For women at hereditary risk for breast carcinoma, risk reduction. Reoperations after Prophylactic Mastectomy with or without Implant Reconstruction

Transcription

1 2152 Reoperations after Prophylactic Mastectomy with or without Implant Reconstruction Sara M. Zion, M.D. 1 Jeffrey M. Slezak, M.S. 2 Thomas A. Sellers, Ph.D. 2 John E. Woods, M.D. 3 Phillip G. Arnold, M.D. 3 Paul M. Petty, M.D. 3 John H. Donohue, M.D. 3 Marlene H. Frost, Ph.D. 4 Daniel J. Schaid, Ph.D. 2 Lynn C. Hartmann, M.D. 4 1 New York Presbyterian Hospital, Weill Cornell Medical Center, New York, New York. 2 Department of Health Sciences Research, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. 3 Department of Surgery, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. 4 Department of Oncology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota. Supported in part by Grant DAMD17-94-J-4216 from the U.S. Department of Defense (LCH, DJS), Grant R from the National Cancer Institute (LCH, TAS, DJS), and funding from the Andersen Foundation (LCH, MHF). The authors are indebted to the participants in this study for their help in addressing an important topic; to Ann Harris, BA, and members of the Survey Research Center for follow-up of the patients; and to biostatistics personnel who entered data. Address for reprints: Lynn C. Hartmann, M.D., Department of Oncology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905; Fax: (507) ; hartmann.lynn@mayo.edu Received June 9, 2003; revision received July 30, 2003; accepted August 4, American Cancer Society DOI /cncr BACKGROUND. The authors characterized the unanticipated reoperations after prophylactic mastectomy, with or without implant reconstruction. METHODS. The surgical cohort was comprised of 1417 women with a family history of breast carcinoma. The women received a prophylactic mastectomy with (bilateral, n 593; contralateral, n 506) or without reconstruction (n 318) at the Mayo Clinic (Rochester, MN) between 1960 and Reoperations and indications for reoperation were compiled from medical records and a patient survey. RESULTS. Three hundred eighteen women received a bilateral (n 39) or contralateral (n 279) prophylactic mastectomy without reconstruction. With a median follow-up of 15 years, 18 women (6%) required reoperation. Most of these reoperations occurred within the first year after prophylactic mastectomy. Five hundred ninety-three women had reconstruction with implants following bilateral propylactic mastectomy. Approximately one-half of the women (52%) required at least 1 unanticipated reoperation during a median follow-up of 14 years. Approximately 39% of all reoperations occurred within 1 year of breast reconstruction and 69% within 5 years. Implant-related issues were the most common cause for reoperation. Some women with breast carcinoma elected to receive contralateral prophylactic mastectomy with therapeutic mastectomy for the affected breast. Five hundred six women received reconstruction with implants. During a median follow-up of 8.8 years, 189 women (37%) required unanticipated reoperation. The most common indication was implant-related issues. The time course of reoperations was similar to that for women in the bilateral group. CONCLUSIONS. Surgical reoperations were fairly common among women who received prophylactic mastectomy with implant reconstruction. Most of the reoperations were implant related. Reoperations were fairly uncommon after prophylactic mastectomy without reconstruction. Cancer 2003;98: American Cancer Society. KEYWORDS: prophylactic mastectomy, reconstruction, reoperations, breast carcinoma. For women at hereditary risk for breast carcinoma, risk reduction strategies include optimal surveillance, chemoprevention, or prophylactic surgery, either mastectomy or oophorectomy. 1 High-risk women include those previously unaffected with breast carcinoma and those who developed a first breast carcinoma and are at high risk for a contralateral event. Some high-risk women elect to undergo bilateral prophylactic mastectomy to avoid a cancer occurrence. Other women, who have developed a first breast carcinoma, elect to undergo a therapeutic ipsilateral mastectomy and a contralateral prophylactic procedure. Both bilateral and contralateral prophylactic mastectomies have been associated with a substantial reduction in the risk of breast carcinoma. 2 5 Women considering surgical options need information about

