In November and December of 2004, hydraulic fluid was accidentally substituted for detergent in one step of a multi-step cleaning and sterilization process of surgical tools at Durham Regional Hospital (DRH) and Duke Health Raleigh Hospital (DHRH), both of which are part of the Duke University Health System (DUHS).

Chain of events leading to the mix-up

Employees of Automatic Elevator, a firm that serviced the hospitals’ elevators, drained the fluid from elevators into containers that had held surgical detergent. The containers were not properly re-labeled or securely stored by Automatic Elevator. Cardinal Health, the firm that provides surgical detergent to DUHS, picked the containers up, restocked and stored them, and then shipped them as detergent back to several hospitals. The fluid was then used in the cleaning process for surgical instruments at DRH and DHRH. None of the fluid was used at Duke University Hospital. The mix-up was not immediately noticed, in part because the detergent our hospitals use and the hydraulic fluid have a similar color and consistency (see picture). While the hydraulic fluid occasionally left the instruments feeling slightly oily, it is not uncommon for instruments to feel slick, as a lubricant is added during the normal cleaning process to prevent rust and ensure the instruments work smoothly during surgery.

Duke’s initial response

Replacing fluid and recleaning instruments:

Once the problem was identified, our hospitals immediately replaced the solution and re-cleaned and sterilized all surgical instruments that had been washed in the hydraulic fluid (reserving a randomly selected group of instruments for analysis).

Disclosures:

Duke also notified Cardinal Health of the mix-up, sent a letter to each of the approximately 3,650 patients who may have had surgery or minor bedside procedures using the instruments in question, and alerted the local news media. The letter included the phone number of a physician they could call with any questions.

External Reviews

Teams of surveyors from the North Carolina Division of Facility Services (DFS), on behalf of the Centers for Medicare and Medicaid Services (CMS), visited both DRH and DHRH in January. The DFS visitors accepted our actions to date on the hydraulic fluid issue as being sufficient, and the hospitals immediately responded with corrective actions to other issues raised by DFS. As a result, CMS has certified that both DRH and DHRH are in full compliance with CMS standards.

Prevention of similar accidents in the future:

To make sure a similar mix-up could not occur again, we put into place a process that ensures that any product used in the cleaning and sterilization of instruments is verified as being correct before the product is distributed throughout a DUHS hospital. We have established systems to track and monitor containers while they are in our possession. Our health system employees have completed an online safety training session on container management. We have also worked with our suppliers to develop processes to alert users when the integrity of a product has been compromised (e.g. once a seal has been broken).

Analysis request:

As usual, Duke infection control experts monitored the infection rates at the hospitals--before, during the period when the fluid was used, and since then. They have found no notable increase in infection rates, or concern that the sterilization was not fully effective.

Duke also requested two independent, expert analyses of the instruments that had been exposed to the hydraulic fluid to determine whether the instruments were sterile and the nature of patients’ exposure, if any, to potentially harmful substances that might have remained on the instruments once they had been through the rinsing and sterilization process.

Results of the sterilization report

The report by Dr. Rutala of the Statewide Program in Infection Control and Epidemiology at UNC-Chapel Hill, which Duke received on June 15, 2005, concluded that the substitution of hydraulic fluid for detergent during the initial step of a multi-stage, intensive cleaning process did not compromise the sterilization process.

Results of the chemical analysis report

We also asked a second group of outside scientists at RTI International (RTI) to determine how much residue would have remained on surgical instruments after they had been rinsed and sterilized. Their report, which we received on June 24, concluded that the residual amount of material was extremely small, 0.08 milligrams on average. Of 11 metals that might have been introduced into the hydraulic fluid by the elevator system, most could either not be detected at all, or were barely detectable, even with sophisticated testing equipment. The one exception was zinc, an element of the original fluid, which was found at a level of 0.0004 milligrams per milligram of fluid.

Duke review

Woodhall Stopford, MD, a Duke toxicologist, reviewed RTI’s analysis of the fluid, along with chemical safety information provided by the manufacturer of the hydraulic fluid. He performed an initial risk assessment and concluded that none of the chemicals was likely to be harmful in the extremely small amounts to which patients were exposed.

Further patient communications

On June 20, 2005, Duke again sent letters to all affected patients sharing with them the results of the sterilization report. The letters also informed patients of the opportunity to receive a medical consultation at no cost from Duke environmental medicine experts.

On June 27, 2005, Duke sent letters to patients informing them of the results of the RTI report, reiterating the availability of the medical consultations, and informing patients of a hotline number they could call with any questions regarding the incident.

On August 3, 2005, Duke sent a follow-up letter to patients communicating the Health System’s ongoing commitment to provide information and resources regarding issues related to the hydraulic fluid incident. The letter included answers to frequently asked questions, and informed patients of how they could access future updates.

Long-term monitoring and next steps

Duke also tracked any unusual symptoms experienced by affected patients via reports from patients, their personal physicians, and Duke environmental medicine consultants (who will not release patient names or other identifiers). The long-term monitoring program was managed by PharmaLinkFHI.