Taking it to the Next Level: Advanced Laparoscopic Techniques
April 18, 2001
America's Center
St. Louis, Missouri
|
Program Chairman |
Bruce D. Schirmer, M.D. |
|
Course Director |
Lee Swanstrom, M.D. |
|
Lab Coordinator |
Thadeus Trus, M.D. |
|
SAGES President |
Nathaniel J. Soper, M.D. |
|
SAGES Executive Director |
Sallie Liesmann Matthews |
SESSION I: NEW TOOLS
SESSION III: NEW TECHNIQUES
Taking it to the Next Level:
Advanced Laparoscopic Techniques
SESSION I: NEW TOOLS
1. ENERGY SOURCES: BIPOLAR, HARMONIC AND OTHER
University of Virginia Health Sciences Center
P.O. Box 800709
Charlottesville, VA 22908-0709
Advanced laparoscopic procedures have become considerably more feasible due to the improvement in surgical instrumentation which has occurred over the past decade. Improved instrumentation alone cannot confer upon a surgeon the ability to perform advanced laparoscopic procedures. Only the requisite acquisition of advanced laparoscopic surgical skills allows the safe performance of such procedures. Nevertheless, tools now available to the laparoscopic surgeon make the steps of dissection, vascular ligation, and tumor ablation significantly easier technically.
The ideal hemostatic instrument for use in either laparoscopic or open surgery would allow the surgeon to grasp and dissect tissue, while at the same time have the capacity to apply an energy source to reliably and rapidly ligate and divide vascular structures with reliable hemostasis and little injury to surrounding tissues. No such tool currently exists; those that do fulfill some or several of these criteria but not all of them.
Electrocautery
The most commonly used form of energy in laparoscopic and open surgical procedures is monopolar electrocautery. Monopolar electrocautery is produced by electrical energy modified by a transformer into sufficiently high frequency (1-2 million Hz) electrical energy that it can pass through the body without tissue injury except heating at the entry and exit points. In the "cutting" mode, electric current is generated at an extremely high frequency which causes disintegration and division of tissue at the cautery tip, with a resultant depth of burn injury of only approximately 0.1 mm. The "coagulating" mode generates electric current at a lower frequency and higher voltage with resultant increased heating of the tissues, greater sealing of small vessels, and a greater depth of tissue injury (1).
Monopolar electrocautery offers an energy source that is excellent for hemostasis of small blood vessels, easy to use during tissue dissection, rapid, accurate, and inexpensive. The major disadvantages are the limitations in terms of size of vessels (< 1 mm) which may be hemostatically divided, the potential for exit site injury, and the potential for energy to jump a tissue gap away from the site of the cautery tip if there is a leak in the energy circuit. This is the major source of "off-camera" injuries which may occur during laparoscopic surgery (2). The estimated incidence of such injuries is 0.2%. Monopolar cautery also produces a large degree of smoke, especially if the tissues are moist, and it is ineffective within a liquid pool.
Bipolar electrocautery
On the whole, general surgeons have underutilized bipolar cautery in their operative tools. Bipolar cautery uses high frequency electric energy to melt and coagulate tissue placed between two directly apposed electrodes. Because the electrodes are closely approximated, low voltages of electricity are adequate. It is highly effective for use in hemostatic ligation of small vessels and vascular tissue, and can be performed with good speed. Studies comparing the use of bipolar cautery forceps to the harmonic scalpel in performing laparoscopic surgery have shown little difference in efficacy, but the bipolar instrument is significantly less costly. Bipolar scissors have also been shown to improve speed and hemostasis over conventional instruments in the performance of abdominal hysterectomy (3). However, in general such scissors are not appropriate for division of larger vessels.
Bipolar electrocautery does eliminate the need for a patient grounding pad, and the current does not pass through the entire body, thus elminating the danger of burns distant to the site of the forceps tips. Its depth of injury to surrounding tissues is minimal (4), and it produces little smoke.
Currently there are laparoscopic bipolar forceps available which coagulate and seal tissue and have a knife available for subsequent division of the tissue between the jaws of the forceps. The instrument is not well designed for dissection of tissues, and there is not a clear signal to the surgeon when the tissues within the jaw have been safely coagulated. In general most bipolar instruments have jaws which are poorly designed for dissection.
Ultrasonic coagulator
The ultrasonic coagulator is an instrument which allows hemostatic division of small to medium blood vessels (up to 3 mm in diameter) safely. The instrument uses ultrasonic energy waves (55,000 Hz generally) which are transmitted to the vessels through the active blade of the scalpel. The energy waves essentially coapt or seal the vessel closed by denaturing the protein in the vessel wall and achieving a plug in the lumen of the vessel.
The major advantages of the ultrasonic coagulator are that it allows safe division of vessels, as well as having the potential to perform dissection due to the shape of the blades. Tissue division and dissection can be simultaneously performed, and where this occurs in vascularized tissue such as omentum, the ultrasonic coagulator is a powerful tool for rapidly and hemostatically dividing such tissue. The blades of the instrument produce very little spread of energy, and so damage to adjacent tissues is low. The blades can become quite hot, and burn injuries are not impossible. The coagulator produces very little smoke, allowing good continuous visualization during dissection. In addition, it can perform hemostatic division under a liquid surface, unlike electrical energy sources. Finally, no energy is transmitted through the patient.
The ultrasonic coagulator has made a major impact on the capability of performing several common laparoscopic operations. It has been shown to clearly improve the speed of performance of a laparoscopic Nissen fundoplication, where it rapidly and safely divides the short gastric vessels (5), and can replace all other energy sources for the performance of this procedure. It is effective in dividing the pancreas (6) and better than monopolar cautery in the performance of resorative proctocolectomy (7). Transverse colon resection, where the omentum had previously been troublesome, is now much more feasible, as it division of smaller vessels within the mesentery of the intestine. The vascular tissue surrounding the adrenal gland is also hemostatically divided with this instrument, preserving the hemostasis for good visualization of the gland parenchymal edge.
The surgeon can use several features of the ultrasonic coagulator for slower and more hemostatic or more rapid and less hemostatic division of tissue. The blade configuration (flat versus sharp), grip pressure (light versus hard) and variable speeds of energy delivery (slow versus fast) all promote slow versus rapid tissue division. The instrument can be obtained in 5 or 10 mm shaft widths, and the blades can be straight or curved.
LigaSure Vessel Sealing System
The most recent addition to the laparoscopist's arsenal of tools which may be used to divide tissue and blood vessels is the LigaSure Vessel Sealing System. The LigaSure uses bipolar electrical energy to totally obliterate the lumen of a vessel, heating it to the point where the vessel walls literally fuse together for a width of several millimeters. The obliterated lumen is readily identifiable once the jaws of the instrument are released. As the energy is being applied, there is an audible signal which informs the surgeon that the obliteration has been completed. The LigaSure can safely seal and divide vessels which are up to 7 mm in diameter. It therefore is the only instrument currently available which can safely and reliably divide larger (4-7 mm) sized vessels. Despite the high degree of heat produced between the jaws of the instrument, it does not produce a large amount of lateral heat, and damage to surrounding tissues is low. Smoke production is also low.
While the LigaSure is clearly the most dependable and safe instrument to use laparoscopically to divide mesentery and other medium sized vessels, its major disadvantage is that it is somewhat slow to use as a dissecting instrument. This is because with each tissue division, the generated heat must be sufficient to weld closed the vessels and tissues, which can be tedious. Then, after the signal is produced and the jaws are opened, the obliterated vessel area must still be divided with a scissors, as the LigaSure cannot divide tissue. The need to introduce and use the scissors each time greatly slows the speed of tissue divison, especially compared to the ultrasonic coagulator (8).
Laser Energy
Laser is an acronym for Light Amplification by Stimulated Emission of Radiation. Laser physics involves the principle that a material becomes energized or excited by an energy source, then gives off the energy as a photon or light particle. By many molecules of the energized substance giving off photons that are in phase together, a laser beam of light energy is formed. Various different materials can be used as the material which is activated, and the laser is named for the material (such as argon, CO2, etc.).
The laser beam is produced when the energized medium is activated and the majority of particles release same-length in phase photons with identical wavelength, called coherent light. Coherent light is produced within a laser chamber by having the chamber lined with mirrors, which reflect the photons back into the center, further activating the medium inside. At one end, a small opening allows a small portion of the light, which must be coherent to be emitted, out of the chamber in a laser beam.
The most commonly used medical lasers have light wave properties designed to accomplish specific tissue injuries. The CO2 laser produces light in the far-infrared spectrum, which is highly absorbed by water. Thus this laser is most effective in vaporizing a thin layer of surface tissue which is rich in water content. The resulting depth of injury is shallow. The Nd:YAG laser, on the other hand, has a light wavelenth in the near-infrared spectrum, and is poorly absorbed by water, resulting in its ability to deeply penetrate into tissues. It is useful for ablating obstructing luminal tumors. The argon and KTP lasers produce wavelengths in the visible spectrum, resulting in light waves absorbed by sustances such as hemoglobin or melanin, making these lasers advantageous to ablate lesions such as hemangiomas.
Lasers may offer selective advantages in certain minimally invasive surgical situations, especially endoscopic interventions such as destruction tumors or bleeding mucosal lesions. Their main disadvantage is that the beam, if not used in a contact fashion, can injure tissues beyond the intended lesion if inappropriately pointed or directed while activated. Lasers are also very expensive, and most commonly performed laparoscopic procedures no longer use laser energy sources as a major tool (9).
Tissue Ablation Energy Sources
While the topic for another entire lecture, the recent use of solid organ lesion ablation, particularly the destruction of hepatic metastases, will be briefly discussed. The two most commonly used energy sources for solid tissue destruction have been cryotherapy, or deep freezing, of tissue, and radiofrequency destruction. Both modalities use probes to deliver the energy into the hepatic parenchyma and achieve tissue destruction. Cryosurgery was first used for hepatic parenchymal destruction, with reasonable efficacy. Problems included difficulty assessing the degree of injury of the ice ball created by the energy tip, as well as protection of surrounding tissues from hypothermic injury. RITA (radiofrequency interstitial tumor ablation) has largely replaced cryosurgery for destruction of hepatic parenchymal lesions in the past few years. Radiofrequency energy is transmitted between two active electrodes placed under ultrasound guidance within the hepatic lesion, and destruction is monitored using visual and ultrasound criteria. Efficacy has been shown for destruction of hepatic lesions; experimental studies suggest this may soon be applicable to renal tumors as well (10).
REFERENCES
1. Odell RC. Laparoscopic electrosurgery. In: Hunter JG, Sackier JM, eds. Minimally Invasive Surgery. New York: McGraw-Hill, 1993: pp 33-41.
2. Voyles CR, Tucker RD. Education and engineering solutions for potential problems with laproscopic monopolar electrocautery. Am J Surg 1992; 164:57-62.
