surgical endoscopy aug1998

7/27/2019 surgical endoscopy aug1998

1/99

Letter to the editor

Telesurgical laparoscopic cholecystectomy

On March 3 1997, a telesurgical laparoscopic cholecystec-tomy was performed for the first time in history at the St.Blasius hospital in Dendermonde, Belgium. The deviceused was the Mona from Surgical Intuitive, MountainView, California, USA.

After clearance from the local ethical committee and

after informed consent had been obtained, the patient, a72-year-old woman with a body mass index of 42 kg/m2

was put under general anesthesia with endotracheal intuba-tion. Four 10-mm and one 5-mm trocar cannulas were in-serted according to Dubois technique [1]. One 10-mm can-nula harbored a straight-looking optical system connected toa three-dimensional camera system. Another 10-mm can-nula was placed to the left of the umbilicus and used for clipplacement only. The 5-mm trocar contained a probe for liverretraction. The two remaining active ports were con-nected to two fully mobile mechanical arms attached to thesiderails of the operating table. They harbored two articu-

lated tools (end effectors): a grasper and an electrocauteryhook commanded by the surgeon sitting at a working con-sole approximately 15 feet away from the patient.

As the surgeon watched the three-dimensional image ofthe operative field, he manipulated two handles that trans-mitted impulses to and from the end effectors via a com-puter interface. Sensory input and downscaling of the sur-geons motions 4 to 1 were secured. The procedure wassuccessfully performed in 82 min, and the patients recoverywas uneventful.

The use of a computer interface between surgeon andpatient has many more advantages than the ability to per-

form operations on a site remote from the patient.The translation of virtual manipulations into commandsto robot arms allows the surgeon to deal with the three mostsignificant shortcomings of laparoscopic surgery: (a) the

reduction in degrees of freedom, (b) the lack of tactile feed-back, and (c) the impaired dexterity attributable to the use oflong and rigid instruments introduced through a fixed pointin the abdominal wall and manipulated in an often awkwardposition [2].

With the present master-slave system, seven degrees of

freedom are acquired because the end effectors have anadditional articulation (wrist) inside the abdominal cavity,perfectly mimicking all motions as they are performed bythe surgeon. Moreover, the system is capable of reducing(downscaling) the virtual manipulations of the operator,hereby eliminating minor flaws such as physiologic tremor.This, combined with the ergonomically optimal position,the visual immersion, and the sensory feedback on the sur-geons part, allow for a more intuitive, nearly perfect sur-gical approach, hence more precision, and thus improvedsafety for the patient. More complex endoscopic operationssuch as microanastomosis in the cardiovascular field will

now become feasible with the same accuracy as in the openchest procedure.

References

1. Dubois F, Berthelot G, Levard H (1991) Laparoscopic cholecystectomy:historic perspective and personal experience. Surg Laparosc Endosc 1:5257

2. Wapler M (1995) Medical manipulators: a realistic concept? Minim

Invasive Ther 4: 261266

J. Himpens1,2

G. Leman1

G. B. Cadiere

2

1Department of Surgery

Saint Blasius Hospital50, KroonveldlaanDendermonde 9200, Belgium2Department of SurgerySaint Pierre University HospitalBrussels, BelgiumCorrespondence to: J. Himpens

Springer-Verlag New York Inc. 1998Surg Endosc (1998) 12: 1091


2/99

Transthoracic induction of a hiatal hernia in domestic swine

F. J. Brody, J. Hunt, J. Sackier

Department of Surgery, Suite 3B, The George Washington University Medical Center, 2150 Pennsylvania Avenue NW, Washington, DC 20037, USA

Received: 22 August 1996/Accepted: 29 January 1997

AbstractBackground: With the common performance of laparoscop-ic Nissen fundoplication for gastroesophageal reflux dis-ease, there is renewed interest in the pathophysiology andpotential histologic consequences of hiatal hernias. How-ever, in vivo model exists that both reliably reproduces thehiatal hernia and is amenable to subsequent laparoscopicrepair.

Methods: A transthoracic approach was used to induce ahiatal hernia surgically in female James pigs (50160 kg; n

5).Results: Hiatal hernias were successfully induced in all pigsand verified with barium swallow, endoscopy, and/or lapa-

roscopy. Laparoscopic reduction and Nissen fundoplicationwere subsequently completed on each animal on postopera-tive day 30. One postoperative death occurred on postop-erative day 4 after thoracotomy.Conclusions: We describe the induction of a hiatal herniavia a transthoracic approach in domestic swine. The hiatalhernia is amenable to subsequent laparoscopic repair, en-abling surgeons to acquire the technical skills required tocorrect this defect in the laboratory. To our knowledge, thisis the first report of a reproducible model of a transthoraci-cally induced hiatal hernia that allows subsequent laparo-scopic repair. We suggest that in addition to refinement of

surgical skills, our model may provide new information toresearchers regarding the potential indications for antirefluxprocedures, as well as the natural history and appropriatemanagement of hiatal hernias.

Key words: Gastroesophageal reflux Pig Laparo-scopic Nissen fundoplication Hiatal hernia Animalmodel

Approximately 33% of individuals with hiatal hernias suffer

from gastroesophageal reflux disease [6, 7]. The vast ma-jority of these individuals (80%) are successfully treated

medically without surgical intervention. The remaining20%, however, require surgical intervention to repair thehiatal hernia and prevent pathophysiologic reflux [1]. Sur-gical interventions such as Nissen fundoplication are suc-cessfully performed laparoscopically. To our knowledge,there is no in vivo model of a transthoracically inducedhiatal hernia that can subsequently be repaired laparoscopi-cally. We sought to test the hypothesis that a hiatal herniacould be transthoracically induced for subsequent laparo-scopic repair in an animal model.

Materials and methods

Five female James (50160-kg) pigs were used. The animals were anes-thetized with intramuscular injections of 1.0 mg/kg of xylazine, 6.0 mg/kgof telazol, and 0.01 mg/kg of atropine. The pigs were intubated and me-

chanically ventilated with anesthesia maintenance using 2.0% isoflourane.Selective intubation of the right mainstem bronchus was performed with aflexible bronchoscope. The animals were positioned on their right side, anda left posterolateral thoracotomy was performed through the sixth inter-costal space. A rib spreader was applied for retraction. The inferior pul-monary ligament was divided, and the left lower lobe of the lung wasgently retracted superiorly to expose the distal esophagus and aorta. Thepleura overlying the esophagus was incised exposing the vagus nerves. APenrose drain was secured around the esophagus, and vagus nerves andcephalad traction was applied. The phrenoesophageal membrane was vi-sualized at the gastroesophageal junction, and circumferential dissectionwith Metzenbaum scissors and electrocautery was performed to incise thisstructure thoroughly. The dissection completely freed the esophagogastric

junction from its diaphragmatic attachments and allowed retraction of thelower esophagus and gastric fundus into the posterior mediastinum. Toallow elevation of the gastric fundus into the chest, the most proximal shortgastric vessel was ligated. The gastric body was secured to the posterioraspect of the left crus of the diaphragm with one suture.

An 18 French red rubber catheter was inserted in the left chest througha stab wound above the eighth rib for evacuation of the pneumothorax. Thechest wall musculature was reapproximated in three layers and the skinclosed with a continuous subcuticular suture. A Heimlich valve was ap-plied to the distal end of the red rubber catheter, and the pig was weanedfrom ventilatory support after discontinuation of anesthesia. The Heimlichvalve was removed and a 60-cc syringe was attached to the red rubbercatheter to ensure complete resolution of the pneumothorax. The red rubber

catheter was removed and the pursestring suture secured. The animal wasextubated and given postoperative analgesia consisting of buprenorphine(0.1 mg/kg intramuscularly.)

The pig was restarted on a liquid diet 6 hours postoperatively, and aregular diet was resumed 12 hours postoperatively. On postoperative dayCorrespondence to: J. Sackier

Surg Endosc (1998) 12: 10611063

Springer-Verlag New York Inc. 1998


3/99

25 an upper gastrointestinal (UGI) series was completed with barium, andon day 30 the animal underwent esophagoscopy and gastroscopy followedby laparoscopic reduction and repair of the hiatal hernia. The pig wassubsequently killed.

Results

The UGI series was completed with the animal in the left

lateral decubitis position and documented the posterior lo-cation of the hiatal hernia (Fig. 1). Hiatal hernias were docu-mented in all animals with either an UGI series or endos-copy (Fig. 2). According to the radiographic and endoscopicfindings, these hernias were all type II paraesophageal her-nias. A standard Nissen fundoplication with crural closurewas completed on all animals without complications.

Discussion

Previous reports regarding hiatal hernias have focused on

the pathophysiologic and mechanical relationship of hiatalhernias to gastroesophageal reflux. After performing dia-phragmatic crural myotomies in cats through an abdominalincision, Mittal and colleagues [4] performed postoperativeesophageal manometry and pH monitoring. They showed ahigher frequency of gastroesophageal refux after crural my-otomy than with control animals. Laparoscopic repair wasnot attempted in this model. Patterson and Kolyn [8] dem-onstrated esophageal shortening in an opossum model afterthe induction of esophagitis secondary to intraluminal per-fusion of hydrochloric acid. Esophageal shortening was as-sociated with a significant decrease in lower esophagealsphincter pressures. On the basis of their study, the authorssuggest that esophagitis with subsequent esophageal short-ening may contribute to the development of hiatal hernias.No attempt was made at operative intervention and repair ofthis abnormality. We could not find any reports of an animal

model with a hiatal hernia or gastroesophageal reflux ame-nable to laparoscopic repair.

We believe that the one postoperative death of a 160-kgpig was attributable to respiratory compromise secondary toa fractured rib that occurred intraoperatively. The fracturedrib probably impaired normal ventilation throughout theanimals enormous thoracic cavity, resulting in respiratoryfailure. On the basis of this complication, we use 40 to50-kg pigs with careful application of the rib spreader in-

traoperatively. Since making this change, we have had nosubsequent morbidity or mortality.

