61

Informacion Luminometria ATP Sector Salud

Embed Size (px)

Citation preview

Page 1: Informacion Luminometria ATP Sector Salud
Page 2: Informacion Luminometria ATP Sector Salud

Rachel Scott Unit 11 Wenta Business Centre Colne Way Watford Hertfordshire WD24 7ND

Manchester PCTMauldeth House

Mauldeth Road WestChorlton

ManchesterM21 7RL

Tel: 0161 958 4003Fax: 0161 958 4040

Email: [email protected]

27th March 2009 We at Manchester PCT use the Hygiena systemSURE to prove that our high standards of cleaning have been met in our Health Centre, Clinics and Care homes. The Hygiena systemSURE is simple to use and provides powerful information for our Facilities Team, Site Managers and of course our cleaning contractors. We generate reports for our infection control team and trust board. It is cost effective, robust and reliable and the extensive support and training we have received has been excellent. Yours sincerely Peter Kevan Contract Manager NHS Manchester PCT

Page 3: Informacion Luminometria ATP Sector Salud
Page 4: Informacion Luminometria ATP Sector Salud
Page 5: Informacion Luminometria ATP Sector Salud
Page 6: Informacion Luminometria ATP Sector Salud

Untitled Document http://www.hygienausa.com/news/6.23.08.html

1 de 1 27/09/2010 12:20 p.m.

Page 7: Informacion Luminometria ATP Sector Salud

SP

EC

IAL

RE

PO

RTHow clean is CLEAN?

New technologies, monitoring practices gain traction

The important link be-tween environmentalcleanliness and infec-

tion prevention has long beenappreciated, but how best toachieve these objectives re-mains a source of ever-chang-ing science and application.So how are hospitals faring in this all-important area andhow are they responding tothe rapid changes in cleaningtechnologies, processes andverifying proper cleaningprocedures are followed?

More than fourout of five are

either already using or plan-ning to use advanced clean-ing technology such as mi-crofiber mops. Nearly one in four are augmenting ob-servation-based audits andperformance goals with new-er technologies such as rapidenvironmental testing withchemical markers that fluo-resce with ultraviolet light, or through environmentalcultures. And many are turning to environmentallyfriendly cleaning products,even as they’re requesting ev-idence-based data to confirmthe value of these products.

These are among the manyfindings of a recent survey of environmental servicesmanagers and infection pre-ventionists. The online surveywas co-sponsored by HealthFacilities Management andMaterials Management in

W W W. H F M M A G A Z I N E . C O M | A U G U S T 2 0 0 9 | 2 3

John Scherberger,director of environ-mental services,pastoral care andguest services atSpartanburg (S.C.)

Hospital forRestorative Care,is an advocate forteamwork between

environmentalservices and infec-

tion prevention.

2009 Infection Prevention+Hospital Cleaning Survey

ARTICLE BY GINA ROLLINS • DATA BY SUZANNA HOPPSZALLERN • PHOTOGRAPHS BY CARROLL FOSTER

Page 8: Informacion Luminometria ATP Sector Salud

Health Care magazines, along with the American Society forHealthcare Environmental Services (ASHES) and the Associa-tion for Professionals in Infection Control & Epidemiology(APIC). The survey underscored that close collaboration be-tween environmental services managers and infection preven-tionists is essential to achieving and monitoring performancestandards for cleanliness of the patient environment, and forintroducing new technologies and practices.

“It’s really a team effort. We can’t operate in silos,” says JohnScherberger, director of environmental services, pastoral careand guest services at Spartanburg (S.C.) Hospital for RestorativeCare. “I have an excellent working relationship with our infec-tion preventionist and nurse managers.”

“The good working relationship between infection preven-tionists and environmental services showed all the way through

the [respondents’] comments,” ob-serves Judene Bartley, vice presidentof Epidemiology Consulting Servicesin Beverly Hills, Mich. Bartley wasthe APIC-appointed expert for thesurvey.

“The survey data supports whatASHES has long believed,” says PattiCostello, ASHES executive director.“Environmental services and infec-tion preventionist professionals areessential partners in providing aclean, safe, comfortable and high-quality patient care environment.

“Pathogenic microorganisms aremore resistant and persistently pres-ent in the environment, staffingbenchmarks for infection prevention-ists and front-line cleaning techni-cians are problematic, the Centersfor Medicare & Medicaid Services(CMS) and insurers are redefining reimbursement and dramatic healthcare reform is on the horizon,”Costello adds. “The combined intel-

lectual capital and collabora-tive efforts between ASHES,APIC and other organizationswill better position our appealfor improved staffing ratios,widely accepted staff compe-tencies and health care clean-ing/disinfection protocols,training, education and qualityassurance as well as sound sci-ence behind the claims of newproducts and technologies.”

The connection between environmental services and in-fection prevention is importantin evaluating new products,training and educating staffabout good hygiene practices,and even in eliminating prod-ucts or procedures.

That was the case at St.Joseph Mercy Hospital in Ann Arbor, Mich. “Our staff asked tostop using disinfectant on the floors because they were develop-ing a cloud on the surface and environmental services was hav-ing to strip and clean them more often,” reports Russ Olmstead,epidemiologist. “We did a literature review and found no evi-dence of increased health care-acquired infections from using de-tergents instead of disinfectants on the floor, so we allowed themto make the change.” The updated Centers for Disease Controland Prevention (CDC) Guideline for Disinfection and Sterilizationin Healthcare Facilities, which can be accessed at www.cdc.gov/ncidod/dhqp/pdf/guidelines/Disinfection_Nov_2008.pdf, also acknowledges that the use of germicidal chemicals to disinfecthospital floors and other noncritical items is controversial.

Monitoring cleanlinessObservation-based audits continue to be the mainstay for monitoring compliance with cleaning standards, with nearly 87 percent of respondents using that method. However, manyhealth care institutions are finding ways to tighten observationprotocols. For instance, Brigham and Women’s Hospital inBoston maintains an electronic database of all rooms that environmental services supervisors access with handheld devices. This helps achieve the quality assurance goal of twoinspections per employee per month.

“When a supervisor takes the handheld, it shows all the em-ployees what that person is responsible for and randomly identi-fies the rooms to be inspected so we don’t have a selection bias,”explains Richard Bass, director of environmental services. Thesystem also has a results reporting feature that can provide sum-maries by employee, supervisor, time period and item inspected.

Other respondents couple observation-based audits with de-tailed check lists to set and measure performance expectations.

Pinckneyville (Ill.) Community Hospital is an example. “I devel-oped a task list that covers basically every item and surface in theroom. They vary depending on whether a patient is in the roomduring the cleaning, staying in the room but not present duringthe cleaning, or discharged,” says Kevin Daugherty, environmen-tal services manager. “We use this for our performance improve-ment standards and supervisors sign off on the lists when theycheck each room.”

2 4 | A U G U S T 2 0 0 9 | W W W. H F M M A G A Z I N E . C O M

Thoroughly cleaning such items as monitors and TV remotes helps prevent health care-associated infections. In the farright photo, a cleaning agent is applied to a microfiber cloth that is then used to clean, which is proper cleaning protocol.

»2009 INFECTION PREVENTION +HOSPITAL CLEANING SURVEY

About thesurvey…

H ealth Facilities Manage-ment, Materials Manage-

ment in Health Care, theAmerican Society for Health-care Environmental Services(ASHES) and the Associationfor Professionals in InfectionControl and Epidemiology(APIC) teamed up to conductthe Infection Prevention &Hospital Cleaning Survey.

A random sample of3,538 infection prevention-ists were contacted to findout what steps hospitals aretaking to optimize cleaningprotocols and reduce hospi-tal-acquired infections. Thesurvey response rate was19.7 percent or 696 com-pleted surveys.

Page 9: Informacion Luminometria ATP Sector Salud

W W W. H F M M A G A Z I N E . C O M | A U G U S T 2 0 0 9 | 2 5

SP

EC

IAL R

EP

OR

T

SOURCE: MMHC/HFM/APIC/ASHES 2009 INFECTION PREVENTION & HOSPITAL CLEANING SURVEY

Top six ways hospitals measure compliance with cleaning standards in patient care areas

Observation-based audit

Patient satisfaction scores on cleanliness of room

Monitor compliance with performance targets for patient area cleaning

Risk-based audit

Environmental culture results

Measure cleaning rates of high-risk objects in patient areas

87%

78%

34%

15%

14%

14%

84%

81%

Step hospitals have taken to optimize environmental services staff performance

Hands-on training in cleaning protocols

Education on transmission of health care-associated pathogens and resultant infection

Ongoing performance feedback

Predefined performance targets for patient area cleaning

Patient interviews by supervisory staff

Well-defined quality management program for patient area cleaning

Use of visually observable feedback tool (e.g., black-light marker, ATP)

Quality control assessments tied to compensation

62%

31%

27%

24%

20%

10%

Cleaning practices and technologies hospitals routinely employ to disinfect patient rooms

Sodium hypochlorite, household bleach

Quaternary ammonium disinfectant

Disinfectant-impregnated wipers

Microfiber cloths

Microfiber mops

Copper and copper-alloy fixtures

Pour bottles to dispense disinfectant

Change cubical curtains after discharge of patients placed under contact precautions

Hydrogen peroxide vapor decontamination system

68%6%

85%1%

77%3%

46%17%

68%

4%

14%

2%

42%4%

57%13%

2%5%

Hospitals using chemicals (e.g., ATP, fluorescing markers) to verify cleaning of the following high-risk objects

Bed rail

Tray table

Nurse call device

Bedside table

Bathroom doorknobs

Toilet seat

Patient telephone

Sinks

Toilet handle

Patient room doorknobs and cabinet pulls

Bathroom light switch

Restroom grab bars

16%

16%

16%

15%

15%

14%

14%

13%

15%

15%

14%

14%

Currently use

Planning to use

Top 11 challenges to cleaning and disinfectionof the patient environment

Pressure to expedite room turns for incoming patients

Assigned responsibility for cleaning mobile objects

High hospital occupancy

Inadequate time to properly clean patient rooms and care areas

Reluctance to clean electronic equipment with saturated cloths

Inadequate staffing levels

Too busy/insufficient time allowed to consistently follow protocols

High turnover rates among environmental services technicians

Inadequate financial resources to invest in cleaning technologies and equipment

Lack of objective microbiologic standards for hospital cleaning

Lack of knowledge of the role specific high-risk objects play in transmitting health care-associated pathogens

42%

41%

35%

32%

32%

31%

28%

26%

26%

20%

20%

Page 10: Informacion Luminometria ATP Sector Salud

Nearly a quarter of the participants reported augmenting observation-based audits and performance goals with newertechnologies such as rapid environmental testing with chemicalmarkers that fluoresce with ultraviolet light, or through envi-ronmental cultures. The introduction of chemical markers inparticular can have a profound effect on cleaning standardsand protocols, according to Bartley.

“It’s a more objective way of determining how well cleaning istaking place, and it can be very dramatic. It’s a shock to find outthat things you thought were clean in reality are not,” she explains.These products are just beginning to gain traction in health care,Bartley adds, and it will be interesting to track their use and im-pact on cleaning in future surveys. (See chart on page 25.)

Just how quickly this area of infection prevention is changingwas evident during the Society for Healthcare Epidemiology of America’s (SHEA) annual scientific session in March. Resultsof a federally funded study were presented, concluding thatrisks of methicillin-resistant Staphylococcus aureus (MRSA) andvancomycin-resistant enterococcus (VRE) transmission could be lessened by immersing cleaning cloths in cleaning solution,educating workers and providing feedback on the removal ofintentionally applied marks visible only under ultraviolet light.

Regardless of the specific compounds used, these visually observ-able feedback tools make for excellent training tools, particularlywith staff who may not be fluent in English. “It immediately sends a message, no matter what language the individual is proficient in,that what they thought they removed is still there,” notes Bartley.

Adopting new technologiesThe survey also revealed that many hospitals are incorporatingthe latest technologies to achieve and monitor performancestandards for cleanliness of the patient environment.

For example, more than 80 percent of respondents indicated

that they already use or plan to use microfiber mops, while 63 percent either already use or plan to use microfiber cloths(see chart on page 25).

Organizations that have switched report multiple benefits.Spartanburg Hospital, a long-term acute care facility, transi-tioned to microfiber mops and cloths about four years ago.“Since then, we’ve had no workers’ compensation injuries, noslips and falls from wet floors, and our waste-water treatmenthas decreased by hundreds of thousands of gallons,” saysScherberger.

Other survey respondents also gave the technology athumbs-up. “It saves labor because it picks up more. The employee still uses a figure-eight motion, but you get a cleanersweep,” explains Bobby Dorsett, director of environmental operations at Scripps Memorial Hospital La Jolla (Calif.).

Microfiber products also minimize cross-transmission of microbes among patients, according to Cecilia DeLoach Lynn,senior manager of sustainable operations for Practice Green-health, a membership and networking organization for institu-tions committed to eco-friendly practices. “Because they don’trequire re-dunking in a mop bucket and using the same wateror mop for two or three rooms without changing, they reallycut down on cross-contamination,” she says.

Copper or copper-alloy fixtures, a technology touted for itsantimicrobial properties, has yet to gain traction among surveyrespondents. Less than 4 percent reported using the fixturesnow, and nearly 94 percent indicated they have no plans to implement the technology. “In laboratory studies copper hasbeen shown to have antimicrobial properties, but it’s a chal-lenge to take lab findings and apply them in a natural healthcare environment,” observes Olmstead. “It’s an interesting technology that’s not quite ready for prime time.”

