Critically ill patients are at risk for developing diarrhea; reported prevalence rates are as low as 6.4%, but most reports vary from 30% to 41%.1,2Clostridium difficile is the most common cause of infectious diarrhea in the United States.3,4C difficile was first described in 1935, and a newer and more virulent strain of the pathogen (BI/NAP1/027) was identified in the late 20th century; this more virulent strain has been implicated in recent increases in the prevalence and mortality of C difficile –associated diarrhea (CDAD).5,6 For example, in Quebec, the incidence increased from 85,000 cases in 1993 to more than 300,000 cases in 2005.7 Similarly, a retrospective review in the United States from 2000 to 2005 found that the prevalence of CDAD increased by 23% per year, and this increased prevalence could not be attributed to improved awareness and more accurate reporting of the condition.8 Several characteristics of the newer strain contribute to more severe morbidity and to increased mortality, including increased secretion of toxins A and B, secretion of a binary toxin, and resistance to fluoroquinolones.4
Once C difficile infection occurs, the risk for spread to other patients is high. Millions of spores are excreted in the stool of infected patients, and the risk for environmental contamination in the room of a patient with CDAD is as high as 50%.9,10 Though multiple factors contribute to the risk for environmental contamination, uncontrolled passage of liquid stool is one of the most significant. C difficile infection is typically associated with high-volume liquid stools, which tremendously increases the risk of acute fecal incontinence (FI),3,11 and resultant environmental contamination.
Prevention and control of CDAD in critically ill patients relies on a bundle of interventions. An updated 2010 clinical practice guideline from the Society of Healthcare Epidemiology of America (SHEA) and Infectious Diseases Society of America (IDSA) divided environmental control of C difficile into 4 categories: (1) interventions for patients, health care workers, and visitors; (2) environmental cleaning and disinfection; (3) restriction of antibiotic use; and (4) use of probiotics.12 Preventive interventions aimed at health care workers, patients, and visitors are focused on treating existing CDAD while preventing its spread to others in the immediate environment. These interventions include meticulous hand hygiene when in contract with a patient with CDAD. All persons coming into contact with a person with C difficile infection are advised to use soap and water rather than alcohol-based cleansers, and to apply and regularly change gloves.12 In a study comparing standard hand hygiene with the use of vinyl gloves for all patient contacts in 2 intensive care units (ICUs), the use of gloves resulted in a 5-fold reduction in the spread of C difficile.13 Preventing environmental spread must also include interventions to effectively disinfect room surfaces and patient care equipment, such as use of hypochlorite-based agents or other cleaners with known antisporicidal properties. In addition, patients with CDAD should have dedicated equipment (such as blood pressure cuffs and stethoscopes) to prevent the spread of C difficile spores via the surfaces of these devices.12
The SHEA/IDSA clinical practice guidelines also recommend containment of liquid stool, especially in the critically ill patient with concomitant FI.12 Multiple methods of stool containment, including absorptive briefs for adults, absorbent bed pads, fecal pouches, and stool management systems, may be used. While all forms of containment and absorptive devices are designed to contain liquid feces, stool management systems have proven particularly useful in critically ill patients with liquid diarrhea, including CDAD. Three devices are available in the United States; all share a number of common features including a retention cuff that is inserted into the rectum above the anal sphincter, connecting tubing and a hub that allows regular changes of containment bags, and a containment bag for stool. Limited research strongly suggests that these devices help prevent incontinence-associated dermatitis and sacral pressure ulcer development in critically ill patients.14,15 However, research on the ability of these devices to prevent environmental spread of C difficile in the ICU is extremely sparse. Jones and colleagues16 evaluated the ability of a stool management system (with and without charcoal filtered collection bags) to prevent contamination over a period of 11 days. The collection bags were filled with simulated fecal effluent inoculated with C difficile, and the external surfaces of the devices (collection bags, catheter and balloon irrigation ports, and connecting tubing) were swabbed 5 times weekly to detect contamination. The collection bags were changed twice during the 11-day period, on days 4 and 8. The study also incorporated a positive control arm consisting of bags with purposely created needle holes. C difficile was not detected on any of the external surfaces of the sealed stool management systems. In contrast, contamination was detected on all of the positive control systems whose integrity had been compromised intentionally with a sterile needle. A fecal management system manufacturer reported results of an in vitro study suggesting that their stool management system prevented environmental contamination over a period of 29 days.17 However, these findings have not been published in a peer-reviewed journal and the methods and statistical analyses used to generate results cannot be adequately judged. No studies completed in a clinical setting were identified that compared the efficacy of various stool management systems for the containment of C difficile spread in the hospital setting.
