January 4, 2013 — Two community hospitals in North Carolina in 2004 unwittingly used a oil-based hydraulic fluid instead of a detergent to “clean” surgical instruments prior to terminal sterilization and their reuse.

These improperly cleaned instruments — slick with hydraulic fluid after sterilization and slippery when handled by surgeons during operations — were used to perform surgery on 3,650 patients between November and December, 2004.

The oily hydraulic fluid was previously used by the hospital to operate its elevators.

In response to this mishap, the federal Centers for Medicare and Medicaid Services (CMS) issued a report in 2005 concluding that these two hospitals had placed patients in “immediate jeopardy” for failing to detect the error, despite complaints from medical staff about slick tools, the Associated Press reported at the time.

In addition, a legal complaint was filed, concluding that the failure by these two hospitals to have cleaned surgical instruments properly using a detergent posed an increased risk of serious patient infections.

A 2005 study

In 2005, Duke University Health System — the health system that owned these two North Carolina hospitals — hired a team of microbiologists from the University of North Carolina to perform a series of laboratory tests to:

  • perform tests that would recreate this infection control mishap in the laboratory setting;
  • evaluate whether the oily hydraulic fluid — coated onto the surfaces of surgical instruments also contaminated with resistant bacteria — prevents terminal sterilization; and
  • assess the risk of disease transmission associated with this mishap.

Known herein as the “Rutala et al. (2005) study,” this team published a letter in 2005 concluding not only that the hydraulic fluid coated on these surgical instruments does not adversely impact the terminal sterilization process, but also that their study provides certain assurance that no patients would become infected as a result of Duke’s error.

The two types of terminal sterilization processes that the Rutala et al. (2005) study evaluated for effectiveness were the same as those used by these two community hospitals in North Carolina in 2004, at the time of the error: (a) steam sterilization; and (b) ethylene oxide (EtO) sterilization (using 100% EtO gas).

Simulated vs. clinical studies

This team’s 2005 microbiological study is called a “simulated in-use” study, because its objective is to recreate Duke’s error.

Whereas this type of study is controlled and performed in a laboratory setting, a “clinical in-use” study, in contrast, is performed in the hospital using instruments contaminated during surgery.

A review

I evaluated the Rutala et al. (2005) study, focusing on its conclusions, scope and methodology. For my evaluation, I reviewed:

1.  most of the reports, studies, letters to patients, correspondence, and newspaper articles that discuss Duke’s mishap;

2.  published reports and investigations of infections associated with the improper cleaning, disinfection, and/or sterilization of surgical instruments;

3.  published studies that evaluate the effectiveness of cleaning, disinfection, and sterilization processes under different conditions; and

4.  cleaning and sterilization standards and guidelines, including guidance documents published by the Food and Drug Administration (FDA).

I performed my review because the Rutala et al. (2005) study provided the primary — if not sole — set of data that officials of the Duke University Health System used to announce to the public that this mishap in 2004 did not pose a risk of infection for any of the more than three thousand potentially affected patients.

My review was limited to the infection risk associated with Duke’s cleaning mishap, however; it did not evaluate the potential for injury due to the potentially toxic exposure of patients to hydraulic fluid during surgery.

Safety, Healthcare Services.
Safety, Healthcare Services.


My review, which is the first and only published analysis to evaluate the Rutala et al.’s (2005) study:

  • identified potentially significant limitations in this study’s design and methodology;
  • determined that some of the protocols, procedures, and methods that this study used lacked some important detail and specificity; and
  • found difficulty interpreting and rationalizing some of this study’s findings and conclusions.

Questions raised

My review raises a number of questions about the Rutala et al. (2005) study, including whether:

— its methodology was sufficiently robust to yield meaningful data that are relevant and applicable to Duke’s 2004 mishap;

— its conclusion that hydraulic fluid used in lieu of detergent to “clean” surgical instruments did not adversely impact the terminal sterilization processes, which used either pressurized steam or ethylene oxide gas, is entirely sound; and

— its assurance that the inadvertent use of hydraulic fluid to clean surgical instruments would not result in any patients becoming infected with a pathogen (including with infectious prions) is reasonable and evidence-based.

