in whomthe injury occurred after conversion.

Statistical analysis. Data from patient character-istics, operation report characteristics and type ofinjury are represented in numbers and percent-ages. Mean values with standard deviations ormedian values with minimum and maximumvalues are presented as appropriate. Comparisonbetween groups was performed with a Chi-squaretest, t test and Mann-Whitney U test, when appro-priate. Data analyses were performed using SPSSsoftware (SPSS, Chicago, IL).

RESULTS

Patient characteristics and conditions. Patientcharacteristics, indication for cholecystectomy, andtype of injury are summarized in Table I. The

present series of patients is representative for thetotal cohort of 800 BDI patients referred to ourcenter; age (P = .58), gender (P = .45), indicationfor surgery (P = .96), type of initial procedure(P = .33), and type of injury (P = .53) did not differ.The majority of patients (n = 303; 57.4%) werereferred from nonteaching hospitals, and the pri-mary indication for surgery was symptomatic chole-lithiasis (n = 382; 72.3%). Table II summarizes thedetails of the surgical procedure. In 57%, theaspect of the gallbladder is documented (n =304). The procedure was complicated owinginflammation in 261 patients (49.4%), limitedview in 94 (17.8%), and instrumental problemsin 14 (2.7%).

Anatomic structures. The operation report men-tions identification of the cystic duct and artery in79.5% of the procedures (n = 420). The commonbile duct is reported to be identified in 39.2%and the right hepatic duct and artery in 7.2%and 9.7% respectively. Anatomic variations are re-ported in 81 (15.3%) operation reports. Generallyunspecified aberrant structures are reported; lessfrequently, an aberrant right hepatic duct is identi-fied or the anatomy is reported to be aberrantowing to a very short or missing cystic duct.

Safety measures and injury classification. TheCVS technique is documented in 33 patients(6.3%) and a complete dissection of Calot’s trian-gle is reported in 87 patients (16.5%). IOC wasperformed in 59 procedures (11.2%), but gener-ally after conversion to open surgery (n = 37), ifBDI was suspected (n = 38) and after repair of com-mon BDI with the use of a T-drain (n = 21).

Table III shows the postoperative diagnosed in-juries according to the Amsterdam, Strasberg,and Bismuth classifications. The use of the CVStechnique or the reported dissection of Calot’s tri-angle was not associated with the Amsterdaminjury classification (P = .47 and P = .50, respec-tively). IOC was more often performed in patientswith type D injury (P = .01), but generally only toconform the suspected injury or to locate the posi-tion of an inserted T-drain instead of a preventivemeasure during surgery.

Conversion and suspected BDI. A laparoscopicprocedure was started in 479 patients (91%), andconversion to open surgery was indicated in 180patients. The reported indication for conversionwas limited view or adhesions in 52.8% of thepatients (95/180), perioperative detected BDI in64 (35.6%), and bleeding in 21 (11.6%). Suspicionof BDI is described in 141 of the operation reports(26.7%). Interventions taken after this suspicionare summarized in Table II. If BDI was diagnosed,

Table I. Patient characteristics

Characteristic n = 528 %

GenderMale/female 163/365 31/69

Age at cholecystectomy (y)Mean (SD) 51 ± 16

Type of referring hospitalAcademic 21 4.0Teaching 204 38.6Non-teaching 303 57.4

Delay before referral (d),median (range)

21 (0–5,029)

Indication for cholecystectomySymptomatic cholelithiasis 382 72.3Acute cholecystitis 59 11.2Cholecystectomy a froid* 54 10.2Status after biliary pancreatitis 14 2.7Other 14 2.7Unknown 5 0.9

Initial procedureOpen 49 9.3Laparoscopic cholecystectomy 299 56.6Laparoscopic cholecystectomywith conversion

180 34.1

*Cholecystectomy performed in delayed setting after acute cholecystitis.

