large mixtures chargedthrough following pneumaticexperiment. sterile swabs were dipped into these...

STUDIES ON STEAM STERILIZATION AND THEEFFECTS OF AIR IN THE AUTOCLAVE

ANSON HOYT, ALBERT L. CHANEY AND KORINE CAVELLDepartment of Bacteriology, School of Medicine, University of Southern California

and the Los Angeles County Hospital, Los Angeles, Cal.

Received for publication August 1, 1938

Underwood (1937) has recently emphasized a number of thelimitations of autoclave sterilization in a series of very practicalarticles. The work herewith reported involves experiments uponthe detrimental effects on efficient steam sterilization of air inthe autoclave.An autoclave of special design was employed. This apparatus

consisted of a small double-jacket sterilizer with separate boiler;the inlet and outlet valves were large so that pressure could bebuilt up or released within 10 seconds. A total time of 30seconds sufficed for the evacuation of over 99 per cent of the airand the attainment of the desired pressure of saturated steam,-afact that was verified by numerous experiments. Figure 1 showsa diagram of the lay-out.

In order to run tests involving known mixtures of air and steamthe discharge line of the autoclave was connected to a modifiedpneumatic trough and in this way the contents could be dis-charged through water, the steam condensed and the air measuredby displacement. Since part of the air contents remained in thesterilizer even after it had reached atmospheric pressure followingthe first discharge of air and steam into the pneumatic trough,it was necessary to sweep out the remaining air by means ofmore steam before making the final measurement. This use ofsaturated steam to flush out residual air had the disadvantage ofexerting an additive sterilizing effect which was quite marked athigh pressures, under which conditions it often sufficed to sterilizespores which should otherwise have remained viable. In experi-

639

on February 19, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

640 A. HOYT, A. L. CHANEY AND K. CAVELL

ments at 20 pounds pressure with no air discharged from theautoclave it was found possible to dispense with this complicatingfactor and to measure the air only at the beginning and end ofeach series of tests. In other experiments at lower pressures,

Chamber Gauye Jacket GaugeC X~~~~~~~~~~~~~C

FIa. 1. DIAGRAMTIC DRAWING OF AUTOCrAVEThe boiler is not shown. The Steam Control for Chamber represents a valve

which passes steam from the jacket into the chamber. The Jacket is alwaysdirectly connected with the boiler, so that the Jacket Gauge represents boilerpressure. The boiler and jacket can carry a full head of steam with the chamberempty. The rectangle within the chamber represents a rack designed to carrytwo rows of test swabs placed horizontally at six different levels indicated by thetwo rows of dots. "T. C." shows the approximate positions of the three thermo-couple leads within the chamber. The left Exhaust Valve is shut and the rightExhaust Valve opened when one desires to empty the autoclave through thepneumatic trough. Both valves are opened to empty the autoclave past thepneumatic trough. The Steam Inlet opens into the middle of the back of thechamber. The exhaust valve opens from the bottom of the front of the chamber.

however, the air was evacuated and measured at the end of eachrun before the test swabs were removed from the autoclave.The percentage of air in the steam was calculated as follows:

The air was actually measured at room temperature and pressurein the pneumatic trough; the volume it would occupy at the


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

STEAM STERILIZATION AND AIR IN AUTOCLAVE

temperature and pressure existing in the autoclave was thencalculated by means of the gas laws. The amount of air inthe steam was then expressed as a percentage,-for instance,25 per cent air means that at the temperature and pressure inthe autoclave the steam and air are respectively 75 per cent and25 per cent by volume.The following is a sample calculation for air volume. The air

collecting cylinder was of the same diameter as the inside ofthe autoclave and the autoclave chamber had a total length of20 inches. Hence 5 inches of air in the cylinder would be A ofthe total volume of the autoclave chamber. At 10 pounds gaugepressure and a temperature of 1100C. this A of the chambervolume would be changed to5 X 14.7 lbs. X 2730 + 110° X 100 per cent or 19 per cent20O 10 lbs. + 14.7 lbs. 2730 ± 250

The temperatures within the autoclave were measured bymeans of three thermocouples which were customarily located atthree different levels, one near the top, another one in the centerand a third close to the bottom of the chamber. The thermo-couples could be connected in turn by means of a multiple switchto a direct reading galvanometer. By maintaining the temp-erature of the constant junction at 100'C. in boiling water thetemperature could be read quickly and accurately. When anyconsiderable percentage of air was retained in the autoclave thelower thermocouple would always record a temperature of a fewdegrees below that registered by the upper two which were inclose agreement. In this case the temperature for correcting theair volume reading was taken from the center thermocouple.

