Management of Contaminated Autologous Grafts in Plastic Surgery

METHODS

We developed a questionnaire for which a Web link was sent via e-mail by the Aesthetic Surgery Education and Research Foundation to members of the American Society of Aesthetic Plastic Surgery (ASAPS). Members were asked to participate in the survey and assured that their responses would be anonymous. The Web link connected surgeons to an 11-question online survey, shown in Figure 1, that was hosted by Survey Monkey (www.surveymonkey.com), a data collection and analysis company. The questionnaire was designed to obtain information on autograft contamination frequency, treatment preferences, clinical outcomes, and patient disclosure.

In conjunction with the survey, PubMed was searched for published literature related to contamination of fat, skin, cartilage, composite tissue, and bone grafts, as well as proposed management strategies and appropriate decontamination techniques.

RESULTS

Although 1900 ASAPS members were invited to participate in this anonymous online survey, only 223 surgeons (12%) submitted the questionnaire. The survey began by asking how long surgeons have been in practice. Of the 223 responders, 87% have been practicing for 10 or more years, with 65% in practice for more than 16 years.

Surgeons were then asked whether they had witnessed or experienced a graft contamination. The 30% of respondents who answered “no” were directed to stop and submit the survey. The 70% who said they had experienced a graft contamination represents at least 157 surgeons; however, there was a 3-surgeon inconsistency, with 160 answering questions that specifically asked about contamination experience. It seems that 3 surgeons who answered “no” on question 2 should have checked “yes” because they did, in fact, have graft contamination experience.

Reasons for the low (12%) response rate are unknown. Surgeons may not have had time to complete it or were not interested in the topic. Another possibility is that surgeons did not want to admit (even anonymously) that they have had a graft contamination incident. The fact that 87% of respondents have been in practice for at least 10 years may reflect a greater willingness to admit unintentional errors by more experienced surgeons. Because of the low response rate, the survey cannot be viewed as representative of all ASAPS members. Nevertheless, the results offer some interesting information about an issue not previously reported in the literature and rarely discussed openly by colleagues.

Among survey respondents, 70.4% (n = 157) reported either witnessing or experiencing graft contamination during a plastic surgery procedure. As shown in Table 1, 33% of surgeons reported 2 occurrences, and almost 26% experienced 4 or more contaminated graft incidents. The reporting of 2 instances by 52 surgeons represents 104 contaminations. If this mathematical exercise is continued and the “more than 5” response is assumed to be 6, then the total number of contaminated graft experiences is at least 426 among the 156 surgeons who answered this question.

Figure 1. Survey sent to plastic surgeons on management of contaminated grafts.

Multiple responses were allowed for many survey questions to capture a fuller range of experience for each surgeon. The most frequent method of contamination, reported by 75% of respondents, was a graft or flap falling on the floor (Table 2). A distant second, at 44%, was exposure to a nonsterile part of the field/drape. All other reasons for contamination shown in Table 2 are equally accidental.

The predominant anatomical site undergoing surgery at the time of autograft contamination was the craniofacial region, at 66% (Table 3). This may be because craniofacial grafts tend to be very small and therefore may be more easily knocked off a table or thrown in the trash. Contamination also occurs in autografts of the lower extremity, breast, trunk, upper extremity, and genitourinary areas.

Table 4 shows the types of contaminated autografts. Skin grafts received the most responses (61%), followed by cartilage grafts at 39%. Nipple-areolar complex grafts and bone grafts were the only other types reported by more than 20% of respondents.

No surgeon abandoned a procedure after graft contamination. Instead, 94% decontaminated the graft and proceeded with their surgical plan (Table 5). Only 7% harvested another graft, 4% used another reconstructive technique, and 2% substituted alloplastic material. The methods of decontamination are shown in Table 6, with the favored techniques being the use of povidone-iodine (polyvinyl pyrrolidone or PVP-I) (54%) or antibiotic solution (50%). Responses to the “other” category for this question included the use of systemic antibiotics or washing the graft with other agents.

More than 98% of respondents who had contaminated graft experience said that its use did not, to their knowledge, lead to an infection. Three decontaminated grafts (1.9%) were associated with clinical infection, but this infection rate is comparable to that reported for clean noncontaminated surgical cases.¹ One surgeon did not specify the location, 1 reported a systemic infection, and the third, a surgical site infection. Two of these surgeons had used PVP-I and 1 used bulb saline with antibiotic solution for decontamination.

Sixty percent of responding surgeons did not disclose the contamination incident to patients. Other actions taken are shown in Table 7, but only 21% said they had informed the patient/family postoperatively. The survey choice of “chart notation/incident report made” (selected by 25% of respondents) was poorly phrased because a chart notation is quite different from submitting an incident report to the surgical facility. One respondent noted that 1 incident was not disclosed because the possibility had been discussed in the informed consent process, but in another case the patient was told because the consent did not mention this possibility.

DISCUSSION

Inadvertent contamination of autografts presents a dilemma for surgeons, yet the plastic surgery literature lacks specific discussions of clinical experience with the management or outcome of contaminated grafts. Although our search for relevant articles focused on autografts, bone and tendon allografts have been studied more than other types, with much of the literature appearing in orthopedic journals. Most of articles described either in vitro experiments or in vivo animal studies of sterilization agents used after intentional microbial inoculation. Extrapolating data from such experiments to clinical practice is difficult.

