EFFECT OF TAMOXIFEN ON ARTERIAL MICROVASCULAR

ANASTOMOSIS
BRENO BEZERRA GOMES DE PINHO PESSOA, M.D.,1* BRUNO BEZERRA MENEZES CAVALCANTE, M.D. (student),2
MUNIQUE PINHEIRO MAIA, M.D. (student),2 HAROLDO H. RIBEIRO FILHO, M.D. (student),2 MARCIA KIYOMI KOIKE, Ph.D.,3
SALUSTIANO GOMES DE PINHO PESSOA, M.D.,1 LUIZ GONZAGA PORTO PINHEIRO, M.D., Ph.D.,4
and EDNA FRASSON DE SOUZA MONTERO, M.D., Ph.D.5

Breast reconstruction after cancer treatment is based on the circulation of pedicle and microvascular flaps. This article aimed to verify the
effect of tamoxifen (TMX) pretreatment in arterial anastomosis in rats. Twenty female Wistar rats were equally divided into two groups.
TMX (0.3 mg/kg) was administered to the experimental group for 2 weeks orally. After this period, the right femoral artery was sectioned
and a terminoterminal anastomosis performed. The same procedure was done in the control group, except that the animals received the
vehicle without TMX. One week later, the femoral arteries were inspected for flow through the anastomosis, and the vessel near it was
sent to light microscopic examination. It was observed mild vasculite in both groups. The intimal thickness and total vessel wall in the
TMX-treated group was significantly higher. A real thrombotic effect upon the arterial vascular anastomosis was not observed, eventhough,
TMX induced intimal hyperplasia. VC 2007 Wiley-Liss, Inc. Microsurgery 27:286–288, 2007.

Breast cancer is the most frequent neoplasm in women.1 the psychological impact and stigmata of the mastectom-
Adjuvant and chemoprevention therapy utilizing the ized women. Flap reconstruction is totally dependent on
selective estrogen receptor modulators (SERM) are the vascular patency. Pedicle thrombosis is a cumbersome
mainstay breast cancer treatment beside surgery.2 Tamox- complication of pedicle and microvascular flaps, and ulti-
ifen is the first SERM in clinical use and the most stud- mately leads to tissue death and jeopardizes the recon-
ied drug in breast cancer prevention and adjuvancy. structive attempt. The concomitant treatment with tamoxi-
Tamoxifen efficacy in reducing breast cancer incidence is fen and reconstructive procedure brought the question
attested in many clinical trials over decades.3–6 about the possibility of a tendency toward pedicle throm-
As a mixed estrogen agonist and antagonist, tamoxi- bosis in microvascular flap transfer. The purpose of the
fen affects a multitude of clinical conditions, notably an present study is to investigate the effect of tamoxifen pre-
increased risk of endometrial cancer and vascular events. treatment in arterial microanastomosis.
Many clinical reports showed an increased risk of venous
and arterial thrombosis in diverse vascular beds as brain
and upper and lower extremity.7–9 The common pathway METHODS
to thrombotic and thromboembolic disease in the setting The study used 20 adult female Wistar rats, weighing
of tamoxifen therapy is not totally understood. Reduced 200 6 20 g, and the rats were divided into two groups
levels of antithrombin III, and protein C and S are impli- with 10 animals each. They were kept upon established
cated in the genesis of vascular events.10,11 Paradoxically, protocol for biomedical research in the experimental sur-
tamoxifen therapy is related to myocardial infarct reduc- gical laboratory of the Federal University of Ceará. All
tion, probably because the pathway to atherogenesis and animals had access to chow and water ad libitum in a
thrombogenesis are affected in different ways.12 12-h light and dark cycle. The animals were weighed
The reconstructive procedures to mammary ablation before the experiments.
are a great advance in the cancer treatment, as it reduces Tamoxifen (0.3 mg/kg) dissolved in sorbitol 10% and
methylcellulose was administered by oral gavage daily to
1
Division of Plastic Surgery, Department of Surgery, Federal University of the experimental group, for 2 weeks prior to the surgical
Ceara, Fortaleza, Brazil
2
Federal University of Ceara, Fortaleza, Brazil
procedures. The same preparation, except for the tamoxi-
3
Clinical Emergency Laboratory, Division of Clinical Medicine, University of fen, was administered to the control group concomitantly
São Paulo (USP), São Paulo, Brazil through the same route.
4
Division of Mastology, Department of Surgery, Federal University of Ceara
(UFC), Fortaleza, Brazil The animals were anesthetized with ketamine (50 mg/
5
Division of Surgical Technique and Experimental Surgery, Department of kg) and benzodiazepine (5 mg/kg) intraperitonially. The
Surgery, Federal University of São Paulo (UNIFESP) São Paulo, Brazil
right groin was prepared with iodinated solution and
*Correspondence to: Breno Bezerra Gomes de Pinho Pessoa, Rua Vicente
de Castro Filho, 1540 Agua Fria Fortaleza, Ceara, Brazil. the femoral vessels dissected. The femoral artery was
E-mail: breno_gpp@yahoo.com.br separated from the vein, and a standard terminoterminal
Received 11 March 2007; Accepted 14 March 2007
Published online 3 May 2007 in Wiley InterScience (www.interscience.wiley.
anastomosis was performed utilizing 11-0 nylon monofila-
com). DOI 10.1002/micr.20357 ment, eight stitches per anastomosis, microsurgical instru-
V
C 2007 Wiley-Liss, Inc.
 Arterial Microvascular Anastomosis 287

