97
INTERPRETING 3-DIMENSIONAL STRUCTURES FROM 2-DIMENSIONAL IMAGES: A WEB-BASED INTERACTIVE 3D MODEL OF THE LIVER TO ENHANCE SURGICAL RESIDENTS' SPATIAL UNDERSTANDING OF STRUCTURAL INTER-RELATIONSHIPS. J.L. Crossingham, J. Jenkinson, N. Woolridge, S. Gallinger, C.A.E. Moulton, G. Tait. Biomedical Communications, Institute of Medical Science, Faculty of Medicine, University of Toronto; Department of Surgery, Division of General Surgery, Toronto General Hospital, Toronto, Ont.
Learning intrahepatic anatomy and developing an understanding of the relationships that exist between the key structures is a difficult process that is required of surgical residents. They need to mentally reconstruct 3D images from available computed tomography (CT) scans, and this may not be an effective way of understanding the liver. A web-based interactive 3D model of the liver was created to facilitate understanding of the complex spatial anatomy of the liver and to help visualize this anatomy in 3D when viewing CT scans. By importing CT scans into Osirix, 3D surface renderings of the liver were obtained. Using these images as reference, anatomic structures were modelled in Cinema4D. This included the liver surface and the intrahepatic structures: portal veins, hepatic veins, hepatic arteries and the biliary system. Users can view common liver anatomy and common variations online in interactive the 3D rotational model to observe the complex interactions of the vascular and biliary systems. This model will be useful for surgical trainees learning the difficult and complicated intrahepatic anatomy and will optimize learning opportunities for all trainees requiring knowledge of liver structures.
98
PERIOPERATIVE ANALYSIS OF LAPAROSCOPIC VERSUS OPEN LIVER RESECTIONS. A.J. Rowe, P.A. Schumacher, A. Buczkowski, A.T. Meneghetti, O.N.M. Panton, C.H. Scudamore, S.W. Chung. Department of Surgery, University of British Columbia, Vancouver, BC.
A prospective series of 18 laparoscopic liver resections performed from 2005 to 2007 was evaluated to determine whether a minimally invasive approach is superior to traditional open liver resection in terms of perioperative clinical outcomes and resource utilization.
These were compared with an equivalent group of 12 consecutive open liver resections undertaken immediately before the introduction of laparoscopic liver resection. The outcomes were evaluated for differences in perioperative morbidity, hospital length of stay and operative costs.
All patients had 1 or more lesions confined to the same side of the liver, including both benign and malignant pathology. There were no perioperative deaths in either group, but patients undergoing open resection had significantly more intraoperative blood loss (473 mL v. 287 mL, p < 0.03) and more postoperative complications (42% v. 6%, p < 0.03), compared with the laparoscopic group. There was a trend toward reduced hospital length of stay in patients undergoing laparoscopic resection (4.3 d v. 5.8 d, p < 0.07). There was no significant difference in skin-to-skin operating time or total time in the operating theatre (138 min for open resection v. 135 min for laparoscopic resection, and 224 min v. 214 min, respectively). The analysis of instrumentation comparing the Echelon stapler and Xcel trocar used for laparoscopic resection and the GIA and TX staplers used for open resection demonstrated that the 2 approaches were equivalent in terms of cost.
Our data suggest that laparoscopic liver resection is associated with less perioperative morbidity than open liver resection and that the 2 approaches are equivalent in terms of resource utilization. For segmental resections, a laparoscopic approach appears to be a superior technique. Further studies are needed to determine whether these results can be generalized to more complex liver resections.
99
TIME TO COMPLICATIONS IN PATIENTS AWAITING LAPAROSCOPIC CHOLECYSTECTOMY. A. Barnes, C.D. Mercer, W.M. Hopman, M. Mercer, D. Jalink. Department of Surgery, Queen's University, Kingston, Ont.
An acceptable wait time has yet to be established for elective laparoscopic cholecystectomy. This prospective analysis examines the relationship between time spent on the waiting list and risk of an adverse event, defined as a complication of untreated biliary disease leading to an emergency department visit.