The barium study seen in Fig. 1 details the posteriorlocation of the hiatal hernia. We did not observe grossesophageal reflux (GER) on barium examination. AlthoughGER was not grossly documented on the UGI series,we plan to investigate potential reflux through pH monitor-ing and esophageal manometry in subsequent trials. Withthe addition of physiologic data from future trials, we hopeto correlate the anatomic role of the crura, phrenoesopha-geal membrane, and the fundoplication to GER in ourmodel.

To our knowledge, this is the first animal model thatallows transabdominal surgical repair of a hiatal hernia viaa laparoscopic approach. In addition to the anatomic accu-racy of the hiatal hernia, our model may provide a physi-ologic paradigm of gastroesophageal reflux disease includ-ing esophagitis and dysplasia. We suggest that the describedanimal model may be used to address the following sub-

jects: (1) the selective use of antireflux procedures withparaesophageal hernias [5], (2) the natural history of low-grade esophageal dysplasia after medical therapy or an an-tireflux procedure [2], and (3) the anatomic relationship ofthe esophagogastric junction in a hiatal hernia and its patho-physiologic association with gastroesophageal reflux dis-ease [3]. Regardless of the outcome of future protocols withpH monitoring and esophageal manometry, this model willcontinue to provide excellent anatomic representation fortraining purposes.

Fig. 1. Under flouroscopy, the posteriorly displaced hernia was visualized (black arrow). The pig was positioned in the left lateral decubitus position.

Fig. 2. With endoscopy, the hiatal hernia was clearly seen under direct vision (white arrow).

1062


4/99

Acknowledgments. The authors thank Drs. Michael Salem and Paul Lin fortheir editorial help with this manuscript and Ethicon Endo-Surgery staff fortheir generous support.

References

1. Barlow AP, Hinder RA, Demeester TR (1989) Principles of 24 hour pHmonitoring and its clinical applications. Gastroenterology 98: A27A29

2. Edwards MJ, et al. (1996) The rationale for esophagectomy as theoptimal therapy for Barretts esophagus with high-grade dysplasia. AnnSurg 223: 585591

3. Mittal RK (1993) Hiatal hernia and gastroesophageal reflux: anotherattempt to resolve the controversy. Gastroenterol (1995) 105: 941943

4. Mittal RK, et al. (1993) Effect of crural myotomy on the incidence and

mechanism of gastroesophageal reflux in cats. Gastroenterol 5: 740747

5. Myers GA, Harms BA, Starling JR (1995) Management of paraesopha-geal hernia with a selective approach to antireflux surgery. Am J Surg170: 375380

6. Ott DJ, Glauser SJ, Ledbetter MS, Chen MY, Koufman JA, GelfandDW (1995) Association of hiatal hernia and gastroesophageal reflux:correlation between presence and size of hiatal hernia and 24-hour pHmonitoring of the esophagus. AJR Am J Roentgenol 165: 557559

7. Ott DJ, Ledbetter MS, Chen MY, Koufman JA, Gelfand DW (1996)Correlation of lower esophageal mucosal ring and 24-hour pH moni-toring of the esophagus. Am J Gastroenterol 91: 6164

8. Patterson WG, Kolyn DM (1994) Esophageal shortening induced byshort-term intraluminal acid perfusion in opossum: a cause for hiatalhernia? Gastroenterol 107: 17361740

1063


5/99

Laparoscopic cholecystectomy under epidural anesthesia in patientswith chronic respiratory disease

K. G. Pursnani,1 Y. Bazza,1 M. Calleja,2 M. M. Mughal1

1 Department of Surgery, Chorley and South Ribble District General Hospital, Preston Road, Chorley, Lancashire PR7 1PP, England2 Department of Anaesthesia, Chorley and South Ribble District General Hospital, Preston Road, Chorley, Lancashire PR7 1PP, England

Received: 11 July 1997/Accepted: 28 October 1997

AbstractBackground: Laparoscopic cholecystectomy (LC) has be-come firmly established as a procedure of choice for gall-stone disease. The procedure usually necessitates generalanaesthesia and endotracheal intubation to prevent aspira-tion and respiratory embarrassment secondary to the induc-tion of pneumoperitoneum. There is a paucity of data in theliterature on the procedure being performed under regional(epidural) anaesthesia, especially in patients with coexistingpulmonary disease and pregnancy, who are deemed highrisk for general anaesthesia. We report our preliminary ex-perience with LC using epidural anaesthesia in patients withchronic obstructive pulmonary disease (COPD).

Methods: We performed LC in six patients (one man andfive women), with a median age of 56 years (range, 3874),under epidural anaesthesia over an 8-month period. All pa-tients were ASA grade III/IV and the mean FEV1/FVC was0.52 (range, 0.40.68), due to chronic asthma (two cases)and COPD (four cases). They were admitted a day prior tosurgery for pulmonary function tests, nebulisers, and chestphysiotherapy. An epidural catheter was introduced at T10/11 intervertebral space, and a bolus of 0.5% Bupivacainewas administered. Depending on the patients pain thresh-

old and the segmental level of analgesia achieved, incre-mental doses of 2 ml of 0.5% Bupivacaine along with bo-luses of intravenous 100 mcg Alfentanil was given to eachpatient. The patients were breathing spontaneously. No na-sogastric tube was inserted, and a low-pressure (10 mmHg)pneumoperitoneum was created. LC was performed accord-ing to the standard technique.

Results: All the patients tolerated the procedure well andmade an uneventful postoperative recovery. Median oper-ating time was 50 min; average length of hospital stay was2.5 days (range, 24). The epidural catheter was removedthe morning after the operation. Only one patient required

postoperative opioid analgesia. Two patients complained ofpersistent shoulder tip pain during surgery and required in-

traoperative analgesia (Alfentanil). There was no change inthe patients cardiorespiratory status, including pO

2and

pCO2, and no complications occurred either intra- or post-operatively.Conclusions: LC can be performed safely under epiduralanaesthesia in patients with severe COPD. Intraoperativeshoulder tip or abdominal pain does not seem to be a majordeterrent and can be effectively controlled with small dosesof opioid analgesia.

Key words: Laparoscopic cholecystectomy Epidural an-aesthesia Chronic respiratory disease

For decades, the management of symptomatic cholelithiasisin high surgical risk patients has remained contentious.Since its advent in 1988, laparoscopic cholecystectomy hasbecome firmly established as a procedure of choice in themanagement of symptomatic cholelithiasis [7, 14, 15]. Theprocedure usually necessitates general anaesthesia and en-dotracheal intubation to prevent aspiration and respiratory

embarrassment secondary to the induction of pneumoperi-toneum. There have been several case reports of successfullaparoscopic cholecystectomy performed under epidural an-aesthesia in pregnant patients during the 3rd trimester [4, 5,17] and patients with cystic fibrosis [6, 11], who are deemedhigh risk for general anaesthesia. However, little has beenreported about the possibility of performing the procedureunder regional (epidural) anaesthesia in patients with sig-nificant pulmonary disease.

Cholecystitis and cholelithiasis in patients with chronicobstructive pulmonary disease (COPD) pose several man-agement problems for the surgeon. Because of the high riskassociated with the induction of anaesthesia in patientswhose pulmonary status is compromised, surgery is some-times delayed or avoided. It is generally agreed that thecondition is best managed conservatively and that surgicalintervention should be reserved for patients who fail to re-Correspondence to: M. M. Mughal

Surg Endosc (1998) 12: 10821084



6/99

spond or develop complications. Laparoscopic cholecystec-tomy (LC) usually necessitates general anaesthesia and en-dotracheal intubation to prevent aspiration and respiratoryembarrassment secondary to the induction of pneumoperi-toneum. Furthermore, in patients with COPD, CO

2pneu-

moperitoneum could have detrimental effects secondary tosplinting of the diaphragm and systemic CO

2absorption.

With the advent of LC and anaesthetic techniques such as

epidural blockage, we have another option that may be safefor many of these patients. We report our preliminary ex-perience with laparoscopic cholecystectomy using epiduralanaesthesia in patients with COPD.

Patients and methods

Six patients (one male, five female), with a median age of 56 years (range,3874), underwent laparoscopic cholecystectomy under epidural anaesthe-sia over an 8-month period in our institution. All patients were AmericanSociety of Anaesthesiologist (ASA) grade III/IV due to chronic asthma(two cases) and COPD (four cases). Spirometric studies (performed with arespiradyne pulmonary function monitor) showed a mean tidal volume of

300 ml, forced vital capacity (FVC) of 1.94 L (50% of predicted value forage, weight, and height) and peak expiratory flow rate (PEFR) of 146 ml/s.Forced expiratory volume in 1 s (FEV1) was 1.02 L (30% of predictedvalue), and FEV1/FVC 100 was 52% (range, 4068%). All patients wereadmitted a day prior to surgery for pulmonary function tests, nebulisers,and chest physiotherapy.

Epidural anaesthesia

The patients were premedicated with 10 mg diazepam 1 h before theprocedure. A 20-gauge epidural catheter via a Tuohy-Huber needle wasintroduced at T10/11 intervertebral space; the tip of the catheter was ad-vanced 3 cm cephalad beyond the tip of the needle. After a test dose of 3ml, an 8 ml bolus of 0.5% Bupivacaine was injected. Depending on the

segmental level of anaesthesia achieved, incremental doses of 2 ml of 0.5%Bupivacaine were administered to reach the desired segmental block. Inmost cases a block up to T4/5 was achieved, as determined by temperaturesensation using ethyl chloride spray. Depending on the patients painthreshold and the amount of shoulder tip pain they experienced, boluses ofintravenous 100 mcg Alfentanil were given to each patient during laparo-scopic cholecystectomy. Along with measuring heart rate and arterial pres-sure, the monitoring also included ECG, pulse oximetry, and expired cap-

nography.