3. Dessole S, Rubattu G, Capobianco G, et al. Utility of bipolar electrocautery scissors for abdominal hysterectomy. Am J Obstet Gynecol 2000; 183:396-9.
4. Barrat C, Capelluto E, Champault G. Intraperitoneal thermal variations during laparoscopic surgery. Surg Endosc 1999; 13:136-8.
5. Underwood RA, Dunnegan DL, Soper NJ. Prospective, randomized trial of bipolar electrosurgery vs. ultrasonic coagulation for division of short gastric vessels during laparoscopic Nissen fundoplication. Surg Endosc 1999; 13:763-8.
6. Takao S, Shinchi H, Maemura K, Aikou T. Ultrasonically activated scalpel is an effective tool for cutting the pancreas in biliary-pancreatic surgery: experimental and clinical studies. J Hepatobiliary Pancreat Surg 2000; 7:58-62.
7. Kusunoki M, Shoji Y, Yanagi H, et al. Current trends in restorative proctocolectomy: introduction of an ultrasonically activated scalpel. Dis Colon Rectum 1999; 42:1349-52.
8. Fried GM. Hemostatic tools for the gastrointestinal surgeon: ultrasonic coagulator vs. bipolar ligation. J Gastrointest Surg 2001; 5(2): 1-3.
9. Laycock WS, Hunter JG. Electrosurgery and laser application. In: MacFadyen BV Jr, Ponsky JL, eds, Operative Laparoscopy and Thoracoscopy. Lippincott-Raven, Philadelphia, 1996; pp 79-91.
10. Zlotta AR, Wildschutz T, Raviv G, et al. Radiofrequency interstitial tumor ablation (RITA) is a possible new modality for treatment of renal cancer: ex vivo and in vivo experience. J Endourol 1997; 11:251-8.
2. Ultrasound: The Surgeon's Hand
Chief of G.I. Surgery, Saint Pierre University Hospital , Brussels (Belgium)
Laparoscopic surgery brings a benefit to the patient yet poses challenges to the surgeon. The axe of vision is not the same than the axe of working. The surgeon needs to manipulate long and sharp instruments through a fix opening under the control of a bidimensional screen and without any tactile sensation.
The body cavity is penetrated by cannulas which cannot be interchanged. Therefore the surgeon needs to move around the patient in order to reach the best position for every step of the procedure.
A computer interface in command of a mechanical system (robot) allows:
1. to recuperate several lost degrees of freedom, thanks to intra-abominal articulations.
2. to obtain better visual control of instrument manipulation thanks to three-dimensional vision
3. to modulate the amplitude of surgical motions by downscaling and stabilisation
4. to work at a distance from the patient.
These features allow improved quality of surgical tasks, not the least thanks to substantially better ergonomics.
The robot (Da Vinci system, Intuitive Surgical, Mountain View, Ca) consists of a console and a surgical cart which supports three articulated robot arms. The surgeon is sitting at the console. He manipulates joystick like handles while observing the operative field through binoculars that provide a three-dimensional image. This computer is capable of modulating data by eliminating physiologic tremor and by downscaling the amplitude of motions by a factor 5 or 3 to one.
The first robot assisted procedure in a man was performed in March 1997 by our team. Since this time 130 patients underwent robot assisted laparoscopic surgery including anti-reflux procedures, cholecystectomies, tubal reanastomosis, gastroplasties for obesity, inguinal hernias, intra-rectal procedures...
1/This study has demonstrated the feasibility for telesurgery on humans in different procedures, without specific morbidity, and in acceptable operative times.
2/ In its present embodiment, the system seems most efficient when involving micro-suturing within the abdomen or in very confined spaces.
3/ improved ergonomic conditions and improved instrument mobility at the distal articulation level seem beneficial in usual abdominal procedures. More research is needed for further improvement in tool shape and optics for this type of approach.
4/ The robotic approach implies new operative strategies, including a specific trocar placement.
Technology assessment has not kept pace with other fields such as pharmaceutical trials and software development
Most technology is assessed by measuring crude safety and efficacy measures along with subjective methods such as focus groups, individual trials, etc.
There is a need for the scientific method in assessing new technology. What are the variables of interest?
Mortality and complication rates are too crude
Operative time or length of stay are better
Measures of individual and team performance are good
Measures of physical and mental workload are best
It is possible to test technology using the scientific method - i.e. hypothesis testing
The field of Human Factors and Ergonomics (HFE) does precisely this: Evaluate individual and team workload and performance in relation to the work environment. The experience learned from the study of office workspaces, the dental operatory, aircraft carrier flight decks, aircraft cockpits, etc. can be used to better assess medical technology
What questions to ask?
1. Is the device safe for the patient and the user?: e.g. electrical, mechanical safety? Does it prevent
you from doing the wrong thing?
2. Does the device do what it is supposed to do effectively?: e.g. cut, tie, hold
3. Is the use of the device intuitive? Can you tell by looking at it and holding it how to use it properly
and safely?
4. How will it work in the real OR environment with other users and equipment present? Does it
complement or simplify what you have, or does it more complexity?
5. What evidence is there that the device is better that what you currently have? Are there any
laboratory or clinical studies of its use? What do those studies measure: Performance, workload,
subjective ratings, clinical outcomes?
1. Berguer, R., Surgery and ergonomics. Arch Surg, 1999. 134(9): p. 1011-1016.
2. Berguer, R., D. Forkey, and W. Smith, Ergonomic problems associated with laparoscopic surgery.
Surg Endosc, 1999. 13: p. 466-468.
3. Davies, R., Science and Surgical Technology. Contemporary Surgery, 2000. 56: p. 714.
Taking it to the Next Level:
Advanced Laparoscopic Techniques
SESSION II: NEW APPROACHES
5. Accessing the Pre- and Retroperitoneum
SYLLABUS NOT AVAILABLE
6. Hand-Assisted Laparoscopic Surgery
University of Massachusetts Medical School, Worcester, MA
Although laparoscopic surgery has been widely embraced for operations such as laparoscopic cholecystectomy, complex laparoscopic operations are less likely to be performed by surgeons. There are a number of reasons for this, including lack of adequate training, lack of surgical instrumentation, absence of tactile feedback, fear of changing the biology of a malignancy, and lack of perceptual/motor coordination. Also, at times it may not seem apparent to the surgeon that the outcome is worth the time and effort expended. For these reasons, the introduction of the hand into the abdominal cavity during laparoscopic surgery may be beneficial.
A precise operative strategy is paramount to successful hand-assisted laparoscopic surgery (HALS). The initial thought to port placement may be to place the hand-assist incision directly over the target structure. However, movements become awkward and uncomfortable and the hand may be difficult to see around and maneuver around with laparoscopic instruments. Instead, the hand should be used like an operating port, where triangulation and forward access are vital to success. The hand should be placed in a neutral position; hyperflexion and hyperextension of the hand can be uncomfortable and fatiguing. In addition, the ability to convert the hand-assist incision to a usable conversion incision is also important. (Figure 1)
The hand can be used to palpate structures, to retract structures atruamtically, or to apply traction. When used properly, it can increase operative efficiency and the surgeon's confidence. Our approach, when possible, is to use the non-dominant hand.
There are several devices on the market in the United States: HandPortTM (Smith & Nephew, Inc., Andover, MA), PneumosleeveTM (Dexterity Inc., Roswell, GA), and IntromitTM (Medtech Ltd., Dublin, Ireland). We prefer the HandPortTM device, which was developed at the University of Massachusetts Medical School in conjunction with Smith and Nephew Endoscopy. Indeed, our enthusiasm with HALS led to this development.
An incision large enough to fit the surgeon's hand (typically 7 cm) is created, and the base ring retractor is placed in the abdomen and inflated. The bracelet is placed over the gloved wrist, and is covered with a second glove (brown: the white glove reflects light from the laparoscope and can impair the visual field). Next, the sleeve is placed over the arm and well-lubricated glove, and the lower end snaps into the bracelet. The hand is placed into the abdomen, and the sleeve is attached to the base ring retractor. Pneumoperitoneum is created. (Figure 2)
The sine qua non of any hand-assist device is that it retains pneumoperitoneum. Devices that rely on adhesives to prevent air leakage have an unsatisfactory failure rate.2 Attempts to place the hand in the abdomen, with the wound cinched closed with sutures also work, but to a limited degree. The premise of the circular retractor placed in the wound and inflated to maintain its position makes it reliable in most circumstances. It may fail in obese patients, but a device with a longer skirt length is under development. Also, placement too close to a bony structure such as the iliac crest or costal margin may cause it to slip out altogether.
Our current management algorithm is based upon the particular patient. Although most of our patients are considered for laparoscopic surgery, some may be unsuitable. While some of these will receive open surgery, others may be suitable candidates for HALS, and thus will be spared an open operation. Also, patients destined for conversion to open surgery may be best served by conversion to HALS. Surgeons are faced with management options ranging from purely laparoscopic to open surgery, with HALS in the middle.
There are multiple potential applications for HALS. Gastrectomy can be complex and time consuming laparoscopically, and an incision may be necessary to remove the specimen. Splenectomy can be an arduous task in cases of splenomegaly; the hand may provide elevation of the spleen to appropriately visualize its attachments. Donor nephrectomy can be performed with warm ischemia times comparable to open nephrectomy. Colon resection shows great promise; an incision is required for specimen extraction, and organ retraction can be difficult in the purely laparoscopic approach. Other procedures, such as pancreatic and liver resections, spine surgery and morbid obesity surgery under investigation, but hold great promise.
With any new technology, one must ask if the additional costs accrued translate into discernible benefit. With HALS, the benefit that probably occurs is a decrease in the operative time. Also, one must ask if the physiologic benefits of a minimally invasive approach are conferred by HALS. Certainly, a 7 cm incision is larger than is required by most laparoscopic operations. However, it is smaller and less morbid than most operations with presumably less postoperative pain and faster return to function. Also, with HALS, there is less instrumentation of the bowel, which may decrease morbidity.
Future considerations for HALS include the invention of hand-directed instruments that can be placed through the device. Also, techniques to allow conventional instruments through the existing devices are under investigation. This may allow the concept of HALS to expand even further.
HALS is a part of the continuum of the surgical options ranging from pure laparoscopic surgery to pure open surgery. Its use may allow for more complex operations to be safely completed with the benefits of a minimally invasive approach.
References
1. Litwin DEM, Novitsky Y, Kercher KW et al. Hand-assisted laparoscopic surgery. Surgical Technology International IX, 2000.
2. Meyers W, et al. A prospective multicenter trial of a minimally invasive technique for complex abdominal surgery. Arch Surg 1999; 134: 477-85.