The purpose of this study was to compare contamination of the immediate environment with C difficile spores and vegetative cells from 2 stool management systems over a period of 30 days in a controlled laboratory setting. Specific research questions were as follows: (1) Does each of the stool management systems prevent contamination with C difficile at the surface of the containment bag and at the hub and interface tubing when containment bags are changed on a daily basis over a period of 30 days? (2) Do daily bag changes result in contamination of the immediate environment over a period of 30 days? and (3) Do differences in system design affect the likelihood of contamination?
This in vitro study was designed to simulate use of a fecal management system in a clinical setting. 2 fecal management systems, 16 systems total, were compared; 8 were labeled device 1 (DigniShield Stool Management Systems, C. R. Bard, Inc, Covington, Georgia), and 8 were labeled device 2 (Flexi-Seal Signal Fecal Management Systems, ConvaTec, Inc, Princeton, New Jersey). Each fecal management system was initially laid flat along the length of a shelf and the inflation cuff was inflated with sterile water based on manufacturer recommendations. The containment bags were hung from the shelf to a level below the flat shelf, with the retention cuff elevated 11 to 12 cm above the shelf to simulate orientation of these devices when placed in the clinical setting (Figure 1). Devices were segregated on opposite sides of the laboratory to prevent cross-contamination.
Stool management systems were filled with sterile, loose canine stool inoculated with a concentration of approximately 106 colony-forming units (CFUs) per milliliter of C difficile (ATCC 42355). The strain was obtained from a toxin B producing clinical isolate cultured from an abdominal wound. Each device was filled on a daily basis (Monday to Friday) with 800 ml; stool was introduced from the cuff end of the device because it most closely resembles flow in the clinical setting. The cuff was then covered to maintain and simulate the closed system obtained when the cuff is placed in the human rectum. Each device was slightly elevated to allow approximately 300 ml of the stool to remain in the tubing to more closely mimic the situation when the devices are placed in a patient with high-volume liquid diarrhea. All devices were filled from the same batch of artificially contaminated stool to ensure a consistent risk of contamination for all devices tested.
Culturing and typing C difficile is technically challenging and current diagnosis is based on enzyme immunoassays for toxins A, B, and glutamate dehydrogenase.5 In order to prepare the inoculum for the stool, a media of Brain Heart Infusion Supplement (BHIS, Becton, Dickinson and Co, Sparks, Maryland) and 0.01% taurocholate was used to grow the bacterial broth culture from the cryovial stock. C difficile agar plates (CDSA) were prepared by adding 8 μg/ml of cefotaxime and 100 μg/ml of cycloserine to BHIS plates and used to culture recovered organisms. All cultures were incubated under anaerobic conditions for 24 to 48 hours.
Sterile swabs were used to recover C difficile organisms or spores from each of the following locations: (1) the front surface of the collection bag, (2) the outer surface of the hub where the collection bag connects to the tubing, and (3) the distal 12 inches of the connecting tubing nearest the collection bag. An absorbent pad was placed under the area where bag changes were conducted and a CDSA plate was placed in close proximity to each system to evaluate environmental contamination via leakage or aerosolization of C difficile cells or spores. Swabs were then streaked onto CDSA plates and the piece of absorbent pad over which the bag was changed was cut out and placed on the agar surface of a CDSA plate. After undergoing a 24- to 48-hour anaerobic incubation period, each plate was assigned an ordinal growth score based on a 6-point ordinal scale where “−” indicated no growth and “+++++” indicated maximal growth (Figure 2). In addition, on days 3, 10, 20, and 30 of data collection, the absorbent tips of the swabs were cut off and placed in a recovery medium that was filtered through a 0.45-μm membrane filter (Pall Corporation, Ann Arbor, Michigan) to allow quantitative analysis of C difficile growth based on CFUs per surface.
