ANZ Journal of Surgery

Background:  Regional and local anaesthetic techniques are thought to improve postoperative pain control and functional outcomes following total knee arthroplasty, potentially leading to a reduction in hospital length of stay.

Objectives:  The objective of this study was to evaluate the reporting quality and discuss the clinical findings of the available literature on these modalities that included length of stay as a study outcome.

Data sources:  The electronic databases Pubmed, Scopus, Medline, Web of Science and Cochrane library were searched using key words.

Review methods:  Eight-hundred and forty-three papers were identified in the search. Fifteen of these met the inclusion criteria. Eight further studies were identified from their reference lists to give a final total of 23 studies that were reported against the consolidated standards of reporting trials (CONSORT) 2001 statement checklist.

Results:  The mean criteria CONSORT score was 17.3/22 (79%). The majority of studies that compared femoral nerve blocks with placebo or conventional pain management modalities failed to demonstrate a significant reduction in length of stay. All studies that compared femoral nerve blocks with epidurals found no significant difference in length of stay. Only half of the studies comparing local anaesthetic techniques to placebo or conventional pain management methods found a significant reduction in length of stay.

Conclusions:  The reporting quality has specific deficiencies in the areas of sample size calculation, randomization whilst there was under-reporting of blinding. Regional and local anaesthetic techniques have not demonstrated a clear reduction in hospital length of stay. Epidurals and femoral nerve blocks have similar impacts on length of stay.

Background:  The histological characteristics of follicular thyroid carcinomas (FTCs) are important predictors of prognosis, and lesions can be classified as either minimally invasive follicular carcinoma (MIFC) or widely invasive follicular carcinoma (WIFC) based on histopathological characteristics. There has been controversy surrounding the histological classification of FTC, which can present challenges to clinicians attempting to deliver accurate prognostic information to their patients. The aim of the present study was to examine cases of metastatic FTC for characteristics that may predict aggressive tumour behaviour.

Methods:  The Monash University Endocrine Surgery Unit database was searched for patients with FTC. The histopathology reports were collated for these patients to confirm the diagnosis of FTC, classify patients into MIFC versus WIFC, and examine for key characteristics such as the capsular and/or vascular invasion. The thyroid specimens from patients with metastatic FTC were examined by reviewing pathologists. It was hypothesized that patients with metastatic disease would likely have WIFC as their primary lesion.

Results:  There were 64 patients with FTC identified during the period of 1997–2009. Of these, 10 patients were found to have metastatic disease. On review of the histopathology, three patients were found to have WIFC,four patients had MIFC and three patients did not have definite features of FTC found in the thyroid gland.

Conclusion:  Currently accepted histological classification of FTC is inadequate and fails to accurately predict patients with distant metastatic disease and a more aggressive clinical course. It is thus the policy of our unit to recommend total thyroidectomy and radioactive iodine ablation for all patients with FTC.

Background:  The purpose of the study was to develop a system of risk stratification, based on clinical and histological factors that would aid prediction of metastasis from cutaneous squamous cell carcinoma (SCC) of the head and neck.

Method:  The method used was a retrospective case control study comparing clinical and histological parameters of 78 patients who developed metastasis with 92 patients who did not develop metastasis over a 5-year period.

Results:  The two ‘absolute’ (highest) risk factors for development of metastatic disease are poor histological differentiation and perineural/lymphovascular infiltration. The three ‘relative’ risk factors are moderate histological differentiation, diameter ≥20 mm and Clark level 5.

Risk stratification:  High-risk lesions have either one of the absolute risk factors or all three of the relative risk factors with a predicted incidence of metastasis of 37%. Intermediate risk lesions have two of three relative risk factors and a predicted incidence of metastasis of 5%. Low-risk lesions have one or none of the relative risk factors and a predicted incidence of meatstasis of 0.3%.

Conclusion:  Ongoing management of patients with histo-pathologically proven invasive SCC of the head and neck should be based upon risk stratification for metastasis.

Backgroud:  The study aims to discuss indications for conversion to thoracotomy in completely thoracoscopic lobectomy.

Methods:  From September 2006 to April 2010, 306 patients (164 men, 142 women, median age 58.1 years, range 15 to 86 years) underwent completely thoracoscopic lobectomy. There were 223 cases of primary lung cancer, 11 other malignant diseases and 72 cases of benign disease. The steps of the thoracoscopic procedures are almost identical to those of traditional open lobectomy, which requires standard mediastinal lymph node dissection for primary lung cancer patients. When conversion to an open procedure is necessary, such as in the presence of lymph node adhesions or metastases and bleeding, operative incisions are extended 12–15 cm towards lower angle of the scapula, retractors are used to separate the ribs, and the procedure is completely under direct visualization.