7. Endoluminal Access Techniques: Laparoscopic Intragastric Surgery (LIGS)
Professor, Department of Endoscopic Surgery
Osaka University Medical School, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan
Eiji Taniguchi, MD, Shuji Takiguchi, MD, Kiyokazu Nakajima, MD
Departments of Endoscopic and Pediatric Surgery, Osaka University Medical School
GENERAL REMARKS
Traditionally, surgical techniques employing endoluminal access involve polypectomy, thermocoagulation, strip biopsy etc using flexible gastrofiberscopy or colonoscopy. Since the surgical instruments are passed into the gut lumen via a small working channel of the scope, procedures realized in this setting have fateful technical limitations. A steady effort, therefore, has been made to improve the surgical flexibility. Endoscopic mucosal resection (EMR), a well-established technology treating early-staged esophageal or gastric cancer, is one of the fruits of this endeavor. Cases with small lesions (i.e. <10 mm in diameter) are best indicated for this technique. EMR is also feasible for lesions > 10 mm, however, piecemeal resection becomes mandatory.
Transanal endoscopic microsurgery (TEM), another promising methodology managing colorectal neoplasms, has gained wider clinical acceptance since its emergence in early 90's. With TEM technique, surgeons can access into the colorectal lumen, and can utilize various instruments under magnified visualization of rigid endoscope. Cases with lesions located within 20 cm from anal verge are candidates for this technique.
With the recent progress in laparoscopic surgery, endoluminal access techniques have gained another "access" route. Modern laparoscopic trocars have enabled surgeons to gain easy access into the gut lumen with minimum enterotomy incisions. Laparoscopic endoluminal surgery (LES), preserves the integrity of the operated organs, and still keeps well-known advantages of minimally invasive techniques. Surgeons are allowed to use their both hands with familiar instruments, and can perform theoretically all endoscopic procedures inside the digestive tract. This novel procedure is, however, rather technically demanding and requires sound endoscopic skills. The author herein describes the basics of typical LES in the stomach: laparoscopic intragastric surgery (LIGS).
INDICATIONS FOR LIGS
LIGS is currently indicated for the treatment of early-staged gastric cancers (i.e. cancers located within mucosal or submucosal layers), benign mucosal or submucosal tumors. Virtually any part of the stomach except for the limited area of the anterior gastric wall, is accessible with our current LIGS technique. Cases with intractable peptic ulceration with bleeding, which is difficult to be controlled with flexible gastrofiberscope, are also potential candidates for this technique. LIGS is also considered as an attractive alternative for patients with preoperative risk factors e.g. advanced age, malnutrition and severe hepatic dysfunction.
PREOPERATIVE PREPARATION
In addition to the routine work-up for general anesthesia, flexible gastrofiberscopy and/or endoscopic ultrasonography are employed in all patients, to identify the location, size and depth, and histology of the gastric lesions. In cases with flat or shallow lesions, markings with metal clips or tattoos are recommended to help identifying the lesions at surgery. Barium swallow is also routinely performed to evaluate the distance or angle between the lesion and the puncture sites. This is essential in planning the trocar layout and evaluating the accessibility prior to surgery.
PATIENT POSITIONING AND ROOM SETUP
The general OR setup is very similar to laparoscopic cholecystectomy in most respects. Patients are placed in the supine head-up tilt position on the operating table. A modified lithotomy position is optional. The operating surgeon generally stands on the left side of the patient, and the first assistant and scrub nurse on the right side. The endoscopist works at the head of the table, outside the sterile field. Usually two videomonitors are placed at the head of the table.
TROCAR POSITIONING
Under general anesthesia, a gastrofiberscope is inserted per orally into the stomach. The gastric lesion is reconfirmed, and the stomach is filled with air as same as routine gastrofiberscopic examination. A naso-gastric tube with balloon is then placed in the duodenum and the balloon is inflated to prevent the air-flow from the stomach to the intestine. While observing the image from the gastrofiberscope on the monitor, the abdominal wall is compressed with surgeon's forefinger to select three adequate points for trocar insertion. It is essential to select these points in which the abdominal wall touches the anterior gastric wall when compressed with a forefinger. Logically, these three points should be about 5 cm or more apart from one another, in order to avoid the instrumental fighting. For lesions located at upper part of the stomach, an initial trocar for optics should be placed near the umbilicus. Remaining two working trocars are to be placed at convenient working angles (i.e. 30 to 40 degrees). For lower gastric lesions, the initial access point should be at the left upper midabdomen.
PERFORMING LIGS
Upon the determination of trocar positioning, a skin incision of 1 to 2 cm is made at the initial puncture site. Two sets of stay sutures are placed on the anterior wall of the stomach through the incision, and a small "cut-down" gastrotomy is made. A 10/12 mm Hasson cannulae is inserted via gastrotomy and inflated to fix the gastric wall to the abdominal wall. Placement of other two working trocars is then accomplished with standard puncture technique. After evaluating various types of trocars, the author's present preference is the 5 mm radially expandable trocar system (Step®, Innerdyne, Inc., Sunnyvale, CA). The use of more miniaturized, 3 or 2 mm trocar systems is optional. All puncture procedures should be directly monitored with gastrofiberscopy.
After insertion and fixation of all three trocars in the stomach, the 10 mm 0 degree videolaparoscope is inserted. The gastric lesion is identified, and the adequate removal margin is determined. The next pivotal step is a submucosal injection of normal saline with epinephrine. This classic maneuver facilitates mucosal resection, and prevents bleeding or perforation. Saline of 30 mL in volume, up to 100 mL, is usually necessitated for this purpose. After obtaining the sufficient mucosal bulging, the removal margin is marked circumferentially with electrocoagulation. Surgery is then performed with double-handed technique, totally within the lumen of the stomach. The resected specimen is extracted per orally using gastrofiberscope, or delivered via trocar after isolation in the plastic bag. Repair of the mucosal defect is optional.
After extraction of the specimen, the each trocar is removed under the guidance of gastrofiberscopy. The initial cut-down site for the laparoscope is meticulously closed. When Step® trocars are used, the author does not routinely repair the puncture sites.
POSTOPERATIVE CONSIDERATION
Patients are encouraged to be fully ambulatory on the following day. Clear liquid is allowed on 1 POD, and the solid diet is resumed after the removal of naso-gastric tube. H2-blocker coverage is routinely required postoperatively. One major possible complication is bleeding, however, this can be mostly avoided by meticulous hemostasis during operation. Other possible complications include unrecognized or delayed gastric perforation (probably due to excessive electrocoagulation or lateral thermal injury) and leakage of the gastric content from the trocar entry sites. These complications can be prevented or minimized, with the routine postoperative naso-gastric decompression.
THE OSAKA EXPERIENCE
From February 1993 to December 2000, the author and colleagues experienced 43 cases of LIGS for mucosal / submucosal gastric cancers. The mean operative time was 153±62 minutes, and the estimated blood loss was 62±129 mL. The mean size of the resected lesion was 36.2±14.1 mm. There was one case with positive surgical margin. Five out of 43 cases (in early series) were converted to open procedures because of technical difficulties. Except for one case with postoperative bleeding requiring transfusion, no major complications were encountered. With mean follow-up period of 39.9±20.6 months (max 7 years), we have experienced no recurrences or metastases.
SELECTED REFERENCES
1. Ohashi S: "Laparoscopic intra-gastric surgery": is it a new concept in lap surgery? (Abstr). Surg Endosc 8: 256, 1994.
2. Ohashi S: Laparoscopic intraluminal (intragastric) surgery for early gastric cancer. A new concept in laparoscopic surgery. Surg Endosc 9: 169-171, 1995.
3. Way LW, Legha P, Mori T: Laparoscopic pancreatic cystgastrostomy: the first operation in the new field of intraluminal laparoscopic surgery (Abstr). Surg Endosc 8: 235, 1997.
4. Spivak H, Hunter JG: Endoluminal surgery. Surg Endosc 11: 321-325, 1997.
5. Taniguchi E, Kamiike W, Yamanishi H, Ito T, Nezu R, Nishida T, Momiyama T, Ohashi S, Okada T, Matsuda H: Laparoscopic intragastric surgery for gastric leiomyoma. Surg Endosc 11: 287-189, 1997.
6. Taniguchi E, Ohashi S, Takiguchi S, Yumiba T, Itoh T, Matsuda H, Nakajima K: Laparoscopic intragastric surgery using a radially expandable sleeve. Surg Endosc 14: 505-507, 2000.
Department of Surgery and Molecular Oncology, Ninewells Hospital and Medical School, University of Dundee.
In the first instance it is important to distinguish glues and adhesives that attach to a surface principally by molecular cross-linking from cements that achieve union through mechanical interlocking [1]. Cements e.g., polymethylmethacrylate, are initially liquid or plastic and hence conform with the irregularities within the underlying substrate (bone) before solidifying on polymerisation producing a mechanical but not a chemical bond. They are not considered further in this review.
Tissue adhesives (glues) used in surgical practice (open and endoscopic) fall into three categories:
The biological glues are of several types
The synthetic glues are further classified into non-resorbable or resorbable. The medical/ surgical use of non-resorbable glues is limited to surface applications, whereas the resorbable (biodegradable) glues can be deployed for both surface applications and internal use. It is likely that glues will eventually replace in whole or in part, sutures and staplers for the approximation and binding of tissue. The financial importance of the topic is illustrated by the existing 2.5 billion $ worldwide market for wound closure. This is currently equally divided between sutures and staplers.
Biological Glues
Fibrin glues
These function by reproducing the latter stages of the normal clotting cascade leading to the formation of a stable fibrin clot from fibrinogen. These glues are used largely as hemostatic agents for bleeding surfaces and vascular anastomoses. There is a wide range of fibrin glue products but in essence they fall into two types: two-component fibrin and cryoprecipitate based glues. The composition of the two-component types is basically the same despite several different product proprietary names, i.e., the glue consists of two solutions (Fibrinogen and Thrombin + Calcium) which are mixed together at the point of application. The many different preparations vary in the source of the fibrinogen (usually human), the thrombin (usually bovine), other clotting factors/ additives. The risk of hepatitis is reduced by donor screening and heat inactivation but the advent of bovine spongiform encephalopathy and the new-variant CJD within many European countries will almost certainly lead to the total exclusion of bovine thrombin in the future. Some two-component fibrin glues such as Tisseel incorporate an anti-protease (aprotinin) to retard the fibrinolytic process and hence prolong the effective sealing period. Others, e.g., Scottish Fibrin Glue are devoid of `additives' and appear to be equally effective. In a comparative experimental study on rat femoral artery microvascular anastomosis increasing the viscosity of Tisseel by the addition of Na hyaluronate resulted in a significantly higher patency rate 20 minutes after completion of the anastomosis and reduced the amount of fibrin that entered the vessels.
FibRx is a light-activated fibrin adhesive consisting of human fibrinogen and thrombin incorporating an inhibitor that retards the polymerisation of fibrinogen by thrombin to form the clot until the mixture is exposed to light. Light exposure results in evaporation of the inhibitor enabling clot formation.
