Annals of the Royal College of Surgeons of England (1990) vol. 72, 388-392
Antiseptic toxicity to breast carcinoma in tissue culture: an ad juvant to conservation therapy? Michele E Lucarotti FRCS
Ian A Silver MA FRCVS
Surgical Registrar
Professor of Pathology
Hugh White MRCPath
David J Leaper MD FRCS
Consultant Histopathologist
Consultant/Senior Lecturer
Judy Deas MSc Research Assistant
University Departments of Surgery and Pathology, Southmead Hospital, Bristol
Key words: Breast carcinoma; Antiseptics; Conservation surgery; Cytology
Of 50 patients who had scrape cytology of the excision cavity after conservative surgery for breast carcinoma, 10 (20%) had malignant cells remaining in the cavity recognised by cytology. Of these patients, 18 had histological evidence of tumour at the resection margin, giving an accuracy of the cytology of 84%, a sensitivity of 56%, and a specificity of 100%. When assayed for cytotoxicity against a breast tumour cell line (MCF7) or human fibroblasts, chlorhexidine gluconate was the most effective of eight antiseptics or antitumour agents (100% cytotoxicity at a 1/10 000 dilution) in killing breast tumour cells and had 70% toxicity to human fibroblasts at the same dilution. Hydrogen peroxide appeared to be the most useful agent overall with 94% cytoxicity to breast tumour cells with only a 12% cytotoxicity to human fibroblasts at a dilution of 1/1 000 000. We suggest that free malignant cells left in the cavity after conservative surgery for breast cancer may be a cause of local recurrence. They can be recognised by scrape cytology at operation and the topical use of antiseptics as cytotoxic agents may be beneficial and warrants further investigation.
The incidence of breast carcinoma in women living in the Western World is rising and remains one of the leading causes of death from cancer (1). Despite many new techniques of surgical or adjuvant treatment, there has
Correspondence to: Mr D J Leaper, University Department of Surgery, Southmead Hospital, Bristol BS1O 5NB
been little improvement in the 10-year survival rate over the past 40 years (2). Any investigation which can improve the adequacy of primary treatment is therefore also likely to reduce the mortality and morbidity from this disease. Conservative surgical procedures have become a viable alternative option to mastectomy for many women who have breast cancer (3). This concept is based on serial randomised clinical trials which have shown that survival and local tumour control after breast conservation are comparable to that achieved with more radical procedures (4-7). However, limited surgery cannot ensure that the cancer is adequately excised and this is one argument against conservative treatment, particularly as there are no guidelines for the extent of surrounding normal breast tissue which should be excised around the tumour. We have therefore examined the postexcision breast cavities intraoperatively, using cytology to assess whether excision was complete, in 81 patients undergoing breast lump excision for suspected or confirmed carcinoma. This preliminary study showed that following conservative surgery malignant cells remained within the excision cavity of some patients, and this finding led us to investigate the use of local cytotoxic adjuvant treatment. We examined the cytotoxic effects of several antiseptic and antitumour agents on a breast tumour cell line (MCF7) and human fibroblasts in tissue culture to determine whether topical intraoperative application of cytotoxic agents might have any therapeutic value before closure of the breast cavity after tumour excision.
Antiseptic toxicity to breast carcinoma
Patients and methods Intraoperative cytology A series of 81 patients undergoing breast lump excision for suspected or confirmed carcinomas were assessed. A final diagnosis of carcinoma was made on histological examination in 50 patients. The median age of the 'malignant' group was 57 years (range 27-80 years) and that of the 'benign' group was 41 years (range 19-75 years). Wooden spatula scrapings were taken from the tumour excision cavity in the breast immediately before skin closure and after haemostasis was secured. Air-dried slides were prepared and a Giemsa stain applied. The slides were then examined for the presence of malignant cells. Cell culture
Primary human fibroblast cell cultures were initiated from a piece of skin taken from the edge of a biopsy removed from one of our patients. Under sterile conditions, the biopsy was washed with phosphate-buffered saline (PBS) containing 400 U/ml penicillin and 200,ugl ml streptomycin, transferred to a Petri dish and finely minced using a scalpel. The resulting tissue was washed twice with PBS, and transferred to a Universal tube containing 10 ml of 0.1% collagenase and 0.05% trypsin in Earle's balanced salt solution. The tissue fragments were digested for 20 min at 37°C on a rotator, after which the supernatant was removed, and fresh enzyme solution applied. Following a further 2 h digestion period, the resulting cell suspension was filtered through sterile gauze and 10% (v/v) fetal calf serum added to inhibit further action of trypsin. The cells were spun at 350g for 10 min, the supernatant discarded, and the pellet of cells resuspended