Data were collected prospectively from 337 patients awaiting elective cholecystectomy. Rate of adverse events was analyzed in relation to duration on the waiting list by applying previously used time intervals, as well as intervals devised by the Ontario Wait Time Strategy for general surgery cases.
Fifty-four patients (16.0%) experienced at least 1 adverse event while awaiting surgery. Using waiting time intervals developed by the Ontario Wait Time Strategy, the highest risk of events occurred between 4 and 12 weeks on the waiting list (odds ratio = 1.2, as compared with the reference category [weeks 0–4]). Patients who experienced events waited an average of 104.7 days for surgery, compared with 118.4 days for the remaining patients (p = 0.452).
Patients appear to be at greatest risk of events after spending 4 weeks on the waiting list, but this risk falls notably after 12 weeks. This evidence supports the Ontario Wait Time Strategy guidelines, which indicate that patients with symptomatic biliary disease should undergo surgery within 12 weeks of presentation. It may also point toward a more indolent course of illness with minimal chance of future complications among patients who do not experience an adverse event within 12 weeks. Time to surgery is similar regardless of whether or not an event is experienced while on the waiting list.
100
HEPATIC INJURY: 10-YEAR REVIEW FROM A PROVINCIAL TRAUMA CENTRE. M.E. Goecke, C.H. Scudamore, A. Buczkowski, R. Simons. Trauma Services, Hepatobiliary Surgery, Vancouver General Hospital, Vancouver, BC.
The majority of patients with liver injury can be managed successfully nonoperatively; others require damage-control surgery (packing), and a few need heroic surgical interventions unfamiliar to many surgeons. Over the last decade, several new approaches to complex hepatic trauma have emerged including damage-control concepts, veno-venous bypass and stapled nonanatomic resection. The purpose of this study is to report on evolving trends in hepatic trauma management, the utility of new surgical technologies and their impact on overall survival.
The trauma registry and discharge abstract database of a level-1 provincial trauma centre were reviewed to identify all patients who suffered traumatic liver injury between January 1, 1997, and August 31, 2007. A retrospective chart review of these patients was completed to determine trends in management strategy (operative or nonoperative, operative approach), predictors of failure of nonoperative management and patient outcomes with subgroup analysis of higher-grade injuries and operative approach.
In total, 605 patients with liver trauma were identified, which accounted for 3% of the total trauma volume; 66% were male with an average age of 37 years and 54% had an abbreviated injury score ≥ 3. Blunt trauma accounted for 83% of cases, of which 81% were motor vehicle collisions. Nonoperative management was attempted in 61%. Mortality from all causes was 19% with a trend toward improved survival over time for higher-grade injuries.
Nonoperative management is successful in the majority of patients with liver trauma. Those requiring operative intervention can usually be stabilized using damage-control techniques, notably packing. New operative approaches appear to offer survival benefit to those few patients requiring more definitive intervention.
101
CLINICAL EPIDEMIOLOGICAL ANALYSIS OF 309 ERCPS PERFORMED IN NOVA SCOTIA IN 2006 FOR CHOLEDOCHOLITHIASIS AND GALLSTONE PANCREATITIS. M. Molinari. Dalhousie University, Halifax, NS.
The ideal management of choledocholithiasis and gallstone pancreatitis is controversial. With the improvement of diagnostic imaging studies of the biliopancreatic duct, retrograde cholangiopancreatography (ERCP) should ideally be performed only as a therapeutic modality. Referral patterns for ERCP and patient management are influenced by resources and location. Limited information is currently available for clinical use and results of ERCP in Atlantic Canada. The aim of this study was to assess the diagnostic, therapeutic and success rates of ERCP performed in a high-volume tertiary medical centre in Nova Scotia.
