Minimally Invasive Surgery - BMJ · 3 McDowell I, Newell C. Measuring health. A guide to rating scales and questionnaires. NewYork: OxfordUniversityPress, 1987. 4 Eastwood RP. Sales

many of the completed forms as you can personally anddraw the interviewers' attention to the problems. Ifyou cannot identify any, look again: you have simplymissed them. At least one of the questions will tum outto be hopeless despite steps 1 to 8; this is par for thecourse and not worth losing sleep over. Take pains toachieve a high response rate, especially in postalsurveys. If the response rate is poor and you areconfident that your questionnaire is not to blame sendout a reminder along with a second and even a thirdcopy of the form. Avoid haranguing non-respondentsbut emphasise to them how important their coopera-tion is to the success of your extraordinarily importantstudy. Flattery works.

(10) Start again-Good research is usually the resultof learning from mistakes. If time and resources (aswell as personal motivation) permit replicate yourstudy at least once. This will allow you to performvalidation, to increase the sample size, and to fine tuneyour questionnaire to the point where you can be proudof it.

You now know about as much theory as you need toget you started on your survey. For those with the timeand inclination, more detailed advice on questionnairedesign can be found in good medical libraries.67 Someaspects of the subject may appear highly technical andcomplicated to the point where you may be deterredfrom going further. As there is no substitute forexperience, however, take a deep breath and jump.The landing will not be as hard as you fear.

1 Bennett AE, Ritchie K. Questionnaires in medicine. A guide to their design and use.London: Oxford University Press (for the Nuffield Provincial HospitalsTrust), 1975.

2 Hulley SB, Cummings SR, eds. Designing clinical research. Baltimore: Williamsand Wilkins, 1988.

3 McDowell I, Newell C. Measuring health. A guide to rating scales andquestionnaires. New York: Oxford University Press, 1987.

4 Eastwood RP. Sales control by quantitative methods. New York: ColumbiaUniversity Press, 1940.

5 Cartwright A. Health surveys in practice and in potential: a critical review of theirscope and methods. London: King Edward's Hospital Fund, 1983.

6 Sudman S, Bradbum NM. Asking questions. London: Jossey Bass, 1982.7 Streiner DL, Norman GR. Health measurement scales. A practical guide to their

development and use. Oxford: Oxford University Press, 1989.

(Accepted 23August 1993)

Minimally Invasive Surgery

General surgery: biliary surgery

This is thefirst in a series ofarticles on minimally invasivesurgery

Middlesex Hospital,London WIN 8AARC G Russell, consultantsurgeon

BMJ 1993;307: 1266-9

R C G Russell

The management of biliary tract disease haschanged completely as a result ofminhimally invasivetreatment. For most patients with gallstones thatcause symptoms a laparoscopic cholecystectomywill treat the condition with minimal morbidity anda short recovery period. If complications areencountered, conversion to a mini-cholecystectomygives results that are nearly as good. Acute chole-cystitis can be treated by percutaneous drainagefollowed either by percutaneous cholecystolitho-tomy or a laparoscopic cholecystectomy. Gallstonesin the bile duct are best treated by endoscopicsphincterotomy with duct clearance. The day of thelarge cholecystectomy scar with its subsequentincisional hernia has gone.

The concepts of minimally invasive surgery in thebiliary tree grew out of the role of the endoscopist intreating gallstones in the bile duct by endoscopicsphincterotomy. From then on the search for a mini-mally invasive technique to deal with stones in thegallbladder took many tums. Techniques to intubatethe gallbladder endoscopically proved difficult, and itis still only rarely possible to negotiate the valves of thecystic duct to gain entry to the gallbladder. A betterapproach proved to be a percutaneous transhepaticpuncture of the liver with the insertion of a catheter,through which solvents could be injected and stonesdissolved. This proved time consuming and technic-ally difficult, and the operation has been largelyabandoned. For draining an acute empyema of thegallbladder, however, percutaneous drainage can be adramatic lifesaving and non-invasive technique.

