BREAST Outcomes Article
Breast Reconstruction with Tissue Expanders: Implementation of a Standardized Best-Practices Protocol to Reduce Infection Rates Ibrahim Khansa, M.D. Russell G. Hendrick, Jr., M.D. Alison Shore, M.D. Joseph Meyerson, M.D. Maelee Yang, B.S. James H. Boehmler IV, M.D. Columbus, Ohio; Houston, Texas; and Chicago, Ill.
cpt
Background: Periprosthetic infection remains a frustrating and costly complication of breast reconstruction with tissue expanders. Although some specific steps have been previously shown to reduce periprosthetic infections, no comprehensive protocol addressing all aspects of preoperative, intraoperative, and postoperative patient management has been evaluated in the literature. The authors’ goal was to evaluate the effectiveness of their protocol at reducing periprosthetic infections. Methods: A comprehensive, best-practices protocol was introduced and implemented in November of 2010. All patients undergoing breast reconstruction using tissue expanders at the authors’ institution in the 5 years before the protocol, and in the 2 years after, were analyzed. Results: Three hundred five patients underwent 456 tissue expander reconstructions in the 5 years before the protocol, and 198 patients underwent 313 reconstructions in the 2 years after. Significantly fewer patients developed periprosthetic infection after protocol (11.6 percent versus 18.4 percent; p = 0.042), and the number of infected tissue expanders trended toward a decrease (9.3 percent versus 13.2 percent; p = 0.097). On multivariate analysis, the protocol significantly reduced the odds of periprosthetic infection (OR, 0.45; p = 0.022). Predictors of infection included obesity (OR, 2.01; p = 0.045) and preoperative breast size larger than C cup (OR, 2.83; p = 0.006). Conclusions: The authors’ comprehensive, best-practices protocol allowed them to reduce the odds of tissue expander infections by 55 percent (OR, 0.45; p = 0.022). The authors were able to identify several potential areas of improvement that may help them lower the rate of infection further in the future. (Plast. Reconstr. Surg. 134: 11, 2014.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, III.
T
issue expansion followed by implant reconstruction remains the most commonly performed postmastectomy breast reconstruction modality.1 Unfortunately, periprosthetic infections remain a common and debilitating complication, often requiring hospital admission for intravenous antibiotics, reoperation for expander removal, and additional procedures to reconstruct the breast.2 Apart from the emotional toll periprosthetic infections may exert on From the Department of Plastic Surgery, The Ohio State University Wexner Medical Center; the Division of Plastic Surgery, Baylor College of Medicine; and Lakeview Plastic Surgery. Received for publication November 30, 2013; accepted January 15, 2014. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000261
the patient, they are also very costly complications from a financial perspective.3 The incidence of periprosthetic infections in the literature ranges from 1 to 35 percent.4 Several potential risk factors for periprosthetic infections have been reported in the literature, including chemotherapy,5,6 smoking,5,7 radiation therapy,8,9 lymph node dissection,8 large breast size,10 higher body mass index,11 immediate reconstruction,11,12 and bilateral reconstruction.12 To our knowledge, no comprehensive protocol that aims to reduce periprosthetic infections by addressing all aspects of preoperative, intraoperative, and postoperative patient care has been
Disclosure: The authors have no financial interest to declare in relation to the content of this article.
www.PRSJournal.com
11
Plastic and Reconstructive Surgery • July 2014 shown to reduce infection rates. Craft et al. have published a description of their algorithm, but its effectiveness at reducing periprosthetic infections was not evaluated.13 Several other authors have attempted to reduce infections by addressing one of the various aspects of preoperative, intraoperative, or postoperative patient care. Pfeiffer et al. demonstrated that pocket irrigation with a first-generation cephalosporin was superior to saline irrigation in breast augmentation.14 Adams et al. found low expander infection rates using triple-antibiotic irrigation solution for expander immersion and pocket irrigation.15 Murray et al. demonstrated a significant reduction in infections by applying mupirocin to drain sites, and by not allowing drains to contact the tissue expander.16 We have previously published our outcomes with 413 tissue expander reconstructions performed between 2001 and 2007.10 Our rate of periprosthetic infection and resulting expander loss (16.5 and 9.4 percent, respectively) prompted us to reexamine our practices critically. At the time, our institution did not have a standardized protocol to minimize periprosthetic infections. The choice of skin preparation varied by surgeon preference. Multiple individuals in the operating room were often allowed to handle the expander. The subpectoral pocket was not universally irrigated with triple-antibiotic solution, and the surgical team did not routinely repeat preparation of the skin or change gloves before expander insertion. The postoperative drain and antibiotic management varied widely by surgeon. In November of 2010, we introduced a comprehensive, standardized protocol that dictated precise measures to be taken in the preoperative, intraoperative, and postoperative settings. Our goal in this study was to evaluate the effects of this protocol on our periprosthetic infection rates, by comparing our results from 2005 to 2010 (before the protocol) to those from 2010 to 2012 (after the protocol).
