Revision surgery for incidental gallbladder cancer: factors influencing operability and further evidence for T1b tumours

Abstract

Background and aim. There is a need to increase the available data on revision radical surgery for incidental gallbladder cancer and to determine factors influencing operability. We aimed to assess the impact of stage of disease (pT) and the type of primary surgery (laparoscopy versus open) on resectability rates. Material and methods. The data of 90 consecutive patients referred to the Tata Memorial Hospital between 1 January 2003 and 30 April 2007 for revision radical surgery for incidental gallbladder cancer were reviewed retrospectively. Results. Of the 90 patients who underwent revision surgery, accurate data on T-stage was available in 76, and of these 76 patients, 44 (57.8%) had prior laparoscopic simple cholecystectomy, while 32 (42.2%) had undergone open surgery. The median time interval between the two surgeries was 2 months (range 4 weeks to 11 months). By T-stage, 23 patients had T1b disease, while 33 and 20 patients had T2 and T3 disease, respectively. Successful revision surgery could be undertaken in 71% of patients (54/76) and 29.6% of these had residual disease confirmed by histopathological examination. Conclusions. T-stage is an important factor in determining operability as confirmed by our study. As the T-stage of the disease increased, the chances of finding residual disease increased, while operability decreased. Furthermore, the case for revision surgery is strengthened because the incidence of lymph nodal disease is high even for pT1b cancers. The type of primary surgery does not affect operability in patients undergoing revision radical surgery for incidental gallbladder cancer.

Introduction

Radical resectional surgery for gallbladder cancer has been consistently associated with the likelihood of long-term survival ¹,²,³. This applies not only in the case of per primum surgery, but also in revision radical surgery for incidental gallbladder cancer ³,⁴,⁵,⁶. The rationale for performing revision surgery for patients who have undergone simple cholecystectomy for invasive gallbladder cancer is based on the premise that as the T stage increases, there is an increased chance of finding disease in the gallbladder fossa as well as in regional lymph nodes ³,⁴,⁵,⁶.

Patients undergoing laparoscopic cholecystectomy have been reported to have a poorer clinical outcome than patients undergoing open surgery ⁶,⁷,⁸,⁹. In a recent retrospective analysis, though, Cucinotta et al. ¹⁰ reported that laparoscopic cholecystectomy has no such adverse effect. However, the number of patients with gallbladder cancer in this series was too small to reach such a conclusion. Stage of the disease is also known to impact on outcome. The likelihood of a good outcome reduces proportionately with advancement of the stage ²,¹⁰,¹¹.

In presenting their experience with gallbladder cancers, Fong et al. ¹² showed that there was no difference in long-term survival after curative resection between those treated with one operation versus those treated with two. However, the impact of the delay between primary surgery and revision surgery on the chances of successful radical resection has not been addressed.

In this study, we attempted to determine whether the type of primary surgery, viz. laparoscopy or open surgery, plays a role in affecting operability. We also addressed the rationale for performing radical surgery based on the T-stage obtained from the histopathological examination of the specimen of the first surgery.

Material and methods

The records of all patients who underwent revision radical surgery for incidental gallbladder cancers referred to the Department of Gastrointestinal Surgical Oncology of the Tata Memorial Hospital, Mumbai between 1 January 2003 and 30 April 2007 were included in the study. Preoperatively, all patients were investigated with routine blood investigations, including blood counts, liver and renal functions, and tumor marker CA 19-9. Preoperative evaluation included contrast-enhanced computed tomography (CT scan) and a review of the histopathology of the gall bladder excised during the first surgery to confirm the diagnosis and stage of the cancer.

All patients (T1b and above) were offered surgical exploration with an intent to perform a revision radical cholecystectomy. The procedure included clearance of the following nodes: cystic, pericholedochal, hepatic hilar, hepatic, retroportal, posterior pancreatoduodenal, and celiac. In addition, a non-anatomical wedge resection of the gallbladder bed (segments IV B and V) was performed. In all patients, the cystic duct stump was identified and revised (with negative margins confirmed by frozen section). Perioperative mortality was defined as death during the hospitalization following surgery or within 30 days of surgery. Total hospital course was defined from the date of surgery until the patient was discharged.

All statistical analyses were performed using SPSS 14.0. Continuous data are expressed as the mean±standard deviation. For purposes of studying significance, the variable of duration between surgeries has been divided into ≤1 month, 2–3 months, and ≥4 months. Significance was calculated by chi-square test and correlation was measured using cross tabs in the SPSS.

