risk assessment in infection control: which risks? •

Risk Assessment in Infection Control: Which Risks? • Author(s): Michael Millar , FRCPathSource: Infection Control and Hospital Epidemiology, Vol. 29, No. 4 (April 2008), pp. 381-382Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiologyof AmericaStable URL: http://www.jstor.org/stable/10.1086/529122 .

Accessed: 16/05/2014 01:09

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaboratingwith JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 193.104.110.108 on Fri, 16 May 2014 01:09:40 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=ucpress

http://www.jstor.org/action/showPublisher?publisherCode=shea

http://www.jstor.org/action/showPublisher?publisherCode=shea

http://www.jstor.org/stable/10.1086/529122?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


infection control and hospital epidemiology april 2008, vol. 29, no. 4

l e t t e r s t o t h e e d i t o r

Outbreak of Burkholderia cepaciaBacteremia Caused by ContaminatedChlorhexidine in a Hemodialysis Unit

t o t h e e d i t o r — Burkholderia cepacia is a widespreadgram-negative environmental bacillus associated with noso-comial infection.1,2 This organism colonizes water supplies,filter membranes, and antiseptic solutions.3-5 Inadequate cath-eter care, defects in membrane integrity, and the reprocessingof dialyzers have all been implicated in outbreaks in hemo-dialysis units.6,7 We describe an outbreak of B. cepacia bac-teremia that occurred in Madrid, Spain, in the Alcobendashemodialysis center.

The Alcobendas hemodialysis center has 14 stations andserves approximately 60 outpatients with chronic renal failureper month. Municipal water passes through cartridge filters,water softeners, carbon filters, and a simple reverse-osmosismembrane unit before being distributed to the dialysis sta-tions in a closed-loop circuit. Dialysis machines are disin-fected according to the manufacturer’s instructions at the endof each dialysis session.

Chlorhexidine was purchased by the hemodialysis centeras a 5% solution and diluted with deionized water to a 2.5%solution, then stored in 250-mL plastic bottles. The solutionwas prepared every 7-10 days by healthcare workers. The 2.5%chlorhexidine solution was used to disinfect the skin of pa-tients before catheter insertion and during follow-up care.

The nosocomial outbreak of B. cepacia bacteremia occurredfrom December 1, 2005, through April 30, 2006. Five out-patients developed symptoms of bacteremia. The overall at-tack rate was 1.6%. An outbreak case was defined by thepresence of fever and the detection of B. cepacia in bloodcultures.

Four of the case patients had long-term central venouscatheters (CVCs) and developed high temperatures withoutsigns of catheter-related infection. The catheters were re-moved from all 4 patients but only 1 catheter was sent to thelaboratory for culture.

The unusually high number of B. cepacia isolates motivatedus to carry out an epidemiological study. Blood samples areusually inoculated into both aerobic and anaerobic media forprocessing with the Versatrek culture system (Trek DiagnosticSystems). The blood samples were cultured on 3 agar plates:sheep blood agar, chocolate blood agar, and brucella agar.The sheep blood agar and chocolate blood agar plates wereincubated at 35�C in an atmosphere containing 5% CO2 for48 hours. The brucella agar was incubated at 35�C in ananaerobic atmosphere for 48 hours.

Environmental cultures were performed on potential

sources of B. cepacia contamination: the deionized water usedto dilute the dialysate concentrate, the postosmosis water, thetap water of the 14 hemodialysis stations, the undiluted 5%chlorhexidine solution, the diluted 2.5% chlorhexidine so-lution, and the povidone-iodine solution. All environmentalsamples were cultivated on B. cepacia–selective agar and in-cubated at 35�C in an atmosphere containing 5% CO2 for 18hours.

