Pseudobacteremia Attributed to Contamination of Povidone-lodine with
Pseudomonas cepacia
RUTH L BERKELMAN, M.D.; SHARON LEWIN, M.D.; JAMES R. ALLEN, M.D.; ROGER L ANDERSON,
Ph.D.; LAWRENCE D. BUDNICK, M.D.; STANLEY SHAPIRO, M.D.; STEPHEN M. FRIEDMAN, M.D.;
PETER NICHOLAS, M.D.; ROBERT S. HOLZMAN, M.D.; and ROBERT W. HALEY, M.D.; Atlanta,
Georgia; and New York, New York
Pseudomonas cepacia was recovered from the blood
cultures of 52 patients in four hospitals in New York over
7 months from April through October 1 9 8 0 .
Epidemiologic investigation in one hospital indicated that
the positive results of blood culture represented
pseudobacteremias and implicated a 1 0 % povidone-
iodine solution used as an antiseptic and disinfectant
(Pharmadine; Sherwood Pharmaceutical Company,
Mahwah, New Jersey) as the source of contamination.
Physicians who drew blood cultures positive for P. cepacia
were more likely to have left povidone-iodine on the skin
before venipuncture (p=0.026) and were more likely to
have applied povidine-iodine to the blood culture bottle
tops and to have left it there while inoculating the blood
culture media ( p = 0 . 0 0 7 ) than those who drew cultures
negative for P. cepacia. Direct inoculation of Pharmadine
into brain-heart infusion broth yielded P. cepacia;
however, 2 weeks after the first cultures, the same
Pharmadine bottles were culture negative. The iodine
concentrations of the contaminated Pharmadine solutions
were similar to those of 1 0 % povidone-iodine solutions
distributed by other manufacturers.
P O V I D O N E - I O D I N E SOLUTION is widely used in hospitals
and other health care institutions as a skin and mucous
membrane antiseptic and as a disinfectant. Unlike other
classes of antiseptics and disinfectants (1), povidone-
iodine solutions have not previously been reported to be
intrinsically contaminated.
In October 1980, City Hospital Center at Elmhurst
reported to the Centers for Disease Control (CDC) that
17 blood cultures obtained over the preceding 3 months
were positive for Pseudomonas cepacia. An epidemiolog-
ic investigation was conducted by hospital personnel, the
New York City Department of Health, and C D C at this
hospital and subsequently at three other New York City
hospitals identified as having had clusters of blood cul-
tures positive for P. cepacia over the previous 6 months.
We describe herein the epidemiologic investigation that
implicated contaminated povidone-iodine solution pro-
duced by one manufacturer (Pharmadine; Sherwood
Pharmaceutical Company, Mahwah, New Jersey) as the
source of the problem and that established that the posi-
tive blood cultures were pseudobacteremias (that is, false-
positive results of blood culture). We also present find-
ings of subsequent laboratory studies that have further
defined the problem.
• F r o m the Hospital Infections Branch a n d Epidemiologic Investigations Labora-
tory Branch, Bacterial Diseases Division, Center for Infectious Diseases, Centers
for Disease Control, Atlanta, Georgia; a n d M t . Sinai Services at City Hospital
Center at Elmhurst, M t . Sinai School of Medicine; N e w York City D e p a r t m e n t of
Health; a n d Bellevue Hospital Center, N e w York University School of Medicine,
New York, N e w York.
32 Annals of Internal Medicine. 1981;95:32-36.
Downloaded From: http://annals.org/ by a Michigan State University Libraries User on 08/14/2018
Materials and Methods
EPIDEMIOLOGIC INVESTIGATION
At City Hospital Center at Elmhurst we reviewed blood cul-
ture log books for January through October 1980 to identify all
Pseudomonas isolates and all bacteriology reports for June
through October to identify isolates of P. cepacia from sites
other than blood. We further reviewed medical records of all
patients with blood cultures positive for P. cepacia and abstract-
ed information on age, sex, underlying illnesses, indications for
blood cultures, and use of intravenous therapy and medications.
We interviewed pharmacy and central supply personnel about
distribution of supplies to the various areas of the hospital and
laboratory personnel about methods for processing and han-
dling blood culture specimens.
We conducted a case-control study at City Hospital Center at
Elmhurst to ascertain whether blood cultures positive for P.
cepacia were significantly associated with the medical service.
