More than 20 years have passed since 1987, when the French surgeon Philippe Mouret first presented the technique of videolaparoscopic cholecystectomy (VLC),1 which subsequently became more widely used as a result of the contributions of Jaques Perissat and Francois Dubois.2 Videolaparoscopy immediately demonstrated its advantages: its less invasive use of small incisions led to less postoperative pain and a better aesthetic result; the enlarged operative field shown on a high-resolution screen was clearly visible to all of the members of the surgical team (unlike the open technique, which sometimes meant that only some members had an optimal view); it caused fewer postoperative intraperitoneal adherences; and it required a shorter period of hospitalization. However, it also had the disadvantages of the lack of tactile sensation and the third dimension of depth.

The initial learning phase was partially hindered by the fact that the instruments were still being developed, but this problem has since been overcome, and the procedure is now reliable and widely accepted, particularly by younger surgeons who have a more technologically refined mentality and whose hand-eye coordination is more practised in using this type of equipment. Nevertheless, the learning process naturally requires a gradual transition from exercises using special simulators to participation as third or second surgeon in order to acquire familiarity with the optics and instruments before undertaking responsibility for the operation under the support and guidance of an expert team (i.e., the learning curve).

Having established that videolaparoscopy is a technique rather than a discipline, it is worth noting that laparoscopic cholecystectomy has become the gold standard treatment of symptomatic gallstones, with the number of complications being inversely proportional to the experience of the operator. Nevertheless, some hospital departments still tend to use the open technique more frequently because of its historical reliability,3 which is above all associated with the classical figure of a traditional surgeon.

When learning how to perform a VLC, it can be helpful to use a laparascopic risk index4,5 that incorporates (A) experience, with scores ranging from 1 (i.e., no previous experience to 4 (i.e., optimal experience); (B) instruments, with scores raging from 1 (i.e., mediocre) to 3 (i.e., good or available); and (C) the patient's clinical status, with scores ranging from 1 (i.e., mildly symptomatic) to 5 (i.e., acute cholecystic symptoms 72 hours after onset). If the laparascopic risk index (defined as A × B ÷ C) is <1.5, a good outcome can be considered unlikely.

As mentioned in the introduction discussion about disadvantages, laparoscopy is characterized by the absence of direct tactile sensations, which makes careful anatomic preparation even more important,1,6,7 because the area of the gallbladder and the extrahepatic biliary system often presents anatomic anomalies and variants of the biliary pathways and vessels.8

There has been a decrease in the occurrence of one of the most feared complications of cholecystectomy (i.e., a lesioned main biliary pathway), but this has not been as great as it was logical to expect: although surgeons are now more experienced, the mean published incidences are still 0.60% to 0.75% in the case of laparoscopy, with variations ranging from 0.2% to 4%9,10 compared with 0.35% to 0.50% in the case of laparotomy.11–17 We have also found that biliary pathway lesions are more frequent when using the laparoscopic technique: 6 lesions (0.58% of 1037 patients) versus 1 lesion (0.51% of 198 patients).

In addition to the causes analyzed below in the Discussion section, the increased number of biliary pathway lesions is also due to the broadening of the indications for laparoscopic cholecystectomy18,19 and its use in complex and complicated cases, which seems to suggest the presence of an intrinsic risk.

As well as the lack of tactile sensation and two-dimensional vision, complications may also be due to technical problems (e.g., thermal insults with direct and indirect damage caused by monopolar energy, which are also dangerous because of the long-term effects of necrosis), equipment defects, and the scissors-like action of the clips.14,20,21 It is useful remember that clips should be the right size and should not be handled once they have been positioned. Nevertheless, most complications seem to be due to misperception rather than the result of a lack of skill or knowledge or errors in judgement.4,5,22

In order to ensure anatomic control, it is essential to identify the elements of Calot's triangle,23 which is formed by the common hepatic duct on the left, the cystic duct inferiorly, and the inferior surface of the liver superiorly (although the original description referred to the cystic artery as the upper side of the triangle). The cystic artery crosses Calot's triangle in 90% of cases and, typically, arrives there after having passed behind the common and right hepatic duct. There is now also a tendency to refer to a cystohepatic triangle,6 which is bordered medially by the common and right hepatic ducts, laterally by the cystic duct, and superiorly by the inferior edge of the liver. Many surgeons agree that, especially in the case of inflammation, when the attempt to dissect and identify these elements exceeds 20 to 30 minutes, it is better to switch to the laparotomic approach.20