Clostridium oedematiens was employed as the main test or-ganism. This bacterium is considered by a number of observersto be one of the most heat-resistant of any of the spore-formingorganisms of surgical importance, an observation which wasconfirmed by one of us (Hoyt, 1934). Two strains were obtainedfrom the American Type Culture Collection (C. oedematien8,No. 681 and C. novyi, No. 3539). Both liquified gelatin butfailed to digest coagulated serum. Both fermented glucose,

641


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

A. HOYT, A. L. CHANEY AND K. CAVELL

dextrin and maltose with the production of acid and gas andboth failed to ferment mannitol, salicin, lactose and sucrose.C. novyi produced acid and gas in glycerol while C. oedenmatiensformed acid only. C. oedematiens proved more virulent for micethan did C. novyi. Their resistance to heat was, for practicalpurposes, identical when both organisms were employed in anumber of the earlier experiments. In later work C. oedematiens,No. 681 was generally utilized but we have considered bothorganisms as different strains of C. oedematiens and they aredesignated as such in our experimental results.A local strain of Clostridium botulinum A was also employed,

mainly to check the experimental technic against a spore-formerof extremely high heat resistance.' The resistance of Clostridiumwelchii and Clostridium tetani was also tested in a few instances.

All bacteria were grown for a minimum of 3 days in chopped-veal infusion broth and each batch of cultures was checkedmicroscopically for spore formation before being utilized in anexperiment. Sterile swabs were dipped into these tubes andthen usually inserted into a rack which held 12 swabs, the rackwas placed in the autoclave and the swabs were directly exposedeither to saturated steam or to mixtures of steam and air. Atthe end of the period of sterilization, the autoclave was rapidlyemptied of steam and each swab was removed and cultured in atube of chopped-veal infusion broth which had recently beenboiled and allowed to cool; the swabs were then discarded and thecultures were overlaid with sterile melted vaseline. Each culturewas labeled with the date and the position of its swab in theautoclave. Control cultures of unheated organisms were madeat the time of each experiment.The cultures were incubated at 370C. for a minimum of one

month before being discarded. This was an arbitary time limitand we realize from the work of Dixon, Burke, Beck and Johnston(1925) and other reports that occasional spores may lie dormantfor much longer intervals than this even when placed in a medium

1 We gratefully acknowledge the cooperation of Dr. Ivan C. Hall of the Schoolof Medicine, University of Colorado in checking this culture for us.

642


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


suitable for growth. In our results occasional tubes inoculatedwith C. botulinum A spores showed growth only after beingincubated for periods ranging up to a month or more but theC. oedematiens cultures almost never became positive after aninterval of more than 12 days.Growth in the cultures was clearly evidenced by the pro-

duction of gas which forced the vaseline away from the surfaceof the liquid. Smears were made from positive tubes chosen atrandom from time to time and stained by Gram's method tocheck the cultures for correct morphology. Very occasionalcultures showed irregular evidences of growth, usually with aminimum of gas production; smears made from most of suchtubes disclosed contaminants, usually staphylococci or diph-theroids. All tubes shown to contain contaminants were omittedfrom the tabulations of experimental results. Experiments wereperformed on 40 different days and involved a total of over3,000 cultures exclusive of unheated controls.

EXPERIMENTAL RESULTS

Saturated steam served as the sterilizing agent in the earlierexperiments and pressures of 6, 10, and 15 pounds correspondingto temperatures of 110, 115, and 121'C. were employed. Thesetemperatures within various parts of the autoclave were checkedby thermocouples during every test and variations from thevalues indicated were negligible. Tables 1, 2 and 3 summarize24 experiments upon the sterilizing effects of saturated steam onspores of Clostridium botulinum A and Clostridium oedematiens.

Tables 1, 2 and 3 indicate that a direct contact of saturatedsteam with spores of C. botulinum A and C. oedematiens requiredthe following minimum times to kill these organisms completely:

C. BOTULINUM A C. OUDUMATIENS

minutes minutes

6 pounds saturated steam, 11000.Over 30 1010 pounds saturated steam, 11500.10 415 pounds saturated steam,12100.4 1

643


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


TABLE 16 pounds saturated steam, 1100 C.

C. BOTULINUM A C. OEDEMATIENS

tm TMOFPercentagesTME0Total Cultures of cultures EXPOSURE Total Cultures of t

cultures showing ahog cultures showing shoftregrowth shwn growth gotgrowth got

minute

30 30 100 2j 24 24 100

't 4- 5 45 38 84

6 41 36 8836 36 100 7i- 8 84 12 14

1 10 69 0 0

126 102 75 11-15 36 0 0108 67 62 16-20 12 0 018 3 17 21-2512 1 9 30

TABLE 210 pounds saturated steam, 1160C.