We used a table of random numbers to select 118 control cul-
tures from the 2360 blood cultures obtained between 30 June
and 4 October 1980. We excluded six cultures from final analy-
sis: two that were positive for P. cepacia, one that was actually a
pleural fluid culture, and three for which service was not re-
corded. We compared the bacteriology reports of the remaining
112 with those of the 17 cultures positive for P. cepacia to
identify differences in services of origin.
A questionnaire was administered to 35 medical and 12 pe-
diatric house staff working during the investigation. Only the
responses of the 30 medical and 10 pediatric house staff who
indicated that they routinely obtained one or more blood cul-
tures per week were included in the epidemiologic analysis. We
compared the blood culturing techniques of the medical house
staff with those of the pediatric house staff and the techniques of
the medical house staff who had drawn the cultures positive for
P. cepacia with those who had not drawn cultures positive for
the organism.
We telephoned infection control personnel at 27 hospitals in
the New York City area, including all those with more than 500
beds, and requested them to review their microbiology laborato-
ry records for 1980 to identify all isolates of P. cepacia. In the
three hospitals that identified five or more blood cultures posi-
tive for P. cepacia, we reviewed the medical records of all pa-
tients with cultures positive for P. cepacia. Techniques and
products used at each of these hospitals were compared to iden-
tify those used in common.
LABORATORY STUDIES
Available blood isolates from three hospitals and isolates re-
covered from povidone-iodine solutions were identified by stan-
dard biochemical testing (2) in the Epidemiologic Investigations
Laboratory Branch, Bacterial Diseases Division, Center for In-
fectious Diseases, CDC. Kirby-Bauer antimicrobial susceptibili-
ty testing (3) was done on all isolates identified as P. cepacia.
The following special laboratory studies also were done.
Isolation of Pseudomonas cepacia from Povidone-lodine So-
lutions: Fifty-four bottles of Pharmadine solution and 11 sets of
Pharmadine swabsticks (Sherwood Pharmaceutical Company)
from 19 manufacturing lots were obtained from the stock of
three of the hospitals investigated and were examined for the
©1981 American College of Physicians
 presence of P. cepacia. Three aliquots (0.02, 0.1, and 0.5 mL) of
Pharmadine solution were chosen for direct culture to simulate
quantities of povidone-iodine that might be introduced into
blood culture media if blood were inoculated through a blood
culture bottle top that had povidone-iodine pooled in its depres-
sion. Each aliquot was separately added to 125 mL screw-cap
flasks containing 50 mL of brain-heart infusion broth with
0.5% beef extract. Swabsticks were aseptically bent and placed
directly into wide-mouth jars containing 150 mL of brain-heart
infusion broth. Broth cultures were incubated at 35 °C and ob-
served for turbidity daily for 7 days. Aliquots of broths showing
turbidity were subcultured onto sheep blood agar and MacCon-
key agar. All nonturbid broths were subcultured on day 7.
Two different culture techniques were used for subsequent
attempts to recover P. cepacia from bottles of solution from
which it had been isolated initially: direct inoculation into
brain-heart infusion broth and Millipore filtration (Millipore
Corporation, Bedford, Massachusetts) of 10 mL povidone-
iodine in 50 mL brain-heart infusion broth. The filter was then
washed with 50 mL of brain-heart infusion broth and was incu-
bated on a blood agar plate.
Survival of Pseudomonas cepacia in Povidone-iodine Solu-
tion: Four strains of P. cepacia were tested to establish their
survival in 10% povidone-iodine solutions: an isolate from
Pharmadine, lot X80290, adapted to deionized water; an isolate
from Pharmadine, lot U80250; an isolate from a clinical speci-
men; and a CDC laboratory stock strain. Two povidone-iodine
solutions were used: Pharmadine, lot X80299, from which via-
ble organisms could no longer be recovered; and Betadine, lot
M793136 (Purdue Frederick Company, Norwalk, Connecticut).