Materials and Methods

We reviewed the data relating to 1037 consecutive patients who underwent VLC during the 17 years from 1993–2009 at 3 General Surgery Units with partially the same and similarly experienced teams in order to ensure that the results were as homogeneous as possible. For a period of 10 years (2000–2009), it was also possible to make a direct comparison between the open and laparoscopic techniques (n  =  950 cholecystectomies, of which n  =  752 were performed laparoscopically); more recently, the latter has entirely replaced the former. The patients included 66 patients (6.37%) whose gallstones were resolved by means of preoperative endoscopic retrograde cholangiopancreatography (ERCP), often with papillotomy and the endoscopic removal of stones; 379 (36.5%) who underwent laparoscopic cholecystectomy because of acute cholecystitis (36.5%); and 28 (2.7%) who underwent urgent surgery (i.e., within 24–48 hours of the acute episode).

In addition to the normal blood tests and instrumental examinations preceding general anesthesia, the patients underwent abdominal ultrasonography, with particular attention being given to the gallbladder, the biliary pathways, and the liver; selected cases (i.e., those with suspect formations and/or complex diseases) also underwent nuclear magnetic resonance cholangiography for purposes of verification. Patients with a history of jaundice or pancreatitis underwent preoperative ERCP (with endoscopic papillotomy and lithomomy, if necessary) in collaboration with an endoscopist and on the basis of precise protocol indications (i.e., common bile duct (BD) dilation; the presence of stones in the main biliary pathway; previous jaundice; and alterations in such parameters as bilirubin, liver enzymes, biliary stasis, and pancreatic enzymes). All of the patients received antibiotic prophylaxis with first-generation cephalosporin.

Technically, the operator stands between the patient's legs. After an initial period using a Veress needle, the first supraumbilical trocar is inserted after the creation of the pneumoperitoneum using the open technique (trocar with balloon and/or Hasson's method); the other 3 trocars are inserted under visual guidance.

Intraoperative choledoscopy and cholangiography were used in selected cases, mainly for organizational reasons, although the latter technique permits a complete study of the biliary pathway.24,25 The number of these procedures tended to decline with the increased use of nuclear magnetic resonance cholangiography, which we believe should be used in the case of any clinical or technologic doubt or suspicion concerning the biliary pathway.

Great care was taken to recover any stones or free clips from the abdomen because of the more theoretical than real possibility that they may cause granulomas or abscesses; in the case of a perforated gallbladder, the bladder was emptied and repeatedly washed with saline solution. With some rare exceptions, the procedure finished with the placement of subhepatic drainage (to be removed after 24 hours if nothing was collected); this has been a routine clinical procedure for some years. In general, and also for reasons of prudence, we prefer to clean up the common BD (CBD) approximately 30 to 40 days before performing a laparoscopic cholecystectomy.

As it is difficult to organize, the endolaparoscopic rendezvous technique was used in only 2 cases of gallstones with the presence of CBD stones and a difficult papilla (0.2% of the patients), and the technique consisted of a simultaneous VLC and ERCP, which was facilitated by inserting a guide wire through the cystic canal until beyond Vater's papilla, and an endoscopic lithotomy and papillosphincterotomy.39

Results

Thirty-four (3.28%) of the 1037 laparocholecystectomies led to complications: 6 BD lesions (0.58%), 18 cases of bleeding (1.74%), 5 cases of biliary pooling (0.49%), 2 intestinal lesions (0.20%), 2 trocar site infections (0.20%), and 1 postoperative occlusion (0.10%). Postoperative histology revealed 1 case of adenocarcinoma (0.10%). The gasless technique was used 4 times, the only complication being 1 mild subcutaneous hematoma. There were 14 coversions as a result of problems connected with the videolaparoscopic method (1.35%), which were mainly due to the presence of adherences or bleeding. To avoid the adherences, it was found that access to the right upper quadrant was useful. Six (3.03%) of the 198 open cholecystectomies led to complications: 1 BD lesion (0.51%), 2 intestinal lesions (1.01%), 2 cases of bleeding of the gallbladder bed (1.01%), and 1 case of biliary pooling (0.51%). Five hundred eighteen of all of the laparoscopic and laparotomic cholecystectomies involved female patients, and 717 involved males. The mean age of the 5 female and 2 male patients in whom a lesion of the main biliary pathway occurred was 46.2 years. Figure 1 shows the percentages of laparoscopic and laparotomic cholecystectomies from 2000–2009.