It became apparent that to gain access to thegallbladder and remove the stones it was necessary topass larger catheters into the gallbladder, and thusthe technique of percutaneous nephrolithotomy wasadapted to the gallbladder. This technique provedsuccessful in removing the gallstones, but a drain hadto be left in the gallbladder to allow the gallbladder toheal round it so that bile did not leak into the peritonealcavity on its removal. This was done 10 days later, after

the biliary tree had been checked radiologically toensure that there were no residual stones. The majordisadvantage of this technique is that up to a third ofpatients are subject to formation of new stones in thegallbladder, and more than half of these patients haveto have their gallbladder removed within three years ofthe original procedure.

After trials with oral dissolution, contact dissolution,extracorporeal lithotripsy, percutaneous cholecysto-lithotomy, and rotary lithotripsy attention is nowfocused almost entirely on cholecystectomy for themanagement of gallstones either by the laparoscopictechnique or by mini-cholecystectomy. Endoscopicretrograde cannulation of the bile duct remainspre-eminent as the method of dealing with a gallstonein the bile duct by minimally invasive technology.

Laparoscopic cholecystectomyThe standard treatment for gallstones in developed

countries is laparoscopic cholecystectomy. This treat-ment was first described in Germany in 1985 but waspublished in an obscure journal and received littlepublic acclaim. Mouret in Lyons, who is both a generaland a gynaecological surgeon, performed the firstpublicised laparoscopic cholecystectomy in March1987. Dubois in Paris, who for a long time had beenadept at minicholecystectomy, progressively replacedthis approach with laparoscopic cholecystectomy fromFebruary 1988.' In June 1988 McKeman and Sayeperformed the first laparoscopic cholecystectomy witha laser to dissect the gallbladder.2 This techniquedeveloped rapidly under the stimulus provided byReddick in Nashville, Tennessee, from October 1988.The world at large became familiar with the tech-

nique when Perrisat from Bordeaux presented a videoof it to the Society of American GastrointestinalEndoscopic Surgeons in April 1989.3 By the spring of1990 the operation was performed in numerous centresin the United Kingdom. In 1992 over 60% of chole-cystectomies performed in the United Kingdom weredone by the laparoscopic method.

1266 BMJ VOLUME 307 13 NOVEMBER 1993

on 13 June 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.307.6914.1266 on 13 N

ovember 1993. D

ownloaded from

INDICATIONS FOR SURGERY

In the past year the number of cholecystectomiesperformed in the United Kingdom and worldwide hasincreased from the 600-800 procedures per millionpopulation per year that had been undertaken, withslight variation from region to region in previous years.It is tempting to suggest that the recent rise is due to anentrepreneurial effort to leam and practise a newtechnique that is of interest to the technically orientedsurgeon. No surgeon would deny such temptation, butthe public is not blameless: patients who previouslyput up with an occasional attack of biliary colicsuddenly appear to be keen to have the minor proce-dure, and patients who have just had their first attackare keen to have the new operation in case severeinflammation makes the old operation necessary. Ianticipate that the present rate of operations will settledown to the previous level or to an even lower numberif present dietary trends continue.The prime indication for cholecystectomy is biliary

colic, with its characteristic and well defined pain.Acute cholecystitis, with its complications, is an urgentindication for cholecystectomy. Indeed, an attack ofgallstone related disease necessitating admission tohospital is an indication for cholecystectomy duringthat admission or soon after to prevent more severesymptoms or complications developing. Jaundice dueto gallstones is a further indication for cholecystectomyafter the gallstones have been treated endoscopicallyexcept in elderly patients, for whom the endoscopicprocedure is sufficient in 85% of cases. Patients withnon-specific dyspepsia or symptoms suggestive ofthe irritable bowel syndrome should only be offeredcholecystectomy with caution. Rigid avoidance ofthe procedure with such patients could deny somesufferers effective treatment, but patients should beclearly told that some symptoms may remain afterthe operation. There is no such thing as the post-cholecystectomy syndrome, but the incidence ofgastrointestinal symptoms after the operation will bevery high if unsuitable patients are given cholecystec-tomy. It can now be stated dogmatically that symptom-less gallstones do not require treatment.There are situations when a surgeon may be un-