For the experiment, 10" to 1010 organisms of each of the four test
strains of P. cepacia were suspended separately in 1 mL deion-
ized water and then were mixed with 10 mL of each of the two
povidone-iodine solutions. For the strain adapted to deionized
water, 2 mL of deionized water containing 1.4 times 106 organ-
isms were mixed with 10 mL of each of the povidone-iodine
solutions. Aliquots of 0.1 mL and 0.5 mL were obtained from
these mixtures at 0, 10, and 60 minutes and at 4 and 24 hours
and were inoculated separately into flasks containing 50 mL
brain-heart infusion broth and either 0.25% or 0.5% or no sodi-
um thiosulfate. Specimens were incubated at 35 °C and were
subcultured as described above.
Measurements of Iodine Concentration in Povidone-iodine
Solutions: Available iodine contents of coded samples of Phar-
madine, Betadine, and Povidine solution (National Pharmaceu-
tical Manufacturing Company, Baltimore, Maryland) were as-
certained by titration with sodium thiosulfate at a commercial
laboratory (MacMillan Research, Marietta, Georgia). The free Figure 1 . Blood cultures positive for Pseudomonas cepacia at four
iodine contents of coded samples of Pharmadine, Betadine, and New York City hospitals, by month of culture, 1 9 8 0 .
Povidine were ascertained by the distribution of iodine between
the aqueous (product) phase and a nonmiscible solvent, hep-
tane, at 25 °C (4) (done by Murray W. Winicov, Kansas City, underlying illnesses and the reasons for obtaining blood
Missouri). All statistical tests used Fisher's exact test; two- cultures varied greatly; the latter included diagnostic
tailed p values are reported unless otherwise indicated. work-up of pyrexia and follow-up of documented infec-
tions at other sites.
Results
Only six of the 14 patients had a maximum tempera-
CLINICAL A N D EPIDEMIOLOGIC F I N D I N G S ture above 38.3 °C the day the blood culture was drawn,
From 30 June to 4 October 1980, P. cepacia was isolat- and no patient had a temperature more than 1 °C higher
ed from 17 blood cultures obtained from 14 patients at than the maximum temperature of the previous day. N o
City Hospital Center at Elmhurst (Figure 1). No isolates patient experienced rigors. Two were hypotensive (systol-
of P. cepacia and only one of a species of Pseudomonas ic blood pressure less than 90 mm Hg) for reasons unre-
other than P. cepacia had been isolated from blood cul- lated to possible septicemia. The median leukocyte count
tures in the preceding 6 months. was 9100/mm 3 (range, 2500 to 30 000/mm 3 ). Although
Fifteen of the blood cultures positive for P. cepacia cultures from other sites were obtained at or near the
were obtained from 12 patients on the medical service; same time as the blood cultures in 12 of the patients,
the other two cultures were taken from two patients on none were positive for P. cepacia. Only four patients were
the hemodialysis unit. The 14 patients were hospitalized treated with antibiotics for possible septicemia as a result
on two of the 10 patient care floors. The median age of of the positive cultures. In summary, the clinical infor-
the 14 patients was 59.5 years (range, 23 to 90 years); mation did not suggest that any of the patients had gram-
eight of the patients were female. Type and severity of negative septicemia.

Berkelman et al. • Pseudomonas Pseudobacteremia 33

Downloaded From: http://annals.org/ by a Michigan State University Libraries User on 08/14/2018

 Table 1 . Iodine Concentrations of 1 0 % Povidone-lodine Solutions
10% Povidone- Availat )le
Iodine Solution Iodine *
fjig/mL f % jug/mL
Pharmadine, lot 10000 1.00
X80299
Pharmadine, lot 10 000 1.00
U80250
Pharmadine, lot 10 200 1.02
V80259
9500 0.95
Betadine, lot
M793163
Povidone, lot 01428 11 100 1.11
* Available iodine equals thiosulfate-titratable iodine.
t One microgram per millilitre equals 1 part per million.
Because the initial epidemiologic findings suggested
pseudobacteremia, we studied the circumstances under
which the positive blood cultures were done and pro-
cessed. Fifteen of the 17 blood cultures positive for P.
cepacia had been obtained from the medical service,
whereas only 54 of 112 (48%) randomly selected blood
cultures came from this service (p = 0.003). Because all
blood cultures were handled and processed in an identical
manner in the laboratory, the clustering of the positive
cultures on the medical service made it highly improba-
ble that contamination of the specimens occurred in the
laboratory. Also, because the same brand of blood cul-
ture medium was distributed as needed to all areas of the
hospital, intrinsic contamination of the medium was an
unlikely explanation for the pseudobacteremias.