Laparoscopic cholecystectomy

Laparoscopic cholecystectomy was performed according to the standardtechnique [7, 14, 15]. Sequential compression device stockings were used

in all patients, and 1.2 g Augmentin (coamoxiclav/clavulinic acid) wasadministered intravenously intraoperatively. The patients were breathingspontaneously, no nasogastric tube was inserted, and a low-pressure (10mmHg) pneumoperitoneum using CO2 was created. A 10-mm trocar wasinserted via the umbilical port to accommodate the laparoscope and visua-lise the peritoneal surface of the abdominal cavity. Trocars were placed inthe anterior axillary line (5 mm), midclavicular line (5 mm), and midepi-

gastrium (10 mm) just beneath the costal margin. The gallbladder wasgrasped through the 5-mm ports. Dissection, clip application, and electro-cauterisation were performed through the 10-mm epigastric port. The gall-bladder was dissected free from the liver bed and the neck deliveredthrough the epigastric port. In some cases, the gallbladder was decom-pressed using a small suction instrument and removed intact from theabdomen.

Results

All the patients tolerated the procedure well and made anuneventful postoperative recovery. The median operating

time was 40 min (range, 3060), and the average length ofhospital stay was 2.5 days (range, 24). The epidural cath-eter was removed the morning after the operation. Only onepatient required postoperative opioid analgesia. Two pa-

tients complained of persistent shoulder tip pain during sur-gery and thus required intraoperative analgesia (Alfentanil).Two patients elected to view the procedure, thereby pro-ducing an enhanced sham surgical response. Both these pa-tients were communicative throughout the procedure. Theydid not express any distress or discomfort. They made anextremely rapid postoperative recovery and were dischargedhome the following day. They gave a positive response ondirect questioning whether they would have a similar pro-cedure done again under the same conditions. There was nochange in the patients cardiorespiratory status includingSpO

2(oxygen saturation) and E

tCO

2(end-tidal CO

2), and

no complications occurred either intra- or postoperatively

(Table 1).

Discussion

In addition to changes to mucociliary transport associatedwith anaesthetic agents, abdominal surgeryparticularlyupper abdominal surgeryis associated in normal individu-als with adverse effects on respiratory mechanics such asfunctional residual capacity (FRC), vital capacity (VC),tidal volume (TV), and closing volume [8, 13]. Becausemucociliary clearance is an important pulmonary defense

mechanism against infection, general anaesthesia using in-halational or intravenous agents may be deleterious to thepatient with COPD undergoing surgical procedure. Further-more, it has been shown that patients with COPD are at riskof developing pulmonary complications after upper abdomi-nal surgery [1, 9]; therefore, these patients may benefit fromlaparoscopic surgery. The goal of anaesthesia managementin these patients should include avoidance of anaestheticsthat depress mucociliary transport, provision of postopera-tive pain relief adequate to prevent deterioration of respira-tory mechanics, and ambulation as early as possible. Epi-dural anaesthesia fulfills all of the above criteria and aids inthe quick and uneventful postoperative recovery of thesepatients.

Langenberg et al. [12] evaluated the feasibility of per-forming LC under locoregional anaesthesia in 25 patientswithout any evidence of respiratory disease. The procedure

Table 1. Cardiorespiratory function after LC performed under epiduralanaesthesia in six patientsa

Mean Preop Intraop Postop

FEV1 (L) 1.02 (0.921.28) 1.12 (0.911.36)FVC (L) 1.94 (1.882.12) 1.98 (1.872.10)PEFR (ml/s) 146 (122180) 142 (128188)EtCO2 (%) 4.5 (3.85.4) 4.8 (4.15.5) 4.6 (4.15.2)SpO2 (%) 86 (8292) 90 (8894) 87 (8594)

Heart rate(beats/min) 92 (84102) 100 (94110) 94 (90102)

Blood pressure(mmHg) 122/80 110/76 118/84

a All values are mean (range), except blood pressure, where only medianvalues are given.

1083


7/99

was successful in 20 patients (80%) using epidural anaes-thesia, allowing satisfactory surgical conditions and rapidpostoperative recovery. In another study [10], 43 of 45 pa-tients (95%) (without COPD) underwent successful LC un-der epidural anaesthesia and intravenous propofol sedation.They also reported excellent operating conditions and ex-ceptionally pleasant postoperative recovery. The presentstudy is the only report of LC being performed successfully

under epidural anaesthesia in patients with COPD.The pneumoperitoneum should progress slowly, and a

low-pressure (


8/99

Case report

Laparoscopic treatment of duodenal carcinoid tumor

Wedge resection of the duodenal bulb under endoscopic control

T. Toyonaga,1 K. Nakamura,1 Y. Araki,2 H. Shimura,1 M. Tanaka1

1 First Department of Surgery, Kyushu University Faculty of Medicine, 3-1-1 Maidashi, Higashi-ku, Fukuoka, 812-8582 Japan2 Third Department of Internal Medicine, Kyushu University Faculty of Medicine, 3-1-1 Maidashi, Higashi-ku, Fukuoka, 812-8582 Japan

Received: 27 January 1997/Accepted: 4 December 1997

Abstract. A 46-year-old man with epigastralgia and slightelevation of urinary 5-hydroxyindole acetic acid (5HIAA)was found to have a well-demarcated carcinoid tumor in theduodenal bulb. The tumor measured 8 mm in size, andshowed submucosal involvement but no metastasis to theliver and regional lymph nodes. After laparoscopic expo-sure and lifting of the duodenal wall around the tumor,wedge resection of the duodenal bulb including the tumorwas performed successfully with a laparoscopic endostaplerunder direct endoscopic control. The postoperative course

of the patient was uneventful. Laparoscopic wedge resec-tion of the duodenum would be an appropriate minimallyinvasive treatment for selected duodenal neoplasms withspecial preoperative assessments and intraoperative consid-erations.

Key words: Laparoscopic surgery Intraoperative endos-copy Duodenal carcinoid Endoscopic ultrasonogra-phy

A gastroduodenal carcinoid tumor is characterized by slowgrowth and low metastatic potential to the liver or lymphnodes until the late stage of the disease [4]. A conventionalmethod of treatment for the duodenal carcinoid has beensurgical excision or endoscopic mucosal resection [3]. Theadvent of laparoscopic surgery has opened a new pathwayfor treating benign and malignant diseases of the gastroin-testinal tract [1]. We report a case of a duodenal carcinoidtumor treated by laparoscopic wedge resection of the duo-denal bulb with the aid of intraoperative endoscopy.

Case report

A 46-year-old Japanese man suffering from epigastralgia after meals for 2months was admitted to the Kyushu University Hospital on June 6, 1996.

Physical examination showed no remarkable findings except for mild epi-gastric tenderness. Laboratory studies showed an elevated level of urinary

5HIAA of 8.3 mg/day (1.06.0), but normal serum concentrations of se-rotonin (0.16 g/ml, normal < 0.35) and 5HIAA (4.2 ng/ml, normal < 6.1).

Duodenal endoscopy and radiology demonstrated a submucosal tumor,Correspondence to: T. Toyonaga

Fig. 1. Upper gastrointestinal series showing a well-demarcated tumorlocated in the duodenal bulb (arrow).



9/99

8 mm in diameter, located in the lesser curvature of the duodenal bulb (Fig.1). Endoscopic biopsy revealed a carcinoid by histology. Endoscopic ul-trasonography demonstrated a hypoechoic tumor in the submucosal layerof the duodenum and no metastasis to periduodenal lymph nodes (Fig. 2).

Computed tomography and ultrasonography of the abdomen revealed nometastasis to the liver or para-aortic lymph nodes. Endoscopic mucosalresection was abandoned because the tumor was located too close to thepylorus.

Under general anesthesia the patient was put in the supine position.After creation of pneumoperitoneum via a Hasson trocar placed by openlaparotomy at a subumbilical region, a 10-mm trocar was inserted on thepararectal line at the right lower quadrant and a 5-mm trocar at the rightupper quadrant just below the costal margin. At laparoscopy, the stomachand duodenum showed no deformity or adhesion. Intraoperative duode-noscopy confirmed the site of the tumor on the posterior wall of theduodenal bulb just distal to the pylorus. The duodenal bulb was mobilizedby devascularizing the lesser curvature from the pylorus to the duodenaldescending portion.

While precisely locating the tumor by duodenoscopy, the laparoscopicsurgeon placed two stitches through all layers of the duodenum longitudi-

nally 5 mm apart from the tumor edges, and lifted up the duodenal wallincluding the tumor (Fig. 3). Then wedge resection of the duodenum in-cluding the tumor with proper margins was performed by the use of a60-mm EndoGIA. Before firing the stapler with its arms closed, wechecked its correct placement by endoscopy. The duodenum was checked

for leaks, and the specimen was retrieved through the right lower 10-mmtrocar. Although the oral side of the staple line was close to the pylorus, nodeformity of the pyloric ring was observed by duodenoscopy. The stapleline was reinforced with seromuscular sutures and covered with the omen-

tum. The patient tolerated the whole procedure, which took 4 h and 20 min.The resected specimen showed a smooth-surfaced whitish submucosal

tumor, which was hard in consistency and 9 8 mm in size (Fig. 4). Ahistologic examination revealed a well-demarcated carcinoid tumor locatedin the submucosal layer. The tumor was composed of small cells withuniform, round, or oval nuclei arranged in a trabecullar and microglandularpattern (Fig. 5).

Postoperative recovery of the patient was uneventful. He was able towalk the next day, and his postoperative urinary 5HIAA was within thenormal range.

Discussion

Therapeutic options for a duodenal carcinoid tumor includesurgical resection and endoscopic excision [3]. Althoughmost pathologists consider all extra-appendiceal carcinoidsas potentially malignant, solitary tumors smaller than 1 cmin diameter and confined to the submucosal layer rarely

Fig. 2. Endoscopic ultrasonogram demonstrating the tumor involving the mucosa and the submucosa (arrow).

Fig. 3. Intraoperative view. The duodenal wall including the tumor is exposed, lifted up, and grasped to simulate and confirm complete excision.