Biochemical identification was performed with the Vitek 2GNI card (bioMerieux Vitek) and Biolog GN2 panels(Biolog). Susceptibility testing was performed with the WiderSystem (Francisco Soria Melguizo). Genotyping of clinicaland environmental B. cepacia strains was performed afterdigestion with the restriction enzyme XbaI with the CHEFDR-III system (BioRad) according to the conditions previ-ously described.8

During the outbreak, B. cepacia was isolated from the bloodof 5 patients treated at the hemodialysis unit and from 2environmental samples. This organism was detected in the2.5% chlorhexidine solution and in the water used to dilutethe chlorhexidine at station 3. Other potential sources ofcontamination yielded no B. cepacia.

All the clinical and environmental isolates were of Burk-holderia stabilis sp. nov. (formerly Burkholderia cepacia gen-omovar IV) and had identical DNA banding patterns, as de-termined by pulsed-field gel electrophoresis (PFGE),suggesting their relation and a common source of infection.The clonality of the 7 isolates is shown in the Figure.

Our findings suggest that the outbreak of B. cepacia bac-teremia in patients receiving long-term hemodialysis at ourinstitution was caused by the use of contaminated alcohol-free 2.5% chlorhexidine solution for skin disinfection beforecatheter insertion and during follow-up care. All the patientsof the hemodialysis unit were exposed to this solution, butonly 5 women, with a median age of 79 years (range, 77�81years), developed infection. Concomitant long-term use of aCVC and advanced age were significant risk factors.

We hypothesize that the water tap of station 3 was con-taminated with the B. cepacia through inappropriate manip-ulation when the 2.5% chlorhexidine solution was preparedby healthcare workers, because samples of water from the restof the distribution system in the hemodialysis unit yieldedno B. cepacia. After we discarded the diluted chlorhexidinesolution and cleaned the tap of station 3 with solutions ofsodium hypochlorite and 80% v/v ethanol, no further caseswere identified.

Environmental bacteria are becoming increasingly involvedin outbreaks. In these cases, molecular identification and typ-ing are fundamental for determining the correct strategy inthe management of epidemics.

On the basis of evidence from our study and others in theliterature, we recommend that hospitals and other hemodi-



378 infection control and hospital epidemiology april 2008, vol. 29, no. 4

figure. Pulsed-field gel electrophoresis (PFGE) of B. cepaciastrains. The l denotes the lambda ladder PFGE weight marker (48.5kbp concatemers). Lane 1, an epidemiologically unrelated strain re-covered from a cystic fibrosis patient; Lane 2, strain retrieved fromthe catheter of patient 2; Lanes 3-7, strains isolated from the bloodof patients 1, 2, 3, 4, and 5, respectively; Lane 8, strain recoveredfrom a sample of the tap water of hemodialysis station 3; Lanes 9and 10, strains recovered from 2 bottles of 2.5% chlorhexidinesolution.

alysis units discontinue the use of diluted alcohol-free chlor-hexidine solution. In fact, the manufacturer’s instructionsrecommend adding alcohol when diluting the chlorhexidinefor skin disinfection before catheter insertion and during fol-low-up care.

acknowledgments

Potential conflicts of interest. All authors report no conflicts of interest rel-evant to this study.

M. P. Romero-Gomez, MD; M. I. Quiles-Melero, MD;P. Pena Garcıa, MD; A. Gutierrez Altes, MD;

M. A. Garcıa de Miguel, MD; C. Jimenez, MD;Sylvia Valdezate, PhD; J. A. Saez Nieto, PhD

From the Departments of Microbiology and Parasitology (M.P.R.-G.,M.I.Q.-M., P.P.G., A.G.A.) and Nephrology (M.A.G.M., C.J.), La PazUniversity Hospital, and the Servicio de Bacteriologıa, Centro Nacional deMicrobiologıa, Instituto de Salud Carlos III, Majadahonda (S.V., J.A.S.N.),Madrid, Spain.

Address reprint requests to M. P. Romero-Gomez, Department ofMicrobiology and Parasitology, La Paz University Hospital, Paseo de laCastellana, 261, 28046 Madrid, Spain ([email protected]).

