PENICILLIN RESISTANCE OF STAPHYLOCOCCUS AUREUS AND ITS CLINICAL
IMPLICATIONS
MAJOR HAROLD H. ] XIUGH, S N . C , A.U.S.
The pioneer investigations of the effect of penicillin
on pathogenic bacteria have established not only that
certain species are inhibited by exposure to the drug
while others are resistant, but in addition that certain
of the penicillin susceptible species notably staphylococcus and pneumococcus—gradually acquire resistance
after variable periods of exposure to the drug in vitro.
Rammelkamp and Maxon2 found that 29 strains of
Staphylococcus aureus were all killed by concentrations
of penicillin between 0.02 and 0.35 units per cc; that is,
these strains showed remarkable constancy in susceptibility to the agent. These investigators and also
McKee and Houch4 have shown that increased resistance followed exposure to penicillin in vitro but that in
all cases prolonged exposure was required. The main
results have been confirmed by other laboratory workers, but the constancy shown by different strains and
the necessity of prolonged exposure to effect penicillin
resistance seemed to be at variance with our experience
in the laboratory of an Army General Hospital. Therefore it was decided to make accurate tests of a series of
strains of a susceptible organism, Staphylococcus
aureus, isolated directly from patients, and to determine
the relation of susceptibility or resistance of the
organism to exposure to penicillin within the patient.
Subsequently, animal inoculations were undertaken
in an effort to get evidence that strains with acquired
resistance could again become susceptible. The major
part of the study was carried on in the laboratory of the
Lovell General Hospital, and it was rechecked and
extended in the 363rd Medical Laboratory while on
overseas service. The writer is grateful to the Commanding Officers of both organizations for encouragement to continue the study.
METHODS
The investigation was concerned with 40 strains
of hemolytic Staphylococcus aureus isolated from
specimens from surgical patients sent to the
laboratory for routine culture. All strains of
Staph, aureus isolated during a period of about
two months beginning April 1, 1944, were set
aside on Loeffler's blood serum, and were tested
in groups of four or more at a time. The method
of testing penicillin resistance was essentially that
devised by Rammelkamp1 for assay of the unit
strength of penicillin samples. A penicillin
solution of standard strength is set up by diluting
a fresh ampule to 20 units per cc. in sterile normal
saline according to the unit value on the label.
This is tested against at least two standard
cultures whose penicillin resistance is known, as
follows: Set up a series of 12 or more sterile Kahn
tubes. 1) Place 0.4 cc. of the initial dilution of
penicillin in the first tube and 0.2 cc. sterile saline
in each of the others. 2) Transfer 0.2 cc. from
tube #1 to tube #2, mix, transfer 0.2 cc. from
#2 to #3, mix and continue to #12 from which
0.2 cc. is discarded. This gives a series of penicillin samples ranging from 20 units per cc. to
0.009 units per cc, which will be diluted by
subsequent additions to final effective dilution's of
4 units per cc. to 0.0019 per cc. 3) Add to each
tube 0.8 cc. of 1 per cent rabbit's blood broth.
(If human blood is used the end points given
below have to be redetermined). 4) Add one
drop of a suspension of a 6 hour broth culture of
the standard Staph, aureus H, previously determined to have about 10,000 organisms per cc.
(A sub-culture of Staph, aureus H was kindly
supplied by the Research Laboratory of Merck
and Co.) 5) Shake, incubate 24 hours and read.
Organisms will grow and produce hemolysis in a
certain number of tubes with the greatest dilutions
of penicillin. The end point is sharp and with
Staph, aureus H growth is inhibited down through
tube #7, that is hemolysis ordinarily occurs in
tube #8 and beyond. This indicates that Staph,
aureus H is inhibited by 0.312 units per cc, but
will grow in the dilution of 0.156 units per cc.