2 Prophylactic Mastectomy: Reoperations/Zion et al both the effectiveness and morbidities of these procedures. 1,6 The majority of women who have prophylactic mastectomy elect breast reconstruction, often with implants. 2 Unanticipated reoperations after implant reconstruction represent one source of morbidity for these patients, but there is a lack of published information on this topic. Gabriel et al. 7 reported a 30% complication rate at 5 years for a group of 92 women who received prophylactic mastectomy and implant reconstruction. In contrast, only 12% of women who had breast implants for augmentation experienced complications at 5 years in their series. In the current study, we provide long-term follow-up of 1099 women who had bilateral or contralateral prophylactic mastectomy and implant reconstruction at the Mayo Clinic (Rochester, MN) and of 318 women who had prophylactic mastectomy, but no reconstruction. We describe the time course of reoperations and the most common indications for them. MATERIALS AND METHODS Cohort Identification We used the Mayo Clinic surgical index to identify patients who had either a subcutaneous or total mastectomy between 1960 and A medical chart review confirmed the procedures that were performed with prophylactic versus therapeutic (i.e., for cancer) intent. This approach has been described in detail previously. 2,4 We studied two groups of women: women unaffected with breast carcinoma who had both breasts removed prophylactically 2 and women with breast carcinoma who had a therapeutic mastectomy and a contralateral prophylactic procedure. 4 Data Collection and Definitions We collected data regarding unanticipated reoperations from the medical records and a patient survey. The survey asked the question, After your prophylactic mastectomy, did you have any complications that required additional breast-related surgery? The respondent was then asked to provide the date and primary reason for any procedures. These responses were validated against information in the Mayo medical record. With this retrospective study, we did not attempt to distinguish between medically necessary and elective (i.e., to improve cosmesis) procedures. Reoperations identified in the medical record review were correlated with the survey data. When a procedure identified in the survey occurred within 2 years of a similar reoperation identified in the medical record review, it was assumed to refer to the same reoperation. All reoperations, whether identified in the medical record, survey, or both, were included. With the subcutaneous mastectomy, the majority of breast tissue ( 90%) is removed, leaving the nipple and areolar TABLE 1 Categories of Indications for Reoperation Implant-related issues Implant rupture Capsular contracture Implant herniation Implant leak, port failure Aesthetic implant issues Change of implant Location pocket revision Change of implant size Change of implant type Silicone implant concerns Implant anxiety Implant replacement Silicone granulomas Postoperative complications Infection Bleed Hematoma Seroma Inadequate perfusion Wound dehiscence/secondary closure Nonimplant aesthetic concerns Scar revision Nipple/areola revision Asymmetry Nodule removal Nodule removal Others Chest wall injury Pain/discomfort complex. With the total mastectomy, the entire breast is removed, including the nipple areolar complex. Any procedure that was not part of the standard breast reconstruction protocol and was performed in an operating room was counted as an unanticipated reoperation. The standard surgical protocol for implant reconstruction included the initial placement of a tissue expander implant, a 3 6-month period for chest wall tissue stretching followed by removal of the filler port, the exchange of the tissue expander implant for a permanent implant (if the expander implant was not itself permanent), and creation of the nipple areolar complex. Reasons for reoperations were collected and grouped into the following categories: immediate postoperative complications, implant-related issues, aesthetic concerns (non implant related), nodule removal, and other (Table 1). Each reoperation could have multiple indications. A reoperation on both breasts at the same time was counted as a single procedure. The current study details the unanticipated reoperations after prophylactic mastectomy. In the contralateral group, we excluded reoperations that occurred on the original site of cancer. This was to avoid confounding secondary to cancer treatment, especially radiation.