PATIENTS AND METHODS After approval by the institutional review board, all patients who underwent postmastectomy breast reconstruction using a tissue expander at the James Cancer Hospital between November of 2005 and November of 2012 were analyzed. Patients were divided into two time periods: patients undergoing surgery between November of 2005 and November of 2010, who were not treated according to the best-practices protocol; and patients undergoing surgery between
12
November of 2010 and November of 2012, who were treated according to the protocol. Patient characteristics, history of irradiation and chemotherapy, operative technique, and postoperative management were evaluated. Univariate analysis to compare the effects of various risk factors and the protocol on infection rates was performed using chi-square analysis. Multivariate binary logistic regression was performed to eliminate the effect of confounders. A value of p < 0.05 was considered indicative of statistical significance. All statistical tests were performed using Minitab 16 statistical software (Minitab, Inc., State College, Pa.). Best-Practices Standardized Protocol In November of 2010, we introduced a comprehensive, standardized protocol to be applied to all patients having tissue expanders inserted into the breast. The protocol is divided into three parts: preoperative, intraoperative, and postoperative. Preoperatively, implant size and type were chosen by the plastic surgeon at the time of consultation based on patient base width and projection preference. Expander models used were Mentor (Mentor Corp., Santa Barbara, Calif.) and Allergan (Allergan, Inc., Santa Barbara, Calif.). Patients were asked to scrub their bodies with chlorhexidine the day before surgery and the morning of surgery. All patients received preoperative intravenous antibiotics (i.e., cefazolin, or clindamycin for penicillin-allergic patients), with the dose adjusted properly according to weight, and timed at least 30 minutes before incision.17 Intraoperatively, the skin preparation and draping were performed using chlorhexidine with the attending plastic surgeon present. In immediate reconstruction cases, the surgical oncologist performed total or modified radical mastectomies through either a traditional elliptical incision or a periareolar incision. Reconstruction followed immediately and was performed by the plastic surgeon. The mastectomy skin flaps were first inspected for hemostasis and viability. Subpectoral pockets were then created using electrocautery. In some cases, a total submuscular pocket was created by elevating a flap of serratus fascia to be sutured to the inferolateral border of the pectoralis major muscle. In other cases, acellular dermal matrix [AlloDerm (LifeCell Corp., Bridgewater, N.J.), Biodesign (Cook Medical, Bloomington, Ind.), or AlloMax (Bard, Covington, Ga.)] was used to support the lower pole. The choice was left to the discretion of the plastic surgeon and was made on a case-by-case basis depending on the quality of the patient’s muscle coverage.
Volume 134, Number 1 • Reduction of Tissue Expander Infections As soon as the tissue expander containers were opened, they were filled with triple-antibiotic solution (50,000 units of bacitracin, 1 g of cefazolin, and 80 mg of gentamicin in 500 ml of normal saline)15 until the tissue expander was completely immersed; then, the container was covered. In patients with a penicillin allergy, dilute povidoneiodine with 80 mg of gentamicin in 500 cc of saline was used. Before tissue expander insertion, the mastectomy pocket was irrigated thoroughly with the antibiotic solution. The skin surrounding each surgical site was then prepared again with povidone-iodine solution and draped with sterile towels. The povidone-iodine was allowed to dry. The plastic surgeon then donned a new pair of talc-free surgical gloves. The tissue expanders were handled only by the surgeon performing the reconstruction. Implants were inserted with as little skin contact as possible. The subpectoral pockets were closed immediately by suturing the inferior border of the pectoralis major muscle to serratus fascia or acellular dermal matrix. Surgical drains were always inserted, and the number of drains depended on the setting (one in delayed reconstruction, two in immediate reconstruction after total mastectomy, or three in immediate reconstruction after modified radical mastectomy). Incisions were closed in multiple layers. Administration of intravenous antibiotics was repeated if the procedure was longer than 4 hours. Depending on plastic surgeon discretion, initial expander fills were performed intraoperatively if tolerated by the mastectomy skin flaps. Postoperatively, intravenous antibiotics (firstgeneration cephalosporin or clindamycin) were continued for 24 hours; then, patients were discharged on an oral first-generation cephalosporin or clindamycin until removal of the final drain. Each drain was removed when its output was less than or equal to 30 ml/day. Expander fills were usually started 2 to 4 weeks postoperatively, and were performed using chlorhexidine preparation of the port site. Patients undergoing adjuvant chemotherapy underwent expansion throughout the duration of chemotherapy. Patients requiring adjuvant radiation therapy underwent rapid expansion followed by radiation therapy. Patients with signs of infection were started on broad-spectrum antibiotics. If patients displayed signs of systemic illness or if the erythema did not improve after 3 to 4 days, they underwent tissue expander explantation. Cultures were obtained intraoperatively to narrow antibiotic coverage. In cases where no purulence was seen intraoperatively, another tissue expander
was placed after washout. In cases where purulence was encountered, no tissue expander replacement was performed until at least 6 months postoperatively. Patients with multiple tissue expander infections were offered autologous reconstruction instead. Infection Tissue expander infection was defined as any instance where antibiotics were started, or where a tissue expander was removed because of erythema, warmth, drainage, fever, tenderness, and/or swelling.10 Cases where mastectomy skin flap necrosis occurred and resulted in expander extrusion without any signs of infection were not defined as infection.