A retrospective observational study was performed over a 1-year period (January 1 to December 31, 2006) at the Queen Elizabeth II Hospital. ERCP data of 565 procedures were reviewed by 2 independent investigators and entered into computerized databases. All data discrepancies were analyzed and resolved. Demographic characteristics, indications for ERCP, interventions, diagnosis and outcomes were captured. Procedures were then divided into 2 groups: therapeutic or diagnostic. Therapeutic ERCPs were defined when successful stone extraction or sphincterotomy for sphincter dysfunction were obtained.
Within the 12-month period, 309 ERCPs were carried out for presumed choledocholithiasis (273, 88.3%) or gall stone pancreatitis (36, 21.7%) on a population with mean age of 62 (standard deviation 17.9) years. Therapeutic ERCP was performed in 175 (64.1%) patients with choledocholithiasis and in 12 (33.4%) individuals with a pre-ERCP diagnosis of pancreatitis. Inability to cannulate the papilla was observed in 8 (2.9%) patients with choledocholithiasis and 4 (11.1%) individuals with a pre-ERCP diagnosis of pancreatitis.
Because ERCP is an invasive procedure, ideally it should be indicated only as therapeutic modality. Our findings suggest that in Atlantic Canada, diagnostic ERCPs are still performed in 35.9% of cases of choledocholithiasis and 66.6% of gallstone pancreatitis. Implementing better referral patterns and patient selection would optimize resources and decrease complication rates for ERCP in the Maritimes.
102
IS BIOPSY NECESSARY IN THE MANAGEMENT OF LIVER LESIONS? P. Renfrew, M. Smith, T. Hamilton, M. Molinari, M.J. Walsh. Department of Surgery, Queen Elizabeth II Health Sciences Centre, Dalhousie University, Halifax, NS.
Our aim was to investigate whether biopsy of liver lesions is necessary in their management. The majority of liver lesions found either incidentally on imaging or with screening follow-up for a history of malignancy are a common occurrence. The management of these lesions can be a challenging clinical problem. This usually involves further imaging, blood work and the formulation of a treatment plan.
We conducted a retrospective chart review of 201 patients presenting with liver lesions or biliary strictures from July 2004 to June 2007. For each patient we assessed whether imaging modalities (ultrasound [U/S], computed tomography [CT], magnetic resonance [MR]), tumour markers (CA 19–9, CEA, AFP) or liver biopsies were conducted. We then examined whether these investigations had significant impact on the treatment for these patients.
Of the 201 patients, 104 (52%) had U/S, 181 (90%) had CT and 55 (27%) had MR imaging. A total of 128 (64%) had blood work for tumour markers, and 29 (14%) patients had liver biopsies. After the initial clinical assessment, 72 (36%) patients required further imaging to facilitate diagnosis and characterization of the lesion before treatment. However, with further imaging, only 2 patients required liver biopsies for diagnosis. There were 109 (54%) patients that received treatment either in the form of a surgical intervention, referral to medical oncology, transarterial chemoembolization (TACE) or radio frequency ablation (RFA). For those receiving treatment, the majority (85 patients) did not have a liver biopsy. Additionally, there was a disparity between the biopsy result and the final pathologic diagnosis in 4 of 29 (14%) cases; in comparison, 16 of 201 (8%) cases were misdiagnosed radiologically.
Radiological modalities and tumour markers are sufficient for diagnosing and managing the vast majority of liver lesions. A hepatobiliary surgeon should be involved in the triage of these lesions, including the need for biopsy. It is a rare occurrence that biopsy of liver lesions is ever necessary.
103
SOCIETAL REINTEGRATION AND ECONOMIC IMPLICATIONS FOLLOWING ORTHOTOPIC LIVER TRANSPLANT. M. Molinari. Dalhousie University. Halifax, NS.