3 2154 CANCER November 15, 2003 / Volume 98 / Number 10 TABLE 2 Characteristics of 1417 Women with Implant Reconstruction or no Reconstruction after Prophylactic Mastectomy at the Mayo Clinic, Implant reconstruction (n 1099) (%) Contralateral Characteristics Bilateral (n 593) All (n 506) SCM (n 311) TM (n 195) No reconstruction (n 318) (%) Median age at PM (yrs) Median length of follow-up (yrs) Type of PM Subcutaneous 561 (95) 311 (61) 311 (100) 15 (5) Total 32 (5) 195 (39) 195 (100) 303 (95) Median no. of previous breast biopsies (range) 1 (0 32) 1 (0 15) 1 (0 15) 0 (0 5) 0 (0 22) Married at time of PM 481 (81) 419 (83) 261 (84) 158 (81) 233 (73) Nulliparous at time of PM 73 (12) 51 (10) 28 (9) 23 (12) 55 (17) SCM: subcutaneous mastectomy; TM: total mastectomy; PM: prophylactic mastectomy. We attempted to contact all patients or their next of kin with a mailed survey and, if necessary, telephone follow-up. Overall, 92% provided responses. Medical record information was available for all. Our protocol and all patient-contact documents were reviewed and approved by the Mayo Clinic institutional review board. Statistical Methods Frequencies of reoperations and indications in each category were generated. Rates of reoperation were determined by dividing the total number of reoperations in a time period by the total number of personyears in that time period. The association between factors and the rate of reoperations was assessed using the Cox regression 8 and Andersen Gill models. 9 The Cox model was used to assess the influence of potential risk factors on the relative risk (RR) of a woman s first reoperation. The Andersen Gill model is an extension of the Cox model that evaluates the influence of potential risk factors on the RR of multiple reoperations. The results of the Cox and Andersen Gill models were similar, with hazard ratios generally differing by less than 10%. Therefore, we focused on the overall experience, counting all reoperations with the Andersen Gill model. The cumulative percent of women receiving a reoperation over time was estimated by the Kaplan Meier method. Observations were censored for women who did not have a reoperation by their last follow-up. RESULTS Patient Cohort A total of 1099 women underwent prophylactic mastectomy with implant reconstruction at the Mayo Clinic between 1960 and Five hundred ninetythree women underwent a bilateral prophylactic mastectomy and 506 women underwent a contralateral prophylactic mastectomy with reconstruction (Table 2). Three hundred eighteen women had no reconstruction after bilateral (n 39) or contralateral (n 279) prophylactic mastectomy. Of the 318 women, 303 underwent a total mastectomy and 15 underwent a subcutaneous contralateral mastectomy. The median follow-up is 14.2 years for the bilateral plus reconstruction group and 8.8 years for the contralateral plus reconstruction group. In the bilateral plus reconstruction group, 561 (95%) women underwent subcutaneous mastectomy and 32 (5%) underwent total mastectomy. In the contralateral plus reconstruction group, 311 (61%) underwent subcutaneous mastectomy and 195 (39%) underwent total mastectomy. In the current study, 92% of the bilateral group and 96% of the contralateral group who underwent reconstruction did so within 2 weeks of prophylactic mastectomy. The median ages at reconstruction were 42 and 46 years, respectively. Factors Associated with Reoperation We performed a multivariate analysis to assess factors associated with reoperation after implant reconstruction (Table 3). Factors included in our model were age at reconstruction, year of reconstruction, marital status at time of surgery (married vs. not married), number of benign breast biopsies before prophylactic mastectomy, parity, and subcutaneous versus total mastectomy (contralateral groups only). In the bilateral group, nulliparous women had significantly fewer reoperations (RR 0.68, 95% confidence interval [CI] ). Although there was a trend for a similar association in the contralateral subcutaneous group, the association was not statisti-

