SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix

The intra-operative blood loss of 50 consecutive gynae-oncology patients undergoing surgery for endometrial, cervical or ovarian cancer was cell salvaged and filtered. In each case blood samples were taken from the effluent tumour vein, a central venous line, the cell saver reservoir, the cell salvage re-transfusion bag after processing but before filtration and from the cell salvage re-transfusion bag after processing and filtration. Samples were examined using immunohistochemical monoclonal antibody markers for epithelial cell lines. Viable, nucleated malignant cells were detected in 2/50 central venous samples, 34/50 reservoir samples and 31/50 unfiltered cell salvaged samples. After passage through a Pall RS leucocyte depletion filter no remaining viable, nucleated malignant cells were detected in any sample. The clinical risks of cell salvage in these circumstances should be reviewed in the light of the risks of allogeneic blood transfusion.

Cell salvage is a well recognised and increasingly utilised technique in many branches of surgery. Patients benefit by avoiding the risks of homologous (allogeneic) blood transfusion [1] and the NHS benefits from conserving the supply of blood and, potentially, reducing costs [2]. Allogeneic transfusion has been shown to be associated with both tumour recurrence [3, 4] and postoperative infections [5], and these associations are not seen with autologous transfusion [6–10]. In the UK pre-operative autologous donation and acute normovolaemic haemodilution are not currently recommended by the National Blood Service except in exceptional circumstances [11].

Until recently, the use of autologous transfusion by intra-operative cell salvage (IOCS) during cancer surgery has been controversial because of the theoretical risk of disseminating disease by re-transfusing malignant cells which may have been aspirated into the cell saver from the surgical field and not subsequently removed by the cell salvage and filtration process. However, in April 2008 the use of cell salvage in urological cancers was endorsed by the National Institute for Health and Clinical Excellence (NICE) [12] who state that there is no evidence for these concerns.

We undertook this observational study to investigate the ability of leucocyte depletion filtration and cell salvage to remove malignant cells during gynaecological cancer surgery. We used mouse monoclonal antibodies against human cytokeratins to label tumour cells of epithelial origin in blood sampled from the patients’ circulation and in blood aspirated from the surgical field during surgery for gynaecological malignancies. The presence of viable, nucleated malignant cells in the final, filtered samples would suggest that the use of cell salvage during surgery for gynaecological malignancy would risk re-transfusion of such cells, with an unknown risk of causing metastatic spread of disease. We aimed to establish whether or not total clearance of all labelled malignant cells was achievable in all cases.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix

After LREC approval, 50 consecutive gynae-oncology patients scheduled for major pelvic surgery for ovarian, endometrial or cervical cancer were recruited. Written informed consent was obtained during discussion with the principle investigator (SC). Detailed written information was given, emphasising that no cell salvaged blood would be returned to the patients, but that the cell saver would be used to collect blood lost during the procedure and that this blood along with two samples of blood from the patients’ circulation, would be tested for the presence of malignant cells. All patients underwent total abdominal hysterectomy with bilateral salpingo-oophorectomy with or without omentectomy, or pelvic clearance as clinically indicated. All patients received combined general and regional anaesthesia with invasive monitoring including direct arterial and internal jugular CVP lines, and were managed postoperatively on HDU or ICU. CVP lines were placed for clinical reasons, not just for sampling. During the procedure all blood loss was retrieved using the Haemonetics Cell Saver 5 (Haemonetics UK Ltd, Leeds, UK). Heparin 30 000 units in 1000 ml of 0.9% saline was used as the anticoagulant for suction of the surgical site, and in addition blood was retrieved from the surgical swabs after washing in 0.9% saline according to standard operating procedures for the machine. The Pall RS leucodepletion filter (Pall Corporation, Portsmouth, UK) was used as a final filtration step after cell salvage processing.

During surgery five 10-ml blood samples were taken and labelled as follows: effluent tumour vein sample taken by the operating surgeon under direct vision prior to clamping the vein; central venous sample from the CVP line; unprocessed blood from the operative field prior to cell salvage processing, taken from the reservoir of the cell salvage machine; cell salvaged blood after processing according to standard operating procedures; and cell salvaged blood after processing and filtration through the leucocyte depletion filter.