willing to undertake laparoscopic cholecystectomy.When the technique was first developed it was thoughtto be inadvisable in patients who were pregnant,grossly obese, or who had just recovered from acutepancreatitis. Ideas have changed, and these conditionsare now considered ideal for the laparoscopic tech-nique. Many surgeons with a keen interest in laparo-scopic surgery will first consider the laparoscopicapproach for all patients who require a cholecystec-tomy. Attempts have been made to identify patientswho are unsuitable for the procedure by preoperativeultrasonography: it is known that patients with a thickwalled gallbladder, a large stone or evidence of inflam-mation in Hartmann's pouch, or a localised abscess willprove difficult, but the laparoscopic approach can stillbe undertaken in most of these patients.

SURGICAL TECHNIQUE

Preoperative preparation-Patients should undergopreoperative ultrasonography for examination of thegallbladder and bile duct. If the gallbladder is highlyinflamed or has thickened walls or there is pus presentthe patient should be wamed that conversion to opencholecystectomy may be necessary. Similarly, if thecommon bile duct is more than 5 5 mm in diameter apreoperative endoscopic cannulation of the bile ductshould be performed to remove any stone present.Patients with tests showing abnormal liver functionand with a history of jaundice should probably havepreoperative endoscopy. Many centres also undertakeintravenous cholangiography before laparoscopic

cholecystectomy to assess anatomy. Patients must beassessed by an anaesthetist to ensure that they are fit forgeneral anaesthesia and a pneumoperitoneum with apressure of 15 mm Hg. Anaesthetised patients areplaced on the operating table either flat or in a modifiedlithotomy position so that the surgeon can standbetween the legs.Equipment-For a pneumoperitoneum a high flow

carbon dioxide insufflator, preferably with a facilityfor recirculating the carbon dioxide through a filter toremove smoke, is needed. Adequate safety alarmsmust be attached to the insufflator so that excessivepressures of carbon dioxide cannot occur in theabdomen. As diathermy is used for much of the dissec-tion and is the main method of stopping haemorrhage,an efficient coagulation diathermy machine isimportant. Means of suction and of flushing theperitoneal cavity with saline are necessary to ensure afield free of blood and other fluids such as bile. A videocamera with its attached light source is vital becauseaccurate surgery cannot be undertaken without clearvision. The new three chip cameras attached to highdefinition television monitors greatly ease the dissec-tion and overcome the problems associated with twodimensional as opposed to three dimensional vision. Abackup facility for each piece of apparatus is needed incase of failure.

Gaining access-Two Kochers forceps are placed onthe skin of the umbilicus, which is everted, and atransverse incision is made in the umbilicus. A Verresneedle attached to the carbon dioxide insufflator ispassed into the incision, the Kochers forceps are liftedup, and the needle is passed through the linea alba andperitoneum into the peritoneal cavity. The formationof the pneumoperitoneum can then be started, andinsufflation continues until the pressure in the intra-abdominal cavity reaches 15-17 mm Hg. A trocar and acannula are then placed through the same umbilicalincision into the abdominal cavity. I prefer non-disposable trocars as the cost of the disposable trocarmakes the procedure uneconomic. The laparoscope, towhich the video camera is attached, is advanced intothe peritoneal cavity, and the gallbladder and peritonealcavity are examined. Provided that the gallbladderseems suitable for laparoscopic removal, a 10 mmepigastric trocar is inserted through the linea alba toenter the peritoneal cavity just to the right side of thefalciform ligament, and 5 mm trocars are placedlaterally near the subcostal margin and in the mid-clavicular line, preferably at the level of the umbilicus,to provide good triangulation between the threeworking ports.