C. BOTUIUYNM A C. OEDEMATIENS

,are ~TIME OFeoetWTota Cultures of cultures EXPOSURE Total Cultures Percentages

cultures showing showing cultures showing shofutrgrowth 'growtht

mninutea30 30 100 1 36 36 10030 30 100 2 120 11 946 42 91 3 120 1 146 32 70 4 36 0 046 22 48 5 36 0 046 13 28 6 36 0 046 14 30 7 36 0 046 6 13 8 35 0 034 3 9 9 24 0 026 0 0 1016 0 0 1112 0 0 12

The times required to sterilize spores of C. oedematiens wereabout one third of the times required for complete killing ofC. botulinum A spores under like experimental conditions. The


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

STEAM STERILIZATION AND AIR IN AUTOCLAVE 645

results here reported on the heat resistance of C. botulinum Aclosely approximate the maximum values obtained by Esty andMeyer in 1922.A check was made in a few experiments upon the relative

resistance of moist and dried spores of C. oedematiens and C. bot-ulinum A. To obtain dried spores, sterile swabs were dippedinto spore-filled cultures of the organisms in question and thenplaced in sterile empty test tubes; these tubes were replugged

TABLE 315 pounds saturated steam, 12j1C.


Totl CltuesPercentages TIEOPecnasTotal Cultures of cultures EXPOSURE Total Cultures of cultures

cultures showing showing cultures sgowinhg showinggrwth grothgrowth growth

minutes

i ~~24 2 535 32 91 1 36 0 018 13 72 1j 24 0 036 8 23 2 24 0 017 3 182536 1 3 315 1 5 325 0 0 412 0 0 430 0 0 512 0 0 66 0 0 76 0 0 56 0 0 9

and incubated at 3700. for one or more days. They were thenexposed to steam in parallel with swabs that had recently beendipped into cultures of the same total length of incubation. Inevery instance it was evident that dried spores were considerablyless resistant to heat than were the moist spores,-a fact whichhas been noted by previous workers (Esty and Meyer, 1922).

Moist spores of locally isolated strains of Clostridium welchiiand Clostridium tetani were also tested for their resistance tosaturated steam in a few instances. With the strains employed


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


C. welchii was definitely less heat-resistant than C. oedematiensbut C. tetani approximated the resistance of C. oedematiens veryclosely.The remaining experiments concerned the effects of residual

air in the autoclave. In work with saturated steam, the spore-covered swabs had been inserted vertically into the sterilizer,which meant that the test organisms were located approximatelyat the center level of the chamber. In the following experimentsthe swabs were inserted horizontally at six different levels rangingfrom near the top to near the bottom of the sterilizing chamber.The upper two levels approximated the upper thermocouple, themiddle two the middle thermocouple and the lower two swabscorresponded to the level of the lower thermocouple. C. oede-matiens was the test organism employed in these experiments.

Unsaturated steam is in no way as efficacious a sterilizingagent as saturated steam. This was definitely manifested invarious ways which will be discussed separately.A. Air in the autoclave lowers the internal temperature to a con-

siderable extent. This lowering of temperature is an unevenphenomenon inasmuch as air is heavier than steam and tends tostratify beneath the steam.

Six experiments were performed with an autoclave pressure of10 pounds and with approximately 25 per cent of air in thesterilizer. As mentioned above, the test swabs occupied sixdifferent levels and temperatures were measured by thermo-couples at 3 different levels in the sterilizing chamber. In allthese tests the three thermocouples in saturated steam indicateda temperature of 115'C.-+1lC., while the temperatures shownwith 25 per cent air approximated 1140C. for the upper thermo-couple, 1130C. for the middle and 108'C. for the lower thermo-couple. The cultural results in the various tests were fairlyuniform and the following protocol is considered as representativeof this entire series of experiments.Table 4 shows that 25 per cent air in the autoclave at a gauge

pressure of 10 pounds caused the time required for sterilizationof any swabs to be at least twice that of the controls in saturatedsteam; this differential between unsaturated and saturated steam

646


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

STEAM STERILIZATION AND AIR IN AUTOCLAVE 647

was increased to at least five times when only the lowest swabswere taken into consideration.Two experiments with 15 pounds pressure and 25 per cent air

bore out the above findings. All control swabs were sterilizedby 1 minute of exposure to 15 pounds of saturated steam whereasover 6 minutes was required to sterilize all swabs when 25 percent air was present at this pressure.

