POTENTIAL INFECTIOUS HAZARDS OF LABORATORY TECHNIQUES

III. VIRAL TECHNIQUES1

MORTON REITMAN, ROBERT L. ALG, WILLIAM S. MILLER, AND NOEL H. GROSS

Camp Detrick, Frederick, Maryland

Received for publication April 23, 1954

Workers engaged in viral or rickettsial in-vestigations are well aware of the numerouslaboratory infections that plague them. In theliterature are reports of laboratory infectionswith, to name a few, Psittacosis, Poliomyelitis,Hepatitis, Equine Encephalomyelitis, Q fever,Epidemic Typhus, Scrub Typhus, and RockyMountain Spotted Fever (Sulkin and Pike,1951). Each laboratory has set up its ownstandard operating procedures (Smadel, 1951).Unfortunately, the standards of safety areusually limited at some points by the fundsavailable for equipment. As a result, in somelaboratories and with some organisms, it is aforegone conclusion that sooner or later many ofthe personnel would become infected with theagent studied. Realization of this fact is adeterrent to otherwise desirable research. Manylaboratories rely heavily on a strong immuniza-tion program. Immunization is desirable whenavailable, but better protection of the laboratoryworker is achieved when immunization is com-bined with a policy of containment of theorganisms. The first step in this direction is adetermination of where and how the infectiousparticles gain access to the environment. Thispaper is the third in a series of papers devoted tostudies on infectious hazards of laboratorytechniques (Reitman et al., 1954a,b).An investigation has been made of the aerosols

produced and of the extent of contaminationduring the following viral and rickettsial tech-niques: (1) diluting viral particles in vaccinebottles, (2) mouse inoculation via the intranasaland intracerebral routes, (3) egg inoculation viathe allantoic and yolk sac routes, (4) inoculationof the chorioallantoic membrane by the windowtechnique, (5) harvesting allantoic and amnioticfluid, and (6) the maceration of mouse braintissue in the Ten Broeck grinder.

1 Presented in part at the 53rd General Meetingof the Society of American Bacteriologists held inSan Francisco, Calif., August 10-14,1953.

Coliphage T-3 was used as the test organism.Air and surface samples were taken by means ofsieve type air samplers (DuBuy and Crisp,1944) and cotton swabs, respectively. Thesampling medium used was tryptose-phosphate-glucose agar for the sieve sampler plates andtryptose-phosphate-glucose broth for the cottonswabs. Air samples were taken at the rate of onecu ft per minute. These plates were then coveredwith a layer of Escherichia coli, strain B, seededtryptose-phosphate-glucose agar (0.7 per centagar) and incubated 18 to 24 hours at 37 C atwhich time plaque counts were made. Surfacesamples taken by cotton swabs moistened withnutrient broth were immersed and shaken inmelted tryptose-phosphate-glucose agar whichwas then seeded with E. coli, strain B, andpoured into tryptose-phosphate-glucose agarplates.The operational area was surrounded by

sieve samplers (figure 1) by placing 3 samplerson the table and suspending 4 samplers 8 to 12inches above the table. Control air and surfacesamples were taken before the beginning of eachoperation.

VACCINE BOTTLE DILUTION METHOD

In the first experiment 9 tenfold dilutions of aphage suspension containing approximately2.6 X 109 infectious particles per ml were madein vaccine bottles containing 9 ml of tryptose-phosphate-glucose broth. Two ml tubercuilinluer-lok syringes equipped with 23 gauge three-quarter inch needles were used to transfer oneml amounts of the diluted suspension from onebottle to another. A fresh syringe and needlewere used for each dilution. In one series therubber diaphragms were swabbed with 70 percent ethanol before making the dilutions, whilethe other series in addition to the initial ethanolswabbing the needle was surrounded by anethanol soaked cotton pledget which served thetwofold purpose of surrounding the point of