The main advantages of all fibrin based glues is their lack of toxicity and complete compatibility such that healing is not disturbed, the material is resorbed by natural fibrinolysis without any inflammatory response [Byrne et al]. It is thus safe even in the CNS and is in established usage for dural closure and repair of CSF leaks [Felman et al 1988]. Portal vein embolisation with lipiodol-fibrin adhesive mixture has been used in the management of primary hepatocellular carcinoma [Matsuoka et al 1984]. Management of bronchopleural fistula is a challenging clinical problem. Laser-assisted cryoprecipitate bonding has been used to treat bronchopleural fistulae by the bronchoscopic approach. This laser assisted soldering produces a reliably strong seal over leaking bronchial stumps [Oz MC et al 1992]. Fibrin glue has been used effectively in the laparoscopic management of patients with blunt hepatic trauma. In one report of 61 patients, 55 patients were successfully treated without recourse to laparotomy [Chen et al 1998]. Thus the selective use of laparoscopy and fibrin glue can reduce the non-therapeutic laparotomy rate among blunt hepatic trauma patients. Two-component fibrin glue has been used to fix the prosthesis during percutaneous endoscopic external ring (PEER) inguinal hernia repair [Waldron et al, 1998]. Two component fibrin glue is very effective in preventing bleeding from surface liver cracks that form during hepatic laparoscopic cryotherapy for in-situ ablation of liver tumours [Cuschieri et al 1995]
Waldron B, Frank T, Cuschieri A. Percutaneous endoscopic external ring hernioplasty using Dundee inguinal canal retractor. Sem Lap Surg 1998; 5: 253-8.
Gelatin-based glues - Hydrogels
These provide alternative resorbable biological glues that have (as a class) greater bonding strength than fibrin based glues. The first such hydrogel glues developed and used clinically were GRF (gelatin + resorcinol + formaldehyde) and GRFG (gelatin + resorcinol + formaldehyde + glutaraldehyde) glues. GRFG has been used in thoracoscopic pulmonary surgery as sealant for the pulmonary parenchyma to prevent pneumothorax [Nomori et al 1999]. Despite efficacy, widespread acceptance of the GRF/G glues has been limited because of their cytotoxicity including high pyrexia resulting from the release of formaldehyde during degradation. Substantial late complications have been reported following GRF use in cardiothoracic surgery [Bingley et al 2000]. The adverse effects of GRF were clearly demonstrated in an experimental study comparing the effects of GRF and fibrin glue (Tissucol) on the rat abdominal aorta. The GRF glue induced destruction of the vascular wall with multiple inclusions of the glue observed in the media. By contrast, the two component fibrin glue preserved the normal architecture of the three arterial layers [Portoghese et al 1992].
In a comparative experimental study of cryoprecipitate glue, two-component fibrin sealant, and gelatin-resorcinol-formaldehyde-glutaraldehyde (GRFG) glue, the cryoprecipitate glue and the two-component fibrin sealant glue were equally effective in controlling bleeding from the aortic and atrial suture lines whereas GRFG did not provide total control at either site. The cryoprecipitate glue and the two-component fibrin sealant caused minimal adhesions in the subpericardial space whereas moderate-to-dense adhesions were present in the GRFG glue group at 6 weeks. The two-component fibrin sealant was completely reabsorbed by 10 days, but cryoprecipitate and GRFG glues took longer. Both fibrin glues exhibited minimal tissue reaction; in contrast to extensive fibroblastic proliferation induced by GRFG glue. Both the two-component and GRFG glues had outstanding adhesive properties whereas the cryoprecipitate glue did not show any adhesive power [Basu et al].
The second generation of gelatin-hydrogel glues are much less toxic as the formaldehyde has been substituted with other cross-linking agents, e.g., carbodiimide or genipin [Sung et al 1999] and poly L-glutamic acid (PGLA). The hemostatic activity of rapidly curable glues composed of gelatin and poly(L-glutamic acid) (PLGA) has been compared with that of two component fibrin glue. The gelatin - PLGA hydrogels were produced in aqueous solution by the addition of water-soluble carbodiimide (WSC). The WSC-catalysed gelatin-PLGA glues exhibited significantly higher hemostatic capability than the fibrin glue in animal experiments involving injured spleens. The gelatin PLGA hydrogels adhered much more strongly to the parenchymal surface of dog spleen than the fibrin counterpart.
Photochemically activated surgical tissue `bonding or soldering' technology involves using photoreactive gelatins and a water-soluble difunctional macromer (polyethylene glycol diacrylate: PEGDA). Photoreactive groups, e.g., ultraviolet light-reactive benzophenone or visible light-reactive xanthene dyes (e.g., fluorescein sodium salt, eosin Y, and rose bengal) are incorporated in the gelatins, which are then suspended in a saline solution containing PEGDA forming a viscous. This forms an adhesive hydrogel within 1min when irradiated with the appropriate light. The resulting gel is tightly adherent to soft tissues such as the liver. Experimentally this photocurative gelatin glue has been used to seal effectively arteriotomies in canine abdominal or thoracic aortas. This glue has great potential application in laparoscopic surgery, as the percutaneous delivery of the glue followed by in situ photogelation will result in prompt, safe and effective haemostasis [Nakaymama et al 1999].
Composite biological glues
The limitation of all fibrin-based and gelatin-hydrogel glues as hemostatic agents is that they are only effective in the absence of active bleeding. This problem has been overcome by the development of hybrid biological sealants that are effective in the control of active or diffuse bleeding. BioGlue is a composite of bovine serum albumin and glutaraldehyde, developed for treatment of acute aortic dissections and for strengthening vascular anastomoses. BioGlue is capable of binding prosthetic materials to tissues in the absence of bleeding and has potential for laparoscopic mesh hernia repair. FloSeal Matrix is composed of pellets of bovine collagen coated with thrombin. These form a matrix that induces strong clot that adheres to the underlying tissue. The resulting clot-matrix swells on absorption of fluid exerting pressure on the bleeding surface. The clot matrix is resorbed in 30 days. Early clinical studies with FloSeal Matrix confirmed effective hemostatic control in patients. CoStasis is a recent composite product of bovine fibrillar collagen and bovine thrombin suspended in calcium chloride solution developed for hemostatic control of diffuse bleeding. The agent is provided premixed in a syringe. Prior to application, fibrinogen from the patient is mixed with the suspension and then sprayed on the bleeding surface.
Polymeric Sealants
These have been developed as bio-sorbable sealants for internal use. FocalSeal-L is based on proprietary polyethylene glycol hydrogel technology. The product is liquid on application but polymerises to a solid gel when exposed to light. FocalSeal-L, is under evaluation for sealing air leaks during pulmonary resections. A randomised, multicentre RCT demonstrated superior air leak sealing than either sutures or staples. FocalSeal-S, is a similar but faster-resorbing hydrogel developed to seal the dura mater in spinal and neurosurgery. In a recent study experimental study in dogs FocalSeal-S prevented CSF dural leaks in 100% of the dogs compared to a leak rate of 35% with sutures. The product is applied in two stages (i) application of primer is followed by (ii) application of the sealant to the exposed dura, after which light irradiation results in polymerisation within in 40-60s. The other potential applications of the polyethylene glycol hydrogel technology is in reinforcement of vascular and gastrointestinal anastomoses/ suture lines.
Cyanoacrylates
Cyanoacrylates (CAs) (Table 1) have been available since 1960 but their clinical use has been limited because of the thermal damage and scarring to the tissues (hence strictures in anastomoses work) they produce and also because of unsubstantiated concerns regarding mutagenicity and carcinogenicity. There is no doubt that CAs are cytotoxic but this varies with the chemistry and the formulation. Experiments have consistently shown that polymerised CAs are cytotoxic to human fibroblasts and inhibit cell proliferation but this cytotoxicity varies considerably between the various cyanoacrylates [ Thumwanit et al 1999, Ciapetti et al 1994].
The endovascular treatment of cerebral arteriovenous malformations requires the use of cyanoacrylate glues especially when flow is very rapid. Isobutyl 2-cyanoacrylate (IBCA) was the agent used initially but this was replaced by Histoacryl which is much less toxic [Brothers et al 1989]. Histoacryl blue was the first cyanoacrylate glue to be used clinically for closure of skin incisions in Europe in the early 1980s. It is n-butyl-2cyanoacrylate monomer coloured by a blue dye. It is now the favoured adhesive embolic agent used by the interventional radiologists and for endoscopic control of bleeding varices and gastroduodenal lesions. Histoacryl and two-component fibrin glue have been used successfully and without complications in the endoscopic sealing of bronchopulmonary fistulae after pulmonary surgery [Inaspetto et al 1994]
Bucrylate (isobutyl 2-cyanoacrylate) has been used for the transcatheter occlusion of the splenic artery in 4 patients with bleeding gastric varices secondary to splenic vein thrombosis, 3 patients with symptoms of hypersplenism, and 8 patients with bleeding esophageal varices secondary to portal hypertension. The splenic artery was completely occluded in 13 patients and partially occluded in 2. In all but one of the patients, functioning splenic tissue was preserved and
no abscess developed. Medical splenectomy with Bucrylate thus appears to be a safe and effective and may preserve the immune function of the spleen [Golman et al 1981]. Cyanoacrylate glues have been used to repair meniscus and tendon injuries [Koukoubis et al 1995, Powell et al 1989], and to seal needle hole bleeding from PTFE during arterial grafting procedures [Citrin el al 1985].
Dermabond was the first synthetic cyanoacrylate used in USA and is now approved by FDA for closure of trauma-induced skin lacerations and small surgical incisions. The chemical formulation of Dermabond is 2-octyl cyanoacrylate monomer. When applied to skin, Dermabond polymerises within 45-60 seconds to form a more pliable bond than Histoacryl, which sloughs off after 7 - 10 days. Under current regulations, the glue is for surface application to the wounds, i.e. not in the base of the wound or between the wound edges. Clinical trials have demonstrated equivalent cosmetic results following healing to that achieved following wound closure by sutures but there appears to be a slightly higher infection rate following the use of Dermabond. Indermil is a related and rival cyanoacrylate glue for wound closure. The glue is delivered by a dedicated pump-driven injection system. Although it is licensed for skin closure only, it has been use for fixation of mesh during open and laparoscopic hernia repair [Jourdan and Bailey 1998]. Resorbable formulation of cyanoacrylate glues are being developed by a number of companies. Biodegradable versions of Dermabond and Indermil are currently undergoing experimental evaluation. The potential applications of biodegradable synthetic glues includes re-enforcement of intestinal and vascular anastomosis, sealing of air and liquid leaks, e.g., CSF leaks following trauma and surgery, pulmonary air leaks, intestinal fistulae. The progress in the absorbable synthetic technology will need to cover a range of specific requirements depending on the indication but the important variable qualities of designer absorbable glues will include: the material's flexibility, setting time, viscosity, toxicity and biocompatibility, bond strength and its in-vivo half-life.