Infect Control Hosp Epidemiol 2008; 29:377-378� 2008 by The Society for Healthcare Epidemiology of America. All rightsreserved. 0899-823X/2008/2904-0018$15.00. DOI: 10.1086/529032

references

1. Humar A, Oxley C, Sample ML, Garner G. Elimination of an outbreakof gram-negative bacteremia in a hemodialysis unit. Am J Infect Control1996; 24:359-363.

2. Lu DC, Chang SC, Chen YC, Luh KT, Lee CY, Hsieh WC. Burkholderiacepacia bacteremia: a retrospective analysis of 70 episodes. J Formos MedAssoc 1997; 96:972-978.

3. Reboli AC, Koshinski R, Arias K, Marks-Austin K, Stieritz D, Stull TL.An outbreak of Burkholderia cepacia lower respiratory tract infection as-sociated with contaminated albuterol nebulization solution. Infect ControlHosp Epidemiol 1996; 17:741-743.

4. Molina-Cabrillana J, Bolanos-Rivero M, Alvarez-Leon EE, et al. Intrin-sically contaminated alcohol-free mouthwash implicated in a nosocomialoutbreak of Burkholderia cepacia colonization and infection. Infect ControlHosp Epidemiol 2006; 27:1281-1282.

5. Ghazal SS, Al-Mudaimeegh K, Al Fakihi EM, Asery AT. Outbreak ofBurkholderia cepacia bacteremia in immunocompetent children caused bycontaminated nebulized sulbutamol in Saudi Arabia. Am J Infect Control2006; 34:394-398.

6. Kaitwatcharachai C, Silpapojakul K, Jitsurong S, Kalnauwakul S. An out-break of Burkholderia cepacia in hemodialysis patients: an epidemiologicand molecular study. Am J Kidney Dis 2000; 36:199-204.

7. Magalhaes M, Doherty C, Govan JR, Vandamme P. Polyclonal outbreakof Burkholderia cepacia complex bacteraemia in haemodialysis patients. JHosp Infect 2003; 54:120-123.

8. Anderson DJ, Kuhns JS, Vasil ML, Gerding DN, Janoff EN. DNA finger-printing by pulsed field gel electrophoresis and ribotyping to distinguishPseudomonas cepacia isolates from a nosocomial outbreak. J Clin Microbiol1991; 29:648�649.

Outbreak of Pseudomonas aeruginosaInfections Associated With ContaminatedWater in a University Hospital in Tunisia

To the Editor—Pseudomonas aeruginosa is a major pathogenthat causes nosocomial infections, particularly in ventilatedand/or immunocompromised patients. This organism isubiquitous in moist environments and is frequently found invarious hospital sites. Strains involved may be spread via thehands of healthcare workers or by an environmental source,such as contaminated water.1-5 The installation of surveillancesystems that detect outbreaks of nosocomial infection is im-portant; sources of infection may be characterized and iso-lated, and modifications in procedures made to stop furtherinfections from occurring. Infection is a frequent event insurgical wards, such as the urology ward, where endoscopyand surgical procedures are common.

We describe an outbreak of P. aeruginosa infection thatoccurred between July and September 2005 in the urologyward in the University Hospital Sahloul in Tunisia, a 548-bed hospital. An investigation of the environment was doneto determine the potential source of infection and to imple-ment control measures to stop the outbreak. The urology



letters to the editor 379

figure. Pulsed-field gel electrophoresis banding patterns ofchromosomal DNA of 15 Pseudomonas aeruginosa isolates from pa-tients and from samples of water used in ureteroscopy. Lane M,molecular size marker; lane LMG, reference strain P. aeruginosaLMG5827; lanes A, B, and C, environmental isolates; lanes 1 to 10,isolates from patients who had undergone ureteroscopy; lanes 11and 12, isolates from 2 other patients in the urology ward duringthe outbreak.

ward is a 60-bed unit situated in the third floor of the hospital.Endoscopic procedures are performed using an ultravioletwater disinfection system; this technology has been in placesince the hospital opened in 1990. Endoscopes were disin-fected with a quaternary ammonium compound and gluter-aldehyde and rinsed with water.