Table II. Operation details documented inoperation notes

Details n = 528 %

Procedures/conditionsEntry procedure 488 92.4Limited view 94 17.8Limited instruments 14 2.7Aspect gallbladder 304 57.6Aspect liver 45 8.5Inflammation 261 49.4

Reason for conversionLimited view 95/180 52.8Bile duct injury 64/180 35.6Bleeding 21/180 11.7Other 1/180 0.01

Reported anatomyCystic duct 489 92.6Cystic artery 430 81.4Cystic duct and artery 420 79.5Common bile duct 207 39.2Right hepatic duct 38 7.2Right hepatic artery 51 9.7Anatomic variations 81 15.3

Safety measuresTriangle of Calot 87 16.5CVS technique 33 6.3IOC 59 11.2

ComplicationsBleeding 172 32.6Bile leakage 198 37.5

Additional clip placement 38 7.2Extensive coagulation 184 34.8Suspicion of bile duct injury 141 26.7InterventionDirect repair/ETE/T-tube 84/141 59.2Biliodigestive anastomosis 24/141 16.9Drainage 23/141 16.8Hemihepatectomy 1/141 0.7None 9/141 6.3

Consultation ofA colleague 98 18.6An external HPB expert by phone 22 4.2

Postoperative policyFurther treatment 94 17.8

IOC intraoperative cholangiography

CVS, Critical view of safety; ETE, end to end anastomosis; HPB, hepatic-pancreatic-biliary; IOC, intraoperative cholangiography.

SurgeryMarch 2014

386 Booij et al

a biliodigestive anastomosis was performed in 24patients (16.9%), direct repair by end-to-end anas-tomosis with (n = 58) or without (n = 26) the use ofa T-tube was performed in 84 patients (59.2%),and in 1 patient a hemihepatectomy was per-formed during the initial procedure owing toextensive injury to the right hepatic artery andright portal vein. No repair was performed in 32patients in whom the suspicion of BDI was docu-mented in the report. In these patients, a drainwas left behind (n = 23) or nothing was done(n = 9) before a postoperative ERCP was orderedor patients were referred to a tertiary center.

Conversion and injury classification. Theoutcome in terms of injury classification in pa-tients in whom the injury occurred before con-version (n = 64) compared with patients in whomBDI occurred after conversion (n = 116) to opensurgery is described in Table IV. In patients inwhom conversion was performed owing to diag-nosed BDI, significantly more often a type Dinjury had occurred compared with patients inwhom the injury occurred after conversion (78%vs 56%; P < .01). Furthermore, the rate of com-plex injuries (Bismuth III–V) increased signifi-cantly if the injury occurred after conversion toopen surgery compared with patients in whomthe transection occurred before conversion(from 42% to 66%; P = .01).

Guideline implications. The analysis of theoperation reports and the use of safety measuresduring LC in BDI patients before and after guide-line implementation are presented in Table V.Operation reports dictated after the guideline im-plementation in 2007 (n = 103) resulted in an in-crease of documented CVS technique from 4% to17% (P < .01) compared with documented CVSin notes before 2007 (n = 425). However, no

Table III. Injury classification according to theAmsterdam, Strasberg and Bismuth classification

Injury classification n %

Type of injuryAmsterdam classificationType A 132 25.0Type B 94 17.8Type C 50 9.5Type D 252 47.7

Strasberg classificationType A 132 25.0Type B 11 2.1Type C 29 5.5Type D 93 17.6Type E 263 49.8

Location of injuryBismuth classification*I 13 2.5II 167 31.6III 111 21.0IV 63 11.9V 42 8.0

*Cystic duct leakage (n = 132) not included.

SurgeryVolume 155, Number 3

Booij et al 387

increase was measured in documentation ofanatomic structures, the dissection of Calot’s trian-gle, or the use of IOC. Of interest is the increasedconsultation of a hepatic-pancreatic-biliary (HPB)colleague during the initial procedure (from 2%to 8%; P < .01), and a reduction in delay in referralto a tertiary center from a median of 25 to 13 days(P < .01). No reduction in primary repair duringinitial operation was found (21% vs 18.5% after2007; P = .54).

DISCUSSION

The present series of patients who underwent acomplicated LC show that safety measures aretaken in the minority of procedures to preventBDI. Although guideline implementation in 2007significantly improved the use of the CVS tech-nique, the assessed use in this selected series is insharp contrast with the reported use of CVS in anational survey. This descriptive analysis also showsthat intraoperative suspicion of BDI does notalways lead to the identification of injury oradditional measurements for classification orrepair, even sometimes without adequate drainageof the abdominal cavity.