RESULTS Patient Demographics Three hundred five patients underwent 456 tissue expander reconstructions between November of 2005 and November of 2010. One hundred ninety-eight patients underwent 313 tissue expander reconstructions between November of 2010 and November of 2012. All reconstructions after November of 2010 were performed according to the best-practices protocol, compared with none before November of 2010. Five breast reconstruction surgeons performed all reconstructions before and after the protocol. Demographic data of the two patient groups are listed in Table 1. Compared with the period before the protocol, there were significantly more immediate reconstructions, bilateral reconstructions, nonsmokers, patients with breast size larger than a C cup, and patients undergoing reconstruction with acellular dermal matrix in the period after the protocol. There were significantly fewer patients with a body mass index greater than 35 in the period after the protocol. Univariate Analysis In the period before the protocol, 56 of 305 patients (18.4 percent) developed an infection, compared with 23 of 198 patients (11.6 percent) after the protocol. The difference was statistically significant (p = 0.042) (Table 2). In the period before the protocol, 60 of 456 tissue expanders (13.2 percent) developed an infection, of which 42 (9.2 percent) had to be removed because of infection, and 18 (3.9 percent) were treated with antibiotics alone. This compared with 29 of 313 tissue expanders (9.3 percent; p = 0.097)
13
Plastic and Reconstructive Surgery • July 2014 Table 1. Baseline Patient Characteristics and Surgical Details No. of Expanders Before the After the Protocol (%)* Protocol (%)† Total no. of expanders Timing Immediate Delayed Laterality Unilateral Bilateral Chemotherapy None Preoperative Postoperative Any Radiation therapy None Preoperative Postoperative Any Tobacco None Former user Current user BMI 25 >30 >35 Prior breast surgery Yes No Breast size ≤C >C Coverage Total submuscular Acellular dermal matrix Mean age, yr
456
p
313
351 (77.1) 105 (22.9)
283 (90.4) 30 (9.6)
30) and preoperative breast size larger than a C cup. Both of these factors were found to be predictors of tissue expander infection in our prior study10 and most likely reflect longer operative times with more risk of contamination and longer mastectomy skin flaps that are at higher risk of necrosis. Unlike our previous study, we did not find radiation therapy to be a risk factor for tissue expander infection. We have become much more selective with irradiated patients. Those whose skin displays signs of radiation injury are now considered poor candidates for tissue expanders in our program and are usually offered autologous reconstruction. Among patients with mastectomy skin flap necrosis, there was no difference in the rate of tissue expander infection between those who had acellular dermal matrix (32.1 percent) and those who had total submuscular coverage (25 percent; p = 0.78). This is contrary to our belief that total submuscular coverage provides increased
Volume 134, Number 1 • Reduction of Tissue Expander Infections protection to the tissue expander compared with acellular dermal matrix. Common pathogens implicated in breast periprosthetic infections include Staphylococcus aureus, Staphylococcus epidermidis, and Pseudomonas aeruginosa.19 Courtiss et al. showed that S. epidermidis was the main organism that grew from cultures milked from the nipples before breast augmentation, suggesting that contamination of the operative field by colonized ductal fluid during surgery might be one mechanism for implant seeding, and arguing in favor of repeating preparation of the skin before the expander is inserted.20 Given the large predominance of Gram-positive cocci in periprosthetic infections, we chose to use first-generation cephalosporins as our preoperative antibiotics and as our antibiotics at discharge. In our study, we observed a shift from predominantly Gram-positive infections before the protocol, to an equal distribution between Gram-positive cocci and Gram-negative bacilli after the protocol. Most of the decrease in Gram-positive infections was attributable to a decrease in infections caused by Staphylococcus epidermidis. Interestingly, we have found an increase in the proportion of bilateral tissue expander infections, most of which were caused by Gram-negative bacilli. Two patients were found to have urinary tract infections with Pseudomonas aeruginosa at the time of their tissue expander infection. This suggests that seeding from a systemic source to tissue expanders is a possible mechanism of tissue expander infection in our patient population. Our study suffers from several limitations: it is retrospective and involves several different plastic surgeons. Although the best-practices protocol was reported as having been followed in all reconstructions performed after November of 2010, it is impossible to confirm that there were absolutely no deviations from protocol. We believe that many intraoperative contaminations probably go unnoticed and eventually lead to infection. A prospective, observational study in which any breach in sterile technique is recorded would elucidate when such contaminations occur. The simple act of introducing a new protocol may have spurred our surgeons to pay closer attention to sterility. Moreover, it is impossible to delineate exactly which steps of the protocol were most critical to its success. In addition, the baseline characteristics of our two populations are different. We did, however, control for all baseline differences using a multivariate logistic regression model, aimed at
eliminating the effect of potential confounders. A randomized controlled trial in which one group of patients is treated according to the protocol, and another group is not, would be the ideal study design to prove the efficacy of a protocol. Although we lowered our rate of tissue expander infection with our protocol, many of the infections that we are still observing are likely avoidable. In addition, we have observed a shift in our infectious pathogens, with the proportion of tissue expanders infected with Gram-positive cocci decreasing and the proportion with Gramnegative bacilli remaining the same. Further improvements to our protocol are likely needed to better address Gram-negative microorganisms. In particular, the possibility of seeding of the tissue expander from a systemic source needs to be investigated further.
CONCLUSIONS By implementing a simple, comprehensive best-practices protocol, we have been able to reduce our odds of tissue expander infections by 55 percent. This has significant implications for health care cost reduction, in addition to patient satisfaction, in the field of breast reconstruction. Our study has allowed us to identify several modifications that may be able to further improve our protocol.
cpt
CODING PERSPECTIVE
This information provided by Dr. Raymund Janevicius is intended to provide coding guidance.