Samples were labelled with monoclonal mouse anti-cytokeratin antibodies to identify cells of epithelial origin, prepared as slides and examined using light microscopy [13–15]. This technique is a highly sensitive and specific system for the identification of cancer cells of epithelial origin [16–19]. Further details of the laboratory methodology are provided in the Appendix.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix

Malignant cells of epithelial origin were clearly identifiable in two of 50 blood samples from the central venous circulation, in 34 of 50 samples from the cell salvage reservoir prior to processing, and in 31 of 50 samples of the processed blood prior to leucocyte depletion filtration (see Table 1). In no case were viable, nucleated malignant cells detectable in the final, filtered sample of cell saved blood.

Table 1.   Detection of malignant cells using anti-human cytokeratin monoclonal antibody labelling in blood sampled from 50 patients undergoing gynaecological cancer surgery.
Label usedEffluent tumour veinsCVP lineCell saver reservoir (before processing)Cell saver after processingCell saver after processing and filtration
MNF116AE1AE3MNF116AE1AE3MNF116AE1AE3MNF116AE1AE3MNF116AE1AE3
Viable nucleated cells detected001263453100
Viable nucleated cells not detected50504948431543185050
Non-viable fragments detected (un-nucleated)0000696501
Samples unusable (processing problem)0000112100

Fragmented cytoplasmic debris from labelled cells was detectable in several of the reservoir and pre-filter samples and in one of the post-filter samples, indicating destruction of the labelled cells during the cell salvage process. Such fragments could not cause tumour spread as they lack a nucleus and cannot divide.

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix

In 50 patients with gynaecological malignancies we found no instance of viable, nucleated malignant cells passing intact through the combined cell salvage and filter system. Of particular note is the finding of labelled epithelial cells in the central venous samples, indicating circulating malignant cells at the time of surgery, which were then undetectable in the final, filtered samples.

When it was first introduced, cell salvage was assumed to be contra-indicated in any situation in which there was a risk of contamination of the re-infused blood – particularly with amniotic fluid, infectious agents and malignant cells [20] and this view has been subsequently restated throughout the literature. As a result the technique was not used in obstetrics, bowel surgery and cancer surgery. These contra-indications, while sensible and intuitive at a time when cell salvage was new, were not supported by any evidence, and are now being revised with the benefit of the results of studies on the ability of cell salvage systems to remove such contaminants. The acceptance of the safety of cell salvage in obstetrics began with in-vitro clearance studies [21, 22] and reports of successful cases in which the technique had been used where there was ‘no alternative’ [23]. A survey conducted by the Obstetric Anaesthetists Association in 2006 [Teig M, Clarke V, Catling S. OAA Survey of UK Maternity Units. Poster Presentation, Obstetric Anaesthesia 2007, Sheffield, UK] showing that 38% of UK maternity units were using cell salvage regularly in 2005 and that 28% had incorporated cell salvage into their major haemorrhage protocol illustrates the extent to which this theoretical contra-indication is now disregarded in practice by clinicians. In a series of 152 penetrating abdominal injuries there was no difference in wound infection rates between patients receiving allogeneic or cell saved blood, and no correlation between organisms grown from the cell saved blood and those causing postoperative pneumonias, bacteraemias or urinary tract infections [24]. This has been confirmed in a recent prospective, randomised controlled trial of penetrating abdominal injuries in which the use of cell salvage at laparotomy led to a significant reduction in allogeneic blood usage with no discernable effect on rates of postoperative infection or mortality [25]. Most surgeons and anaesthetists still avoid salvage from grossly contaminated fields, but procedures involving bowel resection now commonly use cell salvage for at least part of the procedure.

The statement that cell salvage is contra-indicated in malignancy originally came from a Council Report from the American Medical Association in 1986 [26]. It stated that ‘Blood potentially contaminated with malignant cells should not be re-infused because of the risk of producing metastases’. This view has held sway since, despite the fact that in the intervening 22 years there has been no in-vivo evidence to support it. Malignant cells are detectable in the circulation prior to surgery, and surgery itself releases many more cells during tumour manipulation. This has been shown for many tumour types both in patients who do and those who do not develop metastases [27, 28].

