World Journal of Surgery

, Volume 28, Issue 1, pp 80–86

Practice Patterns in Breast Cancer Surgery: Canadian Perspective

Authors

    • Department of Surgery, Dalhousie UniversityCancer Care Nova Scotia
  • Heather McMulkin-Tait
    • Department of Surgery, Dalhousie UniversityCancer Care Nova Scotia
Original Scientific Reports

DOI: 10.1007/s00268-003-7040-6

Cite this article as:
Porter, G. & McMulkin-Tait, H. World J. Surg. (2004) 28: 80. doi:10.1007/s00268-003-7040-6

Abstract

Breast cancer is a common disease, and the surgical management is continually evolving. The objective of this study was to describe the current breast cancer practice patterns among Canadian surgeons. All active General Surgeons (n = 1172), as accredited by the Royal College of Physicians and Surgeons of Canada, were sent a 31-item questionnaire. Anonymous responses were collected and analyzed regarding surgeon demographics, practice, and perceptions regarding surgical care of breast cancer patients. Overall 640 active surgeons responded; of these, 519 (81%) treated breast cancer and formed the study cohort. Practice settings included community (55%), community with university affiliation (28%), and academic (17%). The majority of surgeons (76%) stated that < 25% of their practice was devoted to breast disease, and 42% performed ≤ 2 breast cancer operations/month. Immediate breast reconstruction (IBR) was used by 57% of surgeons. On multivariate analysis, higher surgeon volume of breast cancer cases (p = 0.0008), fellowship training in Surgical Oncology (p = 0.009), community population (p = 0.001), and academic practice setting (p < 0.0001) were independently associated with the use of IBR. Of the 640 surgeons who responded, 79% stated that breast cancer surgery should be performed by “most general surgeons.” In Canada, most breast cancer surgery was performed by general surgeons who did not appear to have an interest, as defined by training or clinical volume, in breast cancer. Although variability regarding specific surgical issues was found among subgroups of surgeons, the majority of respondents felt that most general surgeons should treat breast cancer.

Introduction

Approximately 19,500 new cases of breast cancer are diagnosed and treated in Canada each year. Breast cancer is the most common malignancy and the second most common cause of cancer death among Canadian women [1]. In the majority of cases, surgery is an important component of initial treatment; this surgical management has evolved considerably over the past century. Radical mastectomy, as described by Halsted to include removal of the entire breast, pectoral muscles, and axillary lymph nodes, was initially considered to be standard therapy [2]. Approximately 30 years ago, the concept of breast-conserving therapy with wide local resection of the primary tumor, axillary node dissection, and postoperative breast radiotherapy was shown to be as effective as mastectomy for most women with operable breast cancer [3, 4, 5, 6, 7, 8, 9].

More recently, the use of neoadjuvant chemotherapy in patients with locally advanced tumors [10, 11, 12], the development of sentinel lymph node biopsy and the consequent potential to avoid routine axillary node dissection [13, 14, 15, 16, 17, 18], and the widespread acceptance of immediate breast reconstruction as a therapeutic option for many breast cancer patients requiring mastectomy [19, 20, 21] are examples of important advances for surgeons managing breast cancer patients. Given these changes and options, the purpose of this study was to examine practice patterns related to current surgical management options in breast cancer. More specifically, this study sought to describe the demographics of breast cancer surgery in Canada, and to identify factors associated with the use of immediate breast reconstruction and perceptions regarding its use.

Materials and Methods

A questionnaire was developed for use in the Canadian Breast Cancer Surgery Survey (CBCSS). This survey was developed to ensure clarity and comprehensiveness for the target population. The questionnaire contained demographic information and practice pattern information. In addition, the survey requested specific information from surgeons regarding the use and availability of sentinel lymph node biopsy (SLN Bx) and immediate breast reconstruction (IBR), as well as their opinion on who should provide the surgical care of breast cancer patients. The CBCSS was written in English; no French translation was provided. Questionnaire questions used in this analysis are presented in Appendix 1 .
https://static-content.springer.com/image/art%3A10.1007%2Fs00268-003-7040-6/MediaObjects/03-7040-6fapp1.gif
Appendix 1

.