19357 B reast reconstruction, imme-
diate or delayed, with tissue expander, including subsequent expansion 15777 Implantation of biologic implant (eg, acellular dermal matrix) • Placement of a tissue expander for breast reconstruction is reported with code 19357. • Code 19357 is used for both immediate and delayed breast reconstructions. • Creation of a submuscular pocket is included in the global 19357. This is not considered a “muscle flap,” and it is not appropriate to report code 15734 for creation of the pocket.
17
Plastic and Reconstructive Surgery • July 2014
• Use of acellular dermal matrix is reported with code 15777. • For bilateral procedures report:
19357 Breast reconstruction with tissue expander, right 19357-50 Breast reconstruction with tissue expander, left 15777 Implantation of biologic implant, right 15777-50 Implantation of biologic implant, left • Some payers prefer single line entries:
19357-50 Breast reconstruction with tissue expander, bilateral 15777-50 Implantation of biologic implant, bilataral
James H. Boehmler IV, M.D. Department of Plastic Surgery The Ohio State University Wexner Medical Center 915 Olentangy River Road, Suite 2100 Columbus, Ohio 43212 [email protected]
REFERENCES 1. American Society of Plastic Surgeons. 2012 Plastic Surgery Statistics Report. Available at: http://www.plasticsurgery.org/ Documents/news-resources/statistics/2012-Plastic-SurgeryStatistics/full-plastic-surgery-statistics-report.pdf. Accessed December 14, 2013. 2. Olsen MA, Lefta M, Dietz JR, et al. Rick factors for surgical site infection after major breast operation. J Am Coll Surg. 2008;207:326–335. 3. Olsen MA, Chu-Ongsakul S, Brandt KE, Dietz JR, Mayfield J, Fraser VJ. Hospital-associated costs due to surgical site infection after breast surgery. Arch Surg. 2008;143:53–60. 4. Bennett SP, Fitoussi AD, Berry MG, Couturaud B, Salmon RJ. Management of exposed, infected implant-based breast reconstructions and strategies for salvage. J Plast Reconstr Aesthet Surg. 2011;64:1270–1277.
18
5. Disa JJ, Ad-El DD, Cohen SM, Cordeiro PH,Hidalgo DA. The premature removal of tissue expanders in breast reconstruction. Plast Reconstr Surg. 1999;104:1662–1665. 6. Mitchem J, Herrmann D, Margenthaler JA, Aft RL. Impact of neoadjuvant chemotherapy on rate of tissue expander/implant loss and progression to successful breast reconstruction following mastectomy. Am J Surg. 2008;196:519–522. 7. Goodwin SJ, McCarthy CM, Pusic AL, et al. Complications in smokers after postmastectomy tissue expander/implant breast reconstruction. Ann Plast Surg. 2005;55:16–19; discussion 19–20. 8. Nahabedian MY, Tsangaris T, Momen B, Manson PN. Infectious complications following breast reconstruction with expanders and implants. Plast Reconstr Surg. 2003;112:467–476. 9. Krueger EA, Wilkins EG, Strawderman M, et al. Complications and patient satisfaction following expander/implant breast reconstruction with and without radiotherapy. Int J Radiat Oncol Biol Phys. 2001;49:713–721. 10. Francis SH, Ruberg RL, Stevenson KB, et al. Independent risk factors for infection in tissue expander breast reconstruction. Plast Reconstr Surg. 2009;124:1790–1796. 11. Alderman AK, Wilkins EG, Kim HM, Lowery JC. Complications in postmastectomy breast reconstruction: Two-year results of the Michigan Breast Reconstruction Outcome Study. Plast Reconstr Surg. 2002;109:2265–2274. 12. Armstrong RW, Berkowitz RL, Bolding F. Infection following breast reconstruction. Ann Plast Surg. 1989;23:284–288. 13. Craft RO, Damjanovic B, Colwell AS. Evidence-based protocol for infection control in immediate implant-based breast reconstruction. Ann Plast Surg. 2012;69:446–450. 14. Pfeiffer P, Jørgensen S, Kristiansen TB, Jørgensen A, Hölmich LR. Protective effect of topical antibiotics in breast augmentation. Plast Reconstr Surg. 2009;124:629–634. 15. Adams WP Jr, Rios JL, Smith SJ. Enhancing patient outcomes in aesthetic and reconstructive breast surgery using triple antibiotic breast irrigation: Six-year prospective clinical study. Plast Reconstr Surg. 2006;117:30–36. 16. Murray JD, Elwood ET, Jones GE, Barrick R, Feng J. Decreasing expander breast infection: A new drain care protocol. Can J Plast Surg. 2009;17:17–21. 17. Weber WP, Marti WR Zwahlen M, et al. The tim ing of surgical antimicrobial prophylaxis. Ann Surg. 2008;247:918–926. 18. Kapadia BH, McElroy MJ, Issa K, Johnson AJ,Bozic KJ, Mont MA. The economic impact of periprosthetic infections following total knee arthroplasty at a specialized tertiary-care center. J Arthroplasty 2014;29:929–932. 19. Pittet B, Montandon D, Pittet D. Infection in breast implants. Lancet Infect Dis. 2005;5:94–106. 20. Courtiss EH, Goldwyn RM, Anastasi GW. The fate of implants with infections around them. Plast Reconstr Surg. 1979;63:812–816.