B. A mixture of air and steam at any temperature is not asefficient a sterilizing agent as saturated steam at that same temp-

TABLE 4Test organism-C. oedematiens

10 POUNDS STEAM WITH 25 PER CENT CONTROLSAIR IN AUTOCLAVE 10 POUNDS SATURATED STEAM

Thermooouple readings

Upper 1l4C. Upper, middle andMiddle 11300. lower 11500.Lower 108lC.Cultural results

Number of swabs showingnumberof growth

Time of exposureTotal number Number of swabs

numberosieo of swabs showing growthLower Middle Upper

2 minutes 12 03 minutes 12 0

12 4 4 4 4 minutes12 4 4 0 6 minutes12 4 0 0 8 minutes12 1 0 0 10 minutes

erature. Four experiments bearing on this point were performedand the results of the different tests fell into close agreement.For unsaturated steam 20 pounds gauge pressure was employedand all the air possible (about 42 per cent) was left in the sterilizer.Swabs at different levels were placed in direct contact with thisunsaturated vapor in the test runs. In control experimentssaturated steam at 6 and 10 pounds gauge pressure was used.Table 5 shows the results of a representative experiment.

Table 5 demonstrates two findings of particular interest:-(1)The temperature in the lower part of the autoclave was recorded


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


as 115'C. both when 20 pounds gauge pressure of steam with42 per cent air and when saturated steam at 10 pounds pressurewere utilized for purposes of sterilization. In spite of this factit took over four times as long (10 minutes) to sterilize the testswabs with the unsaturated vapor as was the case (2 minutes)with saturated steam. (2) The temperature recorded with 6pounds of saturated steam was 110'C. Nevertheless the experi-ment indicates that 6 pounds of saturated steam in direct contactwith the test swabs proved to have sterilizing properties which

TABLE 5Test organism-C. oedematiens

CONTROLS20 POUNDS STZAM wrrH 42 PERs CENT

AIR IN AUTOCLAVE 6 pounds saturated 10 pounds saturatedsteam steam

Thermooouple readings

Upe 120°1C.Mpicddle ll9-1C. 1100C. llS°1C.Lower 116°C.

Cultural results

Total Number of swabs showing Total Number Total Numbernumber of growth Time of exposureumber of swabs number of swabsTotalbimso of showing of showingswabs Lower Middle Upper swabs growth swabs growth

2 minutes 12 012 4 0 0 4 minutes12 4 0 0 6 minutes12 4 1 0 8 minutes 12 212 0 0 0 10 minutes 12 0

were as effective as those shown by unsaturated steam at 20pounds pressure and possessing a minimum temperature of 115'C.These findings were borne out by the other experiments.

C. Air in the autoclave cuts down the penetration of steam to aconsiderable extent. Two experiments were performed upon therate of penetration of steam into uniform packages of rubbergloves. Each package consisted of a glove book containing twogloves. Two test swabs were thrust into the finger of eachglove. A cloth wrapper was folded loosely around each package

648


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


and tied with string. One thermocouple lead was inserted asclosely as possible into the center of each package while a secondlead ran to the outside. Only one package was employed ineach test run and it was placed in the center of the sterilizingchamber. Fifteen -pounds gauge pressure of steam with 25 percent air was contrasted with 15 pounds of saturated steam.Table 6 shows the results of one of these experiments. The

thermocouple readings indicate that with unsaturated steam ittook over 3 minutes for the inside of the loosely wrapped pack-

TABLE 6Experiment-Test organism, C. oedematiens

Air in the autoclave cuts down the penetration of steam to a considerableextent.

Uniform glove packets placed in autoclave. Each package contained 4 con-taminated swabs. One thermocouple lead ran to the center of each package,a second lead being placed outside each package.

16 POUNDS STEAM WITH 25 PER CENT CONTROLS: 15 POUNDS SATURATED STEAM

AIR IN AUTOCLAVE IN AUTOCLAVE

Exposure Cultures Growth Exosure Cultures Growth

mm. min.5 4 + 2 4-8 1 + 3 4-3-

11 ~~4-The inner thermocouple took from 3 The inner and outer thermocouples

to 4 minutes to come up to the outer came to the same reading of 1210C.thermocouple reading of 1150C. within one half minute.

ages to come up to the temperature of the free chamber space ofthe autoclave, whereas with saturated steam these temperaturesequalized themselves almost instantly. The cultural results showthat sterilization was completed by saturated steam in 2 minuteswhile unsaturated steam required longer than 8 minutes toaccomplish the same results.