The CBCSS was sent to all accredited General Surgeons within Canada in July 2001. It was accompanied with a cover letter explaining the purpose of the study; a self-addressed, stamped envelope was included to encourage return. The CBCSS was completed in an anonymous manner and an embedded four-digit code within the survey allowed linkage for the investigators to establish response or non-response. A second mail out was conducted to all surgeons who had not returned the questionnaire after 60 days.

Geographic boundaries were established a priori based on population as follows: East (Newfoundland, Nova Scotia, New Brunswick, Prince Edward Island), Central (Ontario, Quebec), and West (Manitoba, Saskatchewan, Alberta, British Columbia, Territories). Population was based on the census population (Statistics Canada, 2001) of the municipality where the surgeons’ practice was located. The presence and type of post-General Surgery fellowship training was established as reported by the respondents, as was whether they treated breast cancer (assumed to involve surgical treatment, not limited to diagnostic procedures), their practice setting (community, community with university affiliation, academic), and volume of breast cancer surgery (number of breast cancer cases/month, percentage of practice devoted to breast disease).

Univariate analyses included the Student’s t-test to compare continuous variables and the χ2 test to compare categorical variables. Two-tailed tests were used whenever the data allowed, and statistical significance was set a priori at p < 0.05. To control confounding, multivariate analysis was performed using logistic regression to assess factors associated with the use of IBR. An independent data manager performed all data entry and data management. Statistical analysis was performed using SPSS software (SPSS Chicago, IL).

Results

Of the 1172 mailed questionnaires, 690 were returned, for a response rate of 59%. Of these, 50 (4%) were not completed by the surgeon because of retirement (n = 28), absence of clinical practice (n = 11), and other (n = 11). Of the 640 completed questionnaires, 79% were returned from the first mail out, and 21% were returned from the second. The distribution of completed questionnaires by province and geographic area, depicted in Table 1, showed a significantly lower response rate from surgeons in Central Canada, largely because of the low response rate obtained from Quebec. Of the 640 surgeons who responded, 519 (81%) treated breast cancer; these form the cohort upon which further analyses are based.
Table 1.

Distribution of completed questionnaires by province and geographic region.

Province

Total

Completed response (%)

 

Geographic area

Total

Completed response (%)

New Brunswick

32

59

}

East

105

71

Nova Scotia

46

76

Prince Edward Island

6

83

Newfoundland

21

76

Quebec

295

40

}

Central

735

49*

Ontario

440

55

Manitoba

51

53

Saskatchewan

32

78

}

West

332

63

Alberta

105

67

British Columbia

140

61

Territories

4

75

*p < 0.001 comparing Central to East and West

Descriptive characteristics of the study cohort (n = 519) are presented in Table 2. The mean age of the study cohort was 47 years (median: 45; range: 30–68), and the mean interval since certification in General Surgery was 15 years (median: 13; range: 0–36). The majority of surgeons worked in a community practice setting (55%) and had not done any fellowship training after General Surgery (70%). Most surgeons (75%) reported that breast cancer comprised < 25% of their clinical practice. Forty-five surgeons (9%) had fellowship training in Surgical Oncology; no “breast fellowships” were reported.
Table 2.

Characteristics of the study cohort (n = 519).

 

n

%

Gender

  

 Male

436

84

 Female

83

16

Geographic area

  

 East

65

12

 Central

274

53

 West

180

35

Fellowshipa

  

 No

363

70

 Yes—Surgical Oncology

45

9

 Yes—Other

111

21

Practice setting

  

 Community

285

55

 Community with university affiliation

146

28

 University

87

17

Population

  

 < 50,000

198

38

 50,000–500,000

189

36

 > 500,000

132

26

Days/week in operating room

  

 ≤1

135

26

 1.5-2.0

302

58

 >2.0

82

16

% practice related to breast disease

  

 < 25%

393

76

 25%–50%

79

15

 > 50%

47

9

Breast cancer cases/month

  

 ≤ 2

217

42

 3–5

190

37

 6–10

75

15

 > 10

37

7

aFellowship training after General Surgery training.

Of the 519 surgeons who reported treating breast cancer, 452 (87%) stated that axillary lymph node dissection was the standard of care in the surgical treatment of most patients with invasive breast cancer; 138 (27%) surgeons performed SLN Bx in breast cancer. Further details regarding the use of SLN Bx are reported elsewhere [22].

