B. AL-SARIREH*, S. SATHAPORN*, A. ROBINS#, D. JENKIN†, W. VASSANASIRI‡, M EL-SHEEMY‡, J. A. JIBRIL‡, D.
CLARK‡ and O. EREMIN*‡
*Section of Surgery, #Department of Immunology and †Department of Pathology, University Hospital, Nottingham and
‡Department of Surgery, Lincoln County Hospital, Lincoln, U.K.
Background and aims: Selective expression of cytokines by tumour infiltrating mononuclear cells has been reported in several tumours. 1,2 The local production of cytokines by such cells within the tumour microenvironment is crucial in mounting an immune response to tumour cells, and the presence of suppressive cytokines my hinder effector responses. Recently, several studies have focused on the cytokine network involved in the tumour microenvironment. 3-6 With regard to this, at least two distinct cytokine patterns are known to be generated by T lymphocytes: type 1 cytokines (TH1), which include IL-2 and IFN-c and have been demonstrated to promote cell-mediated immunity, and type 2 cytokines (TH2), which include IL-4, IL-5, IL-10, and IL-13 and have been shown to suppress cellular immune responses.7 Interleukin-10 has been associated with inhibition of a broad array of immune functions, such as T lymphocyte proliferation, type 1 cytokine production, antigen presentation and macrophage effector functions.8,9 Thus, because of its potential ‘protective’ effects on tumour cells, particularly via inhibition of specific tumour-reactive cytotoxic T lymphocyte,7 IL-10 production and secretion may be reasonably supposed to be up-regulated in cancer patients. In fact, high frequency of IL-10 mRNA and low frequency IL-2 and IFN-c mRNAs have been detected by means of RT-PCR analysis in various tumour samples including breast carcinoma.3,6 Also, increased serum levels of IL-10 have been demonstrated in patients with different histotypes of solid and haematopoietic tumours and these levels have been shown to correlate with extent of disease;8-11 in addition, it has been suggested that IL-10 may be released not only by immune cells but directly by tumour cells because serum levels of this cytokine often correlate with tumour burden, whereas surgical excision of neoplasia may be followed by reduction in IL-10 serum levels.12 These findings indicate that IL-10 is likely to be involved in suppressing the host's antitumour immune responses. To the best of our knowledge, there are few reports regarding IL-10 levels in breast carcinoma patients,3,13 none of which have investigated its source in these patients. Therefore, in this study, we have examined the expression patterns of IL-10 in women with localized operable breast carcinoma. Materials and methods: Sera from 10 women with clinically localized and operable breast cancer (prior to surgery) and from age- and sex-matched healthy controls were collected. Serum levels of IL-10 have been determined by ELISA using specific Quantikine Immunoassay Kits (R&D System, UK). In an attempt to identify the source of IL-10, its expression in paraffin-embedded breast cancer tissue sections (primary tumours and tumour draining lymph nodes) have been examined using standard immunohistochemical method and goat polyclonal antihuman IL-10 antibodies (R&DSystems, UK). Results and discussion: As anticipated, the current study a positive correlation between the presence of breast carcinomas and high IL-10 concentrations in serum, when compared with healthy controls. Similar results recently have been reported in patients with different histotypes of solid and hematopoietic tumours 10, 11, 14-18 suggesting that IL-10 overproduction may be a shared survival strategy of several types of human malignancies. Most authors have suggested that the main source of IL-10 in patients with malignancies is the tumour itself rather than the inflammatory infiltrates based on the following observations: (1) high frequency of IL-10 mRNA has been detected in several tumours3,13 (2) expresion and secretion of IL-10 by various tumour cell lines19, 20 and (3) tumour shrinkage was paralleled by a decrease in IL-10 levels.10 By contrast, in breast carcinoma as we demonstrated in our study using immunohistochemistry, the main source of IL-10 was tumour infiltrating leukocytes mainly macrophages and not tumour cells. Although not all the tumour infiltrating macrophages but significantly considerable number, as identified by morphology and staining with the specific macrophage marker CD68, was demonstrated to express IL-10. This might indicate that certain subset/s of tumour infiltrating macrophages (i.e. suppressor macrophages) are the main source of IL-10 in tumour bearing hosts resulting in induction of immune suppression and subsequently progressive tumour growth.
REFERENCES
OTHER PRESENTATIONS
Periprosthetc fractures of the femur
S.P. KALE and L. READ
Walsall Hospital, Walsall, U.K.