View Figure

• Download as Powerpoint
• Download Figure

Most of the complications were secondary to the presence of a difficult gallbladder, one possible classification of which is that of Nassar et al58 on the basis of the appearance of the gallbladder, the obviousness of the peduncle, and the presence of inflammatory adherences. In relation to operating difficulty, a gallbladder can be grade I (i.e., soft walls, a free peduncle, lax adherences at the infundibulum), grade II (i.e., mucocele, impacted with stones, abundant fat, simple adherences on the body), grade III (i.e., intrahepatic, acute cholecystitis, retraction, fibrotic Hartmann's pouch, impacted stones, anatomic anomalies, short cystic duct, tenacious adherences at the fundus and the duodenum), grade IV (i.e., empyema, gangrene, suspected cancer, unrecognisable peduncle, fibrous and tenacious adherences involving all of the gallbladder and the duodenum).

Figure 2 shows the number and distribution of complications after laparoscopic cholecystectomy.

View Figure

• Download as Powerpoint
• Download Figure

There were 14 conversions: 2 as a result of phlegmon and extensive adherences, 7 as a result of tenacious adherences or difficulties in isolating the hilus, 3 as a result of bleeding and an uncleanable operative field (n  =  1 case of bleeding of the gallbladder bed and n  =  2 of cystic artery bleeding), and 2 as a result of the presence of hepatic cirrhosis.

Two of the 3 intestinal lesions occurred during laparoscopy: in 2 cases (n  =  1 during laparotomy and n  =  1 during laparoscopy) a jejunal loop was lesioned, of which the first was repaired during the course of the operation and the second required laparotomic resection on the third postoperative day; in the third case, the transversal colon was lesioned during laparoscopy and underwent laparotomic repair on the third postoperative day.

Of the 18 cases of bleeding, 7 involved the gallbladder bed (n  =  4 resolved laparoscopically and n  =  3, laparotomically), 7 involved the abdominal wall caused by the trocar (n  =  4 resolved laparoscopically and n  =  3, laparotomically), 2 were sine materia (i.e., no bleeding after the distancing of the clots, n  =  1 laparoscopy and n  =  1 laparotomy), 2 were due to a small lesion of the III hepatic segment caused by the trocar (i.e., hemostasis in laparotomy), and 1 was due to a hematoma of the right lobe (exploratory laparotomy on the third postoperative day, spontaneous resolution).

There were 5 cases of fistulas and biliary pooling after laparoscopic cholecystectomy; 3 resolved spontaneously, and 2 resolved after ERCP and stent placement. An early reintervention after a laparoscopic cholecystectomy was necessary in 1 case because of a curtain bridle between the omentum and a trocar site.

There were 7 lesions of the main biliary pathway [n  =  6 during laparoscopy, (Fig. 3), and n  =  1 during laparotomy]: 1, a deep lesion during laparotomic cholecystectomy, was resolved by means of a Roux-en-Y loop hepatico-jejunostomy during the course of the same operation; 1, a lesion of the common hepatic duct during laparoscopy, required laparotomic reconstruction on a Kehr T-tube; another, a lesion of the choledochus during laparoscopy, also required laparotomic reconstruction on a Kehr T-tube and the endoscopic positioning of a stent because of substenosis at the suture site; 1 tangential lesion of the right hepatic duct occurring during laparoscopy required the endoscopic positioning of a stent and the drainage of a subhepatic abscess under ultrasound guidance; a biliary fistula caused by stenosing odditis (after laparocholecystectomy) was treated by means of laparotomy, choledocholithotomy, duodenotomy, papillotomy and transduodenal plastic surgery on the sixth postoperative day; 1 case of Mirizzi syndrome, which started during laparoscopy, required conversion and a Roux-en-Y loop hepatico-jejunostomy because of a lesion of the main biliary pathway; and 1 direct laparotomic reconstruction on a Kehr T-tube was performed on the third postoperative day after a tangential lesion of the biliary pathway during laparoscopic cholecystectomy.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