Results:  All procedures were performed without significant complications or intraoperative deaths. The average surgical duration was 195 min, and average blood loss was 256 mL with no blood transfusions required. The average chest tube drainage duration was 7.45 days. The average post-operative hospital stay was 10.34 days. There were 27 cases (8.8%) of conversion to open thoracotomy, for the reasons of interference by lymph nodes (n= 18), bleeding (n= 4), inflammatory adhesions of arteries (n= 3) and large size tumours (n= 2).

Conclusion:  Adhesions or lymph node metastases and bleeding were the most important causes of conversion to thoracotomy in completely thoracoscopic lobectomy. Large tumours, fused fissures and dense pleural adhesions can always be managed thoracoscopically.

Background:  The Whipple's procedure (WP) is a major operation that adds a further demand on the body's nutritional reserves and therefore body composition after the effect of pancreatic cancer. The aim was to document changes in body composition changes that occur during the first six months after a WP for a pancreatic cancer malignancy.

Methods:  Twenty-seven (14 males, 13 females) consecutive WP patients had body composition measured at baseline and then at 2, 5, 14 and 26 weeks after surgery. These included; anthropometric measure (weight, skin folds and arm muscle area (AMA)), total body measures of protein (TBP), potassium (TBK), water (TBW) and fat mass (FM). Changes were compared using repeated measures analysis of variance.

Results:  Hospital nutritional care maintained TBP and TBK but at 2 weeks there was a loss of FM (P= 0.037). The nadir of weight loss (P < 0.001) occurred at 5 weeks because of losses of protein (P= 0.007), fat (P < 0.001) and potassium (P= 0.045) but not water. Although weight and FM were still significantly less than baseline measures at 26 weeks weight, TBP, TBK and AMA were not significantly different to preoperative values.

Conclusions:  Although at 6 months, important measures of the metabolically functioning tissue, TBP and TBK, have returned to preoperative values significant losses occurred during the first 3 weeks after discharge from hospital and FM did not return to preoperative values. These results suggest the need to improve post-discharge nutritional care.

Introduction:  Various models have been proposed to effectively provide acute surgical care in Australasia. Recently, General Surgeons Australia (GSA) has published a 12-point plan with guiding principles on this matter. This study describes a model of providing acute general surgical care in a high-volume institution, evaluates clinical outcomes and critically appraises the system against the GSA 12-point plan.

Methods:  The acute care system is qualitatively described with quantitative measures of workload. The outcomes of acute laparoscopic cholecystectomy were used as a proxy of system performance. The system was critically appraised against the GSA 12-point plan.

Results:  Teams are on call once per week with each surgeon on call once per fortnight. The three key elements of acute management – collecting patients, post-acute ward round and operating – are treated as modules. The patient remains under the care of the admitting consultant but is often operated on by another team. From June 2009 to 2010, there were 7429 acute general surgical admissions (mean: 20.4 patients per day) with 2999 acute operations (mean: 8.4 operations per day). The other activities of the department were not compromised. In that time, 388 acute laparoscopic cholecystectomies were performed with a conversion rate of 1.3% and no major bile duct injury. The system is compatible with the GSA 12-point plan.

Conclusion:  This study describes an efficient and safe system for providing acute general surgical care in a high-volume setting with satisfactory clinical outcomes. It is compatible with the GSA 12-point plan.

Background:  The demands of surgical training, learning and service delivery compete with the need to minimize fatigue and maintain an acceptable lifestyle. The optimal balance of working hours is uncertain. This study aimed to define the appropriate hours to meet these requirements according to trainees.

Methods:  All Australian and New Zealand surgical trainees were surveyed. Roster structures, weekly working hours and weekly ‘sleep loss hours’ (<8 per night) because of 24-h calls were defined. These work practices were then correlated with sufficiency of training time, time for study, fatigue and its impacts, and work–life balance preferences. Multivariate and univariate analyses were performed.