Fig. 4. Resected specimen. A well-demarcated tumor with sufficient surgical margins.

Fig. 5. Microphotograph of the carcinoid tumor. The tumor involves the mucosal and submucosal layers of the duodenum (hematoxylin-eosin stain, originalmagnification 2.5). Tumor cells with round or oval nuclei proliferate in a trabecullar and microglandular pattern (inset, original magnification 100).

1086


10/99

have metastasis [2]. Therefore, as shown in this case, en-doscopic ultrasonography is of great value in determiningthe depth of the tumor involvement and in demonstratingthe presence or absence of metastasis to the adjacent lymphnodes [3]. Endoscopic mucosal resection should be at-tempted if the lesion is localized within the submucosa andunaccompanied by lymph node metastasis.

Endoscopic mucosal resection, widely performed in Ja-

pan for excising small gastric tumors, has some limitationswhen applied to duodenal tumors, even when they are con-fined to the mucosa or submucosa [3]. The thin wall andnarrow lumen of the duodenum may lead possibly to higherrisks of perforation and luminal stenosis as well as apparentdifficulties of endoscopic control of the resection procedure.

Considering the location in the duodenal bulb and thesecond portion as seen in most cases [3], duodenal carci-noids may be treated laparoscopically if the tumor is soli-tary, smaller than 1 cm in size, and free of metastasis. Pre-operative assessments by endoscopic ultrasonography andintraoperative endoscopic luminal visualization are essential

as shown in the present case. The latter facilitated the exact

location of the tumor and also ensured the complete resec-tion of the tumor with proper margins. The patient should beinformed about the possibility of conversion to a standardopen procedure in the event of technical difficulties or un-expected findings such as lymph node metastasis.

References

1. Burke AP, Sobin LH, Federspiel BH, Shekitka KM, Helwig EB (1990)Carcinoid tumors of the duodenum: a clinicopathologic study of 99cases. Arch Pathol Lab Med 114: 700704

2. Kaplan EL, Udekwu A (1990) The carcinoid syndromes. In: Friesen SR,Thompson NW (eds) Surgical endocrinology: clinical syndromes, 2nded, Lippincott, Philadelphia, pp 181209

3. Ohgami M, Kumai K, Otani Y, Wakabayashi G, Kubota T, Kitajima M(1994) Laparoscopic wedge resection of the stomach for early gastriccancer using a lesion-lifting method. Dig Surg 11: 6467

4. Yoshikane H, Tsukamoto Y, Niwa Y, Goto H, Hase S, Mizutani K,Nakamura T (1993) Carcinoid tumors of the gastrointestinal tract:evaluation with endoscopic ultrasonography. Gastrointest Endosc

39:375383

1087


11/99

Review article

Active electrode monitoring

How to prevent unintentional thermal injury associated with monopolar electrosurgery

at laparoscopy

T. G. Vancaillie

Department of Endogynecology, Royal Hospital for Women, Barker Street, Randwick NSW 2031, Sydney, Australia

Received: 26 June 1997/Accepted: 10 December 1997

AbstractBackground: In recent years, the use of minimally invasivesurgery (MIS) has expanded to a wide variety of surgicalspecialties. The increased popularity of the procedure, how-ever, has been accompanied by its share of complications,including trocar lacerations and inadvertent thermal injuriesto nontargeted tissues during monopolar electrosurgery.

Methods: A survey on electrosurgical thermal injuries andthree case studies are presented. The new technology of

active electrode monitoring (AEM) is described.Results: AEM eliminates stray currents generated by insu-lation failure and capacitive coupling.Conclusions: To reduce the incidence of injury by monopo-lar electrosurgery at laparoscopy, there is a need for ad-vanced technology, such as AEM. In addition, laparoscopicsurgeons should be encouraged to study the basic conceptsof the biophysics of electrosurgery.

Key words: Laparoscopy Electrosurgery Insulationfailure Capacitive coupling Active electrode moni-toring

In the late 1980s, the development of videolaparoscopy ledto an explosion in the use of minimally invasive surgery(MIS). Beginning with a small number of gynecologicalprocedures, MIS has been applied to a wide array of surgicalspecialties, including gastrointestinal, oncologic, and gen-eral surgery. Survey results indicate that by the year 2000,50% of general surgery procedures and 70% of gynecologyprocedures will be performed via MIS [9].

Laparoscopic surgery is favored by both surgeons andpatients over conventional surgery. Patients usually healfaster and suffer less postoperative pain than with traditional

open surgery, expediting discharge from the hospital andrequiring shorter convalescence. As with open surgery,monopolar electrosurgery is the preferred technique for tis-sue cutting and hemostasis in laparoscopy. Monopolar elec-trosurgery is employed by >85% of surgeons who performlaparoscopic procedures [6].

Monopolar instruments enable the delivery of a signifi-cant level of energy to targeted tissue, accounting for theversatility of the procedure. Surgeons can perform smooth

cuts by using a continuous low-voltage current, fulguratetissue with a damped current, or combine the two functionssimply by varying the current or voltage level delivered tothe tip of the active electrode. These adaptive features ofmonopolar electrosurgery have made it an invaluable tool inboth open and laparoscopic surgery.

Though it is popular, cost-effective, and versatile, thecombination of monopolar electrosurgery and laparoscopycan be dangerous [3, 5, 8]. The reduced field of view in-herent to laparoscopic surgery prevents the surgeon fromdirectly observing any tissue located away from the tip ofthe active electrode. Because of this restricted view, the

surgeon is less likely to detect thermal damage caused bystray energy [5].Electrosurgical burns result from insulation failure or

capacitive coupling. Small defects in the layer of electricalinsulation surrounding the shaft of the active electrode al-low energy to leak from the instrument during surgery.These instrument defects can be hard to detect, even withcareful visual inspection. Repeated handling and passes ofthe electrode shaft through trocars can compromise the in-sulation, as can disinfection and sterilization. Moreover, thehigh voltage associated with certain current modes canstress the insulation over time, making it vulnerable tocracks [4].

Capacitive coupling is electrical current that is estab-lished in tissue or in metal instruments running paralleltobut not directly in contact withthe activated elec-trode. The electromagnetic field around the active electrodeCorrespondence to: T. G. Vancaillie



12/99

created by the alternating current induces electrical energyin any nearby parallel conductor. Capacitive coupling is, atbest, containedit cannot be eliminated [5, 8].

Various other laparoscopic surgical techniques, such asbipolar, laser, and the harmonic scalpel, have been evalu-ated by surgeons in an attempt to circumvent the problem ofelectrosurgical burns during laparoscopic monopolar elec-trosurgery. The clinical efficacy of these techniques, how-

ever, is limited, and they have not been widely adopted. Asurvey conducted at a 1993 conference of the AmericanCollege of Surgeons (ACS), in fact, found that only 12% ofsurgeons who perform laparoscopic surgery use bipolartechniques and only 2% use laser energy.

Surgeon perspectives and case histories

The results of the 1993 ACS conference survey indicatedthat there is a high level of awareness among surgeons ofthe danger of electrosurgical thermal injury to patients.

When questioned, 86% of the 506 responding surgeons ac-knowledged the potential for burns to tissues outside thesurgical field during laparoscopic monopolar electrosur-gery. A number of the surgeons reported firsthand experi-ence with complications resulting from insulation failure orcapacitive coupling, and over half (54%) stated that theyknew of colleagues whose patients had suffered complica-tions [7].

Despite this understanding of the potential for electro-surgical burns from laparoscopic surgery, complicationssuch as direct trocar or needle puncture wounds and instru-ment lacerations are more widely acknowledged in dailypractice by the surgical communityin part, because they

are easier to diagnose and treat when they occur. Inadvertentburn injuries to nontargeted tissues outside the surgeonsview during laparoscopic monopolar electrosurgery, on theother hand, are difficult to diagnose and thus are less wellunderstood.

Symptoms of electrosurgical thermal injury are oftendelayed, making it difficult to determine the etiology of theproblem. The injured area may also be compromised bysecondary infection, making histologic diagnosis complex.In addition, an area of coagulative necrosis may be missedon microscopic examination, or a pathologist may not beaware of the unique histological characteristics of thermal

injury and mistakenly attribute the injury to some othercause, such as trocar puncture or mechanical laceration. Forthese reasons, the prevalence of electrosurgical burns islikely underreported and underestimated by the surgicalcommunity.

Among the consequences of thermal injuries are bowelperforation and peritonitis, which are associated with sig-nificant morbidity and even death. Fecal peritonitis, for ex-ample, has a mortality rate as high as 25% [1].

Recently, we encountered a case involving thermal in-jury to a patient undergoing routine laparoscopic surgery.The patient was a 37-year-old woman who presented withpelvic pain and metrorrhagia. A laparoscopy was performedfor resection of endometriosis. Monopolar electrosurgerywas used to resect the affected area of the rectovaginalseptum. Hemostasis was accomplished by monopolar elec-trodesiccation and fulguration. At the end of the procedure,

a bandlike strip of unintentional thermal damage was notedalong the left side wall that included the inner aspect of theleft ureter, necessitating the placement of a 28-cm 6-Frdouble J stent. Fortunately, the patient did not suffer anylong-term adverse effects from the burn since the injury wasimmediately detected and treated. Not all patients are sofortunate.

Reliable incidence figures on burn injuries during lapa-

roscopic monopolar electrosurgery are difficult to obtaingiven the aforementioned diagnostic challenges. In recentyears, however, an increasing number of case histories haveappeared in the medical literature. The following case his-tory illustrates that complications arising from undetectedburns during monopolar electrosurgery can have serious andlong-term morbidity.

A 38-year-old nurse was seen by a gynecologist for leftlower quadrant pain. The patients surgical history includedwedge resection of the left ovary for endometriosis. Thegynecologist diagnosed pelvic adhesions of the ovary andperformed laparoscopic surgery. Monopolar electrosurgery

was used to cauterize adhesions from the ovary to the pelvicside wall. The power setting of the electrosurgical generatorwas 30 W, and the electrode was activated for 5 sec. Thepatient was discharged from the hospital on the same day ofsurgery, but she was admitted to the emergency room witha low-grade fever and leukocytosis on the 7th postoperativeday.

