Human Urine Is Not Sterile - Shift of Paradigm: Urologia
Human Urine Is Not Sterile - Shift of Paradigm: Urologia
Human Urine Is Not Sterile - Shift of Paradigm: Urologia
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
Michael I. Kogan a Yulia L. Naboka b Khalid S. Ibishev a Irina A. Gudima b
Kurt G. Naber c
Departments of a Urology and b Microbiology, Rostov State Medical University, Rostov on Don, Russia;
c
Technical University of Munich, Munich, Germany
Abstract
Objective: Until recently the generally accepted paradigm Introduction
implied that urine of healthy people is sterile. In the present
study, urine of healthy subjects was investigated by extend- Until recently the generally accepted paradigm im-
ed bacteriological methods. Material and Methods: Three plied that urine of healthy people is sterile [1]. At the same
midstream urine samples from 52 healthy subjects each (24 time, bacterial presence in urine associated with inflam-
females, 28 males; 18–25 years of age) were investigated by matory reactions and clinical symptoms indicated uri-
an extended set of culture media for identification of faculta- nary tract infection (UTI). Lack of clinical symptoms in
tive aerobic (FAB) and nonclostridial anaerobic bacteria the presence of bacteria in urine with or without pyuria is
(NCAB). Ward’s method (Euclidean distance) was used for characterized as asymptomatic bacteriuria (ABU). In the
similarity analysis. Results: The bacterial count of FAB in past the latter was generally considered as an UTI precur-
urine was usually low (≤102 colony-forming units/ml) in both sor because ABU has been found particularly in popula-
groups. In contrast, the bacterial count of NCAB was higher tions more likely to develop pyelonephritis, i.e. individu-
(≥103 colony-forming units/ml), at least in some species, als with diabetes mellitus, pregnancy, obstructive uropa-
with significant differences between genders. The average thy, and past history of the urinary tract [2]. However,
number of bacterial species found was 5.8 in female and 7.1 well-performed randomized clinical studies have docu-
in male urine. Half of the females were assigned to a specific mented that the treatment of bacteriuria of asymptom-
‘female’ microbial spectrum, different from that of males. In atic patients does not provide any benefit for the patient
the mixed-gender clusters, the males showed a greater sim- and therefore treatment is generally not indicated [3].
ilarity among themselves. Conclusions: As also shown by Sometimes ABU may even be protective for recurrent
other investigators, urine of healthy people is normally not UTI when left untreated [4, 5]. Thus, the paradigm of
sterile. The role of the routinely not cultivated bacteria in ABU has completely shifted and nowadays it is consid-
© 2015 S. Karger AG, Basel Kurt G. Naber, MD, PhD, Associate Professor of Urology
0042–1138/15/0944–0445$39.50/0 Technical University of Munich
Karl-Bickleder-Strasse 44c
E-Mail [email protected]
DE–94315 Straubing (Germany)
www.karger.com/uin
E-Mail kurt @ nabers.de
ered a bacterial colonization of the urinary tract, similar The study was conducted among two groups of individuals.
to commensalism at other mucosal sites [6]. Under cer- Group I consisted of 24 healthy sexually active females (18–25
years), and group II consisted of 28 healthy sexually active males
tain circumstances, however, differentiation between (20–25 years). Inclusion and exclusion criteria were the following:
UTI and ABU may be difficult when the diagnosis has to age (25 years or younger); absence of gynecological or urological
be made from nontypical clinical symptoms, such as in disorders and sexually transmitted infections in past history and at
very young children, in patients with indwelling urinary the time of examination; absence of infectious diseases in the pre-
catheter, or in patients with neuropathic bladder due to vious year; no genetic, systemic lesions, autoimmune diseases, or
risk factors (radiation exposure, etc.); no abnormal changes in kid-
spinal cord injuries, with the consequence of overtreat- neys, urinary system, or internal genitalia according to ultrasonog-
ment as well as undertreatment in some cases [7, 8]. Mei- raphy, no medications including antibiotics and corticosteroids
jer-Severs et al. [9] investigated the involvement of an- used within the last 2 months, and standard clinical blood and
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
aerobic bacteria in pregnant women with ABU and tried urine analyses within normal ranges.
to differentiate between invasive (silent renal infection) Midstream clean-catch urine samples were collected from the
individuals for bacteriological examination in sterile disposable
and noninvasive ABU by antibody-coating of aerobic and containers by specifically trained medical staff with special empha-
anaerobic bacteria. Therefore, some questions remain. sis on proper hygienic procedures. The samples were coded with
What essentially is ABU? Can it be even a general condi- an identification number and subdivided in two aliquots: one for
tion of healthy people? Is human urine normally sterile? standard urinalysis (white and red blood cells count, etc.), and the
The Fundamental Human Microbiome Project other for standard and extended bacteriological investigations of
the urine microbiota. From each individual, three urine samples
(http://nihroadmap.nih.gov/hmp/;www/human- were collected, with an interval of 3 days, and analyzed.