Immediate Breast Reconstruction

Among surgeons who treated breast cancer, 296 (57%) claimed IBR was available at their institution. Factors associated with the availability of IBR are shown in Table 3. On univariate analysis, fellowship training, practice setting, population, percentage of the surgeon’s practice related to breast disease, and number of breast cancer cases/month were found to have a significant association with the availability of IBR. The association between the numbers of days the surgeon was in the operating room/week and availability of IBR was also significant, although the highest rate of IBR availability was found among surgeons within the intermediate group, whereas surgeons with abundant access to the operating room (> 2 days/week) had the lowest rate of IBR availability. On multivariate analysis, only fellowship training (p = 0.009), practice setting (p < 0.0001), population (p = 0.001), and number of breast cancer cases per month (p = 0.0008) appeared to be independently associated with the availability of IBR.
Table 3.

Factors associated with surgeons where immediate breast reconstruction (IBR) was available at their institution.

 

% with IBR

p Value univ.

p Value multi.

OR (95% CI)

Surgeon gender

 

NS

NS

 

 Female (n = 83)

65

   

 Male (n = 436)

56

   

Geographic area

 

NS

NS

 

 East (n = 65)

46

   

 Central (n = 274)

58

   

 West (n = 129)

59

   

Fellowship training

 

<0.0001

0.009

 

 No (n = 363)

47

  

1.0

 Yes—Surgical Oncology (n = 45)

91

  

1.9 (0.5–6.4)

 Yes—other (n = 111)

78

  

2.5 (1.2–3.8)

Practice setting

 

<0.0001

<0.0001

 

 Community (n = 285)

38

  

1.0

 Community + university affiliation (n = 146)

71

  

2.5 (1.5–4.0)

 University (n = 87)

97

  

17.3 (5.2–58.3)

Population

 

<0.0001

0.001

 

 < 50,000

29

  

1.0

 50,000-500,000

67

  

3.3 (2.1–5.4)

 > 500,000

83

  

5.5 (3.0–10.1)

Days in operating room/week

 

<0.002

NS

 

 ≤ 1 (n = 135)

53

   

 1.5-2.0 (n = 301)

63

   

 2 (n = 82)

43

   

% of practice dedicated to breast disease

 

<0.0001

NS

 

 < 25% (n = 393)

51

   

 25%–50% (n = 79)

71

   

 > 50% (n = 47)

87

   

No. of breast cancer operations/month

 

<0.0001

0.0008

 

 ≤ 2 (n = 217)

45

  

1.0

 3–5 (n = 190)

56

  

1.4 (0.9–2.3)

 6–10 (n = 75)

76

  

2.5 (1.2–4.9)

 > 10 (n = 37)

92

  

5.3 (1.4–19.7)

univ.: univariate; multi.: multivariate; OR: odds ratio; CI: confidence interval.

In attempting to further examine surgeon perceptions and practices, we asked three specific questions regarding the use of IBR in ductal carcinoma in situ and invasive breast cancer. These questions, and the distribution of responses, are depicted in Figure 1. In patients with indications for a mastectomy, the majority of surgeons “always or usually” discussed IBR, although this rate was higher if the patient had ductal carcinoma in situ (DCIS) versus invasive cancer (69% versus 51%; p < 0.0001). Interestingly, although 51% of surgeons “always or usually” discuss IBR in patients with invasive breast cancer with indications for a mastectomy, only 42% of respondents felt IBR was oncologically safe in patients with invasive cancer without skin involvement.
https://static-content.springer.com/image/art%3A10.1007%2Fs00268-003-7040-6/MediaObjects/03-7040-6f1.gif
Fig. 1.

Surgeon responses (n = 519) to specific questions regarding perceptions and practice of immediate breast reconstruction (IBR). A. In a patient with ductal carcinoma in situ with indications for a mastectomy, do you discuss possible IBR? B. In a patient with invasive breast cancer with indications for a mastectomy, do you discuss possible IBR? C. In patients with invasive breast cancer without clinical skin involvement, do you feel that skin-sparing mastectomy is oncologically safe?

Among surgeons where IBR was available (n = 296), we examined the impact of population, surgeon training, volume of breast cancer surgery, and practice setting as to whether surgeons discussed IBR with patients (Table 4). Surgeons with Surgical Oncology training were significantly more likely to “always or usually” discuss IBR for women with a diagnosis of DCIS or invasive breast cancer and indications for a mastectomy. A similar finding was noted among surgeons in a university practice setting, but only for DCIS.
Table 4.