A CT scan found free air in the abdomen. Exploratorylaparotomy revealed multiple necrotic areas in the distalileum that resembled burns. Several areas of the colonappeared compromised, with one area showing perfora-tion. Peritonitis was localized to the right lower quadrant.Microscopic examination of the small bowel showed focal

full-thickness necrosis. Examination of the large intestinerevealed areas of mucosal ulceration and full-thicknesswall necrosis. During the follow-up surgery, 40 cm of theileum were removed, and a temporary colostomy was per-formed. After the laparotomy, the patient developed awound infection requiring further treatment. She was notsufficiently well to return to normal activities until 6 monthsafter the initial laparoscopic surgery [11].

As this case study suggests, electrosurgical thermal in-jury should be suspected in patients who have undergonemonopolar laparoscopic electrosurgery and who demon-strate symptoms associated with organ perforation or peri-

tonitis. The restricted visual environment in laparoscopyincreases the risk of unseen electrosurgical burns. Electricalinterference on the electrosurgical units video monitor orreduced power at the tip of the electrode are unreliableindicators of electrosurgical burn injury potential, but theyshould arouse the suspicion of the presence of stray cur-rents.

In addition to the potential clinical risks to patients,surgeons performing laparoscopic monopolar electrosur-gery may encounter medicolegal liability. At the 1995 meet-ing of the Society of Laparoendoscopic Surgeons, 13% ofmembers surveyed reported involvement with one or moreactive malpractice cases associated with a laparoscopicelectrosurgical procedure [7].

A malpractice case in 1994 illustrates the liability riskssurgeons face as a result of undetected thermal injury. Inthis case, an Oregon woman underwent laparoscopy in 1994

1010


13/99

for gallbladder removal. One week after the surgery, a lap-arotomy revealed a high-grade stricture of the common he-patic ducta complication that required additional surger-ies for repair and dilation. Following her recovery, the pa-tient sued the surgeon for negligence and received asubstantial damage award. The operating room record indi-cated that significant electrical interference on the videomonitor had hindered completion of the surgery. A witness

for the surgeon testified that these periods of interferencemost likely signaled the existence of stray electric currentsthat produced the burns to the hepatic duct [2].

Physician insurance companies have responded to themedicolegal risk posed by laparoscopic monopolar electro-surgical procedures through rate adjustments and trainingincentives. Some providers of malpractice insurance offertheir members free accredited postgraduate training coursesin electrosurgery and risk management [5].

Active electrode monitoring: a technological advance

in the prevention of electrosurgical burns due tostray currents

The continued reports of clinical and medicolegal problemsdirectly associated with thermal injury during laparoscopicmonopolar electrosurgery reinforces the view that the mostcommon protective measures (e.g., inspection of electrodesof insulation cracks, specialized training for surgical per-sonnel, etc.) have not eliminated the risk of burns to non-targeted tissues during minimally invasive monopolar elec-trosurgery. Even the most able and experienced surgeonswho consistently use strict safety protocols cannot transcendthe immutable conditions of the electrical environment en-

countered during laparoscopic monopolar electrosurgery.An alternative technological solution is necessary to en-

sure patient safety during laparoscopic monopolar electro-surgery. The selection of a particular technology should beevaluated relative to its capability of eliminating inadvertenttissue injury due to stray electrical currents, require a mini-mal amount of training or modification in surgical methods,and offer a cost-effective solution. Moreover, it shouldovercome the current deficiencies in the maintenance andtesting of laparoscopic instruments by ensuring that whensuch instruments do fail, they fail safely.

Active electrode monitoring (AEM) (ElectroScope,

Boulder, CO, USA) meets these criteria. AEM offers theultimate safety in monopolar electrosurgery by combiningadded electrical insulation, conductive shielding, and anelectronic current monitoring system. Stray currents thatmay be released through faulty insulation are absorbed bythe additional electrical insulation and conductive shielding(Fig. 1).

The conductive shielding within the insulation itself be-comes capacitively coupled to the active electrode, insteadof any metal surgical instruments or the patients tissue,eliminating the incidence of tissue burns from capacitivecoupling. The conductive sheath is electrically connected tothe return electrode of the electrosurgical unit, allowing forharmless dissipation of capacitively coupled currents. Ifstray energy levels become sufficiently high to damage non-targeted tissues, the AEM circuit interrupts the flow of en-ergy from the electrosurgical unit and sounds an alarm.

The Emergency Case Research Institute, a nonprofit or-ganization that evaluates medical devices, tested this moni-toring system and concluded that it offers maximum pro-tection against patient injury due to either insulation failureor capacitive coupling [10]. The contribution of AEM to thesafe application of monopolar laparoscopic electrosurgeryhas also been recognized by the American Association ofGynecological Laparoscopists, which has urged surgeons toconsider the use of AEM when performing laparoscopic

monopolar electrosurgery.

Conclusions

Laparoscopic monopolar electrosurgery is a highly versatileand effective tool that is used in a wide variety of surgicalspecialties. Monopolar electrosurgery is by far the superiorand preferred technique for tissue cutting and hemostasis,eclipsing alternative measures. It can, however, place pa-tients at risk for unintended burns to nontargeted tissuesoutside the surgeons view, resulting from stray electricalcurrents associated with insulation failure or capacitive cou-

pling. The clinical and medicolegal risks, combined with theanticipated growth in the number and type of laparoscopicsurgery applications, necessitate a shift in electrosurgicalpractice. This shift should ideally encompass the study ofthe basic concepts in biophysics of electrosurgery and theintroduction of more sophisticated technology.

AEM minimizes the risk of nontargeted tissue burnsassociated with minimally invasive electrosurgery while al-lowing both surgeons and patients to reap the many benefitsof laparoscopic monopolar electrosurgery.

References

1. Berry SM, Ose KJ, Bell RH, Fink AS (1994) Thermal injury of theposterior duodenum during laparoscopic cholecystectomy. Surg En-dosc 8: 197200

2. Golden TR (1993) Laparoscopic cholecystectomy verdict. Trial News 20

Fig. 1. Schematic representation of active electrode monitoring. The activeelectrode is surrounded by three successive layers: a first layer of insula-tion, a conductive sheath, and a second layer of insulation. The conductivesheath captures the current generated by capacitive coupling, which cannotbe avoided, and other stray currents. These stray currents are analyzed bythe active electrode monitor. If the amount or character of the stray currents

exceeds or differs from preset norms, the AEM will shut down the elec-trosurgical generator (ESU).

1011


14/99

3. Grosskinsky CM, Hulka JF (1995) Unipolar electrosurgery in opera-tive laparoscopy. J Reprod Med 40: 549552

4. Luciano AA, Soderstrom RM, Martin DC (1994) Essential principlesof electrosurgery in operative laparoscopy. J Am Assoc Gynecol Lapa-rosc 1: 189195

5. Odell RC (1993) Electrosurgery in laparoscopy. Infert Reprod MedClin North Am 4: 289304

6. Southern Surgeons Club (1991) A prospective analysis of 1518 lapa-roscopic cholecystectomies. New Engl J Med 324: 10731078

7. Tucker RD (1995) Laparoscopic electrosurgical injuries: survey resultsand their implications. Surg Laparosc Endosc 5: 311317

8. Vancaillie TG (1994) Electrosurgery at laparoscopy: guidelines toavoid complications. Gynaecol Endosc 3: 143150

9. Wetter PA (1994) Trends study. Presented at the Annual Meeting ofthe Society of Laparoendoscopic Surgeons, Westin Hotel, Seattle

10. Anonymous (1995) Focus on laparoscopy. Health devices 24: 338

11. Trudy Karl v. Rufus S. Armstrong, M.D. (1993) Fla Jury Verdict Rep14: 4748

1012


15/99

Laparoscopic Collis gastroplasty and Nissen fundoplication

A new technique for the management of esophageal foreshortening

A. B. Johnson, M. Oddsdottir, J. G. Hunter

Department of Surgery, Emory University Hospital, Room H124C, 1364 Clifton Road, N.E., Atlanta, GA 30322, USA

Received: 8 September 1997/Accepted: 17 December 1997

AbstractBackground: The short esophagus increases the difficultyand limits the effectiveness of laparoscopic Nissen fundo-plication. In our experience, 2025% of esophagi judgedby preoperative criteria to be foreshortened will, after dis-section, be insufficiently long to allow 2 cm of esophagus toreside below the diaphragm without inferior distraction (i.e.,tension free). Collis gastroplasty combined with Nissen fun-doplication has become the standard approach for the cre-ation of an intraabdominal neoesophagus and fundic wrap.

Methods: After developing methods of performing totallylaparoscopic stapled gastroplasty in the cadaver lab in 1994,we started applying the technique clinically in 1996. Weperformed 220 laparoscopic antireflux procedures betweenJanuary 1996 and July 1997. Of these 220 patients, 26%were suspected to have esophageal foreshortening based onpreoperative barium studies and/or endoscopy.

Results: After hiatal dissection, nine patients, or 16% ofthose suspected to have esophageal foreshortening and 4%of the entire population, required the laparoscopic Collis-Nissen procedure. There was symptomatic improvement inall patients as assessed by patient-initiated symptom scores.

Conclusions: The management of patients with esophagealforeshortening is a complex problem. We believe that ourtechnique of laparoscopic Collis-Nissen provides an effec-tive means of achieving intraabdominal placement of thefundic wrap while maintaining the benefits of a minimallyinvasive approach.