Since the introduction of LC, there has beenextensive literature about the incidence of BDI,risk factors and preventive measurements.1-3

Shortly after the introduction, injury rates of#2% were reported; more recently, there is a ten-dency to find lower rates around 0.2–0.3%. The

present study is not an attempt to obtain evidencefor the preventive function of CVS or IOC duringLC; however, it is illustrates that in this selectedgroup of BDI patients safety measurement ratesare strikingly low.

The most classical mechanism of BDI ismisidentification of the common bile duct forthe cystic duct.5,23 Technical hazards are excessivetraction on the gallbladder, misuse of cautery, orthe inappropriate use of clips.24 Aberrant anatomyis a local risk factor reported by the initial surgeonin 15% of the operation reports in the present se-ries of BDI patients.

Four years after the Dutch guidelines for LC wasintroduced by publication in the national surgicaljournal, at the annual surgical meeting, androutinely communicated to surgical residents intraining, Buddingh et al3 sent an electronic ques-tionnaire to all Dutch members of the society ofSurgery to investigate the use of safety measure-ments during LC. Ninety-two percent respondedpositive on the question if the CVS technique wasused and documented. This is in contrast withthe presented CVS rates of 6% in the present se-ries. The same questionnaire revealed that IOCwas never performed by 53% of the responders,and only 43% perform IOC in selective cases.This last finding is questionable; the present datashow that in these ‘‘selective cases,’’ IOC is rarelyperformed. The selection of BDI patients in thisstudy generally explains the differences inoutcome; however, safety measurements mightnot be applied as regularly as presented in the sur-vey. It could be hypothesized that a learning curvenot only exists in the operation itself, but inapplying safety measurements in a correct way aswell.

Postoperative dictated operation notes repre-sent the surgeons’ subjective opinion about theprocedure. Documentation of CVS in an operationnote might therefore not represent a true CVS asdescribed by Strasberg.16 Previous groups haveillustrated the discrepant accuracy in documenta-tion of the biliary anatomy on photos, video, orby IOC.7,16 IOC is suggested to be superior for ac-curate documentation; however, the limited use ofthis procedure supports the documentation byvideo or $1 or more photos.7 Although this wayof digital documentation is not obligatory, it willhelp to convince litigation experts and a judge toshow that appropriate safety measures were taken.

Conversion to open surgery is still presented asa safety measure by many surgeons. We previouslydiscussed the issue by presenting a case study inpatients in whom BDI occurred after conversion to

Table IV. Injury classification and location beforeand after conversion (n = 180)

Injuryclassification

Injury occurredbefore conversion,n = 64 (%)

Injury occurredafter conversion,n = 116 (%)

Pvalue

Type D injury(n = 115)

50 (78.1) 65 (56.0) <.01

Bismuth I-II 26 (57.8) 19 (42.2) .013Bismuth III-V 24 (34,3) 46 (65.7)

Table V. Improvements after Dutch guidelines(2007)

VariableBefore,

n = 425 (%)After,

n = 103 (%)P

value

Reported anatomyCystic duct

and artery335 (79) 85 (83) .01

Safety measuresDocumentation

Calot’s triangle66 (16) 21 (21) .22

Documentationof CVS

16 (4) 17 (17) <.01

IOC 50 (12) 9 (9) .38Consult of HPBsurgeon

14 (3) 8 (8) .04

Delay LC – diagnosisBDI, d (SD)Median days (range) 7 (0–4,842) 7 (0–202) .78Delay LC – referraltertiary centerMedian days (range) 25 (0–5,029) 13 (0–637) <.01

BDI, Bile duct injury; CVS, critical view of safety; HPB, hepatic-pancreatic-biliary; IOC, intraoperative cholangiography; LC, laparoscopiccholecystectomy.

