validasi-autoklaf

7/29/2019 validasi-autoklaf

1/8

utoclaveValidation

PHARMACEUTICAL ENGINEERING JULY/AUGUST2002Copyright ISPE 2002

Practical Guide to AutoclaveValidation

by Raymond G. Lewis, PE

In addition topotentialbusinessliabilities, therecan be significantcosts associatedwith an autoclave

validationprocess. Thepracticalexperience thatthis article isbased on mayprovideassistance inensuring aneffective, efficientvalidation

process forsteamsterilization.

Introduction

This a rticle is based on pra ctical experi-

ences gained by the author while con-

ducting hun dreds of val idat ion test r uns

on dozens of a utoclaves of varied ma nufa cture.

I t is primari ly intended that personnel who

perform validation testing on autoclaves may

benefit from th ese experiences, an d th at it will

a ssist in ensu ring a high level of compliance inthe va lidat ion process. The a rticle also may be of

benefit in selecting an appropriate validation

stra tegy a nd/or cycle. Personnel unfam iliar with

stea m sterilizat ion principles or a utoclave vali-

dat ion could use the material as a basic training

tool and it may be a good refresher for more

experienced personnel. A list of definitions a nd

references ar e provided at the end of the a rticle.

Sterility Assurance LevelThe level of microbial inactivation can be de-

scribed by an exponential function, Sterility

Assuran ce Level or SAL. For exam ple, a S AL of

10-6 means t ha t t he probabi l ity of a single viable

microorganism being present on a sterilized

item/product is one in one million a fter th e item

ha s undergone a sterilizat ion process. A SAL of

10-3 means t ha t t he probabi l ity of a single viable

microorganism being present a fter sterilizat ion

is one in one thousa nd.

The SAL required is determined by the in-

ten ded use of th e item/product. S ter ilizat ion

processes associat ed with pa rentera ls and medi-

cal devices tha t pose a significant risk in terms

of the probability and severity of an infection

(e.g., implant s, sterile fluid pat hw ay s, products

intended t o come into conta ct wit h compromised

tissue) generally ha ve been st erilized to an SAL

of 10-6. Medica l device products n ot intend ed to

come int o conta ct wit h brea ched skin or compro-

mised tissue ar e generally sterilized to a SAL of10-3.

The remainder of this article is written as-

suming tha t a SAL of 10-6 is required.

Log ReductionAchieving a 1-log reduction means to decrease

the microbial population by a factor of 10. The

bioburden is the number and type of viable

microorgan isms conta mina ting an it em. A ster-

ilization cycle tha t provides a S AL of 10-6 effec-

t ively means that the microorganisms that

could be present (i.e., bioburden) are killed,

an d a n a dditional 6-log reduction sa fety fa ctor

has been provided. The following provides an

example of a cycle achieving a SAL of 10-6.

B ioburden (worst case) = 134 CFU (colony

forming unit ).

To reduce the microbial populat ion from 134

to 1 = log (134) = 2.13 (i.e., a 2.13-log red uc-

tion is required to reduce the

populat ion from 134 to 1).

Apply ing an addit iona l 6-log

reduction will theoretically

reduce the microbial popu-lation from 1 to 0.000001.

This pr ovides a S AL of 10-6 or

a one in one million prob-

ability of a single surviving

microorganism.

Tot a l l og red uct i on = 2. 13 + 6

= 8.13. Therefore t o provide

a S A L o f 1 0 -6 w i t h a

bioburden of 134 CFU re-

quires a sterilization cycle

th at provides an 8.13 log re-

duction.

Figure 1. Empty chamber

temperature mapping

(Photograph provided courtesy

of Kuhlman Technologies Inc.)


2/8

AutoclaveVali

JULY/AUGUST2002 PHARMACEUTICAL ENGINEERINGCopyright ISPE 2002

Thermal Resistance CharacteristicsThe th erma l resista nce of specific microorga nisms is chara c-

terized by D -va lues an d Z-va lues. A D-value is the t ime in

minutes, at a specific temperature, to reduce the surviving

microbial population by 1-log. A Z-value is the temperature

chang e required to result in a 1-log reduction in D-value.

Other t ime measurement variables perta ining to thermal

resistance are F-values and F o-va lues. An F -value is the

number of minutes t o kill a specified number of microorga n-

isms w ith a specified Z-va lue at a specific tempera tur e. An F o-value is t he num ber of minutes t o kill a specified num ber of

microorganism s with a Z-va lue of 10C (50F ) at a t emperat ure

of 121.1 C (250 F).

Common Misconception and EquivalentSterilization Time

It is not uncommon to encounter the concept that 121.1C

(250F) is the temperature required for steam sterilization.

