What Do I Do When I Drop the Graft?

Abstract & Commentary

Synopsis: Pulsatile lavage with three liters of 2% chlorhexidine solution followed by a saline rinse was found to effectively decontaminate bone-tendon allografts whereas other techniques did not.

Source: Burd T, et al. The effects of chlorhexidine irrigation solution on contaminated bone-tendon allografts. Am J Sports Med 2000;28:241-244.

As the use of autogenous and allogenic tissues in orthopaedic reconstruction procedures becomes more common, there needs to be an established method for disinfecting a graft that has been inadvertently contaminated. This paper sought to determine the most expedient and effective method for disinfecting human bone-tendon allografts.

Burd and associates from the University of Missouri thoroughly addressed this issue with a three-part experimental protocol. The first part involved both beef muscle and human cadaveric tensor fasciae latae tissue samples. Soft tissue samples were used because these are more subject to contamination and more difficult to sterilize. Part 1 was performed as a screening method to select the most promising method of sterilization, which was further tested in parts 2 and 3 of the study. The tissues were all contaminated with a carefully standardized inoculum consisting of four different bacteria: Staphylococcus aureus, Staphylococcus epidermidis, Pseudomonas aeruginosa, and Klebsiella pneumoniae. These bacteria were selected based on their high virulence as well as the fact that these are common bacteria for orthopaedic infections or are found on an operating room floor. To simulate the scenario of dropping a graft on the floor, the bacterial inoculum was allowed to incubate at room temperature for one minute, following which the tissues were swished in sterile saline for about two minutes until the power irrigation equipment was assembled.

At that point several solutions as well as methods of delivery were assessed. These included 0.05% Castile soap and/or 0.03% benzalkonium chloride, triple-antibiotic solution (gentamicin 0.1%, clindamycin 0.1%, and polymixin 0.05%), or 4% chlorhexidine gluconate, each irrigated with pulsatile lavage equipment with a three-liter or one-liter volume. Additionally, successive 30-minute soaks with the combination of 4% chlorhexidine and the triple-antibiotic bath were tested. Normal saline was used as a control.

Burd et al found that all solutions used with just one liter of volume had positive cultures. The only successful decontaminations were with the three liters of power irrigation with chlorhexidine gluconate or with the successive 30-minute soaks of chlorhexidine and the triple-antibiotic solution. Based upon these findings, Burd et al then assessed varying concentrations of chlorhexidine from 0.05% to 4% and found that the 2% solution was as effective as 4% and achieved 100% decontamination.

Lastly, fresh human Achilles bone-tendon allografts were inoculated and then tested following treatment with three liters of a 2% chlorhexidine power wash followed by a saline rinse. Ten out of 10 allografts were effectively decontaminated. All of the control specimens produced positive cultures for this as well as the other parts of the study, confirming the sensitivity of the method for detecting the presence of bacteria, using a combination of broth and plate culturing.

Comment by David R. Diduch, MS, MD

Inadvertently dropping a graft in the operating room is considered a disaster. The options at that point are limited: 1) harvest a new graft from the opposite knee or a different source, 2) use a new allograft, or 3) somehow disinfect the contaminated graft. Harvesting a new graft from the contralateral knee or a new location results in additional morbidity and may alter the perioperative rehabilitation. There may be medical-legal concerns regarding informed consent for taking an additional graft as well as changing to an allograft intraoperatively. Ideally, there would be a method to predictably decontaminate and still use the graft that had been harvested. This paper by Burd et al is a major step toward establishing such a method. Their experimental protocol was carefully planned and effectively addressed the question. They had adequate numbers of specimens involving both animal and human tissue, and methods to assess the presence of bacteria were appropriately sensitive. Sample numbers were sufficient to achieve statistical significance. This is a strong paper with excellent clinical implications.

In summary, Burd et al determined that chlorhexidine was more effective than the other solutions at decontaminating the grafts. Furthermore, they established that a power wash with three liters of chlorhexidine effectively decontaminated the graft. This can be done in 10 to 12 minutes. This is a vast improvement over the previously established method by Goebel et al, which prescribed successive 30-minute washes of chlorhexidine followed by a triple-antibiotic bath.1 Sixty minutes of operating room time, anesthesia time, and potential tourniquet time is dramatically different than a 10- to 12-minute power wash. There appeared to be no advantage to long-term soaks in this study over the power wash.

A similar article in the current issue of Arthroscopy also found 4% chlorhexidine gluconate to be superior to all other solutions tested.2 That study by Molina et al only soaked the graft for 90 seconds; however, contamination occurred by dropping the graft on the floor for 15 seconds as opposed to direct inoculation with bacterial cultures for one minute in Burd et al’s study. It would seem the direct inoculation provides a more difficult challenge to graft sterilization given that only 58% of control, untreated grafts cultured positive after dropping on the floor. Interestingly, in all of these studies, treatment with povidone-iodine solution, which is most commonly used for surgical preps, was ineffective at graft sterilization.

Burd et al stressed that we have not determined the biological effects of using chlorhexidine on human tissues and whether this will alter the mechanical properties of a graft. Because of this, they do not advocate using this technique yet in the clinical setting. They are planning further experiments to assess the mechanical effects on the graft as well as any cellular toxicity. They do recommend rinsing the graft with saline using the same power wash conditions to help remove residual chlorhexidine. Until these studies are available it would be a clinician’s preference as to whether to put this current information to use should the unanticipated disaster of a graft drop occur. Burd et al do mention that intra-articular lavage with chlorhexidine does produce untoward effects on the synovium and articular surfaces in horses, supporting the notion of thoroughly washing the graft after the chlorhexidine treatment.

The standards in place for disinfecting operating room floors generally yield a low bacterial count on a floor that should be below the minimum number of bacteria required to produce infection. However, the devascularized nature of an allograft or even removed autograft reduces the bacterial contamination level required to produce an infection to four times less. As few as 100 colony-forming units (CFU) of S. aureus on allografts have resulted in an infection rate of 50% of allografts implanted.3

As such, an effective method to decontaminate a graft is essential before we consider replacing it in the knee. Burd et al are to be congratulated on their research protocol, and further studies will help us with decision-making in this disturbing scenario which we all hope never to encounter.


1. Goebel ME, et al. Contaminated rabbit patellar tendon grafts. Am J Sports Med 1994;22:387-390.

2. Molina ME, et al. Contaminated anterior cruciate ligament grafts: The efficacy of 3 sterilization agents. Arthroscopy 2000;16:373-378.

3. Cordero J, et al. Influences of bacterial strains on bone infection. J Orthop Res 1996;14:663-667.

The most effective and expeditious way to decontaminate a graft that has been dropped is:

a. use of benzalkonium chloride and Castile soap.

b. 1 L of 4% chlorhexidine wash.

c. 30-minute triple-antibiotic bath soak.

d. 3 L of 2% chlorhexidine power wash.