Data on the impact of orthotopic liver transplant (OLT) on patients' integration in their communities and its financial consequences at a personal and familial level are lacking. Primary aims of this study were to assess the employment rate, productivity and activity index of patients after OLT in Atlantic Canada. From September 2006 to January 2007, a cross-sectional study was performed using validated Work Productivity and Societal Reintegration questionnaires. Participants were interviewed by phone or during follow-up visits and were adults at least 3 years post-OLT, without communication impairments. All data were prospectively collected. Categorical data were analyzed by χ2 and Student's t test for continuous variables; p values less than 0.05 with 2-tail distribution were considered significant. Among the eligible 158 patients, 47 were randomly selected. Forty-five (95%) participated and 2 declined for personal reasons. Fifteen patients (33%) were employed full time and 12 (52%) were employed in the same position as before OLT. Thirteen individuals (28%) were on disability while the remaining were students or housekeepers. Eleven patients (24%) had retired. After OLT, 19 patients (42%) experienced financial restrictions preventing access to care or necessary medications due to significant reduction in their annual personal ($28300 v. $26500, p < 0.005) and household income ($44000 v. $43700, p < 0.05). Overall, 77% of patients reported that OLT had a negative impact on their financial status in comparison with their pre-OLT status (p < 0.005). However, the productivity and activity index showed overall acceptance of their current societal role, and sequlae of OLT had little impact on activities of daily life at this time. In Atlantic Canada, patients undergoing OLT for end-stage liver disease have an employment rate of 33%, consistent with the available literature. Socially, 84% of patients are satisfied with their current role in society and are able to perform most daily activities without difficulty.
104
EARLY SURROGATE END POINTS PREDICT LONG-TERM OUTCOMES IN ORTHOTOPIC LIVER TRANSPLANTATION. M. Boutros, M. Cantarovich, S. Paraskevas, M. Deschenes, P. Ghali, P. Wong, M. Fernandez, P. Chaudhury, P. Metrakos, J. Tchervenkov, J. Barkun. Division of General Surgery, Department of Surgery, Multi-Organ Transplant Program, Division of Gastroenterology, Department of Medicine, McGill University Health Centre, Montréal, Que.
As graft and patient survival continue to improve, there is an increasing need for early end points that are predictive of long-term outcomes in orthotopic liver transplantation (OLT). We propose dynamic changes in renal function as well as the use of clinical benefit states (CBS), a composite measure combining outcomes as “syndromes” rather than isolated complications, to predict long-term outcomes in OLT.
Using the MUHC-Transplant database, we identified all OLT recipients, excluding retransplants. We assessed the association between early changes in renal function and rejection with long-term outcomes. In addition, we created CBS defined as combinations of renal dysfunction (dichotomized creatinine-clearance [CrCl]) and rejection (liver biopsies) post-OLT. All relationships were assessed using Student's t test and χ2 analyses.
Four-hundred and ninety-six OLT recipients, over a 17-year period (1990–2007), had a mean age of 55 years, a mean MELD of 23.2 and 69% were male. Long-term renal function: At 1 month, the best predictor of CrCl at 1 year was a decline in CrCl ≤ 30% (66.1 v. 58.2 mL/min/1.73 m2, p < 0.01). At 6 months, however, the best predictor of CrCl until 4 years was a CBS defined by a decline in CrCl ≤ 20% and mild rejection (69.1 v.62.3 mL/min/1.73 m2, p < 0.023). Graft survival (GS): At 1 month, the best predictor of GS was a decline in CrCl ≤ 30% (OR 1.91, CI 1.13–3.22). However at 3 months, the best predictor of GS was a CBS defined by a decline in CrCl ≤ 50% and moderate rejection (OR 2.67, CI 1.29–5.56). Similarly, the best predictor of patient survival at 1 month is a CBS defined by a decline in CrCl ≤ 40% and moderate rejection (OR 2.02, CI 1.20–3.40). Finally, hepatitis-C recurrence was most strongly predicted by a CBS at 3 months post-OLT defined by a decline in CrCl ≤ 50% and mild rejection (OR 2.04, CI 1.02–4.25).
Certain key declines in early renal dysfunction are associated with long-term graft and patient survival. However, short-term combined end points defined as CBS are most strongly associated with traditional long-term clinical outcomes.
105
QOL EVALUATION OF PATIENTS UNDERGOING TACE FOR HCC. M. Molinari. Dalhousie University, Halifax, NS.