Background: As life expectancy grows and the number of primary and revision arthroplasty procedures continue to rise, so too does the incidence of periprosthetic fractures. The management of these fractures is quite complex, and often involves surgery. A wide variety of surgical techniques are described to treat these types of injuries. Patients and Outcomes: A series of 20 cases of periprosthetic fractures of the femur from one unit in Birmingham is presented. In this series, female patients outnumbered males, the commonest aetiology being trivial trauma or a fall. Thirteen cases were fractures of the femur around previous total hip prosthesis, four cases were in the proximity of hemiarthroplasty implants and three cases were supra-condylar fractures above total knee arthroplasty prosthesis. Eighteen cases were subjected to open reduction and internal fixation with bone grafting done in fourteen patients. It is difficult to quantify outcomes but the majority had a good result at six months using the Beals and Tower criteria. We had a low complication rate with a majority of the patients progressing to graduated full weight bearing. Summary: The classification systems for such fractures, principles of management and various treatment options with our recommendations are outlined with reference to the abnormal biomechanics of the reconstructed extremity.
Prediction of cardiac risk prior in abdominal aortic surgery:
comparison of a revised Goldman cardiac risk index and
MUGA scan
C.D. KARKOS,* ‡ G.J.L. THOMSON,* R. HUGHES, * S.
HOLLIS,‡ J.C. HILL,# and U.S. MUKHOPAHYAY *
Department of *Vascular Surgery and #Nuclear Medicine, Royal
Preston Hospital, Preston, U.K., †Department of Medical
Statistics, University of Lancaster, Lancaster, UK, and
‡University of Thessalia Medical School, Larissa, Greece
Objective: Lee et al1 suggested that in patients undergoing major non-cardiac surgery, a revised Goldman cardiac risk index can identify those at higher risk for cardiac complications. The aim of this study was to test the validity of this model in an independent series of patients undergoing abdominal aortic surgery and to compare the index with the MUGA scan in predicting cardiac complications. Methods: A series of 77 patients that underwent cardiac assessment with MUGA scan prior to elective abdominal aortic reconstruction were retrospectively analysed. The revised index was calculated for each patient after recording the following five preoperative risk factors (RFs): history of ischaemic heart disease, congestive heart failure, cerebrovascular disease, IDDM, and creatinine>177µmol/l. Technetium-99m MUGA scan provided information about the resting left ventricular ejection fraction (EF%) and the presence of regional wall motion abnormalities. Results: Fourteen patients (18%), all with ³2RFs, developed cardiac complications. The index predicted post-operative cardiac events (p=0.008). An abnormal EF% failed to predict a cardiac event (p=0.1). The presence of wall abnormalities, either alone or combined with an abnormal EF%, predicted cardiac complications (p=0.004 and p=0.006, respectively). Patients with a higher index score showed a tendency to have a lower EF% (Spearman's rank correlation =-0.43, p<0.001). Wall abnormalities, with or without an abnormal EF%, were more frequent in patients with a higher score (³2RFs). Combining the index with the EF% or the wall abnormalities could further stratify the cardiac risk in patients with ³2RFs (?2-test for trend, p=0.004 and p=0.0003, respectively). Conclusions: For those undergoing elective abdominal aortic surgery, the revised Goldman Cardiac Risk Index is a simple method of evaluating cardiac risk with minimum resource implications. MUGA scan can offer additional stratification in patients judged by the index to be at high risk.
REFERENCE
Optimal number of transplanted Schwann cells for nerve
regeneration
A. MOSAHEBI
Blond McIndoe Labs, Royal Free and VCL Medical School,
London, U.K.