LiquiBand is a transparent, cyanoacrylate-based topical adhesive for external use skin closure. The formulation (n-butyl-2 cyanoacrylate) has been combined with an antimicrobial agent to minimise the risk of infection in the wound under the layer of the adhesive material. LiquiShield, a related product by the same manufacturer, is a postsurgical wound dressing that is applied after the wound is closed by sutures or staples to provide a waterproof dressing that promotes healing and reduces the risk of infection by eliminating the need for daily wound dressing changes.
Tissue Adhesives based on Protein Engineering
These polymers are still at the experimental stage but show great promise as biocompatible and biodegradable internal sealants. They are based on proprietary protein engineering based on DNA gene technology. The polymers combine synthetic engineered silk and elastin proteins. When mixed with an organic crossing-linking agent, these form a strong, quick-setting, flexible, biocompatible adhesive matrix. An experimental study in guinea pigs showed that the material binds the wound edges. Epithelialization and healing with absorption of the material were observed at 28 days. The wound strength was equivalent to sutured controls.
Conclusions - Usage of Glues in Endoscopic Surgery and Radiological Interventions
Established indications:
Emerging indications:
Table 1: cyanoacrylate medical glues
Proprietary name |
Chemical name |
Usage |
Histoacryl |
n-butyl-2cyanoacrylate monomer coloured by a blue dye |
Skin closure embolic agent |
Dermabond non-resorbableDermabond resorbable |
2-octyl cyanoacrylate monomer |
Skin closureInternal seal and tissue approximation/ fixation |
Indermil non-resorbableIndermil resorbable |
n-butyl-2 cyanoacrylate |
Skin closureMesh hernioplasty, internal seal and tissue approximation/ fixation |
LiquiBand |
n-butyl-2 cyanoacrylate + antibiotic |
Antibiotic protected wound closure |
LiquiShield |
n-butyl-2 cyanoacrylate + antibiotic |
Postsurgical wound dressing |
Krazy Glue |
Ethyl-cyanoacrylate glue |
Sealing sutured vascular anastomosis |
Mediglue |
Ethyl-2-cyanoacrylate |
Skin closure |
Bucrylate |
Isobutyl-2 cyanoacrylate |
Transcatheter embolisation and medical splenectomy |
References
1.Donkerwolcke M, Burny F, Muster D. Tissues and bone adhesives--historical aspects. Biomaterials 1998;19:1461-6.
2.Byrne DJ, Hardy J, Wood RA, McIntosh R, Cuschieri A. Effect of fibrin glues on the mechanical properties of healing wounds. Br J Surg 1991;78:841-3.
3.Feldman MD, Sataloff RT, Ballas SK. Autologous fibrin tissue adhesive for cerebrospinal fluid leaks: a controlled study of neurotoxicity. Am J Otol 1988;9:302-5.
4.Matsuoka T, Nakatsuka H, Kobayashi N et al. Portal vein embolization for hepatoma with lipiodol-fibrin adhesive mixture. Nippon Igaku Hoshasen Gakkai Zasshi 1984 25;44:1411-3.
5.Oz MC, Williams MR, Moscarelli R et al. Laser bonding of secondary bronchi with solvent--detergent-treated cryoprecipitate. J Clin Laser Med Surg 1992;10:105-7.
6.Cuschieri A, Crosthwaite G, Shimi S, Pietrabissa, Joypaul V, Tait I, Naziri W. Hepatic cryotherapy for liver tumours: development and clinical evaluation of a high-efficiency multineedle probe system for open and laparoscopic use. Surg Endosc 1995; 9: 483-9.
7.Waldron B, Frank T, Cuschieri A. Percutaneous endoscopic external ring hernioplasty using Dundee inguinal canal retractor. Sem Lap Surg 1998; 5: 253-8.
8.Chen RJ, Fang JF, Lin BC et al. Selective application of laparoscopy and fibrin glue in the failure of nonoperative management of blunt hepatic trauma. J Trauma 1998;44:691-5.
9.Wadstrom J, Wik O. Fibrin glue (Tisseel) added with sodium hyaluronate in microvascular anastomosing. Scand J Plast Reconstr Surg Hand Surg 1993;27:257-61.
10.Nomori H, Horio H, Morinaga S, Suemasu K. Gelatin-resorcinol-formaldehyde-glutaraldehyde glue for sealing pulmonary air leaks during thoracoscopic operation. Ann Thorac Surg 1999;67:212-6.
11.Bingley JA, Gardner MA, Stafford EG, et al. Late complications of tissue glues in aortic surgery. Ann Thorac Surg 2000 Jun;69:1764-8.
12.Portoghese M, Acar C, Jebara V et al. Changes in the vascular wall induced by surgical glues. Experimental study. Presse Med 1992;21:1154-6.
13.Sung HW, Huang DM, Chang WH, et al. Gelatin-derived bioadhesives for closing skin wounds: an in vivo study. J Biomater Sci Polym Ed 1999;10:751-71.
14.Otani Y, Tabata Y, Ikada Y. Hemostatic capability of rapidly curable glues from gelatin, poly(L-glutamic acid), and carbodiimide. Biomaterials 1998;19:2091-8.
15.Basu S, Marini CP, Bauman FG et al. Comparative study of biological glues: cryoprecipitate glue, two-component fibrin sealant, and "French" glue. Ann Thorac Surg 1995;60:1255-62.
16.Nakayama Y, Matsuda T. Photocurable surgical tissue adhesive glues composed of photoreactive gelatin and poly(ethylene glycol) diacrylate. J Biomed Mater Res 1999;48:511-21
17.Thumwanit V, Kedjarune U. Cytotoxicity of polymerized commercial cyanoacrylate adhesive on cultured human oral fibroblasts. Aust Dent J 1999;44:248-52.
18.Ciapetti G, Stea S, Cenni E et al. Cytotoxicity testing of cyanoacrylates using direct contact assay on cell cultures. Biomaterials 1994;15:63-7.
19.Brothers MF, Kaufmann JC, Fox AJ, Deveikis JP. n-Butyl 2-cyanoacrylate--substitute for IBCA in interventional neuroradiology: histopathologic and polymerization time studies. AJNR Am J Neuroradiol 1989;10:777-86.
20.Inaspettato G, Rodella L, Laterza E, et al. Endoscopic treatment of bronchopleural fistulas using N-butyl-2-cyanoacrylate. Surg Laparosc Endosc 1994;4:62-4.
21.Goldman ML, Philip PK, Sarrafizadeh MS et al. Intra-arterial tissue adhesive for medical splenectomy in humans. Radiology 1981;140:341-9.
22.Koukoubis TD, Glisson RR, Feagin JA et al. Augmentation of meniscal repairs with cyanoacrylate glue. J Biomed Mater Res 1995;29:715-20.
23.Powell ES, Trail IA, Noble J. Non-suture repair of tendons. J Biomed Eng 1989;11:215-8.
24.Citrin P, Doscher W, Wise L, Margolis IB. Control of needle hole bleeding with ethyl-cyanoacrylate glue (Krazy Glue). J Vasc Surg 1985;2:488-90.
25.Jourdan IC, Bailey ME. Initial experience with the use of n-butyl 2-cyanoacrylate glue for the fixation of polypropylene mesh in laparoscopic hernia repair. Surg Laparosc Endosc 1998;8:291-3.
26.Vanholder R, Misotten A, Roels H, Matton G. Cyanoacrylate tissue adhesive for closing skin wounds: a double blind randomized comparison with sutures. Biomaterials 1993;14:737-42.
Taking it to the Next Level:
Advanced Laparoscopic Techniques
SESSION III: NEW TECHNIQUES
9. Endoscopic Thyroidectomy
New York, NY
In the late 19th century, Theodor Kocher developed the technique of thyroidectomy, an operation that requires a transverse cervical incision and the creation of myocutaneous flaps to gain access to the thyroid. Although conventional thyroidectomy can be performed safely with little morbidity, this approach leaves an undesirable scar on the anterior surface of the neck and may decrease neck mobility in the immediate postoperative period. Until recently, there had been few major refinements in the surgical approach to the thyroid gland. However, with the advent of laparoscopic surgery in the latter part of the 20th century, endoscopic techniques have been developed for surgery of the neck1-5. Endoscopic thyroidectomy is a new minimally invasive technique that permits thyroid excision without the potentially unfavorable cosmetic results of conventional thyroidectomy.
Patient Selection
Thyroidectomy is one of the most common operations performed throughout the world. Since the neck offers a limited working space for endoscopic maneuvering, multiple factors play a role in patient selection, including the type of pathology, thyroid size, and body habitus.
The most common indication for endoscopic thyroidectomy is the presence of a solitary, nonfunctioning thyroid nodule. The preoperative evaluation should include a detailed history and physical examination, indirect laryngoscopy, and thyroid function tests. Ultrasonography, which I routinely perform on all patients undergoing thyroidectomy, provides useful information on the size and location of thyroid pathology and permits evaluation of the contralateral thyroid lobe. Fine needle aspiration (FNA) plays an integral role in patient selection for endoscopic thyroidectomy. Patients in whom FNA demonstrates indeterminant cytology are excellent candidates for an endoscopic approach, as surgery can be both diagnostic and therapeutic. Patients with a preoperative diagnosis of carcinoma are not offered an endoscopic approach, as endoscopic thyroidectomy may lead to inadequate staging. Currently, patients with atypical or suspicious cytology are not appropriate candidates, since there is a higher incidence of malignancy in nodules with atypical features. Furthermore, the effects of CO2 insufflation on malignant disease have not been investigated in endoscopic neck surgery. As more experience as gained, it may be possible to broaden patient selection to include patients with thyroid carcinoma. With follicular or Hürthle cell neoplasms, patients are thoroughly counseled about the possibility of open completion thyroidectomy if frozen section analysis or final pathology yields a diagnosis of carcinoma. Other indications for endoscopic thyroidectomy include a solitary toxic nodule, recurrent thyroid cysts, and small multinodular goiters.
There are numerous exclusion criteria for endoscopic thyroidectomy. Since the neck offers a limited working space, endoscopic thyroidectomy should not be attempted in patients with thyroid nodules great than 3cm. Other contraindications include a large multinodular goiter, a history of prior neck surgery, and morbid obesity, as a short, wide neck will compromise maneuverability. In Graves' disease, the thyroid gland can be markedly enlarged and highly vascular, which may obscure endoscopic exposure. Finally, elderly patients with co-morbid medical problems may not tolerate CO2 insufflation as well as younger patients; therefore, conventional open thyroidectomy is best recommended for this patient population.
Surgical Technique
After inducing general endotracheal anesthesia, the patient is placed on the operating table in the supine position with the neck slightly extended. Anatomical landmarks are outlined with a marking pen, including the sternal notch, the midline, the anterior border of the sternoclediomastoid muscle (SCM), and the external jugular veins. The head is slightly rotated to maximize access to the ipsilateral neck.