Two randomized controlled studies have shown that for patients who are not candidates for hepatic resection or liver transplantation for hepatocellular carcinoma (HCC), significant survival benefit is obtained by the use of transcatheter arterial chemoembolisation (TACE). Still, studies on the quality of life (QOL) of these patients are scarce.
The aim of this study was to assess QOL of a prospective cohort of patients treated by TACE for HCC in North America.
From January 2006 until September 2007, a consecutive cohort of 26 patients was diagnosed with nonresectable HCC. TACE was performed by selective chemoembolisation of the tumour(s) with doxorubicin and polyvinyl alcohol (PVA) particles every 3 months. The World Health Organization Quality of Life (WHOQOL)-BREF questionnaire was used to collect data on the QOL of all patients before undergoing TACE and every 3 months after treatment. Patients' demographic characteristics, etiology of underlying cirrhosis, tumour size and location were collected prospectively.
The average population age was 60.4 (standard deviation [SD] 7.5) years old; 23 were males and 2 females. Average serum α-feto protein was 137 μg/L (SD 43 378), tumour size was 7.1 cm (SD 3.9) and the average number of TACE sessions was 1.2 (range 1–3). Overall QOL score before TACE was 89 (SD 13.7), at 3 months was 96 (SD 8.1), at 6 months was 91 (SD 15.7) and at 9 months was 107 (SD 17.6). Before TACE, the average physical health scores was 20 (SD 4.5), at 3 months was 21 (SD 4.1), at 6 months was 22 (SD 3.7) and was 25 (SD 2.8) at 9 months (p = NS). The interval psychological health, social relationships and environment scores are reported in Table 1 (p = NS).
Table 1
In this study, the overall QOL of patients undergoing TACE for HCC remained stable over time and appears not to be affected by the number of embolic sessions performed. These findings confirm results from previous studies from Asian centres.
106
BILE DUCT INJURIES ASSOCIATED WITH LAPAROSCOPIC CHOLECYSTECTOMY: TIMING OF REPAIR DETERMINES LONG-TERM OUTCOMES. S.C. Chow, A.K. Sahajpal, E. Dixon, P.D. Greig, S. Gallinger, A.C. Wei. Hepatobiliary and Pancreatic Surgical Group, Division of General Surgery, Toronto General Hospital, Mount Sinai Hospital, University of Toronto, Toronto, Ont.
Bile duct injury is a serious complication following laparoscopic cholecystectomy (LC). The objective of this study is to report a large institutional experience with laparoscopic cholecystectomy–associated bile duct injury (LC-BDI).
Patients who underwent a surgical repair of a LC-BDI between 1992 and 2007 were identified by retrospective chart review at 2 university-affiliated hospitals. Risk factors for postoperative complications were analyzed by univariate analysis.
Sixty-nine patients were identified, 24 men and 45 women, with a median age of 49 years. Thirteen immediate (within 72 hours of LC), 34 intermediate (between 72 hours and 6 weeks) and 22 late (after 6 weeks) LC-BDI repairs were performed. The LC-BDIs were Strasberg type: A in 1 (1%), D in 2 (3%), E1 in 22 (32%), E2 in 16 (23%), E3 in 22 patients (32%), E4 in 4 (6%) and E5 in 2 patients (3%). Forty-one hepaticojejunostomies (59%), 24 choledochojejunostomies (35%), 3 right hepatic lobectomies with biliary reconstruction (4%) and 1 primary common bile duct repair (1%) were performed. The overall morbidity and perioperative mortality rates were 30% (21 patients) and 1% (1 patient), respectively. Twelve patients (17%) developed short-term complications; the most common was cholangitis, which occurred in 7 patients (10%). There was no significant relationship between timing of LC-BDI repair and perioperative morbidity (p = 0.95). The most frequent long-term complication was biliary stricture, which occurred in 10 patients (14%). Patients repaired in the intermediate period were more likely to develop biliary stricture than patients repaired in the immediate or late periods (p = 0.03).