Background and aim: Inadequate peripheral nerve regeneration is a major cause of morbidity in surgical practice. This may happen following trauma or tumour resection. Nerve graft is the current method of microsurgical repair of nerve deficits, which leads to poor functional recovery. It is well established that Schwann cells, glial cells of the peripheral nerves, are essential for nerve regeneration. The average number of Schwann cells in the peripheral nerve graft has been estimated as 20x106/ml. The aim of this study was to establish the optimal number of Schwann cells needed for neuronal regeneration, which may explain the poor results, obtained by current mode of surgical repair. Methods: Purified cultured Schwann cells were obtained using recombinant glial growth factor and cells were genetically marked by lacZ transduction for identification from the host. A 10-mm gap in the rat sciatic nerve was repaired using a bioresorbable conduit polyhydroxybutyrate either without Schwann cells or with varying concentration of syngeneic Schwann cells at 20, 40, 80 or 160x106/ml. There were 6 animals in each group, the animals were sacrificed at 3 weeks post repair and conduits harvested. Immunohistochemistry was used to look for axonal regeneration (Pan neurofilament) and Schwann cells (S100). Results: Transplanted Schwann cells could be identified from hosts at all the concentrations and were seen to be related to the regenerating axons. The axonal regeneration was increased when conduits contained Schwann cells. This increase continued up to 80x106/ml and there was a decrease in regeneration distance at 160x106/ml, compared with the preceding concentration. Conclusion: The optimal Schwann cell concentration for axonal regeneration was 80x106/ml, which is 4 folds higher then the number of Schwann cells present in the nerve graft. Tissue engineering of nerve conduits seeded with cultured Schwann cells may represent a new therapeutic strategy to achieve better clinical results in peripheral nerve repair.
Case control study on nutritional support to post-operative
patients with PEG and NG tube feeding
M.M. REDDY
Monklands Hospital, Airdrie, U.K.
Aim: To compare the two modes of nutritional support, percutaneous endoscopic gastrostomy (PEG) feeding and nasogastric tube (NG) feeding, for patients who underwent major oropharyngeal cancer resection reconstruction and radiotherapy. Patients and methods: Fifty patients with oral cancer involving various sites like tongue, floor of the mouth, soft palate, retromolar trigone, who were treated at two centres are included in this study. Among them 25 patients were fed by NG tube at Monklands District General Hospital and another 25 patients were fed by PEG tube at Falkirk Royal Infirmary, postoperatively. The haemoglobin, total protein, and albumin levels of all patients were measured both pre-operatively and postoperatively at the time of discharge. Results: The patients with PEG tube feeding had a greater weight gain and nearly same preoperative levels of haemoglobin and total protein, at the time of discharge when compared with patients with NG tube feeding. One patient who had PEG done died due to peritonitis. The patients with NG tube feeding had oesophageal irritation, tube displacements and blockage. The complications with NG tube feeding were relatively more frequent than with PEG tube feeding. Conclusion: Percutaneous endoscopic gastrostomy tube feeding is a safe and effective method of providing long term enteral nutrition to oropharyngeal cancer patients and offers advantages over NG feeding.
Defective dendritic cells in patients with breast cancer
S. SATTHAPORN *‡, B. AL-SARIREH *, A. ROBINS #, W.
VASSANASIRI †, M. EL-SHEEMY †, J.A. JIBRIL †, D.
CLARK †, D. VALERIO † and O. EREMIN *†
*Section of Surgery and #Department of Immunology, Queen's
Medical Centre, University of Nottingham, and †Department of
Surgery, Lincoln County Hospital, Lincoln, U.K. ‡ Funded by the
Royal Thai Army
Introduction: Dendritic cells (DCs) play a crucial role in presenting antigens to T lymphocytes and inducing cytotoxic T cells, including those directed against tumours. The function of DCs have been studied in patients with breast cancer in order to understand the factors leading to failure of an effective systemic and loco-regional anti-cancer host response and resultant progressive tumour cell growth. Methods: DCs were obtained from the peripheral blood (PB) and lymph nodes (LNs) of women with operable breast cancer using immuno-magnetic bead selection. The stimulatory capacity of DCs in the allogeneic mixed leukocyte reaction (MLR) and autologous T cell proliferation test purified protein derivative (PPD) as stimulator), the expression of surface markers and the production of cytokines in vitro by DCs from patients with breast cancer and normal healthy donors (controls) were determined and compared. Results: Using the above methodology, 70-75% purified DCs (PB and LNs) were isolated. PBDCs and LNDCs from patients with breast cancer demonstrated a significantly lower capacity to stimulate in an MLR, compared with PBDCs from controls (p<0.05). Also, antigen-driven autologous T cell proliferation in patients with breast cancer had a significantly decreased ability to respond to PPD, when compared with PBDCs from controls (p<0.05). T cells from breast cancer patients, however, responded as well as control T lymphocytes in the presence of control DCs. PBDCs and LNDCs from patients with breast cancer expressed significantly lower levels of HLADR and CD86, and induced decreased amounts of interleukin-12 secretion in vitro, compared with DCs from normal donors (p<0.05). Conclusion: These data suggest a defective, switched off DC function in patients with operable breast cancer. These may be important factors inducing inhibition of anti-cancer host defences and resulting in progressive growth seen in patients with breast cancer.