549

~~~~MORTON REITMAN ET AL.

Figure 1. Air sampling by means of sieve type air samplers during intranasal inoculation of mice

with T-3 coliphage.

withdrawal and of swabbing the diaphragm

after the needle was removed.

Table 1 shows that when no attempt was made

to surround the needle with a pledget, a small

amount of aerosol was produced, as shown by air

sampling, in 6 out of 10 tests. The number of

plaques obtained ranged from 10 to 1. The use

of an ethanol soaked pledget entirely elimninated

the aerosol production.

Table 2 shows that dilutions in which con-

TABLE 1

T-3 coliphage recovered from the air* during and

after making tenfold dilutions of phage

suspensions in vaccine bottles

Number

~~Total Number of Plaques

of Tests No cotton pledget Ethanol cotton

around needle pledget used

1 10 0

1 9 0

1 3 0

3 1 0

4 0 0

*Rubber diaphragm of the vaccine bottle was

swabbed with 70 per cent ethanol before the

needle was inserted.

All controls were negative for the test virus.

taminated diaphragms occurred ranged from10-9 to 10-3 when no pledget was used, and 56out of 90 diaphragms were contaminated with thetest viruses. The pledget reduced diaphragm

TABLE 2T-3 coliphage recovered from rubber diaphragm*

after making tenfold dilutions of phagesuspensions in vaccine bottles

Highest Dilution in Which ContaminatedDiaphragm Found

Numberof Tests No cotton Nubr Ehnlcto

pledget around ofmtest ptanledgettuseneedle oftss pegtud

1 10'9 1 1-

2 10-7 3 10-21 10-6 3 10-14 10-5 3 01 10-41 107-3

Total Number of Contaminated Diaphragms

561 lit

*Rubber diaphragm of the vaccine bottle wasswabbed with 70 per cent ethanol before theneedle was inserted.

t Out of 90.All controls were negative for the test virus.

550 [VOL. 68

....

INFECTIOUS HAZARDS OF LABORATORY TECHNIQUES

contamination to 11 out of 90, and no diaphragmwas found to be contaminated beyond the 10 4

dilution.

ANIMAL INOCULATION

Intranawal route. In the next experiment 10mice were anesthetized with ether and inoculatedintranasally with 0.05 ml of coliphage T-3. Aseries of 10 tests was done, each test usingincreasing 10-fold dilutions of phage. Table 3shows a summary of the results obtained. Theextent of aerosol production was directly propor-tional to the concentration of the inoculum.Starting off with undiluted phage suspension,each mouse receiving an inoculum of approxi-mately 1.5 X 108 phage particles, a large amountof aerosol was produced which decreased sharplywith each 10-fold dilution until it was practicallynegligible at the 10- dilution level (inoculum1.5 X 10w). Hand and table contamination wasmuch greater than air contanation. Since themouse was held in the operators' hand, and it inturn rested on the table, these areas receivedthe brunt of aerosol contamination and any of thedrippings that may have inadvertently con-

taminated the mouse fur.Intracerebral inoculation. In the next experi-

ment 10 mice were anesthetized with ether andinoculated intracerebrally with 0.03 ml of a

phage suspension by means of a 27 gauge one-

quarter inch needle. Each mouse received ap-proximately 3.3 X 107 particles. In series A the

TABLE 3Summary of aerosols produced during inoculation

of mice with varying amounts of T-S coli-phage via the intranasal route

Average Num-ber of Phage

Particles Recov- Extent of ContaminationNumber of ered Per Test as Shown by Means of

Phage Particles as Shown by Cotton SwabsInoculated Plaque Countsper Mouse Obtained by

Means of SieveType Air Hand TableSamplers

1.5 X 108 270 heavy heavy1.5 X 107 19.9 heavy heavy1.5 X 106 8.6 heavy heavy1.5 X 106 2.4 moderate moderate1.5 X 104 0.5 light light1.5 X 103 0.1 trace trace

* For each dilution, 10 tests were made with10 mice per test.

TABLE 4Recovery of T-S coliphage during and after