Insufficiently challenging instruments

My review identified three salient limitations with the Rutala et al. (2005) study.

First, the types of surgical instruments that the Rutala et al. (2005) study used as surrogates during its tests — namely, stainless steel scalpel handles and plastic syringe barrels — are not sufficiently complex in physical design to be apt or to yield results applicable to Duke’s mishap.

Indeed, the FDA requires that evaluations of the effectiveness of sterilization processes for use in the clinical setting, like the Rutala et al. (2005) study, be performed using surgical instruments whose designs feature internal surfaces and are significantly more physically complex and challenging to sterilize.

Stainless steel scalpel handles and plastic syringe barrels do not feature any internal surfaces that are challenging to sterilize, as required by the FDA to evaluate sterilization effectiveness.

Examples of complex surgical instruments that the Rutala et al. (2005) study might have used instead include:

  • laparoscopic forceps with long shafts and serrated and hinged jaw assemblies;
  • reusable biopsy forceps with tightly wound, spring-like metal coils surrounding a long and inaccessible cable;
  • and haemostatic clamps with locking and overlapping, mated blades; implantable orthopedic screws with threaded crevices.

In short, while this 2005 study may be used to provide some information about sterilization effectiveness, its results apply only to a subset (and limited number) of surgical instruments that are comparable, or less complex, in design than stainless steel scalpel handles and plastic syringe barrels.

Confidential Quality and Safety Healthcare Services, Legal Reviews for Hospitals, Manufacturers and the Public:  Click here to read about Dr. Muscarella’s quality and safety services committed to educating and helping hospitals (and manufacturers and patients) reduce the risk of healthcare-associated infections, including those linked to contaminated duodenoscopes.

Clinical tests not performed

Second, the FDA requires that the effectiveness of a sterilization process intended for use in hospitals be evaluated using not only simulated in-use tests — Rutala et al.’s (2005) study meets this criterion — but also clinical in-use tests.

Although clinical tests provide crucial data, none were performed — a significant limitation that would call into question any conclusions that the Rutala et al. (2005) study might advance about the risk of infection associated with Duke’s mishap in 2004.

According to the FDA, the application of results from simulated in-use tests to actual clinical in-use conditions are at best “limited” (FDA, 2003).

Further, my review found that some of the Rutala et al. (2005) study’s data and conclusions are difficult to interpret and rationalize.

A more complete discussion of my review is provided, in detail, in an unpublished study.

Creutzfeldt-Jakob disease (CJD)

Third, published studies discuss the potential for contaminated instruments to pose an increased risk of patient-to-patient transmission of prions during certain types of surgical procedures.

Briefly, prions are infectious, rogue agents reported to be responsible for human transmissible spongiform encephalopathies, such as Creutzfeldt-Jakob disease (CJD).

These rare neurodegenerative diseases (for which there currently is no cure) are invariably fatal and characterized by brain deterioration, rapidly progressive dementia, the inability to stand or walk, and involuntary jerking movements.

Surgical instruments that may contact specific neurological tissues of patients suffering from a transmissible spongiform encephalopathy, such as the brain, spinal cord, and cornea, are considered at an increased risk of contamination with prions.

A number of reports highlight concerns over the possible transmission of CJD during surgery, with hospitals or health officials notifying affected patients of the infection risk. For example:

  • The Novant Health Forsyth Medical Center notified 18 neurosurgery patients in 2014 of their potential exposure to CJD, because the surgical instruments were not subjected to “enhanced” sterilization procedures as prescribed by the federal Centers for Disease Control and Prevention CDC to prevent the transmission of CJD;
  • Health officials in New Hampshire notified eight patients in 2013 who had brain surgery of their potential exposure to prions, which can survive standard sterilization;
  • Emory University Hospital in Atlanta (Georgia) notified more than 500 patients in 2004 of the potential for their exposure to surgical instruments contaminated with CJD;
  • The Hotel-Dieu Hospital in Windsor, Ontario temporarily closed its operating rooms in 2001 to prevent the transmission of CJD during surgery; and
  • Tulane University Hospital in New Orleans, La. reported in 2000 that eight surgical patients may have been infected with CJD.