On going greenHospitals are beginning to implement environmen-tally preferred cleaning initiatives, but the healthcare industry is advancing cautiously for good reasons, according to Bartley.

“We’ve been slower to adopt the products becausewe know there are certain chemicals we have touse,” she says. “Until there are safer materials andmore science behind their use, we’re left with infec-tion preventionists doing risk assessments to deter-mine where it’s safe to use [environmentally friend-ly] materials and where it’s required to use harsherones.” (See sidebar at left.)

The most common environmentally preferred ac-tions among survey respondents included orienta-tion and training about the hazards, use and dispos-al of cleaning products (64 percent), policies on lim-iting exposure to chemicals (54 percent) and onenvironmentally preferable cleaning (57 percent),use of pre-diluted disinfectant systems (53 percent)and use of equipment that doesn’t negatively impact indoor air quality (52 percent).

At Brigham and Women’s Hospital, the environ-mental safety department vets all new products,and use of the more toxic chemicals is limited tostaff members with special training. In addition, the hospital uses fewer products than in the past.“The vast majority of our staff is of low skill-level

2 6 | A U G U S T 2 0 0 9 | W W W. H F M M A G A Z I N E . C O M

»2009 INFECTION PREVENTION +HOSPITAL CLEANING SURVEY

Putting ‘green’in perspective

A s hospitals seek ways to tran-sition to environmentally pre-ferred practices and solutions,

teamwork between infection prevention-ists and environmental staff must cometo the fore, according to Cecilia DeLoachLynn, senior manager of sustainable oper-ations for Practice Greenhealth, an organ-ization promoting eco-friendly practices.

“They need to look at effectivenessfirst before putting on the green filter,”she says. “Being sure of all the dataand evidence takes close collaboration.”

Lynn is encouraged by the responseson environmentally preferred practices inthe Infection Prevention & Hospital Clean-ing Survey, particularly the use of mi-crofiber mops and cloths. “We’ve seenthat technology really take hold in a lot offacilities,” she says. “It’s a win-win for in-fection prevention and the environment.”

Practice Green-health encouragesthe use of prod-ucts that have

been certified asgreen. “We recom-mend third-party certifi-

cation of green cleanersbecause it holds the product to a stan-dard and worker safety is ensured,” sheexplains. “We recognize that it might bedifficult for small companies in the inno-vation stage to afford the cost of obtain-ing green certification, but unfortunatelyothers will slap a green label on a prod-uct that’s not really green.” Two organiza-tions that certify health care products asgreen include Green Seal in the UnitedStates and the EcoLogo program basedin Canada.

For facilities seeking to adopt environ-mentally preferred technologies, more in-formation is available through the Website of the Green Guide for Health Careat www.gghc.org. n

Page 11: Informacion Luminometria ATP Sector Salud

SP

EC

IAL R

EP

OR

Tand we’ve dropped the numberof products we work with toabout five or six versus dozensI’ve seen used in other hospi-tals,” Bass says. “Our bottles arecolor-coded and bilingual.”

Like Brigham and Women’s,many facilities reported stream-lining the number of productsused and minimizing worker exposure to chemicals.

For instance, SpartanburgHospital for Restorative Carenow uses only four products,and all are green-certified exceptthe disinfectant. “All our chemi-cals are in a dispenser and envi-ronmental services staff does notget involved in the mixing,” saysScherberger. “They press one button for a lot of the chemical,another for less, and I can check that their bottles are the prop-er dilution based on the color. We also stopped using triggerspray bottles. They’re all pull-top so housekeepers saturatetheir cloths rather than spraying in the environment.”

Some facilities, including Curry General Hospital in GoldBeach, Ore., have made the switch to all sustainable products.“We already used products from this distributor, so we re-searched their green products to make sure they killed MRSAand the like and would not be irritating to our staff—we have alot of people with allergies,” explains Kim Sharp, environmen-tal services manager.

Challenges and opportunitiesEven as hospitals have made strides in implementing new tech-nologies and practices, they still face challenges in achievingperformance improvement objectives. Of respondents, 42 per-cent cited high hospital occupancy rates and the need to expe-dite room cleaning for new patients as major challenges tocleaning patient care areas, and they have employed a numberof mechanisms to address this ongoing dilemma.

St. Joseph Mercy Hospital redeployed environmental servicesstaff from nonpatient care areas, and asked staff in those loca-tions to pick up the slack by taking on tasks normally left toenvironmental services such as emptying trash and recyclingbins, according to Olmstead.

Scripps Memorial uses a teletracking system to keep nursing,environmental services, emergency and bed control staff up-to-date on the status of rooms. “It gives everyone information onunoccupied beds in real time,” says Dorsett. When a bed be-comes unoccupied, nursing staff punch in codes via the tele-phone, which sends a page to appropriate staff. Environmentalservices staff use the same system to notify other departmentswhen they begin and end the cleaning of each room. “This way,everybody at any given moment knows how many rooms areavailable, and how many are in the process of being cleaned. Ithelps with the anxiety around turnaround times,” he explains.

Perhaps the most crucial element of achieving performancestandards for cleanliness of the patient environment is theoverall approach taken in training and supporting environmen-tal services staff.

“People can be trained to mop a floor, but if they don’t under-

stand the reasoning behindit, you won’t get the effortneeded,” notes Daugherty.His staff receives ongoing,detailed education about thesource of microbes and whatit takes to clean surfaces ap-propriately. At SpartanburgHospital, environmentalservices employees are as-signed to specific units andattend staff meetings forthose areas. “They’re part ofthe team, and that providespositive reinforcement,”Scherberger explains. “Envi-ronmental services has to understand they’re just as important as any other part

of the team. We’re the first line of defense in infection control.”

Testing for cleanlinessThe survey also revealed that the use of rapid environmentaltesting with substances that fluoresce under UV light is juststarting to be adopted in health care. However, experts predictthat adoption will become widespread soon.

While some technologies, such as adenosine triphosphate

W W W. H F M M A G A Z I N E . C O M | A U G U S T 2 0 0 9 | 2 7

John Scherberger, director of environmental services, pastoral care andguest services at Spartanburg (S.C.) Hospital for Restorative Care, trains

two environmental services personnel on how to test for cleanliness.

Page 12: Informacion Luminometria ATP Sector Salud

(ATP) bioluminescence meters, have been used for years to detect microbes in the food industry, the science behind themhas only been applied recently to the patient care setting.

A leader in the field is Philip Carling, M.D., clinical professorof medicine at Boston University. Carling sought a way to improve on standard methods of verifying the cleanliness ofpatient care areas, primarily observation-based audits and lesscommonly, culturing of selected surfaces. He came up with amixture that people jokingly refer to as “Carling’s Goo,” whichis a transparent, easily cleaned, environmentally stable mark-ing solution that fluoresces when exposed to UV light.

Carling and a consortium of participating hospitals havedemonstrated in a series of studies that certain high-risk ob-jects such as tray tables, toilet handles, bathroom light switchesand bathroom and room doorknobs typically are not well-cleaned. Research shows that out of 14 high-risk objects, the lowest mean rates of cleaning were found for toilet door-knobs and toilet handholds (both 28 percent), bedpan cleaners (25 percent), room doorknobs (23 percent) and bathroom lightswitches (20 percent).

“Our findings are consistent that, on average, critical objectsare only being cleaned about 50 percent of the time, and thatthere are several objects that are almost universally overlooked,”Carling explains. “We’ve uncovered a systemic problem with thethoroughness of environmental health in hospitals because therewas not a system for simple measurement in the past.”

Subsequent research has shown that with a focused perform-ance improvement initiative, including education and feedback

to staff, the average rate of cleaning can be increased from lessthan 50 percent to about 77 percent. “Once the cleaning staffunderstands they’re critical to patient safety and are shownthey’re not doing well in certain areas, that’s when the para-digm shift occurs,” says Carling.

Use of a solution like Carling’s can be an eye-opener even forfacilities that have overall high rates of cleaning compliance,like Nebraska Medical Center in Omaha, which participated in Carling’s studies.

“The program is fantastic. It’s helped us identify various opportunities for improving our processes,” says Mark Rupp,M.D., professor of infectious diseases at the University of Nebraska and director of epidemiology at Nebraska MedicalCenter. “We found there were objects and surfaces that no onewas taking responsibility for cleaning. Many were electronicsuch as a computer mouse. This enabled us to go item-by-itemand figure out who would be responsible for cleaning them.”Rupp also is president of SHEA.

The utility of fluorescent marking is obvious even to facilitiesjust beginning to use the solutions such as Scripps Memorial.“That’s really exciting news, and it’s going to revolutionize theindustry,” Dorsett says. “We’ll be able to tell what’s clean andwhat’s not.” HFM

Gina Rollins is a Silver Spring, Md.-based freelance writer who frequently cov-

ers health care topics. Suzanna Hoppszallern is senior editor of data and re-

search for Health Facilities Management’s sister publication, Hospitals &

Health Networks.

2 8 | A U G U S T 2 0 0 9 | W W W. H F M M A G A Z I N E . C O M

»2009 INFECTION PREVENTION +HOSPITAL CLEANING SURVEY

Determined to control infection, healthcare

providers have a secret weapon. It’s under

their feet and where it matters. Vinyl.

For over 50 years, medical facilities have

relied on this remarkably durable mate-

rial to create a safe, healing environment.

Used in flooring, wall coverings and more,

this impervious surface is easy to clean

and disinfect, without compromising

affordability and aesthetic appeal. For

buildings that must perform 24/7, there’s

only one choice—vinyl.

VINYL MATTERS.Learn more at www.vinylindesign.com/hc

Page 13: Informacion Luminometria ATP Sector Salud

About Cleanroom Technology | Contact us | Newsfeed

Home News Technical Articles Companies & Products Events Jobs & People

Areas

Cleaning agents

Clothing and PPE

Construction

Containment

Decontamination

Furniture andEquipment

Healthcare

Hospital acquiredinfections

HVAC

Hygiene

Medical devices

Microbiology

Nanotechnology

Pharmaceuticals

Regulation

Sterilisation

29-Feb-2008

Professor Chris Griffith

To sample or not to sampleCleanrooms | Monitoring | Hospital acquired infections | Healthcare | Foodmanufacturing |

The food industry believes the contamination of the inanimate food processingenvironment is likely to lead to the same organisms contaminating the foods being

produced

Can the health sector learn from the food industry when it comes toimproving hygiene? Professor Chris Griffith, head of the Food Research &Consultancy Unit at the University of Wales Institute, argues yes

Healthcare associated infections (HCAIs) and the role of cleaning in their preventionhave, and continue to receive a high level of media attention. However, the link

between environmental cleanliness and HCAIs is the subject of debate.1,2,3 The UK

guidelines for infection control4 recommend that the hospital environment “must bevisibly clean” with routine bacteriological sampling rarely indicated unless there is

an outbreak of infection.5 The rationale behind this approach is based on theassumption that the inanimate hospital environment is of “little importance in the

spread of endemic infections but may occasionally have a role in outbreaks”.5

This belief and approach is in direct contrast toviews held by the food industry, which believesthe contamination of the inanimate foodprocessing environment is important and is highlylikely to lead to the same organisms

contaminating the foods being produced.6,7 Thefood industry in the 1980s and 1990s attractedthe type of adverse media attention concerningcleanliness and risk management currently beingreceived by hospitals.

In response, the food industry put in place arange of corrective risk management strategiesincluding Hazard Analysis Critical Control Points(HACCP), which could be combined with

independent third party auditing in relation to standards and practices.8 While thefood manufacturing industry is not perfect, hygiene standards and practices haveimproved as a result of the measures taken.

Although not identical, there are interesting parallels in the survival, transfer andspread of pathogens in both food and healthcare environments. It has been argued

in healthcare3 that it is the patients that contaminate the environment and not thereverse. If transmission one way is possible, it is difficult to believe the reversecannot take place and studies have shown that reducing environmental

contamination can lead to a reduction in infection rates9 and that contaminants from

the environment can contaminate patients and ultimately lead to an infection.10

But the problem is complex and likely to involve many additional factors, includinghand hygiene compliance rates, patient movement and bed occupancy levels.

The potential for an environmental pathogen to contaminate foods is well recognisedand has led to a much more organised and scientific approach to environmental

monitoring.6 Environmental monitoring (see Table 1) is used in the food industry forplant commissioning, validation and bench marking of cleaning methods, as well asthe more routine monitoring and verification of cleaning and is predicated on thebelief that isolation of pathogens from the environment is a cause for concern.

Another difference between the monitoring of cleaning in food and healthcareenvironments is how it is undertaken (Table 1). It is highly unlikely a major foodprocessor would rely solely on visual measurement of cleanliness and would, forBritish Retail Consortium (BRC) certification, be required to have documented

Related Articles

A simple guide to how aerosol particle counterswork

Particle loss in transport tubing

Get calibration up to ISO standard

Energy management

The devil is in the details

Related News

Merck opens €50m Material Research Centre

BM Polyco acquires medical supplies companyBodyguards

Bioquell invests for future growth

AsepticSure virtually eliminates food-bornepathogens

Vaisala probes monitor humidity in criticalenvironments

Products and Press

UKAS calibration accreditation for PMT (GB) Ltd

Multi-talent for biocontamination control

Thermo Fisher Scientific demonstratesaccelerated method development for food safetyand environmental testing

New Alcorub sanitising gel

Introducing TimestripPlus

Contamination control, design, equipment and supplies

To sample or not to sample http://www.cleanroom-technology.co.uk/technical/article_page/To_samp...