Pediatrics was the only service other than medicine
that obtained more than 10% of the hospital's blood cul-
tures; none of the blood cultures positive for P. cepacia
had been obtained on this service. Analysis of the ques-
tionnaire surveys indicated that the blood culturing tech-
niques reported by house staff on the medical service dif-
fered from those reported by house staff on the pediatric
service in 2 ways: Pediatric house staff often obtained less
than the 10 m L blood, usually drawn by medical house
staff; and medical house staff differed in their use of povi-
done-iodine. Eleven of 30 medical house staff but none of
10 pediatric house staff reported doing venipuncture
without removing the povidone-iodine used for skin anti-
sepsis (p = 0.038). Also, 13 of 30 medical house staff
versus two of 10 pediatric house staff reported using povi-
done-iodine to disinfect the tops of the blood culture bot-
tles before inoculation of a blood specimen (p = 0.27).
Seven medical house staff, including medical students,
interns, and residents, were documented to have drawn
12 of the 15 blood cultures positive for P. cepacia on the
medical service; the persons who drew the remaining
three contaminated cultures could not be identified. Five
of the seven house staff who drew the contaminated cul-
tures versus five of the 23 medical house staff who did not
draw contaminated cultures did venipuncture without re-
moving the povidone-iodine from the skin (p = 0.026).
Six of seven house staff who obtained positive cultures
and seven or 23 house staff who did not draw contami-
nated cultures used povidone-iodine to disinfect the tops
of the blood culture bottles (p = 0.025). Moreover, five
34 July 1981 • Annals of Internal Medicine • Volume 95 • Number 1
Downloaded From: http://annals.org/ by a Michigan State University Libraries User on 08/14/2018
of the seven physicians who obtained positive cultures
allowed the povidone-iodine to pool in the depression in
Free Iodine
the stopper of the blood culture bottle and inoculated
directly through the pool of povidone-iodine, whereas
only three of the 23 who were not known to have drawn a
1.2
contaminated culture used this technique (p = 0.007).
1.5 Thus, epidemiologic evidence implicated povidone-iodine
as the source of the pseudobacteremias.
1.4 Through our telephone survey we identified three other
0.6
hospitals in New York City with apparent clusters of P.
cepacia bacteremias (Figure 1); the positive cultures had
1.3 occurred from April through October 1980. Review of
the patients' medical records indicated that few of the
patients had clinical features consistent with septicemia
and that pseudobacteremia had been strongly suspected
by the physicians of most of these patients. Each of the
four hospitals identified as having clusters of pseudobac-
teremias attributed to P. cepacia purchased Pharmadine
for use as an antiseptic or disinfectant; no other brand of
iodophor was available in these hospitals.
One patient at each of two hospitals was identified as
having cultures positive for P. cepacia from sites other
than blood. One patient had the organism isolated from
an abdominal wound that had been treated with Pharma-
dine wet-to-dry dressings. A second patient had a sputum
culture positive for P. cepacia after use of Pharmadine for
care of her tracheostomy site.
MICROBIOLOGIC RESULTS
Of the 59 isolates of P. cepacia obtained from blood
cultures by the four New York City hospitals, 17 isolates
from three hospitals were sent to C D C ; all were con-
firmed as being P. cepacia. Antimicrobial susceptibility
testing showed all isolates to be sensitive to trimetho-
prim-sulfamethoxazole and nalidixic acid and resistant to
gentamicin, nitrofurantoin, cephalothin, ampicillin, tetra-
cycline, and colistin. Susceptibility to chloramphenicol,
kanamycin, amikacin, and carbenicillin varied among the
isolates. Biotypes were identical.
Pseudomonas cepacia in pure culture was isolated
from three previously unopened 16-ounce bottles of
Pharmadine solution, lot X80299, manufactured 10 days
before culture, and from one previously opened 16-ounce
container of Pharmadine solution, lot U80250, manufac-
tured 4 months before culture. Pseudomonas cepacia was
also recovered from lot U80250 9 days after the first cul-
tures, but it could not be isolated from lots X80299 and
U80250 9 days and 16 days, respectively, after the first
positive cultures. Sixteen other lots of Pharmadine solu-
tion were cultured, but no viable organisms were recov-
ered. One set of Pharmadine swabsticks, lot X80290,
manufactured 1 month before culture, yielded P. cepacia.