intracerebral inoculation of mice with3.8 X 107 particles

Sieve Surface Samples with Cotton SwabsType

Samples Site of Technicians' TableSape injection hands TalSeries

Aveeno.rae No. contam- No. contam- No. contam-plaques inated inated inatedPer 10 No. injected No. tests No. testsmice

A 1.1 100/100 4/10 6/10B 0.2 53/100 3/10 2/10C 0.3 6/100 1/10 2/10

A-Inoculation site swabbed with 70 per centethanol before inoculation.B-Inoculation site swabbed with 70 per cent

ethanol before and after inoculation. Needlesurrounded by 70 per cent ethanol soaked cottonpledget.

C-Inoculation site swabbed by 70 per centethanol before inoculation, swabbed with 2 percent tincture of iodine after inoculation. Needlesurrounded by 70 per cent ethanol soaked pledget.

* Ten mice per test.

injection site was swabbed with 70 per centethanol prior to injection; in series B the injec-tion site was swabbed with ethanol before andafter injection, and the needle was surrounded byan ethanol soaked cotton pledget; and in series Cthe injection site was swabbed with ethanol be-fore injection, with two per cent tincture ofiodine after injection, and the needle was sur-rounded by an ethanol soaked pledget.Table 4 shows that the intracerebral inocula-

tion of 100 mice produced only slight aerosols asshown by recovery of viral particles by the sievesamplers. The average recovery of phage pereach 10 mice injected was 1.1 in series A, 0.2 inseries B, and 0.3 in series C. However, the site ofinoculation was grossly contaminated in 100per cent of the animals injected when no attemptwas made to decontaminate the inoculation site.This was reduced to 53 per cent when 70 percent ethanol was used to decontaminate the siteand the needle was surrounded by an ethanolpledget. The use of 2 per cent tincture of iodinetogether with an ethanol pledget reduced con-tamination to 6 per cent. The extent of con-tamination of the technician's hand and the

19541 551

MORTON REITMAN ET AL.

table top likewise was reduced by the use of thedisinfectants and ethanol pledget.

EGG INOCULATION

Allantoic and yolk sac inoculation. In the nextexperiment 12 day old embryonated eggs wereinoculated in the allantoic sac and in the yolk sac.Shells were swabbed with 2 per cent tincture ofiodine before inoculation, and 0.25 ml of phagewas inoculated into the allantoic sac with a 23gauge three-quarter inch needle. The inoculationhole was then sealed with Duco cement. Surfacesamples were taken of the egg shell, technicians'hands, and egg tray. Ten eggs were inoculatedper test. In the 10 tests that were made, phagewas recovered in the air samples in very smallamounts in only 2 tests. Only 4 out of 100 eggshells were contaminated. Phage was recoveredfrom the technicians' hands in lare amountsafter 2 tests, and no recoveries were made fromthe tray.Yolk sac inoculation was accomplished by

introducing 0.25 ml of phage suspension with a23 gauge 11 inch needle through the air sac ofiodine swabbed eggs. Egg shells were sealed asdescribed before. Table 5 shows that as in theallantoic method only a slight aerosol was pro-duced as judged by air sampling. Only 4 plaquesappeared on the plates from one test. However,shell contamination was considerable.When this experiment was repeated using an

ethanol pledget around the needle, only 1 plaquewas obtained on the air sampling plates and only4 out of 100 eggs were contaminated.

ChorioaUantoic membrane inoculdion. Aninoculum of 0.05 ml containing 1 X 10'/ml wasdropped onto the chorioallantoic membrane, andthe opening was then sealed with scotch tape.

TABLE 5Recovery of T-X coliphage from egg shells after

allantoic and yolk sac inoculation with T-8

Number of Egg Found to be Contaminated After

Yolk sc inoculationAllantoic scinoculation Without ethanol With ethanol

pedget pledget

4/100t 23/100 4/100