1 de 3 27/09/2010 12:15 p.m.

Page 14: Informacion Luminometria ATP Sector Salud

evidence of a monitoring programme and its results.

Table 1: Comparison of approaches to monitoring cleaning efficacy betweenfood and healthcare industries

FOOD ENVIRONMENT HEALTHCARE ENVIRONMENT

Isolation of pathogens from environmentalsurfaces causes concern

Isolation of pathogens fromenvironmental surfaces may or maynot cause concern

Environmental surface sampling in foodmanufacturing used as part of a preventativestrategy

Environmental surface sampling likelyto be used only in response to anoutbreak

Range of surface sampling techniques used,including visual, microbiological and rapidmethods in coordinated and integratedapproaches

Assessment of cleaning efficacydominated by visual inspections(ICNA, PEAT, Healthcare Commission)

Monitoring cleanliness in the food industry starts with visual assessment (if visuallydirty there is no point in other forms of measurement), but may then involve themeasurement of residual soil using adenosine triphospahte (ATP). ATP is found inliving cells and cell debris and is widely used in the food industry.6 While used bythe top five world food manufacturing companies it has had only limited use in

healthcare.11,12,13 These studies have shown clearly that visual assessment alone isan inaccurate measure of cleanliness and produces an verly optimistic indication ofcleaning efficiency. Such findings should question the value of the development of

elaborate methods of assessing cleanliness based solely on visual assessment.14

ATP provides a very rapid (10-20 seconds) measure of residual surface soil i.e.cleanliness, if cleaning is defined as the removal of soil. ATP can be used on its ownand as a preliminary step prior to microbiological sampling. ATP and microbiologicalassessments measure different things (the latter measures only residual viable

organisms). Although sometimes undertaken,15 there is therefore little value intrying to directly correlate one with another.

For a strong correlation to exist, the ratios between organic debris and micro-organisms would need to be constant and there are many reasons why this might

not be the case.6 Of much greater benefit is comparing the results obtained usingboth approaches in relation to pass/fail limits following properly implemented

cleaning.12,13 Surfaces passing ATP are much more likely to correlate with andachieve microbiological limits than those assessed visually (Table 2).

Table 2: Surfaces passing visual assessment: Differences in failure ratesafter cleaning, between visual and two other assessment methods

ATP (% failure rates) ACC* (% failure rates)

Hospital A

Paediatric 77 45

Surgical 82 51

Hospital B

Paediatric 93 62

Surgical 83 84

Hospital D

Paediatric 77 73

Surgical 91 77

*ACC= Aerobic Colony Count

There is no ideal method for monitoring cleaning efficiency and in reality, as theymeasure different things ATP and microbiology should be used together as part ofan integrated cost-effective assessment strategy.6 This considers not just the costsof testing but the costs of cleaning (which can be considerable and consider stafftime, chemicals/equipment costs etc) as well as failure costs. Cleaning shoulddeliver value for money (quality in relation to cost), and ATP can help to achieve

this.6

However, not all ATP instruments are the same and some are more sensitive andreproducible than others and benchmark values obtained with one instrument/testare not applicable to values from other instruments/test combinations. Benchmarkvalues after cleaning have been proposed and can be used in routine monitoring toimprove the management of cleaning and in adopting a more scientific approach todecisions on cleaning frequency, as well as the validation of new cleaning methods.

Evidence is growing that poor cleaning can result in increased environmental surface

contamination and this may contribute to pathogen reservoirs16,17 and some casesof nosocomical infections caused by specific pathogens. An improved, more scientificassessment strategy may help to achieve better value for money from cleaning inhealthcare as with the potential benefit of a reduction in infection rates caused bysome pathogens.

References

1. Boyce JM. J Hosp Infect 2007; 65 (Supp2): 50-54. doi:10.10.1016/S0195-6701(07)60015-2

2. Dettenkofer M, Spencer RC. J Hosp Infect 2007; 65 (Supp2): 55-57.doi:10.1016/S0195-6701(07)60016-4

3. Fraise AP. J Hosp Infect 2007; 65 (Supp2): 58-59.doi:10.1016/S0195-6701(07)60017-6

4. Pratt RJ, Pellowe CM, Wilson JA, et al. J Hosp Infect 2007; 65 (supp 1): 1-29.doi: 10.1016S0195-6701(07)60002-4

5. Ayliffe GAJ, Fraise AP, Geddes AM, Mitchell K. Control of Hospital Infections: a

To sample or not to sample http://www.cleanroom-technology.co.uk/technical/article_page/To_samp...

2 de 3 27/09/2010 12:15 p.m.

Page 15: Informacion Luminometria ATP Sector Salud

Read all about it

Get further insightful and informative articles like this one sentto you every month with a great value subscription toCleanroom Technology.

practical Handbook. 4th edn. London: Hodder Arnold; 2002.

6. Griffth CJ. Improving surface sampling and detection of contamination, Handbookof Hygiene Control in the Food Industry. Cambridge: Woodhead Publishing; 2005.

7. Cordier JL. Assessing Microorganisms in Food and Factory. 3rd IAFP EuropeanSymposium on Food Safety. Rome: October 2007.

8. British Retail Consortium. Global Standard for Food Safety. Issue 5. TSOStationary Office.

9. Denton M, Wilcox MH Parnell P, et al. J Hosp Infect 2004; 56: 106-110.

10. Hardy KJ, Oppenheim BA, Gossain S, Gao F, Hawkey PM. Infect Cont HospEpidemiol 2006; 27: 127-132.

11. Griffith, C.J., Cooper, R.A., Gilmore, J, Davies, C and Lewis M. (2000). Journal ofHospital Infection. 45(1): 19-28.

12. Griffith C.J., Obee P., Cooper R.A., Burton H.F., Lewis M (2007) Journal ofHospital Infection. June 66(4): pp352-359

13. Cooper R.A., Griffith C.J., Malik R.E., Obee P. and Looker N (2007) AmericanJournal of Infection Control. 35 (5): pp 338-341

14. National Patient Safety Agency Report. The National Specifications forCleanliness in the NHS: a framework for setting and measuring performanceoutcomes. NHS; April 2007.

15. Department of Health. Evaluation of ATP bioluminescence swabbing as amonitoring and training tool for effective hospital cleaning. Crown Copyright; 2007.

16. Sattar SA. Journal of Hospital Infection. 56, Supplement 2, April 2004: pp64-69

17. Sexton T, Clarke P, O'Neill E, Dillane T and Humphreys H (2006) Journal ofHospital Infection, 62(2): 133-260, February.

About Cleanroom Technology About HPCi Media Limited Terms and Conditions Privacy

To sample or not to sample http://www.cleanroom-technology.co.uk/technical/article_page/To_samp...

3 de 3 27/09/2010 12:15 p.m.

Page 16: Informacion Luminometria ATP Sector Salud

Silliker, Inc., Food Science Center Report RPN: 13922

December 11, 2009

Revised January 21, 2010

Performance Evaluation of Various ATP Detecting Units

Prepared for:

Steven Nason Hygiena

941 Avenida Acaso Camarillo, CA 93012

Tel: 805-388-8007 [email protected]

Prepared by:

Brian Kupski

Operations Supervisor [email protected]

Erdogan Ceylan, Ph.D.

Research Director [email protected]

Cynthia Stewart, Ph.D.

General Manager [email protected]

The entire content of this REPORT is subject to copyright protection. Copyright © 2009-10 Silliker, Inc. All rights reserved. The contents of this REPORT may not be copied other than for use by non-for-profit organization, and appropriate reference with all copyright notices stated. The REPORT may not be copied, reproduced or otherwise redistributed. Except as expressly provided above, copying, displaying, downloading, distributing, modifying, reproducing, republishing or retransmitting any information, text or documents contained in this REPORT or any portion thereof in any electronic medium or in hard copy, or creating any derivative work based on such documents, is prohibited without the express written consent of Silliker, Inc. Nothing contained herein shall be construed as conferring by implication, estoppel or otherwise any license or right under any copyright of Silliker, Inc., or any party affiliated with Silliker, Inc.

Copyright © 2009-10 Silliker, Inc.

1

Page 17: Informacion Luminometria ATP Sector Salud

Executive Summary In early 2009 Silliker Group Corporation was commissioned to perform an extremely comprehensive study to examine the performance of several commercially available ATP (Adenosine Triphosphate) bioluminescence hygiene / sanitation monitoring systems. The properties of the ATP assay are well suited to determinations of cleanliness, with cleanliness being defined as the absence of organic (derived from life) material. Clean surfaces have little to no ATP, while dirty surfaces have ATP and perhaps live microbial cells. The result of an ATP test that is available in minutes permits the immediate assessment of the sample condition and whether additional cleaning action is required. The traditional method of determining cleanliness is the aerobic plate count. This test requires 2 days to complete. This procedure is limited in the type of microorganisms it can detect and does not detect organic residue.

The study was designed to assess system performance when challenged with varying levels of pure ATP, food residues and microbes. There were 4 key study phases. Each phase of the study was designed to measure system performance under specific conditions against parameters chosen to approximate real world environmental situations and are vital technical measurements. It is important to understand that the combination of these factors and no single factor alone can clearly define system overall performance. Not all study phases include all systems. 1. Detection of pure ATP

Swabs and solution

2. Detection of ATP from pure microbial cultures

Escherichia coli, Lactobacillus plantarum, Pseudomonas aeruginosa, Salmonella

Typhimurium and Staphylococcus aureus and one yeast culture- Saccharomyces

cerevisiea

3. Detection of ATP from food

Ground beef, Milk (pasteurized 2% low fat, Orange Juice (pasteurized without

pulp),Salad (bagged mixed salad greens)

4. Detection of ATP from food soiled stainless steel surfaces

Ground Beef, Milk

At each stage of the study and for each combination of factors, 10 replicate sub-samples were tested at each test parameter. A total of >5000 data points were generated and these were analyzed mathematically to describe the performance of the tests in terms of Linearity, Repeatability, Sensitivity, and Accuracy. A key factor in the study of performance was the use of clone reagent swabs. These are reagent swabs that are formatted to operate with multiple instruments but are not proprietary to the instrument manufacturer. The study data indicates that the use of clone swabs is acceptable. Due to differences in RLU scales, data output (pass / fail settings based on RLU scale) and

Copyright © 2009-10 Silliker, Inc.

2

Page 18: Informacion Luminometria ATP Sector Salud

instrument linearity, the user must understand individual system nuances to successfully convert from a proprietary reagent swab to a clone. The key findings of the study are

1. 1. The linear correlation coefficient values for the ATP solution, microorganisms, and foods showed that the Log10 RLU readings and Log10 dilutions were linearly correlated. All Linear coefficient calculations can be found in tables 7 and 8 of the final report and were generally >0.9. However at low ATP levels , the Neogen and Charm systems lost linearity and could not detect <10 fmols ATP.

2. To quantify repeatability, the coefficient of variation (CV%) for each dilution of the ATP solution, microorganisms and foods were calculated. A high number of the average CV% values lies over the 10% to 35% range, which is reasonable for this type of assay and these types of studies. However the CV of individual systems for ATP detection ranged from 9% to 123%. The Hygiena SystemSURE with the Hygiena Supersnap swab had 9% CV whereas Neogen Accupoint and Charm Pocketswab had 123% and 86% CV respectively.

3. High RLU values do not confer a greater sensitivity to a system. Sensitivity depends greatly on several factors within the each system including both instrument and reagent output as well as the background of the system. The systems with large RLU scales such as Neogen and Charm systems also showed limit of detection of 10.0 fmols compared to other systems with a limit of detection of 1 fmol.

4. Clone swabs (Hygiena Snapshot) consistently improved the instrument performance by reducing the background noise and improving both sensitivity and repeatability of 3M CleanTrace, Charm Novalum and BioControl MVP systems.

5. The mean calculated sensitivity (or limit of detection) of the most and least sensitive systems differed by approximately 60 fold. The Hygiena SystemSure with the Hygiena Supersnap swab was the most sensitive reagent / instrument combination with a limit of detection of 0.17fmols. (data in tables 11 and 12)

6. When tested against the microbial cultures, the overall best extraction index included Hygiena Supersnap swab and Hygiena Snapshot swab. Several systems were included but no significant difference was observed.( tables 16, 17 and 18 ) The limit of detection was typically 10,000 to 100,000 bacteria / ml.

7. When tested against the food samples, the overall best extraction index included Hygiena Snapshot swab. Several systems were included but no significant difference in extraction rates was observed. ( tables 16, 17 and 18 ). However there was a difference in the sensitivity of the systems when testing foodstuffs. This was independent of the type of foodstuff and was similar to the sensitivities determined for ATP. The most sensitive system for the detection of food residues was Hygiena SystemSURE Plus with Supersnap swab and the least sensitive systems were the Neogen Accupoint and Charm Pocketswab.

The report that follows this summary contains a large amount of detailed experimental data. It should be thoroughly reviewed to fully understand the depth of the experiments and the conclusions drawn from that data.

Copyright © 2009-10 Silliker, Inc.

3

snason
Highlight
snason
Highlight
snason
Highlight
snason
Highlight
snason
Highlight
Page 19: Informacion Luminometria ATP Sector Salud

Objective

The purpose of this study was to evaluate commercial ATP units targeted for use in the food industry. The study was intended to provide comparative data and not as a comprehensive evaluation or review.

Background ATP is a molecule that is essential and common to all plant, animal and microbial cells. ATP may persist long after the cells have died. Measurement of ATP requires only a few minutes and is based upon the firefly luciferase bioluminescence assay.