Detection of tumour cells in peripheral blood has been greatly facilitated by the use of immunohistochemical techniques using monoclonal mouse anti-human cytokeratin antibodies to label cells of epithelial origin [14]. Cytokeratins (CKs) are a family of water-soluble complex polypeptides that form the cytoskeleton of epithelial cells, and are fundamental markers of epithelial differentiation. Twenty distinct CK polypeptides have been revealed in various human epithelia [15]. Cells labelled by the antibodies display a cytoplasmic staining pattern. Using an immunohistochemical labelling method, Hansen and colleagues found that tumour cells were detectable in the peripheral blood of 26% of surgical patients at the time of tumour surgery [29]. There is, however, debate about whether such circulating tumour cells are of pathological significance [30–32]. Schoppmeyer in 2006 [32] using a similar immunohistochemical labelling method to ours, prospectively studied 47 patients with colonic cancer before and during surgery. Cytokeratin-labelled (CK+) cells were detected in the bone marrow of 26 patients (55%) and in blood samples of 14 patients (30%) before surgery and 11 patients (23%) during surgery with a median detection rate of 1 (range, 1–14) CK+ cell per 4 × 106 mononuclear cells. The detection of CK+ cells was not predictive for extrahepatic recurrence or overall survival after a mean observation time of 43 months.

The question of whether cell salvage with or without filtration can remove tumour cells in-vitro has been extensively investigated since 1975 using evolving and increasingly sophisticated detection techniques. Where filters are not used, most studies showed tumour cells passing through the cell saver [29, 33–36] though some types of cancer cells appeared to be removed by centrifugation, washing and a standard 40 micron filter [37]. In the 1990s there were several in-vitro clearance studies using early leucocyte depletion filters which claimed to show total removal of malignant cells [38–46]. These studies were summarised by the 1998 Consensus Conference on Autologous Transfusion [47], which concluded that ‘in-vitro data suggests that the combination of ICS and LDF gives virtually 100% protection from infusion of tumour cells’. The problem with these studies is the sensitivity of the method used to detect malignant cells. In 1986 Fidler and Talmadge [48] demonstrated that in experimental mice, pulmonary metastases of melanoma can result from a single cell in the lung parenchyma. This work has been quoted by Hansen as indicating that nothing short of a detection system with 100% sensitivity, showing removal of every tumour cell should be accepted as evidence of safety [49]. Hansen states that irradiation of cell salvaged blood could provide a practical solution in malignancy surgery as blood irradiated at 50Gy would not contain viable, clonogenic tumour cells, whereas cell-saved blood that has undergone leucocyte depletion filtering inevitably would, even if the cells were undetectable [50]. Blood irradiation is impractical as most centres do not have a blood irradiator in theatre or on the hospital site. Use of a distant facility would involve delays in availability and could introduce a source of error in the checking and transport of blood that is removed from the immediate patient vicinity and is outwith the clerical structure of the blood bank administration.

The sensitivity and specificity of the two anti-cytokeratin monoclonal antibodies we used in our study – MNF116 and AE1/AE3, have been studied by several authors. MNF116 is an extract of splenic cells from nude mice grafted with human breast carcinoma tissue, and behaves as a broad spectrum reagent reacting with intermediate and low molecular weight keratins, corresponding to cytokeratins 5, 6, 8, 17 and probably 19 [16]. It labels epithelial tissue from simple glandular to stratified epithelium and is used clinically as a tool for the identification of normal and neoplastic cells of epithelial origin. Goddard showed that MNF116 can identify 100% of specific squamous cell carcinomas (epidermal, cervical and bronchial) and 93.5% of epithelial tumours in general (breast, colon, gastric, prostate, renal, transitional, teratoma and pleomorphic adenoma) [14]. Non-epithelial cells are not labelled by the antibody, except for some weak reactivity with leiomyosarcoma in up to 17% of cases [14]. AE1/AE3 is a cocktail of two monoclonal antibodies that are obtained by immunising mice with human callus keratins, and is used to identify two epitopes present on a majority of epithelial CKs [17, 18]. It will, therefore, reliably identify cells of simple and stratified epithelial origin. Listrom and Dalton studied the performance characteristics of AE1/AE3 on a mixture of epithelial neoplasms, lymphomas, melanomas and sarcomas and found 100% staining of the 34 epithelial neoplasms, but with only weak staining of transitional cell carcinomas and unreliable staining of small cell carcinomas [19]. The authors recommended use of the agent as part of a panel of antibodies for identification of these tumour types. No false positive staining was found in the 25 non-epithelial neoplasms studied. Using both MNF116 and AE1/AE3 together will therefore label the majority of CKs and provide a highly sensitive and specific system for the identification of intact, viable, epithelial cancer cells.