in a small amount of culture medium consisting of RPMI 1640 supplemented with 2 mM Lglutamine, 10% fetal calf serum, 200 U/ml penicillin and 100 Fig/ml streptomycin. The cells were seeded into tissue culture flasks at a density of 2.5 x 105 cells/ml, and incubated at 37°C in a 95% air/5% CO2 humid atmosphere. After 3 days, the cultures were washed with sterile PBS and fresh culture medium applied. When the fibroblasts reached confluency they were deemed suitable for cytotoxicity experiments. MCF7 cells (derived from pleural effusion of human breast carcinoma ATTC HTB22) were a gift from Dr P Moncourrier, Universit6 de Montpelier, France. Cells were maintained in Eagle's minimum essential medium (MEM) supplemented with 10% fetal calf serum, 1% MEM non-essential amino acids, 2 mM L-glutamine, 200 U/ml penicillin and 100 [tg/ml streptomycin. Both cell types were routinely subcultured and prepared for experimentation using a solution of 0.2% EDTA and 0.05% trypsin (1:250) in PBS. Cytotoxic agents Agents tested were: chlorhexidine gluconate 5% w/v; hydrogen peroxide 3%; povidone-iodine 7.5% w/v
389
(Betadine®); chlorhexidine gluconate 0.015% w/vcetrimide 0.15% w/v (Savlodil®); Eusol substitute 25%; noxythiolin 2.5% (Noxyflex®); mercuric chloride 0.1% and taurolidine (Taurolin8) 2%. Cytotoxicity assay MCF7 or human fibroblasts were seeded at a density of 1 x 104 into 96-well microtitre plates and grown overnight in the appropriate culture medium. Concentrations of several different aqueous antiseptics ranging from stock to 10-6 stock solution concentration were prepared using culture medium; 200 1d of each solution was added to the cells and incubated for a further 24 h. A colorimetric cytotoxicity assay was used to measure the number of viable cells remaining after incubation with each antiseptic. Following incubation, the cells were washed, fixed with 5% formalin, and stained with 1% methylene blue in 0.01 M borate buffer (pH 8.5). After rinsing with distilled water, the dye fixed on the residual cells was eluted using 0.1 N hydrochloric acid and the optical density read on a spectrophotometer at 665 nm. The absorbence of the eluted dye is proportional to the number of viable cells present so that the percentage cytotoxicity is: (
O.D. cells + antiseptic solutions 1 O.D. cells + culture medium
where O.D. is the optical density.
Results Intraoperative cytology Of the 50 patients with histologically confirmed carcinoma, 14 (28%) also had evidence of incomplete excision histologically (defined as extension of tumour to the resection margin of the specimen). Of these, 10 had cellular cytology smears adequate for examination, of which seven were malignant. In the remaining three cellular smears benign cells only were found, and one of these patients had a second operation based on the histological findings, but no further malignancy was found in the second specimen. Only 13 (26%) of the patients could therefore be judged to have had an incomplete excision at the first operation. Five further patients (10%) had histological evidence of lymphatic or vascular invasion, despite a histologically complete resection margin. One of these had an inadequate smear and three had malignant smears. One had benign cells only. This gave an accuracy of 84%, a sensitivity of 56% and a specificity of 100%. If the inadequate smears are discounted, the sensitivity was 71%. The inadequate smears were all made early in the study, which suggests that the technique of taking scrapes with the wooden spatula had to be learned by the surgeon performing the operation.
390
M E Lucarotti et al.
toxicity to MCF7 cells was observed at dilutions greater than 1/1000 from that of the standard stock concentration. In contrast to this, Fig. 2 demonstrates that human fibroblast cultures were not as sensitive as MCF7 cells to low concentrations of antiseptics and thus significant toxicity was primarily seen at dilutions less than 1/1000. The most effective antiseptic was chlorhexidine gluconate which was 100% toxic to MCF7 cells and 95% toxic to human fibroblasts at a 1/10 000 dilution. However, it would appear that hydrogen peroxide and Eusol substitute are more specifically cytotoxic to MCF7 cells than to human fibroblast cultures. When applied at a 1/1000 dilution, Eusol substitute killed 88% of MCF7 cells, but
100 9 -Log
0
dLLutuon
-2- ChLorhex,oine
---
-|V- Betocine
-O-
H_j&ogen
PeroxWde
753
5ovt.docL
(a) 3
0
-)0- HLd-ogen PeroxLde
-2- ChLorhextdne -V- Betodne
-O-
SavLod.L
(a)
-Log oLutLOn --
OEusoL
-2- Mercur'Lc
Chtor
de
X Noxyflex -2O- Tour-oLLn
(b) Figure 1. (a, b) Effects of antiseptics cells in vitro.
Jn on
MCF7 breast tumour
Xj
.S
No patient in the benign group (31) or in the completely excised group with no vascular or lymphatic invasion had a malignant cell smear. When the data were combined, the overall accuracy of the technique was 90% with the same sensitivities and specificities. -Log
-2- EusoL ---V Mercuroc
Cytotoxicity of antiseptics All the antiseptics tested in this study appeared to be toxic to both MCF7 cells and human fibroblast cultures in vitro. From the results shown in Fig. 1, significant
dlutoLon --*-
Noxyflex
DC)orLde --- Toacron
(b) Figure 2. (a, b) The effects of antiseptics on human fibroblasts in vitro.