Key words: Hiatal hernia Paraesophageal hernia Gas-troesophageal junction Esophageal stricture Collisgastroplasty Laparoscopic Nissen fundoplication

The shortened esophagus not only increases the difficultybut also limits the effectiveness of laparoscopic Nissen fun-

doplication. It has long been known to complicate the workof anti-reflux surgery and paraesophageal hernia repair.Esophageal foreshortening is found more frequently in as-sociation with a gastroesophageal (GE) junction that is >5cm above the hiatus on barium swallow, esophageal stric-ture, type III (mixed) paraesophageal hernia, and Barrettsesophagus (with or without stricture) [8]. Even when thesepreoperative findings are noted, it is often difficult topredict which patients will have a truly foreshortenedesophagus, because the esophagus, when adequately mobi-

lized and transposed to the anterior hiatus, is still longenough to allow the GE junction to reside below the hiatuswithout tension. Previous investigators have demonstratedthat only 20% of esophagi believed preoperatively to beforeshortened will prove to be foreshortened intraopera-tively [11].

Collis gastroplasty in combination with complete or par-tial fundoplication has become the standard approach tocreate an antireflux valve in patients with esophageal fore-shortening. Collis originally described the performance ofgastroplasty through a thoracoabdominal incision; however,it is now usually performed with a transthoracic approach

and followed by a partial (Belsey) or complete (Nissen)fundoplication [2]. Traditional teaching has emphasized theneed for extensive mediastinal dissection in order to ad-equately mobilize the esophagus for a tension-free transtho-racic repair [1]. Though it provides adequate exposure, tho-racotomy subjects patients to significant pain and morbid-ity. The desire to avoid inconvenience to the patient led tothe development of abdominal gastroplasty techniques bySteichen and Henderson and Marryatt to manage the short-ened esophagus [4, 10].

Two descriptions have been published of thoracoscopicCollis gastroplasty combined with laparoscopic fundoplica-tion [3, 11]. Because of the additional requirements of tho-racoscopy (e.g., double-lumen anesthesia, additional video-endoscopic equipment, chest preparation), we have devel-oped a laparoscopic approach to esophageal lengthening.Our technique and results are described in this report.Correspondence to: J. G. Hunter

Surg Endosc (1998) 12: 10551060



16/99


17/99


18/99

comes clear that esophageal lengthening is necessary, the10-mm laparoscope port (above and to the left of the um-bilicus) is replaced with a 12-mm trocar to accommodatethe linear cutting stapler. A 48-Fr dilator is placed to cali-brate the width of the gastric tube (Fig. 2). A burn mark isthen placed 3 cm inferior to the angle of His and 1 cm fromthe dilator on the anterior wall of the stomach to indicate theexit point of the anvil of a 21-mm circular stapler. This

procedure will create a 4-cm neoesophagus.A 2-cm vertical mini-laparotomy is made just to the left

and slightly inferior to the xiphoid and divided downthrough the peritoneum with electrocautery. The mini-laparotomy is dilated with a large hemostat. A 2-0 Prolenesuture on a Keith needle is attached to the open hole on theplastic skewer of a 21-mm circular cutting anvil (CLH-21;Ethicon Endosurgery, Cincinnati, OH, USA) and poppedinto the peritoneal cavity through this mini-laparotomy. Theskin is closed with towel clips in order to maintain thepneumoperitoneum.

The anvil is placed in the lesser sac by elevating the well

mobilized gastric fundus. The greater curvature of the stom-ach is held anteriorly with two graspers. The Keith needleand attached anvil are then passed through the posterior wallof the stomach, exiting the anterior wall of the stomach atthe burn mark 1 cm away from the dilator (Fig. 3). The bodyof the 21-mm circular stapler is introduced through the xi-phoid incision, docked with the anvil, and fired to create asealed window through both gastric walls (Fig. 4). The lapa-roscope is shifted to the left subcostal position, and a 30-mmlinear cutting stapler (Ethicon Endosurgery) is insertedthrough the primary trocar near the umbilicus and firedadjacent to the dilator (Fig. 5).

The staple line is oversewn with two running verticalmattress sutures of 2-0 braided nylon, one starting at the GE

junction and one starting at the distal margin of the stapleline on the fundus. This suture reinforcement of the stapleline may not be essential, but it provides additional securityagainst leak or bleeding. The sutures are tied to each otherwhen they meet in the region of the circular staple line.After an appropriate crural closure, a 2-cm floppy Nissenfundoplication is sutured. The fundic staple line lies behindthe esophagus, with its apex becoming the middle point ofthe fundic suture line to the right of the esophagus (Fig. 6).

Clinical experience

Our technique of laparoscopic Collis-Nissen was performedin nine patients between January 1996 and July 1997 (Table1). Large hiatal hernias were present in all patients. Twopatients had intrathoracic stomachs, and both of these pa-tients had esophageal strictures requiring dilation. There hasbeen symptomatic improvement in all patients in whom wehave utilized this technique; however, one patient suffered arecurrence of moderate dysphagia from a distal esophagealstricture present before the operation. This patient has re-sponded to dilation and standard doses of omeprazole, adosage that did not control his heartburn preoperatively. Webelieve that the additional length of this patients esophagusmay be contributing to symptom control, since endoscopic

examination suggests that the fundoplication may havecome apart.

Between January 1996, when we performed the firstCollis gastroplasty, and July 1997, we did a total of 220laparoscopic antireflux procedures. Of this population, 58patients (26%) were suspected to have esophageal fore-shortening prior to surgery. Of these 58, nine patients (16%of those suspected, 4% of entire population) required esoph-ageal lengthening with a laparoscopic Collis gastroplasty.As compared to a group of patients that underwent laparo-scopic fundoplication, operative time was longer and lengthof stay was longer, but there was no additional morbidity.Using a five-point patient-initiated symptom score pre- andpostoperatively (0 no symptoms, 1 rare symptoms, 2

moderate symptoms, 3 severe symptoms, 4 incca-pacitating symptoms), we found no differences in clinical

response between these patients and our control group of300 patients undergoing Nissen fundoplication (Table 2) [6].

Discussion

Laparoscopic Nissen fundoplication has become a routineprocedure; many centers have reported >200 procedures. Inmost centers, Collis gastroplasty is not considered neces-sary. Our development of the Collis gastroplasty techniquestarted in the cadaver lab shortly after we had reached 100

procedures and was not applied until we had reached 400procedures. We believed that this additional procedure wasoccasionally necessary because of the high rate of para-esophageal herniation following laparoscopic Nissen fundo-plication reported by others as well as ourselves (range,38%) [6, 13]. One of the avoidable causes of postoperativeparaesophageal hernia is the need to ensure adequate esoph-ageal length to allow the GE junction to reside in the ab-domen without tension. When we specifically looked for acause of paraesophageal herniation among the 3% of ourpatients that developed this problem, esophageal foreshort-ening contributed in less than a third of patients (1%). Whileit may be enticing to attribute the high frequency of post-operative dysphagia reported during the learning curve oflaparoscopic fundoplication to a short esophagus and para-esophageal herniation, this does not appear to be the case. Inmost cases, persistent postoperative dysphagia occurs be-

Table 1. Patient demographics

Patientno.

Age/sex Cause

O.R.time(min)

Lengthof stay(days) Complications

1 70/F Large HH/stricture 285 6 atelectasis2 35/M HH/stricture 351 3 none3 65/M Para HH/stricture 289 2 none4 83/M Large Para HH 394 4 A-fib

5 48/M Large HH/esophagitis 332 2 none6 67/M Large HH/stricture 210 3 none7 69/F Para HH/stricture 232 2 none

8 68/M Para HH 287 2 none9 57/F HH/stricture 269 3 none

Para, paraesophageal; HH, hiatal hernia

1058


19/99

cause the fundoplication has been misformed. In our expe-rience, most of these patients have had the Rosetti modifi-cation of the Nissen fundoplication [12].

The key to managing patients with esophageal fore-shortening is making an accurate intraoperative determina-tion that the esophagus is truly foreshortened. After dissect-ing 46 cm up into the mediastinum with the GE junction

retracted inferiorly, the esophagus is transposed to the an-terior hiatus and released. If it springs back to the diaphragmor above, it is too short and should be lengthened. Preop-eratively, a short esophagus was predicted in 16% of pa-tients who met liberal preop criteria; this group represents4% of our entire population. In another series, a 14% inci-dence of esophageal foreshortening was predicted by pre-operative criteria, of which 9% (1.2% of the patient popu-lation) required esophageal lengthening with a thoraco-scopic Collis gastroplasty [11].

With our technique, considerable time may be requiredto oversew the gastric staple lines. The staple line of thegastroplasty may not need to be oversewn; however, it adds

additional security in preventing gastric leakage, since theendoscopic linear cutting staples are only 3.5 mm long asopposed to the 4.8-mm staple length generally used on thestomach. A running vertical mattress suture has been mosteffective for oversewing these staple lines.

Two other minimally invasive techniques of performingCollis gastroplasty have been described [3, 11]. In the firsttechnique, the usual subdiaphragmatic and mediastinal dis-section for laparoscopic fundoplication is performed withlaparoscopic visualization, followed by right thoracoscopyand placement of the linear stapler for the gastroplastythrough a second port. This maneuver facilitates the place-

ment of the stapler at the angle of His in the proper orien-tation. Crural closure and fundic wrap are then completedlaparoscopically [11]. In the second technique, the entireprocedure is performed with left thoracoscopic access. Thegastroplasty is created using a noncutting linear stapler, fol-lowed by fundic wrap and reduction below the diaphragm[3]. Although these techniques recapitulate the standardopen surgical techniques, they require single-lung ventila-tion, chest preparation, two video carts, thoracotomy privi-leges, and the increased pain associated with thoracoscopy.These disadvantages can be avoided if laparoscopic Collisgastroplasty is performed instead.

Certainly, other traditional approaches for the perfor-mance of Collis gastroplasty should not be neglected. Al-though most surgeons perform this procedure through a leftthoracotomy, the technique performed by Steichen througha laparotomy yields equivalent results [10]. Although our

preferred access is laparoscopic in patients without previousoperation, in one patient we elected to perform Collis gas-troplasty through a laparotomy because the patient had pre-viously undergone open fundoplication. It is generally mostexpedient to perform redo surgery through a laparotomy ifthe first operation was performed through a laparotomy.Another very acceptable approach is to convert a laparo-

scopic Nissen to an open Collis-Nissen when esophagealforeshortening is discovered intraoperatively. It is more im-portant that the operation be performed correctly than thatlaparoscopic access be maintained. Because we developedthis technique in the cadaver lab and practiced multipletimes in both the pig and human cadaver, it was not neces-sary for us to convert any of our cases.