Attempts to recover P. cepacia from povidone-iodine so-
lutions after laboratory inoculation of organisms from
broth culture have not been successful.
The available iodine concentration of the povidone-io-
dine solutions tested ranged from 9500 to 11 100 jag/mL
(0.95% to 1.11%) (Table 1). The free iodine concentra-
tions of samples of three lots of Pharmadine ranged from
1.2 to 1.5 juLg/mL compared with 0.6 to 1.3 juig/mL for
 samples of two povidone-iodine solutions manufactured
by other companies.
Discussion
Pseudomonas cepacia is a motile nonfermentative
gram-negative bacillus able to use a large number of sub-
strates over a wide range of temperatures; it has a diverse
geographic distribution and has been isolated from soil,
rivers, and vegetation (5). More recently, the hospital en-
vironment has become an important source for isolation
of P. cepacia; the organism has been recovered from irri-
gating fluids, anesthetics, ultrasonic nebulizers, deter-
gents, and disinfectants, including chlorhexidine and
quaternary ammonium compounds (6). Pseudomonas ce-
pacia has only infrequently caused disease in humans;
bacteremia associated with contaminated intravenous
fluid (7) and blood pressure transducers (8), and endocar-
ditis in heroin addicts (9), pneumonitis, urinary tract in-
fections, and wound infections caused by contaminated
solutions have all been described (6).
Among the clusters of pseudobacteremias previously
described (10, 11), several have been attributed to con-
taminated benzalkonium chloride or other antiseptics (1).
Contamination of iodophor products, however, has not
been described previously and represents a newly identi-
fied source of pseudobacteremia. False-positive blood cul-
tures may have several adverse consequences including
additional laboratory studies, administration of unneces-
sary and potentially dangerous antibiotics, and prolonga-
tion of hospital stay.
Contaminated products in hospitals may cause true in-
fections as well as false-positive laboratory tests. Povi-
done-iodine solutions are used for many purposes, includ-
ing skin antisepsis before surgery, cleansing and packing
of wounds and ulcers, treatment of vaginitis, and disin-
fection of patient care items. Although only a few pa-
tients in this outbreak were identified as being colonized
with P. cepacia secondary to use of contaminated Phar-
madine and no patient had a clinically adverse effect
clearly established, the potential exists for patients to de-
velop serious infection from use of contaminated povi-
done-iodine solution.
Iodine in solution has a broad spectrum of antimicrobi-
al activity against both gram-negative and gram-positive
vegetative bacteria, fungi, viruses, protozoa, and even
bacterial spores (12); however, elemental iodine solutions
have numerous side effects (such as staining, odor, and
skin or mucous membrane irritation) that have limited
their usefulness in patient care practices. Iodophors,
complexes of iodine and a surfactant or nonsurfactant
carrier molecule, were developed to reduce these side ef-
fects. Povidone-iodine solution, USP, is the common
name for iodine in complex with the nonsurfactant com-
pound polyvinylpyrrolidone in a 10% solution.
We tried to learn how the povidone-iodine solution
had become contaminated and what factors had allowed
survival of organisms in the solution. Discussions with
the manufacturer of Pharmadine indicated that the povi-
done-iodine solution was prepared in a multistage process
that included mixing commercially available polyvinyl-
Downloaded From: http://annals.org/ by a Michigan State University Libraries User on 08/14/2018
pyrrolidone-iodine powder with water. The water source
was a municipal supply that was then processed through
resin deionizers at the plant. Although significant con-
tamination of municipal water supplies with Pseudomo-
nas would be unlikely (13), microbial contamination of
resins is a well-described problem (14). In this outbreak
P. cepacia probably proliferated on the deionizing resin
in the water system and contaminated the water supply
and possibly manufacturing surfaces within the plant. Af-
ter the problem was identified, a Millipore filtration sys-
tem with a 0.22-jam filter was installed between the
deionizing resin and the mixing tanks; this system is be-
ing evaluated to ascertain whether it will prevent bacteri-
al contamination of the water supply used in manufactur-
ing povidone-iodine products.