SurgeryMarch 2014

388 Booij et al

open surgery.25 In those cases, conversion did notlead to more accurate assessment of the biliaryanatomy, while the extent of the injury increased.Owing to the current decline of experience inopen cholecystectomies we do not advocate toconvert in difficult cases, if experience is lacking.In these situations, subtotal cholecystectomy orpercutaneous drainage can be performed.26 Theanalysis in the present series of BDI patientsshow that common bile duct transection stilloccurred in 56%. More interestingly, however, isthe observation that if conversion is performedbecause of other reasons than detected BDI, therate of segmental injuries increases significantly.Segmental injury caused during LC is known as adiagnostic and therapeutic challenge associatedwith worse outcome.23,27

In cases with suspected BDI during surgery,different interventions were taken as described inthe result section. Of interest are the 32 patients inwhom only drainage or no intervention was per-formed during the initial operation. In thesepatients, a HPB surgeon was consulted in 31%(n = 10) of the operations, and referral to a tertiarycenter was made within a median of 5 days (range,0–708). The majority (n = 24) of these patients suf-fered from a type D injury, which was repaired bybiliodigestive anastomosis in 21 patients, hemihe-patectomy in 2 (of which 1 patient was referredafter 708 days after a biliodigestive reconstructionduring the initial operation), and a rendezvousprocedure and subsequent stenting in 1 patient.

In a situation in which perioperative bileleakage is owing to (extensive) tissue loss, inparticular in patients with more proximal lesionsat the bifurcation or intrahepatically, we previouslyadvised not to perform a primary repair.28 In thissituation, adequate drainage of the upper rightabdomen is strongly advised and the patientshould be referred for elective reconstruction.Referral to tertiary center in this situation has apositive effect on outcome.19

In conclusion, safetymeasures to prevent BDI arescarcely reported in the present series of patients

with a complicated LC. Guideline implementationattributes to better documentation, but more regu-lar application is mandatory. The present studyshows that IOC is rarely performed in TheNetherlands, even in this selected group of difficultcases. Based on these findings, we advocate todocument safety measures more properly, performIOC in experienced hands, and to consult acolleague or HPB surgeon in a timely manner toadjust intraoperative plans and avoid further injury.

REFERENCES

1. Strasberg SM. A teaching program for the ‘culture of safetyin cholecystectomy’ and avoidance of bile duct injury. J AmColl Surg 2013 May 22 [Epub ahead of print].

2. Wauben LS, van Grevenstein WM, Goossens RH, van derMeulen FH, Lange JF. Operative notes do not reflect realityin laparoscopic cholecystectomy. Br J Surg 2011;98:1431-6.

3. Buddingh KT, Hofker HS, ten Cate Hoedemaker HO, vanDam GM, Ploeg RJ, Nieuwenhuijs VB. Safety measures dur-ing cholecystectomy: results of a nationwide survey. World JSurg 2011;35:1235-41.

4. Buddingh KT, Morks AN, ten Cate Hoedemaker HO,Blaauw CB, van Dam GM, Ploeg RJ, et al. Documenting cor-rect assessment of biliary anatomy during laparoscopic cho-lecystectomy. Surg Endosc 2012;26:79-85.

5. Way LW, Stewart L, Gantert W, Liu K, Lee CM, Whang K,et al. Causes and prevention of laparoscopic bile duct in-juries: analysis of 252 cases from a human factors and cogni-tive psychology perspective. Ann Surg 2003;237:460-9.

6. Flum DR, Cheadle A, Prela C, Dellinger EP, Chan L. Bileduct injury during cholecystectomy and survival in Medi-care beneficiaries. JAMA 2003;290:2168-73.

SurgeryVolume 155, Number 3

Booij et al 389

7. Buddingh KT, Weersma RK, Savenije RA, van Dam GM,Nieuwenhuijs VB. Lower rate of major bile duct injuryand increased intraoperative management of commonbile duct stones after implementation of routine intraoper-ative cholangiography. J Am Coll Surg 2011;213:267-74.

8. Flum DR, Dellinger EP, Cheadle A, Chan L, Koepsell T. In-traoperative cholangiography and risk of common bile ductinjury during cholecystectomy. JAMA 2003;289:1639-44.