The important clinical question is whether cancer patients receiving cell salvaged blood are likely to get more recurrences and have an increased mortality. This does not seem to be the case. In 1998 the Edinburgh Consensus [47] concluded that there were ‘many small, retrospective, non-randomised in-vivo studies which all support ICS in malignancy’. All the in-vivo evidence has shown that cancer patients receiving cell salvaged blood do not get increased rates of recurrences or increased mortality, and there has been considerable clinical use of cell salvage in all types of cancer surgery – particularly urology, over the last 20 years. Up to 2003 the evidence consisted of a small number of case reports and observational studies [51–59], and has been criticised as a result [50]. However, they do show no difference in recurrence rate with intra-operative cell salvage or a better outcome compared with allogeneic transfusion.

The 1998 Edinburgh Consensus [47] stated that ‘there is no prospective randomised controlled trial (RCT) using modern ICS techniques and modern leucocyte depletion filters’. The reason for this is the large number of patients that would be needed for a properly powered RCT. For example, in 1992 Harrison showed that in colorectal surgery patients, assuming 40% survival at 5 years, 1000 patients would be needed to show a 10% change in survival with cell salvaged blood [60]. In 1999 Moir et al. showed a significant improvement in head and neck cancer recurrence rate in a group of 165 patients with autologous transfusion compared with those receiving allogeneic transfusion [6]. More recently Neider and co-workers retrospectively reviewed 1038 patients undergoing radical prostatectomy between 1992 and 2003 and found no difference in biochemical recurrence on five-year follow-up between the patients receiving cell salvaged blood and those that received no blood [7]. In 2007 the same group reported on long term survival in a group of 378 patients undergoing radical prostatectomy between 1992 and 2005 and found no difference in disease–specific or overall survival between those who received cell salvaged blood and those who did not [8]. A recent prospective study of 49 patients undergoing surgery for hepatocarcinoma has also shown no difference in recurrence rate between those who did and did not receive cell salvaged blood [9].

In contrast, there is evidence that allogeneic transfusion is independently associated with an increased rate of both postoperative infection [5] and disease recurrence [4]. In 1992 Wooley and colleagues [3] studied 143 patients with head and neck squamous carcinoma and showed that the odds ratio for recurrence at 5 years for patients receiving allogeneic blood was 3.2 (95% CI 1.5–6.9, p < 0.004). They then conducted a meta-analysis of the six available studies and found that the combined odds ratio for recurrence after allogeneic blood was 2.6 (95%CI 1.9–3.7, p < 0.0001). More recently, Takemura has shown improved survival after oesophagectomy for T3 and T4 stage oesophageal cancer in patients receiving autologous blood as compared with those receiving allogeneic blood. They also showed that natural killer cell activity remained higher in the autologous group (p < 0.05) and suggested this as a possible mechanism influencing survival [10]. Furthermore, there is evidence that the increased risk for both infection [5] and disease recurrence [61–63] is dose related, with the increased risk of recurrence starting at a threshold of 3–4 units of allogeneic blood transfusion, depending on tumour type.

There is, therefore, evidence for benefit in terms of less disease recurrence and less infection where cell salvage has been used during cancer surgery. Several studies report total removal of tumour cells by modern leucocyte depletion filters, despite the theoretical limitations of the detection methods. It seems probable that either no malignant cells are re-transfused with cell salvage, or that the transfused cells are no more clinically significant than cells that can be detected routinely in the circulation in cancer patients.