In our survey, the most common source of contamination resulted from a graft falling on the floor. Several investigations have cultured grafts intentionally dropped and left on an operating room (OR) floor for as little as 15 seconds. Although one study found no positive results of cultures on contaminated samples,² others reported that between 58% and 96% of dropped grafts became contaminated.^3–5 As proof of the risk to which patients may be exposed, 90% of rabbits receiving uncleansed contaminated grafts developed infections.⁶

The literature agrees that a dropped graft can be safely used if sterilized before placement. Among survey respondents, more than 94% did decontaminate and use the graft. Harvesting another graft is often not a practical option and/or can cause additional morbidity. Substitute alloplastic material may not be available, and obtaining an allograft can take days; furthermore, either of these approaches would require patient consent. Thus, abandoning a procedure with the idea of rescheduling later is impractical and probably not in the best interests of patients.

RECOMMENDATIONS

The literature contains many conflicting results about which antimicrobial solution is most effective, which concentration or volume of a solution is better, or how long grafts should be exposed to a solution. One point is clear that an autograft can be contaminated in the average OR within seconds. We must therefore assume that a clinical infection may result from graft contamination, and 3 survey respondents did report infections following contamination incidents. It is therefore incumbent upon the surgeon to deal with such incidents in an appropriate manner.

Prevention should be the first priority. The algorithm shown in Figure 2 outlines perioperative steps for reducing the chance of graft contamination. The process analysis should include OR staff members to determine how graft contamination has occurred or might occur. All OR staff members should be explicitly alerted at the beginning of a case, during the mandatory “time-out” period, that an autologous graft will be used, and they should be aware of a graft’s location at all times. To minimize the chances of contamination, place the graft in a labeled, sterile container with a closed lid, and place it on the widest table available and away from the table edge or instruments. In addition, limit handling and exchanges of a graft between OR personnel to reduce the chance of dropping or exposure to a nonsterile area. Include the graft in the surgical count during the checkout process when a surgical technician is relieved, and have the surgical assistant confirm with the surgeon before discarding any tissue.

The written and verbal informed consent process for a procedure that will employ a graft seeks permission to harvest. This provides an ideal opportunity to convey the potential for graft contamination so that patients will be aware, in advance, that accidents can happen.

Figure 2. Perioperative autograft harvest algorithm.

The Figure 3 algorithm presents the steps to take if an autograft does become contaminated. Before decontamination, culture the graft to find out what organisms might have contaminated it. Recently, experts have recommended culturing all grafts, including allografts, before implantation so that appropriate antibiotics can be prescribed should infection develop.²² Include Clostridium in the culture because it has been implicated in deaths of patients who received tissue bank allografts.²³ In addition, the Centers for Disease Control and Prevention recommends all allografts be sterilized just before implantation.²⁴

We located no studies on the decontamination of fat grafts, but the limited added morbidity from harvesting more fat precludes the need for decontamination. Skin, cartilage, and composite tissue should be sterilized with at least 1 L of 4% CHG, preferably with low-pressure pulse lavage. Several studies recommend 3 L of decontamination solution, but this large volume does not seem necessary for the very small grafts typically used in plastic surgery. Furthermore, the addition of low-pressure lavage should reduce the need for larger volumes and for prolonged soaking times. We recommend reharvesting cancellous bone, but cortico-cancellous bone may be decontaminated with low-pressure lavage and 1 L of triple antibiotic solution, the only option not found to be cytotoxic to bone. Regardless of the decontamination agent used, rinse the graft with normal saline as the final step.

Figure 3. Intraoperative algorithm for contaminated autograft.

Figure 4. Postoperative algorithm for contaminated graft.

The Figure 4 algorithm outlines a process for postoperative management. We advocate full disclosure to the patient. If the informed consent process is handled properly, patients are aware of and prepared for the possibility of contamination. Failure to disclose the incident can increase the likelihood of a malpractice suit.^25,26 A focus group of patients identified what they want to hear following a medical error: full disclosure; clinical impact; causative factors; methods of future error prevention; and emotional support, including an apology.²⁷

Many patients cite their desire to prevent similar adverse events as justification for a malpractice claim,^28,29 yet a discussion of ways to prevent future errors is not a routine part of disclosure.³⁰ Therefore, root cause analysis along with an incident report should be part of postoperative process for preventing future incidents. Furthermore, engaging in the quality assurance review process is protected, privileged communication and non-discoverable in a legal proceeding as a way to encourage well-intentioned healthcare providers to report adverse events for quality improvement purposes.

CONCLUSION

Typically, incidents of graft contamination are handled according to a surgeon’s personal preference or a particular institution’s policies. Because our survey results suggest that autologous graft contamination may not be as rare as we’d like to believe, this article proposes the graft decontamination procedures outlined in the presented algorithms. Perhaps, the most important point to emerge from our literature review is that PVP-I is not recommended for graft decontamination by many investigators. Disclosure, reporting, and privileged quality improvement should not be overlooked as part of the prevention process.