Among patients with DCIS or invasive breast cancer with indications for a mastectomy, the percentage of surgeons who “usually or always” discuss IBR (includes only surgeons with IBR available at institution; n = 296).

 

DCIS (%)

p Value

Invasive breast cancer (%)a

p Value

Fellowship training

 

0.002

 

0.008

 No (n = 169)

78

 

57

 

 Yes—Surgical Oncology (n = 41)

98

 

83

 

 Yes—other (n = 86)

81

 

70

 

Practice setting

 

NS

 

0.04

 Community (n = 108)

79

 

59

 

 Community + university affiliation (n = 104)

79

 

61

 

 University (n = 84)

88

 

75

 

Population

 

NS

 

NS

 < 50,000 (n = 58)

79

 

57

 

 50,000–500,000 (n = 128)

79

 

67

 

 > 500,000 (n = 110)

85

 

65

 

No. of breast cancer operations/month

 

NS

 

NS

 ≤ 2 (n = 98)

79

 

67

 

 3–5 (n = 107)

78

 

56

 

 6–10 (n = 57)

88

 

65

 

 > 10 (n = 34)

91

 

79

 

DCIS: ductal carcinoma in situ.

a:No skin involvement.

Patterns of Breast Cancer Practice

Of the 519 respondents who treated breast cancer, 408 (79%) felt “most General Surgeons” should perform breast cancer surgery. Of those who thought that breast cancer surgery should be performed predominantly by specific surgeons, 85 (80%) felt that such surgeons should have fellowship-level training in breast surgery/surgical oncology or have a “major interest” in breast cancer surgery. Only three surgeons (3%) thought that breast cancer surgery should be performed exclusively by surgeons with fellowship-level training.

Most respondents (71%) responded that, if given the option, they would like the proportion of their practice related to breast disease to remain the same, and 19% and 10% wished such a proportion to increase and decrease, respectively. A significantly higher proportion of surgeons in a university or community with university affiliation practice setting wished the proportion of their clinical practice related to breast disease to decrease compared to surgeons practicing in a community setting (15% versus 6%; p = 0.008).

Discussion

Variability in many aspects of breast cancer treatment and outcome has been previously described in the literature. Much of this work was carried out in the United States and has focused on factors associated with the use of breast-conserving surgery [23, 24, 25, 26] and the use of adjuvant therapy [27, 28, 29]. However, there remains a relative paucity of literature examining actual, present-day surgical practice patterns. The combination of advances in the surgical treatment of breast cancer and the high frequency of this diagnosis make understanding such practice patterns important, particularly in a government-funded health care system such as that in Canada.

This study found that the surgical treatment of breast cancer was not “subspecialized” in Canada. Most surgeons involved in the treatment of breast cancer had no further training beyond a General Surgery residency (70%) and had practices where breast disease comprised < 25% of their clinical work (76%). A significant proportion of surgeons involved in breast cancer care (42%) performed ≤ 2 breast cancer operations per month.

However, over the past 10 years, there have been a significant number of publications examining the impact of training and disease-specific volume of the surgeon on outcome in several surgically treated malignancies [30]. Improved short-term and long-term outcomes among patients of higher volume or specialty-trained surgeons have been documented in many malignancies, including pancreatic [31, 32, 33], esophageal [34], ovarian [35], and colorectal cancer [36, 37, 38, 39]. Specifically in breast cancer, two British studies have demonstrated improved survival among patients of higher volume surgeons (> 30 cases/year) and among patients of “specialist surgeons” [40, 41]. In New York State, 5-year survival was improved across all stages among patients treated at higher volume hospitals [42]. Although the present study did not examine outcomes, it did identify significant variations in surgeon practices and perceptions according to volume and training.

The appropriate response to demonstrated volume outcome and training outcome relationships remains controversial. One approach would be to regionalize care to specific “centers of excellence,” as has been suggested for the surgical treatment of pancreatic, esophageal, and rectal cancer [32, 34, 36]. However, the incidence of breast cancer makes regionalization of surgical treatment difficult and likely untenable. Furthermore, the overwhelming majority of respondents felt that “most General Surgeons” should perform breast cancer surgery, and most respondents (71%) were content with the proportion of their practice related to breast disease; the geographic and population characteristics of Canada support this approach. For these reasons, education initiatives may play an important role in reducing variability and ensuring that current surgical approaches are available to all patients. Such an approach has been used successfully in the treatment of rectal cancer [43].