Freeing the gallbladder-Grasping forceps are putinto the lateral trocar and placed on the fundus of thegallbladder, which is then lifted up with the liver toexpose the rest of the gallbladder and the subhepaticspace (fig 1). Adhesions are dissected away from thegallbladder to expose the whole organ. The secondgrasping forceps is placed on Hartmann's pouch andpulled downwards to expose the triangle of Calot. Thediathermy hook is passed through the epigastriccannula, and a careful inspection of the anatomy ismade to define the position of the duodenum, thecommon bile duct, and the artery and its cysticduct branch. The peritoneum at the lower border ofthe gallbladder is then divided both anteriorly andposteriorly from the cystic duct to the liver bed (fig 2).These dissections are deepened until the branches ofthe cystic artery are found (fig 3): the cystic arteryalmost always breaks up into numerous branchesbefore reaching the gallbladder. Individual branchescan be coagulated by diathermy, or the main trunk canbe clipped. The principle of this dissection is to remainas close to the gallbladder as possible. When thetriangle of Calot has been completely exposed and the

BMJ VOLUME 307 13 NOVEMBER 1993 1267

on 13 June 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.307.6914.1266 on 13 N

ovember 1993. D

ownloaded from

FIG 1-Fundus of gallbladder is lifted up to show FIG 2-Adhesions separated by hook dathenmy and now FIG 3-Tnangle of Calot being exposed by dividingomentum attached to lower border of slighdy inflamed duodenum being separatedfromn gallbladder by division of peritoneum over lower border of gallbladder away fromgallbladder adhesions bile duct and on gallbladder wafl

diathermy hook can be passed from the front dissectionto the back dissection the cystic duct is cleared oftissueso that it can be clipped (fig 4). At this stageperoperative cholangiography can be done. Only 50%of surgeons think that this is necessary because by thisstage of the operation the main anatomy has beendefined and ifthe triangle of Calot has been completelyexposed there is little danger ofdamage to the commonhepatic or right hepatic ducts. When the cystic ducthas been exposed two absolok clips are placed near tothe common bile duct and a ligaclip is placed near tothe gallbladder (fig 5). The cystic duct is divided, andthe gallbladder is then dissected from the gallbladderbed until it is free of all attachment to the liver. Somesurgeons do this by laser, but diathermy is adequateand much cheaper.

Extracting the gallbladder-The gallbladder isextracted through the umbilicus (fig 6). If the gall-stones are large these have to be removed separatelythrough the umbilicus with De Jardin forceps. Largestones and big or thick walled gallbladders can providetechnical difficulties, especially if the gallbladder hasbeen torn or punctured during a tedious dissection.The preferred technique is to place the gallbladder ina specially constructed bag, which is brought outthrough the umbilical incision, and to break up thestones with forceps in the bag so that the debris can beremoved without contamination of the peritonealcavity. When the gallbladder has been extracted gas issucked from the abdomen, and the wounds are closedwith a 5/0 subcuticular suture.

Conversion to open surgerv-Probably the mostdifficult technical situation which the laparoscopistencounters is bleeding. If the cystic artery or the righthepatic artery is damaged the application of clips in theabsence of clear vision can lead to the inappropriateapplication of the clip to the common bile duct.Similarly, diathermy without clear anatomical defini-tion can cause damage to the biliary tree and subse-quent formation of a fistula. A large stone or grossinflammation within the triangle of Calot can lead to adifficult dissection so that it is almost impossible to

dissect the cystic duct safely. If any difficulty isencountered during the procedure conversion to amini-cholecystectomy is safe, proper, and appropriate:persisting for many hours trying to dissect difficultanatomy is not. The differences between mini-chole-cystectomy and laparoscopic cholecystectomy are sosmall that excessive operative times can no longer bejustified. A conversion rate of 5% is acceptable, andhigher and lower rates are probably equally inappro-priate.

Postoperative care-Recovery is rapid after thisprocedure, and the patient is usually up on the eveningof the day of the operation and drinking normally.Intravenous fluids are stopped either on the eveningafter the operation or the following morning. Morethan half of patients leave hospital within 24 hours ofthe operation while the rest leave within three days.Complications are rare if the procedure is done care-fully, but if a complication is suspected immediateultrasonography of the abdomen should be undertakento look for collections of fluid in the subhepatic space orbleeding. If fluid is seen a drain should be insertedpercutaneously with ultrasound or x ray guidance. Ifbile is found in the drainage fluid an endoscopiccannulation of the bile duct should be done within 24hours. If contrast medium leaks from the stump of thecystic duct during endoscopic cholangiography it maybe due to the clip slipping or to a small biliary radical inthe gallbladder bed, but endoscopic sphincterotomyinvariably solves the problem. Narrowing of the bileduct requires immediate balloon dilatation or stentingto salvage a difficult situation, while if there istransection of the duct the appropriate operativeprocedure is indicated. Bleeding is rare after thisoperation, and in my experience it can usually bemanaged conservatively. Wound infections are minorand rarely cause problems.