This underst a nding is not ent irely correct. E xtensive empirica l

studies w ere conducted an d one of th e critical va ria bles (tem-

perature) was pre-selected. It is not surprising that the tem-

perature selected was an obvious round number in the tem-

perat ure ra nge of interest (250F). The F o-value equa tion can

be used to determine the relative sterilization time at other

tempera tur es as per th e following (with Z-va lue = 10 C):

F o = 10(T 121. 1) /10

where T = tempera ture ( C) and F o = equivalent steri l iza-

tion time (min.)1

Table A provides some exam ples a nd the relat ionship follows

in gra phical form in F igure 2.

As is demonstrated by the data above, sterilization can be

achieved using any of these temperatures. The lower the

tempera tur e the longer th e sterilizat ion cycle required. This is

an importa nt concept t o consider because th ere are occas ions

where the temperature needs to be carefully selected. Anexample is a l iquid that cannot withsta nd high temperatures.

Ideal ly , the highest temperature tha t the load can w ithsta nd

is selected, since this will provide the shortest possible cycle.

Variables Required to Determine anIdeal Sterilization Cycle

An ideal sterilization cycle presumes an ideal sterilizing

environment (i.e. , satur a ted stea m w ith n o air). The ideal cycle

can be determined with t he followin g thr ee va ria bles: bioburden,

D-value, and required SAL. The following provides some ex-

amples:

a) G iven: B ioburden = 75 CFU , D-value = 0.5 min./log at

121.1 C, Required S AL = 10-6Then: Log (75) = 1.88

Log Reduction r equired = 1.88 log + 6 log = 7.88 log

Idea l Cy cle at 121.1C (250 F) = (7.88 log)(0.5 min./

log) = 3.94 min ut es

b) G iven: B ioburden = 1,215 CFU, D-value = 1.6 min./log a t

121.1 C, Required S AL = 10-6

Then: Log (1215) = 3.08

Log Reduction r equired = 3.08 log + 6 log = 9.08 log

Idea l Cy cle at 121.1C (250 F) = (9.08 log)(1.6 min./

log) = 14.53 minut es

Overkill ApproachDetermining the bioburden and D-value for all i tems to be

sterilized in a load ca n be quit e time consuming an d costly. As

a result , for i tems tha t a re not heat sensit ive, an overki ll

a pproach is generally employed.

An overkill approach avoids collecting bioburden and D-

va lue dat a by a ssuming w orst-case conditions. A bioburden of

106 of a h ighly heat resista nt spore forming bacteria (Ba cillus

stearothermophilus) is utilized. The D-value at 121.1C for

these bacteria is generally slightly above 2 minutes, andth erefore using 2.5 minutes is a good w orst-case va lue.

With a bioburden of 106, to a chieve a SAL of 10-6 requires a

12 (6 + 6) log reduction. Un der ideal conditions, t he length of an

overkill ster ilizat ion cycle a t 121.1C is th erefore (12 log)(2.5

min ./log) = 30 minut es.

Bioburden and D-Value ApproachFor i tems that are heat sensi t ive and cannot withstand an

overkill approach, it is necessary to collect bioburden and

possibly D-value data. This will dramatically shorten the

sterilizat ion cycle required. For example, if th e bioburden is

low (e.g., 10 CFU ) and even m odera tely resista nt (e.g., D-value

= 0.5), an id eal 30-minu te overkill cycle at 121.1 C can be

repla ced by a n id ea l cycle of 3.5 minu tes (7 log x 0.5 min ./log).

Alterna tively, the sterilizat ion t emperat ure could be reduced

Figure 2. Equivalent sterilization time.

Figure 3. Ideal cycle time.


3/8

utoclaveValidation


to 112C and yet only require slightly less than a 30 minute

ideal cycle. I t may be a signi ficant a dvant age to reduce the

sterilizat ion t emperat ure a nd/or time.

A compromise approach m a y sometimes be utilized wh ere

bioburden da ta is collected, but D -value st udies ar e not per-

formed. A worst case D-value of 2.5 could then be employed.

This approach will provide a somewhat shortened cycle and

avoids the time and cost of D-value studies. Following our

example with a bioburden of 10 CFU , the ideal cycle a t 121.1C

can be shortened from a 30-minute overkill cycle to a 17.5-

min ut e cycle (7 log x 2.5 m in./log).

Figure 3 shows the st eriliza tion time required a t 121.1C for

an ideal cycle to achieve a SAL of 10-6 at varying levels ofbioburden (D-va lue = 2.5 min.).

Vacuum and Non-Vacuum CyclesPreviously, this article has addressed ideal cycles that pre-

sume an idea l sterilizing environment. In term s of the length

of cycle required, one can only a pproach idea l cycles for item s

tha t a re easily st erilized. Most often, items/loa ds w ith less

tha n ideal conditions ar e encount ered.