As can be seen in the following data, the published frequencies of the different complications vary widely9,10,12,26–32: BD lesions, 0.2% to 4%; intraoperative bleeding, 0.47% to 1.9%; postoperative bleeding, 0.08% to 5%; intestinal lesions, 0.1% to 3.9%; trocar site infections, 0.45% to 8%; and conversions, 0.9% to 2.9%. Most of the complications become apparent during the procedure or in the subsequent 48 hours (e.g., biliary losses, bleeding, intestinal lesions). Table 1 summarizes the percentages of complications in our case series.

Table 1 Percentages of complications

View Fullscreen

The vascular complications associated with VLC that are most frequently reported in the literature include cystic artery bleeding, vascular lesions of the hepatic peduncle, bleeding of the hepatic bed of the gallbladder or the hepatic parenchyma, and bleeding of an omental vessel or the vessels of the abdominal wall (caused by a trocar). Some bleedings are not due to any particular structural lesion but are caused by coagulation defects, particularly in patients with cirrhosis or portal hypertension; in the case of such defects, the videolaparoscopic indication must be carefully weighed against the risk.

In the case of bleeding from the cystic artery during the course of VLC, an attempt at hemostasis without conversion can be made, provided that the operative field can be cleaned (also by means of the addition of an additional trocar and the use of gauze, without positioning clips blindly or by using monopolar coagulation); however, in these cases, conversion may be the best solution. Conversion is necessary and must be rapid in the presence of major bleeding.33

Bleeding is favored by inadequate anatomic exposure, the presence of an acute inflammatory state, pre-existing liver disease, the extent of adherencies, and the association of coagulopathies. Trocar-induced bleeding (i.e., parietal, visceral, or, rarely, major vascular bleeding) can be minimized by using the open technique to insert the first trocar and by inserting the others under the direct visual guidance of the telecamera.

Postoperative bleeding can be insidious and should, therefore, be frequently monitored by means of clinical checks and hemochrome testing (because silent drainage can be misleading). Bleeding at the trocar insertion site can be minimized by means of careful checking with the optic after the removal of the individual trocars by momentarily occluding the hole with a finger or damp gauze with low-pressure pneumoperitoneum.

Lesions of the main biliary pathway are among the major complications of laparoscopic cholecystectomy because of the difficulty of repair and the impact they have on the patient, who remains psychically and mentally tried regardless of their course.14,34 Possible sequelae are a long hospital stay, a worse quality of life, greater morbidity, and, occasionally, mortality.35–37

Biliary pathway lesions (i.e., BD injuries) can be divided into 4 classes by using the Stewart-Way classification22,38: class I: incision (incomplete transection) of the CBD; class II: lateral lesion of the common hepatic duct; class III: transection of the CBD or common hepatic duct; and class IV: lesions of the right hepatic duct or the right hepatic segmentary duct. Biliary lesions can be divided into 5 types according to Strasberg et al21: type A: biliary losses from a minor duct that maintains continuity with the CBD; types B and C: discontinuities in a part of the biliary tree (often involving an anomalous right hepatic duct; type B are ductal occlusions and type C sections without occlusion); type D: lateral lesions of the main biliary pathway (usually partial lesions); and type E: circumferential lesions of the main biliary pathway.

Another important classification of biliary lesions is that of Bergman et al39: type A—biliary loss as a result of the dehiscence of the cystic stump or the section of peripheral ducts with an intact BD; type B—biliary loss as a result of laceration of the BD, with or without associated stenoses; type C—BD stenosis without biliary loss; and type D—complete sectioning or excision of a segment of the biliary pathway.

However well the repair is performed, the future onset of cholangitis is possible, also in the form of recurrent episodes that may lead to biliary pathway stenoses or cholestatic liver cirrhosis.14,40

If not recognized intraoperatively, biliary pathway lesions are more frequently detected as biliary fistulas than as dilation as a result of complete obstruction (with pain and rapidly deteriorating liver function). Endoscopic retrograde cholangiography is the most important examination for detecting a lesion; percutaneous cholangiography is less frequently needed.