OUTCOME

Critics of this operation point to the large number ofcomplications that, according to hearsay, occurs withthis procedure. There is now, however, good evidence

FIG 4-Triangle of Calot widey opened and cystic duct FIG 5-Cystic duct now cleaned and two absolok dips FIG 6-Freed galbladder being puled out throughbeing cleared oftissue placedproximaly with a ligaclip on gallbladder umbilicus

BMJ VOLUME 307 13 NOVEMBER19931268

-s ",Wqw-,"

-..

on 13 June 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.307.6914.1266 on 13 N

ovember 1993. D

ownloaded from

from Europe and the United Kingdom that the rate ofcomplication of the laparoscopic method is at least asgood as that of open cholecystectomy if not better. Arecently published American review of 77600 casesreported that laparotomy for the treatment of acomplication was required by 1-2% of patients.4 Themean rate of bile duct injury was 0-6%, and theseinjuries occurred mainly during the first 100 cases.Postoperative bile leak was recognised in 0 3%, andonly 18 of the 33 postoperative deaths resulted fromoperative injury. Furthermore, the long term rate ofcomplication was reduced by the low number ofcomplications associated with the wound, such asinfection and incisional hernia. As already mentioned,patients return to work much more rapidly than afteropen surgery.

Gallstones in the bile ductThe management of stones in the common bile duct

is still a controversial subject. About 5%-10% ofpatients who have a cholecystectomy will have a stonein the bile duct. This may present with jaundice or bediagnosed at the time of cholecystectomy by operativecholangiography. Only a half of surgeons regularlyperform operative cholangiography at cholecystec-tomy; a quarter undertake cholangiography selectively,usually because of a history of jaundice, the presenceof an enlarged bile duct, or a test showing abnormalliver function; and a quarter never perform theprocedure. If a gallstone is found by cholangiography itcan either be left for the patient to undergo anendoscopic sphincterotomy later, or the stone can beremoved during the laparoscopic procedure. Onemethod is to dilate the cystic duct and then either seeand retrieve the stone by means of a cholangioscopepassed through the hole in the cystic duct or removethe stone by basketing techniques with radiologicalguidance. Alternatively, some surgeons directly incisethe duct as at open surgery and remove the stone. Thistechnique has the advantage that the whole duct can beexplored more easily, but there is more contaminationand the defect has to be sutured, which is difficult toperform laparoscopically.

Others argue that an endoscopic sphincterotomy ispreferable. This technique is safe in competent hands,it is thorough in that good clearance of the whole ductis achieved, and it prevents stasis within the bile ductleading to formation of another stone. Initial worriesabout the long term effects of sphincterotomy have sofar proved groundless.

Extracorporeal shock wave lithotripsyExtracorporeal shock wave lithotripsy was

pioneered in Munich and was first used in 1980 to treatkidney stones.5 Experimental work suggested that thistechnique would also fragment gallstones.67 In theprocedure shock waves are generated, focused, andtransmitted to the body through a liquid medium,usually water, that has an acoustic impedance similarto that of body tissues. The shock waves spreadthrough the body with little energy loss and thereforeminimal tissue damage. The stone must be accuratelypositioned at the focal point of the waves for the releaseof the high energy on impact, which creates mechanicalstresses and leads to fragmentation of the stone. Afterfragmentation of the gallstone in the gallbladder thespontaneous discharge of stone fragments is unlikely tobe complete. These fragments must negotiate thecystic duct and its valves of Heister, the common bileduct (which has no peristalsis), and the sphincter ofOddi to reach the duodenum. Emptying of the normalgallbladder is slow and incomplete, and in patientswith gallstones it is considerably impaired. Adjuvant

treatment is therefore given in the form of oral bileacids to dissolve the fragments and to allow themto negotiate the pathway to the intestine. The bestresults suggest that 75%-95% of stones are fragmentedcompletely, but it takes up to 18 months for 91% ofpatients to become clear of stones.8 The limitedinformation on recurrence of gallstones suggests that athird of patients will have a recurrence in three years.This technique is probably not useful for managinggallstones, and the rare instance of a stone stuck in thecommon bile duct, which is difficult to remove byendoscopic sphincterotomy, is likely to be its onlyindication.