There are three basic types of cycles as follows:

a) Ha rd Goods (Vacuum):

Suitable for items easy to sterilize since air removal and

steam penetra t ion a re highly effect ive. Examples are ma ny

types of glasswa re a nd lar ge diameter piping. A typical har d

goods cycle may draw one vacuum prior to introducing

steam, r eaching t he desired st eril izat ion tempera ture, a nd

beginning th e sterilizat ion dw ell period. A typical pressure

vs. time gra ph for a h ar d goods cycle is shown in Figure 4.

b) Wra pped G oods (Vacuum):

Ut ilized for items difficult t o sterilize since air removal a nd

steam penetrat ion are harder to achieve. Examples are

gowns, long lengths of tubing, a nd ta nks/vessels/appa ra tus

wit h sm a ll inlet/outlet port s a nd/or vent filt ers. A ty pical

wr apped goods cycle may d ra w t hree or more vacuums prior

to reaching the desired sterilization temperat ure an d begin-ning the sterilization dwell period. A post sterilization

vacuum also is usual ly draw n to evacuate the steam from

the load items. Often t he length of time to pull an d release

the va cuums exceeds the length of the sterilizat ion dw ell. A

typical pressure vs. time gr a ph for a wr apped goods cycle is

shown in Figur e 5.

c) Liquids/G ra vity Displacement (Non-Vacuum):

Items that contain l iquids general ly cannot have a deep

vacuum pulled or th e liquid w ill be draw n out of the item.

Liquid cycles generally just heat up an d cool down a nd do

not utilize vacuums. These items may require a lengthy

cycle time especially where the liquid volume is large

becau se the length of time required to heat up an d cool down

the liquid may be considerable. Another term for a liquid

cycle is gravity displacement as the air is displaced by

gra vity (i.e. , removing air by int roducing st eam int o the top

of a chamber an d displacing the air , w hich is heavier tha n

stea m, by removing the a ir from the bottom of the cha mber).

A typical pressur e vs. time gra ph for a liquids cycle is shown

in Figure 6.

Basic Validation ApproachInstallation Qualif ication (IQ )

The IQ process is int ended to demonstra te th a t th e aut oclave

a s insta lled meets a ll specifica tions, is inst alled properly, andthat the supporting programs needed for ongoing operation

(e.g., standard operating procedures, maintenance program,

etc.) are in place.

An IQ ma y include the following checks:

Mecha nical Equipment Speci ficat ions (chamber, valves,

traps, s trainers, f i l ters , regulators, vacuum pump, heat

excha nger, condenser, etc.)

Control and Ins t rumenta t ion S peci fi ca t ions (programmable

logic cont roller, opera tor in terf a ce, printer /recorder, cont rol

valves, transducers, pressure and temperature transmit-

ters, resistance temperature devices, switches, level sen-

sors, int erlocks, ph otocells, etc.)

Si te Specificat ions/Ut i li t ies (power, grounding, surge pro-

tector, unint erruptible power supply, breakers, wa ter, air,

clean steam, plant steam, drain, shutoff /isolation valves,

electrical disconnect switches, etc.)

Dra wings Veri ficat ion (P &ID, mechan ical , electrical)

Const ruct ion Mat er ia l s/Mater ia l s in P roduct C ontac t

Approval D ocumentat ion (e.g. , pressure vessel, electrical ,

etc.)

C h a ng e/S pa r e P a r t s

B i ll of Ma t er ia ls

Vendor Speci fi ca t ion Sheets

P u rch a se Or der s

F a ct ory Per forma nce Te st s

C ommis s ion i ng Repor t P r ev en t iv e Ma i nt en a n ce P r og ra m

Sta ndard Opera t ing Procedures (opera t ing , maintenance ,

calibration)*

O per a t in g a n d M a in t en a n ce Ma n u a ls

P iping Insta l lat ion Verificat ion (slope, dead legs)

Weld Inspection/Sur face Roughness Documenta tion/Met-

allurgical Documentation

Control System Documentat ion (system configurat ion/block

diagr a m, flow sheets, display/report layouts, r equired inter-

lock considerations, general process limits, conditions for

operat ing over ran ge, har d copy a nd electronic applicat ion

code listing, t iming dia gra m, syst em security, input /output

point l is t ing, dat a monitoring, alarms, softwa re inventory

Table A. Equivalent sterilization time.