Antiseptic toxicity to breast carcinoma only 6% of human fibroblasts. Similarly, hydrogen peroxide at a dilution of 1/1 000 000 was 94% toxic to MCF7 cells but only 12% toxic to human fibroblasts.
Discussion The reported incidence of local recurrence in recent studies of conservative breast surgery, with adjuvant radiotherapy, varies greatly due to different treatment regimens and patient selection. A NSABP trial reported an 8% local recurrence at 5 years for segmental mastectomy and radiotherapy, but excluded patients with positive histological margins who underwent mastectomy (8). Montague (9) reported a 5% 5-year local recurrence, but more than 10% of patients had in situ disease only. Although there is no evidence that local recurrence influences long-term survival it could nevertheless be more devastating psychologically to the patient than the original diagnosis, and is evidence to her of primary treatment failure. It also necessitates, in many cases, more extensive surgery. There is no clinical data from controlled trials to support any specific surgical technique, in particular how wide a margin of normal breast tissue should be included around the tumour. The goal of conservative surgery is to achieve local control and inadequacy of surgical excision and intralymphatic extension correlate with the risk of recurrence (10-12). The multicentricity of breast cancer has been much debated as a possible contraindication to breast-conserving therapy. Holland (13) found in consecutive cases that 17% had multifocal disease. Gump (14) in another study found an incidence of 19%. Both are higher than the 10% seen in the NSABP trial (8). However, the natural history of such subclinical lesions is not known (7) and their prevalence may be overestimated in pathological studies of selected patients. Most recurrences also occur close to the previous excision site after breast conserving treatment which favours the cause of recurrence being due to inadequate local excision. The technique of intraoperative cytological assessment of the excision cavity may alert the surgeon to incomplete tumour excision at the time of surgery and if more extensive surgery is not undertaken may act as a prognostic index. Adjuvant treatment may be adjusted accordingly. The technique is simple and can produce a result quicker than frozen section. In our opinion and from these early results the sensitivity of the technique might improve with experience. The possibility of implantation of exfoliated tumour cells into newly cut tissues was postulated by Ryall in 1908 (15). Today various precautions have been recommended in colorectal surgery to prevent implantation of exfoliated tumour cells. These include peroperative irrigation of the intestinal lumen with a cytotoxic agent. The incidence of suture-line recurrence has been shown to be reduced using such precautions (16,17). Umpleby and Williamson have demonstrated that topical chlorhexidine-cetrimide and povidone-iodine have effective
391
cytoxicity against extraluminal tumour cells shed from colorectal carcinomas (18). It is accepted that postoperatiuve radiotherapy decreases the rate of local recurrence in breast cancer and that lower doses of radiotherapy result in higher recurrences (19). In this study we have found that hydrogen peroxide and Eusol substitute show a preferential cytotoxic action towards MCF7 (breast tumour) cells than to human fibroblasts when applied at appropriate concentrations in vitro. This suggests that it may be practicable to sterilise the tumour excision bed by local irrigation using a cytotoxic agent in a way analogous to the current practice in colorectal surgery. In this study we have shown that malignant cells are frequently present in the cavities left after conservative surgery for breast carcinoma and can be harvested by scrapings and recognised by cytology. The cytotoxic effects of antiseptics in tissue culture suggest that they may be a useful adjunct to conservation therapy.
References I Kelsey JL. A review of the epidemiology of human breast cancer. Epidemiol Rev 1979;1:74-109. 2 Baum M. Carcinoma of the breast. In: Taymor S ed. Recent Advances in Surgery 10. New York: Churchill Livingstone, 1980:241-57. 3 Harris JR, Heilman S, Kinne DW. Limited surgery and radiotherapy for early breast cancer. N Engl J Med 1985;313: 1365-8. 4 Atkins H, Hayward JL, Klugman DJ, Wayne AB. Treatment of early breast cancer. A report after ten years of clinical trial. Br Med Jf 1972;2:423-9. 5 Veronesi V, Saccozzi R, Del Vecchio M et al. Comparing radical mastectomy with quadrantectomy in patients with small cancers of the breast. N Engl J Med 198 1;305:6-1 1. 6 Sarrazin D, Le M, Ronesse J et al. Conservative versus mastectomy in breast cancer tumours with macroscopic diameter of 20 millimetres or less. Cancer 1984;53: 1209-13. 7 Fisher B, Bower M, Margoles R et al. Five year results of a ransomised clinical trial comparing total mastectomy and sequential mastectomy with or without radiation in the treatment of breast cancer. N EnglJ7 Med 1985;312:665-73. 8 Fisher ER, Sass R, Fisher B et al. Pathologic findings from the National Adjuvant Breast Project (protocol 6) II Relation of local breast recurrence to multicentricity.