We considered several hypotheses as possible explana-
tions for survival of P. cepacia in the povidone-iodine
solution. First, we hypothesized that the product was de-
fective and contained a low concentration of iodine active
against microorganisms. We measured the iodine concen-
tration of several povidone-iodine solutions. The "avail-
able" or sodium thiosulfate titratable iodine is the most
commonly obtained measurement of iodine content of
povidone-iodine solutions (12). Both Pharmadine and po-
vidone-iodine solutions by other manufacturers had an
available iodine content of approximately 1% (10 000
jag/mL), conforming with USP standards for a 10% po-
vidone-iodine solution (Table 1).
The available iodine is known to be divided between
iodine in complex (iodine bound by the carrier molecule
and iodides) and "free" iodine, which may be as little as
1.0 juLg/mL (15, 16). The free iodine in solution has been
well established to possess microbicidal activity (12, 17);
in contrast, there is little evidence to substantiate micro-
bicidal activity of the complexed iodine. Therefore, the
free iodine concentration may be a more accurate mea-
sure of bactericidal activity than available iodine. Mea-
surements of free iodine in the three samples of Pharma-
dine solution were similar to those of two other
povidone-iodine solutions analyzed. Thus, we could not
show by chemical measurements of iodine content that
the contaminated Pharmadine solution was less bacteri-
cidal than other marketed solutions.
Our second hypothesis was that organic or inorganic
debris had mechanically protected P. cepacia organisms
from Pharmadine. Although a small quantity of debris
was present on the filter after Millipore filtration, we
were unable to identify constituents of the debris other
than numerous bacteria.
Finally, we considered that the strains of P. cepacia
isolated from Pharmadine possessed a greater iodine tol-
erance than other strains. Few studies of microbial resist-
ance to povidone-iodine solutions have been published. In
one study a minimal inhibitory concentration of iodine
for a wild strain of P. cepacia was reported to be 16 jag/
mL (18); in another study organisms were reported not to
develop resistance to povidone-iodine after serial passages
through that solution (19). These studies, however, might
not be accurate for two reasons. First, these experiments
were based on measurements of available iodine concen-
Berkelman et al. • Pseudomonas Pseudobacteremia
35
 trations although available iodine has not been shown to
be responsible for the inhibitory and bactericidal activity
of povidone-iodine (15). Second, because the iodine in a
10% povidone-iodine solution may be in a tighter com-
plex than that of a more dilute preparaton, dilution of
povidone-iodine may paradoxically increase the concen-
tration of free iodine and possibly increase its bactericidal
activity ([4], W I N I C O V M W . Personal communication).
In this outbreak P. cepacia may have developed toler-
ance to the levels of free iodine present in povidone-io-
dine; tolerance of P. cepacia to levels of free iodine higher
than those found in currently marketed povidone-iodine
solutions has been described ( F A V E R O MS. Personal
communcation). Although we could not document toler-
ance to 10% povidone-iodine solutions in the laboratory,
we used organisms isolated from Pharmadine that had
been passed through artificial media; partial or complete
loss of iodine resistance after subculturing in broth has
been documented (20).
Although we favor the hypothesis that the P. cepacia
was tolerant to levels of free iodine found in povidone-
iodine solution, further work is needed with organisms
comparable to those free growing in deionized water.
Further work is also needed to ascertain whether free
iodine alone accounts for the microbicidal activity of the
solution.
Iodophors fill a unique and useful role in current medi-
cal practice. Confirmation of bacterial contamination of
povidone-iodine solution, however, has raised serious
questions about the way in which these products should
be manufactured and used. More complete information is
needed on the mechanism of the microbicidal activity of
iodophors and the efficacy of the iodophor products.
A C K N O W L E D G M E N T S : The authors thank Mr. Murray W. Winicov,
West Agro-Chemical, Inc., for measuring free iodine concentrations and for
his discussions with the authors on povidone-iodine chemistry; Ms. Betty W.
Holland and Ms. Claudine B. Bryant, Hospital Infections Laboratory Sec-
tion, Centers for Disease Control, for their assistance with the laboratory
investigations; Lilla Lyon, M.D., New York City Department of Health, for
her assistance with the epidemiologic investigation; and Ms. Ruth W. Slade
for her excellent secretarial support.
• Requests for reprints should be addressed to Ruth L. Berkelman, M.D.;
36 July 1981 • Annals of Internal Medicine • Volume 95 • Number 1
Downloaded From: http://annals.org/ by a Michigan State University Libraries User on 08/14/2018
Hospital Infections Branch, Center for Infectious Diseases, Centers for Dis-
ease Control; Atlanta, G A 30333.