Figure 1. Photomicrograph show-

ing intimal hyperplasia in the ta-
moxifen group (A: HE 3100).
Detail (B: 3400). [Color fig-
ure can be viewed in the online
issue, which is available at www.
interscience.wiley.com.]

ments, and D.F. Vasconcelos surgical microscope with RESULTS

253 magnification in both groups. All anastomosis were
Wound Inspection
performed by the same surgeon.
One week later, the surgical wound was inspected for All groin wounds were examined. Healing score was
healing complication. The criteria described by McNamara similar in both groups (tamoxifen and vehicle groups ¼
et al.13 were utilized as follows: Grade 0 ¼ complete Grade 4). There was no wound complication (Grades 4
wound dehiscence; Grade 1 ¼ 50–75% ulceration; Grade and 5) in 100% and 70% of the animals from vehicle
2 ¼ 25–50% ulceration; Grade 3 ¼ <25% ulceration; and tamoxifen groups, respectively. Three animals in the
Grade 4 ¼ raised, healing normally; and Grade 5 ¼ tamoxifen group had partial wound dehiscence (Grades 0,
almost invisible. Briefly, the wound was inspected for the 1, 2).
dehiscence percentage.
At this occasion, the rats were anesthetized as Gross Vascular Assessment
described earlier, and the groin opened for macroscopic
After the wound inspection, the femoral artery was
and empty-and-refill test assessment. The inspection con-
dissected and observed for flow. All animals from both
sisted of visualizing pulsatile flow through the anastomo-
groups had pulsatile flow through anastomosis. The
sis site, and the test was performed by gently milking the
empty-and-refill test was also positive.
distal end of the anastomosis. After that, the site of vas-
cular anastomosis was ressected and included in paraffin
for histologic analysis. Histologic Analysis
Inflammatory infiltrate as well as mild vasculites were
observed in both groups, but it was more intense in the
Histologic Analysis tamoxifen group. The intimal thickness in the tamoxifen
The vascular histology was studied under light mi- group was significantly higher than in the vehicle group
croscopy and hematoxilin and eosin stain. Perivascular (18.2 6 8.5 lm vs. 6.1 6 1.9 lm, respectively;
infiltrate, endothelial lesions, and lumen thrombus were expressed as mean 6 SD; P ¼ 0.036, Mann–Whitney
observed. Rank sum test). The difference was reflected in the total
Vascular morphometric digital measurements were vessel wall thickness, as well (tamoxifen group 153.3 6
performed by using a 203 objective. Intimal, medium, 19.9 lm vs. 122.1 6 17.2 lm for vehicle group; P ¼
and adventitial layer thicknesses were identified and 0.043; Mann–Whitney Rank sum test) (Fig. 1).
measured using a Java image-processing program (Image
J 1.31v, NIH, USA). Results are expressed as mean
DISCUSSION
(standard error of mean, SEM). Statistical analysis was
performed using t test or Mann–Whitney Rank sum test. Chemopreventive and adjuvant therapy with tamoxi-
Data were considered statistically significant when P < fen is widely used today in pre- and postmenopausal
0.05. women with breast cancer and other solid tumors. Estab-
Microsurgery DOI 10.1002/micr
288 Pessoa et al.