A 0.5cm incision is made at the sternal notch and the cervical fascia is opened under direct vision. The subplatsymal space is developed along the anterior border of the ipsilateral SCM. A purse-string suture is placed in the subcutaneous tissue and a 5mm trocar is inserted. Carbon dioxide is insufflated to a pressure of 12 mmHg, but once an adequate working space has been developed, the insufflation pressures are decreased to 8-10 mmHg. A 0 degree 5mm endoscope is used to perform the initial dissection by gently advancing the endoscope along the anteromedial border of the ipsilateral sternocleidomastoid muscle, a technique that allows development of an avascular space.
Once an adequate working space has been developed, the camera is changed to a 30 degree 5mm endoscope (or alternatively, a 45 degree 5mm endoscope). Additional trocars are inserted under direct vision including a 2mm trocar at the midline above the cricoid cartilage, a 2mm trocar at the midportion of the ispilateral SCM, and a 5-10mm trocar superolaterally along the anterior border of the SCM. The latter trocar site, which is located in a cosmetically favorable location, is used to extract the specimen at the conclusion of the operation.
The carotid artery is identified and the space between the lateral border of the strap muscles and the medial edge of the carotid artery is developed. The strap muscles are not divided but rather retracted anteromedially. The thyroid lobe is visualized and mobilized using a combination of sharp and blunt dissection. Cautery is never used in the deeper tissue planes. The middle thyroid vein is ligated using 5mm clips or the 5mm ultrasonic scalpel. Using blunt dissection, the recurrent laryngeal nerve and parathyroid glands are identified and carefully dissected free of their attachments to the thyroid gland. The inferior thyroid artery is mobilized at its junction with the recurrent laryngeal nerve and ligated with the 5mm clip applier. With the recurrent laryngeal nerve in full view, the inferior thyroid artery is divided.
The superior pole vessels are isolated, clipped, and divided. The superior laryngeal nerve is identified and protected during this maneuver. In a similar manner, the inferior pole vessels are clipped and divided. Alternatively, the ultrasonic scalpel can be used to divide the inferior thyroid vasculature. After releasing the anteriomedial attachments of the recurrent laryngeal nerve by blunt dissection, the 5mm ultrasonic scalpel is used to divide the ligament of Berry. The recurrent laryngeal nerve must be in full view during this maneuver. Finally, the isthmus is divided using the ultrasonic scalpel.
A small sac is constructed by removing the thumb portion of a surgical glove and a purse string suture is fashioned. The specimen is placed in the sac and extracted through the superolateral trocar site. Steri strips are used to close the incisions.
Mount Sinai Experience
From October 1998 to Dec 2000, 24 patients with a solitary or dominant thyroid nodule underwent endoscopic thyroidectomy at the Mount Sinai Medical Center. Informed consent was obtained from all patients.
There were 22 females and 2 males with a mean age of 43 years (17-66 years). Cervical ultrasound was performed in all patients to precisely define the dimensions and location of the thyroid nodule and to exclude the presence of nonpalpable contralateral disease. Indications for surgery included follicular neoplasm (n=12), indeterminate cytology (n=8), recurrent thyroid cyst (n=2), Hürthle cell neoplasm (n=1), and toxic thyroid nodule (n=1). The mean nodule diameter was 2.7 cm (0.6-7cm). Operative procedures included left thyroid lobectomy (n=11), left subtotal thyroidectomy (n=2), right thyroid lobectomy (n=5), right subtotal thyroidectomy (n=2), and isthmusectomy (n=4).
Twenty-one of 24 cases were successfully completed endoscopically. In 1 patient with a large 7cm cyst, the procedure was completed through a small ipsilateral incision. The mean operating time was 213 minutes (100-330 min). There were no major complications, but 3 patients developed mild hypercarbia and 1 patient had an incidental parathyroidectomy. Final pathology yielded a diagnosis of follicular adenoma (n=14), Hürthle cell adenoma (n=1), oncocytic adenoma (n=1), thyroid cyst (n=2), multinodular goiter (n=3), and papillary thyroid carcinoma (n=3). Two of the patients with papillary carcinoma underwent open completion thyroidectomy without evidence of residual cancer. The third patient had only a small focus of papillary carcinoma and refused completion thyroidectomy. When compared to conventional thyroidectomy, patients undergoing endoscopic thyroidectomy had a significantly superior cosmetic result (p<0.005) and a quicker return to normal activity (p<0.05), but there was no difference in analgesic requirement.
Discussion
Proper patient selection plays an important role in endoscopic thyroidectomy. Patients with a long, narrow neck and a solitary thyroid nodule less than 3cm in diameter are well suited for endoscopic thyroidectomy. The use of an energy source (ie cautery, ultrasonic scalpel) is to be avoided in the deeper tissue planes, especially in proximity to the recurrent larygneal nerve. If the surgeon does not unequivocally identify the recurrent layrgneal nerve and parathyroid glands, the operation should be converted to an open, conventional approach. We routinely submit the specimen for frozen section analysis. If a diagnosis of thyroid carcinoma is established, an open completion thyoidectomy is performed.
Although conventional open thyroidectomy can be performed with few complications, this approach leaves a visible scar on the anterior surface of the neck in a cosmetically unfavorable location. The endoscopic approach provides a superior cosmetic result when compared to conventional thyroidectomy and results in a quicker return to normal activity. Endoscopic thyroidectomy provides fantastic magnification of thyroid anatomy, including the recurrent laryngeal nerve, superior laryngeal nerve, and the parathyroid glands. In addition, since muscle is not divided during endoscopic thyroidectomy, there is less tissue trauma, resulting in a quicker return to normal activity5. The long duration of surgery is the primary disadvantage of endoscopic thyroidectomy, however, this should decrease as additional experience is gained. The results of our initial experience support the use of endoscopic thyroidectomy in carefully selected patients. Prior experience with endoscopic parathyroidectomy is recommended.
References
1. Hüscher CSG, Chiodini S, Napolitano C, Recher A: Endoscopic right thyroid lobectomy. Surg Endosc 11:877, 1997
2. Yeung GHC. Endoscopic surgery of the neck. Surg Laparosc Endosc 1998; 8:227-232.
3. Bellantone R, Lombardi CP, Raffaelli M, et al. Minimally invasive, totally gasless video-assisted thyroid lobectomy. Am J Surg 177:342-3, 1999
4. Shimizu K, Akira S, Jasmi AY, et al. Video-assisted neck surgery: endoscopic resection of thyroid tumors with a very minimal neck wound. J Am Coll Surg 188:697-703, 1999
5. Inabnet WB, Gagner M. How I do it: endoscopic thyroidectomy. J Otolaryngol 2001; 30:1-2.
Leon C. Hirsch Professor of Clinical Surgery
Weill Cornell Medical College
Director, Minimal Access Surgery Center
New York Presbyterian Hospital
525 E. 68th St., Rm. F-763, Box 203
New York, NY 10021
Surgery on any segment of the gastrointestinal tract often requires the creation of an anastomosis. In open surgery that anastomosis is routinely completed with either suturing or stapling techniques. Both techniques have been standardized for many years. During laparoscopic surgery on the gastrointestinal tract, a similar option of either suturing or stapling exists. However, an additional decision must be made as to whether to complete the anastomosis intracorporeally or extracorporeally.
After laparoscopic resection of a portion of the gastrointestinal tract, most surgeons use a minilaparotomy for extraction of the specimen. Creation of the anastomosis through this minilaparotomy with open surgical technique is appropriate in most cases because it is faster, easier, and does not seem to alter the outcome compared to completing the anastomosis intracorporeally. However, there are circumstances when it is desirable to create the anastomosis intracorporeally, even when there is a minilaparotomy for specimen removal. The most common reason why the minilaparotomy could not or should not be used to create an anastomosis extracorporeally is that the ends of the bowel will not reach through the minilaparotomy. This may be due to the location of the minilaparotomy or may be due to anatomic factors such as a short or obese mesentery. In these situations, intracorporeal completion of the anastomosis is necessary.
When the anastomosis is to be created without a resection, there is little, if any, justification for not completing the anastomosis intracorporeally. This can be accomplished with any of several techniques. Just as in open surgery, surgeons can complete the anastomosis with suturing techniques, stapling techniques, or a combination of the two. Suturing techniques are feasible but require a great deal of skill with laparoscopic suturing and tying. Stapling techniques can use either the endoscopic linear cutting stapler or a circular stapler. Alternatively, the linear stapler can be used for part of the anastomosis and the remainder can be sutured. Following are brief descriptions of these techniques and lists of examples of appropriate procedures for each anastomotic technique.
Theoretically any intraabdominal anastomosis can be sutured laparoscopically. Single layer or double layer suturing is appropriate depending on the actual anastomosis and the skill and experience of the surgeon. Many surgeons are now performing a sutured gastrojejunostomy during laparoscopic gastric bypass. Choledochoduodenostomy or choledochojejunostomy are other examples where sutured anastomoses are either necessary or desirable. The actual techniques simply require an appropriate setup to orient the scope, ports, and anastomosis in a line with the surgeon and the monitor.
There are two basic types of stapling techniques that can be used. One uses the circular stapler and the other uses the linear cutting stapler. Use of the circular stapler requires access either through an orifice (such as the mouth or anus) or through an enlarged port site in the abdominal wall. Either the stapler or the anvil of the stapler is then passed into the lumen of the gastrointestinal tract through an orifice or an enterotomy. Examples of this technique include esophagogastrostomy after esophagectomy, gastrojejunostomy as a part of gastric bypass for severe obesity, Billroth I anastomosis after antrectomy, and colorectal anastomosis after sigmoid resection or low anterior resection of the rectum.
The other technique uses the linear cutting stapler to create a side to side anastomosis. Appropriate procedures for this anastomotic technique include gastrojejunostomy (Fig. 1), Billroth II anastomosis after antrectomy, and duodenojejunostomy. A functional end-to-end technique is very similar to the side-to-side technique and is commonly used for jejunojejunostomy (either as a part of a Roux-en-Y or after an enterectomy) and ileocolostomy. In every case there is a defect in the anastomosis where the stapler entered the lumen of the bowel. This can be closed with laparoscopic stapling techniques using either the linear cutting stapler or the linear noncutting stapler. Alternatively, this remaining defect can be sutured with laparoscopic sututing techniques.
References
1.Fowler, D.L. Palliative surgery in upper digestive cancer: laparoscopic gastroenterostomy. In Minimally Invasive Surgery in Gastrointestinal Cancer. Ed. Cuesta M.A. and Nagy, A.G. Churchill Livingstone, pp. 123-30, 1993.
2.Uddo, J.F. Laparoscopic right henmicolectomy with intracorporeal anastomosis. In Atlas of Laparoscopic Surgery. Ed. Ballantyne, G.H. W. B. Saunders Company, pp. 325-41, 2000.