The surgical management of patients with esophagealforeshortening is a complex problem. In patients with ex-tremely poor esophageal motility and a tight stricture, seg-mental esophagectomy is often the optimal therapy [8].When esophageal motility is poor and the esophagus is shortbut there is no stricture, the Collis gastroplasty may be

combined with a posterior partial fundoplication. Less de-finitive approaches, such as gastropexy with crural closureand mediastinal positioning of the fundoplication, result inunacceptably high recurrence rates and postoperative dis-comfort. Most authors agree that in order to obtain the bestresults, the fundoplication following gastroplasty must beplaced below the diaphragm and under no tension [9]. Webelieve that our technique of laparoscopic Collis-Nissenprovides an effective means of managing patients with aforeshortened esophagus while also offering the advantagesof a minimally invasive approach.

References

1. Adler RH (1990) Collis gastroplasty: origin and evolution. Ann ThoracSurg 50: 839842

2. Collis JL (1957) An operation for hiatus hernia with short oesophagus.Thorax 12: 181188

3. Demos NJ, Kulkarni VA, Arago A (1994) Video-assisted transthoracichiatal hernioplasty using stapled, uncut gastroplasty and fundoplica-tion. Surg Rounds xx: 427436

4. Henderson RD, Marryatt GV (1985) Transabdominal total fundopli-cation gastroplasty to control reflux. A preliminary report. Can J Surg28: 127129

5. Hunter JG, Champion JK (1996) Laparoscopic Nissen fundoplication.In: Endosurgery. Churchill Livingstone, New York & London

6. Hunter JG, Trus TL, Branum GD, Waring JP, Wood WC (1996) Aphysiologic approach to laparoscopic fundoplication for gastroesopha-geal reflux disease. Ann Surg 223: 673687

Table 2. Typical symptoms pre- and postoperatively of Collis-Nissen patients (n 9) and population undergoing laparoscopic fundoplication (n 300)1 year following operation

Preop Postop

Collis-Nissenn 9)

Nissen(n 253)

Collis-Nissen(n 9)

Nissen(n 253)

SSSa 01 2 34 01 2 34 01 2 34 01 2 34Heartburn 56% 0 44% 21% 22% 70% 89% 11% 0 92% 4% 4%

Regurgitation 49% 22% 33% 56% 13% 31% 100% 0 0 95% 3% 2%Dysphagia 78% 11% 11% 62% 13% 31% 89% 11% 0 88% 7% 5%

a Symptom Severity Score (SSS): 0 no symptoms, 1 rare symptoms, 2 moderate symptoms, 3 severe symptoms, 4 incapacitating symptoms

1059


20/99

7. Oddsdottir M, Laycock W, Champion K, Hunter JG (1995) Laparo-scopic esophageal lengthening procedure [abstract]. Surg Endosc 9:621

8. Pearson FG (1995) Peptic esophagitis, stricture, and short esophagus.In: Esophageal surgery. Churchill Livingstone, New York

9. Pearson FG, Todd TR (1987) Gastroplasty and fundoplication forcomplex reflux problems: long-term results. Ann Surg 206: 473481

10. Steichen FM (1986) Abdominal approach to the Collis gastroplastyand Nissen fundoplication. Surg Gynecol Obstet 162: 273275

11. Swanstrom LL, Marcus DR, Galloway GQ (1996) Laparoscopic Collisgastroplasty is the treatment of choice for the shortened esophagus.Am J Surg 171: 477481

12. Trus TL, Cornwell M, Waring JP, Galloway K, Hunter JG (1998)Patterns of failure and results of redo fundoplication. Surg Endosc(in press)

13. Watson DI, Jamieson GG, Devitt PG, Mitchell PC, Game PA (1995)

Paraesophageal hiatus hernia: an important complication of laparo-scopic Nissen fundoplication. Br J Surg 82: 521523

1060


21/99

Laparoscopic surgery for abdominal aortic aneurysms

Technical elements of the procedure and a preliminary report of the first 22 patients

J. K. Edoga, K. Asgarian, D. Singh, K. V. James, J. Romanelli, S. Merchant, D. Romano, B. Joostema, J. Street

Departments of Surgery, Anesthesia, and Nursing, Morristown Memorial Hospital, Morristown, NJ 07960, USA

Received: 23 June 1997/Accepted: 11 December 1997

AbstractBackground: Laparoscopic surgery for infrarenal aortic an-eurysms is based on the principle of retroperitoneal exclu-sion of the aneurysm sac with aortofemoral or aortoiliacbypass.

Methods: Of 22 patients who met the selection criteria, 20successfully underwent laparoscopic aortic surgery at Mor-ristown Memorial Hospital between February and October1997. Technical elements and steps of this operation aredescribed and illustrated.

Results: Within 30 days of surgery, 2 patients died and 9

had various major and minor perioperative complications.As a group, the laparoscopic patients had less postoperativepain, needed fewer hours of ventilator support, had shorterintensive care unit (ICU) and hospital lengths of stay, andresumed diet and normal activity earlier than the historicalnorms for patients undergoing transabdominal or retroperi-toneal aortic resections at the same institution.Conclusions: These early observations suggest that the lap-aroscopic treatment of infrarenal abdominal aneurysms mayhave several significant potential benefits. Long-term re-sults and randomized prospective studies with patientsmatched by risk stratification will be needed to confirm

these impressions.

Key words: Laparoscopic Retroperitoneal Transab-dominal Minimally invasive potential benefits

The posterolateral retroperitoneal approach to surgery onthe abdominal aorta, as initially described by Williams et al.[12] and more recently by Darling and his colleagues [3],has been associated with several physiologic advantageswhen compared with the traditional transperitoneal proce-

dure [2]. In 1995, Gregorio Sicard [10] published the first

randomized prospective trial comparing the retroperitonealwith the transabdominal approach to the aorta for routineinfrarenal aortic reconstruction. His findings demonstratedthat the retroperitoneal approach was associated with

Fewer postoperative complications Shorter ICU stay Shorter hospital stay Decreased cost.

However, long-term incisional pain was clearly greater

in the retroperitoneal group.In our small but rapidly growing experience, the lapa-

roscopic technique for the treatment of abdominal aorticaneurysms appears to contribute the proven benefits ofminimally invasive surgery while adding none of the currentdrawbacks and limitations of the endovascular approach [6,7, 11].

Our laparoscopic procedure is based on principles ofretroperitoneal aneurysm exclusion and bypass described byShah and his associates in 1991 [9] and their long-termresults reported by Resnikoff et al. in 1996 [8]. However,performance of his operation through laparoscopic ports

and/or small videoscopically guided incisions has the po-tential benefits of

Decreased postoperative pain Reduced postoperative ventilator dependence Shorter ICU stay Reduced hospital length of stay Significantly reduced total cost Safe early resumption of normal activity [1].

Our experience with the laparoscopic approach has alsodemonstrated that the minimally invasive nature of the op-eration does not appear to positively influence the outcomein high-risk patients and that, as with the traditional openprocedures, the risk of perioperative death is still largelydetermined by the number and severity of comorbid condi-tions [4, 5].

Correspondence to: John K. Edoga, Northwest Surgery P.A., 95 MadisonAvenue, Morristown, NJ 07960, USA

Surg Endosc (1998) 12: 10641072



22/99


23/99

arteries, which by our present criteria for patient selection could con-stitute contraindications to this surgical approach.

In keeping with our practice with patients undergoing open operationsfor abdominal aortic aneurysms, we prescribed mechanical as well as an-tibiotic bowel preparation for all patients on the day before surgery. Allpatients were admitted in a fasting state on the morning of surgery except

patients 4, 7, and 20, for whose cases we felt the comorbid conditionswarranted preoperative hospitalization for extra preparation. Prophylacticparenteral antibiotics were given preoperatively in all cases. We routinelyused intermittent compression boots for DVT prophylaxis.

Anesthesia protocol. General endotracheal anesthesia is induced with thepatient in supine position. Selection of the most appropriate anestheticagent is left to the discretion of the anesthesiologist. We use perioperativeepidural analgesia only in patients with marginal pulmonary reserve toreduce the need for narcotic pain control. As in open surgery, we enhancediuresis during surgery with mannitol (0.5 gm/kg bolus) given intrave-nously once the retroperitoneal access is gained and followed by a con-tinuous infusion of mannitol at 5 gm/h for the remainder of the operation.A low-dose (3 to 5 g/kg/min) infusion of dopamine is started in some

cases and continued for 12 to 24 h after surgery if there is preoperativeevidence of renal insufficiency such as an elevated serum creatinine level,

or if the urine output decreases significantly after application of the aorticcross-clamp. Radial artery and pulmonary artery catheters are placed in allpatients before induction of general anesthesia in keeping with our practice

for traditional open operations. In addition to continuous systemic andpulmonary arterial pressure readouts, frequent cardiac output and indexmeasurements are recorded. Arterial blood gasses, hemoglobin levels,platelet counts, and plasma potassium determinations are checked regularlyand corrective actions taken as indicated. Following administration of hep-arin, activated clotting time (ACT) measurements are performed every halfhour, and additional heparin boluses are given as needed to maintain the

ACT at twice the baseline value. Despite relatively prolonged cross-clamptimes, in our early experience, the need for pressors and afterload reducersto maintain cardiac indices after aortic cross-clamping appears to havebeen dictated by the presence and severity of coronary artery disease andleft ventricular dysfunction. In our experience, only 2 of the 20 patients (#4and #9) have required this pharmacologic support of cardiac function afterapplication of the aortic cross-clamp. Routine use of the cell-saver to save

and wash shed blood for reinfusion has significantly reduced the need forblood bank transfusions.