Results

During the period between 1 January 2003 and 30 April 2007, 90 patients were subjected to an exploratory laparotomy with intent to perform completion revision surgery. However, data on the preoperative pT-stage could be obtained only in 76 patients (since all these patients had their primary cholecystectomy elsewhere, and we were therefore dependent on those slides/blocks for analysis). Of the 76 patients, there were 20 males (26.3%) and 56 females (73.6%). The mean patient age was 49±10.7 (24–73) years.

Fifty-four (71%) patients had successful revision radical surgery, and in 16 of these 54 (29.6%) residual disease was found on histopathological examination of the gallbladder fossa and/or lymph nodes. In the 22 patients in whom resection was not performed, metastatic disease was detected after exploration (not reported in preoperative imaging), and this was confirmed on frozen section. (see Table I for intra-operative factors including American Society of Anesthesiologists (ASA) grade, number of lymph nodes dissected, median operating time, and blood loss.

Table I. Intra-operative factors.

Characteristic	Values (n=90)
American Society of Anesthesiologists (ASA) grade	I–67
	II–22
	III–1
Mean lymph nodes dissected (range)	6±4 (1–23)
Median operating time	180 min (range 40–335)
Median blood loss	300 cc (range 50–1000)

The morbidity rate was 9.2% (7/76); 5 patients had wound infections, 1 had a seroma of the cutaneous wound, and 1 had a Common hepatic duct perforation necessitating T-tube insertion. There was no mortality in the entire study period.

The first surgery (laparoscopic or open) and its influence on outcome

In 44 patients, the first surgery (simple cholecystectomy) had been performed laparoscopically, while 32 patients underwent open surgery. Of the 44 patients who had undergone laparoscopic surgery, 33 (75%) were operable at the time of revision surgery, while 11 (25%) were found to be inoperable. Of the patients who had undergone prior open surgery, 21 (64%) were operable, while 11 (36%) were deemed inoperable based on the presence of metastatic disease. When the type of first surgery was compared to the likelihood of a successful outcome at the time of revision surgery, the results did not attain statistical significance (p=0.373). Table II indicates the pattern of failure (disease spread) at the time of revision surgery in the group in whom the revision surgery could not be undertaken due to distant disease spread. There appears to be no difference in the patterns of failure based on the first surgery.

Table II. Patterns of disease spread in the inoperable patients compared to the factors under study, viz. type of primary surgery, time interval between the two surgeries and T-stage of the disease (n=22).

Patterns of failure	GB fossa	Lymph nodes	Peritoneum	Wound	Liver metastasis
Type of primary surgery
Laparoscopy (11)	2	6	6	1	1
Open (11)	1	7	8	1	1
Time interval
≤ 1 month (5)	1	4	2	0	0
2–3 months (14)	2	7	8	1	1
≥ 4 months (3)	0	2	2	1	1
T-stage
T1b (5)	1	5	4	0	0
T2 (9)	2	4	6	1	0
T3 (8)	0	4	4	1	2

Stage of the disease and relation to outcome

The predominant histology was adenocarcinoma, which was seen in 96% of the patients (see Table III). By pT-stage, 23 patients had T1b disease, while 33 and 20 patients had T2 and T3 disease, respectively. Of the 54 patients who were operable, 16 (pT1b = 3, pT2 = 6, pT3 = 7) patients had residual disease in the gallbladder fossa, lymph nodes, cystic duct margin, or the postoperative cutaneous scar. Table IV gives the distribution of residual disease in these 16 patients. In the inoperable patients (Table II), the incidence of peritoneal disease appeared to increase with advancing T-stage. Eight patients with T1b stage demonstrated the presence of tumor spread to the lymph nodes, while 3 of these 8 patients also had residual disease in the gallbladder fossa.

Table III. Histopathological characteristics of the primary tumor.

Characteristic	Total no. (n=76)
Adenocarcinoma
Well differentiated	10 (13.1%)
Moderately differentiated	39 (51.3%)
Poorly differentiated	24 (31.6%)
Papillary carcinoma	1 (1.3%)
Adenosquamous carcinoma	2 (2.6%)

Table IV. Comparison of pattern of residual disease with the T-stage in the 16 patients who underwent curative resection.