The acute gallbladderLaparoscopic cholecystectomy is difficult in the

presence of complications such as empyema or perfora-tion, and in elderly and unfit patients with acutecholecystitis a laparoscopic or open approach may bedangerous. Minimally invasive treatment can solve theproblem of acute cholecystitis: a puncture of the acutegallbladder can easily be made with ultrasound guid-ance and with minimal upset to a patient under localanaesthesia. Once a needle is in place a guide wire canbe inserted, and a drain can be placed over theguidewire. After an acute gallbladder has been drainedthere is usually immediate resolution of symptoms,and the systemic effects resolve rapidly. It is thenpossible to proceed to laparoscopic cholecystectomyor, if the patient is unfit or elderly, percutaneouscholecystolithotomy under local anaesthesia. Withx ray guidance the tract is dilated with a 9 or 12French gauge catheter, a cholangioscope is pushed intothe gallbladder, and the stones are shattered by meansof contact lithotripsy. The gallbladder is then flushedout to completely remove the stone fragments. A smalldrain is inserted for repeat cholangiography two tothree days later, and if the gallbladder and biliary treeare clear of stones the drain is removed. Most patientswill have no further trouble from their gallbladder,which will shrink to a small size.

1 Dubois F, Icard P, Berthelot G, Levard H. Coelioscopic cholecystectomy.Preliminary report of 36 cases. Ann Surg 1990;211:60-2.

2 McKernan JB, Saye WB. Laparoscopic general surgery. J Med Assoc Ga1990;79:157-9.

3 Perissat J, Collet DR, Belliard R. Gallstones: laparoscopic treatment, intra-corporeal lithotripsy followed by cholecystostomy or cholecystectomy-apersonal technique. Endoscopy 1989 [suppl 1];21:373-4.

4 Deziel DJ, Millikan KW, Economou SG, Doolas A, Ko S-T, Airan MC.Complications of laparoscopic cholecystectomy: a national survey of 4,292hospitals and an analysis of77,604 cases. AmJ Surg 1993;165:9-14.

5 Chaussy C, Schmiedt E, Jocham D, Brendel W, Forssmann B, Walther V. Firstclinical experience with extracorporeally induced destruction of kidneystones by shock waves. J Urol 1982;127:417-20.

6 Brendel W, Enders G. Shock waves for gall stones: animal studies. Lancet1983;i:1054.

7 Sauerbruch T, Delius M, Paumgatener G, Holl J, Wess 0, Weber W, et al.Fragmentation of gallstones by extracorporeal shock waves. N Engi J Med1986;314:818-22.

8 Sackman M, Delius M, Sauerbruch T, Holl J, Weber W, Ippisch G, et al. Shockwave lithotripsy of gallbladder stones. The first 175 patients. N Engl J Med1988;318:393-7.

Correction

Arthroscopic surgery compared with supervised exercisesin patients with rotator cuffdisease (stage II impingementsyndrome)A typesetting error and an editorial error occurred in this paper byJens Ivar Brox and colleagues (9 October, p 899). In table III themedian scores for function at 6 months should have read 21-0 forplacebo laser treatment and 26-0 for supervised exercises (not 15-0and 25-0, respectively, as published). The third sentence of thethird paragraph of the discussion should have read: "Verbal painratings as part of the Neer shoulder score were always madeduring the clinical examination, but patients scored the pain theyhad experienced during the previous week on nine point scalesbefore the examination. Verbal pain ratings may therefore havebeen biased."

BMJ VOLUME 307 13 NOVEMBER 1993 1269

on 13 June 2020 by guest. Protected by copyright.

http://ww

w.bm

j.com/

BM

J: first published as 10.1136/bmj.307.6914.1266 on 13 N

ovember 1993. D

ownloaded from