The properties of the ATP assay are well suited to determinations of cleanliness, with cleanliness being defined as the absence of organic (derived from life) material. Clean surfaces have no ATP, while dirty surfaces have ATP and perhaps live microbial cells. The result of an ATP test that is available in minutes permits the immediate assessment of the sample condition and whether additional cleaning action is required. The traditional method of determining cleanliness is the aerobic plate count. This test requires 2 days to complete. This procedure is limited in the type of microorganisms it can detect and does not detect organic residue.

ATP bioluminescence systems are available from a number of commercial companies. Measures of how these systems perform under controlled conditions will be helpful to customers as well as manufacturers that must make informed decisions.

Materials and Methods

ATP Monitoring Systems and Devices The study was conducted using two sets of ATP monitoring systems and swabs. In the first set (Set 1), the performance of five different commercially available ATP monitoring systems was evaluated using eight different commercially available swabs (Table 1). In the second set (Set 2), the performance of three different commercially available ATP monitoring systems of Set 1 was evaluated using four different commercially available swabs (Table 2).

Table 1. First set of ATP monitoring system and swab combinations used in the sensitivity studies ATP monitoring system Swab

Hygiena Snapshot SBC 1575 Biocontrol Lightning MVP Unit Lightning Clean Trace 3M Clean Trace NG Luminometer Hygiena Snapshot SPXL 1333 Hygiena Snapshot CH 1616 Charm Sciences novaLUM Unit Pocketswab Plus

Hygiena SystemSURE Unit Hygiena Ultrasnap Neogen Accupoint Unit Neogen Accupoint

Copyright © 2009-10 Silliker, Inc.

4

Page 20: Informacion Luminometria ATP Sector Salud

Table 2. Second set of ATP monitoring system and swab combinations used in the sensitivity studies ATP monitoring system Swab Biocontrol Lightning MVP Unit Hygiena Snapshot SBC 1575 Clean Trace NG Luminometer Hygiena Snapshot SPXL 1333

Hygiena Supersnap Hygiena SystemSURE Unit Ultrasnap

Test Matrices Test samples included water spiked with ATP as a measure of system sensitivity, microbial cultures and food soil representing organic residue likely to be present in environmental samples. The test samples, as described below, were prepared and analyzed by the ATP monitoring systems and swabs listed in Tables 1 and 2.

Part A: Detection of Pure ATP The sensitivity of the ATP monitoring systems was first compared using water samples spiked at various levels of ATP (Table 3). In order to evaluate the influence of swab materials on ATP, the spiked water samples were tested either by depositing a pre-determined amount of samples directly onto the swab bud or into the activated reagent in the swab device. For the blank (control) samples, a commercially available ultra pure sterile water product was used (Rockland Inc., Gilbertsville, PA).

Table 3. ATP spiked water used in the sensitivity studies Recovery method ATP level (femtomole)

0 0.1 1 5 10 100

ATP recovery from swab

1,000 0 ATP recovery in solution 100

Copyright © 2009-10 Silliker, Inc.

5

Page 21: Informacion Luminometria ATP Sector Salud

1. ATP recovery from swab

a. 10 µL of sample containing ATP levels at 0 (blank), 0.1, 1, 5, 10, 100, 1000 femtomoles (fmoles) was pipetted directly onto the appropriate swab bud.

b. The swab device was activated. c. The swab was placed in the ATP unit. d. The measurement was started and the relative light units (RLU) result was

read and recorded. e. This procedure was repeated with 10 replicates of each dilution on each

swab device using each ATP system and swab combination of Set 11 and Set 2.

f. Test results were reported as follows: ATP recovery from swab 1000

fmoles 100

fmoles 10

fmoles 5

fmoles 1

fmoles 0.1

fmoles 0

fmoles Replicates 1-10

2. ATP recovery in solution a. The swab device was activated without adding the sample. b. The swab was removed from the activated device. c. 10 µL of sample containing ATP at the level of 0 (control blank) and 100

fmol was added into the reagent in the swab device. d. The swab was placed in the device. e. The swab device was placed in the ATP unit. f. The measurement was started and the RLU result was read and recorded. g. The ATP in solution tests were only performed on the following ATP

system and swab combinations of Set1. i. Biocontrol Lightning MVP unit using Lightning swab,

ii. Clean Trace NG Luminometer using Clean Trace swab, iii. Charm Science unit using Pocketswab Plus swab, iv. Hygiena SystemSURE unit using Hygiena Ultrasnap, v. Neogen Accupoint was omitted per client’s request

h. The ATP in solution tests were performed on all the ATP systems and swabs of Set 2.

i. This procedure was repeated with 10 replicates of each dilution on each swab device using each ATP system and swab combination of Set 11 and Set 2.

j. Test results were reported as follows: ATP recovery in solution

100 fmoles 0 fmoles (blank)

Replicates 1-10

Copyright © 2009-10 Silliker, Inc.

6

1 ATP systems and swabs of Set 1 were tested three times for each replicate, providing 30 readings

Page 22: Informacion Luminometria ATP Sector Salud

Part B: Detection of ATP from pure microbial cultures grown in broth

Five bacterial cultures, Escherichia coli, Lactobacillus plantarum, Pseudomonas aeruginosa, Salmonella Typhimurium and Staphylococcus aureus and one yeast culture- Saccharomyces cerevisiea were obtained from the Silliker Inc., Food Science Center culture collection (FSC-CC) (Table 4). The bacterial cultures were cultivated in 10 mL of tryptic soy broth (TSB) and incubated at 35°C for 18-24 h. S. cerevisiea was cultivated in 10 mL of sabourand dextrose broth and incubated at 30°C for 48 h. After incubation, each culture was washed once with sterile deionized water by centrifugation at 8000 rpm for 20 min and reconstitution with sterile deionized water. The cell level in each dilution of the bacterial cultures was determined by plating serial dilutions on tryptic soy agar (TSA) incubated 35°C for 24 h. The cell level in each dilution of the yeast cultures was determined by plating serial dilutions on potato dextrose agar (PDA) incubated 25°C for 5 d.

Table 4. Bacterial strains used in the sensitivity studies conducted with pure cultures Culture Source FSC-CC Number Escherichia coli Chicken broth 1809 Lactobacillus plantarum Juice drink 998 Pseudomonas aeruginosa Soil 2606 Saccharomyces cerevisiea ATCC MYA 658 2847 Salmonella Typhimurium USDA culture 1860 Staphylococcus aureus Potatoes 1561

Each culture was serially diluted 10-fold up to 10-6 with sterile deionized water and analyzed by each ATP system and swab combination of Set 1 and Set 2 by pipetting 10 µL of culture dilution directly onto each swab bud, placing the swab device into the ATP unit and reading the RLU result. After testing, the cell level in each dilution was determined by plating serial dilutions of the bacterial cultures on tryptic soy agar (TSA) and the yeast culture on potato dextrose agar PDA). The TSA plates were incubated at 35 ± 1°C for 24 ± 2 h and the PDA plates were incubated at 30 ± 1°C for 48 h prior to enumeration. Test results were reported as follows:

ATP from pure culture 0 dilution

(undiluted) 1:10

dilution1:100

dilution1:1,000 dilution

1:10,000dilution

1:100,000 dilution

1:1,000,000dilution

Replicates 1-10

Part C: Detection of ATP from Food Food samples that are commonly available from commercial supermarkets were used to represent a range of product groups for this portion of the study (Table 5). Liquid food samples (orange juice and milk) were diluted using ATP-free sterile water (v/v) in the following ratios: full strength liquid (0 dilution); 1:10; 1:100; 1:1000; and 1:10,000. Solid food samples (ground beef and salad greens) were first stomached using 10 g of sample in 90 ml ATP-free sterile water and then diluted using ATP-free sterile water (w/w) in the following ratios: 1:10 (stomached samples); 1:100; 1:1,000; and 1:10,000. All test samples were shaken by hand for 2 min for homogenization.

Copyright © 2009-10 Silliker, Inc.

7

Page 23: Informacion Luminometria ATP Sector Salud

Table 5. Food samples used in the sensitivity studies Food Ground beef Milk (pasteurized 2% low-fat) Orange juice (pasteurized without pulp) Salad (bagged mixed salad greens)

Ten replicates of each food suspension were analyzed by each ATP system and swab combination of Set 1 and Set 2 by pipetting 10 µL of food suspension dilution directly onto each swab bud, placing the swab device into the ATP unit and reading the RLU result. Test results were reported as follows:

ATP from Food 1:1 (full strength liquid;

0 dilution) 1:10 1:100 1:1,000 1:10,000

Replicates 1-10

Part D: Detection of ATP from Food Soiled Stainless Steel Surfaces Some brands of swabs are wet, and stay wet, during the intended shelf life while other brands of swabs are dry. Test results may vary due to wet and dry swabs, and wet and dry surfaces. Therefore, in addition to the food suspension dilutions tested, stainless steel surfaces soiled with ground beef and milk were tested (Table 6).

Table 6. Food suspension dilutions tested on stainless steel surface in the sensitivity studies. Food Dilution

1:10 Ground beef 1:1,000 1:1 Milk 1:1,000

Five hundred (500) µL of food suspension from the 1:10 and 1:1,000 dilutions of the ground beef and 500 µL of food suspension from the 1:1 (full strength liquid; 0 dilution) and 1:1,000 dilutions of the milk and were spread evenly onto individual 4x4 in2 stainless steel surfaces and immediately tested after preparation by swabbing each swab bud over the stainless steel surface, placing the swab device in the ATP unit and reading the RLU result. An additional set of stainless steel surfaces were prepared as described above and allowed to dry at room temperature for 18-24 h. After drying, each individual stainless steel piece was swabbed by each swab bud and analyzed by each ATP unit and reading the RLU result. Ten replicates of each dilution were tested. Test results were reported as follows:

ATP from Food Soiled Stainless Steel Surface 1:1 (full strength

liquid; 0 dilution) 1:10 1:1,000

Replicates 1-10

Copyright © 2009-10 Silliker, Inc.

8

Page 24: Informacion Luminometria ATP Sector Salud

Data Analysis

Linearity:

The linear correlation coefficient (r) measures the strength and the direction of a linear relationship between the sample variables. The value of r for samples at varying ATP levels and food samples at varying dilution levels against the corresponding RLU readings was calculated to determine the linear relationship.

Repeatability

Repeatability means the level of agreement between successive results obtained with the same method on the same test sample. The ATP measurement repeatability was expressed as a coefficient of variation (CV%), which is the standard deviation (SD) expressed as percentage of the mean (i.e. CV % = 100 × SD/mean).

Sensitivity

Limit of detection (LOD) in this study was calculated as the ratio of the mean of a true blank with three standard deviations to RLU per fmoles (mean +3 sd/RLU per fmoles). Some systems such as the SystemSure, the blanks can run at 0, the next significant RLU is used as the lowest detection limit. Charm and Neogen instruments have a built in algorithm, which discounts some of the RLU measured and therefore they do not display an RLU value for less than 10 fmoles. For these systems an LOD value of 10 fmoles is stated.

Relative Light Unit (RLU) per femtomole

The RLU per femtomole values were calculated by dividing RLU readings to corresponding ATP levels. In order to minimize the variability, the average RLU per femtomole was calculated using the first three ATP dilutions (i.e. 1,000, 100 and 10 fmoles).

Comparison of index (extraction index) for microorganisms and food samples

The lowest detectable concentration levels for microbial and food samples are presented as the lowest dilution at which ATP could be detected. The extracted fmole per dilution was calculated as the ratio of the RLU reading without background to the RLU per fmole value. The extracted fmole per dilution values that show greater than 1 ATP fmole are counted as extracted. The comparison using the extracted ATP rather than RLU normalizes the data making analysis comparable and extraction levels more relevant to each system.

Copyright © 2009-10 Silliker, Inc.

9

Page 25: Informacion Luminometria ATP Sector Salud

Results and Discussion

Test results are reported in relative light unit (RLU) readings in this study. The sensitivity of each of the 12 ATP monitoring systems of Set 1 and Set 2 was tested using aqueous ATP solutions, cell suspensions and exudates of food samples. Each ATP monitoring system uses a different measurement scale. The test results of ten replicates of seven dilutions of the ATP solution, seven dilutions of cell suspensions, five dilutions of liquid food samples (i.e. milk and orange juice), four dilutions of solid food samples (ground beef and salad) and two dilutions of food exudates on stainless steel surfaces are shown in Appendix A.

The formulation Hygiena swab products used in Set 1 and Set 2 differ in order to show the effects of extractant on subsequent detection and test performance. Hygiena products used in Set 2 are those supplied routinely on a commercially basis.

Linearity

Log10 RLU values of ten replicates were plotted against Log10 dilutions of the test matrices (Appendix B). The best-fit line is represented by the solid line and the 95% confidence limits presented as dashed lines. The linear correlation coefficient (r) of the best-fit line was determined to measure the linearity of ATP monitoring systems. A value of 1.0 represents a perfect fit of the regression line to the data. Values greater than 0.8 indicate the curve fits the data very well.

The linear correlation coefficient values of the ATP monitoring systems tested for RLU over the range of dilutions of the ATP solution, microorganisms, and foods are summarized in Tables 7 and 8. All regressions were significant. A total of 132 correlation coefficients were calculated. All correlation coefficient values with the exception of an outlier value of 0.643 determined by the Neogen Accupoint with Neogen Accupoint swab for E. coli, were greater than 0.8 and provided strong evidence that the Log10 RLU readings and Log10 dilutions were linearly correlated.