Technical elements of the procedure

Patient positioning. The patients torso is lifted and rotated so that the leftshoulder is elevated about 60 degrees from the horizontal. The left upperextremity is supported on an arm board on the right side of the table. Thepelvis is left in as neutral a position as possible to facilitate surgical accessto both femoral regions. (In very obese patients, the operating table has to

Figs. 1. AD Scatograms showing outcome comparisons between a single-surgeon series of open and laparoscopic aneurysmectomy patients. (),Laparoscopic AAAs; (), open AAAs.

1066


24/99

be banked severely to the left so the panniculus can be rolled away from theright side when the right groin incision is made.) The table is then extendedapproximately 30 degrees before the bean bag is molded and placed onsuction to maintain the patient position. The left thigh is elevated on eitherpillows or a folded bean bag to relax the left psoas muscle. Generouspadding is used to alleviate pressure in areas of bony prominences and

along the edges of the actuated bean bag (Fig. 2).

Initial retroperitoneal access. Initial retroperitoneal access is gainedthrough a 1.5-cm muscle-splitting incision in the left flank. This incision islocated just posterior to the anterior axillary line halfway between the ribcage margin and the iliac spine. A retroperitoneal tunnel is then created bydigital dissection and directed to pass posterior to the inferior pole of theleft kidney. A kidney-shaped preperitoneal distension balloon OMS-PDBS2 (Origin Medsystems, Menlo Park, California) is then placed com-pletely within the tunnel and inflated under visual inspection by means ofa zero-degree laparoscope inserted through the shaft of the distension bal-loon. Inflation is continued until a sufficient retroperitoneal space has beencreated and familiar structures (e.g., the psoas muscle with its broad ten-don, the genitofemoral nerve along its medial border, the left iliac artery,and the crossing left ureter as well as the left gonadal vein) are clearlyvisible through the inflated balloon. The dissecting balloon is then re-

moved, and the freshly created space is accessed with a blunt-tipped bal-

loon trocar (OMS-T10BT or OMS-T10BTS (Origin Medsystems), depend-ing on the thickness of the abdominal wall. The retroperitoneum is insuf-flated with CO2 up to a maximum cavity pressure of 15 mm of mercury.The CO2 flow rate is kept at a constant maximum to compensate foranticipated gas loss during insertion and manipulation of laparoscopic in-struments.

The insertion of additional trocars through the sites shown in Fig. 2 iscarried out under videoscopic observation, the main operating ports beinginserted first, then followed, in due course, by the insertion of the leftkidney retraction port once the left kidney has been mobilized. The aorticcross-clamp port is created much later during the gasless phase of theoperation.

Vascular dissection. The vascular dissection is begun at the level of the

left common iliac artery. The common iliac artery is mobilized by circum-ferential dissection, and if an aneurysm of the common iliac is present, theiliac bifurcation or the distal extent of the iliac aneurysm must be exposed.(Because stapling is used to exclude the iliac arteries, the iliac vein must bedissected free of the posteromedial arterial wall (Fig. 3) to avoid inadver-

tent inclusion of the vein in the staple line.) The left kidney is then mo-bilized and held rotated in a cephalad and counterclockwise direction withan extra-hand balloon retractor (Origin Medsystems) to expose the perire-nal portion of the abdominal aorta. This rotated position of the left kidneyheld with an extra-hand balloon retractor can in some cases be maintainedby attaching the handle of the balloon retractor to a laparoscopic utility belt(SMC Surg-Med Devices Inc., Morristown, NJ). The aortic neck is thendissected circumferentially by clipping and dividing the lumbar tributariesof the left renal vein, the surrounding lymphatics, and the lumbar arteriesand veins. The left renal artery must be identified in its new locationanterior to the aorta because of the rotation of the left kidney. The proximal2 cm of the aneurysm is similarly freed anteriorly and posteriorly. Care istaken to minimize the manipulation of the aneurysm sac during this dis-section to avoid potential distal embolization of thrombus or atheromata.

We have used two approaches to expose the right iliac artery. If thereis aneurysmal involvement of the right common iliac artery, we feel it issafer to use a counter incision in the right lower abdominal quadrant for a

retroperitoneal exposure of the right common iliac artery and its branches.However, if the right iliac artery is normal, it can be safely exposed cir-cumferentially simply by continuation of the dissection already begun inthe left retroperitoneal space. The aortic bifurcation is identified, and thecourse of the right iliac artery is followed. We have been able to expose the

Table 2. Outcome comparisons: laparoscopic vs open

Laparoscopic(n 20)

Last 12laparoscopic(n 12)

Open(n 100)

Age 72.2 (6288) 72.3 (6285) 71.6 (5291)% Male 75.00% (15/20) 75.00% (9/12) 76.0% (76/100)AAA size 5.98 cm 5.73 cm 6.12 cm (3.912.0)X-Clamp time 2 h, 26 min 2 h, 5 min 1 h, 36 min

Anesthesia time 7 h, 48 min 7 h, 18 min 5 h, 1 minICU days 2.45 days 1.50 days 3.22 daysLength of stay 6.20 days 5.92 days 9.97 days

Days NPO 2.15 days 1.08 days 5.35 daysDays on vent 0.50 days 0.42 days 2.18 daysEBL 1713 ml 1825 ml 1387 mlOR crystalloid 7595 ml 8333 ml 6524 mlOR crystalloid/h 977 ml/h 1170 ml/h 1288 ml/hOR colloid 275 ml 250 ml 303 mlOR cell saver 844 ml 727 ml 719 mlOR PRBCs 1.95 U 1.92 U 1.31 UOR plts 0.80 U 0.00 U 0.60 UOR FFP 0.40 U 0.18 U 0.18 UHospital PRBCs 2.75 U 1.92 U 1.46 UHospital plts 6.50 U 1.33 U 0.90 U

Hospital FFP 2.50 U 1.25 U 0.91 UMortality rate 10.0% (2/20) 0.00% (0/12) 4.00% (4/100)

Fig. 2. Depiction of the patient position on the operating table.

1067


25/99

entire right iliac artery videoscopically down to the right inguinal ligament,even in very obese patients. This method greatly facilitates subsequentpassage of the crossover limb of the bifurcated graft. During creation ofthis tunnel, the first assistant maintains an upward lift of the posteriorperitoneum with a Pennington grasper or similar 1 cm-wide device. Theright ureter must be clearly identified not only to avoid injury, but also tomake sure that the right limb of the graft passes posterior to the ureter.

Right and left groin incisions are then made to expose the femoralarteries. In four patients who had normal external iliac arteries and nofemoral artery disease, we stayed out of the groins and obtained retroper-itoneal exposure of the external iliac arteries through incisions above theinguinal ligaments. The tunnels are then completed between these incisionsand the left retroperitoneal operative space. Once completed, ease of sub-sequent access through the crossover tunnel into the right incision is en-sured by placing a 12-in. Penrose drain within it.

The gasless phase

Gasless laparoscopic techniques are used for the remainderof the operation because pneumoretroperitoneum is difficultto maintain when tunnels to the incisions used to expose thefemoral or external iliac arteries have been completed andcertainly cannot be maintained once the graft server is in-troduced into the operative cavity through the left outflowincision.

To achieve a seamless transfer from the gas to the

gasless phase, a sequence of steps must be executed. Therotation of the left kidney is maintained with the extra-handretractor. CO

2is discontinued. The laparoscopic trocars are

removed from the initial access and the two main operativeincisions. The initial access incision is widened to 2 cm, thelaparolift device (fan, fan-lift/retractor, or air-lift; OriginMedsystems) is inserted, and the mechanical lift is used toreacquire the same exposure obtained during the gas phase.The laparoscope is reintroduced through the same port asthe lift. The main operative channels are then reaccessedwith disposable rubber thoracoscopy access ports, which arecut longitudinally to permit the surgical instruments agreater range of movement. The aortic cross-clamp incisionand tunnel is made through the 10th intercostal space andmaintained with a thoracoscopy port.

Heparin (100 U/kg IV bolus) is then given, and 3 to 4min later the iliac arteries are stapled distal to all aneu-

rysms using endoscopic T.A. 30 stapling devices (U.S. Sur-gical Corp., Norwalk, CT). (We have found that the 3.5-mmstaples are adequate for most vessels. When we use the4.8-mm staples, we have shoed the devices with PTFE orbovine pericardium strips.) The aortic neck is cross-clamped just distal to the renal arteries. We use a speciallydesigned laparoscopic aortic cross-clamp (Scanlan Interna-tional Catalogue #9909-912-13 St. Paul, MN). Previouslyidentified large lumbar arteries are clipped. If the inferiormesenteric artery is patent but not dominant, it is ligated.

At this point, the surgeon must ascertain the absence of

pulsation in the aortic sac. After the stapling of the commoniliac arteries and the application of the cross-clamp to theaortic neck, the aneurysm sac becomes floppy and can beeasily and safely maneuvered to seek and ligate any remain-ing lumbar branches. The aneurysm sac is partially openedand evacuated. (Continued bleeding from the aneurysm sacbeyond the usual initial flash is indicative of a missed lum-bar branch, which should be sought out and clipped.) Aportion of the aneurysm adjacent to the neck should beexcised and the remaining collapsed sac closed with a lap-aroscopic T.A. 60 stapling device (U.S. Surgical Corp.,Norwalk, CT). We have found that excising a portion of the

aortic sac near the neck greatly facilitates the laparoscopicsuturing required for the proximal aorta to graft anastomo-sis, especially when the aneurysm sac is very large or buck-led to the left. The neck is then prepared for the graft anas-tomosis by circumferential incision and definition. Anyneeded endarterectomy is performed at this juncture. In ourexperience so far, only five patients have not required end-arterectomy of the aortic neck.

An appropriately sized bifurcated woven Dacron graft isselected and loaded into a graft server (SMC Surg-MedDevices). The graft server is introduced into the operativecavity through the left groin or suprainguinal incision and i

Documents

surgical endoscopy aug1998