Although few reported cases suggest that contaminated surgical instruments might have infected patients with prions during surgery, these five healthcare facilities realized the concern and therefore notified the affected patients of the infection risk.

Notwithstanding these five hospitals establishing an apparent precedent requiring the notification of patients potentially exposed to prions, the Rutala et al. (2005) study makes no mention of prions or whether Duke’s cleaning error might have facilitated their transmission during surgery.

As a result of prions being very resistant to conventional sterilization methods — including the two sterilization processes the Rutala et al. (2005) study evaluated: ethylene oxide gas and steam sterilization, — the CDC recommends enhanced cleaning and sterilization for surgical instruments potentially contaminated with prions, to minimize the risk of infection.

Because the Rutala et al. (2005) study did not address the potential for the transmission of prions during surgery, reasonable questions may be raised about the completeness of this study’s findings and its conclusion that Duke’s mishap did not pose a risk of infecting any of the 3,650 affected patients with a disease — including a prion disease.

To be sure, the possibility cannot be ruled out that some of Duke’s improperly cleaned surgical instruments might have been contaminated with prions, potentially exposing some of the 3,650 affected patients to these rogue infectious agents.

It is unclear why the Rutala et al. (2005) study overlooked this documented risk of prion infections.

Recommended reading: This article is to be read along with another of my articles, “Study of an Outbreak of Pseudomonas aeruginosa Following Arthroscopy in Texas in 2009.

A second study performed in 2008

Three years later in 2008, these same investigators performed a second set of tests in a laboratory setting — known herein as the “Rutala et al. (2008) study“– once again to simulate Duke’s mishap.

The Rutala et al. (2008) study

Like the first 2005 study, this 2008 study’s conclusions reaffirmed that Duke’s hydraulic-fluid cleaning mishap in 2004 — namely, the presence of hydraulic fluid on the contaminated surfaces of surgical instruments — did not prevent sterilization, whether achieved using pressurized steam or ethylene oxide gas, and, therefore, that Duke’s reprocessing mishap did not pose a risk of infection to patients.

Ironically, however, the methodology of the Rutala et al. (2008) study suggests an acknowledgement that, as my analysis, above, points out, some aspects of the design of their earlier 2005 study lacked rigor.

For example — unlike the Rutala et al. (2005) study, which my review concluded used surgical instruments that were too simple in design — this team’s second 2008 study used instead significantly more complex instruments as surrogates.

Namely, the second study simulated Duke’s cleaning error using:

  • a Kerrison rongeur (featuring a closed clasper),
  • a Stille-Luer rongeur (with a closed hinge),
  • a vaginal speculum (with a thumb screw), and
  • hemostatic forceps (closed hinge).

As they had three years earlier, these researchers reported in their 2008 study that: “Thus, our experiments (in 2008), which used much higher numbers of vegetative bacteria (than in 2005), provided assurance that infection would not result from instruments coated with hydraulic fluid and then sterilized.”

This 2008 study was funded, in part, by Duke University Medical Center.

Ethylene oxide gas vs. pressurized steam

Both the 2005 and 2008 studies performed by Rutala and his colleagues concluded that — no matter whether using pressurized steam or ethylene oxide gas — the sterilization of surgical instruments coated with hydraulic fluid was successful.

The results of these investigators achieved using pressurized steam are easier to understand than their results achieved using ethylene oxide gas. Particularly when sterilizing complex surgical instruments, such as hemostatic forceps with a closed hinge, or reusable endoscopic biopsy forceps used to sample tissues, other studies have routinely and consistently found steam sterilization to be superior in effectiveness to ethylene oxide gas, especially in the presence of a soil, like hydraulic fluid.

For example, Yoon et al. (2012) compared the effectiveness of ethylene oxide gas to pressurized steam for the sterilization of complex reusable endoscopic biopsy forceps.  Their results are traditional and as expected: steam sterilization was found to be effective, whereas vegetative bacteria (e.g., E. coli), which are delicate and ordinarily easily and rapidly destroyed by virtually all types of sterilizing agents, were found to have survived on a few of the biopsy forceps exposed to ethylene oxide gas.