4 Prophylactic Mastectomy: Reoperations/Zion et al TABLE 3 Risk factors for Reoperation by Mastectomy Type, Mayo Clinic, a Contralateral groups Risk factors Bilateral group All SCM TM Nulliparity 0.68 ( ) 0.77 ( ) 0.70 ( ) 0.97 ( ) Age at breast reconstruction b 0.92 ( ) 0.95 ( ) 0.94 ( ) 0.94 ( ) Yr of breast reconstruction b 1.22 ( ) 1.07 ( ) 1.00 ( ) 1.15 ( ) No. of previous breast biopsies 1.00 ( ) 1.06 ( ) 1.03 ( ) 1.20 ( ) Married at time of reconstruction 0.96 ( ) 0.80 ( ) 0.70 ( ) 1.24 ( ) SCM vs. TM 0.69 ( ) 1.09 ( ) SCM: subcutaneous mastectomy; TM: total mastectomy. a Multiple reoperations analyzed by the Andersen Gill model. b Risk ratios provided for a 10-year increment. TABLE 4 Description of Unanticipated Reoperations by Mastectomy Type and Reconstruction Status, Mayo Clinic, Implant reconstruction (n 1099) (%) Contralateral Characteristics Bilateral (n 593) All (n 506) SCM (n 311) TM (n 195) No reconstruction (n 318) (%) No. of women requiring reoperations (48) 317 (63) 181 (58) 136 (70) 300 (94) (52) 189 (37) 130 (43) 59 (30) 18 (6) (26) 142 (28) 94 (30) 48 (25) 13 (4) 2 79 (13) 28 (6) 21 (7) 7 (4) 3 (1) 3 45 (8) 14 (3) 11 (4) 3 (2) 2 (1) 4 35 (6) 5 (1) 4 (1) 1 ( 1) 0 Total no. of reoperations Percent with at least one reoperation by 1 yr yrs yrs Median time to first reoperation (mos) among those with reoperation SCM: subcutaneous mastectomy; TM: total mastectomy. cally significant. We analyzed the year of reconstruction in two ways as a continuous variable and by era. By era, women were grouped into the following time frames: pre-1975, , , , and In the bilateral group, there was a significant trend for women in more recent years to have more reoperations. However, this trend was not noted in the other groups. There was no consistent association between the number of previous breast biopsies and reoperation frequency. Within the contralateral reconstruction group, there was no difference in reoperation frequency between the subcutaneous and total mastectomy groups. Because the factors in Table 3 were only weakly associated with reoperations, and the distributions of these factors were similar across the surgical groups (Table 2), the following statistical comparisons of rates of reconstruction were not adjusted for the risk factors in Table 3. Number and Timing of Reoperations Bilateral Plus Reconstruction Group Of the 593 women, 311 (52%) had at least 1 unanticipated reoperation and 27.5% had their first reoperation by the first year (Table 4). One hundred fifty-two women (26%) had a single reoperation and 159 (27%) had 2 or more reoperations. A total of 605 reoperations were performed. Among women requiring a reoperation, the median time from reconstruction to first reoperation was 10.4 months (range, 1 day 23 years;

5 2156 CANCER November 15, 2003 / Volume 98 / Number 10 FIGURE 1. Likelihood of reoperation by type of mastectomy. Percent of women requiring at least one reoperation by time since prophylactic mastectomy (PM). SCM: subcutaneous mastectomy; TM: total mastectomy. FIGURE 2. Rate of unanticipated reoperations per 100 woman-years by time since prophylactic mastectomy, quarterly for the first 2 years and annually thereafter. SCM: subcutaneous mastectomy; TM: total mastectomy. Fig. 1). Twenty-six percent of reoperations occurred within 6 months of reconstruction. Rates of reoperations per woman-year were high in the immediate postoperative period and decreased rapidly to a fairly constant low rate after 2 years (Fig. 2). Number and Timing of Reoperations Contralateral Plus Reconstruction Group Total mastectomy was performed more commonly in this group of women than in the bilateral group (39% vs. 5%, respectively; Table 2). Therefore, we evaluated the experience in the contralateral group overall and by type of mastectomy within the contralateral group. Of the entire contralateral group of 506 women, 189 (37%) required at least one reoperation, 22.4% by the first year. Among women who underwent a reoperation, the median time to first reoperation was 7.8 months (Table 4). One hundred forty-two women (28%) had 1 reoperation and 47 (9%) women had 2 or more reoperations. Total mastectomy Fifty-nine of the 195 (32%) women who underwent a contralateral total mastectomy plus reconstruction required 75 reoperations during a median follow-up of

6 Prophylactic Mastectomy: Reoperations/Zion et al years. Forty-eight women (25%) undwent 1 reoperation and 11 (6%) had more than 1 reoperation. Among women who underwent reoperations, the median time to first reoperation was 9.6 months (range, 5 days 10 years). The time course of reoperations was similar to that in the bilateral plus reconstruction group, although the rates were somewhat lower in the first year (Fig. 2). Subcutaneous mastectomy Of the 311 women who underwent a contralateral subcutaneous mastectomy, 130 (43%) required 185 reoperations during follow-up. Ninety-four (31%) underwent 1 reoperation and 36 (12%) underwent more than 1 reoperation. Among women who received reoperations, the median time to first reoperation was 7.1 months (range, 1 day 10 years). The time course of these events was similar to that of the bilateral plus reconstruction group (Fig. 2). Number and Timing of Reoperations No Reconstruction Thirty-nine women who underwent bilateral prophylactic mastectomy did not receive reconstruction. Eight of these women (21%) required 14 reoperations during a median follow-up of 21 years. Thirty-six percent of all their reoperations occurred within 1 