Results:  The response rate was 55.3% with responders representative of the total trainee body. Trainees who worked median 60 h/week (interquartile range: 55–65) considered their work hours to be appropriate for ‘technical’ and ‘non-technical’ training needs compared with 55 h/week (interquartile range: 50–60) regarded as appropriate for study/research needs. Working ≥65 h/week, or accruing ≥5.5 weekly ‘sleep loss hours’, was associated with increased fatigue, reduced ability to study, more frequent dozing while driving and impaired concentration at work. Trainees who considered they had an appropriate work–life balance worked median 55 h/week.

Conclusions:  Approximately, 60 h/week proved an appropriate balance of working hours for surgical training, although study and lifestyle demands are better met at around 55 h/week. Sleep loss is an important determinant of fatigue and its impacts, and work hours should not be considered in isolation.

The Australian and New Zealand Society for Vascular Surgery has incorporated a constitutional change to administer a self-funded compulsory vascular surgery audit since January 2010. This is a bi-national quality assurance activity that captures all procedures performed in both countries. Data is collected at two points in the clinical admission; at operation and at discharge and data entry is via the Internet. Security is stringent and confidentiality is guaranteed by Commonwealth privilege. Data privacy is maximized by encryption. The application is flexible and administered by a dedicated administrator with a help-desk facility. Reports are available to provide real-time feedback of user performance compared with the peer group data in key categories of arterial surgery. A structured hierarchy for data management has been established to assess four main categories of performance: mortality after aortic surgery, stroke and death after carotid surgery, patency and limb salvage after infrainguinal bypass and patency after arteriovenous access for haemodialysis. Data is analysed using risk-adjustment techniques and an algorithm for management of underperformance has been followed. Data validation has been performed. The outcomes in all categories have been of a high standard and correction of erroneous data in a single statistical outlier has negated underperformance. The audit has captured only 65% of the estimated procedures in Australia in the first year, but data quality is good. The feasibility of a complete compulsory bi-national audit has now been established and will be the benchmark for other craft groups in the current environment of accountability.

Background:  The study aims to assess the initial experience of laparoscopic radical prostatectomy (LRP) in a regional centre in Australia which includes Fellowship training during our first 50 cases.

Methods:  Data were collected prospectively from our first 50 consecutive patients who underwent LRP for localized prostate cancer between September 2009 and October 2010. All cases were performed or supervised by the primary surgeon. Patient details, operative details, complications, early oncological and functional outcomes were analysed.

Results:  The median age was 65 (45–76) years and median preoperative prostate-specific antigen was 7.5 (2.5–23) ng/mL, with palpable disease present in 48%. Using D'Amico's risk stratification, 14%, 74% and 12% were in low, intermediate and high-risk categories, respectively. Forty percent of cases were training cases with a median of 5 (2–8) of 10 operative steps performed by the Fellow. There was one open conversion and no rectal injuries. Mean operative time was 288 (175–440) min and with blood transfusion rate of 6%. Mean length of stay was 2.5 (1–6) days. Positive surgical margin rates for pT2 and pT3 disease were 14% and 52%, respectively, although for the last 25 cases they were 7% and 30%, respectively. Continence rate was 86% at 6 months, and 45% and 33% of preoperatively potent patients were potent after bilateral and unilateral nerve preservation at 6 months.

Conclusion:  LRP has been safely introduced in a regional centre with establishment of a Fellowship training programme, with early results comparable with other open, laparoscopic and robotic series.

Background:  Positron emission tomography/computed tomography (PET/CT) using F18-fluorodeoxyglucose has been shown to be valuable in the management of malignant disease. The aim of this study is to investigate the impact of this technique on the management of patients with resectable pancreatic tumours.

Methods:  Thirty-six patients with 37 potentially resectable pancreatic tumours on diagnostic CT imaging underwent PET/CT scans. Operative findings, histological reports and/or clinical follow-up served as standard of reference. The impact of PET/CT on patient management was estimated by calculating the percentage of patients whose treatment plan was altered due to PET/CT.

Results:  Pancreatic adenocarcinoma was diagnosed in 30 patients, neuroendocrine tumours in 3, mass-forming pancreatitis in 3 and serous cystadenoma in 1. The median standard uptake (max) value was 5.0 (range 2.2–12.0). Sensitivity and specificity of detecting extrapancreatic metastatic disease were 73% and 100%, respectively. Three occult liver metastases were detected at laparotomy following negative PET/CT. PET/CT findings influenced the management of 8 (22%) patients – 3 with liver metastases, 3 with bone metastases, 1 with lymph node metastases and 1 by identifying the benign appearance of the pancreatic tumour.