Explaining their results, Yoon et al. (2012) acknowledged that sterilization with ethylene oxide gas can fail due to the complex design of biopsy forceps, which can prevent the sterilizing agent from contacting all of the instrument’s contaminated internal surfaces. These authors, therefore, recommend that reusable biopsy forceps be sterilized using pressurized steam.

In apparent contrast to Yoon et al.’s (2012) findings, however, Rutala et al. (2008), like the Rutala et al.’s (2005) study, reported that “hydraulic fluid did not affect the ability to sterilize the surgical or medical instruments with ethylene oxide.”

Also in contrast with the findings of Rutala et al. (2005) and Rutala et al. (2008), a study by Tosh et al. (2011) reported that debris on the surfaces of complex surgical instruments can cause even steam sterilization to fail, resulting in disease transmission and bacterial outbreaks.

Reconciliation of the differences between, on the one hand, the Rutala et al.’s (2005) and (2008) studies both concluding that ethylene oxide sterilization of complex, soiled medical instruments was successful, and, on the other, the findings of Yoon et al. (2012) and others — including Muscarella (1998) and Alfa et al. (1996), each of whom reported that ethylene oxide gas is prone to failure if the instrument is sufficiently complex in design, and especially if the instrument is contaminated with debris (like hydraulic fluid) — warrants further attention and discussion.


My review identified a number of limitations with the scope, design and methodology of Rutala et al.’s (2005) study, including their study’s use of surrogate instruments that were too simple in physical design to pose an adequate challenge and to comply with FDA recommendations.

Consequently, this study’s publicized assurance, in 2005, that the inadvertent use of hydraulic fluid by these two North Carolina hospitals to “clean” surgical instruments did not interfere with the sterilization process and, therefore, did not pose a risk of patient infection is questioned.

My more complete review of the Rutala et al. (2005) study is provided in an unpublished study.

More research is recommended to better define the effectiveness of pressurized steam, compared to ethylene oxide gas, and the limitations of both for the sterilization of surgical instruments soiled with debris including hydraulic fluid.

As a consequence of the limitations of the Rutala et al. (2005) study (and of this same team’s 2008 study), the possibility cannot be ruled out that some of the 3,650 patients directly affected by Duke’s hydraulic-fluid mishap in 2004 were infected with pathogenic microorganisms (and potentially exposed to prions) — notwithstanding the conclusions of Rutala et al. in 2005 and 2008.

An independent re-evaluation of Duke’s infection control mishap might better define the true infection risk.

Article by: Lawrence F Muscarella PhD. Posted January 4, 2013; updated: January 29, 2016.

Email: Larry@LFM-HCS.com; Twitter: @MuskiePhD


  1. Alfa MJ, DeGagne P, Olson N, et al. Comparison of ion plasma, vaporized hydrogen peroxide, and 100% ethylene oxide sterilizers to the 12/88 ethylene oxide gas sterilizer. Infect Control Hosp Epidemiol 1996 Feb;17(2):92-100.
  2. Duke University Health System. Letter to patients. June 20, 2015.
  3. Food and Drug Administration. Guidance on the Content and Format of Premarket Notification [510(k)] Submissions for Liquid Chemical Sterilants and High Level Disinfectants. (2003).
  4. Muscarella LF. The impact of Hydraulic Fluid on the Steam Sterilization of Surgical Instruments. [An unpublished study]
  5. Muscarella LF. Instrument design and cross infection. AORN J 1998 Mar;67(3):552-3, 556.
  6. Rutala WA, et al. Letter of results to Duke University Health System. Date: June 15, 2015.
  7. Rutala WA, et al. Impact of an oil-based lubricant on the effectiveness of the sterilization processes. Infect Control Hosp Epidemiol 2008 Jan;29(1):69-72.
  8. Tosh PK, et al. Outbreak of Pseudomonas aeruginosa Surgical Site Infections after Arthroscopic Procedures: Texas, 2009. Infect Control Hosp Epidemiol 2011;32(12):1179-1186.
  9. Yoon JH, Yoon BC, Lee HL, et al. Comparison of sterilization of reusable endoscopic biopsy forceps by autoclaving and ethylene oxide gas. Dig Dis Sci 2012 Feb;57(2):405-12.

Leave a Reply

Your email address will not be published.