Gallbladder cancer is a disease characterized by its poor prognosis and late diagnosis [1, 2, 3, 4]. In our center, 74% of the early forms of tumors are detected after histologic examination of the cholecystectomy specimen [1]. The introduction of laparoscopic cholecystectomy (LC) drew attention to the existence of recurrence sites related to the laparoscopic method itself [5, 6]. Metastatic wound implants most commonly arose from trocar sites [5, 6, 7]. Hence concern has been expressed that LC could adversely affect the prognosis. Because of the high incidence of gallbladder cancer in Chile (12/100,000 inhabitants) the diagnosis of incidentally finding cancer in a “routine” cholecystectomy specimen is common [1, 3]. In view of these facts, the possible effect of laparoscopy on the prognosis represents a cause of concern among surgeons.

In the present study we evaluated the role of laparoscopy in the survival of patients with gallbladder cancer. We did so by comparing patients in whom cancer was detected after LC with a group of patients with similar clinical characteristics in whom cancer was diagnosed after open cholecystectomy (OC).

Materials and Methods

The medical records, imaging data, operative records, and pathologic findings were reviewed in 24 patients with a gallbladder cancer discovered after or during LC. To analyze the role of laparoscopy, these patients were compared with a series of 40 patients with gallbladder cancer detected after OC. Thus for every patient whose cancer was detected after or during LC, two patients with a cancer detected after or during an open procedure were recruited. To make the two groups comparable, patients were matched by level of wall invasion, age, and whether they had undergone reoperation. These variables were chosen because of their relation with the prognosis [1, 4]. Control patients were chosen correlatively from our database of gallbladder cancer patients. Because patients with in situ and mucosal infiltration who underwent LC and OC are alive and free of disease, we did not study this group. Furthermore, because the number of patients with muscular invasion was small, they were also excluded from the survival analysis. Follow-up data were obtained by personal contact with the patients or from their clinical records. In cases in which the gallbladder was completely resected, the surgical specimen was examined using stepwise sectioning of the whole gallbladder.

Tumors were classified according the level of wall invasion. The categories were in situ, mucosal, muscular, subserosal, and serosal invasion. The American Joint Commission for Cancer/International Union Against Cancer (AJCC/UICC) classification was employed [8]. Patients were managed according to a previously reported protocol [1, 9, 10]. Thus for those with subserosal or deeper infiltration, a second operation with the aim of resecting potentially involved tissue was offered. The median follow-up ranged from 2 to 96 months.

The cumulative survival rate was calculated according the Kaplan-Meier method. To compare the survival curves for the two procedures, we performed a Cox proportional hazard model, adjusting the standard error for clustering on the matched variable. Recurrence was analyzed using Fisher’s exact test. Differences between variables were considered statistically significant when p < 0.05.

Results

The 24 patients with gallbladder cancer detected after the laparoscopic procedure included 4 men and 20 women aged 36 to 77 years. The preoperative diagnosis in these patients was cholelithiasis in 22, gallbladder polyp in 1, and acute cholecystitis in 1.

Concerning the level of wall invasion, 2 patients had an in situ tumor (AJCC/UICC T0). 2 a mucosal tumor (T1a), 1 a muscular tumor (T1b), 13 a subserosal tumor (T2), and 6 a serosal lesion (T3) (Table 1).

Table 1. Patients who underwent laparoscopic cholecystectomy in whom cancer was diagnosed after or during the procedure.
Full size table

All of the patients with in situ or mucosal tumors are alive and free of disease after a follow-up of 20 to 60 months. The patient with muscular invasion is also alive after 8 months of follow-up.

Among the patients in the LC group, 10 who had subserosal compromise underwent reoperation with the aim of resecting potentially compromised areas. Of these 10 patients, 4 had recurrences that precluded resection. The recurrences were observed at the site lateral trocar hole placement in one patient, in the prerenal fasciae in another, and in the lymph nodes in two patients (Table 2). Among these four patients, those with recurrences in the prerenal fasciae, at the trocar holes, and in the paraaortic lymph nodes were dead after 9, 16, and 16 months of follow-up, respectively. The other patient is still alive but with a follow-up of only 2 months.

Table 2. Recurrences in patients undergoing reoperation.
Full size table

Of the six patients with serosal compromise, five were converted to open surgery because of intraoperative suspicious of tumor. The other patient completed the laparoscopic procedure, and the diagnosis was established after studying the cholecystectomy specimen. Three of these patients underwent reoperation, but in all three cases the tumors were unresectable because they were not suitable for a curative procedure.

The 40 patients whose cancers were detected after an open procedure comprised the control group. The group included 26 patients with subserosal compromise, 12 with serosal invasion, and 2 with muscular invasion.