In conclusion, using a sensitive and specific immunohistochemical label for cytokeratin on epithelial cell-line tumour cells we have shown that no viable malignant cells are detectable in blood salvaged and filtered using the Haemonetics Cell Saver 5 and Pall RS filter from 50 patients undergoing surgery for gynaecological cancers. The theoretical risk of re-infusion of salvaged blood containing tumour cells that are not detectable by current methods must be balanced against the recent encouraging clinical studies of disease recurrence rates and mortality in patients receiving salvaged blood, and against the well documented risks of allogeneic transfusion.

Acknowledgement

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix

This study was funded by an AAGBI Research Grant. The authors did not receive sponsorship from any commercial sources.

Potential conflict of interest

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix

In 2005 Dr Catling gave a lecture on cell salvage for obstetrics at Redhill Hospital, for which her expenses were paid by Haemonetics Ltd.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix
  • 1
    Serious Hazards of Transfusion (SHOT). http://www.shotuk.org (accessed 20 April 2008).
  • 2
    Better Blood Transfusion. Appropriate Use of Blood. London: Department of Health, 2002.
  • 3
    Woolley AL, Hogikyan ND, Gates GA, Haughey BH, Schechtman KB, Goldenberg JL. Effect of blood transfusion on recurrence of head and neck carcinoma. Retrospective review and meta-analysis. Annals of OtoRhinoLaryngology 1992; 101: 72430.
  • 4
    Jones KR, Weissler MC. Blood transfusion and other risk factors for recurrence of cancer of the head and neck. Archives of Otolaryngology – Head and Neck Surgery 1990; 116: 3049.
  • 5
    Kinoshita Y, Udagawa H, Tsutsumi K, et al. Usefulness of autologous blood transfusion for avoiding allogenic transfusion and infectious complications after esophageal cancer resection. Surgery 2000; 127: 18592.
  • 6
    Moir MS, Samy RN, Hanasono MM, Terris DJ. Autologous and heterologous blood transfusion in head and neck cancer surgery. Archives of Otolaryngolology- Head and Neck Surgery 1999; 125: 8648.
  • 7
    Neider AM, Carmack AJK, Sved PD, Kim SS, Manoharan M, Soloway MS. Intraoperative cell salvage during radical prostatectomy is not associated with greater biochemical recurrence rate. Adult Urology 2005; 65: 304.
  • 8
    Neider AM, Manoharan M, Yang Y, Soloway MS. Intraoperative cell salvage during radical cystectomy does not affect long term survival. Adult Urology 2007; 69: 8814.
  • 9
    Muscari F, Suc B, Vigouroux D, et al. Blood salvage autotransfusion during transplantation for hepatocarcinoma: does it increase the risk of neoplastic recurrence? Transplant International 2005; 18: 12369.
  • 10
    Takemura M, Osugi H, Higashino M, Takada N, Lee S, Kinoshita H. Effect of substituting allogenic blood transfusion with autologous blood transfusion on outcomes after radical oesophagectomy for cancer. Annals of Thoracic and Cardiovascular Surgery 2005; 11: 293300.
  • 11
    James V. A National Blood Conservation Strategy for NBTC and NBS. Report from the Working Party on Autologous Transfusion and the Working Party on Alternatives to Transfusion to the NBS sub-group on the appropriate use of blood. 2004. http://www.dh.gov.uk/en/Publichealth/Healthprotections/Bloodsafety/Bloodsafetygeneralinformation/index/htm (accessed 20 June 2008).
  • 12
    Intraoperative Red Blood Cell Salvage During Radical Prostatectomy or Radical Cystectomy. London: NICE, 2008. http://www.nice.org.uk/guidance/index.jsp?action=download&o=40387.IPG258 (accessed 25 April 2008).
  • 13
    Levine TS, Rolfe KJ, Crow J, et al. The use of cytospin monolayer technique in the cytological diagnosis of vulval and anal disease. Cytopathology 2001; 12: 297305.
  • 14
    Goddard MJ, Wilson B, Grant JW. Comparison of commercially available cytokeratin antibodies in normal and neoplastic adult epithelial and non-epithelial tissues. Journal of Clinical Pathology 1991; 44: 6603.
  • 15
    Moll R. Cytokeratins as markers of differentiation in the diagnosis of epithelial tumours. In: HermannH, HarrisJR, eds. Subcellular Biochemistry Volume 31. New York: Plenum Press, 1998: 20562.
  • 16
    Moll R, Franke WW, Schiller DI. The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell 1982; 31: 1124.