Repeatability

To quantify repeatability, the coefficient of variation (CV%) for each dilution of the ATP solution, microorganisms and foods were calculated (Appendix C). The CV% values indicate the amount of variation. The higher CV% values represent greater variation and hence less repeatability. The CV% values increased as the limits of detection were approached. This is expected because closer to the detection limit there is much less ATP to measure and there is more variability in the measurement. The CV% were erratic between dilutions of the test matrices and ranged from 2% to 316%.

For comparative purposes, the average CV% of the ATP solution data, microbial cultures and food samples was calculated. All CV% were then averaged for each ATP monitoring system (Table 9-10). The average coefficient of variation values ranged from a low of 6% by the BioControl Lightning MVP with Hygiena Snapshot SBC 1575 swab to a high of 186% by the Neogen Accupoint with Neogen Accupoint swab. The relative frequency distribution of the average CV% values is presented in Figure 1. A high number of the average CV% values lies over the 10% to 35% range, which is reasonable for this type of assay and these types of studies.

Copyright © 2009-10 Silliker, Inc.

10

The natural variation of biological assays such as ATP bioluminescence combined with the variation from sample collection and operator handling during hygiene monitoring applications means that the test results do not have the same precision as other analytical methods. The results from ATP hygiene measurements are used as a rapid qualitative

Page 26: Informacion Luminometria ATP Sector Salud

assessment of cleaning and results are typically expressed in board bands of Pass , Caution or Fail that typically equate to 10 – 100fmols of ATP. The ATP hygiene monitoring application is not intended to be used as a precise determination of ATP content. The trending of RLU or Pass / Fail results are much more meaningful in routine manufacturing operations.

Sensitivity

ATP Solution

The average RLU readings, standard deviations and CV% values for dilution of the ATP solution, microorganisms and food samples are summarized in Appendix C. The average background readings from the system (i.e. reagents outputs in the 0.0% ATP blank solution) for the 10 replicate test samples were 0.0 RLU for the swab devices of Charm Science with Pocketswab Plus swab (Set 1), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1) (Appendix C-Table C1) and the Hygiena SystemSure with Hygiena Supersnap swab (Set 2) and Hygiena SystemSure with Hygiena Ultrasnap swab (Set 2) (Appendix C Table C2).

The average background reading for the Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575 swab, BioControl Lightning MVP with Lightning swab, Clean Trace NG Luminometer with Clean Trace swab, Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab, and Hygiena SystemSure with Hygiena Ultrasnap swab of Set 1 were 142.17, 283.17, 4.0, 0.83, and 0.67 RLU, respectively (Appendix C-Table C1). The average background reading for the Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575swab and Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab were 199 and 1.90 and RLU, respectively (Appendix C-Table C2). ATP analysis of the ATP solutions showed that the calculated LODs ranged from 0.17 fmoles to 10 fmoles (Tables 11 and 12). The mean calculated LODs of the most and least sensitive systems differed by approximately 60 fold. The Hygiena SystemSure with Hygiena Supersnap swab was the most sensitive, while the Charm Science with Pocketswab Plus swab and the Neogen Accupoint with Neogen Accupoint swab were the least sensitive systems.

The RLU output and range shown of different systems varies considerably because the RLU is not a standard unit of measurement and is unique to each test system. High RLU values do not confer a greater sensitivity to a system and this is shown in Table 13 that summarizes the performance characteristics of the tests systems as supplied commercially.

ATP detection and recovery was variable between systems (Figure 2). BioControl had a high recovery of ATP but it was also highly variable (+/- 37%). Hygiena ATP recovery was high (92%) with good repeatability (9% CV). The Charm Science with Pocketswab recovered only 57% ATP with 20% variability, and Clean Trace NG Luminometer systems recovered only 52% ATP with a variability of 10%. A reduced recovery of ATP means that the accuracy of the system is also reduced.

Hygiena SnapShot is designed to be used with other luminometers such as 3M Clean Trace NG Luminometer, BioControl MVP and Charm Sciences novaLUM. Table 14 shows snapshot performance for ATP detection compared to other systems and their corresponding swabs.

Copyright © 2009-10 Silliker, Inc.

11

Page 27: Informacion Luminometria ATP Sector Salud

Snapshot increases the performance of other luminometers and detects lower levels of ATP by;

• Increasing the linearity of ATP response • Reducing the background and giving similar or greater RLU output per unit of

ATP • Reducing the variation and thereby increasing repeatability and consistency of

ATP detection • Increasing the extractability of ATP and thereby increasing the accuracy of the

measurement • Improving the sensitivity of the system • Similar results were also obtained with the detection of foodstuffs.

Microorganisms

Data for six different microbial cultures using various ATP systems and swab devices are presented in Appendix C-Table C3 through C14. During the course of testing the dilutions, the lower detection limit was observed; hence not all 10-fold dilutions were analyzed by each swab device. The analysis of each culture used the RLU per femtomole calculation to normalize the RLU measured by each system to femtomoles. This normalization is required to bring all measured RLUs onto a similar scale; this scale can then be easily compared device to device and instrument to instrument. Comparisons using RLUs is difficult due to the differing scales used and the variable machine and reagent background RLUs which do not contribute to the measured signal. Hence, the normalization of the data to RLU per femtomole is required for accurate comparative.

Escherichia coli

E. coli had a culture level of 9.59 log10 CFU/ml for the first set of swab devices (Set 1) analyzed and 9.08 log10 CFU/ml for the second set of swab devices (Set 2) tested. The Biocontrol Lightning MVP with Hygiena Snapshot 1575 swab (Set 1), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1), Hygiena SystemSure with Hygiena Ultrasnap swab (Set 1 and 2), and Neogen Accupoint with Neogen Accupoint swab (Set 1) were the least sensitive swab devices analyzed for the detection of ATP from E. coli as these swab devices were able to detect the E. coli at the 1:100 diluted culture level, and not when the culture was subsequently diluted (Appendix C-Table C3, Appendix D Table D2). The Biocontrol Lightning MVP with Lightning swab (Set 1), Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 2), Clean Trace NG Luminometer with Cleantrace swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot 1333 swab (Set 1 and 2), Charm Science with Pocketswab (Set 1) and Hygiena SystemSure with Hygiena Supersnap (Set 2) appeared to be the most sensitive swab devices analyzed as these were able to detect ATP from E. coli at the next dilution (1:1,000) tested (Appendix C-Tables C3 and C4, Appendix D Table D2).

Lactobacillus plantarum

L. plantarum had a culture level of 6.45 log10 CFU/ml for the first set of swab devices (Set 1) analyzed and 9.48 log10 CFU/ml for the second set of swab devices (Set 2) tested. The Biocontrol Lightning MVP with Lightning swab (Set 1), Hygiena SystemSure with Hygiena Ultrasnap swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1) were the least sensitive swab devices analyzed for the detection of ATP from L. plantarum, as these swab devices were able to detect L. plantarum at the 1:10 diluted culture level (Appendix C-Table C5, Appendix D Table D2). The Biocontrol MVP with Hygiena Snapshot 1575 swab (Set 1), Clean Trace NG Luminometer with Cleantrace

Copyright © 2009-10 Silliker, Inc.

12

Page 28: Informacion Luminometria ATP Sector Salud

swabs (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot 1333 swab (Set 1), Charm Science with Hygiena Snapshot 1616N swab (Set 1), Charm Science with Pocketswab (Set 1), and Hygiena SystemSure with Hygiena UltraSnap swab (Set 2) could extract ATP from the next dilution (1:100). The most sensitive swabs were Biocontrol Lightning MVP with Hygiena Snapshot 1575 swab (Set 2), Clean Trace NG Luminometer with Hygiena Snapshot 1333 swab (Set 2) and Hygiena SystemSure with Hygiena Supersnap swab (Set 2), as these were able to detect ATP from L. plantarum at the 1:1,000 dilution level. (Appendix C-Tables C5 and C6, Appendix D Table D2).

Pseudomonas aeruginosa

P. aeruginosa had a culture level of 8.32 log10 CFU/ml for the first set of swab devices (Set 1) analyzed and 8.52 log10 CFU/ml for the second set of swab devices (Set 2) tested. The Clean Trace NG Luminometer with Cleantrace swab (Set 1), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1), Charm Science with Pocketswab Plus swab (Set 1), Hygiena SystemSure with Hygiena Ultrasnap swab (Set 1 and 2), Neogen Accupoint with Neogen Accupoint swab (Set 1) and Hygiena SystemSure with Hygiena Supersnap swab (Set 2) were the least sensitive swab devices analyzed for the detection ATP from P. aeruginosa, as these swab devices were able to detect P. aeruginosa at the 1:100 diluted culture level, and not when the culture was subsequently diluted (Appendix C-Tables C7 and C8, Appendix D Table D2). The Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 1 and Set 2), Biocontrol Lightning MVP with Lightning swab (Set 1) and Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 1 and 2) appeared to be the most sensitive swab devices analyzed as these were able to detect ATP from P. aeruginosa at the 1:1,000 diluted culture level (Appendix C-Tables C7 and C8, Appendix D Table D2).

Saccharomyces cerevisiae

S. cerevisiae had a culture level of 7.49 log10 CFU/ml for the first set of swab devices (Set 1) analyzed and 7.74 log10 CFU/ml for the second set of swab devices (Set 2) tested. The Charm Science with Hygiena Snapshot CH 1616 swab (Set 1), Charm Science with Pocketswab Plus swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1) were the least sensitive swab devices analyzed for the detection of ATP from S. cerevisiae as these swab devices only were able to detect S. cerevisiae at the 1:1,000 diluted culture level, while all other systems could detect ATP from S. cerevisiae at the 1:10,000 diluted culture level (Appendix C-Tables C9 and C10, Appendix D Table D2).

Salmonella Typhimurium

S. Typhimurium had a culture level of 7.96 log10 CFU/ml for the first set of swab devices (Set 1) analyzed and 9.08 log10 CFU/ml for the second set of swab devices (Set 2) tested. The Biocontrol Lightning MVP with Hygiena Snapshot 1575 swab (Set 1), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1) were the least sensitive swab devices analyzed for the detection of ATP from S. Typhimurium as these swab devices were able to detect ATP from S. Typhimurium at the 1:10 diluted culture level, while the Biocontrol Lightning MVP with Lightning swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 1), Charm Science with Pocketswab (Set 1) and Hygiena SystemSure with Hygiena Ultrasnap swab (Set 1 and 2) could detect ATP from S. Typhimurium at the 1:100 diluted culture level. The most sensitive systems were Clean Trace NG Luminometer with Cleantrace swab (Set 1), Biocontrol Lightning MVP with Hygiena Snapshot 1575 swab (Set 2), Clean Trace NG Luminometer with Hygiena

Copyright © 2009-10 Silliker, Inc.

13

Page 29: Informacion Luminometria ATP Sector Salud

Snapshot SPXL 1333 swab (Set 2) and Hygiena SystemSure with Hygiena Supersnap swab (Set 2) as these swab devices were able to detect ATP from S. Typhimurium at the 1:1,000 diluted culture level (Appendix C-Tables C11 and C12, Appendix D Table D2).

Staphylococcus aureus

S. aureus had a culture level of 8.23 log10 CFU/ml for the first set of swab devices (Set 1) analyzed and 8.76 log10 CFU/ml for the second set of swab devices (Set 2) tested. The Biocontrol Lightning MVP with Snapshot 1575 swab (Set 1), Biocontrol Lightning MVP with Lightning swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 1), Charm Science with Pocketswab (Set 1), Hygiena SystemSure with Hygiena UltraSnap swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1) were the least sensitive swab devices analyzed for the detection of ATP from S. aureus, as these swab devices were able to detect ATP from S. aureus at the pure culture level, and no detection was observed at any lower levels of S. aureus (Appendix C-Table C13, Appendix D Table D2). The Clean Trace NG Luminometer with Cleantrace swab (Set 1) and Hygiena SystemSure with Hygiena UltraSnap swab (Set 2) could detect ATP from S. aureus at the 1:10 diluted culture level, while Charm Science with Hygiena Snapshot CH 1616 swab (Set 1), Biocontrol Lightning MVP with Hygiena Snapshot 1575 swab (Set 2), Clean Trace NG Luminometer with Hygiena Snapshot 1333 swab (Set 2) were able to detect ATP from S. aureus at the 1:100 diluted culture level. The most sensitive swab device analyzed for the detection of ATP from S. aureus was the Hygiena SystemSure with Hygiena Supersnap swab (Set 2) as it was able to detect ATP from S. aureus at the 1:1,000 diluted culture level (Appendix C-Tables C13 and C14, Appendix D Table D2).

Compendium Extraction Index

To fully evaluate how the systems perform across the range of bacteria, the lowest level from each system for each bacterium can be assessed by analyzing at which dilution level in each dilution series 1 femtomole of ATP can be extracted above the blank values. This relationship is then tabulated in Table 15.

The systems with the overall best extraction index include Hygiena SystemSURE with Hygiena Supersnap swab (Set 2), Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 2) and Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 2). The overall extraction index is -3.00 (which is a mean extraction level of 1:1,000) across all bacteria measured. The next best systems are in the -3.00 to -2.00 range (i.e. 1:100 to 1:1000 dilution region). These systems include Clean Trace NG Luminometer with Clean Trace swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 1), Hygiena SystemSURE with Hygiena Ultrasnap swab (Set 2), Biocontrol Lightning MVP with Lightning swab (Set 1), Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 1), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1) and Charm Science with Pocketswab Plus swab (Set 1). With the other systems with extractions below the 1:100 dilution across all bacteria includes Hygiena SystemSURE with Hygiena Ultrasnap swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1).