Handling of Materials and Quality Control
All data collectors were trained in data collection procedures prior to initiation of the study. Data collectors also practiced soap- and water-based hand hygiene prior to contact with all study materials. Sterile gloves were worn throughout all test procedures, including manipulation of the fecal management systems and swabbing of all surfaces. Bag changes were performed in accordance with the Instructions for Use packaged with each device. Data collectors who obtained swabs from the devices could not be blinded to the device identity owing to difference in device design and appearance. Investigators who ranked colonization were blinded to the identity of the device from which the samples were collected.
Data were analyzed via the SAS software program, version 9.3 (Cary, North Carolina). Group comparisons were based on 2 techniques. Swabs and CDSA plates placed near the devices were analyzed throughout the data collection period and analysis was completed based on a dichotomous and ordinal scale for each day of data collection. A dichotomous outcome variable (growth vs no growth) was used to analyze contamination on a daily basis. In addition, the magnitude of growth was analyzed based on the 6-point ordinal scale illustrated in Figure 2. Devices were also compared on days 3, 10, 20, and 30 by measuring CFUs per device surface. Logistic regression analysis was used to analyze growth over time; day and study devices were included in the model. The Generalized Estimating Equation (GEE) was used as a model for analysis of dichotomous variables. When observations showed no growth, the Cochran-Mantel Haenszel test was used to compare study devices. Statistical significance was declared when P < .05.
Daily analysis based on dichotomous variables yielded 168 observations. When C difficile growth on the surface of collection bags was compared, 20.8% of device 1 surfaces were contaminated throughout the data collection period as compared to 83.9% of device 2 surfaces. Analysis of these data based on the ordinal scale described in Figure 2 revealed that more than half of device 2 measurements were ranked as maximum growth as compared to 1 maximum growth obtained from device 1. Comparison of growth on the tubing/hub interface found that 20.8% of device 1 systems were contaminated versus 86.3% of systems in the device 2 group. Measurement of the absorbent pad revealed that 0.5% of device 1 measurements were contaminated as compared to 61.9% of measurements from device 2. Comparison of CDSA plate growth found that no device 1 plates were contaminated as compared to 21.4% of device 2 plates. Table 1 summarizes rates of contamination for device 1 versus device 2.
Quantitative measurement based on CFU/device counts completed at days 3, 10, 20, and 30 yielded 32 observations for each system. Significantly less contamination was observed for device 1 compared to device 2 in all locations. Table 2 summarizes contamination rates for device 1 versus device 2.
Although evidence is limited, current research strongly suggests that stool management systems greatly reduce or eliminate leakage of liquid stool into the environment. The majority of these reports have focused on clinical consequences of exposure of the skin to liquid stool produced by diarrhea and FI. Benoit and Watts15 reported on use of stool management systems in a quality improvement project in a 21-bed surgical ICU; they found that use of stool management systems reduced the incidence of stage II or higher pressure ulcers from 43% to 12.5% over the 6-month surveillance period. Keshava and colleagues18 reported results of a before-after study involving use of a stool management system for a group of 20 patients with severe perineal skin damage (n = 13) or perineal burn wounds (n = 7); they compared perianal disease activity scores (perineal skin inflammation and damage, presence of a fistula, functional impact score, and volume of perianal discharge), mean frequency of perineal dressing changes, and bed linen changes. Insertion of the bowel management system reduced mean perineal skin activity scores from 14 to 6.4 (P < .001), mean perineal dressing changes from 3.3 to 1.5 changes per day, and mean bed linen changes from 9.3 to 1.2 per day. Pittman and colleagues19 reported results of a randomized controlled trial comparing 3 methods of management of fecal incontinence involving liquid stool in 6 ICUs located in the Midwestern United States. Subjects were randomly allocated to 3 groups: a stool management system, nasopharyngeal trumpet placed into the rectum and attached to a drainage system, and usual care (use of absorptive pads with regular skin cleansing after fecal incontinence episodes). Patients allocated to the stool management system had significantly lower scores on the incontinence-associated dermatitis and its severity instrument (P < .001) than subjects receiving usual care; occurrence rates of pressure ulcers did not vary among the 3 groups.
