Gallbladder cancer (GBC) is a relatively rare neoplasm and is considered to be an aggressive and highly lethal disease. It is the most frequently occurring malignancy of the biliary tract and the fifth most common gastrointestinal cancer [1]. Only 30% of gallbladder carcinomas are suspected preoperatively, and the remaining 70% are usually discovered incidentally by pathological examination during or after surgery [2].

In recent years, laparoscopic cholecystectomy (LC) has become the accepted gold standard management for gallbladder disease. With the advantages of a shorter hospital stay, decreased post-operative pain and earlier resumption of normal activities compared to open cholecystectomy, this procedure has now become routine in the treatment of benign gallbladder disease worldwide. LC that is performed for benign gallbladder disease rarely results in a diagnosis of unexpected gallbladder carcinoma. Incidental gallbladder carcinoma (IGBC) is defined as carcinoma of the gallbladder suspected for the first time during cholecystectomy, or accidentally found on histological examination of gallbladder. The estimated incidence of IGBC diagnosed by cholecystectomy is 1–2% [3]. Although patients with GBC have a poor prognosis, most IGBC tumors tend to be at an early stage of development. Hence, radical second surgery is performed in most patients with IGBC, and improved prognosis is usually reported [4–7]. However management of IGBC is a difficult issue in the absence of established guidelines. The primary aim of our study was to evaluate the incidence, clinicopathological characteristics and outcome of patients with IGBC.

The study was approved by Ethics Committee at Iwakuni Clinical Center. Between January 2008 and January 2015, 352 patients underwent LC at Iwakuni Clinical Center. Among these patients, 8 (2.3%) were diagnosed with IGBC. Their clinical presentation, ultrasonography findings, preoperative diagnosis, intraoperative findings, histopathological records and surgical outcomes were reviewed. We excluded from the study all patients with suspected pre-operative malignancy. All LCs were executed using the standard 4 trocar technique, and EndBag protected gallbladder (GB) extraction. Patients were staged according to the American Joint Committee on Cancer 7th edition tumor node metastasis clinical staging system for GBC [8]. We have advocated additional radical surgery for patients with pT2 or pT3 tumors [4, 5]. Regarding the surgical radicalization procedure, we performed hepatic S4a + S5 resection with hepatic peduncle lymphadenectomy [4, 5, 9]. The completeness of the resection was classified as follows: R0, no residuals in the hepatic margins; R1, a microscopically positive margin; and R2 macroscopic residuals in the hepatic margin.

According to the pathological TNM classification system, the tumor stages were as follows: pT1a (3 cases); pT2 (4 cases); and pT3 (1 case). The pathological characteristics of the tumors were as follows: well differentiated adenocarcinoma (6 cases); moderately differentiated adenocarcinoma (1 case); and poorly differentiated adenocarcinoma (1 case).

Three patients with pT1a disease had no additional surgery, and there was no recurrence. Two patients with pT2 disease underwent radical surgery and there was no recurrence. The remaining three patients with pT2 disease (2 cases) or pT3 disease (1 case) refused radical surgery. All patient had no adjuvant therapy (chemotherapy or chemoradiation). Follow up of both patients undergoing radical surgery and refusing radical surgery is blood examination including tumor marker every 2 months and CT imaging studies every 4 months. The median follow-up time was 24 (range 11–80) months. All patients are still alive and two of the three patients who refused radical surgery developed recurrence (liver metastases and recurrence in the peritoneum, respectively). The remaining patients were disease-free during follow-up. We did not find any patients with port-site recurrence during the follow-up period. Patients with pT2/T3 received no adjuvant chemotherapy or chemoradiation post-operatively.

The incidence of IGBC during or after LC has been reported to be 0.19–3.3% [5, 10–12]. Because of the increased use of LC and difficulty regarding the preoperative diagnosis of GBC, IGBC has become more frequent [11, 13]. In the present study, the rate of occurrence of IGBC was 2.3%.