  • 17
    Tseng SCG, Jarvinen MJ, Nelson WG, Huang JW, Woodcock-Mitchell J, Sun TT. Correlation of specific keratins with different types of epithelial differentiation: monoclonal antibody studies. Cell 1982; 30: 36172.
  • 18
    Woodcock-Mitchell J, Eichner R, Nelson WG, Sun TT. Immunolocalization of keratin polypeptides in human epidermis using monoclonal antibodies. Journal of Cell Biology 1982; 95: 5808.
  • 19
    Listrom MB, Dalton MW. Comparison of monoclonal antibodies MAK-6, AE1/AE3 nd CAM-5.2. American Journal of Clinical Pathology 1987; 88: 297301.
  • 20
    Dzik WH, Sherbarne B. Intraoperative blood salvage: medical controversies. Transfusion Medicine Review 1990; 4: 20833.
  • 21
    Waters JH, Biscotti MD, Potter PS, Phillipson E. Amniotic fluid removal during cell salvage in the caesarean section patient. Anesthesiology 2000; 92: 15316.
  • 22
    Catling SJ, Williams S, Feilding AM. Cell salvage in obstetrics: an evaluation of the ability of cell salvage combined with leucocyte depletion filtration to remove amniotic fluid from operative blood loss at caesarean section. International Journal of Obstetric Anesthesia 1999; 8: 7984.
  • 23
    Catling SJ, Freites O, Krishnan S, Gibbs R. Clinical experience with cell salvage in obstetrics – four cases from one UK centre. International Journal of Obstetric Anesthesia 2002; 11: 12834.
  • 24
    Ozmen V, McSwain NE, Nichols RL, Smith J, Flint LM. Autotransfusion of potentially culture-positive blood in abdominal trauma:preliminary data from a prospective study. Journal of Trauma 1992; 32: 369.
  • 25
    Bowley DM, Barker P, Boffard KD. Intraoperative blood salvage in penetrating abdominal trauma: a randomised, controlled trial. World Journal of Surgery 2006; 30: 107480.
  • 26
    Autologous Blood Transfusions. Council on Scientific Affairs, American Medical Association Report. Journal of the American Medical Association 1986; 256: 237880.
  • 27
    Mori M, Mimori K, Ueo H, et al. Molecular detection of circulating solid carcinoma cells in the peripheral blood: the concept of early systemic disease. International Journal of Cancer 1996; 68: 73943.
  • 28
    Denis MG, Lipart C, Leborgne J, et al. Detection of disseminated tumour cells in peripheral blood of colorectal cancer patients. International Journal of Cancer 1997; 74: 5404.
  • 29
    Hansen E, Wolff N, Knuechel R, Ruschoff J, Hofstaedter F, Taeger K. Tumor cells in blood shed from the surgical field. Archives of Surgery 1995; 130: 38793.
  • 30
    Salsbury AJ. The significance of circulating cancer cells. Cancer Treatment Review 1975; 2: 5572.
  • 31
    Fidler IJ. Cancer metastasis. British Medical Bulletin 1991; 47: 15777.
  • 32
    Schoppmeyer K, Frühauf N, Oldhafer K, Seeber S, Kasimir-Bauer S. Tumor cell dissemination in colon cancer does not predict extrahepatic recurrence in patients undergoing surgery for hepatic metastases. Oncology Reports 2006; 15: 44954.
  • 33
    Yaw PB, Sentany M, Link WJ, Wahle WM, Glover JL. Tumour cells carried through autotransfusion: contra-indication to intraoperative blood recovery? Journal of the American Medical Association 1975; 231: 4901.
  • 34
    Homan B, Zenner HP, Schauber J, Ackerman R. Tumour cells carried through autotransfusion: are these cells still malignant? Acta Anaesthesiologica Belgica 1984; 35: 519.
  • 35
    Dale RF, Kipling RM, Smith MF, Collier D, Smith PJ. Separation of malignant cells during autotransfusion. British Journal of Surgery 1988; 75: 581.
  • 36
    Karczewski DM, Lema MJ, Glaves D. The efficiency of an autotransfusion system for tumour cell removal from blood salvaged during cancer surgery. Anesthesia and Analgesia 1994; 78: 11315.
  • 37
    Fujimoto J, Okamoto E, Yamanaka N, et al. Efficacy of autotransfusion in hepatectomy for hepatocellular carcinoma. Archives of Surgery 1993; 128: 10659.
  • 38
    Miller GV, Ramsden CW, Primrose JN. Autologous transfusion: an alternative to transfusion with banked blood during surgery for cancer. British Journal of Surgery 1991; 78: 7135.
  • 39
    Edelman MJ, Potter P, Mahaffey KG, Frink R, Leidich RB. The potential for re- introduction of tumour cells during intraoperative cell salvage: reduction of risk with use of the RC 400 leucocyte depletion filter. Urology 1996; 47: 17981.
  • 40