Cancer 1986;57:1717-24. 9 Montague ED. Conservative surgery and radiation therapy
in the treatment of operable breast cancer. Cancer 1984;53: 700-4. 10 Aspegren K, Holmberg L, Adami HO. Standardisation of the surgical technique in breast conserving treatment of mammary cancer. Br J Surg 1988;75:807-10. 11 Levene MB, Harris J, Hellman S. Treatment of carcinoma
of the breast by radiation therapy. Cancer 1977;6:2840-5. 12 Barr LC, Brunt AM, Goodman AG, Drury A, Phillips RH, Ellis H. The primary management of breast cancer: is breast conservation feasible? Ann R Coil Surg EngI 1989; 71:390-3. 13 Holland R, Veling SHJ, Mravunach M, Hendrics JHCL. Histologic multifocality of Tis, T1-2, breast carcinomas.
392
14
15 16
17
M E Lucarotti et al.
Implications for clinical trials of breast conserving surgery. Cancer 1985;56:979-90. Gump FE, Shikora S, Habib DV, Kister S, Logerto P, Estabrook A. The extent and distribution of cancer in breast with palpable tumours. Ann Surg 1986;204:384-90. Ryall C. The technique of cancer operations with relation to the danger of cancer infection. Br Med J 1908;2: 1005-8. Keynes WM. Implantation from the bowel lumen in cancer of the large intestine. Ann Surg 1961;153:357-64. Southwich HW, Harridge WM, Cole WE. Recurrence at the suture line following resection for carcinoma of the
colon. Incidence following preventive measures. Am J Surg
1962;103:86-9. 18 Umpleby AC, Williamson RCN. The efficacy of agents employed to prevent anastomotic recurrence in colorectal carcinoma. Ann R Coll Surg Engl 1984;66:192-4. 19 Hayward J. Trials of wide excision and radiation therapy at the breast unit, Guy's Hospital, London. In: Zander J, Baltzer J, eds. Early Breast Cancer, Berlin: Springer Verlag, 1985:253-63.
Received 22 May 1990
Assessor's comment This is a very well-designed and executed experimental project. From the in vitro experiment they have shown that hydrogen peroxide and Eusol substitutes show a preferential cytotoxic action towards MCF7 tumour cells and human fibroblasts when applied in appropriate concentrations. They have also harvested tumour cells from the cavity following conservative management of breast cancer. Local antiseptic applied to the cavity would, therefore, be a useful adjuvant to conservative therapy.
However, because of the multicentricity of breast cancer it may be difficult to confer the same benefit in vivo, and hopefully they will now proceed to a controlled, double-blind trial in vivo. F GIVEN MSc MCh FRCSI Professor of Surgery Regional Hospital Galway, Ireland
Notes on books Difficult Problems in General Surgery edited by John L Sawyers and Lester F Williams. 545 pages, illustrated. Yearbook Medical Publishers, Chicago. 1989. £56.50. ISBN 0 8151 7544 2
The intention of this book is to integrate into one volume the experiences of distinguished surgeons in dealing with complicated problems of general surgery. The authors have been selected because of their particular expertise in the problem they discuss. Three-quarters of the book is devoted to surgery of the gastrointestinal tract and includes such problems as short bowel syndrome, low rectal carcinoma, biliary fistula, liver injuries and retained common bile duct stones. There are also sections on vascular surgery, recurrent groin hernia, burns, acute necrotising fasciitis and intra-abdominal sepsis. Each chapter can be read independently of the others. There can be few general surgeons who would not profit and learn from reading this book.
Monitoring in Anesthesia and Critical Care Medicine edited by Casey D Blitt. 2nd edition. 903 pages, illustrated. Churchill Livingstone, New York. 1990. £75.00. ISBN 0 443 08635 4
This new edition of a textbook first published 5 years ago continues to have as its goal the dissemination of detailed information regarding the monitoring of the anaesthetised
patient and the patient in the Intensive Care Unit. It is designed to be a reference book. All chapters have been updated to reflect advances and new chapters have been added including sections on pulse oximetry, transoesophageal echocardiography and monitoring and patient safety.
Textbook and Colour Atlas of the Temporomandibular Joint edited by John E deB Norman and Paul Bramley. 262 pages, illustrated. Wolfe Medical Publications, London. 1990. £95.00. ISBN 0 7234 A
1518 8
This is a handsome book. Produced in large format, it contains a wealth of high-quality colour photographs as well as an extensive text covering all aspects of the diseases, disorders and surgery of the temporomandibular joint. The early part relates to anatomy and surgical access, together with pathology, medical diseases and imaging. The authors then go on to discuss arthroscopy and arthroscopic surgery in some detail, followed by dislocation, ankylosis and the surgery of internal derangement. There follows an extensive section on growth problems and the volume concludes with a section on unusual surgical diseases and disorders. An important volume that will surely find a place on the shelves of every maxillofacial unit as well as in the private libraries of a large number of maxillofacial surgeons.