Conclusion:  PET/CT achieves a significant diagnostic impact in detecting extrapancreatic metastatic disease. F18-fluorodeoxyglucose PET/CT appears to be useful in assessing suspicious pancreatic masses.

Skull base surgery has gone through significant changes with the development of extended endoscopic endonasal approaches over the last decade. Initially used for the transphenoidal removal of hypophyseal adenomas, the endoscopic transnasal approach gradually evolved into a way of accessing the whole ventral skull base. Improved visualization, avoidance of brain retraction, the ability to access directly tumours with minimal damage to critical neurosurgical structures as well lack of external scars are among its obvious benefits. However, it presents the surgeons with a number of challenges, including the need to deal endoscopically with potential arterial bleeding, complicated reconstruction requirements as well as the need for a true team approach. In this review drawing from our experience as well as published series, we present an overview of current indications, challenges and limitations of the expanded endonasal approaches to the skull base.

Introduction:  Serious complications can result when casts are used for bone immobilization following fracture. Adequate patient information regarding cast care and possible complications is vital for prevention. This study examines the effectiveness of verbal and written patient information regarding cast safety.

Methods:  Patients (n= 109; age ≥18 years) from three Western Australian teaching hospitals were interviewed using a custom-designed questionnaire. Patients' understanding of cast care and possible complications were tested by recall of seven categories of information, notably: pain, swelling, cast care, itching, neural signs, vascular signs/symptoms, exercise/rest. A follow-up phone call (3–8 weeks after initial interview) was conducted to elicit complications and determine information recall.

Results:  Written information was received by 62% of patients; however, overall, only 35% claimed to have read the information provided. Of these, the highest recall was in four of seven information categories. A high proportion of those given only verbal information had poor recall (≤2 categories, defined a priori). In contrast, patients who also received written information had better recall (three or more categories, defined a priori, P= 0.031). Four of the 109 patients developed complications attributable to the cast (three had pressure ulcers and one had a deep vein thrombosis).

Conclusion:  Patient recall of information concerning cast care and possible complications was no more than 60%. The provision of written information was associated with a significantly higher awareness of possible complications. The results indicate that for fracture care, the delivery and use of information protocols can be greatly improved.

Background:  Conventional laparoscopic Nissen fundoplication (CLF) is generally considered the surgical approach of choice for gastro-oesophageal reflux disease. Robotic-assisted laparoscopic fundoplication (RALF) has recently been introduced into laparoscopic clinical practice with the aim of improving surgical performance by eliminating tremors and fatigue. A meta-analysis of randomized clinical trials (RCTs) was performed to compare RALF and CLF.

Methods:  Medline, Embase, ISI Web of Knowledge CPCI-S and The Cochrane Library were searched and the methodological quality of included trials was evaluated. Outcomes evaluated were intraoperative, dysphagia, flatulence, antisecretory medication, satisfaction with intervention, operation time, hospital stay and total cost. Results were pooled in meta-analyses as risk ratios and weighted mean differences (WMD).

Results:  Of 221 patients in six RCTs, 111 were allocated to CLF and 110 to RALF. RALF prolonged total time necessary to carry out fundoplication (WMD 3.17 (95% confidence interval. 2.33–4.00) min; P < 0.00001, χ2P = 0.25, I2 = 24%). Operation complication, antisecretory medication, satisfaction with intervention, the time needed for hiatal dissection, the time from incision to completion of sutures, the total operation time and total cost were similar in both groups.

Conclusion:  Clinical outcomes from RALF were comparable to CLF approach, but RALF prolonged the operation time. Currently, CLF should be routinely used as costs are lower.

Background:  Incidence rates of colorectal cancer (CRC) in New Zealand rank among the highest worldwide. Internationally, there has been evidence of a shift in colon cancer from left- to right-sided. The objective of this study was to determine trends in left- and right-sided colon and rectal cancers incidence by sex, age and ethnicity.

Methods:  Using datasets created by linking data from the New Zealand Cancer Registry to the census data, we analysed a total of 47 694 CRCs from 1981 to 2004. Cancers were divided into right-sided colon (cecum to the splenic flexure); left-sided colon (descending and sigmoid colon); and rectal (rectosigmoid junction and rectum).

Results:  Left- and right-sided colon, and rectal cancer incidence rates increased by 13–20% among men. In women, colon cancer rates increased by 25% for right-sided cancers, decreased by 8% for left-sided cancers and remained unchanged for rectal cancers. This corresponds with an increase in right-sided cancers from 57% to 65% of total colon cancers in women. The incidence of all CRCs increased at a faster rate among Māori than non-Māori.