Many authors believe that the complete sectioning of the CBD, which is rare in laparotomy, is the most frequent biliary pathway lesion during laparoscopy and is often related to an erroneous conviction that it is the cystic duct. The main reason for a complete section of the main biliary pathway is a failure to recognize it, so it occurs “without any doubt or hesitation”14; it is important to stress that this type of lesion is often caused by expert surgeons, who may tend to pay less attention as a result of a sort of technical habit.

According to the classification of Clavien,41 the lesions can also be divided on the basis of their impact on the clinical course: grade I: lengthen the postoperative period but do not require treatment (e.g., suppuration at a trocar insertion site); grade IIA: lengthen the postoperative period and require conservative treatment but do not leave sequelae (e.g., moderate hematic or biliary loss; the use of endoscopy for a papillotomy or the placement of a temporary stent); grade IIB: require a second laparoscopy or a laparotomy but without sequelae (e.g., dehiscence of the cystic duct, a bleeding hepatic bed, tangential BD lesion); grade III: complex lesions; and grade IV: death.

The most frequent cause of an iatrogenic biliary pathway lesion is a failure to recognize the anatomy of the hepatic peduncle (37% of cases),8 followed by inflammatory alterations of the gallbladder (23%) and anatomic anomalies (13%).3,42 A failure to identify the CBD can lead to its confusion with the cystic duct, with the subsequent positioning of clips and its sectioning; excessive traction on the gallbladder can lead to the high resection of the common hepatic duct and/or a lesion of the right hepatic artery; the presence of an aberrant right hepatic duct (2% of patients) can become an additional cause of mistaken anatomy.

The picture is complicated by possible pericholecystitis or a scleroatrophied gallbladder (a difficulty in finding a plane of cleavage in the gallbladder bed, modified local anatomy); liver cirrhosis associated with the biliary disease makes hemostasis and cleavage difficult. The presence of Mirizzi syndrome is to be greatly feared.43

The severity of the inflammation of the Calot's triangle also needs to be borne in mind, to the point that a scoring system has been created14 on the basis of the complexity of the local situation: grade I: no inflammation (score 0); grade II: moderate inflammation, with the structures easily identifiable and separable (score 1); grade III: moderate inflammation, with the structures still reasonably clearly identifiable and separable (score 2); grade IV: severe inflammation, in which the structures are difficult to separate but are identifiable (score 3); and grade V: severe inflammation, in which the structures are very difficult to identify and separate, and conversion to open surgery is necessary (score 4).

Various techniques have been proposed as a means of avoiding problems in identifying the cystic duct and main biliary pathway14,44: lateral rather than cephalic traction of the infundibulum is preferable45; the incision of the ventral and dorsal serous matter of Calot's triangle must be followed by careful observation of the anatomy; a critical view of safety has been proposed,20,21,46,47 which consists of completely dissecting the fat and fibrous tissue from Calot's triangle and isolating the cystic artery and duct by attacking the gallbladder in the lowest part of its adhesion to the cholecystic bed (a maneuver that is not easy in the presence of inflammatory tissue or a large infundibular stone). The junction between the cystic duct and gallbladder (infundibular technique) must always be precisely identified21,45 before applying any clips or making any incision; it is better to use a laparoscope with a 30-degree view45; in particular cases, it has also been proposed to use a fundus-first technique that foresees the preliminary ultrasonographic dissection of the gallbladder bed before isolating the structures of the peduncle in order to avoid any bleeding caused by a failure to ligate the cystic artery beforehand.48

When the anatomy cannot be precisely identified, it is necessary to consider a conversion to laparotomy26; some authors suggest the use of laparoscopic echography49 and intraoperative cholangiography.50 In expert hands, intraoperative echography50,51 has proved useful in detecting stones in the main biliary pathway, in recognizing the anatomy, and for dissection in the presence of Mirizzi syndrome; furthermore, unlike intraoperative cholangiography, it has the advantage of not requiring the previous isolation of the cystic duct.51,52