Antiseptic toxicity to breast carcinoma in tissue culture: an ad juvant to conservation therapy? Michele E Lucarotti FRCS
Ian A Silver MA FRCVS
Surgical Registrar
Professor of Pathology
Hugh White MRCPath
David J Leaper MD FRCS
Consultant Histopathologist
Consultant/Senior Lecturer
Judy Deas MSc Research Assistant
University Departments of Surgery and Pathology, Southmead Hospital, Bristol
Key words: Breast carcinoma; Antiseptics; Conservation surgery; Cytology
Of 50 patients who had scrape cytology of the excision cavity after conservative surgery for breast carcinoma, 10 (20%) had malignant cells remaining in the cavity recognised by cytology. Of these patients, 18 had histological evidence of tumour at the resection margin, giving an accuracy of the cytology of 84%, a sensitivity of 56%, and a specificity of 100%. When assayed for cytotoxicity against a breast tumour cell line (MCF7) or human fibroblasts, chlorhexidine gluconate was the most effective of eight antiseptics or antitumour agents (100% cytotoxicity at a 1/10 000 dilution) in killing breast tumour cells and had 70% toxicity to human fibroblasts at the same dilution. Hydrogen peroxide appeared to be the most useful agent overall with 94% cytoxicity to breast tumour cells with only a 12% cytotoxicity to human fibroblasts at a dilution of 1/1 000 000. We suggest that free malignant cells left in the cavity after conservative surgery for breast cancer may be a cause of local recurrence. They can be recognised by scrape cytology at operation and the topical use of antiseptics as cytotoxic agents may be beneficial and warrants further investigation.
The incidence of breast carcinoma in women living in the Western World is rising and remains one of the leading causes of death from cancer (1). Despite many new techniques of surgical or adjuvant treatment, there has
Correspondence to: Mr D J Leaper, University Department of Surgery, Southmead Hospital, Bristol BS1O 5NB
been little improvement in the 10-year survival rate over the past 40 years (2). Any investigation which can improve the adequacy of primary treatment is therefore also likely to reduce the mortality and morbidity from this disease. Conservative surgical procedures have become a viable alternative option to mastectomy for many women who have breast cancer (3). This concept is based on serial randomised clinical trials which have shown that survival and local tumour control after breast conservation are comparable to that achieved with more radical procedures (4-7). However, limited surgery cannot ensure that the cancer is adequately excised and this is one argument against conservative treatment, particularly as there are no guidelines for the extent of surrounding normal breast tissue which should be excised around the tumour. We have therefore examined the postexcision breast cavities intraoperatively, using cytology to assess whether excision was complete, in 81 patients undergoing breast lump excision for suspected or confirmed carcinoma. This preliminary study showed that following conservative surgery malignant cells remained within the excision cavity of some patients, and this finding led us to investigate the use of local cytotoxic adjuvant treatment. We examined the cytotoxic effects of several antiseptic and antitumour agents on a breast tumour cell line (MCF7) and human fibroblasts in tissue culture to determine whether topical intraoperative application of cytotoxic agents might have any therapeutic value before closure of the breast cavity after tumour excision.
Antiseptic toxicity to breast carcinoma
Patients and methods Intraoperative cytology A series of 81 patients undergoing breast lump excision for suspected or confirmed carcinomas were assessed. A final diagnosis of carcinoma was made on histological examination in 50 patients. The median age of the 'malignant' group was 57 years (range 27-80 years) and that of the 'benign' group was 41 years (range 19-75 years). Wooden spatula scrapings were taken from the tumour excision cavity in the breast immediately before skin closure and after haemostasis was secured. Air-dried slides were prepared and a Giemsa stain applied. The slides were then examined for the presence of malignant cells. Cell culture
Primary human fibroblast cell cultures were initiated from a piece of skin taken from the edge of a biopsy removed from one of our patients. Under sterile conditions, the biopsy was washed with phosphate-buffered saline (PBS) containing 400 U/ml penicillin and 200,ugl ml streptomycin, transferred to a Petri dish and finely minced using a scalpel. The resulting tissue was washed twice with PBS, and transferred to a Universal tube containing 10 ml of 0.1% collagenase and 0.05% trypsin in Earle's balanced salt solution. The tissue fragments were digested for 20 min at 37°C on a rotator, after which the supernatant was removed, and fresh enzyme solution applied. Following a further 2 h digestion period, the resulting cell suspension was filtered through sterile gauze and 10% (v/v) fetal calf serum added to inhibit further action of trypsin. The cells were spun at 350g for 10 min, the supernatant discarded, and the pellet of cells resuspended in a small amount of culture medium