Some studies have reported that risk factors for IGBC are as follows: sex (female); obesity; age >65 years; cholelithiasis; polypoid lesions; Asian or African American; and an elevated alkaline phosphatase level [14–17]. In our study, three patients were female and their median age was 64 years. All of our patients with IGBC were Asian. The symptoms related to IGBC can be relatively nonspecific, with the early symptoms mimicking those of GB stones or cholecystitis. GB stones are found in 70–98% of patients with GBC. Cholelithiasis causes chronic irritation and inflammation of the GB, which leads to the development of mucosal dysplasia and subsequent carcinoma; it takes a long time for the promotion of tumor proliferation and the occurrence of malignancy [18]. In our study, pain in the right hypochondrium was present in four patients. An additional three patient had no symptoms. None of our patients presented with a clinical profile that differed from that of typical patients with GB stones or cholecystitis. In the early stages, it is difficult to differentiate between GBC and cholecystitis, because thickening of the GB wall is a feature of both diseases [19].

Frozen sectioning should be performed when surgeons suspect malignancy in intraoperative finding [20, 21]. In our study, the intraoperative assessment of the GB using frozen sections revealed GBC in one patient with GB wall thickening and cholecystitis. Consequently, we switched from LC to open cholecystectomy. The efficacy of frozen sections was poor with respect to the diagnosis of carcinoma in situ [20, 21]. Kwon et al. also found that diagnosis using frozen sections did not reliably detect the carcinoma in situ and the depth of invasion of GBC; thus, it should not be considered a definitive diagnostic procedure [22]. The most appropriate approach for diagnosis at present is to macroscopically examine the GB mucosa during surgery, and to perform frozen sectioning for any suspected lesion.

The survival rate of IGBC patients is high with a median survival rate of 21.2–60 months [5, 10, 11]. All patients in our studies were alive at a mean follow-up time of 24 months. One reason for this may be early diagnosis, as in the present study in which most of the cases involved stage 1 or 2 tumors, thereby resulting in better prognoses. However, we could not find any common characteristic that could differentiate these GB carcinomas from benign tumors in laboratory and sonographic findings. The difficulty in early diagnosis of GBC results from its poor specificity and ambiguity regarding clinical symptoms [23–25]. Tumor differentiation is an independent prognostic factor that affects survival [5]. In the present study, six of eight tumors were well differentiated adenocarcinomas.

Surgical resection is the only curative therapy for GBC [15, 26, 27]. Determining the therapeutic approach for GBC in accordance with disease stage is supported by most authors [27, 28]. There is a consensus that simple cholecystectomy or LC is an adequate treatment for pT1a. pT2 and higher stages should be treated using additional radical surgery using hepatic S4a + S5 resection with hepatic peduncle lymphadenectomy [9, 15, 29, 30]. For patients with T1b disease, the correct approach is still debated and some authors recommend hepatic gallbladder bed resection with hepatic peduncle lymphadenectomy [31, 32]. In the present study, the survival of patients with pT1 disease was favorable after LC. Two patients with pT2 disease underwent additional radical surgery and had good outcomes. Other patients with pT2 and pT3 disease refused additional radical surgery and later developed recurrence.

Laparoscopic cholecystectomy appears to have resulted in the earlier discovery of GBC in some patients, resulting in an increased probability of successful treatment [33]. Precise surgery is essential in obtaining a good outcome from IGBC treatment [4]. The present study was retrospective and involved a small number of patients. A study involving a considerably larger sample size will be required to evaluate the surgical outcome and effectiveness of radical surgery for IGBC.

The incidence of IGBC detected during laparoscopic cholecystectomy in this single center study was found to be 2.3%. No association could be found with risk factors that have been reported by other authors. Although the number of patients enrolled was small, our findings further support the suggestion that GB carcinoma may be curable if diagnosed at an early stage as IGBC. pT1b or greater tumor stages identified on LC may benefit from additional radical resection.