We have previously reported the Canadian experience with SLN Bx in breast cancer [22]. To summarize, we found that only 27% of surgeons performed SLN Bx, and its use was much more common among surgeons with an interest in breast cancer surgery (those with a Surgical Oncology fellowship or performing a high volume of breast cancer surgery). In addition, lack of resources appeared to be the most significant limitation to more widespread use of the technique. In the present analysis, analogous findings were noted with IBR—fellowship training and higher number of breast cancer cases per month were associated with surgeons who had IBR available. Although this study did not attempt to discern the reason for IBR use or non-use, surgeon perception appeared to play a significant role based on responses noted in Figure 1.

In the past, many concerns regarding the use of IBR in breast cancer, particularly those related to the “oncologic safety” of the procedure, have limited its use [44]. More recently, many studies have demonstrated that immediate reconstruction does not appear to be associated with higher recurrence rates, nor does it impair the detection of local recurrence or delay the administration of systemic chemotherapy [19, 20, 45]. However, IBR appears to remain an underused option in the surgical management of breast cancer. The National Cancer Database, a joint project of the American College of Surgeons and the American Cancer Society, found that only 8.3% of mastectomy patients underwent immediate reconstruction [46]. Factors associated with the use of IBR have been reported to include patient age, income, and ethnicity, as well as tumor stage, hospital type, and geographic location. Although specific tumor features such as skin involvement or the presence of multiple positive axillary lymph nodes requiring postmastectomy radiotherapy may be explain some of these factors, patient and physician perceptions and preferences likely account for most of the variability. The results of the present study support this hypothesis, particularly in identifying demographic characteristics of surgeons associated with the use of immediate breast reconstruction.

This study identified four factors independently associated with surgeons who have IBR available: fellowship training in Surgical Oncology, increasing number of breast cancer cases/month, university practice setting, and population of the surgeons. Not surprisingly, these factors support the hypothesis that surgeons with more of an interest in breast cancer (either by breast cancer volume or by fellowship training) and those with more subspecialty resources are more likely to have the more complex multimodality treatment available. The univariate association of number of days/week in the operating room and the use of SLN Bx is interesting; surgeons with the greatest access to the operating room (> 2 days/week) had the lowest rate of IBR availability. We cannot conclusively explain this finding, although its lack of association on multivariate analysis would suggest significant confounding with other independently significant factors such as fellowship training in Surgical Oncology and practice setting. As an example, our study did demonstrate an association between community practice setting and a greater number of days/week in the operating room (data not shown).

Response bias is a potential limitation of most surveys. The 59% response rate in this study approaches the conventional acceptable standard of 60% [47]. Although we asked all surgeons to complete the survey regardless of their involvement in breast cancer surgery, certain subspecialties appeared relatively underrepresented as respondents (critical care, transplantation) suggesting a non-response bias involving surgeons not performing breast cancer surgery. Otherwise stated, this study may not accurately quantify, and possibly overestimates, the proportion of accredited Canadian General Surgeons who perform breast cancer surgery. However, such a bias would be unlikely to affect the findings among surgeons involved in the care of breast cancer patients. For logistical reasons, the questionnaire was administered only in English, which likely explains the relatively low response rate and consequent questionable generalizability of this study’s findings to Quebec surgeons. In addition, although this study reports the use of specific breast cancer surgical techniques, it does not provide any information regarding the quality control of such surgery. For example, the benefit of IBR is likely highly related to the cosmetic results obtained; these are clearly unevaluable in this survey. Finally, the issue of the importance of patient demand was not examined in this study; this concept is very difficult to evaluate in the context of the government-funded Canadian health care system.

In conclusion, this study found that, in Canada, most breast cancer surgery was performed by General Surgeons who did not appear to have a defined interest in breast cancer, as measured by either fellowship training or volume of breast cancer within their practice. Significant variations were identified in the availability and use of IBR; surgeons with fellowship training in Surgical Oncology, higher volumes of breast cancer cases/month, and university practice settings were more likely to have IBR available. Surgeons with Surgical Oncology training were also more likely to offer IBR for both DCIS and invasive cancer. Despite this variability most respondents felt that all General Surgeons should treat breast cancer. Survey research like that presented in this article is important to define present practice patterns and to identify areas where relatively new surgical concepts and/or technologies require further development.