consisting of RPMI 1640 supplemented with 2 mM Lglutamine, 10% fetal calf serum, 200 U/ml penicillin and 100 Fig/ml streptomycin. The cells were seeded into tissue culture flasks at a density of 2.5 x 105 cells/ml, and incubated at 37°C in a 95% air/5% CO2 humid atmosphere. After 3 days, the cultures were washed with sterile PBS and fresh culture medium applied. When the fibroblasts reached confluency they were deemed suitable for cytotoxicity experiments. MCF7 cells (derived from pleural effusion of human breast carcinoma ATTC HTB22) were a gift from Dr P Moncourrier, Universit6 de Montpelier, France. Cells were maintained in Eagle's minimum essential medium (MEM) supplemented with 10% fetal calf serum, 1% MEM non-essential amino acids, 2 mM L-glutamine, 200 U/ml penicillin and 100 [tg/ml streptomycin. Both cell types were routinely subcultured and prepared for experimentation using a solution of 0.2% EDTA and 0.05% trypsin (1:250) in PBS. Cytotoxic agents Agents tested were: chlorhexidine gluconate 5% w/v; hydrogen peroxide 3%; povidone-iodine 7.5% w/v
389
(Betadine®); chlorhexidine gluconate 0.015% w/vcetrimide 0.15% w/v (Savlodil®); Eusol substitute 25%; noxythiolin 2.5% (Noxyflex®); mercuric chloride 0.1% and taurolidine (Taurolin8) 2%. Cytotoxicity assay MCF7 or human fibroblasts were seeded at a density of 1 x 104 into 96-well microtitre plates and grown overnight in the appropriate culture medium. Concentrations of several different aqueous antiseptics ranging from stock to 10-6 stock solution concentration were prepared using culture medium; 200 1d of each solution was added to the cells and incubated for a further 24 h. A colorimetric cytotoxicity assay was used to measure the number of viable cells remaining after incubation with each antiseptic. Following incubation, the cells were washed, fixed with 5% formalin, and stained with 1% methylene blue in 0.01 M borate buffer (pH 8.5). After rinsing with distilled water, the dye fixed on the residual cells was eluted using 0.1 N hydrochloric acid and the optical density read on a spectrophotometer at 665 nm. The absorbence of the eluted dye is proportional to the number of viable cells present so that the percentage cytotoxicity is: (
O.D. cells + antiseptic solutions 1 O.D. cells + culture medium
where O.D. is the optical density.
Results Intraoperative cytology Of the 50 patients with histologically confirmed carcinoma, 14 (28%) also had evidence of incomplete excision histologically (defined as extension of tumour to the resection margin of the specimen). Of these, 10 had cellular cytology smears adequate for examination, of which seven were malignant. In the remaining three cellular smears benign cells only were found, and one of these patients had a second operation based on the histological findings, but no further malignancy was found in the second specimen. Only 13 (26%) of the patients could therefore be judged to have had an incomplete excision at the first operation. Five further patients (10%) had histological evidence of lymphatic or vascular invasion, despite a histologically complete resection margin. One of these had an inadequate smear and three had malignant smears. One had benign cells only. This gave an accuracy of 84%, a sensitivity of 56% and a specificity of 100%. If the inadequate smears are discounted, the sensitivity was 71%. The inadequate smears were all made early in the study, which suggests that the technique of taking scrapes with the wooden spatula had to be learned by the surgeon performing the operation.
390
M E Lucarotti et al.
toxicity to MCF7 cells was observed at dilutions greater than 1/1000 from that of the standard stock concentration. In contrast to this, Fig. 2 demonstrates that human fibroblast cultures were not as sensitive as MCF7 cells to low concentrations of antiseptics and thus significant toxicity was primarily seen at dilutions less than 1/1000. The most effective antiseptic was chlorhexidine gluconate which was 100% toxic to MCF7 cells and 95% toxic to human fibroblasts at a 1/10 000 dilution. However, it would appear that hydrogen peroxide and Eusol substitute are more specifically cytotoxic to MCF7 cells than to human fibroblast cultures. When applied at a 1/1000 dilution, Eusol substitute killed 88% of MCF7 cells, but
100 9 -Log
0
dLLutuon
-2- ChLorhex,oine
---
-|V- Betocine
-O-
H_j&ogen
PeroxWde
753
5ovt.docL
(a) 3
0
-)0- HLd-ogen PeroxLde
-2- ChLorhextdne -V- Betodne
-O-
SavLod.L
(a)
-Log oLutLOn --
OEusoL
-2- Mercur'Lc
Chtor
de
X Noxyflex -2O- Tour-oLLn
(b) Figure 1. (a, b) Effects of antiseptics cells in vitro.
Jn on
MCF7 breast tumour
Xj
.S
No patient in the benign group (31) or in the completely excised group with no vascular or lymphatic invasion had a malignant cell smear. When the data were combined, the overall accuracy of the technique was 90% with the same sensitivities and specificities. -Log
-2- EusoL ---V Mercuroc
Cytotoxicity of antiseptics All the antiseptics tested in this study appeared to be toxic to both MCF7 cells and human fibroblast cultures in vitro. From the results shown in Fig. 1, significant
dlutoLon --*-
Noxyflex
DC)orLde --- Toacron
(b) Figure 2. (a, b) The effects of antiseptics on human fibroblasts in vitro.