Temperature Fo Equivalency to 121.1C (250F)

115C (239F) 0.25 min. 1 minute at 115C provides the same lethality as 0.25 minutes at 121.1C


121.1C (250F) 1 min. 1 minute at 121.1C provides the same lethality as 1 minute at 121.1C

122C (251.6F) 1.23 min. 1 minute at 122C provides the same lethality as 1.23 minutes at 121.1C



4/8

AutoclaveVali


and version, software configurations, parameter listings,

softwa re development a nd t esting records, chan ge control,

vendor qua lification, modular softw ar e development docu-

ments, deta iled m odule functiona l specificat ions, et c.)

Ins t rumenta t ion and Input/Output Dry Loop and Wet Loop

Checks**

P I D Tun in g**

I n st r um en t C a lib ra t i on s**

* Operat ing Procedures can only be final ized after Perfor-

mance Qualifications tests are completed when vali-da ted load configurat ions a nd cycles are known.

** Note: in some approaches, these checks a re capt ured as

initial Operational Qualification activities.

Operat ional Q ualif icat ion (OQ)

The OQ process is intended to demonstrate that the compo-

nents of the a utoclave operate properly and tha t t he aut oclave

is deemed rea dy for performa nce or load test ing.

An OQ may include the following checks:

Opera t iona l Tests (opera t or/supervisory/ma intena nce

modes, doors, abort a nd emergency stop, a larm s, progra m-

mable parameters, menu navigation, security, power-up

an d shut down, opera tor interfa ce display checks, interlock

override cont rol, procedure select/sta rt cont rol, step a d-

van ce control, switch a nd int erlock tests, etc.)

Po w er Los s Recovery Tes t

S ou r ce C od e R ev iew

F ilt er S t er ili za t i on

Leak/Air Removal/St eam P enetra tion/Vacuum Hold Test*

J a cket Ma ppin g

S a t ur a t ed S t ea m C heck

E m pt y Ch a m ber Tes ts

* The Bowie Dick test is designed to test a ir removal , the

a bsence of air leaks a nd stea m penetra tion into a porous

load . It u ses a t est pa ck of fabr ic wit h specific dimensions

or there ar e commercial, use once packs a va ilable. It ha sbeen widely employed in Europe. In North America, a

Vacuum H old Test ha s often been employed. Eur opea n

St an da rd E N 554 specifies tha t if a sterilizat ion process

includes air removal from the product, a st eam penetra -

tion test shall be carried out at the commencement of

each da y the a utoclave is used. Although a va cuum hold

test ma y be less sensit ive than a Bowie Dick test , the

aut hor assumes tha t a vacuum hold test can be consid-

ered as a sat isfactory al ternative i f s tric t acceptance

criteria a re applied. This assu mption is based on steam

penetra tion/letha lity in t he worst case load items being

demonstra ted and t hat the vacuum hold test therefore

demonstrates absence of leaks and that the val idated

condit ions t hat resul ted in letha l i ty ar e being met on anongoing ba sis.

Empty Chamber Distribut ion Tests (Figure 1)

The ba sic objective is to show the chamber provides a uniform

sterilizing environment. In the opinion of the author, cold

spots in au tocla ves ar e ra rely encount ered. Sometimes cold

therm ocouples a re misint erpreted a s cold spots (refer to fol-

lowing section Tips).

Three consecutive successful runs are performed for each

cycle type with typical a cceptan ce criteria as per the following:

Throughout the dwel l t ime, a l l tempera tures measured in

Figure 4. Hard goods cycle.

Figure 5. Wrapped goods cycle.

the chamber are within a 3C ba nd (steri lizat ion temperature

+ 3C ).2 Note: th e dwell set-point -1C /+ 2 C is often used.

Throughout the dwell t ime a l l tempera tures measured in

the chamber do not f luctua te by more than 1C. 2

Throughout the dwel l t ime, a l l tempera tures measured in

the cham ber do not differ from each other by more than 2C .2

The s team i s a t a t empera ture corresponding to i t s vapor

pressure.2

The interva l of t ime between the a t t a inment of the s ter i li za-

t ion temperature in the hottest and coldest parts of the

chamber does not exceed 15 seconds for chambers of not

more than 800L and not to exceed 30 seconds for larger

chambers.2

Timed measurements shal l be control led to an a ccuracy of

1%.2

Required pre-cert i f icat ion and post-cert i f icat ion of the dat a

logger ensures th at the temperature m easurement system

is accura te to within 0.5C.

The vacuum hold test should achieve a vacuum level of 2.5

psia (with va cuum pump) an d ma intain t he vacuum (with-

out further vacuum being initiated) within 0.4 psi over a

period of five minutes.

Performance Qualificat ion (PQ)

Loaded Chamber Steam Penetration Tests

Loaded chamber steam penetrat ion r uns a re then conducted

on every load. Note: this is a very time consuming process,

especially if you have a significant number of items to be

sterilized. It is necessa ry to determine which load it ems are th e

most difficult to sterilize and which location(s) within the

items presents the w orst-case conditions.