Résumé.

Le cancer du sein est une maladie fréquente, et la prise en charge évolue continuellement. Le but de cette étude a été de décrire la pratique courante parmi les chirurgiens canadiens. Tous les chirurgiens généraux actifs (n = 1172), accrédités par le Royal College of Physicians and Surgeons of Canada, ont reçu un questionnaire composé de 31 items. Les réponses en ce qui concerne les données démographiques des chirurgiens, leur type d’exercice et leurs perceptions en ce qui concerne la prise en charge des patientes porteuses de cancer de sein ont été collectées et analysées de façon anonyme. Au total, 640 chirurgiens actifs ont répondu; parmi ceux-ci, 519 (81%) prenaient en charge des cancers du sein et constituaient la cohorte d’étude. Parmi ces chirurgiens 55% exerçaient uniquement en ville, 28%, en ville avec un rattachement universitaire, alors que 17% étaient universitaires. La plupart des chirurgiens (76%) ont dit que < 25% de leur pratique était dédiés aux maladies du sein, et 42% réalisaient ≤ 2 opérations par mois sur le sein. Une reconstruction mammaire immédiate a été la règle chez 57% des chirurgiens. Les facteurs indépendants de reconstruction mammaire immédiate, retrouvés par analyse multivariée, comprenaient le volume opératoire élevé des cancers traités (p = 0.0008), une formation en Oncologie Chirurgicale (p = 0.009), une exercice urbaine (p = 0.001), et une affiliation universitaire (p < 0.0001). Des 640 chirurgiens qui ont répondu, 79% ont dit que la chirurgie du sein pourrait être réalisée par la «plupart des chirurgiens généraux». Au Canada, la grande majorité de la chirurgie du cancer du sein est réalisée par des chirurgiens généraux qui n’ont pas un intérêt particulier dans le cancer du sein défini par une formation ou un volume clinique élevé. Bien qu’une certaine variabilité a été retrouvée parmi un sous-groupe de chirurgiens en ce qui concerne les problèmes spécifiquement chirurgicaux, la majorité des réponses était en faveur d’une prise en charge des cancers du sein par des chirurgiens généraux.

Resumen.

El cáncer de seno es una enfermedad común y su manejo quirúrgico está en continuo desarrollo. El propósito del presente estudio fue describir los patrones actuales de práctica en el manejo del cáncer mamario por los cirujanos canadienses. Se envió un cuestionario de 31 tópicos a todos los cirujanos generales activos (n = 1172) acreditados por el Royal College of Physicians and Surgeons del Canadá. Se recolectaron en forma anónima las respuestas para su análisis según la demografía de los cirujanos, su práctica y su percepción en relación al cuidado de los pacientes con cáncer de seno. Respondió un total de 640 cirujanos activos; entre ellos, 519 (81%) tratan el cáncer mamario y constituyen la cohorte del estudio. Los lugares de práctica fueron: comunidad (55%), comunidad con afiliación universitaria (28%) y académico (17%). La mayoría de los cirujanos (76%) manifestaron que < 25% de su práctica se dedicaba en enfermedades del seno y 42% realizaba n ≤ 2 operaciones por cáncer de seno mensuales. Reconstrucción inmediata se usa por el 57% de los cirujanos. En el análisis multivariado, un alto volumen de casos de cáncer (p = 0.0008), entrenamiento formal en Cirugía Oncológica (p = 0.0009), población comunitaria (p = 0.001) y práctica en el medio académico (p < 0.0001) aparecieron como factores independientemente asociados con el uso de reconstrucción inmediata. De los 640 cirujanos que respondieron, 79% manifestaron que la cirugía del cáncer mamario debe ser realizada por la “mayoría de los cirujanos generales”. En Canadá la mayor parte de la cirugía por cáncer mamario es practicada por cirujanos generales que no parecen tener interés especializado, a juzgar por entrenamiento formal o volumen clínico, en el cáncer de seno. Aunque se encontró variación en cuanto a temas quirúrgicos específicos en el subgrupo de cirujanos, la mayoría opinó que la mayor parte de los cirujanos generales deben tratar el cáncer de seno.

Copyright information

© Société Internationale de Chirurgie 2003