3.Brune, I.B. Gastroenterostomy. In Endosurgery. Ed. Toouli, J, Gossot, D., Hunter, J.G. Churchill Livingstone, New York, pp. 391-6, 1996.
11. Solid Organ Resection
(Liver, Partial Splenectomy, Pancreas)
Laparoscopic Liver Resection
Extensive experience in advanced laparoscopic and hepatobiliary surgery is mandatory prior to embarking on laparoscopic liver surgery. Even among advanced laparoscopic surgeons, laparoscopic liver surgery has remained relatively uncharted territory, despite recent advances in laparoscopic instrumentation and techniques.
Several series of laparoscopic liver resection for primary and secondary cancers have been reported in both cirrhotic and noncirrhotic livers. Laparoscopic management of patients with intra-abdominal cancer is controversial, and the liver is no exception. Accepted indications for laparoscopic liver resection include: solitary nonparasitic cysts in any location (although resection versus cyst fenestration is controversial), and benign solid tumors and hydatid cysts in anterolateral Couinaud segments 2, 3, 4b, 5 and 6. Experience with hydatid cysts limited in North America, and laparoscopic excision should be attempted only by surgeons experienced with their management.
The patient is positioned in the inverted-Y position with the surgeon standing between the legs, and the assistants to the patient's sides. The videolaparoscope is placed at the umbilicus, two 10-mm ports are placed around the umbilicus in the usual triangulated fashion, and a subxiphoid trocar is used for a liver retractor or the irrigation and suction device. This set-up is sufficient for a simple wedge resection, but for larger segmental resections, two additional trocars are recommended to enable two surgeons to work simultaneously ("four-handed" technique).
As in open liver surgery, the initial steps of resection are mobilization and vascular control. The hepatoduodenal ligament is dissected and a tourniquet is placed around the porta hepatis to allow for a Pringle maneuver (Figure 1). For a left lateral segmentectomy, the falciform and left triangular ligaments are divided until the suprahepatic inferior vena cava (IVC) is identified. The liver is retracted inferiorly and the junction of the left hepatic vein and the IVC is carefully exposed with gentle blunt dissection. The hepatic vein is then ligated but not divided by passing a silk suture around it and tying it intracorporeally (Figure 2).
The first surgeon fractures the liver parenchyma, exposing the bile ducts and vessels for the second surgeon to control. The laparoscopic ultrasonic surgical aspirator (CUSA) allows for meticulous division of hepatic parenchyma and exposure of the vascular and biliary pedicles. Small vessels and biliary radicals may be divided and sealed with ultrasonic shears, which can alternatively be used to divide the parenchyma directly, without prior fracture. Larger vessels may require clips, and linear cutting vascular staplers are recommended for division of the hepatic veins and large segmental vascular and biliary pedicles (Figure 2). A simpler technique, although tedious, is to use the ultrasonic shears for the entire dissection of the liver parenchyma, including small vessels and bile ducts.
Raw surface bleeding and bile leakage can be controlled by a variety of techniques. Laparoscopic argon beam coagulation provides excellent hemostasis for raw, oozing surfaces. However, special precautions must be taken when using the argon beam laparoscopically, as deaths from argon gas emboli have occurred. Published safety recommendations advise using alarmed monitoring of intraabdominal pressure, leaving one port vent open, and limiting argon flow to a maximum of 4 liters per minute in order to prevent intra-abdominal over-pressurization and subsequent argon gas embolism.
Once the raw surface of the liver is relatively dry, wide application of fibrin sealant may improve hemostasis and prevent bile leaks.
Laparoscopic Partial Splenectomy
A decade has passed since the first laparoscopic splenectomy was reported, and it is now generally accepted as the procedure of choice for routine splenectomies, such as those for Idiopathic Thrombocytopenic Purpura (ITP). As surgeon experience and technology advance, more challenging operations have become feasible laparoscopically.
Partial splenectomy is performed infrequently, by either the open or laparoscopic approach. Preservation of splenic tissue and function is favored over splenectomy whenever possible, in order to decrease the risk of post-splenectomy sepsis. The indications for partial splenectomy include trauma, hamartoma, and diagnosis and resection of nonparasitic cysts or other splenic tumors. It has also been described in the management of type I Gaucher disease, cholestreyl ester storage disease, chronic myelogenous leukemia, thalassemia major, and for staging Hodgkin disease.
Several cases of laparoscopic partial splenectomy have recently been reported in the literature. Patient positioning and operative field set-up is identical to that for laparscopic splenectomy. The patient is in the right lateral decubitus position, and the trocars are placed in the left upper quadrant, along the costal margin. The segment of spleen to be excised is first devascularized by either clipping and cutting, or using bipolar electrocautery or ultrasonic coagulating shears.
While some surgeons have successfully carried out laparoscopic splenectomy using monopolar electrocautery to transect devascularized splenic tissue the ultrasonic shears may facilitate division of the splenic parenchyma. Laparoscopic linear cutting staplers have been successfully employed for splenic transection in a large animal model, but since the porcine spleen is much thinner than the human spleen, the clinical utility of this technique is unexplored. A series of open partial splenectomies has been reported using a transverse stapler, but the open staplers open much wider than the laparoscopic counterparts.
The splenic specimen is removed in an impermeable bag, and the raw surface of the remaining spleen is inspected for hemostasis. Bleeding from the cut edge of the transected solid organ can be controlled with a variety of hemostatic techniques as described above for laparoscopic liver resections.
Laparoscopic Pancreatic Resection
The number of laparoscopic pancreatic resections reported in the surgical literature has been remarkably low. However, early reports of laparoscopic distal pancreatectomy and enucleation of islet cell tumors suggest that patients may benefit from the laparoscopic technique. The postoperative recovery appears to be shorter than open surgery, with no apparent increase in complications compared.
Michel Gagner first described laparoscopic pancreaticoduodenectomy in 1992, and he concluded from his early results that there did not appear to be a benefit over the open operation. A hand-assisted laparoscopic pancreaticoduodenectomy, a hybrid between laparoscopic and open surgery, is a recent minimally invasive alternative. These procedures should be done only in the setting of prospective research protocols.
The indications for laparoscopic pancreatic resection (LPR) include inflammatory conditions such as chronic pancreatitis and pancreatic pseudocysts, and benign or premalignant solid or cystic lesions that can be surgically treated with distal pancreatectomy or islet cell tumor enucleation. The most common indications for LPR are benign cystic lesions (primarily cystadenomas) and islet cell tumors (primarily insulinomas).
Distal pancreatectomy with splenic preservation should be considered only if the benign pathology is not adherent to the splenic artery or vein. Results of open series are conflicting with regard to the post-operative morbidity of spleen-preserving procedures. Splenic preservation may be a reasonable option in selected patients at low risk for pancreatic fistula, such as patients with a firm pancreas from chronic pancreatitis.
Laparoscopic surgery of the body or tail of the pancreas is performed via a four- or five-trocar approach. The patient is placed in either the inverted-Y or the semi-lateral decubitus position. The pancreas is exposed in the lesser sac through a wide opening in the gastrocolic ligament, preserving the gastroepiploic artery. The splenic flexure of the colon must be well mobilized, and whether or not the spleen is preserved, it is important that the short gastric vessels be divided. This entire dissection is performed using the ultrasonic coagulating shears and is facilitated by superior retraction of the stomach and lateral counter-traction of the gastrocolic ligament.
Congenital or acquired adhesions between the posterior surface of the stomach and the anterior surface of the pancreas are sharply divided. Adherence of the pancreas to either the stomach or colon can indicate malignant disease, and should prompt consideration of conversion to open surgery. Once the body and tail of the pancreas are clearly visualized in the retroperitoneum, intraoperative laparoscopic ultrasonography of the pancreas and liver are performed.
Ultrasound is useful for localizing pancreatic lesions and their relationship to the pancreatic duct and planned site of transection (aim for a 2-cm gross margin). A 75% distal pancreatectomy is commonly performed, in which the pancreatic transection site overlies the superior mesenteric vein. The liver can also be evaluated sonographically for metastatic disease.
The posterior peritoneum is incised along the inferior and superior borders of the body and tail of the pancreas. The pancreas is then freed using sharp and blunt dissection from the posterior abdominal wall in this relatively avascular retroperitoneal plane. Dissection is continued medial to the inferior mesenteric vein (IMV), which may be preserved or divided depending on the planned resection and the patient's anatomy.
The splenic artery is identified in its tortuous course along the superior border of the pancreas. The peritoneum overlying the splenic artery is opened, and a right-angled dissector is used to complete the isolation of the artery. The posterior aspect of the pancreas is further mobilized to ascertain involvement of the splenic vessels by the lesion. The decision whether to excise or preserve the spleen must be made before proceeding any further with the operation.
Laparoscopic Distal Pancreatectomy with Splenectomy
The splenic artery is divided at the planned line of pancreatic transection, using either clips reinforced with pre-tied ligatures, or an endoscopic stapler (2.5 mm by 45 mm) reinforced with clips. The pancreas and splenic vein are divided as a single unit with an endoscopic stapler. Ligation of the splenic artery and short gastric vessels significantly decreases arterial inflow to the spleen, which facilitates its mobilization from body to tail, in retrograde fashion (Figure 3).
Alternately, the spleen and pancreas may be mobilized from lateral to medial prior to division of the splenic vessels and the pancreas. Ligamentous attachments of the spleen are divided, and the posterior spleen and tail of the pancreas are mobilized from the retroperitoneum, working dissecting from lateral to medial, in an antegrade direction along the pancreas (Figure 4). These two approaches are analogous to the "hanging spleen" (vessels taken early) versus "suspended pedicle" (attachments taken early) approaches to laparoscopic splenectomy, and choice depends on surgeon preference and location of the lesion. The latter technique is almost identical to the dissection for a laparoscopic left adrenalectomy.
Laparoscopic Distal Pancreatectomy with Splenic Salvage
The initial exposure and retroperitoneal dissection are the same as in pancreatectomy with splenectomy. The peritoneum on the tail of the pancreas is gently grasped with 5-mm atraumatic forceps, and retracted anteriorly and inferiorly to expose the transverse branches of the splenic artery and vein. These vessels are divided with the ultrasonic shears or 5-mm titanium clips until the desired length of pancreas has been attained. Mobilization of the entire pancreatic tail and body, up to the portal vein, can be achieved in this manner. The pancreas is transected with a vascular endoscopic linear stapler.
Laparoscopic Enucleation of Islet Cell Tumors
Enucleation of a benign islet cell tumors should be performed only if a clear margin of normal pancreatic tissue exists between the tumor and the pancreatic duct and adjacent major vessels. Otherwise, a formal pancreatic resection such as distal pancreatectomy is advised. Resection of a tumor in close proximity to the pancreatic duct carries a high risk of ductal injury and pancreatic leak.