Attitudes of surgical research fellows in the UK - a national
questionnaire survey
N.B. TEO*, C.S. SEOW#, C. WILSON#, W.D. GEORGE#
*Pathology Department, University of Liverpool, Duncan
Building, Liverpool, #Surgical Department, University of
Glasgow, Western Infirmary, Glasgow, U.K.
Background and aims: A research fellowship in surgery is a stepping stone to higher surgical training and a successful career in academic surgery. Our aim was to evaluate the nature of the research fellowships and the attitudes of the research fellows towards research in surgical training. Methods: Questionnaire surveys of all research fellows working in an academic department of surgery in the UK. Main outcome measures include the nature of the post, principal source of funding, clinical commitment, attitudes of research fellows towards the role of research in surgical training, aspirations in academic surgery and the current situation in comparison with guidelines in the Calman Report. Results: The response rate was 74% (91 out of 123 surveyed). The majority of the research fellows are male (83.5%), in possession of a surgical diploma (95.6%), junior trainees (63.7% were SHO's) working on projects of a minimum 2 years duration (79.1%) with an element of basic science (81.5%). Of the research fellow posts available by specialty, the commonest were colorectal surgery, vascular surgery and breast surgery. Of the intended specialty, the top three choices were colorectal surgery, vascular surgery and upper gastrointestinal surgery. Grant-awarding bodies (including the Royal Colleges of Surgeons), hospital trusts and private industry each accounted for about a third of the funding of the surgical research fellowships. Pursuit of research projects is mostly driven by a need for career advancement (93.4%). Few trainees, (6.6%) view the intellectual challenge as the sole reason for embarking on surgical research; 41.8% of the respondents have expressed an interest in academic surgery. Majority (85.8%) felt that the desirable duration spent in research should be between 12 to 24 months. Conclusion: More funding is required for surgical research into clinically oriented projects. Research, as yet, is to be incorporated into the higher surgical training system as recommended by the Calman Report.
A risk modelling study for carotid endarterectomy
G. KUHAN
Vascular Laboratories, Hull Royal Infirmary, Hull, U.K.
Background: A 30 day stroke or death rate of 3-10% has been reported in patients undergoing carotid endarterectomy (CEA). The aim of this study was to identify the risk factors that influenced the outcome and to develop a logistic regression model that can aid in the comparative audit of CEA. Methods: 836 CEAs performed by four vascular surgeons from 1992 to 1999 were analysed; 95 procedures were excluded because of incomplete data. The median age of the population was 69 years (range 38-86) and the male to female ratio was 1.6. Sixty-seven possible risk factors (46 pre-operative and 21 operative) were collected and of these fifteen risk factors were selected for the analysis. Multiple logistic regression was used to model the effect of risk factors on the 30-day stroke/ death rate. The outcome after risk adjustment was compared for surgeons and two vascular units using this model. Results: The overall 30-day stroke/death rate was 3.9% (29/741). Regression modelling identified heart disease, diabetes and stroke at presentation as significant risk factors. A risk score of 1 was assigned to each of these risk factors to produce a logistic regression equation: ln (P/1-P)= 0.9294(Risk score) - 4.2726. (P= probability of occurrence of stroke/death within 30 days.) A risk score of greater than equal to two identified the higher risk group of patients as shown in the table.
| Risk score | 30 day Stroke/Death risk(%) | 95% confidence intervals |
|---|---|---|
| 0 | 1.4 | 0.9 - 2.1 |
| 1 | 3.4 | 1.3 - 9.0 |
| 2 | 8.3 | 3.2 - 22 |
| 3 | 18.8 | 10.2 - 35.3 |
The observed 30 day stroke/death rate for four vascular surgeons were 2.9, 4.0, 4.2 and 4.3, respectively. The variation among the surgeons was not statistically significant after risk adjustment using the above model. The observed outcomes for the two vascular units were 3.8 and 4.2, respectively. The difference between the 2 units was not statistically significant after risk adjustment. Conclusions: Multiple logistic regression models can successfully identify patients at higher risk from CEA, and aid in the comparative audit of surgeons and vascular units.