Twelve patients were diagnosed with P. aeruginosa urinarytract infection. Ten of those patients had undergone a ure-teroscopic procedure 24 to 72 hours before diagnosis of in-fection. Isolates were recovered from urine specimens.

An environmental survey was done in order to determinepotential reservoirs or sources of contamination. Water wassampled at 3 points related to the ureteroscopic equipment:the water outflow from the sterilizer, the water reservoir con-tainer, and the irrigation tube in contact with the patient.Samples of solutions used for disinfection of the equipmentand swab samples of environmental surfaces, cystoscopes,ureteroscopes, and resectoscopes were cultured.

Strains recovered from patients and from environmentalspecimens were genotyped by macrorestriction of genomeDNA with SpeI, followed by pulsed-field gel electrophoresis(PFGE).6,7 Banding patterns obtained were analyzed by visualinspection according to criteria described by Tenover et al.8

The 10 strains recovered from patients who had undergoneendoscopy showed the same PFGE banding pattern as isolatesobtained from water outflow, reservoir, and irrigation spec-imens. The 2 strains recovered from patients who had notundergone endoscopy were different in antibiotype profileand in PFGE pattern (Figure).

This outbreak of P. aeruginosa infection was caused by theuse of inadequately disinfected water used in ureteroscopicprocedures. In our hospital, water for bladder irrigation ispassed through an ultraviolet disinfection system. This systemfailed to disinfect the water and its use resulted in the out-break. The use of commercial sources of sterile water was notfeasible because of cost. An alternative system consisting offiltration with 3 filters (with pore sizes of 25 mm, 0.5 mm,and 0.2 mm) was recommended by the hygiene departmentand implemented. After the outbreak and the resulting im-plementation of the new water system, we performed routineand systematic surveillance for 3 months of the environment,equipment, water sources, and patients who had undergoneureteroscopy. We found no further episodes of patient in-fection with P. aeruginosa or other bacteria and no contam-ination of the environment with the epidemic strain.

In conclusion, we found that the source of P. aeruginosathat caused the outbreak was the water used for bladder ir-rigation during endoscopic procedures. Molecular typing sys-tems were important in the detection of the outbreak andconfirmation of the environmental source. The water disin-fection system was changed and rigorous surveillance con-firmed that further infections with the epidemic strain wereprevented.

acknowledgments


Wejdene Mansour, PhD; Olfa Bouallegue;Houyem Said, MD; Safia Dahmen, PhD;

Noureddine Boujaafar

From the Unite des Infections a Bacteries Multiresistantes aux Anti-biotiques, Laboratoire de Microbiologie (W.M., O.B., S.D., N.B.), and theService d’Hygiene Hospitaliere (H.S.), CHU Sahloul, Sousse, Tunisia.

Address reprint requests to Wejdene Mansour, Unite des Infections aBacteries Multiresistantes aux Antibiotiques, UR/29/04, Laboratoire deMicrobiologie, CHU Sahloul, 4054 Sousse, Tunisie ([email protected]).

Infect Control Hosp Epidemiol 2008; 29:378-380� 2008 by The Society for Healthcare Epidemiology of America. All rightsreserved. 0899-823X/2008/2904-0019$15.00.DOI: 10.1086/529588

references

1. Widmer AF, Wenzel RP, Trilla A, Bale MJ, Jones RN, Doebbeling BN.Outbreak of Pseudomonas aeruginosa infections in a surgical intensive careunit: probable transmission via hands of health care worker. Clin InfectDis 1993; 16:372-376.