DISCUSSION

It has long been realized that saturated steam under a pressureof 10 pounds or more is an extremely efficient sterilizing agent.

649


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


Direct contact with 10 pounds of saturated steam will probablykill all surgically significant spores within 6 minutes and shouldeven destroy the highly resistant spores of Clostridium botulinumwithin 20 minutes. When the pressure of saturated steam isincreased to 15 pounds, complete sterilization should be obtainedin approximately one third of the times mentioned above. Inactual practice a margin of safety considerably above thesetimes is customarily allowed for a number of reasons among whichare the following:

1. It is at least conceivable that some spores may exist whichare more heat-resistant than any yet reported.

2. Penetration of saturated steam into various packages orbundles may take considerable time even in a perfectly func-tioning autoclave.

3. Air is not always eliminated perfectly from the free chamberspace in the average autoclave. Even when such elimination ofair from the free chamber space has occurred, it often happensthat local pockets of air may be trapped within the materialsthat are being sterilized. Local areas of unsaturated steam willresult and sterilization in these local areas will be retarded.

4. Spores may become coated by various oily materials andthus be somewhat protected from free contact with saturatedsteam. The work of Dickson, Burke, Beck and Johnston (1925)appears to show that this is a point of considerable practicalimportance.The experiments here reported indicate that the complete

saturation of steam throughout an autoclave is a most importantfactor in the process of sterilization. If dilution with air takesplace the efficiency of the sterilizing process is considerablyreduced and this reduction is not merely a function of a reductionin temperature. It was clearly shown that a mixture of steamand air at a given temperature is a less efficient sterilizing agentthan is saturated steam at that same temperature.Our findings in this respect differ markedly from those recently

published by Savage (1937). This author placed dry spores indry glass tubes and inserted bits of cotton wool as sources ofsteam. These tubes were then evacuated to varying degrees,

650


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

STflAM STERILIZATION AND AIR IN AtTTOCLAVE

sealed and heated for definite intervals at varying temperaturesin a glycerol bath. Cultures were subsequently made and theresults showed that residual air in the tubes had no effect on thesterilizing process. One should note, however, that our experi-mental set-up approximates actual working conditions in anautoclave much more closely than was the case in the experimentswhich Savage reported.

SUMMARY AND CONCLUSIONS

Experiments were performed in an autoclave on various spore-forming anaerobic bacteria to compare the sterilizing effects ofsaturated steam with mixtures of air and steam.The autoclave was constructed to permit extremely rapid

entrance and discharge of steam, direct temperature readingscould be made from three levels in the chamber and the air con-tent of the chamber could be quickly determined. The chieforganisms employed were Clostridium oedematiens and Clostridiumbotulinum A.The following thermal death times were determined for direct

contact with saturated steam:


minutes minutes

6 pounds saturated steam, 11000.Over 30 1010 pounds saturated steam, 11500.10 415 pounds saturated steam,12100.4 1

Dried spores were found to be less resistant than moist spores.Air in the autoclave was shown to lower the efficiency of

sterilization in the following ways:A. Air in the autoclave lowers the internal temperature. This

is an uneven phenomenon inasmuch as air is heavier than steamand tends to stratify beneath the steam.

B. A mixture of air and steam at any temperature is not asefficient a sterilizing agent as saturated steam at that sametemperature.

C. Air in the autoclave cuts down the penetration of steam toa considerable extent.

651


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


REFERENCESDICKSON, E. C., BURKE, G. S., BECK, D., AND JOHNSTON, J. 1925 Studies on

the thermal death time of spores of Clostridium botulinum. Jour.Infec. Dis., 36, 472-483.

ESTY, J. R., AND MEYER, K. F. 1922 The heat resistance of the spores of B.botulinus and allied anaerobes. Jour. Infec. Dis., 31, 650-663.

HoYT, A. 1934 Studies on rubber glove sterilization and the use of sterilityindicators. Jour. Lab. and Clin. Med., 19, 832-390.

SAVAGE, R. M. 1937 Experiments on the sterilizing effects of mixtures of airand steam, and of superheated steam. Quarterly Jour. and YearBook of Pharmacy, 10, 451-462.

UNDERWOOD, W. B. 1937 Surgical sterilization. Hospital Management, 44,(Aug.) 17-18, (Sept.) 16-19, (Oct.) 18-21, 42, (Nov.) 18-21, 42.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

large mixtures chargedthrough following pneumaticexperiment. sterile swabs were dipped into these...

Documents