Cultures of other organisms or other strains, are
made up in the same way and substituted in
step 4). Thus a strain of Streptococcus hemolyticus used as the standard for penicillin assays
in Dr. Keefer's laboratory at the Evans Memorial
Hospital in Boston was less resistant and gave
hemolysis only in tube #12. In order that tests
of penicillin resistance may be comparable it is
necessary that the same penicillin solution be used,
or one which has been assayed with the same
culture against it. In addition it is essential that
culture suspensions be of the same age, grown on a
similar medium, and of approximately similar
densities. When these precautions are followed
consistent results are secured, and differences
in the end points are a measure of penicillin
resistance of the organism. Thus Staph, aureus H
PENICILLIN RESISTANCE OF STAPHYLOCOCCUS AUREUS
447
resists about 16 times as much penicillin as the
tested strains of Strep, hemolyticus. It turns
out that various strains of Staph, aureus show
an even greater range of penicillin resistance than
that. The general results were frequently rechecked by the cup method of assay described by
Foster and Woodruff,5 which in our hands gave
less precise results.
Because of absence on overseas duty the
excellent papers of Gallardo6' 7 did not come to
hand until many months after this paper was
written and submitted to the office of the Surgeon
General. Therefore I have had no opportunity
to try out his original variation of the cup method
of assay. In general however the tube method
described above gave more exact readings than
any of the standard cup procedures. In spite of
the differences in method our results are in substantial agreement with his.
The tests in duplicate were read at the end of
24 hours incubation and then refrigerated for a
second 24 hour period when the readings were
checked. Transfers were than made from the
tube with the greatest concentration of penicillin
in which growth occured to fresh broth cultures.
These cultures were left for a week at room temperature and then retested. Frequently an
increase in penicillin was noted in the second as
compared with the first set of tests and this was
interpreted as being due to the exposure of the
organisms to penicillin in vitro. Frequent retests
of the original cultures unexposed to penicillin
gave the same values for penicillin resistance as
the originals, thus confirming the interpretation.
Five cultures were selected for mouse passage
after one or more penicillin exposures. Mice
were inoculated intraperitoneally with one cc. of a
24 hour broth culture of the strain selected. In
four tests the mice died within 18 hours, and in the
fifth the mouse was moribund and was killed at
24 hours. In all cases, Staph, aureus was isolated
from the peritoneal exudate in pure culture.
This was retested for penicillin resistance as
before.
culture was taken it was noted from the patient's
chart whether he had or had not received penicillin
therapy, and later a follow-up recorded the
subsequent history of the infection. Following
isolation of the culture of Staph, aureus a coagulase
test was done, since this is considered a good
indication of the virulence of the strain. Finally
the penicillin resistance in the first and usually
two successive subcultures after exposure to
penicillin was recorded. On first isolation the
range of penicillin resistance in vitro ran from a
tolerance of at least 4 units to inhibition by as
little as 0.002 units, or a difference in sensitivity
of more than 2000 times. Far from being constantly susceptible to penicillin, 28 per cent of the
strains on first isolation from patients in an Army
Hospital were not inhibited by 4 units per cc,
while 41 per cent may be called resistant. The
data are given in table 1, test # 1.
PENICILLIN RESISTANCE ON FIRST ISOLATION
In tables 2 and 3 the initial data were correlated
with the results of the first tests of penicillin
resistance. Table 2 summarizes the relation of
previous penicillin therapy to the reaction of the
culture isolated. In three cases the previous
history was not available. Table 3 records the
correlation of penicillin susceptibility of the
strain with the success of the treatment according
The details of the tests of penicillin resistance
on first isolation of the 40 strains of Staph, aureus
cannot be given because of lack of space. The
following data will suffice. The sources of the
strains were surgical cases, wound infections,
osteomyelitis, ulcers, and one case of bacterial
endocarditis following a wound. At the time the
ACQUIRING OF PENICILLIN FASTNESS
In table 1 the last three columns show the
effects on penicillin resistance of exposure to one
and to two 48 hour treatments with penicillin in
vitro. Of the 18 strains so tested, only 3 (or
17 per cent) were completely resistant at the
beginning, while after one exposure 5 (or 28
per cent) and after two exposures 11 (or 61 per
cent) resisted the maximum dose of 4 units per cc.
In general, nearly all strains became resistant
following two 48 hour exposures to penicillin and
the trend was toward this condition even after a
single exposure. If we set arbitrarily the boundary between resistance and susceptibility at 0.5
units per cc. and call all strains inhibited through
tube 4 resistant and all those inhibited at 5 or
higher as susceptible, we find only 4 strains of
18 remain susceptible after two exposures. Thus,
while most strains were easily made penicillin
fast, there was nevertheless a small number which
retained susceptibility after repeated exposures
to the drug.