2. Bukholm G, Tannaes T, Kjelsberg AB, et al. An outbreak of multidrug-resistant Pseudomonas aeruginosa associated with increased risk of patientdeath in an intensive care unit. Infect Control Hosp Epidemiol 2002; 23:441-446.




3. Berrouane YF, McNutt LA, Buschelman BJ, et al. Outbreak of severePseudomonas aeruginosa infections caused by a contaminated drain in awhirlpool bathtub. Clin Infect Dis 2000; 31:1331-1337.

4. Bou R, Aguilar A, Perpinan J, et al. Nosocomial outbreak of Pseudomonasaeruginosa infections related to a flexible bronchoscope. J Hosp Infect2006; 64:129-135.

5. Kikuchi T, Nagashima G, Taguchi K, et al. Contaminated oral intubationequipment associated with an outbreak of carbapenem-resistant pseu-domonas in an intensive care unit. J Hosp Infect 2007; 65:54-57.

6. Kokis VM, Moreira BM, Pellegrino FLPC, et al. Identification of an im-ipenem-resistant Pseudomonas aeruginosa clone among patients in a hos-pital in Rio de Janeiro. J Hosp Infect 2005; 60:19-26.

7. Pellegrino FL, Teixeira LM, Carvalho MD, et al. Occurrence of a multi-drug-resistant Pseudomonas aeruginosa clone in different hospitals in Riode Janeiro, Brazil. J Clin Microbiol 2002; 40:2420-2424.

8. Tenover F, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNArestriction patterns produced by pulsed-field gel electrophoresis: criteriafor bacterial strain typing. J Clin Microbiol 1995; 33:2233-2239.

Modified Measles in a Healthcare WorkerAfter Return From Travel

To the Editor—Measles is a highly communicable but vac-cine-preventable infectious disease. It has been reported thatmodified measles, usually presenting with mild symptoms oreven no symptoms, could occur in measles-vaccinated in-dividuals during a measles outbreak.1 Because of the proteanclinical presentations of modified measles, it is difficult toraise clinical suspicion and/or make a rapid diagnosis withoutknowing the history of exposure. Early detection of modifiedmeasles would be a great advance in the interruption of dis-ease transmission.

During a measles epidemic, healthcare settings can becomea high-risk environment for measles transmission. Infectedhealthcare personnel can shed the measles virus particlesduring the prodromal period before clinical characteristicsappear2 and thus transmit the measles virus to susceptiblecoworkers, patients, and family members.3 Moreover, sus-ceptible patients in hospitals are often vulnerable individualswho will suffer severe complications of measles.4 We describea case of modified measles in a healthcare worker who hadreturned from travel to another country.

A 26-year-old male doctor, an intern, visited Tokyo, Japan,from May 19 to May 25, 2007, where a measles outbreak hadoccurred.5 On June 2, after his return to Taiwan, he presentedwith fever and arthralgia, followed by a progressive pustule-like skin rash, which initially appeared on his trunk and faceon the third day of illness and then extended to the extrem-ities. He was hospitalized the next day. Isolation precautionsfor airborne pathogen were implemented during medicalcare. The diagnosis of modified measles was made on thebasis of the following findings: (1) the absence of classicmanifestations of measles, such as cough, conjunctivitis, co-

ryza, or Koplik’s spot; (2) travel to an area where measles isendemic; (3) a self-report of 2-dose measles vaccination inchildhood; and (4) the presence of measles IgG in serumobtained at the acute stage of infection and of IgM in serumobtained at the convalescent stage. He was discharged withoutsequelae after 7 days’ hospitalization.

During the prodromal phase of the patient’s disease, 2ambulatory patients and 25 medical personnel had close con-tact with the patient. The majority of these medical personnel(23 [92%]) recalled prior measles vaccination. Serologicaltests were performed on 32 staff members who cared for thepatient, and 28 (87.5%) had detectable measles IgG in serum.No subsequent case of measles was identified.