The two patients with muscular invasion are alive after 23 and 50 months of follow-up, respectively. Of the 26 patients with subserosal tumors, 20 underwent reoperation; 18 of them were able to undergo resection, but the other 2 had infiltration outside the area to be resected. Although the proportion of patients whose lesions could be resected in the OC group was higher, differences did not reach statistical significance (p = 0.141). On the other hand, among the 12 patients with serosal infiltration, 6 underwent reoperation; and of these 6, only 2 were able to undergo resection (p = 0.50). The rest of the patients had a recurrence that precluded resection with a curative aim (Table 2).

The survival analysis did not show a higher risk of death for those who underwent laparoscopic surgery compared with those undergoing open surgery. Using Kaplan–Meier survival estimates, patients who underwent laparoscopic surgery had a risk of death of 1.22 (hazard ratio) compared with those undergoing open surgery. However, this was not statistically significant [p = 0.598; 95% confidence interval (CI) 0.58–2.571].

Figure 1 shows survival curves of patients according to the cholecystectomy technique. After 60 months of follow-up, 34.9% of the LC patients and 35.4% of the OC patients were alive (p > 0.05). Among the patients with infiltration restricted to the subserosal layer, survival was no different for those who underwent LC or OC (hazard ratio 1.01; p = 0.97; 95% CI 0.36–2.38) (Fig. 2). Analysis of those with invasion of the serosal layer also did not show any differences in survival for the LC and OC patients (hazard ratio 1.64; p = 0.38; 95% CI 0.54–4.92) (Fig. 3).

Fig. 1.
figure 1

Survival curves of patients with subserosal and serosal infiltration undergoing cholecystectomy by a laparoscopic procedure (LP) versus those who undergoing an open procedure (OC). p = 0.598.

Full size image
Fig. 2.
figure 2

Survival curves of patients with subserosal infiltration undergoing cholecystectomy by a laparoscopic procedure compared with those undergoing cholecystectomy by an open procedure. p = 0.974.

Full size image
Fig. 3.
figure 3

Survival curves of patients with serosal infiltration undergoing cholecystectomy by a laparoscopic procedure compared with those undergoing cholecystectomy by an open procedure. p = 0.38.

Full size image

Discussion

The most frequent ways for gallbladder cancer to spread are local invasion, regional lymph node involvement, and distant metastasis [11]. Recurrences at the level of scars or trocar holes and by peritoneal infiltration are not common in patients with early-stage gallbladder cancer, although they have been described in those who have undergone LC [5, 6, 7, 12, 13, 14, 15].

Intraoperative perforation of the gallbladder with spillage of bile seems to be an important event in the seeding of cells. According to the data of Z’graggen et al., the incidence of port-site recurrence increased from 9% in patients without intraoperative perforation to 40% in those in whom perforation could be demonstrated [13].

Although this was not a randomized trial, to distinguish the real effect of laparoscopy on the prognosis, we designed the study with the aim of comparing the same categories of patients according to the cholecystectomy technique used. The analysis of patients with early-stage tumors, such as mucosal or in situ lesions, showed that laparoscopy did not influence their outcome. Patients with small tumors undergoing laparoscopy had an excellent prognosis, as all were alive without evidence of recurrence. Although the total number of these patients is small, this fact is important because it shows that laparoscopy per se is not responsible for the unfavorable prognosis. Although it was not recorded in this study, bile spillage may be responsible for recurrences even with small tumors. This fact stresses the value of using meticulous technique in all LC cases.The analysis of patients with subserosal and serosal invasion who underwent reoperation showed a higher recurrence rate for patients who underwent the laparoscopic procedure. Although differences did not reach statistical significance, studying a larger number of patients might confirm the impression that patients with subserosal or serosal invasion undergoing LC are at a higher risk of recurrence than are those undergoing OC.

Among those with a potentially resectable gallbladder cancer, patients with subserosal infiltration had an intermediate prognosis. Survival analysis of patients who underwent LC did not show any difference in the prognosis compared with those who underwent OC.

Finally, those with serosal infiltration represent a poor-prognosis group independent of the operative method. These patients rarely survive longer than 1 year after the diagnosis, although exceptions occur when a large number of patients are evaluated. Although it was not statistically observed in this study, laparoscopy itself probably influences the outcome in this group of patients. Laparoscopic maneuvers can exfoliate malignant cells from the gallbladder surface, favoring implantation in the peritoneum. This fact is evident in reports that show glandular cells in 40% of laparoscopic instruments and in the filtrate of exhausting carbon dioxide [16, 17]. Other experimental studies demonstrated that pneumoperitoneum significantly increased implantation at trocar sites and tumor growth in the peritoneum [18, 19].