year of prophylactic mastectomy and 65% within 5 years. A larger group of 279 women had contralateral prophylactic mastectomy and no reconstruction. Of these, 10 women (4%) required 11 reoperations on the prophylactic side during a median follow-up of 13 years. Nine of these reoperations (82%) occurred within 1 year of prophylactic mastectomy. In the no reconstruction group, the rate of reoperation was greater among women with bilateral (P 0.01) and subcutaneous (P 0.01) mastectomies. The RR for reoperation in the bilateral versus contralateral group was 7.9 (95% CI ). For subcutaneous versus total mastectomy, including all women in the bilateral and contralateral groups, the RR was 19.5 (95% CI ). However, because of the small numbers, these estimates may not be reliable, as indicated by the wide 95% CI values. Comparisons of Reoperation Rates The implant reconstruction groups had significantly more reoperations than the no reconstruction group. The bilateral plus reconstruction group had a risk ratio of 13.0 (95% CI ) versus the no reconstruction group. The contralateral plus reconstruction group had a risk ratio of 7.7 (95% CI ). To compare the rates of reoperation between the bilateral and contralateral plus reconstruction groups, we restricted our analyses to the groups of women that had subcutaneous mastectomy because only 5% of the bilateral group had a total mastectomy. The rate of first reoperation was not statistically elevated in the bilateral group (hazard ratio 1.22, 95% CI ). However, when considering the cumulative effects of all reoperations, the rate of reoperation was statistically greater for the bilateral group (hazard ratio 1.55, 95% CI ). To evaluate the effect of type of surgery, we compared the rates of reoperation between the subcutaneous and total mastectomy groups within the contralateral reconstruction group. The rates of reoperation did not appear to differ statistically between these 2 groups for either the first reoperation (hazard ratio for subcutaneous contralateral mastectomy 1.24, 95% CI ) or the cumulative rate of all reoperations (hazard ratio 1.21, 95% CI ). Indications for Reoperation The indications for reoperation were categorized as postoperative complications, implant-related issues, aesthetic concerns (non implant related), nodule removal, and other. The specific indications incorporated within each category are listed in Table 1. We also tabulated the indications by type of surgery (Table 5). In each group with implant reconstruction, regardless of the type of mastectomy, the most frequent indication for reoperation was implant related, accounting for greater than one-half of all reoperations. Postoperative complications were responsible for 9 12% of all reoperations in the reconstruction groups. Five to 10% of women in the reconstruction groups underwent reoperation to remove nodular tissue. Essentially, there was no major difference in the type of indication for reoperation in the different implant reconstruction groups. In the no reconstruction group, there were 25 indications for 25 reoperations: postoperative complications in 7 patients (28%), aesthetic concerns in 9 patients (36%), and nodule removal in 9 patients (36%). Implant Removal In the bilateral plus reconstruction group, implants were removed from 194 of 593 (33%) and later replaced and implants were removed from 23 women (4%) and not replaced. In the contralateral subcutaneous group, implants were removed and replaced in 74 of 311 women (24%). In the contralateral total mastectomy group, implants were removed and replaced in 29 of 195 women (15%). The time course of implant removal and replacement mirrored that of all reoperations (Fig. 3). We calculated the proportion of women whose original implants remained in place at 5 years, 10 years, 15 years, and 20 after placement. For all women in both the bilateral and contralateral recon-

7 2158 CANCER November 15, 2003 / Volume 98 / Number 10 TABLE 5 Indications for Reoperations by Mastectomy Type and Reconstruction Status, Mayo Clinic, Implant reconstruction (n 1099) (%) Contralateral No. of women Bilateral (n 593) All (n 506) SCM (n 311) TM (n 195) No reconstruction (n 318) (%) No. of reoperations Total no. of indications Postoperative 135 (12) 29 (10) 19 (9) 10 (12) 7 (28) Implantrelated 688 (59) 161 (54) 113 (53) 48 (59) 0 Nodule removal 118 (10) 26 (9) 22 (10) 4 (5) 9 (36) Nonimplant aesthetic issues 179 (15) 67 (23) 52 (25) 15 (18) 9 (36) Other 38 (3) 12 (4) 7 (3) 5 (6) 0 SCM: subcutaneous mastectomy; TM: total mastectomy. FIGURE 3. Implant survival. The percent of implants remaining in place by time since initial placement for each type of surgery. SCM: subcutaneous mastectomy; TM: total mastectomy. struction groups (original n 1099), those figures are 81%, 75%, 70%, and 62%, respectively (Fig. 4). DISCUSSION Prophylactic mastectomy is considered by some women at high risk for breast carcinoma. This procedure is associated with substantial risk reduction. 