Antiseptic toxicity to breast carcinoma only 6% of human fibroblasts. Similarly, hydrogen peroxide at a dilution of 1/1 000 000 was 94% toxic to MCF7 cells but only 12% toxic to human fibroblasts.
Discussion The reported incidence of local recurrence in recent studies of conservative breast surgery, with adjuvant radiotherapy, varies greatly due to different treatment regimens and patient selection. A NSABP trial reported an 8% local recurrence at 5 years for segmental mastectomy and radiotherapy, but excluded patients with positive histological margins who underwent mastectomy (8). Montague (9) reported a 5% 5-year local recurrence, but more than 10% of patients had in situ disease only. Although there is no evidence that local recurrence influences long-term survival it could nevertheless be more devastating psychologically to the patient than the original diagnosis, and is evidence to her of primary treatment failure. It also necessitates, in many cases, more extensive surgery. There is no clinical data from controlled trials to support any specific surgical technique, in particular how wide a margin of normal breast tissue should be included around the tumour. The goal of conservative surgery is to achieve local control and inadequacy of surgical excision and intralymphatic extension correlate with the risk of recurrence (10-12). The multicentricity of breast cancer has been much debated as a possible contraindication to breast-conserving therapy. Holland (13) found in consecutive cases that 17% had multifocal disease. Gump (14) in another study found an incidence of 19%. Both are higher than the 10% seen in the NSABP trial (8). However, the natural history of such subclinical lesions is not known (7) and their prevalence may be overestimated in pathological studies of selected patients. Most recurrences also occur close to the previous excision site after breast conserving treatment which favours the cause of recurrence being due to inadequate local excision. The technique of intraoperative cytological assessment of the excision cavity may alert the surgeon to incomplete tumour excision at the time of surgery and if more extensive surgery is not undertaken may act as a prognostic index. Adjuvant treatment may be adjusted accordingly. The technique is simple and can produce a result quicker than frozen section. In our opinion and from these early results the sensitivity of the technique might improve with experience. The possibility of implantation of exfoliated tumour cells into newly cut tissues was postulated by Ryall in 1908 (15). Today various precautions have been recommended in colorectal surgery to prevent implantation of exfoliated tumour cells. These include peroperative irrigation of the intestinal lumen with a cytotoxic agent. The incidence of suture-line recurrence has been shown to be reduced using such precautions (16,17). Umpleby and Williamson have demonstrated that topical chlorhexidine-cetrimide and povidone-iodine have effective
391
cytoxicity against extraluminal tumour cells shed from colorectal carcinomas (18). It is accepted that postoperatiuve radiotherapy decreases the rate of local recurrence in breast cancer and that lower doses of radiotherapy result in higher recurrences (19). In this study we have found that hydrogen peroxide and Eusol substitute show a preferential cytotoxic action towards MCF7 (breast tumour) cells than to human fibroblasts when applied at appropriate concentrations in vitro. This suggests that it may be practicable to sterilise the tumour excision bed by local irrigation using a cytotoxic agent in a way analogous to the current practice in colorectal surgery. In this study we have shown that malignant cells are frequently present in the cavities left after conservative surgery for breast carcinoma and can be harvested by scrapings and recognised by cytology. The cytotoxic effects of antiseptics in tissue culture suggest that they may be a useful adjunct to conservation therapy.
References I Kelsey JL. A review of the epidemiology of human breast cancer. Epidemiol Rev 1979;1:74-109. 2 Baum M. Carcinoma of the breast. In: Taymor S ed. Recent Advances in Surgery 10. New York: Churchill Livingstone, 1980:241-57. 3 Harris JR, Heilman S, Kinne DW. Limited surgery and radiotherapy for early breast cancer. N Engl J Med 1985;313: 1365-8. 4 Atkins H, Hayward JL, Klugman DJ, Wayne AB. Treatment of early breast cancer. A report after ten years of clinical trial. Br Med Jf 1972;2:423-9. 5 Veronesi V, Saccozzi R, Del Vecchio M et al. Comparing radical mastectomy with quadrantectomy in patients with small cancers of the breast. N Engl J Med 198 1;305:6-1 1. 6 Sarrazin D, Le M, Ronesse J et al. Conservative versus mastectomy in breast cancer tumours with macroscopic diameter of 20 millimetres or less. Cancer 1984;53: 1209-13. 7 Fisher B, Bower M, Margoles R et al. Five year results of a ransomised clinical trial comparing total mastectomy and sequential mastectomy with or without radiation in the treatment of breast cancer. N EnglJ7 Med 1985;312:665-73. 8 Fisher ER, Sass R, Fisher B et al. Pathologic findings from the National Adjuvant Breast Project (protocol 6) II Relation of local breast recurrence to multicentricity.