Large differences were observed in the ATP results from different species of microorganism and these were lower than expected. This may be a reflection of species difference and size or the effect of the culture preparation and sample storage during testing. The limit of detection for most systems was 105 – 106 bacteria CFU/ml which equates to 103 – 104 bacterial per swab. Similarly 103 – 104 yeast/ml which equates to 101

Copyright © 2009-10 Silliker, Inc.

14

Page 30: Informacion Luminometria ATP Sector Salud

– 102 yeasts per swab. However the prime purpose of the ATP hygiene monitoring application is to detect product residue after cleaning because residues are a direct objective measurement of cleaning efficiency, and the level of ATP in foodstuffs is far greater than that of microbes. The ATP test is not intended to be a replacement for microbiological tests. The post-cleaning standard for bacteria on surfaces is typically 100 – 500 CFU per 100 cm2 which is equivalent to 100 – 500 CFU per swab that is clearly not detectable by the ATP test as shown above.

Food Samples

Raw Ground Beef

The Biocontrol Lightning MVP with Lightning swab (Set 1) was the least sensitive swab device analyzed for the detection of ATP for raw ground beef as this swab device only was able to detect this food suspension at the 1:10 dilution level, while Biocontrol Lightning with Hygiena Snapshot 1575swab (Set 1), Clean Trace NG Luminometer with Cleantrace swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshots 1333 swab (Set 1 and Set 2), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1), Charm Science with Pocketswab Plus swab (Set 1), and Neogen Accupoint with Neogen Accupoint swab (Set 1) could detect ATP at the 1:100 dilution level. Hygiena SystemSure with Hygiena Ultrasnap swab (Set 1 and Set 2) and Hygiena SystemSure with Hygiena Supersnap swab (Set 2) were able to detect ATP for raw ground beef at the 1:1,000 dilution level. The most sensitive swab device was BioControl Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 2) as this swab device was able to detect ATP from raw ground beef at the lowest (1:10,000) dilution tested (Appendix C-Tables C15 and C16, Appendix D Table D1).

Milk

The Neogen Accupoint with Neogen Accupoint swab (Set 1) was the least sensitive swab device analyzed for the detection of ATP from pasteurized 2% low fat milk as this swab device only was able to detect this food suspension at the 1:10 dilution level, while the BioControl Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 1), BioControl Lightning MVP with Lightning swab (Set 1), Clean Trace NG Luminometer with Clean Trace swab (Set 1), Charm Science with Hygiena Snapshot CH 1616 swab (Set 1), Charm Science with Pocketswab Plus swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 2) and Hygiena SystemSURE with Hygiena Supersnap swab (Set 2) could detect ATP from this food suspension at the 1:100 dilution level. Hygiena SystemSure with Ultrasnap swab (Set 1) and BioControl Lightning MVP with Hygiena Snapshot 1575 swab (Set 2) were able to detect ATP at the 1:1,000 dilution level. The most sensitive swabs were Clean Trace NG Luminometer with Hygiena Snapshot 1333 swab (Set 1) and Hygiena SystemSure with Hygiena Ultrasnap swab (Set 2) as these devices were able to detect ATP from pasteurized 2% low fat milk at the 1:10,000 dilution level (Appendix C-Tables C17 and C18, Appendix D Table D1).

Orange Juice

The Charm Science with Pocketswab Plus swab (Set 1), Hygiena SystemSure with Ultrasnap swab (Set 1) and Neogen Accupoint with Neogen Accupoint swab (Set 1) were the least sensitive swab devices analyzed for the detection of ATP from orange juice containing no pulp as these swab devices only were able to detect this food suspension at the 1:1,000 dilution level, while all other systems analyzed were able to detect ATP for orange juice at the lowest dilution (1:10,000) tested (Appendix C-Tables C19 and C20, Appendix D Table D1).

Copyright © 2009-10 Silliker, Inc.

15

Page 31: Informacion Luminometria ATP Sector Salud

Mixed Salad Greens

The Charm Science with Hygiena Snapshot CH 1616 swab (Set 1) was the least sensitive swab device analyzed for the detection of ATP from orange as this swab device only was able to detect this food suspension at the 1:100 dilution level, while Clean Trace NG Luminometer with Cleantrace swab (Set 1), Charm Science with Pocketswab Plus swab (Set 1), Hygiena SystemSure with Hygiena Ultrasnap swab (Set 1), Neogen Accupoint with Neogen Accupoint swab (Set 1), and Clean Trace NG Luminometer with Snapshot 1333 (Set 2) could ATP from bagged mixed salad greens at the 1:1,000 dilution level (Appendix C-Table C21, Appendix D Table D1). The most sensitive swab devices analyzed for the detection of ATP for bagged mixed salad greens were the BioControl Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 1), BioControl Lightning MVP with Lightning swab (Set 1), Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 1), Biocontrol Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 2), Hygiena SystemSURE with Hygiena Supersnap swab (Set 2) and Hygiena SystemSURE with Hygiena Ultrasnap swab (Set 2) as these were able to detect ATP from salad greens at the lowest dilution (1:10,000) tested (Appendix C-Tables C21 and C22, Appendix D Table D1).

Wet versus Dry Food Soil

All swab devices analyzed were able to detect ATP from the wet and dry soiled stainless steel surfaces from the different food suspensions of raw ground beef and pasteurized 2% low fat milk (Appendix C-Tables C23-C26). The RLU reading of the wet and dry soiled stainless steel surfaces were higher than that of food suspension at the same dilution levels. This may be attributed to the difference in volumes used for food suspensions (i.e. 10 μL) and stainless steel coupons (i.e. 500 μL).

Snapshot improved sample recovery from dry surfaces (Figure 3). This is attributed to snapshot’s saturated swab bud and extractant that ensures good recovery of sample and increase RLU output compared to suppliers own swab.

Overall Comparison

For overall comparative purposes, the average extraction index of each microbial and food sample, and the ATP monitoring system were calculated. When tested against the microbial cultures, the Hygiena SystemSure system with Hygiena Supersnap swab (Set 2), BioControl Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 2) and Clean Trace NG Luminometer with Hygiena Snapshot SPXL 1333 swab (Set 2) appeared to be the most sensitive ATP monitoring system analyzed as they were able to detect ATP from the microbial cultures at higher dilution levels compare to all other systems (Table 15). Most RLU output due to ATP derived from microbial cultures was highest in S. cerevisiae and lowest in S. aureus.

When tested against the food samples, the Hygiena SystemSure with Hygiena Ultrasnap swab (Set 2) and the BioControl Lightning MVP with Hygiena Snapshot SBC 1575 swab (Set 2) appeared to be the most sensitive ATP monitoring systems analyzed as they were able to detect ATP from the food samples at higher dilution levels compare to all other systems (Table 16). Most RLU output due to ATP derived from food residues was highest in orange juice and lowest in ground beef.

Each system with generic or clone swabs can be tabulated and graded according to performance from each section, aqueous ATP detection, ATP recovery from swab, extraction of ATP from microbial cultures and extraction of ATP from food stuffs. The

Copyright © 2009-10 Silliker, Inc.

16

Page 32: Informacion Luminometria ATP Sector Salud

Copyright © 2009-10 Silliker, Inc.

17

comparison is shown in Table 17, each category ranks the systems using the most current versions of each system either swab or instrument.

The Hygiena swabs collectively are more sensitive to ATP and better at detecting low level food and cultures than all other systems.

Page 33: Informacion Luminometria ATP Sector Salud

Table 7. Correlation coefficient of ATP monitoring systems of Set 1

Microorganism Food

ATP Unit Swab Device

ATP Solution

Escherichia coli

Lactobacillus plantarum

Pseudomonas aeruginosa

Saccharomyces cerevisiea

Salmonella Typhimurium

Staphylococcus aureus

Ground beef Milk Orange

juice Salad

Biocontrol Lightning

MVP

Hygiena Snapshot SBC 1575

0.987 0.986 0.955 0.995 0.984 0.947 0.868 0.951 0.972 0.981 0.998

Biocontrol Lightning

MVP Lightning 0.982 0.931 0.928 0.986 0.989 0.921 0.810 0.945 0.960 0.996 0.992

Clean Trace NG

Luminometer Clean Trace 0.988 0.993 0.975 0.989 0.988 0.974 0.920 0.890 0.974 0.997 0.995

Clean Trace NG

Luminometer

Hygiena Snapshot

SPXL 1333 0.988 0.987 0.985 0.992 0.997 0.990 0.984 0.855 0.986 0.984 0.996

Charm Science

Hygiena Snapshot CH 1616

0.982 0.984 0.998 0.995 0.984 0.945 0.997 0.954 0.909 0.971 0.937

Charm Science

Pocketswab Plus 0.949 0.972 0.997 0.984 0.983 0.983 0.998 0.990 0.982 0.986 0.986

Hygiena SystemSURE

Hygiena Ultrasnap 0.988 0.962 0.991 0.974 0.979 0.980 ND a 0.855 0.986 0.984 0.996

Neogen Accupoint

Neogen Accupoint 0.976 0.643 0.970 0.972 0.928 0.967 0.961 0.995 0.920 0.938 0.985

a Not determined; only 0 dilution had RLU readings.

Copyright © 2009-10 Silliker, Inc.

18

Page 34: Informacion Luminometria ATP Sector Salud

Table 8. Correlation coefficient of ATP monitoring systems of Set 2

Microorganism Food

ATP Unit Swab Device

ATP Solution

Escherichia coli

Lactobacillus plantarum

Pseudomonas aeruginosa

Saccharomyces cerevisiea

Salmonella Typhimurium

Staphylococcus aureus

Ground beef Milk Orange

juice Salad

Biocontrol Lightning

MVP

Hygiena Snapshot SBC 1575

0.990 0.999 0.986 0.989 0.995 0.997 0.986 0.964 0.976 0.991 0.933

Clean Trace NG

Luminometer

Hygiena Snapshot

SPXL 1333

0.992 0.998 0.978 0.991 0.982 0.997 0.952 0.978 0.988 0.993 0.997

Hygiena SystemSURE

Hygiena Supersnap 0.987 0.987 0.990 0.991 0.980 0.996 0.992 0.984 0.986 0.950 0.986

Hygiena SystemSURE

Hygiena Ultrasnap 0.989 0.974 0.989 0.991 0.991 0.989 0.968 0.966 0.985 0.978 0.985

Copyright © 2009-10 Silliker, Inc.

19

Page 35: Informacion Luminometria ATP Sector Salud

Table 9. Coefficient of variation (CV%) of ATP solution, microorganisms and food samples when tested by different ATP monitoring systems-Set 1

ATP Unit Swab Device APT

Solution

Escherichia coli

Lactobacillus plantarum

Pseudomonas aeruginosa

Saccharomyces cerevisiea

Salmonella Typhimurium

Staphylococcus aureus

Ground beef Milk Orange

juice Salad

Ground beef

soiled surface

Milk soiled

surface Average (Range)2

Biocontrol Lightning

MVP

Hygiena Snapshot SBC 1575

19 20 18 24 53 20 20 19 21 14 12 35 15 22 (12-53)

Biocontrol Lightning

MVP Lightning 39 62 22 51 31 49 30 15 52 20 20 24 32 34

(15-62)

Clean Trace NG

Luminometer

Clean Trace 26 24 20 24 27 27 23 15 18 15 10 31 23 22

(10-31)

Clean Trace NG

Luminometer

Hygiena Snapshot

SPXL 1333 27 22 16 31 25 20 24 22 26 17 10 30 29 23

(10-31)

Charm Science

Hygiena Snapshot CH 1616

68 7 8 16 16 18 14 15 22 9 19 26 15 19 (7-68)

Charm Science

Pocketswab Plus 86 45 14 31 30 15 15 22 31 17 17 19 22 28

(14-86)

Hygiena SystemSURE

Hygiena Ultrasnap 59 28 15 51 31 20 11 31 40 17 17 58 50 33

(11-59)

Neogen Accupoint

Neogen Accupoint 123 186 30 36 63 142 113 78 19 19 17 33 27 68

(17-186)

2 CV% Average does not include 0 and NA readings

Copyright © 2009-10 Silliker, Inc.

20

Page 36: Informacion Luminometria ATP Sector Salud

Table 10. Coefficient of variation of ATP solution when tested by different ATP monitoring systems-Set 2

ATP Unit Swab Device APT

Solution

Escherichia coli

Lactobacillus plantarum

Pseudomonas aeruginosa

Saccharomyces cerevisiea

Salmonella Typhimurium

Staphylococcus aureus

Ground beef Milk Orange

juice Salad

Ground beef soiled surface

Milk soiled surface Average

(Range)3

Biocontrol Lightning

MVP

Hygiena Snapshot SBC 1575

10 11 10 44 28 19 6 17 48 16 19 21 25 21 (6-48)

Clean Trace NG

Luminometer

Hygiena Snapshot

SPXL 1333

15 18 14 35 17 20 16 27 19 13 18 36 44 23 (13-44)

Hygiena SystemSURE

Hygiena Supersnap 9 18 17 41 14 17 18 34 32 18 18 28 34 23

(9-41)

Hygiena SystemSURE

Hygiena Ultrasnap 28 15 13 30 18 12 14 13 158 13 17 23 32

30 (12-158)

3 CV% Average does not include 0 and NA readings

Copyright © 2009-10 Silliker, Inc.