References
1. K A S L O W RA, M A C K E L DC, M A L L I S O N G F . Nosocomial pseudobac-
teremia: positive blood cultures due to a contaminated benzalkonium
antiseptic. JAMA. 1976;236:2407-9.
2. W E A V E R RE, T A T U M WW, H O L L I S D G . The Identification of Unusual
Pathogenic Gram-Negative Bacteria (Elizabeth O. King). Atlanta: Cen-
ter for Disease Control; 1972.
3. B A U E R AW, K I R B Y W M M , SHERRIS JC, T U R C K M. Antibiotic suscep-
tibility testing by a standardized single disk method. Am J Clin Pathol.
1966;45:493-6.
4. WINICOV MW, SCHMIDT W, inventors. West Laboratories, Inc., assign-
ees. Germicidal compositions and methods for preparing the same. U.S.
patent 3,028,299. 1962 April 3. 16 p. Int CI 167-17.
5. H U G H R, G I L A R D I GL. Pseudomonas. In: L E N N E T T E EH, B A L O W S A,
HAUSLER WJ J R , T R U A N T JP, eds. Manual of Clinical Microbiology.
3rd ed. Washington, D.C.: American Society for Microbiology;
1980:288-317.
6. EDERER GM, MATSEN JM. Colonization and infection with Pseudomo-
nas cepacia. J Infect Dis. 1972;125:613-8.
7. M E Y E R GW. Pseudomonas cepacia septicemia associated with intrave-
nous therapy. Calif Med. 1973;119 (July):15-8.
8. W E I N S T E I N RA, E M O R I T G , A N D E R S O N RL, S T A M M WE. Pressure
transducers as a source of bacteremia after open heart surgery: report of
an outbreak and guidelines for prevention. Chest. 1976;69:338-44.
9. N O R I E G A ER, R U B I N S T E I N E, SIMBERKOFF MS, R A H A L JJ J R . Sub-
acute and acute endocarditis due to Pseudomonas cepacia in heroin
addicts. Am J Med. 1975;59:29-36.
10. WEINSTEIN RA, STAMM WE. Pseudoepidemics in hospital. Lancet.
1977;2:862-4.
11. C E N T E R FOR DISEASE CONTROL. Nosocomial pseudobacteremia. Mor-
bid Mortal Weekly Rep. 1980;29:243-4,249.
12. F O O D AND D R U G ADMINISTRATION. O-T-C topical antimicrobial prod-
ucts. Fed Reg. 1978:1218,1231-6.
13. H E R M A N LG. Sources of the slow-growing pigmented water bacteria.
Health Lab Sci. 1976;13:5-10.
14. S T A M M JM, E N G E L H A R D WE, PARSONS JE. Microbiological study of
water-softener resins. Appl Microbiol. 1969;18:376-86.
15. RODEHEAVER G , TURNBULL V, EDGERTON M T , K U R T Z L , EDLICH
RF. Pharmocokinetics of a new skin wound cleanser. Am J Surg.
1976;132:67-74.
16. HARVEY SC. Antiseptics and disinfectants; fungicides; ectaparasiticides.
In: G I L M A N A G , G O O D M A N LS, G I L M A N A, eds. Goodman and Gil-
man's The Pharmacological Basis of Therapeutics. 6th ed. New York:
Macmillan Publishing Co., Inc.; 1980:973.
17. A L L A W A L A NA, RIEGELMAN S. The properties of iodine in solutions of
surface-active agents. J Am Pharm Assoc. 1953;42:396-401.
18. P R I N C E H N , N O N E M A K E R WS, N O R G A R D RC, P R I N C E DL. Drug re-
sistance studies with topical antiseptics. J Pharm Sci. 1978;67:1629-31.
19. H O U A N G ET, G I L M O R E OJA, R E I D C, S H A W EJ. Absence of bacterial
resistance to povidone iodine. J Clin Pathol. 1976;29:752-5.
20. FAVERO MS, D R A K E CH. Factors influencing the occurrence of high
numbers of iodine-resistant bacteria in iodinated swimming pools. Appl
Microbiol. 1966;14:627-35.