Conclusion:  We identified a left- to right-sided shift in colon cancer limited to women over the age of 65. While Māori trends in site distribution parallel those of their non-Māori counterparts, the rapid increase in Māori incidence rates is noteworthy. It is unclear why such shifts in CRC site distribution are occurring.

Background:  The need for surgical excision of benign papillary lesions diagnosed on core biopsy remains debatable. This lack of consensus arises because although there is a possibility of histological underestimation, there are as yet no reliable predictors of malignancy. We therefore aimed to evaluate the incidence of histological underestimation in our practice, and to identify factors that predict for this, in order to reduce unnecessary surgery without missing out on possible malignancy.

Methods:  Retrospective review of 106 patients diagnosed with a papillary lesion on percutaneous image-guided core biopsy was performed between 1 January 2005 and 31 December 2008. The presence of atypia on core biopsy and the presence of malignancy in the surgical specimen were correlated with standard clinical, radiological and pathological features.

Results:  Histological underestimation occurred in 15 of 81 patients (19%). Malignancy was more likely when atypia was present in the core biopsy (P= 0.04, OR 5.17). Otherwise, a final diagnosis of malignancy was not correlated with any clinical or radiological features (P > 0.05). The presence of atypia was also not correlated with any clinical or radiological features.

Conclusion:  In our study, 19% of patients with a benign papillary lesion diagnosed on core biopsy were found to have atypical ductal hyperplasia or malignancy following surgery. In view of this, together with the absence of reliable predictive factors for malignancy, we recommend surgical excision of all papillary lesions diagnosed on core biopsy.

Background:  Most potentially preventable haemorrhagic deaths occur within 6 h of injury. Conventionally, blood component therapy delivery is measured by 24-h cumulative totals and ratios. The study aim was to examine the effect of a massive transfusion protocol (MTP) on early (6 h) balanced component therapy.

Methods:  An 88-month retrospective clinical study at a level 1 trauma centre was undertaken, examining consecutive trauma patients receiving ≥10 units of packed red blood cells (PRBCs) within 24 h, before (pre-MTP) and after implementation of MTP. Demographic data, injury severity score (ISS), abbreviated injury scale (AIS), shock parameters, coagulation profile, the need for surgical intervention (<24 h), mortality and intensive care unit length of stay were collected. The ratios of blood products given by 6 h, by 24 h and the time between administrations of components was collected and analysed.

Results:  Pre-MTP and MTP patients had similar demographics, shock severity and initial laboratory findings. Despite MTP patients having had a higher ISS (42 ± 12 versus 36 ± 12, P < 0.05) and AIS head score (2.6 ± 1.8 versus 1.6 ± 2.0, P < 0.05), there was no difference in mortality. Area under the curve (AUC) of the MTP period showed earlier delivery of higher median ratios of fresh frozen plasma (FFP)/PRBC (P= 0.004). Similar findings were found for cryoprecipitate/PRBC and platelet/PRBC ratios. By 24 h, the AUC for FFP/PRBC ratios were no different.

Discussion:  Implementation of MTP resulted in earlier balanced transfusion. The difference between the FFP/PRBC ratios of the two types of resuscitations levelled by 24 h. The efficacy of component therapy delivery should be measured earlier than 24 h.

Background:  Trastuzumab increases disease-free and overall survival in HER-2-positive, early breast cancer. In 2007, the National Breast and Ovarian Cancer Centre recommended that patients with HER-2 positive cancers (node positive or node negative tumours >1 cm) be offered adjuvant trastuzumab with chemotherapy. The aim of this study was to evaluate recent trends in trastuzumab therapy in Australia and New Zealand.

Methods:  Following data were obtained from the National Breast Cancer Audit for patients treated between 2006 and 2008: tumour size, number of cases recorded per surgeon per year, location of hospital, HER-2 receptor status, age, lymph node status, chemotherapy and trastuzumab treatment.

Results:  Data were available from 23 290 patients. During the study period, the percentage of breast cancers tested for HER-2 rose from 77% to 91%. Patients over 70 had fewer HER-2 tests than their younger counterparts. Fourteen percent of tumours were HER-2 positive; the proportion treated with trastuzumab in 2006, 2007 and 2008 was 50%, 66% and 74%, respectively. Significantly more node-positive patients (77%) were given trastuzumab than node-negative patients (52%). All the patients prescribed trastuzumab also received chemotherapy. Patients under 70 years, patients treated in Australia and patients treated by higher caseload surgeons were more likely to be prescribed trastuzumab than those over 70, patients in New Zealand and patients treated by lower caseload surgeons.