2 5 However, there are significant disadvantages with this approach, including the loss of breast tissue and cutaneous sensation, the requirement for a major surgical intervention, and the irreversibility of the decision. A source of potential problems for women who elect prophylactic mastectomy is the need for subsequent unanticipated reoperation. To the best of our knowledge, there has been a relative lack of information on the long-term likelihood of such problems in this patient population. Therefore, we studied the reoperation experience in our group of women who had undergone either bilateral or contralateral prophylactic mastectomy with and without implant reconstruction. In the bilateral plus reconstruction group, 52% of women underwent reoperation over a median followup of 14.2 years. Approximately one-half of them had a single reoperation and the other half had multiple reoperations. Ninety-five percent of these women had had subcutaneous mastectomy. The median time from reconstruction to first reoperation was 10.4 months. The contralateral plus reconstruction group included women who had undergone either subcutaneous (61%) or total (39%) prophylactic mastectomy.

8 Prophylactic Mastectomy: Reoperations/Zion et al FIGURE 4. Implant survival. The percent of implants remaining in place by time since initial placement. In the entire contralateral plus reconstruction group, 37% of women underwent reoperation during a median follow-up of 8.8 years. A single reoperation was performed for 142 of these 189 women (75%; Table 4). Within the contralateral plus reconstruction group, the likelihood of reoperation was similar in the subcutaneous and total mastectomy groups. Among women who had prophylactic mastectomy, but no reconstruction, 6% required reoperation. Most reoperations in all the surgical groups occurred fairly early after the initial operation. After 2 years, the rates of reoperation tapered and remained relatively constant at a low level (Fig. 1). We grouped indications for reoperation into postoperative complications, implant-related issues, nonimplant aesthetic concerns, nodule removal, and other. In the no reconstruction group of 318 women, 18 women (6%) had an unanticipated reoperation. Twenty-eight percent of reoperations were for postoperative problems, 36% for aesthetic concerns, and 36% to remove nodular tissue (Table 5). In the implant reconstruction groups, postoperative complications represented approximately 10% of all indications, regardless of the type of mastectomy. Approximately 50 60% of indications concerned implants. These included a wide range of problems such as rupture or leakage of implant contents or capsular contracture. Capsular contracture is the most common problem associated with breast implants. 10 This refers to the body s natural reaction to a foreign body, namely, the development of a fibrous scar around the device. When the process progresses and causes excessive firmness of the breast mound, it is referred to as contracture. This does not represent an absolute indication for reoperation but may well be addressed surgically to improve cosmesis. Removal of nodular tissue represented 4 10% of indications. These reoperations could have been performed to assess a possible tumor, to improve cosmesis, or for both purposes. Aesthetic concerns, not related to implants, comprised 15 23% of the listed indications. These included scar revision and nipple areola revision. It is important to acknowledge the limitations of a retrospective analysis like ours. Some reoperations were medically necessary, but many were driven by patient preference about their cosmetic result. The same degree of capsular contracture might be tolerated by one woman but not another. We could not reliably distinguish between medically necessary and elective procedures with our study design and thus tallied all reoperations. We attempted to identify the factors associated with a higher likelihood of reoperation. First, undergoing reconstruction with implants greatly increases the likelihood of reoperation compared with prophylactic mastectomy without reconstruction. However, the majority of women want reconstruction, especially high-risk women who have not had cancer and choose a bilateral prophylactic procedure. Among women who received a bilateral prophylactic mastectomy plus reconstruction, nulliparity was associated with a reduced number of reoperations. The breast changes associated with pregnancy may lead to more ptosis and a higher likelihood for reoperation after reconstruction, which may explain the fewer reoperations