PENICILLIN RESISTANCE IN RELATION TO DATE OF
TREATMENT AND OUTCOME
448
HAROLD H. PLOUGH
to the judgment of the medical officers in charge;
that is, whether the lesion was healed or still active
at the date of compilation.
was naturally somewhat smaller than were
considered in the previous table. Not all of the
patients were treated with penicillin, and in
TABLE 1
PENICILLIN SENSITIVITY OF STRAINS OF STAPHYLOCOCCUS AUREUS ON ISOLATION AND AFTER
EXPOSURE IN VITRO
TEST
NO.
After isolation all tests
la
0
1 2
<4
4
2
(28%)
(41%)
13
1
1
9 10 11 12
3 4 5 6 7
Penicillin units per cc.
l 0.5 0.25 0.125 0.06 0.03 0.015 0.008 0.004 0.002
2 0
1
2
3
3
6
3
3
41
After isolation including only (17%)
3
1
strains tested 3 times
1
1 0
0
1
2
3
4
1
1
18
Re-tested after one exposure
to penicillin
(28%)
1
5
1
0 0
0
2
4
2
1
1
1
18
Re-tested after two exposures (61%)
to penicillin
11
1
2
0 0
0
1
1
2
0
0
0
0
18
Note: a) Tube #0 means that the strain grew in all dilutions, or in other words it was not inhibited by 4
units per cc. Tube % 1 records the number of strains inhibited by 4 units per cc. but growing in all others, and
so on. b) Strains were arbitrarily called "resistant" which grow in tube #4 or above while they were labelled
"susceptible" if they failed to grow in tube # 5 or below.
It seems quite clear from table 2 that strains
isolated from patients who have not received
penicillin up to the time the cultures were taken
are very likely to be penicillin susceptible. Conversely strains from patients who have had penicillin before the culture was taken have an excellent chance of being resistant if they still persist.
It is legitimate to combine the totals of those who
received treatment after culture with those who,
according to the records, have received no treatment at all. When this was done it appears that
5 out of 38 or 13 per cent were resistant to penicillin even though they have not been exposed to
it, at least while in the body of the patients from
whom they were isolated. Stated in another way
these data suggest that penicillin fastness of
Staph, aureus usually results from treatment with
penicillin if the infecting organism survives, but a
small number of invading organisms may be
expected to be penicillin fast even in untreated
patients.
The data bearing on penicillin sensitivity and
the outcome of treatment are summarized in
table 3. The total number of strains considered
TABLE 2
PENICILLIN FASTNESS OF STRAINS or STAPHYLOCOCCUS
AUREUS IN RELATION TO DATE OF
FIRST TREATMENT
TREATED
No treatment
Total
NO.
PENICILLIN SENSITIVITYSTK
AINS
Susceptible
Resistant
9
1
Susceptible
Resistant
4
13
Susceptible
Resistant
7
4
38
certain cases where penicillin was used surgical
failures or infection with more than one organism
complicated the final picture. Such negative or
doubtful cases are not included in the summary.
In all 25 cases of uncomplicated Staphylococcus
infection were considered in the judgment of the
medical officers to have either healed, or failed to
449
PENICILLIN RESISTANCE OF STAPHYLOCOCCUS AUREUS
heal more easily. Penicillin resistance is again
indicated as correlated with failure to heal. In
so far as the coagulase reaction is a test of virulence,
or invasive power, it is less important in patients
treated with penicillin than is penicillin resistance.
It is of interest that the coagulase reaction does
not change in strains which have acquired penicillin fastness in vitro. It remained constant in
all cases tested.
TABLE 4
STILL
ACTIVE
TOTAL
COAGULASE REACTION AND PENICILLIN RESISTANCE
IN RELATION TO OUTCOME
HEALED
heal after penicillin therapy. In the summary of
table 3 the type of treatment whether systemic or
local is shown also. While the numbers are
small, the general trend seems unmistakable.