There a re tw o commonly used methods for determining t he

worst-case items/locations, t hermocouples, an d stea m integra -

tors. Steam integrators a re commercial ly ava i lable strips tha t

provide a q ua nt ita tive indica tion of the exposure to stea m. The


5/8

utoclaveValidation


am ount of stea m exposure can be determined by measuring

the movement of a chemical indicator on the integrator

strip. The author recommends utilizing steam integrators

since they are designed to measure steam exposure and

therm ocouples can result in misleading da ta (i .e. , measur -

ing temperat ure without ta king into account whether t here

is any air present).

Determining which load items are the most difficult to

sterilize and which location(s) within the items presents the

worst-ca se conditions can be a da unt ing ta sk. With a lar ge loadcontaining a wide variety of different types of items, the

number of possible test locat ions seems t o approach infinity. I t

also can be difficult t o get t he th ermocouple a nd/or stea m

integra tor into the item wit hout ad versely affecting the item's

ability t o be sterilized a nd/or ruining t he item (a concern w ith

expensive items).

One must evaluate an item on a case-by-case basis and

determine how best to challenge the item. Often th e item must

be sealed somehow to return th e i tem to a sta te tha t represents

equivalency with respect to stea m penetra tion. No a tt empt will

be made t o provide an exha ustive commenta ry h ere, but rat her

provide a few basic techniques for answering questions that

inevitably arise:

What is the most di ff icult point to steril ize in a h ose of

uniform diameter? Common sense can somtimes assist,

dicta ting in this insta nce tha t th e most difficult to sterilize

point is in th e center of the hose.

How do you get a 10-foot length of thermocouple and/or

stea m int egrat or into th e middle of a 50-foot hose? You can

put a slice/cut into t he middle of the hose an d insert the

th ermocouple/int egra tor th rough t he slice. Note: the cut

must be sealed or you will not be challenging the hose

properly. You can use silicon to seal t he cut. Alterna tively,

if tw o 25-foot lengt hs of th e hose are a va ilable you ca n join

the two lengths with a connector and insert the thermo-

couple into the connector. The connector then must besealed. The a dva nta ge here is tha t you dont ruin t he 50-foot

hose. The connector technique ca n be used for sma ll diam-

eter tubing w here the hose is too sma ll to insert a therm o-

couple an d/or stea m integra tor.

What i s the wors t-case loca t ion w i thin a bot t le, f lask , or

cylinder? This ha s been shown to be in the cent er, near, but

not at the bottom.

How can you minimize the number of runs requi red to

cha l lenge a load? Using st eam int egrators can h elp mini-

mize the nu mber of runs r equired to challenge a load. There

are a limited number of thermocouples available, but as

many integrators as desired can be placed in the load.

Load Configurat ionsAnother va riable of concern is w hether fixed load configura-

tions or flexible load configurations are desired. A fixed load

configuration means that the load to be sterilized will be

identical for all future processing runs and that the load is

placed in the chamber in exactly the sa me wa y for al l future

processing ru ns.

In t he opinion of the aut hor, the location of a n item in th e

chamber d oes not influence its ability t o be sterilized (ass um-

ing tha t t he location chan ge does not involve a chan ge in load

density). This observation is based on the experiences of the

author in conducting hundreds of validation test runs on

dozens of aut oclaves of varied ma nufa cture. However, one

should proceed as if the location within the autoclave is a

va riable of concern. One can eliminat e this var iable by rota ting

the i tems within a load from run to run and t hereby at tempt to

demonstrate positional equivalency.

For most load s, aga in in the opinion of the au thor ba sed on

experience, the number of items in the chamber does not

influence an items ability to be sterilized (unless the load

becomes so dense th a t st eam penet ra tion/circulat ion becomes

a n issue). One should proceed as if t his is a var iable of concern.You can su ccessfully validat e a load w hile encompas sing th is

si tuat ion by performing minimum a nd ma ximum load studies.

The followin g provides a n exa mple of fixed vs. flexible loa d

configurations:

E xa m ple loa d:

- t hree (3) f la s k s

- four (4) graduated cy l inders

- 24 plas t i c bot t les w i th vent f i lters

F ixed Load/Fixed Pos it ion:

In this situation, all of the load items are placed in the

au tocla ve, each t ime in the sam e position for each item, a nd

a diagram of the load configurat ion is avai lable in theprocedures so tha t t he opera tors can reproduce the load for

every processing run. This situat ion w ill require the least

va lidation run s, but offers no flexibility in load configura-

tion.