Measles is still a major health problem because of its world-wide prevalence and its changing epidemiologic pattern incountries where measles vaccine is widely used.2 Despite highlevels of vaccine coverage, measles outbreaks still occur be-cause of the accumulation of susceptible, unvaccinated per-sons and/or of persons without an adequate immunologicalresponse to measles vaccine. The diagnosis of measles duringthe present vaccine era has been complicated by the changein the age incidence of measles, the alteration of disease man-ifestations resulting from previous immunization, and theapparently sporadic occurrence of measles cases.6 Cases ofmodified measles, which is characterized by an atypical ormild clinical presentation in a vaccinated patient, have beenobserved during a sustained outbreak.1 The transmission ofmeasles to patients exposed to sick healthcare workers hasalso been documented.7 The highly contagious nature of themeasles virus also underscores the need for appropriate in-fection control measures to reduce the risk of nosocomialtransmission. In the investigation we describe, the delay indiagnosis and confirmation of the index case was problematic,as it resulted in a delay in contact tracing and follow-up.8

Acquisition of communicable diseases by healthcare per-sonnel during travel poses a potential threat of nosocomialoutbreak. At the present time, there are few rational rec-ommendations for preventing travel-associated illness amonghealthcare personnel.9 We suggest that healthcare workers bescreened for measles IgG antibodies during their occupationalhealth assessment, and nonimmune and uninfected individ-uals should be vaccinated. In our institution, studies are on-going to assess the level of measles immunity in healthcareworkers to determine strategies for measles screening andvaccination. To obtain adequate documentation of previousmeasles vaccination or immunity to measles for a large num-ber of hospital employees when an acute case of measlesoccurs may be impractical.7 Thus, vaccination of all employ-ees under these circumstances seems appropriate.7,10 In par-ticular, measles vaccination status should be confirmed orupdated at the time of employment. Moreover, informationregarding healthcare personnel’s travel to areas where measlesis endemic should be regularly documented to allow evidence-based decisions about infection control policy to be made.



letters to the editor 381

acknowledgments


Nan-Yao Lee, MD; Hsin-Chun Lee, MD;Chia-Ming Chang, MD; Chi-Jung Wu, MD;

Nai-Ying Ko, RN, PhD; Wen-Chien Ko, MD

From the Division of Infectious Diseases (N.Y.L., H.C.L, C.M.C., C.J.W.,W.C.K.), the Department of Internal Medicine (N.Y.L., H.C.L., C.M.C.,C.J.W., N.Y.K., W.C.K.), Center for Infection Control, National Cheng KungUniversity Hospital, and the Departments of Medicine (H.C.L., W.C.K.) andNursing (N.Y.K.), Medical College, National Cheng Kung University, Tainan,Taiwan.

Address reprint requests to Wen-Chien Ko, MD, Department of InternalMedicine, National Cheng Kung University Hospital, No. 138, Sheng Li Road,704, Tainan, Taiwan ([email protected]).


references

1. Edmonson MB, Addiss DG, McPherson JT, Berg JL, Circo SR, Davis JP.Mild measles and secondary vaccine failure during a sustained outbreakin a highly vaccinated population. JAMA 1990; 263:2467-2471.

2. Mulholland EK. Measles in the United States. N Engl J Med 2006; 355:440-443.

3. Foulon G, Cottin JF, Matheron S, Perronne C, Bouvet E. Transmissionand severity of measles acquired in medical settings. JAMA 1986; 256:1135-1136.

4. Kaplan LJ, Daum RS, Smaron M, McCarthy CA. Severe measles in im-munocompromised patients. JAMA 1992; 267:1237-1241.

5. Infectious Disease Surveillance Center, National Institute of InfectiousDisease, Tokyo, Japan. Measles update in Japan as of week 21 (from 21to 27 May 2007). 2007. Available at: http://idsc.nih.go.jp/disease/mea-sles_e/idwr200721.html. Accessed February 10, 2008.