While findings reported by these authors suggest that use of a stool management system reduces or prevents cutaneous complications of fecal incontinence compared to alternative containment strategies such as use of absorbent pads, they address outcomes other than environmental contamination with C difficile, which is the basis for this study. A literature review did not identify any clinical studies evaluating C difficile contamination or cross infection in patients using a stool management system or any other containment product, such as absorbent pads. The study by Jones and associates16 suggests that a stool management system may have the potential to prevent environmental contamination in an in vitro setting. These researchers also evaluated the ability of an absorbent pad to contain stool inoculated with C difficile in an in vitro setting over a 5-day period. Two absorptive pads were sandwiched between plates of glass and the absorptive surface of the pad was exposed to the inoculated stool medium. C difficile transmission was evaluated by swabbing the nonabsorptive surface of the pad on days 1, 2, and 5 followed by transfer to prepared agar plates and anaerobic incubation for a period of 48 hours. Analysis revealed no contamination of the nonabsorptive surface of the absorptive pads; however, considerable spread of the stool was noted laterally across the absorptive surfaces of the pads, suggesting the possibility of contamination via the pad to the bed linen and immediate environment, especially when patients experienced the high-volume diarrhea often associated with CDAD.
C difficile infection control relies on multiple strategies, including containment of the liquid stool containing vegetative cells and spores.12 This study focused on one strategy, use of a stool management system for prevention of contamination in patients with CDAD. Findings demonstrate that use of a stool management system significantly reduced environmental spread of C difficile in a laboratory setting. This study also extended the results of a previous study by Jones and colleagues16 by extending the data collection period to 30 days and by completing daily bag changes throughout this period. This design element was deemed particularly important because it more closely simulates the clinical environment where stool management systems may be left in situ for up to 29 days based on the manufacturers' recommendations, and where daily bag changes are typically performed.12 Study results also reveal statistically significant differences in the performance of 2 stool management systems in an in vitro study setting. The highest magnitude differences were in the rates of contamination of the anterior surface of the bag, tube/hub interface where the tubing and bag connect, and absorbent pads placed underneath the device during bag changes. These differences suggest that the principal difference between the devices may be the interface between the hub and connecting tubing. Reducing contamination from this source is especially important since this is the point where stool management systems are regularly opened in order to change collecting bags when they become filled with stool.
Several limitations of this study should be considered when interpreting findings. Data were collected in an in vitro setting and findings cannot be directly applied to the clinical environment such as an ICU. All study procedures were completed by Innovotech, Inc, an independent company hired by the sponsoring company (C. R. Bard, Inc). In addition, the simulated clinical model used in this study could not adequately reproduce the seepage of small amounts of stool around the device and onto the perianal skin; this is frequently observed in the clinical setting and is an important factor to be considered when evaluating environmental contamination.
Stool management systems have been shown to protect critically ill and burn patients from perianal and perineal skin damage associated with exposure to liquid stool. However, evidence concerning their potential to prevent environmental spread of C difficile in patients with CDAD is especially sparse. Findings from this in vitro study demonstrate that stool management systems can limit or prevent environmental contamination with C difficile. Findings also demonstrate statistically significant differences in the 2 systems compared in this study; we believe these differences are most likely attributable to differences in the tubing/collection bag interface, the point where stool management systems are most often disconnected as collection bags are changed.