* By means of cotton swabs.t Number contaminated.

Number inoculated

A total of 100 eggs were inoculated, 10 per test.No recoveries of phage were made on air samplerplates. Swabbing of egg shells revealed heavycontamination of the shell area around thewindow in two instances and light contamina-tion in one. There was no contAmination of theoperator's hand or the egg tray in which the eggswere held during inoculation.

HARVESTNG

Allantoic fluid. Embryonated eggs were inocu-lated with 0.5 ml of a 1 X 109/ml phage suspen-sion; the shell was decontaminated with 70 percent ethanol, chilled for two hours, and thenharvested. Five eggs were harvested for eachtest. Air was sampled for 10 minutes. The al-lantoic fluid was removed via the air sac byaspirating with a 13 gauge needle and 5 mlsyringe. Results are shown in table 6. A total of57 phage particles was recovered on the air sam-pling plates during the harvesting of 50 eggs.The egg shells and egg trays were found to beheavily contaminated with the test virus. Fifty-three phage particles were recovered from thehands of the operator.

Amniotic fluid. Embryonated eggs wereinoculated through the air sac with 0.5 ml of a1 X 109/ml phage suspension by means of a 1ij/inch needle. The shell was decontaminated with70 per cent ethanol, chilled two hours, and

TABLE 6Phage particles recovered during harvesting of

allantoic fluid

TestNo."

12345678910

Number ofPhcge Parti-

on AirSpgPlates

110241830352

Recover of Phag Parti-cles from Surfacest

Hand1 Egg tray

002201820632

1TNTCTNTC

54TNTCTNTC

220

TNTCTNTC

Shell

TNTCTNTCTNTC127

TNTCTNTCTNTCTNTCTNTCTNTC

Phage CountPer St

5.5 X 1071.7 X 10'3.9 X 1072.9 X 10'2.2 X 1072.4 X 10'1.0 X 1082.1 X 1081.7 X 10'2.3 X 10'

* Five harvestings per test.t One swab was used per test.t Count of pooled allantoic fluid from 5 eggs.TNTC-Too numerous to count.

[voL. 68552

INFECTIOUS HAZARDS OF LABORATORY TECHNIQUES

TABLE 7Phage particles recovered during harvesting of

amniotic fluidNumberof Phs'g'e Recovery of Phage Particles

Test Recov- frOm Surfacet Phage CountNo.* ered on Per MI:Air-- _

Sampling Ha-d Egg tray ShellPlates

1 0 TNTC TNTC TNTC 1.1 X 1082 0 0 TNTC TNTC 8.6 X 1073 0 3 TNTC TNTC 2.3 X 1084 0 3 TNTC TNTC 2.1 X 1085 7 2 TNTC TNTC 1.4 X 1086 0 0 43 TNTC 4.1 X 10'7 0 0 TNTC TNTC 3.7 X 1058 1 0 TNTC 8 5.5 X 1089 0 0 TNTC TNTC 1.6 X 10510 2 TNTC TNTC TNTC 6.0 X 107

* Five harvestings per test.t One swab was used per test.t Count of pooled amniotic fluid fromTNTC-Too numerous to count.

then harvested. Five eggs were hanremoving the shell over the air sac, Clshell membrane, decanting the allaninto 70 per cent ethanol, pulling up themembrane with forceps, and aspiriamniotic fluid with a 13 gauge inch n5 ml syringe. Results are shown in titotal of 10 phage particles was rece

TABLE 8Recovery of phage particles during the ma

mouse brain tissue in a Ten Broecki

Test No.

1

2345678910

Number ofPhage

ParticleRecoveredon AirSamplngPlates

1

229

143149911

Recovery of Phage PartSurfaces

Hand

0

0

0

0

0

0

0

30

5

Table

0

0

0

0

0

0

0

0

0

0

* One swab was used per test.TNTC-Too numerous to count.

means of air sampling in the 10 tests. The eggtrays and egg shelLs were heavily contaminatedwith the test virus while the hands of the operatorwere found to be heavily contaminted in 2 outof 10 tests.

Maceration of mouse brain tissue. One normalmouse brain was placed into the bottom of aTen Broeck grinder, and 0.25 ml of a 2.5 X 108/mlphage supension was added. The tissue wasthen ground for 3 minutes. The average phagecount of the homogenized tisue was 3.3 X 107/g.Table 8 shows that a total of 83 phage particleswas recovered by air sampling during themaceration of 10 mouse brains. The plungerhandle was found to be conta ted in 6 out of10 tests while the operator's hands were con-taminated in 2 tests.