6. Welliver RC, Cherry JD, Holtzman AE. Typical, modified, and atypicalmeasles. Arch Intern Med 1977; 137:39-41.

7. Perlino CA, Parrish CM. Response to a hospitalized case of measles ata medical school affiliated hospital. Am J Med 1991; 91:325S-328S.

8. Weston KM, Dwyer DE, Ratnamohan M, et al. Nosocomial and com-munity transmission of measles virus genotype D8 imported by a re-turning traveller from Nepal. Commun Dis Intell 2006; 30:358-365.

9. Bolyard EA, Tablan OC, Williams WW, Pearson ML, Shapiro CN, Deitch-mann SD. Guideline for infection control in healthcare personnel, 1998.Hospital Infection Control Practices Advisory Committee. Infect ControlHosp Epidemiol 1998; 19:407-463.

10. Willy ME, Koziol DE, Fleisher T, et al. Measles immunity in a populationof healthcare workers. Infect Control Hosp Epidemiol 1994; 15:12-17.

Risk Assessment in Infection Control:Which Risks?

To the Editor—The focus of infection control professionals(ICPs) is on the control of infection risks. ICPs usually work

within a geographically defined setting, such as a hospital,with services organized to control risks within that definedsetting. ICPs have to consider both the risks associated withinfection and those associated with control strategies, whichmay themselves have a significant adverse impact on indi-viduals or groups. For example, isolation of hospitalized pa-tients may be associated with non–infection-related adverseconsequences.1

The importance of dimensions of well-being apart fromthose directly associated with infection is well illustrated byan example of an infection control dilemma posed in therecent article by Bryan et al.2(p1079) We are asked: “Should apostpartum woman being treated for a breast abscess due tomethicillin-resistant Staphylococcus aureus (MRSA) be al-lowed to visit her infant in a busy neonatal intensive careunit (NICU) in which MRSA has not yet emerged as a sig-nificant problem?” The risks include the potential for infec-tion to damage the infant’s health, to threaten the continu-ation of breastfeeding, and also to damage other dimensionsof well-being related to mother-infant attachment. These risksalso threaten other infants who may be in the NICU at thetime, as well as in the future, if MRSA becomes endemic.

If we take a very broad definition of health, such as thatof the World Health Organization (WHO)—a state of com-plete physical, mental, and social well-being and not merelythe absence of disease or infirmity—then risks related to in-fection, breastfeeding, and mother-infant bonding can beconsidered risks to health. Many would argue that the WHOdefinition is impractically inclusive (eg, Saracci3). Even so, ifwe consider that ICPs have a responsibility to consider theoverall well-being and interests of patients, then we should stilltake into account the risk of an adverse impact of controlstrategies on mother-infant bonding.

Some of the recently published work on public health ethics(eg, that of Powers and Faden4) has drawn attention to di-mensions of well-being outside of a narrow definition ofhealth, referring specifically to health, respect, attachment,personal security, reasoning, and self-determination. Nuss-baum5 has defined 10 capabilities derived from the question:“What activities are…definitive of a life that is truly human?”The list of capabilities comprises life (normal life span); bodilyhealth; bodily integrity; senses, imagination, and thought;emotions; practical reason; affiliation; relationships with otherspecies; play; and control over one’s environment. Nussbaum5

argues that we should give priority to ensuring that everyoneachieves a minimum standard of capability in all of thesedimensions.