F ixed Load/Variable Pos it ion:

In this situation, all of the load items are placed in the

au toclave, but t he location of the item in the a utoclave can

vary an d only a l is t of the load i tems is required for the

procedures. The validation runs must demonstrate posi-

tional equivalency by rotating the items from location to

location during t he test r uns. It ma y be possible to accom-

plish th is with t he same number of val idat ion runs a s a bove

and offers the operators some flexibility in loading theautoclave. This can be an advantage especially for large

loads containing numerous items.

Var iable Load/Variable Pos it ion:

In this situation, any or all of the load items (i.e. , any

combin a tion of from 0 to 3 fla sks, from 0 to 4 cylinder s, from

0 to 24 bottles) can be pla ced in t he a utoclave in an y position

in the a utoclave and only a maximum load l ist is required

for the procedures. The valida tion runs mu st demonstr at e

positional equiva lency by rota ting t he items from location to

location during t he test run s. The valida tion runs a lso must

demonstra te tha t t he cycle is adequate for both a maximum

Figure 6. Liquids cycle.


6/8

AutoclaveVali


loa d an d minimum load configurat ion. The minimum load

tests a re done with only one item in the aut oclave, tha t item

being the load item demonstra ted a s being the most difficult

to sterilize. This meth od will require the grea test n umber of

validation runs, but offers the operators a great deal of

flexibility in loadin g th e au toclav e. This can be a s ignifican t

advantage in many si tuat ions.

Loaded Chamber Biological Challenge Tests

After determining the worst-case items and worst-case loca-tions within items, these items are then challenged with

biological indica tors (spore strips a nd/or vials for pla cement

wit hin liquids). A thermocouple should be placed along w ith

each indicator, as the temperature data wi l l be required to

extra polat e the cycle to achieve the SAL of 10-6.

Tests are conducted until a cycle time results in three

consecutive runs w here th e biological indica tors show n o growt h.

If it is important to achieve the shortest possible cycle, this

process can consume a great deal of time as t o determine t he

success/failure point likely r equires obta ining fa iled t est re-

sults along with successful test results. In addition, it takes

time to determine wh ether th e indicators exhibit growt h (a fter

tw o days of incuba tion you can be reasonably confident w hether

there is growth or not in most cases). If a few minutes of

possibly unnecessa ry t ime added t o the cycle is not a significa nt

issue, it can be a dva nta geous to at tempt t o predict a cycle time

that you feel will pass. This can save considerable time and

validation costs.

Once one ha s a chieved thr ee consecutive run s resulting in

no growth and therefore demonstrating a 6log reduction

(a ssuming y ou were using indicat ors of 106 spores /st rip), th e

following equa tions/example show h ow to extra polat e the full

cycle requ ired t o achieve th e SAL of 10-6:

La = [12 x (Fo/R)] - Fo

w he re La = t he a d di t iona l l et ha l it y (F o) required

12 = used to extra polat e a 12-log reductionFo = the minimum accumula ted F o va lue from the bio-

logica l cha llenge run s a t t he end of the cycle

R = the log reduction demonstra ted (i .e. log [spore

population])

Fi = 10(T-121.1)/10

w h er e F i = t h e in st a n t a n eou s F o value

T = the minimum temperature expected during the

additional lethality period (Note: this tempera-

ture should be taken a s the t emperat ure achieved

at the end of the dwell period at the challenge

location where the minimum a ccumula ted F o value

resulted)

Ta = La/Fi

where Ta = the addit iona l t ime required

C = Ta + D

w he re C = t ot a l d w el l per iod t i me req u ired

D = the dwell period t ime which resul ted in the dem-

onstrated reduction

Exa mple Calculat ion:

The biological cha llenge run s wer e performed using spore

strips tha t w ere enumerat ed at 1.21 x 106 spores /st rip.

Ther efore R = log (1,210,000) = 6.08

The minimum accumulated F o value (at the end of the

cycle) from th e biological challenge ru ns w as 30.2 minutes.

Therefore Fo = 30.2 minut es

La = [12 x (30.2/6.08)] - 30.2 = 29.4 min ut es

The temperature in the coldest item at the end of the

dwell period w as 119.4 C

Therefore T = 119.4C

Fi = 10(119.4-121.1 )/10 = 0.676

Ta = La /Fi = 29.4/0.676 = 43.5 minu tes

The biological challenge runs were conducted with a

dw ell period of 45 minu tes. Therefore D = 45 minu tes

C = 43.5 + 45 = 88.5 minutes (note: this num ber should be

rounded up)

Therefore t he dw ell period must be 89 minutes t o achieve

a 12-log reduct ion.

Three consecutive successful biological challenge runs

are performed for each load with typical acceptance

criteria consistent with the empty chamber distribution

test a cceptan ce criteria a nd a ll biological indicators used

during the t est cycle must show negat ive growth .