Hypomethylation of chromosome 1 heterochromatin
correlates with frequent 1q copy number gain in hepato-cellular carcinoma
P.B.S. LAI*, N.WONG#, W.-C. LAM#, E. PANG#, J.W.Y.
LAU* and P.J. JOHNSON#
*Department of Surgery, #Department of Clinical Oncology,
Prince of Wales Hospital, Chinese University of Hong Kong,
Hong Kong SAR, China
Background: We have previously shown by comparative genomic hybridization (CGH) that there is a high incidence of 1q copy number gain in hepatocellular carcinoma (HCC). A similar finding was also reported in other cancers, such as breast cancer, ovarian carcinoma, multiple myeloma, and Wilm's tumour. Cytogenetic study on these tumours has indicated frequent 1q unbalanced translocations and isochromosome formation. Hypomethylation of the heterochromatic region on chromosome 1 (1qh) has been postulated as the underlying cause in the frequent 1q copy number gain detected in many solid tumours. In this paper, we have, therefore, investigated the methylation status on 1qh in HCC with an aim to correlating DNA hypomethylation with copy number gain detected. Material and method: Thirty-two histologically confirmed HCC's were studied. Hypomethylation of satellite DNA on 1qh was analysed by Southern blotting. DNA extracted from tumour was digested by methyl-sensitive restriction enzyme BstB1 and probed against the satellite 2 (Sat2), a major DNA component of the chromosome 1 heterochromatin. Liver and sperm DNA were used as methylated and hypomethylated standards, respectively. In parallel, CGH analysis was carried out on HCC cases studied. Results: A correlation between 1q copy number gain and the corresponding 1qh methylation status in 32 HCC cases studied was investigated and results were as follows:
| Hypomethylation +ve | Hypomethylation -ve | |
|---|---|---|
| Chromosome 1q gain +ve | 20 | 0 |
| Chromosome 1q gain -ve | 4 | 8 |
Fisher's Exact test; P = <0.0001
Conclusion: We have found a strong correlation between the hypomethylated repetitive sequences on 1qh and 1q copy number gain in HCC. It is possible that such hypomethylation alters the interaction between the CpG-rich satellite DNA and chromatin proteins, resulting in heterochromatin decondensation, breakage and finally the aberrant 1q found.
Comparison of surgeons and sonographers in ultrasound
measurement
P. GALLAGHER*, S. ELLIOTT# and R. CHARNLEY*
The Departments of *Surgery and #Radiology, Freeman Hospital
Newcastle-upon-Tyne, U.K.
Introduction: Surgeon performed ultrasound is being used with increasing frequency. A method of assessing one aspect of sonography, measurement error, is presented. Patients and Methods: Nineteen adult patients undergoing abdominal ultrasonography had triplicate measurements of gallbladder dimensions performed by a consultant radiologist and a surgical trainee. The gallbladder is relatively easy to locate with ultrasound, but accurate measurement requires precise probe positioning and manipulation. Each observer examined subjects in immediate succession using the same ultrasound machine. Observers were blinded until the end of the study. Intra-observer error rates were calculated for each dimension using analysis of variance techniques, and expressed as within-subjects standard deviation and a repeatability coefficient (the value that exceeds the difference between replicate measurements in 95% of cases). Results: For both intra- and inter-observer errors the estimated within-subject standard deviation and the repeatability were comparable between the trainee and experienced sonographer for all dimensions (Table 1).
Table 1: Intra-observer error
| Dimension | Observer* | Estimated# | Repeatability(cm) |
|---|---|---|---|
| Length | Sonographer (6.3cm) | 0.22 | 0.6 |
| Surgeon (6.1cm) | 0.17 | 0.5 | |
| Height | Sonographer (2.4cm) | 0.14 | 0.4 |
| Surgeon (2.7cm) | 0.14 | 0.4 | |
| Width | Sonographer (2.6cm) | 0.12 | 0.3 |
| Surgeon (2.5cm) | 0.14 | 0.4 |
*(mean measurement of dimension included for reference); #within-subjects standard deviation
Conclusion: Surgeons are not radiologists but do use ultrasound as a technical adjunct. The ability to measure accurately is one essential skill that must be developed. This comparison study demonstrates that specific facets of ultrasonography can be learnt and assessed by surgeons. An individual can repeat it over time to appraise improvement, or as a comparison between observers.
©2001 The Royal College of Surgeons of Edinburgh, J.R.Coll.Surg.Edinb.