1. Izaguirre Guerricagoitia L, Truchuelo Aragon A. Prevalence of diarrhea investigation group. Prevalence of diarrhea in critical care units of Spain. Enferm Intensiva. 2011;22(2):65–73.
2. Kelly TWJ, Patrick MR, Hillman KM. Study of diarrhea in critically ill patients. Crit Care Med. 1983;11:7–9.
3. Beitz JM. Fecal incontinence
in acutely and critically ill patients: ostomy wound management. Ostomy Wound Manage. 2006;52(12):56–58, 60, 62-66.
4. Rupnik M, Wilcox MH, Gerding DN. Clostridium difficile
infection: new developments in epidemiology and pathogenesis. Nature. 2009;7(7):526–537.
5. Bobo LD, Dubberke ER. Recognition and prevention of hospital associated enteric infections in the intensive care unit. Crit Care Med. 2010;38(8)(suppl):S324–S334.
6. Bartlett JG. Historical perspectives on studies of Clostridium difficile
and C difficile
infection. Clin Infect Dis. 2008;46(suppl 1):S4–S11.
7. Pepin J, Valiquette L, Cossette B. Mortality attributable to nosocomial Clostridium difficile
–associated disease during an epidemic caused by a hypervirulent strain in Quebec. Can Med Assoc J. 2005;173:1037–1042.
8. Zilberberg MD, Shorr AF, Kollef MH. Increase in adult Clostridium difficile
–related hospitalizations and case-fatality rate, United States, 2000-2005. Emerg Infect Dis. 2008;14:929–931.
9. Madan R, Petri WA Jr. Immune responses to Clostridium difficile
infection. Trends Mol Med. 2012;18(11):658–666.
10. Best EL, Fawley WN, Parnell P, Wilcoxon MH. The potential for airborne dispersal of Clostridium difficile
from symptomatic patients. Clin Infect Dis. 2010;50:1450–1457.
11. Garcia CB, Binks R, De Luca E, et al. Prevalence, management and clinical challenges associated with acute fecal incontinence
in the ICU and critical care settings: the FIRST cross-sectional descriptive survey. Intensive Crit Care Nurs. 2012;84:242–250.
12. Cohen SH, Gerding DN, Johnson S, et al. Society for Healthcare Epidemiology of America. Infectious Diseases Society of America. Clinical practice guidelines for Clostridium difficile
infection in adults: 2010 update by the Society of for Healthcare Epidemiology in America (SHEA) and the Infectious Disease Society of America (IDSA). Infect Control Hosp Epidemiol. 2010:31:431–455.
13. Johnson S, Gerding DN, Olson MM, et al. Prospective, controlled study of vinyl glove use to interrupt Clostridium difficile
nosocomial infection. Am J Med. 1990;88(2):137–140.
14. Wilczweski P, Grimm D, GIanakis A, Gill B, Sarver W, McNett M. Risk factors associated with pressure ulcer development in critically ill traumatic spinal cord injury patients. J Trauma Nurs. 2012;19(1):5–10.
15. Benoit R Jr, Watts C. The effect of a pressure ulcer prevention program and the bowel management system in reducing pressure ulcer prevalence in an ICU setting. J Wound Ostomy Continence Nurs. 2007;34(2):163–175.
16. Jones S, Towers V, Welsby S, Wishin J, Bowler P. Clostridium difficile
containment properties of a fecal management system: an in vitro
investigation. Ostomy/Wound Manag. 2011;57(10):38–49.
18. Keshava A, Renwick A, Stewart P, Pilley A. A nonsurgical means of fecal diversion: the Zassi bowel management system. Dis Colon Rectum. 2007;50(5):1–5.
19. Pittman J, Beeson T, Terry C, Kessler W, Kirk L. Methods of bowel management in critical care: a randomized controlled trial. J Wound Ostomy Continence Nurs. 2012;39(6):633–639.