Tips1. If you are going to draw a vacuum(s), ensure that the load

i tems can withst and the va cuum(s). You dont w ant to be the

person w ho has to report tha t t he new $10,000 ta nk is now

as f l a t a s a pancake .

2. Rota te thermocouples from run to run. This avoids misin-

terpreting th ermocouples tha t read slightly lower tempera-

tu res (i.e., cold th ermocouples) as cold spots or cold items.3. Label the thermocouples by number using a sma ll s trip of

aut oclave ta pe. This wil l great ly assist w i th ensuring tha t

you ar e properly recording w ha t t hermocouple wa s placed in

each location and will save validation time.

4. I f you are performing a large number of test run s (e.g . , over

the course of severa l w eeks), st rike a compromise between

post-calibra tion verification of t hermocouples a fter every

run a nd a t t he end of the entire test ing period. I f you wai t

unt il the end of th e testing period, you run the risk tha t a ll

of the runs ar e of no value due t o not meeting th e verification

acceptance criteria. If you verify after every run, you will

ad d considera bly to th e lengt h of time required t o complete

the test ing. The author has found that performing the

verifica tion every few run s or every few da ys is a rea sonablecompromise.

5. B e caut ious w ith th e acceptance cri teria you employ for

post-calibration of thermocouples. If the criterion is too

tight (e.g., all thermocouples must meet the acceptance

criteria), you may lose a lot of runs if one or two thermo-

couples cease functioning or a re outside of the tempera tur e

tolerance aft er the ru n(s).

6. Take great care with documenting the val idat ion test runs.

The documenta tion should include: a dia gra m showing t he

location of all load items wit hin th e aut oclave cham ber, the

items conta ining th ermocouples, integra tors a nd biological

indica tors, t he precise loca tion/num ber of ea ch ther mo


7/8

utoclaveValidation


couple, int egrat or an d biologica l indicat or with in each item,

the printout from the da ta recorder, the printout or chartfrom the autoclave, the time that the dwell period begins

an d ends (a s per the dat a r ecorder time), an d the results for

each integrator or indicator. Each document should be

clearly la beled with t he dat e, test ru n number, etc. If you fail

to generat e good documenta tion wh ile conducting the ru ns,

you w il l not be able to recover when a nalyzing the da ta /

putting together the report, and you will end up with

inadequat e or poor qua l ity da ta to support the va l idat ion

process.

7. A therm ocouple should alwa ys be placed beside the dra in

tempera ture sensor (usual ly a drain temperat ure sensor is

used to control the temperat ure wit hin th e aut oclave).

Cautions1. If you are using a non-vacuum cycle to sterilize a n on-liquid

load, you are taking a significant risk. Some regulatory

bodies will simply not allow processing of non-liquid loads

with non-va cuum cycles.

2. Some regulatory bodies ar e extremely concerned tha t a l l

points within the load achieve sterilization temperature

when sta rt ing the dw ell period. This may mean t ha t you are

not draw ing enough va cuums or t hat modificat ions to th e

items being st erilized a re necessary to a llow more efficient

steam penetrat ion.

3. I f you are not using biological indicat ors to val idat e your

cycle, you a re ta king a significant risk. Using t emperat ure

dat a alone means tha t you ar e assuming ideal condit ions

wh ere it is not justified.4 . I f you are placing a smal l quant i ty o f water w i thin load

items to assist with sterilization, you must have appropri-

at e procedura l cont rols in place to ensure ongoing consis-

tency with the am ount of wa ter present during th e val ida-

tion runs and all subsequent processing runs.

SummaryThe requirements to validate steam sterilization processes

ha ve been document ed for ma ny yea rs. For example, perha ps

the most historically significant reference guide, the PDA

Technical Monograph No. 1 Validat ion of St eam St erilization

Cycles wa s published in 1978. Nonetheless, stea m st erilization

validation remains a significant issue to regulatory bodies,

par ticularly for processes associat ed with h igh risk in terms of

the probability and severity of an infection. Failure to ad-

equa tely addr ess this requirement can place the public at risk

an d lead to regulat ory citat ions/a ction.

In addition to potential business liabilities, there may be

significan t costs a ssociat ed with th e validat ion process. Large

numbers of time consuming and costly test runs may be

required, and if appropriate consideration is not given to

employing t he correct a pproach, unnecessary ongoing opera -

tional costs may result.

It is hoped tha t th e pra ctical experience tha t th is document

is based on will provide assistance in ensuring an effective,

efficient va lidat ion process for stea m sterilization an d tha t the

end result provides th e best possible validat ed cycle to meet th e

needs of the specific application.

DefinitionsSAL: sterility assurance level.

SAL of 10-6: th e proba bility of a single viable microorga nismbeing present is one in one million.

Bioburden: th e number /ty pe of viable microorga nism s con-taminat ing an i tem.