    Wiesel M, Gudemann C, Staehler G. Tumour cell separation by cell saver and membrane filter passage. Infusionstherapie 1991; 18: 1434.
  • 41
    Wiesel M, Gudemann C, Staehler G. Separation of urologic tumour cells from cell saver blood using a membrane filter; a new method in autotransfusion? Urologe 1992; 31: 1825.
  • 42
    Torre GC, Ferrari M, Favre A, Razetta F, Borgonov G. A new technique for intraoperative blood recovery in the cancer patient. European Journal of Surgery and Oncology 1994; 20: 56570.
  • 43
    Muller M, Kuhn DF, Hinrichs B, et al. Is the elimination of osteosarcoma cells with intraoperative ‘mesh autotransfusion’ and leucocyte depletion filters possible? Anaesthetist 1996; 45: 8348.
  • 44
    Kongsgaard UE, Wang MY, Kvalheim G. Leucocyte depletion filtration removes cancer cells in human blood. Acta Anaesthesioloica Scandinavica 1996; 40: 11820.
  • 45
    Perseghin P, Vigano M, Rocco G, Della Pona C, Buscemi A, Rizzi A. Effectiveness of leucocyte filters in reducing tumour cell contamination after intraoperative cell salvage in lung cancer surgery. Vox Sanguinis 1997; 72: 2214.
  • 46
    Park KL, Kojima O, Tomoyoshi T. Intra-operative autotransfusion in radical cystectomy. British Journal of Urology 1997; 79: 71721.
  • 47
    Update Consensus Conference on Autologous Transfusion. Edinburgh: Royal College of Physicians of Edinburgh, 1998.
  • 48
    Fidler IJ, Talmadge JE. Evidence that intravenously derived murine pulmonary melanoma metastases can originate from the expansion of a single tumor cell. Cancer Research 1986; 46: 516771.
  • 49
    Hansen E. Failed evidence of tumour cell removal from salvaged blood after leucocyte depletion. Transfusion Medicine 2006; 16: 2134.
  • 50
    Hansen E. Cell salvage in the presence of malignancy. Transfusion Alternatives in Transfusion Medicine 2003; 5: 4727.
  • 51
    Klimberg IW, Sirios R, Wajsman Z, Baker J. Intraoperative autotransfusion in urologic oncology. Archives of Surgery 1986; 121: 13269.
  • 52
    Hart OJ, Klimberg IW, Wajsman Z, Baker J. Intraoperative autotransfusion in radical cystectomy for carcinoma bladder. Surgery, Gynecology and Obstetrics 1989; 168: 3026.
  • 53
    Hart OJ, Klimberg IW, Wajsman Z, Baker J. Intraoperative autotransfusion in patients with invasive bladder cancer: long term follow up. Journal of Urology 1989; 141: 478A.
  • 54
    Pisters LL, Wajsman Z. Use of predeposit autologous blood and intraoperative autotransfusion in urologic cancer surgery. Urology 1992; 40: 2115.
  • 55
    Zulim RA, Rocco M, Goodnight J. Intraoperative autotransfusion in hepatic resection for malignancy: is it safe? Archives of Surgery 1993; 128: 20611.
  • 56
    Fujimoto J, Okamoto E, Yamanaka N, et al. Efficacy of autotransfusion in hepatectomy for hepatocellular carcinoma. Archives of Surgery 1993; 128: 10659.
  • 57
    Connor JP, Morris PC, Alagoz T, Anderson B, Bottles K, Buller RE. Intraoperative autologous blood collection and autotransfusion in the surgical management of early cancers of the uterine cervix. Obstetrics and Gynecology 1995; 86: 3738.
  • 58
    Chiusano MA, Finkelstein LH, Mene M, Ginsburgh PC. Use of predeposited autologous blood plus intraoperative autotransfusion compared with use of predeposited autologous blood alone in radical retropubic prostatectomy. Journal of the American Osteopathic Association 1996; 96: 1604.
  • 59
    Davis M, Sofer M, Gomez-Martin O, Bruck D, Soloway MS. The use of cell salvage during radical retropubic prostatectomy: does it influence cancer recurrence? British Journal of Urology International 2003; 91: 4746.
  • 60
    Harrison S, Steele RJC. Predeposit autologous blood transfusion in patients with colorectal cancer: a feasibility study. British Journal of Surgery 1992; 79: 3557.
  • 61
    Dresner SM, Lamb PJ, Shenfine J, Hayes N, Griffin SM. Prognostic significance of peri-operative blood transfusion following radical resection for oesophageal carcinoma. European Journal of Surgical Oncology 2000; 26: 4927.
  • 62
    Taniguchi Y, Okura M. Prognostic significance of perioperative blood transfusion in oral cavity squamous cell carcinoma. Head and Neck 2003; 25: 9316.
  • 63
    Szakmany T, Dodd M, Dempsey GA, et al. The influence of allogenic blood transfusion in patients having free-flap primary surgery for oral and oropharyngeal squamous cell carcinoma. British Journal of Cancer 2006; 94: 64753.