lished evidence about the potential for thrombosis and arterial vascular anastomosis was not observed, even
thromboembolism is well documented. However, the though tamoxifen induced intimal hyperplasia.
exact mechanism for the thromboembolic events is not
totally understood. Reduced level of antithrombin III and
deficiencies in factor V and protein C are implicated in REFERENCES
the genesis of these events. Although the majority of
1. Surveillance, Epidemiology, and Results (SEER). SEER cancer statistics
case-report deals with venous thrombosis, a randomized review, 1975–2000. Bethesda, MD: National Cancer Institute; 2003.
study and some case reports pointed an increased risk in Available at: http://www.seer.cancer.gov/ (accessed July 25, 2006).
the arterial thrombosis in pre- and postmenopausal 2. Jordan VC. The development of tamoxifen for breast cancer therapy.
In: Jordan VC, editor. Long-Term Tamoxifen Treatment for Breast
women receiving tamoxifen.7,14 Cancer. Madison: University of Wisconsin Press; 1994. pp 3–26.
The oral administration of tamoxifen is well tolerated 3. Cuzick J, Powels T, Veronesi U, Forbes J, Edwards R, Ashley S,
by the animals and ensures reliable levels of medication. Boyle P. Overview of the main outcomes in breast-cancer prevention
trials. Lancet 2003;361:296–300.
The pretreatment period of 2 weeks was based in the 4. Freedman AN, Graubard BI, Rao SR, McCaskill-Stevens W, Ballard-
maximal life span of rats to simulate a period of chronic Barbash R, Gail MH. Estimates of the number of US women who could
administration of tamoxifen. The period superior to 6 benefit from tamoxifen for breast cancer chemoprevention. J Natl Can-
cer Inst 2003;95:526–532.
month of tamoxifen use is associated with a great inci- 5. Mauriac L. Aromatase inhibitors: Effective endocrine therapy in the
dence of thromboembolic events.11 early adjuvant setting for postmenopausal women with hormone-
It was observed on our study a tendency for wound responsive breast cancer. Best Pract Res Clin Endocrinol Metab
2006;20 (Suppl 1):S15–S29.
breakdown in the group treated with tamoxifen. This 6. Scottish Breast Cancer Trials Committee. Adjuvant tamoxifen in the
data is not corroborated by previous works. McNamara management of operable breast cancer: The Scottish trial. Lancet
et al. studied the effect of tamoxifen on endothelial 1987;2:171–175.
7. Deshmukh N, Tripathi SP. Thrombosis of tibial arteries in a patient
growth factor in vitro and did not find a real wound receiving tamoxifen therapy. Cancer 1995;76:1006–1008.
healing impairment.15 Mancoll et al. found an inhibitory 8. McDonald CC, Alexander FE, Whyte BW, Forrest AP, Stewart HJ.
effect on keloid fibroblast.16 Even though tamoxifen re- Cardiac and vascular morbidity in women receiving adjuvant tamox-
ifen for breast cancer in a randomised trial.The Scottish Cancer Tri-
tarded wound healing, it did not decrease the wound ten- als Breast Group. BMJ 1995;311:977–980.
sile strength. 9. Wilson CB, Lambert HE, Scott RD. Subclavian and axillary vein
Breast reconstruction with free flaps is a reliable thrombosis following radiotherapy for carcinoma of the breast. Clin
Radiol 1987;38:95–96.
and a common option after mastectomies.17 The pedi- 10. Weitz IC, Israel VK, Liebman HA. Tamoxifen-associated venous
cle flow to the flaps is the determinant of success thrombosis and activated protein C resistance due to factor V Lei-
and failure of the reconstructive attempt. Our data did den. Cancer 1997;79:2024–2027.
11. Cushman M, Constantino JP, Bovill EG, Wickerham DL, Buckley
not demonstrate an increase in arterial thrombosis after L, Roberts JD, Krag DN. Effect of tamoxifen on venous thrombosis
microvascular repair. We could not find macrosco- risk factors in women without cancer: The breast cancer prevention
pic differences in the vascular anatomy and external trial. Br J Haematol 2003;120:109–116.
12. Bonani B, Joahansson H, Guerrieri-Gonzaga A, Sandri MT, Mariette
observation of flow between treated and nontreated F, Lien EA, Decensi A. Effect of tamoxifen at low doses on ultra-
groups. sensitive C-reactive protein in health woman. J Thromb Haemost
In the histologic analysis, we did not observe intralu- 2003;1:2149–2152.
13. Mcnamara DA, Walsh TN, Kay E, Bouchier-Hayes DJ. Neoadjuvant
minal thrombus. A perivascular inflammatory infiltrate antiangiogenic therapy with tamoxifen does not impair gastrointesti-
was observed in both group and it did not differ with nal anastomotic repair in the rat. Colorect Dis 2003;5:335–341.
14. Saphner T, Tormey DC, Gray R. Venous and arterial thrombosis in
each other. The difference in intimal thickness points to- patients who received adjuvant therapy for breast cancer. J Clin
ward a possible tamoxifen effect favoring arterial throm- Oncol 1991;9:286–294.
bosis. 15. McNamara DA, Harmey J, Wang JH, Kay E, Walsh TN, Bouchier-
Hayes DJ. Tamoxifen inhibits endothelial cell proliferation and
attenuates VEGF-mediated angiogenesis and migration in vivo. Eur J
Surg Oncol 2001;27:714–718.
16. Mancoll JS, Macauley RJ, Phillips LG. The inhibitory effect of ta-
CONCLUSION moxifen on keloid fibroblast. Surg Forum 1996;47:717–720.
17. Garvey PB, Buchel EW, Pockaj BA, Casey WJ, Gray RJ, Hernandez
Our data showed that tamoxifen induced a tendency JL, Samson TD. DIEP and pedicled TRAM flaps: A comparison of
toward inflammation. A real thrombotic effect upon the outcomes. Plast Reconst Surg 2006;117:1711–1719.

Microsurgery DOI 10.1002/micr