Overkill Approach: a steri l izat ion approach based on as-suming worst-case conditions (a bioburden of 106 of a highly

heat r esista nt ba cteria) .

Log Reduction:reduce the surviving microbial population by

1 log or decrease th e surviving populat ion by a factor of 10.

12-Log Reduction:th e log reduction required a chieving over-

kill and a SAL of 10-6.

CFU: colony-forming unit.

D-value:time in minut es, at a specific tempera tur e, to reducethe surviving microbial population by 90%(one logarithmic

reduction).

Z-value: temperature change required resulting in a 1-logreduction in D -value.

F-value: the n umber of minut es to kill a specified number of

microorganisms wit h a specified Z-value a t a specific tempera -

ture.

Fo-value: the n umber of minut es to kill a specified num ber ofmicroorganisms w ith a Z-value of 10 C (50 F) at a t emperat ure

of 121.1 C (250 F).

1 Fo:the equ ivalent of 1 minute a t 121.1C (250F).

Dwell Period:the t ime period th at begins w hen th e aut oclave

temperature has reached the set-point and ends when thetimer has expired.

Worst case items: i tems in the load which are the mostdifficult to sterilize (as det ermined by steam penetra tion stud-

ies).

Worst case location: the locat ion within a n i tem tha t is the

most difficult t o sterilize (as determined by stea m penetra tion

studies).

Gravity Displacement: a method of removing air by intro-ducing steam into the top of a chamber and displacing th e air ,

...steam sterilizat ion validat ion remains a significant issue to regulat ory bodies,

part icularly f or processes associat ed with high risk in t erms of t he probabil it y

and severit y of an inf ection.


8/8

AutoclaveVali

JULY/AUGUST2002 PHARMACEUTICAL ENGINEERING

which is heavier than steam, by removing the air from the

bottom of the cha mber.

Vacuum Cycle: a sterilization cycle that draws one or morevacuums t o remove air prior to star ting t he dwell period.

Pre-vacuum: a vacuum drawn prior to start ing the dwellperiod t o remove air.

Post-vacuum: a vacuum drawn after the dwell period has

finished to remove stea m.

Hard Goods Cycle:a sterilizat ion cycle designed for items forwh ich a ir removal is not difficult an d therefore genera lly one

pre-vacuum is dr aw n.

Wrapped Goods Cycle: a sterilization cycle designed for

items for which a ir removal is difficult a nd t herefore generally

thr ee or more pre-va cuums are dra wn .

Liquids Cycle:a cycle designed for liquid loads th at generally

uses gravi ty displacement ra ther tha n draw ing a vacuum.

Bowie Dick Test:a test designed to verify tha t a n a utoclaves

vacuum phase is removing a sufficient amount of air prior tothe introduction of steam into the chamber and tests for air

leaks into the chamber.

Empty Chamber Tests:tests with an empty cha mber essen-tial ly designed to demonstra te th at an aut oclave provides a

uniform sterilizing environment .

Steam Penetration Tests:load ed cha mber tests designed to

determine the w orst-case items a nd w orst-case locations wit hin

a load.

Biological Challenge Tests: load ed cha mber tests designedto challenge the w orst-case locat ions (within worst case items)

wit h biologica l indicators t o demonst ra te t he effectiveness of asterilizat ion cycle.

Steam Integrators: commercially available indicators thatprovide an indication of exposure to steam .

Fixed Load: a load configurat ion where the quanti ty and

location of items w ithin t he chamber a re fixed.

SIP: steam-in-place or sterilize-in-place (often used inter-changea bly a lthough t he level of microbial destr uction achieved

ma y differ).

References

1. Validat ion of Steam Steri l izat ion C ycles: PDA TechnicalMonograph No. 1 (1978).

2. Steri l izat ion of Medical Devices Val idat ion a nd Routine

Control of Steri lizat ion by Moist Heat : European Sta ndar d

EN 554 (1994).

About the AuthorRaymond G. Lewis, PE is a Val idat ion Man-ager for Industria l Design a nd Construction

Inc. (IDC) in Portland, Oregon. In this posi-

tion, he is responsible for int erna tional cG MP

regulat ory issues involved in the d esign, con-

struction, a nd val idat ion of pharma ceutical ,

biotech, a nd medica l device fa cilities. He h olds

a degree in chemical engineering from the

Universi ty of Saskatchewan and a Cert i f icate in computerscience from th e Un iversity of Regina . Lewis ha s 16 year s of

experience in the pha rm a ceutical/biotechnology indu str y in

va lidat ion, engineering, production, fa cility services, and com-

puter opera tions.

IDC, 2020 SW 4th Ave., 3rd Floor, Portland, OR, 97201.

Documents

validasi-autoklaf