Breast Reconstruction with Tissue Expanders: Implementation of a Standardized Best-Practices Protocol to Reduce Infection Rates Ibrahim Khansa, M.D. Russell G. Hendrick, Jr., M.D. Alison Shore, M.D. Joseph Meyerson, M.D. Maelee Yang, B.S. James H. Boehmler IV, M.D. Columbus, Ohio; Houston, Texas; and Chicago, Ill.
cpt
Background: Periprosthetic infection remains a frustrating and costly complication of breast reconstruction with tissue expanders. Although some specific steps have been previously shown to reduce periprosthetic infections, no comprehensive protocol addressing all aspects of preoperative, intraoperative, and postoperative patient management has been evaluated in the literature. The authors’ goal was to evaluate the effectiveness of their protocol at reducing periprosthetic infections. Methods: A comprehensive, best-practices protocol was introduced and implemented in November of 2010. All patients undergoing breast reconstruction using tissue expanders at the authors’ institution in the 5 years before the protocol, and in the 2 years after, were analyzed. Results: Three hundred five patients underwent 456 tissue expander reconstructions in the 5 years before the protocol, and 198 patients underwent 313 reconstructions in the 2 years after. Significantly fewer patients developed periprosthetic infection after protocol (11.6 percent versus 18.4 percent; p = 0.042), and the number of infected tissue expanders trended toward a decrease (9.3 percent versus 13.2 percent; p = 0.097). On multivariate analysis, the protocol significantly reduced the odds of periprosthetic infection (OR, 0.45; p = 0.022). Predictors of infection included obesity (OR, 2.01; p = 0.045) and preoperative breast size larger than C cup (OR, 2.83; p = 0.006). Conclusions: The authors’ comprehensive, best-practices protocol allowed them to reduce the odds of tissue expander infections by 55 percent (OR, 0.45; p = 0.022). The authors were able to identify several potential areas of improvement that may help them lower the rate of infection further in the future. (Plast. Reconstr. Surg. 134: 11, 2014.) CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, III.
T
issue expansion followed by implant reconstruction remains the most commonly performed postmastectomy breast reconstruction modality.1 Unfortunately, periprosthetic infections remain a common and debilitating complication, often requiring hospital admission for intravenous antibiotics, reoperation for expander removal, and additional procedures to reconstruct the breast.2 Apart from the emotional toll periprosthetic infections may exert on From the Department of Plastic Surgery, The Ohio State University Wexner Medical Center; the Division of Plastic Surgery, Baylor College of Medicine; and Lakeview Plastic Surgery. Received for publication November 30, 2013; accepted January 15, 2014. Copyright © 2014 by the American Society of Plastic Surgeons DOI: 10.1097/PRS.0000000000000261
the patient, they are also very costly complications from a financial perspective.3 The incidence of periprosthetic infections in the literature ranges from 1 to 35 percent.4 Several potential risk factors for periprosthetic infections have been reported in the literature, including chemotherapy,5,6 smoking,5,7 radiation therapy,8,9 lymph node dissection,8 large breast size,10 higher body mass index,11 immediate reconstruction,11,12 and bilateral reconstruction.12 To our knowledge, no comprehensive protocol that aims to reduce periprosthetic infections by addressing all aspects of preoperative, intraoperative, and postoperative patient care has been
Disclosure: The authors have no financial interest to declare in relation to the content of this article.
www.PRSJournal.com
11
Plastic and Reconstructive Surgery • July 2014 shown to reduce infection rates. Craft et al. have published a description of their algorithm, but its effectiveness at reducing periprosthetic infections was not evaluated.13 Several other authors have attempted to reduce infections by addressing one of the various aspects of preoperative, intraoperative, or postoperative patient care. Pfeiffer et al. demonstrated that pocket irrigation with a first-generation cephalosporin was superior to saline irrigation in breast augmentation.14 Adams et al. found low expander infection rates using triple-antibiotic irrigation solution for expander immersion and pocket irrigation.15 Murray et al. demonstrated a significant reduction in infections by applying mupirocin to drain sites, and by not allowing drains to contact the tissue expander.16 We have previously published our outcomes with 413 tissue expander reconstructions performed between 2001 and 2007.10 Our rate of periprosthetic infection and resulting expander loss (16.5 and 9.4 percent, respectively) prompted us to reexamine our practices critically. At the time, our institution did not have a standardized protocol to minimize periprosthetic infections. The choice of skin preparation varied by surgeon preference. Multiple individuals in the operating room were often allowed to handle the expander. The subpectoral pocket was not universally irrigated with triple-antibiotic solution, and the surgical team did not routinely repeat preparation of the skin or change gloves before expander insertion. The postoperative drain and antibiotic management varied widely by surgeon. In November of 2010, we introduced a comprehensive, standardized protocol that dictated precise measures to be taken in the preoperative, intraoperative, and postoperative settings. Our goal in this study was to evaluate the effects of this protocol on our periprosthetic infection rates, by comparing our results from 2005 to 2010 (before the protocol) to those from 2010 to 2012 (after the protocol).
PATIENTS AND METHODS After approval by the institutional review board, all patients who underwent postmastectomy breast reconstruction using a tissue expander at the James Cancer Hospital between November of 2005 and November of 2012 were analyzed. Patients were divided into two time periods: patients undergoing surgery between November of 2005 and November of 2010, who were not treated according to the best-practices protocol; and patients undergoing surgery between
12
November of 2010 and November of 2012, who were treated according to the protocol. Patient characteristics, history of irradiation and chemotherapy, operative technique, and postoperative management were evaluated. Univariate analysis to compare the effects of various risk factors and the protocol on infection rates was performed using chi-square analysis. Multivariate binary logistic regression was performed to eliminate the effect of confounders. A value of p < 0.05 was considered indicative of statistical significance. All statistical tests were performed using Minitab 16 statistical software (Minitab, Inc., State College, Pa.). Best-Practices Standardized Protocol In November of 2010, we introduced a comprehensive, standardized protocol to be applied to all patients having tissue expanders inserted into the breast. The protocol is divided into three parts: preoperative, intraoperative, and postoperative. Preoperatively, implant size and type were chosen by the plastic surgeon at the time of consultation based on patient base width and projection preference. Expander models used were Mentor (Mentor Corp., Santa Barbara, Calif.) and Allergan (Allergan, Inc., Santa Barbara, Calif.). Patients were asked to scrub their bodies with chlorhexidine the day before surgery and the morning of surgery. All patients received preoperative intravenous antibiotics (i.e., cefazolin, or clindamycin for penicillin-allergic patients), with the dose adjusted properly according to weight, and timed at least 30 minutes before incision.17 Intraoperatively, the skin preparation and draping were performed using chlorhexidine with the attending plastic surgeon present. In immediate reconstruction cases, the surgical oncologist performed total or modified radical mastectomies through either a traditional elliptical incision or a periareolar incision. Reconstruction followed immediately and was performed by the plastic surgeon. The mastectomy skin flaps were first inspected for hemostasis and viability. Subpectoral pockets were then created using electrocautery. In some cases, a total submuscular pocket was created by elevating a flap of serratus fascia to be sutured to the inferolateral border of the pectoralis major muscle. In other cases, acellular dermal matrix [AlloDerm (LifeCell Corp., Bridgewater, N.J.), Biodesign (Cook Medical, Bloomington, Ind.), or AlloMax (Bard, Covington, Ga.)] was used to support the lower pole. The choice was left to the discretion of the plastic surgeon and was made on a case-by-case basis depending on the quality of the patient’s muscle coverage.
Volume 134, Number 1 • Reduction of Tissue Expander Infections As soon as the tissue expander containers were opened, they were filled with triple-antibiotic solution (50,000 units of bacitracin, 1 g of cefazolin, and 80 mg of gentamicin in 500 ml of normal saline)15 until the tissue expander was completely immersed; then, the container was covered. In patients with a penicillin allergy, dilute povidoneiodine with 80 mg of gentamicin in 500 cc of saline was used. Before tissue expander insertion, the mastectomy pocket was irrigated thoroughly with the antibiotic solution. The skin surrounding each surgical site was then prepared again with povidone-iodine solution and draped with sterile towels. The povidone-iodine was allowed to dry. The plastic surgeon then donned a new pair of talc-free surgical gloves. The tissue expanders were handled only by the surgeon performing the reconstruction. Implants were inserted with as little skin contact as possible. The subpectoral pockets were closed immediately by suturing the inferior border of the pectoralis major muscle to serratus fascia or acellular dermal matrix. Surgical drains were always inserted, and the number of drains depended on the setting (one in delayed reconstruction, two in immediate reconstruction after total mastectomy, or three in immediate reconstruction after modified radical mastectomy). Incisions were closed in multiple layers. Administration of intravenous antibiotics was repeated if the procedure was longer than 4 hours. Depending on plastic surgeon discretion, initial expander fills were performed intraoperatively if tolerated by the mastectomy skin flaps. Postoperatively, intravenous antibiotics (firstgeneration cephalosporin or clindamycin) were continued for 24 hours; then, patients were discharged on an oral first-generation cephalosporin or clindamycin until removal of the final drain. Each drain was removed when its output was less than or equal to 30 ml/day. Expander fills were usually started 2 to 4 weeks postoperatively, and were performed using chlorhexidine preparation of the port site. Patients undergoing adjuvant chemotherapy underwent expansion throughout the duration of chemotherapy. Patients requiring adjuvant radiation therapy underwent rapid expansion followed by radiation therapy. Patients with signs of infection were started on broad-spectrum antibiotics. If patients displayed signs of systemic illness or if the erythema did not improve after 3 to 4 days, they underwent tissue expander explantation. Cultures were obtained intraoperatively to narrow antibiotic coverage. In cases where no purulence was seen intraoperatively, another tissue expander
was placed after washout. In cases where purulence was encountered, no tissue expander replacement was performed until at least 6 months postoperatively. Patients with multiple tissue expander infections were offered autologous reconstruction instead. Infection Tissue expander infection was defined as any instance where antibiotics were started, or where a tissue expander was removed because of erythema, warmth, drainage, fever, tenderness, and/or swelling.10 Cases where mastectomy skin flap necrosis occurred and resulted in expander extrusion without any signs of infection were not defined as infection.
RESULTS Patient Demographics Three hundred five patients underwent 456 tissue expander reconstructions between November of 2005 and November of 2010. One hundred ninety-eight patients underwent 313 tissue expander reconstructions between November of 2010 and November of 2012. All reconstructions after November of 2010 were performed according to the best-practices protocol, compared with none before November of 2010. Five breast reconstruction surgeons performed all reconstructions before and after the protocol. Demographic data of the two patient groups are listed in Table 1. Compared with the period before the protocol, there were significantly more immediate reconstructions, bilateral reconstructions, nonsmokers, patients with breast size larger than a C cup, and patients undergoing reconstruction with acellular dermal matrix in the period after the protocol. There were significantly fewer patients with a body mass index greater than 35 in the period after the protocol. Univariate Analysis In the period before the protocol, 56 of 305 patients (18.4 percent) developed an infection, compared with 23 of 198 patients (11.6 percent) after the protocol. The difference was statistically significant (p = 0.042) (Table 2). In the period before the protocol, 60 of 456 tissue expanders (13.2 percent) developed an infection, of which 42 (9.2 percent) had to be removed because of infection, and 18 (3.9 percent) were treated with antibiotics alone. This compared with 29 of 313 tissue expanders (9.3 percent; p = 0.097)
13
Plastic and Reconstructive Surgery • July 2014 Table 1. Baseline Patient Characteristics and Surgical Details No. of Expanders Before the After the Protocol (%)* Protocol (%)† Total no. of expanders Timing Immediate Delayed Laterality Unilateral Bilateral Chemotherapy None Preoperative Postoperative Any Radiation therapy None Preoperative Postoperative Any Tobacco None Former user Current user BMI 25 >30 >35 Prior breast surgery Yes No Breast size ≤C >C Coverage Total submuscular Acellular dermal matrix Mean age, yr
456
p
313
351 (77.1) 105 (22.9)
283 (90.4) 30 (9.6)
30) and preoperative breast size larger than a C cup. Both of these factors were found to be predictors of tissue expander infection in our prior study10 and most likely reflect longer operative times with more risk of contamination and longer mastectomy skin flaps that are at higher risk of necrosis. Unlike our previous study, we did not find radiation therapy to be a risk factor for tissue expander infection. We have become much more selective with irradiated patients. Those whose skin displays signs of radiation injury are now considered poor candidates for tissue expanders in our program and are usually offered autologous reconstruction. Among patients with mastectomy skin flap necrosis, there was no difference in the rate of tissue expander infection between those who had acellular dermal matrix (32.1 percent) and those who had total submuscular coverage (25 percent; p = 0.78). This is contrary to our belief that total submuscular coverage provides increased
Volume 134, Number 1 • Reduction of Tissue Expander Infections protection to the tissue expander compared with acellular dermal matrix. Common pathogens implicated in breast periprosthetic infections include Staphylococcus aureus, Staphylococcus epidermidis, and Pseudomonas aeruginosa.19 Courtiss et al. showed that S. epidermidis was the main organism that grew from cultures milked from the nipples before breast augmentation, suggesting that contamination of the operative field by colonized ductal fluid during surgery might be one mechanism for implant seeding, and arguing in favor of repeating preparation of the skin before the expander is inserted.20 Given the large predominance of Gram-positive cocci in periprosthetic infections, we chose to use first-generation cephalosporins as our preoperative antibiotics and as our antibiotics at discharge. In our study, we observed a shift from predominantly Gram-positive infections before the protocol, to an equal distribution between Gram-positive cocci and Gram-negative bacilli after the protocol. Most of the decrease in Gram-positive infections was attributable to a decrease in infections caused by Staphylococcus epidermidis. Interestingly, we have found an increase in the proportion of bilateral tissue expander infections, most of which were caused by Gram-negative bacilli. Two patients were found to have urinary tract infections with Pseudomonas aeruginosa at the time of their tissue expander infection. This suggests that seeding from a systemic source to tissue expanders is a possible mechanism of tissue expander infection in our patient population. Our study suffers from several limitations: it is retrospective and involves several different plastic surgeons. Although the best-practices protocol was reported as having been followed in all reconstructions performed after November of 2010, it is impossible to confirm that there were absolutely no deviations from protocol. We believe that many intraoperative contaminations probably go unnoticed and eventually lead to infection. A prospective, observational study in which any breach in sterile technique is recorded would elucidate when such contaminations occur. The simple act of introducing a new protocol may have spurred our surgeons to pay closer attention to sterility. Moreover, it is impossible to delineate exactly which steps of the protocol were most critical to its success. In addition, the baseline characteristics of our two populations are different. We did, however, control for all baseline differences using a multivariate logistic regression model, aimed at
eliminating the effect of potential confounders. A randomized controlled trial in which one group of patients is treated according to the protocol, and another group is not, would be the ideal study design to prove the efficacy of a protocol. Although we lowered our rate of tissue expander infection with our protocol, many of the infections that we are still observing are likely avoidable. In addition, we have observed a shift in our infectious pathogens, with the proportion of tissue expanders infected with Gram-positive cocci decreasing and the proportion with Gramnegative bacilli remaining the same. Further improvements to our protocol are likely needed to better address Gram-negative microorganisms. In particular, the possibility of seeding of the tissue expander from a systemic source needs to be investigated further.
CONCLUSIONS By implementing a simple, comprehensive best-practices protocol, we have been able to reduce our odds of tissue expander infections by 55 percent. This has significant implications for health care cost reduction, in addition to patient satisfaction, in the field of breast reconstruction. Our study has allowed us to identify several modifications that may be able to further improve our protocol.
cpt
CODING PERSPECTIVE
This information provided by Dr. Raymund Janevicius is intended to provide coding guidance.
19357 B reast reconstruction, imme-
diate or delayed, with tissue expander, including subsequent expansion 15777 Implantation of biologic implant (eg, acellular dermal matrix) • Placement of a tissue expander for breast reconstruction is reported with code 19357. • Code 19357 is used for both immediate and delayed breast reconstructions. • Creation of a submuscular pocket is included in the global 19357. This is not considered a “muscle flap,” and it is not appropriate to report code 15734 for creation of the pocket.
17
Plastic and Reconstructive Surgery • July 2014
• Use of acellular dermal matrix is reported with code 15777. • For bilateral procedures report:
19357 Breast reconstruction with tissue expander, right 19357-50 Breast reconstruction with tissue expander, left 15777 Implantation of biologic implant, right 15777-50 Implantation of biologic implant, left • Some payers prefer single line entries:
19357-50 Breast reconstruction with tissue expander, bilateral 15777-50 Implantation of biologic implant, bilataral
James H. Boehmler IV, M.D. Department of Plastic Surgery The Ohio State University Wexner Medical Center 915 Olentangy River Road, Suite 2100 Columbus, Ohio 43212 [email protected]
REFERENCES 1. American Society of Plastic Surgeons. 2012 Plastic Surgery Statistics Report. Available at: http://www.plasticsurgery.org/ Documents/news-resources/statistics/2012-Plastic-SurgeryStatistics/full-plastic-surgery-statistics-report.pdf. Accessed December 14, 2013. 2. Olsen MA, Lefta M, Dietz JR, et al. Rick factors for surgical site infection after major breast operation. J Am Coll Surg. 2008;207:326–335. 3. Olsen MA, Chu-Ongsakul S, Brandt KE, Dietz JR, Mayfield J, Fraser VJ. Hospital-associated costs due to surgical site infection after breast surgery. Arch Surg. 2008;143:53–60. 4. Bennett SP, Fitoussi AD, Berry MG, Couturaud B, Salmon RJ. Management of exposed, infected implant-based breast reconstructions and strategies for salvage. J Plast Reconstr Aesthet Surg. 2011;64:1270–1277.
18
5. Disa JJ, Ad-El DD, Cohen SM, Cordeiro PH,Hidalgo DA. The premature removal of tissue expanders in breast reconstruction. Plast Reconstr Surg. 1999;104:1662–1665. 6. Mitchem J, Herrmann D, Margenthaler JA, Aft RL. Impact of neoadjuvant chemotherapy on rate of tissue expander/implant loss and progression to successful breast reconstruction following mastectomy. Am J Surg. 2008;196:519–522. 7. Goodwin SJ, McCarthy CM, Pusic AL, et al. Complications in smokers after postmastectomy tissue expander/implant breast reconstruction. Ann Plast Surg. 2005;55:16–19; discussion 19–20. 8. Nahabedian MY, Tsangaris T, Momen B, Manson PN. Infectious complications following breast reconstruction with expanders and implants. Plast Reconstr Surg. 2003;112:467–476. 9. Krueger EA, Wilkins EG, Strawderman M, et al. Complications and patient satisfaction following expander/implant breast reconstruction with and without radiotherapy. Int J Radiat Oncol Biol Phys. 2001;49:713–721. 10. Francis SH, Ruberg RL, Stevenson KB, et al. Independent risk factors for infection in tissue expander breast reconstruction. Plast Reconstr Surg. 2009;124:1790–1796. 11. Alderman AK, Wilkins EG, Kim HM, Lowery JC. Complications in postmastectomy breast reconstruction: Two-year results of the Michigan Breast Reconstruction Outcome Study. Plast Reconstr Surg. 2002;109:2265–2274. 12. Armstrong RW, Berkowitz RL, Bolding F. Infection following breast reconstruction. Ann Plast Surg. 1989;23:284–288. 13. Craft RO, Damjanovic B, Colwell AS. Evidence-based protocol for infection control in immediate implant-based breast reconstruction. Ann Plast Surg. 2012;69:446–450. 14. Pfeiffer P, Jørgensen S, Kristiansen TB, Jørgensen A, Hölmich LR. Protective effect of topical antibiotics in breast augmentation. Plast Reconstr Surg. 2009;124:629–634. 15. Adams WP Jr, Rios JL, Smith SJ. Enhancing patient outcomes in aesthetic and reconstructive breast surgery using triple antibiotic breast irrigation: Six-year prospective clinical study. Plast Reconstr Surg. 2006;117:30–36. 16. Murray JD, Elwood ET, Jones GE, Barrick R, Feng J. Decreasing expander breast infection: A new drain care protocol. Can J Plast Surg. 2009;17:17–21. 17. Weber WP, Marti WR Zwahlen M, et al. The tim ing of surgical antimicrobial prophylaxis. Ann Surg. 2008;247:918–926. 18. Kapadia BH, McElroy MJ, Issa K, Johnson AJ,Bozic KJ, Mont MA. The economic impact of periprosthetic infections following total knee arthroplasty at a specialized tertiary-care center. J Arthroplasty 2014;29:929–932. 19. Pittet B, Montandon D, Pittet D. Infection in breast implants. Lancet Infect Dis. 2005;5:94–106. 20. Courtiss EH, Goldwyn RM, Anastasi GW. The fate of implants with infections around them. Plast Reconstr Surg. 1979;63:812–816.