Infections with penicillin sensitive strains of
Staphylococci are more likely to be killed and
resistant ones are more likely to survive penicillin
therapy. This is of course only what is to be
expected, but it does focus attention on the
importance of penicillin resistance, especially
since the data already presented show that it is of
widespread occurrence. It is clear from table 3
also that a certain number of susceptible strains
Positive
coagulase
Positive
TILL TOTAL
HEALED ASC
coagulase
TIVE
Sensitive
organism
Resistant
organism
5
2
13
1
Negative
coagulase
Negative
Negative
coagulase
Sensitive
organism
Resistant
organism
5 J
1
3
2
0
>
7j
TABLE 3
SUMMARIES OF PENICILLIN SENSITIVITY AND TYPE or
TREATMENT IN RELATION TO OUTCOME
OF CASES
Sensitive
Systemic
organism... treatment
Sensitive
organism... Local treatment
Resistant
Systemic
organism... treatment
Resistant
organism;.. Local treatment
5
•8
2
3
2
1
'
8
0
1
4
> 4 12
9
16
12
25
• 12 13
RECOVERY OF PENICILLIN SENSITIVITY IN RESISTANT
STRAINS
16 25
9
Total
The data presented in table 2 add further proof
that most strains of Staph, aureus easily acquire
of Staphylococci do not respond to penicillin penicillin resistance if exposed to this antibiotic.
therapy, and an occasional resistant strain may. Is it possible for these strains with induced
penicillin resistance to return to the susceptible
PENICILLIN SENSITIVITY OF STAPH. AUREUS AND condition? Rammelkamp and Maxon found that
VIRULENCE
it was not, but it is clear that the strains described
Bacteriologists have found a rather high in this study gave a result different from theirs.
Some light on this problem is thrown by a study
correlation between the virulence of strains of
staphylococci and the ability to coagulate blood of comparisons of two different strains isolated
plasma. A test of this ability is run routinely in from the same patient. Four such repeat tests
many hospital laboratories on all staphylococci were made in the course of this study although this
isolated from lesions. Since the results of this was realized only after the completion of the
test for each of the strains of Staph, aureus tabulation. Of the original strains numbers 2
studied were recorded the same 25 cases con- and 5, 8 and 12, 21 and 34, 25 and 37 show two
sidered in the previous paragraph are summarized cultures from four different patients. The first
and correlated with penicillin sensitivity in table 4. two pairs were resistant in each of the tests which
It is immediately apparent that there is no were made three days and a week apart respeccorrelation whatever between the ability to tively. One patient had received penicillin before
coagulate plasma (coagulase positive) and sensi- as well as after the first culture and the second
tivity to penicillin, nor is there any indication had never received any treatment with it. The
that infections with coagulase negative strains third pair ( #21 and 34) shows a susceptible strain
450
HAROLD H. PLOUGH
in the first isolation made before treatment, and a
resistant strain after treatment of 2 million units.
In the fourth case two isolations from the same
patient ( #25 and 37) were made four weeks apart.
This patient had received one million units of
penicillin at another army hospital before the
culture was taken, and this culture was resistant.
He received no additional penicillin at this hospital,
and the. second culture was susceptible. This
suggests that while some strains may remain
resistant while they remain in the body, a resistant
strain may become susceptible after an interval
during which there is no contact with penicillin.
In order to get further information on this
point four strains were selected for mouse inoculation. They were numbers 3, 15, 32, and 34 in the
series. The first two were susceptible when
isolated and resistant after three exposures to
penicillin in vitro. These resistant strains from
the final isolation were injected into mice, and
on the death .of the animals reisolated from the
peritoneal exudate and retested on penicillin.
The first (#3) remained resistant while the
second (#15) had become susceptible following
mouse passage. The last two strains were
resistant on first isolation. On reisolation from
the mouse (# 32) was still resistant, and the
other ( #34) was susceptible. This last case is of
special interest for it is the second number of the
third pair discussed in the previous paragraph.
This organism on first isolation from the patient
( #21) was susceptible. It was resistant six weeks
later after a course of treatment ( # 34), and now
after mouse passage it has become susceptible
again. This last case suggests strongly that
penicillin fastness may be lost again after it has
been induced by penicillin therapy.
CLINICAL SIGNIFICANCE OF THE DATA
The clinical implications of the data here
reported are apparent and can be briefly stated.
Since a number of strains' of staphylococcus
encountered in patients in army hospitals are
resistant to penicillin or may become so, it seems
essential to give as large doses as possible at the
very beginning of treatment. Serum levels
reported by Rammelkamp and Keefer3 as well as
those determined by the writer do not exceed one
unit per cc. following a dose of 25,000 units.