Conclusions:  Trastuzumab-prescribing trends conform to the published guidelines. However, older patients and those with HER-2 positive, node-negative tumours >1 cm may be undertreated in some cases.

Background:  Bimodal electric tissue ablation (BETA) incorporates the process of electrolysis into radiofrequency ablation (RFA) to increase the size of tissue ablation. This study investigated whether BETA could increase the efficacy of the Cool-Tip RF system (Covidien, Boulder, CO, USA) to produce larger ablations. It also investigated whether applying electrolysis only during the pretreatment phase (called electrochemical treatment (ECT)/RFA group) is as effective as BETA (where electrolysis was used during both the pretreatment and RFA phases).

Methods:  A Cool-Tip RF system (Covidien) was used to test three types of ablations (RFA, BETA, and ECT/RFA) in a pig liver model. In BETA, 9 V of direct current was provided for 10 min, after which the RF generator was started and both electrical circuits were allowed to run concurrently. In ECT/RFA, however, the direct current circuit was switched off after 10 min of pretreatment and only RFA was performed as described above. Ablation sizes were measured in three dimensions.

Results:  The size of ablations (transverse diameter A and B) produced by BETA and ECT/RFA was significantly larger compared with standard RFA (P < 0/001). BETA also created larger ablations compared with ECT/RFA (P < 0.001).

Conclusion:  BETA could improve the efficacy of the Cool-Tip RF system (Covidien) to achieve larger ablations. The increased tissue hydration improved delivery of electrical energy to the tissues and delayed the process of desiccation, thus allowing the ablation process to continue for longer periods of time to produce larger ablations. BETA could be used to treat larger liver tumours more effectively than standard RFA.

Background:  The history of military surgery claims many forebears. The first surgeon-soldiers were Homer's Machaon and Podalirius, followed a thousand years later by the Roman surgeons-general, Antonius Musa and Euphorbus; and later, e.g. Ambrose Paré, John Hunter and Sir John Pringle; and the 19th century innovators, Dominique-Jean Larrey (France), Friedrich von Esmarch (Prussia) and the Russian, Nikolai Pirogoff. The singular feature that distinguished modern military surgery from its earlier practice was the use of gunpowder. It was one of two inventions (the other was printing) that by the empowerment of individuals, lifted Western humankind from the medieval to the modern era.

Methods:  Research of primary and secondary archives.

Results and conclusion:  Gunpowder was first used in European warfare at Algeceras (1344–1368). Hitherto, the destruction of tissue had been the result of (relative) low-energy wounding with tissue damage caused by incisional or crushing wounds. The founder of modern surgery, Master John of Arderne (1307–1380), wrote of his experience gained as a military surgeon on the battlefield at Crecy (1346). Following Crecy, Arderne was the only chronicler who described the origins of the Prince of Wales's feathers as a royal and later commercial symbol, and the motto ‘Ich Dien’, ‘I serve’, as that of hospitals in the Western World. Later advances in military surgery incorporated both clinical experimentation and the innovation of new systems of pre-hospital battlefield care.

Background:  To meet Australia's future demands, surgical training in the private sector will be required. The aim of this study was to estimate the time and lost opportunity cost of training in the private sector.

Methods:  A literature search identified studies that compared the operation time required by a supervised trainee with a consultant. This time was costed using a business model.

Results:  In 22 studies (34 operations), the median operation duration of a supervised trainee was 34% longer than the consultant. To complete a private training list in the same time as a consultant list, one major case would have to be dropped. A consultant's average lost opportunity cost was $1186 per list ($106 698 per year). Training in rooms and administration requirements increased this to $155 618 per year. To train 400 trainees in the private sector to college standards would require 54 000 training lists per year. The consultants' national lost opportunity cost would be $137 million per year. The average lost hospital case payment was $5894 per list, or $330 million per year nationally. The total lost opportunity cost of surgical training in the private sector would be about $467 million per year. When trainee salaries, other specialties and indirect expenses are included, the total cost will be substantially greater.

Conclusion:  It is unlikely that surgeons or hospitals will be prepared to absorb these costs. There needs to be a public debate about the funding implications of surgical training in the private sector.