Appendix

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgement
  7. Potential conflict of interest
  8. References
  9. Appendix
Preparation and concentration of samples by cytology lab

Each sample was transferred to appropriately labelled universal containers and centrifuged at 3000 rpm for 10 min and then the supernatant/plasma was removed and discarded. Fifteen millilitre of 10% Pharmlyse was added to each sample, mixed and allowed to stand at room temperature for 30 min. The samples were then centrifuged at 3000 rpm for 10 min and the supernatant again discarded. The remaining cells were then re-suspended in 20 ml of phosphate buffered saline and centrifuged again at 3000 rpm for 10 min. These last two steps were repeated, then three or four drops of the resulting re-suspended cell pellet was used to prime three cytospin chambers containing three appropriately labelled slides for each sample. These were spun at 1500 rpm for 10 min in a cytospin centrifuge. The slides were then fixed with acetone.

Immunohistochemistry technique

A Ventana Benchmark XT fully automated machine (Ventana Medical Systems Inc, Tucson, Arizona, USA) was used to stain all slides, each case generating 15 slides if all samples were collected. Barcode labels were printed using the patient trial number as identification.

Each sample was stained with Cytokeratin Pan (MNF116) and Cytokeratin AE1/AE3 at a dilution of 1 : 100, along with a negative control (DAKO-CK Pan, Cat. No- MO821) and CK AE1/AE3 (Cat. No- M3515) (DAKO Ltd, Ely, UK). A composite (breast, bowel and skin) control section was stained for both antibodies in every run to verify antibody validity.

Antibody retrieval was accomplished with Ventana Protease 1 for 10 min, and detection was with the Ventana ‘I View’ Detection kit, which employs Avidin Biotin Complex technique and uses 3,3 diaminobenzidene-tetrahydrochloride as its chromagen.

After staining the slides were washed, counterstained with haematoxylin and examined using light microscopy.