Yet some resistant strains will grow in a concentration of over four units per cc. in a test tube.
It would seem that still larger doses than 25,000
units should be given at the beginning in order to
secure a higher blood level which might inhbit
some of the more resistant strains. While it is
clear that infections have been cleared by long
continued treatment with penicillin, the laboratory
data do not justify this method since strains of
staphylococcus may be made resistant by even a
forty-eight hour exposure to penicillin in a test
tube. It would appear that a short course of
maximal injections would be more successful
than an extended one. Especially there would
seem to be little advantage in a tapering course of
injections.
The results apparently establish the fact that
certain strains of staphylococcus may recover
sensitivity after a period in the body when they
are not in contact with penicillin. This suggests
that cases of infection which are difficult might be
cleared more easily if a period of some days to
weeks were allowed to elapse without treatment
and then the maximal dose repeated. A succession of maximal treatments separated by a long
intervening period without penicillin therapy
might be more effective than an extended series of
injections.
Finally in all infections with Staph, aureus which
fail to clear after the initial course of treatment
the strain should be isolated and its penicillin
sensitivity tested in the laboratory. Either the
method here recorded or that of Gallardo are
easily available.
SUMMARY OF CONCLUSIONS
1. 40 strains of Staphylococcus aureus isolated
from lesions of patients in an army hospital have
been tested for penicillin sensitivity, and over
40 per cent can be called resistant. They range
from complete tolerance of 4 penicillin units per cc.
to inhibition by 0.002 units per cc.
2. Most, but not all, strains isolated from
patients who have received penicillin treatment
are resistant. Most but not all strains from
patients who have not received penicillin are
sensitive.
3. Sensitive strains of Staph, aureus were
made resistant after one or two exposures of "48
hours to penicillin in vitro.
4. A study of case histories shows that sensitive
strains are more frequently eliminated by penicillin treatment, either local or systemic, while
resistant strains more frequently fail to heal.
5. Virulence of the strain as measured by the
PENICILLIN RESISTANCE OF STAPHYLOCOCCUS AUREUS
coagulase test bears no relation to penicillin
sensitivity. Penicillin resistance appears to be a
more important index of failure to heal following
penicillin therapy than is the coagulase test.
6. Successive isolations from the same patients
and mouse inoculations suggest that some penicillin resistant strains of Staph, aureus in vivo
may become sensitive again after a period without
penicillin exposure.
7. Certain modifications in current practice in
penicillin therapy for wounds infected with Staph,
aureus appear to be suggested by these laboratory
studies, a) A maximal initial injection of penicillin without any tapering off dose seems most
likely to clear the wound, b) If the initial
injections fail to clear the infection a period of rest
without penicillin may be advisable, and then a
second maximal dose, c) In cases of Staph,
aureus infections which are refractory to penicillin,
a test of the penicillin sensitivity of the organism
may indicate the probable success of continued
penicillin therapy.
451
REFERENCES
1. RAMMELKAMP, C. H.: A method for determining the
concentration of penicillin in body fluids and
exudates. Proc. Soc. Exper. Biol. & Med., 51:
95, 1942.
2. RAMMELKAMP, C. H.,
AND MAXON, THELMA:
Resistance of Staphylococcus aureus to the action
of penicillin. Proc. Soc. Exper. Biol. & Med.,
51: 386, 1942.
3. RAMMELKAMP, C. H., AND KEEPER, C." S.
The
absorption, excretion and distribution of penicillin. J. Clin. Investigation, 22: 425, 1943.
4. MCKEE, CLARA, M., AND HOUCH, CAROL L.:
Induced resistance to penicillin of cultures of
staphylococci, pneumococci, and streptococci.
Proc. Soc. Exper. Biol. & Med., 53: 33, 1943.
5. FOSTER, JACKSON W., AND WOODRUFF, H. BOYD:
Procedure for the cup assay of penicillin. J.
Bact., 47: 27, 1944.
6. GALLARDO, EDWARD: Sensitivity of bacteria from
7.
infected wounds to penicillin. I. Method of
assay. War Med., 6: 86, 1944.
: II. Results in one hundred and twelve
cases. War Med., 7: 100, 1945.