	Patterns of failure
T-stage	Gallbladder fossa	Lymph nodes	Incision/Port-site	Cystic duct
T1b (3)	2	3	0	0
T2 (6)	2	3	0	1
T3 (7)	5	6	2	2

Timing of second surgery

We broke down the continuous variable of time interval between the two surgeries into ≤1 month, 2–3 months, and ≥4 months. Table V gives the objective correlation between the time groups and the likelihood of operability. On comparing the time interval versus operability, there does not appear to be a significant difference between the time interval and operability. Table III gives the pattern of disease failure based on the time intervals. The incidence of peritoneal disease and the likelihood of finding lymph nodal disease appear to increase with increasing duration between the two surgeries.

Table V. Comparison of time interval between two surgeries with operability.

	Time interval between two surgeries
	≤1 month	2–3 months	≥4 months
Operable	14 (73.6%)	31 (68.8%)	9 (75%)
Inoperable	5 (26.4%)	14 (31.2%)	3 (25%)

Discussion

The incidence of gallbladder cancer in India has been estimated to be as high as 48/100,000 (Delhi) ¹³. In India, the disease is seen to afflict patients at a much younger age than in the rest of the world ¹⁴.

Effect of laparoscopic cholecystectomy versus open

Patients undergoing laparoscopic cholecystectomy have been reported to have a poorer clinical outcome than patients undergoing open surgery ⁶,⁷,⁸,⁹,¹⁵. Port site recurrences have been found in 17% of individuals following laparoscopic cholecystectomy ¹⁶. On closely examining these studies we find that the numbers of patients studied have been rather small. In fact, the study by Wakai et al. ⁶ has actually compared survival after laparoscopic simple cholecystectomy versus revision radical surgery in T2 disease only! On the other hand, De Aretxabala et al. ¹⁷ hinted that laparoscopic cholecystectomy may not actually have any adverse effect on survival.

The concerns that laparoscopy leads to tumor cell dissemination in the peritoneal cavity, as well as tumor cell implantation in the wounds, seem to have receded. Laparoscopy may not, therefore, affect operability adversely.

In the absence of significant evidence on the better outcome following laparoscopic surgery, we believe that it is not the laparoscopy or open technique that influences outcomes, but factors such as stage of the lesion and intra-operative events such as inadvertent perforation of the gall bladder with spillage of bile and improper specimen retrieval. Recently, the Johns Hopkins Group concluded that laparoscopic cholecystectomy may actually help to detect cancer earlier and improve overall survival (in the case of incidental gallbladder cancer) ¹⁸.

Effect of stage of disease

Controversy surrounds additional surgery for p T1b and T2 tumors. Some authors in the past have suggested that no further surgery is required after simple cholecystectomy ⁶,¹⁹,²⁰; this based on good results in patients with p T1b after simple cholecystectomy ²¹,²² as well as the absence of nodal disease in patients with T1 disease ²³,²⁴. However, it now seems that the advocates for radical revision surgery even for T1b and T2 disease ²⁵,²⁶,²⁷,²⁸,²⁹ are being proved correct based on studies examining lymphovascular invasion ³⁰,³¹,³².

We routinely perform revision radical surgery even for T1b tumors. This has been supported by our study reporting the presence of nodal disease in 35% patients with T1b disease at the time of revision radical surgery. In addition, we also report the increased risk of lymph nodal and peritoneal disease corresponding to the increase in T-stage of the disease.

It is clear from our study that increasing pathological stage does have a negative impact on the likelihood of operability at the time of second surgery. It also correlates well with the chances of finding residual disease. This has been seen to correlate with outcomes in previous studies ²,¹⁰,¹¹.

Impact of the time-interval between the two surgeries

The impact of the duration between primary surgery and revision on the likelihood of a successful radical second surgery has not been addressed by previous authors. It is logical to assume that with increased delay the chances of successful resection would decrease proportionately. When we attempted to divide this time interval into three groups, viz. durations of <1 month, 2–3 months, and >3 months, we found that there was no direct correlation between operability and time interval, although the pattern of disease in the inoperable patients clearly shows a trend toward finding peritoneal and lymph nodal disease as duration increased.

Conclusions

Laparoscopic cholecystectomy does not negatively influence the operability of revision surgery nor does it increase the possibility of finding residual disease compared to open surgery. Revision radical surgery should be performed whenever feasible in patients diagnosed with stage pT1 b and above. Increase in the T-stage is associated with an increased chance of finding residual disease at the time of revision radical surgery. And finally, the time interval between the primary surgery and the revision radical surgery in the case of incidental gallbladder cancer does not seem to affect operability.