The Nuffield Foundation has recently published guidanceon public health ethics in which they argue in favor of astewardship model, stating that, as stewards, we have a specialobligation to protect the most vulnerable.6(p144) The NuffieldFoundation defines vulnerability as “lacking capacity to makeinformed judgments for oneself, being socially or economi-cally disadvantaged, or…[having] other factors that contrib-




ute to a lack of autonomy.” Not only are infants in the NICUamong the most vulnerable, according to this definition, butthey are also extremely vulnerable with respect to both in-fection and the adverse consequences of infection. Almost allof the capabilities “definitive of a life that is truly human”(as defined by Nussbaum5) are at risk for infants in the NICUif infection spreads. Control strategies that limit mother-in-fant interactions also put these capabilities at risk, and thiscase illustrates the importance of considering all dimensionsof well-being both when trying to optimize control strategiesand when defining risks. Uncontrolled spread of MRSA hasthe potential to compromise other mothers and infantsthrough infection, through interference with breastfeeding,and through interference with attachment.

The relevance of the observation that “MRSA has not yetemerged as a significant problem” depends on the risk ofcross-infection to other infants. Currently, the emphasis inbiomedical ethics is to support individual rights, autonomy,and self-determination. Only when there is a threat of harmto others “can [power] be rightfully exercised over any mem-ber of a civilized community, against his will, to prevent harmto others.”7(p13) The case for restricting access of the motherto her baby is strengthened as the possibility of harm to otherinfants increases (eg, through the spread of MRSA).

The broad range of risks from infection to the well-beingof infants requires that we do what we can to minimize therisk of preventable infection. With adequate infrastructureand sufficient staff, it should be possible to lessen the degreeof risk of harm to others that is associated with allowing adegree of contact between a mother and her infant. Subop-timal levels of staffing and infrastructure are risk factors forincreased mortality in NICU, according to the UK NeonatalStaffing Group Study.8 Staffing and infrastructure in NICUs,even in affluent countries, may be below recommended stan-dards.9 It is striking that, in the United Kingdom, only 3.8%of NICUs are achieving national standards for nursing staffworking in the NICU.10

If the goal of the ICP is to control infection risks in amanner that best serves the overall well-being and interests of

patients (within that defined setting), then risk managementstrategies must take account of both infection and noninfec-tion risks to well-being. NICU infants illustrate the diversityof dimensions of well-being at risk from preventableinfection.

Michael Millar, FRCPath

From Barts and The London NHS Trust, London, United Kingdom.Address reprint requests to Michael Millar, FRCPath, Barts and The

London NHS Trust, 3rd Floor, Pathology and Pharmacy Building, 80 NewarkStreet, London E1 2ES, UK ([email protected]).


references

1. Stelfox HT, Bates DW, Redelmeier DA. Safety of patients isolated forinfection control. JAMA 2003; 290:1899-1905.

2. Bryan CS, Call TJ, Elliott KC. The ethics of infection control: philo-sophical frameworks. Infect Control Hosp Epidemiol 2007; 28:1077-1084.

3. Saracci R. The world health organisation needs to reconsider its definitionof health. BMJ 1997; 314:1409-1411.

4. Powers M, Faden R. The Moral Foundations of Public Health and HealthPolicy. Oxford, England: Oxford University Press; 2006.

5. Nussbaum MC. Frontiers of Justice: disability, nationality, species mem-bership. Cambridge, MA: Harvard University Press; 2006.

6. Public Health: ethical issues. London, England: Nuffield Council on Bio-ethics; 2007.

7. Mill JS. On liberty. In: Collini S, ed. On Liberty and Other Essays. Cam-bridge, England: Cambridge University Press; 1989:13.

8. UK Neonatal Staffing Study Group. Patient volume, staffing, and work-load in relation to risk-adjusted outcomes in a random stratified sampleof UK neonatal intensive care units: a prospective evaluation. Lancet2002; 359:99-107.

9. Standards for hospitals providing neonatal intensive and high dependencycare. 2nd ed. London, England: British Association of Perinatal Medicine;2001.

10. BLISS Web page. Special delivery or second class: are we failing specialcare babies in the UK? Available at: http://www.bliss.org.uk/pdfs/baby-report_updateweb.pdf. Accessed January 10, 2008.



risk assessment in infection control: which risks? •

Documents