Conclusions

We emphasize here that laparoscopic cholecystectomy does not influence the prognosis in patients with small tumors, particularly among patients with less than subserosal wall infiltration. Patients with subserosal infiltration represent an interesting group in whom a high incidence of recurrence might be demonstrated if a larger number of patients with longer follow-up are studied. Those with serosal invasion might comprise a group in whom the safety of laparoscopy should be examined. Therefore we suggest conversion to an open procedure for patients in whom gallbladder cancer is recognized prior to or during laparoscopy. For patients who have already undergone LC and have been diagnosed as having gallbladder cancer after the specimen is examined histologically, a poor prognosis is likely. This group may benefit from adjuvant chemotherapy with or without irradiation. Clinical trials are warranted to address this issue.

Résumé

De nombreuses publications suggèrent que les récidives du cancer de la vésicule biliaire sont plus fréquentes et d’un plus mauvais pronostic après cholécystectomie coelioscopique (CC) qu’après cholécystectomie par voie traditionnelle (CT). Le but de cette étude a été de comparer la survie des patients ayant eu une intervention sous cœlioscopie par rapport à ceux ayant été opérés par CT. Vingt-quatre patients ayant un cancer de la vésicule biliaire détecté après CC ont été comparés à 40 patients consécutifs porteurs d’un cancer de la vésicule biliaire détecté après CT. Les patients ont été appariés selon le degré d’invasion pariétale, l’âge et selon qu’ils ont eu une réintervention ou seulement une cholécystectomie. La série a été composée de deux patients ayant une tumeur in situ, deux cas de tumeur de la muqueuse, un patient ayant une invasion de la musculaire, treize patients ayant une invasion de la sous-séreuse et six patients présentant une invasion de la séreuse. Une récidive a été constatée chez quatre des dix patients ayant une invasion de la sous-séreuse: ils ont été réopérés. En revanche, dans le groupe opéré par CT, des 26 patients porteurs d’une invasion de la séreuse, 20 ont eu une intervention itérative, mais seulement deux ont eu une récidive. Des six patients ayant une invasion de la sous-séreuse, trois patients, dans le groupe CC, ont eu une intervention itérative. Tous avaient des récidives qui interdisaient la résection. Des 12 patients qui ont eu une invasion de la séreuse dans le groupe CT, six ont eu une ré-opération et quatre ont récidivé. Les courbes de survie globale n’ont pas montré de différences lorsque les patients ont été comparés selon le type de procédé réalisé. De même, dans l’analyse des patients selon le degré d’invasion pariétale, il n’y avait aucune différence significative en ce qui concernait la survie. Bien qu’il existe plusieurs publications démontrant un plus mauvais pronostic pour les patients porteurs de cancer de la vésicule biliaire opérés sous cœlioscopie par rapport à la chirurgie traditionnelle, cette étude ne montre pas une différence de survie significative entre les deux voies d’abord. Bien que le taux de récidive soit plus élevé chez le patient opéré sous coelioscopie, il n’y avait pas de différence en ce qui concernait la survie.

Resumen

La existencia de reportes que muestran un peor pronostico en aquellos pacientes en los que el diagnóstico de cáncer de la vesícula biliar (C.V.B) es efectuado luego de una colecistectomía Laparoscópica (C.L.), motivó la realización de un estudio que pretendía comparar un grupo de pacientes en los que el diagnostico fue efectuado luego de una C.L. con otro en el que el diagnóstico fue efectuado luego de una Colecistectomía abierta (C.A.). Veinte y cuatro pacientes portadores de un C.V.B. diagnosticado luego de una C.L. fueron comparados con 40 pacientes portadores de un C.V.B. diagnosticados luego de una C.A. Los pacientes fueron pareados de acuerdo a la edad, nivel de infiltración de la pared y si fueron reoperados o no. La serie estuvo compuesta por dos pacientes con un tumor in situ, dos pacientes con un tumor mucoso, un paciente con invasión muscular, 13 con compromiso sub seroso, y 6 con invasión serosa. Recurrencias fueron observadas en 4 de 10 pacientes con invasión sub serosa que fueron sometidos a una reintervención. Por otra parte, en el grupo de pacientes sometidos a una C.A., de 26 pacientes, 20 fueron reoperados y solo dos tuvieron tumor residual. De los 6 pacientes con invasión serosa, tres en el grupo de C.L fueron sometidos a una reoperación, los tres tuvieron tumor residual. De los 12 pacientes en el grupo de C.A. 6 fueron reoperados, en 4 observándose tumor residual. El análisis de las curvas de sobreviva no mostró diferencias cuando los pacientes fueron comparados de acuerdo al método en que fueron colecistectomizados. Cuando este análisis fue efectuado en sub grupos de acuerdo al nivel de invasión tampoco fueron observadas diferencias. A pesar de la existencia de reportes que muestran un peor pronóstico para los pacientes portadores de un C.V.B. detectado luego de una C.L. este análisis no mostró dichos resultados. Aunque existe un mayor numero de tumor residual en los pacientes sometidos a una C.L. esto no se tradujo en una menor sobreviva.