21

Page 37: Informacion Luminometria ATP Sector Salud

Figure 1. Relative frequency distribution for CV% values

0

5

10

15

20

25

30

>5 6-10 11-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60 > 60

CV% Range

Rel

ativ

e Fr

eque

ncy

(Per

cent

age)

Dis

trib

utio

n

APT Solution Microbial Culture Food Soil

Copyright © 2009-10 Silliker, Inc.

22

Page 38: Informacion Luminometria ATP Sector Salud

Table 11. Relative Light Unit (RLU) per femtomole (fmole) and limit of detection (LOD) values of Set 1

ATP Unit Swab Device RLU/fmole LOD (fmole)

Biocontrol Lightning MVP

Hygiena Snapshot SBC 1575 552 0.60

Biocontrol Lightning MVP

Lightning 698 1.10

Clean Trace NG Luminometer

Clean Trace 5.4 1.30

Clean Trace NG Luminometer

Hygiena Snapshot SPXL 1333 6.8 0.42

Charm Science Hygiena Snapshot CH 1616 582 5.0

Charm Science Pocketswab Plus 218 10.0

Hygiena SystemSURE Hygiena Ultrasnap 1.0 1.0

Neogen Accupoint Neogen Accupoint 12 10.0

Copyright © 2009-10 Silliker, Inc.

23

Page 39: Informacion Luminometria ATP Sector Salud

Table 12. Relative Light Unit (RLU) per femtomole (fmole) and limit of detection (LOD) values of Set 2

ATP Unit Swab Device RLU/fmole LOD (fmole) Biocontrol Lightning

MVP Hygiena Snapshot SBC

1575 825 0.40

Clean Trace NG Luminometer

Hygiena Snapshot SPXL 1333 9.0 0.39

Hygiena SystemSURE Hygiena Supersnap 6 0.17

Hygiena SystemSURE Hygiena Ultrasnap 1.0 1.0

Copyright © 2009-10 Silliker, Inc.

24

Page 40: Informacion Luminometria ATP Sector Salud

Table 13: Summary of ATP performance characteristics of 5 commercial detection systems

Linearity Output (RLU) Variability Sensitivity System

(r) Blank

(Background at zero ATP)

Maximum (at 1000 fmols ATP ) (CV%) Limit of detection

(fmols ATP)

BioControl MVP with Lightning swab 0.982 283 975,941 39 1.1

3M Clean Trace NG Luminometer with CleanTrace swab 0.988 4 7382 26 1.3

Charm Science novaLUM with Pocketswab Plus 0.949 0 418,517 * 86 10.0

Hygiena SystemSURE Plus with Ultrasnap swab 0.988 0 1589 28 1.0

Hygiena SystemSURE Plus with Supersnap swab 0.987 0 4949 9 0.17

Neogen AccuPoint With Accupoint swab 0.976 0 15,649 * 123 10.0

* does not detect below 10 fmols at which level the instrument shows 0 RLU.

Copyright © 2009-10 Silliker, Inc.

25

Page 41: Informacion Luminometria ATP Sector Salud

Figure 2: Recovery of ATP by different ATP detection systems

0

20

40

60

80

100

120

140

160

Hygiena SystemSURE BioControl MVP Charm Sciences novaLUM

3M Clean Trace NG Luminometer

ATP recovery (%)

Copyright © 2009-10 Silliker, Inc.

26

Page 42: Informacion Luminometria ATP Sector Salud

Copyright © 2009-10 Silliker, Inc.

27

Table 14: SnapShot performance in different luminometer compared to manufacturers own swabs

Linearity Output (RLU) Variability Sensitivity System

(r) Blank

(Backgroundat zero ATP)

Maximum (at 1000 fmols ATP ) (CV%) Limit of detection

(fmols ATP)

BioControl MVP with Lightning swab 0.982 283 975,941 39 1.1

BioControl MVP with Snapshot swab 0.990 199 927,161 10 0.4

3M Clean Trace NG Luminometer with CleanTrace swab 0.988 4 7382 26 1.3

3M Clean Trace NG Luminometer with Snapshot swab 0.992 2 12620 15 0.42

Charm Science novaLUM With Pocketswab Plus 0.949 0 418,517 86 10.0

Charm Science novaLUM with

Snapshot swab 0.982 0 783,031 68 5.0

Page 43: Informacion Luminometria ATP Sector Salud

Figure 3: Snapshot sample recovery from dry surfaces compared with other swabs

0500

10001500200025003000350040004500

BioControl 3M Clean Trace NG 3M Clean Trace NG Meat residue Meat residue Milk residue

RLU output

Supplier’s swab Hygiena snapshot

Copyright © 2009-10 Silliker, Inc.

28

Page 44: Informacion Luminometria ATP Sector Salud

Table 15. Summary of lowest dilution levels for RLU output due to ATP derived from microbiological culture dilutions tested by different ATP monitoring systems

ATP Unit Swab Device Order of best extraction (mean of lowest dilutions detected)

Hygiena SystemSURE (Set 2)

Hygiena Supersnap -3.00

Biocontrol Lightening MVP (Set 2)

Hygiena Snapshot SBC 1575 -3.00

Clean Trace NG Luminometer (Set 2)

Hygiena Snapshot SPXL 1333 -3.00

Clean Trace NG Luminometer (Set 1) Clean Trace -2.50

Clean Trace NG Luminometer (Set 1)

Hygiena Snapshot SPXL 1333 -2.33

Hygiena SystemSURE (Set 2)

Hygiena Ultrasnap -2.17

Biocontrol Lightening MVP (Set 1) Lightening -2.17

Biocontrol Lightening MVP (Set 1)

Hygiena Snapshot SBC 1575 -2.00

Charm Science (Set 1)

Hygiena Snapshot CH 1616 -2.00

Charm Science (Set 1) Pocketswab Plus -2.00

Hygiena SystemSURE (Set 1)

Hygiena Ultrasnap -1.83

Neogen Accupoint (Set 1)

Neogen Accupoint -1.50

Copyright © 2009-10 Silliker, Inc.

29

Page 45: Informacion Luminometria ATP Sector Salud

Table 16. Summary of lowest dilution levels for RLU output due to ATP derived from food samples suspensions tested by different ATP monitoring systems

ATP Unit Swab Device Order of best extraction (mean of lowest dilutions detected)

Hygiena SystemSURE (Set 2)

Hygiena Ultrasnap -3.75

Biocontrol Lightening MVP (Set 2)

Hygiena Snapshot SBC 1575 -3.75

Clean Trace NG Luminometer (Set 1)

Hygiena Snapshot SPXL 1333 -3.50

Hygiena SystemSURE (Set 2)

Hygiena Supersnap -3.25

Hygiena SystemSURE (Set 1)

Hygiena Ultrasnap -3.00

Biocontrol Lightening MVP (Set 1)

Hygiena Snapshot SBC 1575 -3.00

Clean Trace NG Luminometer (Set 2)

Hygiena Snapshot SPXL 1333 -2.75

Clean Trace NG Luminometer (Set 1)

Clean Trace -2.75

Biocontrol Lightening MVP (Set 1) Lightening -2.75 Charm Science (Set 1) Pocketswab Plus -2.50 Charm Science (Set 1)

Hygiena Snapshot CH 1616 -2.50

Neogen Accupoint (Set 1)

Neogen Accupoint -2.50

Copyright © 2009-10 Silliker, Inc.

30

Page 46: Informacion Luminometria ATP Sector Salud

Table 17 Compendium Ranks of ATP Hygiene Monitoring Swabs for ATP Detection, ATP Recovery from swabs, Microbial Extraction and Food Extraction Levels

ATP Limit of Detection

ATP Recovery from Swab

ATP Extraction from Microbial Cultures

ATP Extraction from Foodstuffs

Hygiena Supersnap

BioControl Lightning (100%) Hygiena Supersnap Hygiena Ultrasnap

Hygiena Snapshot 1333

Hygiena Ultrasnap (93%) Hygiena Snapshot 1575 Hygiena Snapshot 1575

Hygiena Snapshot 1575

Charm Pocketswab (57%) Hygiena Snapshot 1333 Hygiena Snapshot 1333

Hygiena Ultrasnap

3M Cleantrace (52%) 3M Cleantrace Hygiena Supersnap

BioControl Lightning Hygiena Ultrasnap 3M Cleantrace

3M Cleantrace BioControl Lightning BioControl Lightning Hygiena

Snapshot 1616 Hygiena Snapshot 1616N Hygiena Snapshot 1616N

Charm Pocketswab Charm Pocketswab Charm Pocketswab

Neogen Accupoint Neogen Accupoint Neogen Accupoint

Copyright © 2009-10 Silliker, Inc.

31

Page 47: Informacion Luminometria ATP Sector Salud

Monitoring Hand Cleanliness Using systemSURE II™ ATP Hygiene System

The importance of hand washing is a well-known fact. Reports indicate that inadequately washed hands are a factor in up to 40% of food-related outbreaks of illness. The low compliance rate (30-40% for Foodservice operators) is probably due to the lack of a rapid and simple method to assess whether effective hand washing has occurred, even if operators have been fully trained in correct hand-washing procedures. Hygiene monitoring based on ATP (adenosine tri-phosphate) bioluminescence is a simple method that can be used as part of a hand-washing-monitoring program to obtain results in real time. ATP is the universal energy molecule found in all living cells. The combination of ATP with the enzyme luciferase produces light that can be measured in a luminometer. The amount of light is proportional to the amount of ATP and is expressed in Relative Light Units (RLUs). The greater the level of ATP, the higher the RLU value, the dirtier the hand. Measurement of ATP can be used: a) during induction training to show the effectiveness of good hand-washing technique and b) to monitor efficacy of hand washing by swabbing clean hands immediately after washing (before hands come into contact with anything). Our validation data suggests that a reduction in ATP levels of greater than 75% is achievable following effective hand washing (requires two samples, one before and one after cleaning -- data not shown). For routine monitoring, Hygiena set a single Pass/Fail limit that would only require a single swabbing device per employee. Company employees washed their hands with soap and water for approximately 20 seconds and dried them with paper towels. The palm of the dominant hand* was swabbed using Ultrasnap™ and the devices measured in the systemSURE II luminometer. A Pass/Fail limit of 60 RLU was used. If the result was higher than 60 RLU, the volunteer was asked to rewash his/her hands for retesting.

Procedure:

Page 48: Informacion Luminometria ATP Sector Salud

Table 1. Routine monitoring of the ATP levels on employees hands immediately post washing.

Volunteer RLU Result Action Yes/No Retest RLU

1 7 No -

2 88 Yes 21

3 23 No -

4 21 No -

5 245 Yes 61a

6 19 No -

7 23 No -

8 112 Yes 14

9 24 No -

10 72 Yes 15

11 24 No -

12 130 Yes 81a

13 28 No -

14 16 No -

15 30 No -

16 20 No -

17 36 No -

18 34 No -

19 25 No -

20 27 No -

* Only the dominant hand was swabbed for each volunteer Bold = failed results and "a"= retraining of hand-washing procedure required

For assessment of ATP levels immediately after hand washing, we found that there is a minimum RLU level that is attainable using an effective washing technique (see below). Studies show that the RLU value after hand washing is almost always below 100 RLU and below 60 RLU in most cases. Hygiena recommends setting a realistic Pass/Fail limit depending upon individual circumstances; e.g., type of food being handled, frequency of hand washing and type of soap/sanitizer used. Soaps vary in their effectiveness in reducing ATP levels. Before introducing an ATP hand-washing program we recommend testing the soap used before and after a thorough correct hand-washing procedure (minimum of 75% reduction in ATP level required). Hand-Washing Procedure It is a published fact that people tend to wash their hands in such a way that soiling and transient microorganisms are not removed from all areas of the hands equally. Hands should be washed frequently and thoroughly throughout the workday, especially after they have been exposed to sources of contamination. Proper hand-washing technique includes:

• Use warm water, sufficient amounts of soap/cleanser, and wash for 20-30 seconds • Wash up to the forearms • Use a nail brush to clean under fingernails • Rinse with hands opened down into the sink • Dry hands and arms thoroughly • Use the paper towel to turn off the water and discard

Wash Hands After • Blowing nose • Coughing/sneezing • Restroom and coffee breaks • Personal grooming • Touching unsanitary surface

For More Information Visit: www.hygienaUSA.com

Page 49: Informacion Luminometria ATP Sector Salud

Final Hygiena International Ltd Press Release HYG06 PRESS RELEASE HYG06

SYSTEMSURE HELPS IN FIGHT AGAINST HOSPITAL ASSOCIATED INFECTION

Hygiena International Ltd have joined forces with the North Tees and Hartlepool NHS

Foundation Trust to provide them with a new weapon for their armoury in the fight against

healthcare associated infections (HAI).

The Hygiena SystemSURE Plus ATP monitoring system is now being widely used by Trust

Quality Monitoring Officers, Ward Matrons and Domestic Supervisors to check the

cleanliness of patient areas within the hospitals. Previously, the Trust has relied wholly on

visual inspections of cleaning standards; now the handheld, lightweight Hygiena instruments

can provide a numerical result in seconds to show how clean or dirty a surface is. This

enables the Trust to monitor the cleaning effectiveness of both the hospital environment and

of the many types of patient equipment in use.

The Hygiena SystemSURE Plus instruments were originally introduced by the Trust’s

Infection Prevention and Control Department and have been used for handwashing training

for several years. Use of the tool has now spread throughout both hospitals and staff and

Managers are happy to use it to help eliminate the risk of infection in their areas.