The most time consuming step of the operation is often localization of the tumor. Preoperative studies can be helpful, but regardless of their findings, laparoscopic intraoperative ultrasound (LIOUS) is routinely used to confirm the location and relationship to the pancreatic duct and peripancreatic vessels. LIOUS may also help to differentiate benign from malignant tumors. A distinct, clear margin between the tumor and pancreatic parenchyma is readily identified in benign tumors, whereas malignant insulinomas may have a more irregular or unclear margin suggesting local invasion.
Once the islet cell tumor has been localized, the dissection is carried between the tumor and the normal parenchyma using an electrocautery hook or ultrasonic coagulating shears. The pancreatic vessels feeding the tumor are ligated with medium to large titanium clips. Laparoscopic suturing skills are required to do these procedures, as suture ligatures may be required for hemostasis. Once the tumor is completely enucleated, it is extracted in a small sterile bag through the 12-mm port. A closed suction drain is placed in the lesser sac at the enucleation site.
Selected References
1. Poulin EC, Thibault C, DesCoteaux JG, Cote G. Partial laparoscopic splenectomy for trauma: technique and case report. Surg.Laparosc.Endosc. 1995; 5:306-310.
2. Seshadri PA, Poenaru D, Park A. Laparoscopic splenic cystectomy: a case report. J.Pediatr.Surg. 1998; 33:1439-1440.
3. Katkhouda N, Mavor E. Laparoscopic management of benign liver disease. Surg.Clin.North Am.2000.Aug.;80.(4.):1203.-11. 80:1203-1211.
4. Katkhouda N, Hurwitz M, Gugenheim J, et al. Laparoscopic management of benign solid and cystic lesions of the liver. Ann.Surg. 1999; 229:460-466.
5. Patterson EJ, Gagner M, Salky B, et al. Laparoscopic pancreatic resection: single-institution experience of 19 patients. 2001; (UnPub)
6. Patterson EJ, Salky B. Laparoscopic pancreatic resections. In Cameron JL., ed. Current Surgical Therapy. St. Louis: Mosby, 2000;
Figure Legends
12. Endoscopic Esophagectomy
Clinical Professor of Surgery, Oregon Health Sciences University
Director, Dept of Minimally Invasive Surgery Legacy Health System
Correspondence: LL Swanstrom MD, 501 N. Graham, Ste 120, Portland, OR 97227
phone: (503) 288-6167; fax: (503) 288-3437
Endoscopic Esophagectomy:
Total esophagectomy is indicated for end stage degenerative disorders of the esophagus (undilatable strictures, sigmoid dilatation of the esophagus, severe motility disorders), high grade Barrett's dysplasia and for cancer. Esophagectomy represents one of the greatest surgical insults to patients as it involves extensive dissection, involves multiple body cavities, impacts several organ systems and is usually performed on nutritionally compromised patients.
Laparoscopic esophagectomy is a broad term that includes a variety of approaches which are intended to minimize the physiologic impact of esophageal resections.
Gerhard Buess and associates first described an endoscopic assisted technique in 1990. This technique relied on a specially designed video mediastinascope which allowed dissection of the upper esophagus under direct vision via the sternal notch. Distal mobilization, gastric division and reconstruction was done by open means. This technique never was widely adopted because of the cost of the instrumentation and the difficulties in operating through the narrow sternal notch. The fact that a major laparotomy was required to complete the procedure was also a determent.
Several authors have described using a thoracoscopic approach where the esophageal dissection and mediastinal lymph node dissection is performed under videoscopic control. The patient is subsequently repositioned supine and an open laparotomy is used to perform the gastric mobilization, placement of feeding tubes and other adjuvant procedures. The primary argument for this approach was that it spared the patient the most morbid element of the traditional Ivor Lewis approach; the thoracotomy incision. Only anecdotal reports with one or two cases have been published with this approach and they describe substantial hospital stays and no long-term follow-up. It is therefore difficult to judge if any benefit is derived from this approach.
Another approach has been to perform a staging laparoscopy and, if the tumor is resectable, to mobilize the stomach and perform gastric division laparoscopically. Jagot then describes performing a thoracotomy for an en block mediastinal resection with subsequent thoracic or cervical anastamosis. A final variation involves making a subzyphoid incision following a staging laparoscopy and gastric dissection to allow a hand to be inserted to perform a standard "blind" transhiatal blunt esophagectomy.
There are several potential patient benefits to be gained from minimally invasive esophagectomy. These include: Decreased surgical stress response (with subsequent improvement in healing and cancer cure rates), avoidance of an ICU admission, more rapid hospital discharge, less pain and fewer complications related to the access. To maximize these advantages it is probably desirable to perform the esophagectomy totally endoscopically. Two techniques have been described which replicate the standard open approaches of either an Ivor Lewis technique or transhiatal blunt esophagectomy.
Thoracoscopic/laparoscopic technique
A combination laparoscopic and thoracoscopic technique was first described by Luketich in 1997. This technique involves standard right thoracoscopy with the patient in left lateral decubitus position and with 2-lumen endotracheal ventilation. Four thoracoports are used to mobilize the mediastinal esophagus and perform a mediastinal node dissection. The azygos vein is typically divided with an endoscopic vascular stapler and ultrasonic coagulating shears are used to divide the vascular attachments while retracting the esophagus with an encircling Penrose drain. The patient is then turned supine and laparoscopy is performed with a 5 port technique. The stomach and duodenum are mobilized, the gastric tube is tailored using multiple firings of the endoscopic stapling device and a jejunal feeding tube is placed. If desired, a pylormyotomy can be performed as well.
Luketich described better outcomes for the thoracoscopic/laparoscopic approach when compared to the laparoscopic transhiatal technique. The average OR times were 7.25 hours vs. 7.75 for the laparoscopic, 0 vs. 2 days in the ICU and 17 vs 4 days hospital stay. Because of this, this group favors the combination laparoscopic and thoracoscopic approach.
Laparoscopic transhiatal approach
At institutions where an open transhiatal esophagectomy is the standard practice a laparoscopic transhiatal approach is preferred. This procedure is an extension of techniques developed for advanced antireflux surgery such as the Collis procedure which required extensive esophageal mobilization. First described by DePaula, this technique uses a 5 port approach with the trocars placed in the same pattern but somewhat lower in the abdomen than for an antireflux procedure. Extra long scopes and instruments are needed for the mediastinal dissection. The duodenum is Kocherized and the lesser curvature mobilize using the ultrasonic coagulating shears. Likewise, the greater curvature is freed up by dividing the short gastric vessels and gastro-colic omentum being careful to preserve the marginal gastro-epiploic artery and vein. Retrogastric and retroduodenal attachments are completely divided. The stomach is then divided to create a narrow gastric tube using multiple firings of the endoscopic linear stapler. The esophageal hiatus is then opened by dividing the phrenoesophageal ligament in the dissection of the distal esophagus is carried out under direct vision. Proximal dissection of the esophagus is accomplished by inserting small retractors and the scope inside the mediastinum. This dissection is continued until the carina is reached. A left cervical incision is made and the esophagus surrounded with a Penrose. The proximal mediastinal esophagus can be mobilized with sharp and blunt dissection until it connects with the mediastinal dissection plane. The proximal gastric tube is attached to the divided esophagus and the specimen removed through the cervical incision. Cancer specimens should be placed in a tissue retrieval sack before removal to prevent tumor seeding of the wound. A standard anastamosis is performed in the neck under direct vision. A laparoscopic feeding jejunostomy is placed and, if desired, a pyloromyotomy is performed.
DePaula's original series of 12 patients, which included both benign indications and cancer, showed amazing results, with operative times of 256 minutes and a relatively short length of stay (7.6 days). Complications were noted in 25% and were mostly minor and transient. Our experience has shown this approach to be much longer than the open (6.7 hours vs. 4.8) and the hospital stay to be only slightly shortened (6.8 vs. 9). On the other hand, only 20% of patients required admission to the ICU and the average return to normal activity post discharge was remarkably short, generally 2 - 3 weeks. One patient has been converted to open because of dense retrogastric adhesions and one patient had a recurrent laryngeal nerve injury requiring PTFE injection of the vocal cord.
Discussion
The majority of esophagectomies are done for esophageal cancer and the majority of these procedures will not result in cure because of gross or occult metastatic disease at the time of surgery. It has been shown, however, that esophagectomy can provide good palliation for these patients and, equally, quality of life is better in patients with a failed esophagus due to benign causes. The problem has been that standard surgery is tremendously morbid to these high risk patients due to pulmonary compromise, pain, prolonged recovery, wound complications and healing problems. Endoscopic esophagectomy offers the potential to minimize these problems but the endoscopic procedures are extremely difficult, requiring advance laparoscopic skills and long OR times.
The choice between the laparoscopic transhiatal or the thoracoscopic/laparoscopic approach is currently unresolved. The obvious drawback to the thoracoscopic approach is the need to reposition the patient for laparoscopy, the large numbers of access ports needed, the additional pain caused by the thoracoscopy and the added complexity of the anesthesia. On the other hand, the exposure offered by the thoracoscopy makes this part of the procedure very straight forward and permits the performance of a good en block mediastinal resection which some feel may improve cancer cure rates. Laparoscopic transhiatal resections have fewer trocars, less pain and probably less pulmonary compromise. On the other hand, the mediastinal dissection is tedious, difficult and requires special long instruments to perform.
The goal of providing a less morbid "cure" to patients with severe esophageal disease is a laudable one and one that laparoscopy may provide a partial answer to.
Selected readings
1. Swanstrom LL, Hansen PD. Laparoscopic total esophagectomy. Arch Surg 1997; 132:943-949.
2. Jagot P, Sauvanet L, Berthoux L, Belghiti J. Laparoscopic mobilization of the stomach for oesophageal cancer. BJS 1996; 83:540-542.
3. DePaula AL, Hashiba K, Ferreira EAB, Anania de Paula R, Grecco E. Laparoscopic transhiatal esophagectomy with esophagogastroplasty. Surg Lap & Endosc 1995; 5:1-5.
4. Watson DI, Davies N, Jamieson GG. Totally endoscopic Ivor Lewis esophagectomy. Surg Endosc 1999; 13:293-297.
5. Fernando HC, Christie NA, Luketich JD. Thoracoscopic and laparoscopic esophagectomy. Semin Thorac Cardiovasc Surg 2000; 12:195-200.
6. Krasna MJ, Mao YS, Sonett J, Gamliel Z. The role of thoracoscopic staging of esophageal cancer patients. Eur J Cardiothorac Surg 1999; 16:s31-s33.
7. Nguyen NT, Follette DM, Wolfe BM, Schneider PD, Roberts P, Goodnight JE. Comparison of minimally invasive esophagectomy with transthoracic and transhiatal esophagectomy. Arch Surg 2000; 135:920-925.