Cancer 1986;57:1717-24. 9 Montague ED. Conservative surgery and radiation therapy
in the treatment of operable breast cancer. Cancer 1984;53: 700-4. 10 Aspegren K, Holmberg L, Adami HO. Standardisation of the surgical technique in breast conserving treatment of mammary cancer. Br J Surg 1988;75:807-10. 11 Levene MB, Harris J, Hellman S. Treatment of carcinoma
of the breast by radiation therapy. Cancer 1977;6:2840-5. 12 Barr LC, Brunt AM, Goodman AG, Drury A, Phillips RH, Ellis H. The primary management of breast cancer: is breast conservation feasible? Ann R Coil Surg EngI 1989; 71:390-3. 13 Holland R, Veling SHJ, Mravunach M, Hendrics JHCL. Histologic multifocality of Tis, T1-2, breast carcinomas.
392
14
15 16
17
M E Lucarotti et al.
Implications for clinical trials of breast conserving surgery. Cancer 1985;56:979-90. Gump FE, Shikora S, Habib DV, Kister S, Logerto P, Estabrook A. The extent and distribution of cancer in breast with palpable tumours. Ann Surg 1986;204:384-90. Ryall C. The technique of cancer operations with relation to the danger of cancer infection. Br Med J 1908;2: 1005-8. Keynes WM. Implantation from the bowel lumen in cancer of the large intestine. Ann Surg 1961;153:357-64. Southwich HW, Harridge WM, Cole WE. Recurrence at the suture line following resection for carcinoma of the
colon. Incidence following preventive measures. Am J Surg
1962;103:86-9. 18 Umpleby AC, Williamson RCN. The efficacy of agents employed to prevent anastomotic recurrence in colorectal carcinoma. Ann R Coll Surg Engl 1984;66:192-4. 19 Hayward J. Trials of wide excision and radiation therapy at the breast unit, Guy's Hospital, London. In: Zander J, Baltzer J, eds. Early Breast Cancer, Berlin: Springer Verlag, 1985:253-63.
Received 22 May 1990
Assessor's comment This is a very well-designed and executed experimental project. From the in vitro experiment they have shown that hydrogen peroxide and Eusol substitutes show a preferential cytotoxic action towards MCF7 tumour cells and human fibroblasts when applied in appropriate concentrations. They have also harvested tumour cells from the cavity following conservative management of breast cancer. Local antiseptic applied to the cavity would, therefore, be a useful adjuvant to conservative therapy.
However, because of the multicentricity of breast cancer it may be difficult to confer the same benefit in vivo, and hopefully they will now proceed to a controlled, double-blind trial in vivo. F GIVEN MSc MCh FRCSI Professor of Surgery Regional Hospital Galway, Ireland
Notes on books Difficult Problems in General Surgery edited by John L Sawyers and Lester F Williams. 545 pages, illustrated. Yearbook Medical Publishers, Chicago. 1989. £56.50. ISBN 0 8151 7544 2
The intention of this book is to integrate into one volume the experiences of distinguished surgeons in dealing with complicated problems of general surgery. The authors have been selected because of their particular expertise in the problem they discuss. Three-quarters of the book is devoted to surgery of the gastrointestinal tract and includes such problems as short bowel syndrome, low rectal carcinoma, biliary fistula, liver injuries and retained common bile duct stones. There are also sections on vascular surgery, recurrent groin hernia, burns, acute necrotising fasciitis and intra-abdominal sepsis. Each chapter can be read independently of the others. There can be few general surgeons who would not profit and learn from reading this book.
Monitoring in Anesthesia and Critical Care Medicine edited by Casey D Blitt. 2nd edition. 903 pages, illustrated. Churchill Livingstone, New York. 1990. £75.00. ISBN 0 443 08635 4
This new edition of a textbook first published 5 years ago continues to have as its goal the dissemination of detailed information regarding the monitoring of the anaesthetised
patient and the patient in the Intensive Care Unit. It is designed to be a reference book. All chapters have been updated to reflect advances and new chapters have been added including sections on pulse oximetry, transoesophageal echocardiography and monitoring and patient safety.
Textbook and Colour Atlas of the Temporomandibular Joint edited by John E deB Norman and Paul Bramley. 262 pages, illustrated. Wolfe Medical Publications, London. 1990. £95.00. ISBN 0 7234 A
1518 8
This is a handsome book. Produced in large format, it contains a wealth of high-quality colour photographs as well as an extensive text covering all aspects of the diseases, disorders and surgery of the temporomandibular joint. The early part relates to anatomy and surgical access, together with pathology, medical diseases and imaging. The authors then go on to discuss arthroscopy and arthroscopic surgery in some detail, followed by dislocation, ankylosis and the surgery of internal derangement. There follows an extensive section on growth problems and the volume concludes with a section on unusual surgical diseases and disorders. An important volume that will surely find a place on the shelves of every maxillofacial unit as well as in the private libraries of a large number of maxillofacial surgeons.