The timing of LC-BDI repair was identified as an important determinant of long-term outcome. Intermediate period (between 72 hours and 6 weeks) LC-BDI repairs were associated with the development of postoperative biliary stricture. Thus LC-BDI repairs should be undertaken either in the immediate (within 0–72 hours) or late (> 6 weeks) period after cholecystectomy.
107
A RETROSPECTIVE ANALYSIS OF HEPATOCELLULAR CARCINOMA PATIENTS: TREATMENT MODALITIES AND SURVIVAL. P.A. Schumacher, J. Powell, A.J. Rowe, A. Buczkowski, S. Ho, C.H. Scudamore, A. Weiss, S.W. Chung. Vancouver General Hospital, University of British Columbia, Vancouver, BC.
Existing treatment modalities for hepatocellular carcinoma (HCC) include resection, radiofrequency ablation (RFA), ethanol injection (EI), chemotherapy (CTx) and transarterial chemoembolization (TACE). This study was performed to evaluate the survival benefit of these modalities as primary or salvage therapy.
A retrospective review was conducted on 251 consecutive patients treated for HCC between 1996 and 2006 at Vancouver General Hospital (VGH) and the BC Cancer Agency (BCCA). Data on 247 patients for whom there was a complete data set were analyzed. Data were retrieved from clinical charts and information systems from VGH and BCCA. All patients underwent primary treatment by resection, RFA, EI, CTx, TACE or observation. A subset with persistent or recurrent disease underwent salvage therapy by 1 of these modalities. Survival analysis was performed using standard statistical methods.
Mean overall survival was 76.8 months. Factors associated with poorer survival were presence of symptoms at diagnosis, chronic HCV versus HBV infection, lack of antiviral therapy in early TNM stage HBV/HCV patients, portal venous thrombosis (PVT), poorer Child–Pugh status and higher TNM stage (all p < 0.001). Among primary treatment modalities, survival was comparable for resection, RFA and EI and significantly poorer for CTx, TACE or observation. Among salvage treatment modalities, RFA of dominant lesions was associated with improved survival (Table 1).
Table 1
These results underscore the value of early detection of HCC in at-risk patients. The data suggest that patients may be stratified based on tumour stage and underlying liver function to curative intent or disease control strategies to optimize survival and minimize operative risk.
108
HEPATIC RESECTION IN CANADA FROM 1995 TO 2004: RATE, GEOGRAPHIC VARIATION AND HOSPITAL VOLUMES. R. McColl, X. You, E. Dixon. Department of Surgery, University of Calgary, Calgary, Alta.
Liver resection is the only curative therapy for hepatic malignancy, both primary and secondary. Despite this, the rates of hepatic resection across Canada are unknown. This study sought to describe patient characteristics and crude outcomes along with the rate and regional variation of hepatic resection in Canada, its provinces and census divisions from 1995 to 2004.
Discharge data from all hospitals across Canada except Quebec were obtained from the Canadian Institute for Health Information for 1995–2004. All patients undergoing a hepatic resection were identified using ICD 9 and 10 codes. Rates and regional variations in rates of hepatic resection were calculated and reported by province and census division using the postal code conversion file. Baseline demographics, patient characteristics and outcomes are reported.
The national age-and sex-adjusted hepatic resection rate per 100 000 people aged ≥ 18 years increased from 3.22 in 1995 to 5.86 in 2004. Provincial rates in 2004 varied from a low of 2.88 in Prince Edward Island to a high of 8.91 in the Territories. For census divisions, rates varied even more from a low of 0 in 76 divisions to a high of 94.85. There were 247 hospitals performing hepatic resections across Canada, with a range of 1–1185 cases per hospital. Eighty-nine percent of cases took place at high-volume centres (defined as those hospitals with case volumes in the 4th quartile). There was a progressive decline in the in-hospital mortality rate over the study period.
There is significant regional and geographic variation in the rates of hepatic resection across Canada. Disparity in access to centres performing hepatic resection may partially explain these results. Our study also demonstrates a pattern of regionalization which may be due to growing evidence that high-volume centres have superior outcomes for complex procedures.