9 2160 CANCER November 15, 2003 / Volume 98 / Number 10 among nulliparous women. In the bilateral group, there was a significant trend for women in more recent years to undergo more reoperations. This was not observed in the other patient groups, however. There was also a suggestion in all groups that younger women were more likely to have reoperation, but these associations were not statistically significant. The number of previous breast biopsies, marital status, and subcutaneous versus total mastectomy (within the contralateral group) were not associated with the likelihood of unanticipated reoperation. How do these data on reoperations compare with other studies? As described earlier, we included all unanticipated reoperations in the current study, regardless of type or indication. Other studies have focused on specific types of problems, such as implant concerns only or postoperative complications. 7,11 14 Gabriel et al. 7 tabulated implant-related reoperations in 92 women who underwent prophylactic mastectomy. At 5 years, 30.4% of women in their study had an implant-related complication that required surgery. The most common problem was capsular contracture. Studies that have evaluated implants that were removed have shown an increasing likelihood of implant leakage or frank rupture with increasing implant age. 11,12 This has led to clinical concerns that the need for reoperations for implant reconstruction would increase with time. We did not observe such a time trend in our series. Some studies have suggested that implants are likely to require replacement every 5 15 years. 15 We found that 62% of women retained their original implant at 20 years. In the current era, reconstruction with autologous tissue is a common procedure. 10 Unanticipated reoperations occur less commonly after autologous tissue reconstruction. 16,17 The current study focused solely on implant reconstruction or no reconstruction. Therefore, these results cannot be extrapolated to women who are candidates for autologous tissue reconstruction. Prophylactic mastectomy represents one option for reducing the risk of breast carcinoma. When followed with implant reconstruction, there is a significant likelihood of subsequent reoperation, usually for implant-related issues. These data may help to inform decision-making for women considering prophylactic mastectomy and reconstruction. REFERENCES 1. Hartmann LC, Sellers TA, Schaid DJ, et al. Clinical options for women at high risk for breast cancer. Surg Clin North Am. 1999;79: Hartmann L, Schaid D, Woods J, et al. Efficacy of bilateral prophylactic mastectomy in women with a family history of breast cancer. N Engl J Med. 1999;340: Meijers-Heijboer H, van Geel B, van Putten WLJ, et al. Breast cancer after prophylactic bilateral mastectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med. 2001;345: McDonnell SK, Schaid DJ, Myers JL, et al. Efficacy of contralateral prophylactic mastectomy in women with a personal and family history of breast cancer. J Clin Oncol. 2001;19: Hartmann LC, Sellers TA, Schaid DJ, et al. Efficacy of bilateral prophylactic mastectomy in BRCA1 and BRCA2 mutation carriers. J Natl Cancer Inst. 2001;93: Stefanek M, Hartmann L, Nelson W. Risk-reduction mastectomy: Clinical issues and research needs. J Natl Cancer Inst. 2001;93: Gabriel S, Woods J, O Fallon W, et al. Complications leading to surgery after breast implantation. N Engl J Med. 1997;336: Cox DR. Regression models and life-tables (with discussion). J R Stat Soc [Ser B]. 1972;34: Andersen PK, Gill RD. Cox s regression model for counting processes: a large sample study. Ann Stat. 1982;10: Bostwick III, J. Breast reconstruction following mastectomy. CA Cancer J Clin. 1995;45: Rohrich RJ, Adams WP Jr., Beran SJ, et al. An analysis of silicone gel-filled breast implants: diagnosis and failure rates. Plast Reconstr Surg. 1998;102: de Camara DL, Sheridan JM, Kammer BA. Rupture and aging of silicone gel breast implants. Plast Reconstr Surg. 1993;91: O Brien W, Hasselgren P-O, Hummel RP, et al. Comparison of postoperative wound complications and early cancer recurrence between patients undergoing mastectomy with or without immediate breast reconstruction. Am J Surg. 1993; 166: Holzgreve W, Beller FK. Surgical complications and followup evaluation of 163 patients with subcutaneous mastectomy. Aesthetic Plast Surg. 1987;11: Hughes K, Papa M, Whitney T, et al. Prophylactic mastectomy and inherited predisposition to breast carcinoma. Cancer. 1999;86(Suppl): Kroll SS, Evans GRD, Reece GP, et al. Comparison of resource costs between implant-based and TRAM flap breast reconstruction. Plast Reconstr Surg. 1996;97: Kroll SS, Baldwin B. A comparison of outcomes using three different methods of breast reconstruction. Plast Reconstr Surg. 1992;90: