Commentary on: Post-Mastectomy Surgical Pocket Irrigation With Triple Antibiotic Solution vs Chlorhexidine Gluconate: A Randomized Controlled Trial Assessing Surgical Site Infections in Immediate Tissue Expander Breast Reconstruction

Even in the cacophony of opinions, viewpoints, and personal preferences that makes up the specialty of plastic surgery, almost everyone agrees: implant infections are bad. Infectious complications following implant-based breast reconstruction increase the risk of reconstructive failure,¹ decrease patient satisfaction,² and are an expensive drain on limited healthcare resources.³ In contrast, there is little agreement about the best way to prevent infections. A survey of infection prevention practice patterns published by Gowda et al in 2017 showed the utter lack of consensus among plastic surgeons regarding measures such as preoperative skin decolonization, implant handling, and postoperative antibiotics.⁴

Similar heterogeneity exists regarding use of antibiotic irrigation solution. A 2019 practice survey conducted by Epps et al of 407 members of the American Society of Plastic Surgeons identified over 30 distinct antibiotic irrigation solutions in use by respondents.⁵ Although the most commonly used irrigant was a triple antibiotic solution (TAS), variations of single-antibiotic, betadine, and chlorhexidine (CHG)-based irrigants were also used. The diversity of personal preference on display in this survey reflects the lack of good data to guide clinical decision-making. In this study,⁶ Dr Nguyen and colleagues attempt to chip away at our collective ignorance by answering a simple question: which works better to reduce implant infections—TAS or a CHG-based irrigant?

The authors are to be commended for their excellent efforts to this end. In this blinded, randomized controlled trial, 88 patients undergoing immediate bilateral breast reconstruction with placement of tissue expanders underwent intraoperative pocket irrigation with TAS on one side and Irrisept CHG irrigant on the other. The trial was stopped early due to unavailability of bacitracin as a TAS component. Overall, the infection rates were 4.5% (5 infections) in the TAS group and 8.0% (7 infections) in the CHG group. There were 2 major infections in the TAS group and 6 in the CHG group. The authors conclude that although no statistically significant difference between the 2 solutions was found, the trend towards a lower major infection rate in the TAS group merits consideration of TAS as the better of the 2 options.

This is a clear, focused, and succinct study with one major advantage over any previously published data on this topic, namely, the prospective, randomized, blinded study design. By having each patient serve as their own control, the authors have neutralized the adverse effects of systemic patient factors such as smoking, diabetes, and increased BMI on infection rate. Importantly, the internal control design also allowed for elimination of breast surgeon as a potential confounder; the quality of mastectomy skin flaps is one of the biggest contributors to postoperative complications including infection. In this study, there was no significant difference in rates of mastectomy skin flap necrosis between the 2 treatment groups. No previous study has attempted to take on the irrigation question with such a rigorous methodology.

Despite its excellent design, this study has several shortcomings. Most importantly, 88 falls far short of the number of patients that would be needed to detect a true difference between the treatment groups. By the authors’ own power analysis, 359 patients would need to have been included to disprove the null hypothesis; however, through no fault of their own, the trial was halted early due to lack of availability of bacitracin, a key component of TAS. With 88 patients, the difference in infection rates between the TAS and CHG groups did not achieve statistical significance; unfortunately, the study also lacked sufficient power to definitively conclude that the 2 solutions are equivalent. The authors’ assertion that the trend toward fewer major infections in the TAS group merits consideration of a change in practice unfortunately does not stand up to statistical scrutiny.

The technique with which the irrigation solutions were used in this study also merits further discussion. As a necessary baseline condition, if we claim that an antibiotic solution works to prevent implant infection by inhibiting or killing causative bacteria, it must have bacteristatic or bactericidal activity against those organisms at the concentrations and with the dwell times used in practice. Epps et al included an in vitro analysis of TAS efficacy compared to sterile saline against several common implant infection pathogens including Staphylococcusepidermidis, Enterococcus faecalis, Escherichia coli, and Pseudomonasaeruginosa.⁵ These authors found that at 1-minute dwell time, TAS was no more effective than sterile saline against any of the organisms analyzed. Similarly, kinetic time-kill data available on the Irrisept CHG irrigant website,⁷ and supported by published preclinical data,⁸ report a full log difference in colony reduction when comparing 1- and 5-minute dwell times against some Enterococcus and methicillin-resistant Staphylococcus aureus species. In the current study, the authors state that they employed a 1-minute dwell time for each solution; presumably the solution was then evacuated from each pocket. Although a small amount of each irrigant likely remained in contact with the tissues even after evacuation, the 1-minute dwell time described in this study was significantly less than the time described to achieve the maximum effect for both solutions. If the in vitro efficacy data for these solutions are accurate, then one must consider whether antibiotic irrigation performed under these conditions has any significant bacterial inhibitory effect, or whether the observed reduction in infection is rather simply due to “dilution of the pollution.”

I freely admit my own bias—in clinical practice, both I and my partners use the CHG-containing Irrisept as our irrigation solution of choice. Perceived advantages include convenience (Irrisept comes prebottled, sterilely packaged, prediluted, and ready for use), broad spectrum of activity and effectiveness against common pathogens (Gram positive, Gram negative, and fungi), and very low rates of allergy/hypersensitivity to the active ingredients. The mastectomy cavity is thoroughly irrigated, and implants and tissue expanders are soaked in Irrisept for 5 minutes prior to placement. To increase the dwell time of the solution in the pocket, additional irrigant is added to the pocket during closure and the drain is clamped; the drain bulb is connected only after closure is complete. Total dwell time therefore ranges from 10 to 20 minutes. In addition to preclinical data showing efficacy of CHG irrigation solution against common bacteria⁸ and effectiveness against biofilms,⁹ there are some clinical data comparing TAS with CHG following breast reconstruction with implants. In a retrospective review of 690 consecutive patients undergoing implant-based breast reconstruction, Merceron et al compared the complication rates of patients who had pocket irrigation with TAS vs those who had pocket irrigation with CHG solution.¹⁰ The rate of overall complications, including major and minor infections, was significantly lower in the CHG group (6.4% vs 12.7%); however, there were some important differences between the study groups, including age, type of mastectomy, and exposure to adjuvant chemotherapy and radiation. Until the publication of Nguyen et al’s present article,⁶ the paper by Merceron et al represented one of only a few attempts at a head-to-head comparison of 2 commonly used irrigation solutions, although the design of this study left much to be desired.

One of the most interesting aspects of Nguyen et al’s study is the way in which it represents a microcosm of breast reconstruction practice and its tendency towards individual eccentricities. To reduce infection, many plastic surgeons follow a series of prescribed steps before, during, and after implant cases that they adhere to with a reverence and attention to detail usually reserved for religious rituals. One recent example is the canonical “Fourteen-Point Plan” published by Adams et al; ¹¹ however, plastic surgeons in all different practice settings have adopted and adapted various interventions into their own custom protocols. The reason for the popularity of this “scattershot” approach to infection prevention is simple: it works. Multiple studies have demonstrated the effectiveness of a standardized, multifaceted infection prevention protocol in implant-based breast reconstruction.^12-14 Similar to “care bundles” used to tackle difficult problems such as ventilator-associated pneumonia, these protocols combine multiple interventions based on the best available evidence.¹⁵ In this paper, Nguyen et al, in addition to pocket irrigation, discuss their use of preoperative antibiotics, re-prepping and draping of the surgical field after mastectomy, and use of outpatient antibiotics for 2 weeks following reconstruction.⁶ The overall low infection rate described in this study is likely due to the synergistic effects of these multiple interventions rather than the strength of a single component.

To the question of which irrigation solution is best, does this study move the ball forward? I think the answer is no … but also yes. Although Nguyen et al’s sample size was insufficient for the reader to draw definitive conclusions, the rigorous methodology of this prospective, randomized trial earns these results the title of “best available evidence.” Whether the findings warrant a widespread change in practice remains, unsurprisingly, a topic for debate. Nevertheless, I commend the authors for doing difficult work that moves the needle away from Quirky Personal Preference and towards Best Practice.

Disclosures

The author declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The author received no financial support for the research, authorship, and publication of this article.

REFERENCES