9. HobbsMS,MaiQ,KnuimanMW, FletcherDR,Ridout SC. Sur-geon experience and trends in intraoperative complicationsin laparoscopic cholecystectomy. Br J Surg 2006;93:844-53.

10. Waage A, Nilsson M. Iatrogenic bile duct injury: a popula-tion based study of 152 776 cholecystectomies in the Swed-ish Inpatient Registry. Arch Surg 2006;141:1207-13.

11. Stewart L, Way LW. Bile duct injuries during laparoscopiccholecystectomy. Factors that influence the results of treat-ment. Arch Surg 1995;130:1123-8.

12. Strasberg SM, Brunt LM. Rationale and use of the criticalview of safety in laparoscopic cholecystectomy. J Am CollSurg 2010;211:132-8.

13. Overby DW, Apelgren KN, Richardson W, Fanelli R. Societyof American Gastrointestinal and Endoscopic Surgeons.SAGES guidelines for the clinical application of laparo-scopic biliary tract surgery. Surg Endosc 2010;24:2368-86.

14. Widdison AL, Norton S, Armstrong CP. Open cholecystec-tomy in the age of the laparoscope. Ann R Coll Surg Engl1995;77:256-8.

15. Visser BC, Parks RW, Garden OJ. Open cholecystectomy inthe laparoendoscopic era. Am J Surg 2008;195:108-14.

16. Wauben LS, Goossens RH, Lange JF. Differences betweenattendings’ and residents’ operative notes for laparoscopiccholecystectomy. World J Surg 2013;37:1841-50.

17. Emous M, Westerterp M, Wind J, Eerenberg JP, van GelovenAA. Registering the critical view of safety: photo or video?Surg Endosc 2010;24:2527-30.

18. Plaisier PW, Pauwels MM, Lange JF. Quality control in lapa-roscopic cholecystectomy: operation notes, video or photoprint? HPB 2001;3:197-9.

19. de Reuver PR, Rauws EA, Bruno MJ, Lameris JS, Busch OR,van Gulik TM, et al. Survival in bile duct injury patients afterlaparoscopic cholecystectomy: a multidisciplinary approachof gastroenterologists, radiologists, and surgeons. Surgery2007;142:1-9.

20. Bergman JJ, van den Brink GR, Rauws EA, de Wit L, Ober-top H, Huibregtse K, et al. Treatment of bile duct lesions af-ter laparoscopic cholecystectomy. Gut 1996;38:141-7.

21. Strasberg SM, Hertl M, Soper NJ. An analysis of the prob-lem of biliary injury during laparoscopic cholecystectomy.J Am Coll Surg 1995;180:101-25.

22. Bismuth H, Lazorthes F. 83rd Congress of the French Surgi-cal Society (Paris, 21-24 September 1981). Second report.Operative injuries of the common biliary duct. J Chir(Paris) 1981;118:601-9.

23. Lillemoe KD, Petrofski JA, Choti MA, Venbrux AC, Ca-meron JL. Isolated right segmental hepatic duct injury: adiagnostic and therapeutic challenge. J Gastrointest Surg2000;4:168-77.

24. Strasberg SM. Avoidance of biliary injury during laparo-scopic cholecystectomy. J Hepatobiliary Pancreat Surg2002;9:543-7.

25. Booij KA, de Reuver PR, van Delden OM, Gouma DJ. Con-version has to be learned: bile duct injury following conver-sion to open cholecystectomy. Ned Tijdschr Geneeskd 2009;153:A296.

26. Henneman D, da Costa DW, Vrouenraets BC, van Wagens-veld BA, Lagarde SM. Laparoscopic partial cholecystectomyfor the difficult gallbladder: a systematic review. Surg En-dosc 2013;27:351-8.

27. Perini RF, Uflacker R, Cunningham JT, Selby JB, Adams D.Isolated right segmental hepatic duct injury following lapa-roscopic cholecystectomy. Cardiovasc Intervent Radiol2005;28:185-95.

28. de Reuver PR, Grossmann I, Busch OR, Obertop H, van Gu-lik TM, Gouma DJ. Referral pattern and timing of repair arerisk factors for complications after reconstructive surgeryfor bile duct injury. Ann Surg 2007;245:763-70.