It is a powerful tool, which has widespread applications throughout the healthcare sector to

monitor areas such as Sterile and Catering Services Departments.

Page 50: Informacion Luminometria ATP Sector Salud

Press release Hygiene Monitoring Press Release - 30th July 2008 - 2 (Page 2 of 4)

A monthly programme of hygiene inspections using the Hygiena SystemSURE Plus

commenced within North Tees and Hartlepool NHS Trust in May 2007 and every clinical in-

patient area now has routine swabbing undertaken on a monthly basis.

The Trust has also been proactive in using the handheld instruments for other more diverse

purposes such as swabbing clogs within Theatre, Doctor’s pens and stethoscopes and the

‘pods’ that are used to deliver various goods between wards and departments within the

Trust.

The palm sized SystemSURE instrument works by detecting the levels of adenosine

triphosphate (ATP), a biochemical found in all living organisms and biological residues. If

ATP is detected on a cleaned surface it means that the cleaning was not effective and the

surface is potentially a hazard for the spread of germs. The ATP method is quick, simple

and easy to do. A swab is taken of the area to be tested and is then inserted into the hand

held instrument. The result is interpreted automatically and can be displayed within a matter

of seconds as a simple “pass”, “caution” or “fail” display.

The Trust’s Quality & Performance Manager Sue Shannon said, “We are delighted to be

working with Hygiena to introduce this system. The ATP monitor provides immediate

feedback on the efficiency of cleaning and assists us in identifying areas that could have

previously been overlooked; such as the underside of tables and down the side of chairs,

where it is difficult to physically see contamination.

The results give us the evidence we need as a Quality Monitoring Department to tackle any

problem areas and to produce reports for Ward Matrons on the cleanliness levels in their

Page 51: Informacion Luminometria ATP Sector Salud

Press release Hygiene Monitoring Press Release - 30th July 2008 - 2 (Page 3 of 4)

areas. More importantly it enables the Domestic Services Department to provide assurance

that areas are as clean and as germ free as possible following cleaning, which is something

we were previously unable to do, having relied purely on visual inspection.

We have also used the hygiene devices to check the efficiency of various methods of

cleaning. As a result of this, we have invested heavily in microfibre technology over the past

twelve months as it has proven to be the most efficient method of cleaning virtually every

patient contact surface.

The Hygiena SystemSURE Plus is successfully marketed throughout 80 countries

worldwide. Combined with their Ultrasnap sampling devices, it has become the premier

system for hygiene monitoring not only in the medical and healthcare markets, but also in

food, cosmetics and pharmaceutical manufacture, together with water treatment when

combined with their Aquasnap sampling device. The instrument incorporates photodiode

technology, and offers up to an 80% saving compared to competitors equipment. Simple

keypad operation with LCD screen display is combined with storage capacity for 250 sites, a

choice of 20 sampling plans and up to 50 named system users.

Further information is available from: Hygiena International Ltd,

Unit 11, Wenta Business Centre, Colne Way, Watford, Hertfordshire WD24 7ND

Telephone: 01923 818821 Fax: 01923 818825 e-mail: [email protected] www.hygiena.net

(approximately 550 words including photographic annotation)

Continued …..

Page 52: Informacion Luminometria ATP Sector Salud

Press release Hygiene Monitoring Press Release - 30th July 2008 - 2 (Page 4 of 4)

GENERAL INFORMATION

SUPPLIER: Hygiena International Ltd

PRESS RELEASE: Number

HYG06

Type

General

Release Date

Immediate READER RESPONSE INQUIRIES TO:

Mr Martin Easter, General Manager of Hygiena International Ltd

PRESS OR PUBLISHING INQUIRIES TO: Steve Lloyd

Enterprise Marketing Services Ltd The Coach House 1 Dunstall Road Barton under Needwood Staffordshire DE13 8AX Tel: 01283 713185 Fax: 01283 716172 e-mail: [email protected]

ENCLOSURES:

REFERENCES: Qty

Code

SPLCHK

NOTE:

If the press release text and/or illustration are required in a digital format, please contact Enterprise Marketing Services Ltd for this to be supplied either by e-mail or on CD.

Photograph of a Quality Monitoring Officer using a handheld device on a patient bedside surface to be inserted here please. Suggested photographic annotation HYGIENA INTERNATIONAL LTD PRESS RELEASE HYG06 TBA – when illustration has been determined.

END

Page 53: Informacion Luminometria ATP Sector Salud

Title: Effects of sanitizers on Ultrasnap ATP Sample Testing Device Effective Date: June 2001 Summary • The majority of common sanitizers (5 µl at working strength) do not significantly affect

the results obtained with the Ultrasnap ATP sample testing device

• Ultrasnap should not be used where an acid sanitizer has not been completely removed by rinsing, as it may affect the signal

Purpose

To determine the effects of low levels of sanitizers on the results from the Ultrasnap ATP sample testing device. Procedure Sample Preparation: ATP (disodium salt) was diluted in sterile pyrogen-free water (PFW) from a certified stock solution (2x10-2 Moles) to a concentration of 2x10-7 Moles. 10 µl of this dilution was pipetted onto the Ultrasnap swab tip. Sanitizers were diluted in water to their recommended working strength and 5 µl amounts were pipetted onto the Ultrasnap swab tip (separately from the water or ATP solution). Assay methods Ultrasnap device activity was measured as follows: 1. Remove swab from swab tube 2. Pipette 10µl sterile pyrogen-free water or ATP solution directly onto the centre of the end

of each swab tip 3. Pipette 5µl sanitizer (diluted to working strength) or water (as a control) onto another area

of the swab tip 4. Replace swab tube and break the blue snap valve in two directions 5. Squeeze reagent reservoir twice to dispense the reagent 6. Shake gently to bathe the swab in the reagent for 10 seconds 7. Measure activity by inserting the device in to the luminometer This document and its contents are confidential and the exclusive property of Hygiena LLC. This document is not to be reproduced in any form whatsoever, without prior written permission from Hygiena LLC.

Page 54: Informacion Luminometria ATP Sector Salud

Materials Ultrasnap devices: Validation Batch

Lot Number FA-05-15-01-1

Luminometer: Prototype systemSURE II

Serial Number A6CW145K

Sanitizers: Neutral surfactants, slight free caustic Quat and surfactant Quat and Glutaldehyde Alcohol, Quat and Biguanide Amphoteric Neutral surfactants, free caustic Free caustic Neutral surfactants, phosphoric acid Hydrochloric acid

Kleencare Detergent Panel: AF123 or Topmaxx 123 DS607 or Triquart Super DS620 DS646 DS696 or Triquart AM NF421 or Topmaxx 421 NN4488 or MIP Betol SF520 or Topmaxx 520 SN570

ATP standard 2 x 10-7 M Results ATP Response to Sanitizers Effects The results from the Ultrasnap device are relatively unaffected by the majority of sanitizers evaluated which were chosen to include sanitizers incorporating a wide variety of active ingredients. Under the conditions of this experiment, some sanitizers gave a slight decrease in signal and other sanitizers a slight increase. The hydrochloric acid sanitizer at the concentration used in this experiment decreased the signal the most of those tested. However, it is not known whether this is a realistic test; sanitizers should be removed from a surface by thorough rinsing, and the amounts used in this experiment may be unnecessarily high. Table C6.1. Effects of sanitizers (5 µl at working strength) on response to ATP for the Ultrasnap hygiene monitoring device. Results are expressed as a percentage of the corresponding “no sanitizer” control.

Sanitizer Code / Name Active Ingredient RLU (% control) AF123 or Topmaxx 123 Neutral surfactants, slight free caustic 103% DS607 or Triquart Super Quat and surfactant 110%

DS620 Quat and Glutaldehyde 113% DS646 Alcohol, Quat and Biguanide 73%

DS696 or Triquart AM Amphoteric 77% NF421 or Topmaxx 421 Neutral surfactants, free caustic 87% NN4488 or MIP Betol Free caustic 114%

SF520 or Topmaxx 520 Neutral surfactants, phosphoric acid 79% SN570 Hydrochloric acid 65%

This document and its contents are confidential and the exclusive property of Hygiena LLC. This document is not to be reproduced in any form whatsoever, without prior written permission from Hygiena LLC.

Page 55: Informacion Luminometria ATP Sector Salud

Figs. C6.1-2. Effects of sanitizers (5 µl at working strength) on response to ATP for the Ultrasnap hygiene monitoring device. Graphs show individual results from 6 replicate tests.

Fig. C6.1. Ultrasnap: effect of sanitizers on ATP response

0

200

400

600

800

1000

1200

PFW AF123 5% DS607 1% DS620 1% DS646 Neat

RLU

Fig. C6.2. Ultrasnap: effect of sanitizers on ATP response

0

200

400

600

800

1000

1200

1400

PFW DS696 1% NF421 3% NN4488 1% SF520 3% SN570 1%

RLU

Conclusions • The majority of commonly used sanitizers (5 µl at working strength) do not significantly

affect the results obtained with the Ultrasnap hygiene monitoring device

• Ultrasnap should not be used where an acid sanitiser has not been completely removed by rinsing, as it may affect the signal

This document and its contents are confidential and the exclusive property of Hygiena LLC. This document is not to be reproduced in any form whatsoever, without prior written permission from Hygiena LLC.

Page 56: Informacion Luminometria ATP Sector Salud

Rapid tests of cleanliness deliver improvements and value

One of the top priorities of the NHS operating framework is improving cleanliness and reducing healthcare-associated infections. New technology is providing a way to simply measure the quality of

cleaning, the delivery of service levels and to ensure value for money. This technology, known as ATP bioluminescence, provides instant results and valuable management information that supports the concept of continuous measurable improvement.

Cleaning can mean different things to different people, anything from general tidiness to absolute sterility, so effective, consistent training is essential. The NHS Cleaning Manual recommends effective cleaning systems and procedures, and their implementation fulfils the requirements of the Health and Social Care Act and Care Quality Commission registration. The aims of the NHS Cleaning Manual are to provide best practice guidance on cleaning techniques and advice on defining responsibilities, scheduling work, measuring outcomes, and reporting and driving improvements. Teamwork and documentation are essential to provide assurance and evidence of due diligence, as it is sometimes said that if it is not written down then it did not happen.

The NHS has previously relied on visual assessments of surface cleanliness, but judging cleaning efficacy this way is subjective, and of questionable validity. However, the new, revised version of the NHS Cleaning Manual now recognizes that hygiene monitoring by ATP bioluminescence can provide an additional tool to monitor the delivery of cleaning services.

The ATP bioluminescence technology is a well–established technology with more than 30 years of proven use in several industries. It has been given category 1 status under the Rapid Review Panel, and is also used in sterile services department for cleaning verification of washer disinfectors and endoscopes.

The SystemSURE Plus ATP detection systems (Hygiena International) are being used in over 200 hospitals for several applications, including as a post-cleaning verification and monitoring tool from wards, in sterile service departments and catering facilities. It also makes a very good training and awareness tool for hand hygiene. It is simple and easy to use for non-technical staff everywhere, and most importantly, it has provided an objective yardstick to yield quantitative numerical data about cleaning levels attained.

In a case study conducted in the North Tees and Hartlepool NHS Foundation Trust shown on Health Exec TV (and also seen at http://www.hygiena.net/ind-healthcare.html), Kevin Oxley (Director of Operations) explains that they have introduced the Hygiena SystemSURE Plus to help combat the increase in HCAI:

‘We are seeing a growth in antibiotic-resistant bacteria and therefore we need to be able to validate our cleaning process to ensure that we can stop the spread of infections. Our cleaning scores have certainly improved since the introduction of the Hygiena SystemSURE Plus, and wehave seen a corresponding decrease within the number

of infections in our patients, so we feel strongly that it’s helping combat the increase that we are seeing elsewhere within the NHS. The system is being used across the Trust by our quality monitoring officers, by our ward matrons and by our domestic supervisors, so that we can have instantaneous feedback on the standards of cleaning at ward level. The reports generated by Hygiena SystemSURE Plus are issued to all the relevant people, e.g. nursing manager and matrons, so that we can all work together as a team to rectify any problems that occur.’

Working with link workers, Emma Davis (infection prevention and control nurse) said that the SystemSURE Plus:

‘highlighted to staff on the wards that it is actually really important they clean the equipment regularly and with the appropriate products. SystemSURE Plus is also incorporated into our hand hygiene training.’

It is now possible to detect invisible contamination, and have rapid meaningful information to enable managers to monitor the delivery of cleaning services and ensure value for money. Effective cleaning is a keystone of infection control, yet it is frequently taken for granted and viewed as burdensome on finances. However, it has been shown that providing effective cleaning services reduces infection rates and produces a cost–benefit analysis of >£56 000 per ward, per annum (Rampling et al, 2001). The cost of failure is not only measured as human suffering and additional medical costs, but there are also severe financial penalties for Trusts that can run into millions. Getting it right makes good sense: clinically and economically.

Rampling A, Wiseman S, Davis L et al (2001) Evidence that hospital hygiene is important in the control of methicillin-resistant Staphylococcus aureus. J Hosp Infect 49(2): 109–16

4

ADVERTORIAL

Housekeeping

MatronsCQCInfection

preventionand control

IC_advertorial.indd 20 6/11/09 09:34:12

Page 57: Informacion Luminometria ATP Sector Salud
Page 58: Informacion Luminometria ATP Sector Salud
Page 59: Informacion Luminometria ATP Sector Salud
Page 60: Informacion Luminometria ATP Sector Salud
Page 61: Informacion Luminometria ATP Sector Salud