DISCUSSION

The inherent hazard in the use of vaccinebottles for making dilutions of infectious ma-

5 eggs. terials lies in the contamination of the rubberdiaphragm. This contamination was greatly

vested by reduced by the use of an ethanol soaked cotton

atting the pledget around the needle and by swabbing the

Ltoic fluid diaphragm with alcohol. The production of slightamitic{lulaerosols by the vaccine bottle method was com-

^ Lotic pletely eliminated by the use of the ethanoldtig the pledget around the needle.Leedle and Intranasal inoculation of mice produced largeable 7. A amounts of aerosols together with extensivevered by contamination of the environment. The only

solution to this problem would seem to lie in theuse of a ventilated cabinet to contain the organ-

ceration of ism and the use of ventilated cages to house theIrinder inoculated animals. The safest method for intra-tices fo cerebral inoculation involved the use of an

ethanol soaked pledget around the needle to-plw~ gether with decontamination of the injection sitehanleof with two per cent tincture of iodine.gindr Allantoic and yolk sac inoculation of embryo-

4 nated eggs produced only slight aerosols, butTNTC shell contamination by the latter method was

0 considerable. This could be quite a problem when74 egg incubators equipped with fans are used.0 Shell contamination was greatly reduced by the53 use of an ethanol soaked pledget around the0 needle.0 Inoculation of the chorioallantoic membrane93 by the window technique appears to be a rela-18

tively safe procedure except for the possibility ofshell contamination which occurred in twoinstances.

55319541

MORTON REITMAN ET AL;

Harvesting of infected allantoic and amnioticfluids appears to be a hazardous procedure.There was heavy surface contamination of theegg trays, shells, and hands of the operator.Maceration of tissue in the Ten Broeck grinder

produced aerosols of the test virus. Surface con-tamination was confined to the plunger handle.The rewlts of these experiments indicate the

need of carrying out viral techniques in a protec-tive ventilated cabinet (Wedum, 1953).

The vaccine bottle method of making dilutionsof virus produced slight aerosols which wereehminated by surrounding the syringe needlewith a 70 per cent ethanol soaked cotton pledget.Fifty-six out of 90 rubber diaphragms were foundto be contaminated with the test virus, coliphageT-3. Intrnasl inoculation of mice is an ex-tremely hazardous procedure. Large amounts ofaerosols were produced by this technique.Intracerebral inoculation of mice producedsurface contamination of the injection site.Insignificant aerosols were produced by allantoic,yolk sac, and chorioallantoic (window technique)

inoculation of embryonated eggs, but shellcontamination by the former two techniques wasconsiderable. Harvesting of infected egg fluidsproduced extensive contamination of the workingarea.

REFERENCESDuBuy, H. G., AND Cmsp, L. R. 1944 A sieve

device for sampling airborne microorganisms.Public Health Repts., (U. B.) Suppl. no.184, 1-39.

RITMAN, M., Moss, M. L., HARSTAD, J. B., ALO,R. L., AND GRoss, N. H. 1954a Potentialinfectious hazards of laboratory techniques.I. Lyophilization. J. Bacteriol., 68, 541-544.

RIITAN, M., Moss, M. L., HAEsTAD, J. B., AwG,R. L., AND GRoss, N. H. 1954b Potentialinfectious hazards of laboratory techniques.II. The handling of lyophilized cultures.J. Bacteriol., 68, 545-48.

SMADEL, J. E. 1951 The hazard of acquiringvirus and rickettsial diseass in the labora-tory. Am. J. Public Health, 41, 788-795.

SULIN, E. S., AND PIKE, R. M. 1951 Survey oflaboratory acquired infections. Am. J.Public Health, 41, 769-781.

WEDUM, A. G. 1953 Bacteriological safety.Am. J. Public Health, 48,1428-1437.

554 [Vo>. as