Lesion prevention therefore involves recognizing all of the elements of the gallbladder peduncle in Calot's triangle. In addition to the cystic artery and duct, these include the following: cystic arterioles of Calot; the cystic lymph node of Mascagni (near the infundibulum and the entry of the cystic artery into the cystic duct); the possible presence of a right hepatic artery (18%); the possible presence of an aberrant right hepatic artery arising from the superior mesenteric artery and from which the cystic artery originates in 83% of cases; the possible presence of right sectorial or segmentary hepatic ducts which converge in the BD beneath the hilar convergence or which outlet in the cystic duct (the so-called hepaticocystic canals); the possible presence of the right hepatic duct in the case of low extrahilar convergence3; or the confluence of the cystic duct in the right hepatic duct (which favors erroneous clipping). In practice, no structure should be clipped or sectioned until it has been surely identified. The use of an electrocauter near already positioned clips should be avoided in order to avoid causing necrotic areas as a result of heat transmission.

In the case of a lesion of the main biliary pathway, there is no doubt that it is important to recognize the lesion itself as soon as possible (and one should always be suspected), to check the anatomy and entity of the lesion, and, if the team is expert in hepatobiliary surgery, to make a primary repair with conversion to laparotomy. After performing intraoperative cholangiography38 in order to make an anatomic evaluation of the damage, the repair can be direct on a prosthesis or T-tube (e.g., in the case of a tangential lesion without any loss of substance). If this is not the case, a Roux-en-Y loop hepaticojejunostomy may be necessary within 24 to 48 hours of the lesion occurring, if possible.42

Approximately two thirds of the lesions are suspected after VLC: the symptoms are related to the type of lesion and, in particular, include jaundice in the case of complete BD obstruction or a biliary fistula (onsite drainage) or choleperitoneum in the case of the complete or partial sectioning of the BD. The jaundice onsets early, appearing in the first 2 postoperative days, with an increase in the indices of stasis, gamma-glutamil-transjeptidasi (gamma-GT), and alkaline phosphatase.

Ultasonography is very useful in such cases, together with endoscopic retrograde cholangiography. The increase in bilirubin per se is not a reason for urgency unless there is the onset of fever,42 and, if possible, the procedure should programmmed.

If the drainage is well positioned, a biliary fistula is formed in the case of biliary loss; if the drainage has already been removed or is badly positioned, biliary ascites develops, and the patient may be asymptomatic or may complain of vague abdominal pains even days afterward; ultrasonography is a great diagnostic help also in this case.

If the biliary pooling becomes infected, the patient rapidly develops a septic state, and repair is very difficult.53–55 In this situation, it is better to wash and drain the abdominal cavity (even laparoscopically) in order to be able to study the lesion more clearly and acquire as much information as possible concerning the extent of the damage. If the loss is due to a peripheral duct or the dehiscence of the cystic duct, with an integral biliary pathway, an endoscopic sphinterotomy is often sufficient, with the removal of any stones from the biliary pathway, external drainage, and the positioning of a stent (most useful in the case of small tangential lesions).

The long-term result of the repair of this type of lesion greatly depends on the first decisions taken and the first provisions made42; the presence of an expert surgeon, an endoscopist, and a radiologist provides the best guarantee of a good result. In reality, only one third of the lesions are recognized during VLC.56–58 If the lesion is complex and the operator is not capable of making a difficult repair, it is better to position drains in contact with the biliary loss and send the patient to a center specialized in hepatobiliary surgery.42

In conclusion, our experience tends to mirror that of other surgeons but leads us to believe that, even after 20 years, closed cholecystectomy has not yet reached the stage of providing the homogeneous results of traditional laparotomic surgery. Laparoscopy is a technique that could still undergo some technologic instrumental evolution and that requires dedicated progressive training; the aim is to reach an equilibrium between the complications of the 2 methods. The literature and our personal experience leads us to believe that, in the preoperative interview with the patient and in obtaining informed consent, it is correct to specify the real advantages, disadvantages, and risks of the 2 procedures.

Whenever possible, the BD lesions were treated by reconstructing the biliary pathway, with hepaticojejunostomy being reserved for just a few cases. The rationale underlying this choice arises from our conviction that, in the case of a future stenosis, there is still the possibility of intervening and performing a later hepaticojejunostomy, which becomes a relatively simple procedure when the biliary pathway is dilated. The patients who were treated by means of reconstruction, sometimes supported by subsequent endoscopic dilation and the positioning of 1 or 2 stents, have not needed any additional treatment.