microbiome.org) initiated in 2008 by the US National
Institutes of Health using detection of bacterial 16S Bacterial Identification
rRNA has considerably shaken the existing concepts, In total, bacteriological tests of 156 urine samples were per-
which imply ‘sterility’ of various human biotopes, and formed using a standard and extended set of culture media for
identification of FAB and NCAB (MacConkey agar, HiCrome
also extended the knowledge of the microbial spectrum Candida differential agar, HiCrome Enterococcus faecium agar
in the gastrointestinal and urogenital tracts, skin, and base, HiCrome Aureus agar base, blood agar prepared using the
other human sites [10]. Published work has covered not Mueller-Hinton agar base with addition of sheep red blood cells,
only the bacterial microbiome, but also the viral micro- Bacteroides bile esculin agar, Schaedler agar, Schaedler broth, MRS
biome of a healthy human organism [11]. Complete agar). HiCrome nutritional media (HiMedia, India) were used in
this investigation. Culture media inoculated with sampling mate-
genome sequencing, however, is a labor-intensive and rial allowing quantification to 10 colony-forming units (CFU) per
expensive method, which is not available in routine milliliter (ml) were incubated under aerobic (24–48 h) and anaer-
clinical practice. That is why careful bacteriological obic (48–72 h) conditions (10% СО2, 10% H2, 80% N2). Microor-
examination is still considered the gold standard of mi- ganisms were identified in accordance with their morphological
crobiological diagnostics but, in our opinion, has insuf- and hemolytic properties. Smears were prepared from colonies
grown in culture media and heat fixed for Gram staining according
ficiently studied urogenital infectious and inflammatory to standard protocol. Using oil immersion microscopy (×900), 10–
diseases – an opportunity for further improvement. 15 representative fields were reviewed. Final identification of mi-
The purpose of this work was to use bacteriological croorganisms was carried out in accordance with their biochemi-
methods for identification of microbiota in the urine of cal properties using entero-, staphylo-, and anaerotests (Lachema,
healthy individuals using an extended set of culture me- Czech Republic).
After identification of the microorganisms, dendrograms were
dia for cultivation of facultative aerobic (FAB) and non- generated for clustering the differences of the urinary microbes
clostridial anaerobic bacteria (NCAB). evaluated separately for their spectrum and concentration. Ward’s
method (Euclidean distance) was used for similarity analysis [12].
Computing was performed in R (version 3.1; R Foundation for
Statistical Computing, Vienna, Austria).
Subjects and Methods
FAB
CNSc 83.3 103 89.3 102
Corynebacterium sp. 75.0 102 78.6 102
Enterobacteriaceae 16.7 102 10.7 102
S. aureus 16.7 102 10.7 102
102 102
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
Enterococcus sp. 12.5 50.0
Micrococcus sp. 12.5 102 0 0
Streptococcus sp. 8.3 102 0 0
Candida sp. 33.3 102 0 0
Bacillus sp. 20.8 102 0 0
NCAB
Lactobacillus sp. 83.3 104 0 0
Peptococcus sp. 75.0 103 21.4 102
Propionibacterium sp. 58.3 104 10.7 102
Eubacterium sp. 41.7 105 78.6 103
Peptostreptococcus sp. 41.7 103 50.0 102
Bacteroides sp. 25.0 104 21.4 103
Veillonella sp. 16.7 103 10.7 102
Prevotella sp. 12.5 102 0 0
Actinomyces sp. 8.3 102 0 0
Megasphaera 0 0 21.4 102
Mobiluncus sp. 0 0 10.7 103
Fusobacterium sp. 0 0 10.7 102
healthy females and males contained different variations der clusters, the males showed a greater similarity among
of FAB and NCAB (table 1). themselves than with the females. The majority of females
The FAB pattern in group I was considerably broader (66.6%) of the mixed-gender clusters were the last to be
than the FAB pattern of group II. The dominant clusters added, i.e. had higher levels of dissimilarity.
in both groups were represented by coagulase-negative The bacterial count of FAB in urine was low in both
staphylococci and Corynebacterium sp. Urine of healthy groups (102 CFU/ml), except for CNS in the female group.
men contained no Micrococcus sp., Streptococcus sp., In contrast, the bacterial count of Eubacterium sp., Lacto-
Candida sp., or Bacillus sp., which were, however, present bacillus sp., Propionibacterium sp., Bacteroides sp., Pepto-
in female urine. In male and female urine there was a wide coccus sp., Peptostreptococcus sp., and Veillonella sp. in
pattern of 9 NCAB genera, but with certain differences group I and Eubacterium sp., Bacteroides sp., and Mobi-
between the dominant clusters. In group I, NCAB were luncus sp. in group II was ≥103 CFU/ml. The concentra-
mostly represented by Lactobacillus sp., Peptococcus sp., tions of urinary bacteria of most females (83.0%) signifi-
and Propionibacterium sp., and in group II by Eubacte- cantly differed from those in males (fig. 2). In the mixed-
rium sp. Urine of healthy women contained no Mega- gender clusters, the males again demonstrated greater
sphaera, Mobilluncus sp., or Fusobacterium sp., while similarity among themselves.
urine of healthy men contained no Lactobacillus sp., Pre- In table 2 the results of two triple microbiological tests,
votella sp., or Actinomyces sp. These differences are re- one from female and one from male urine, using the ex-
flected in the dendrograms (fig. 1). Exactly one half of all tended set of culture media are shown as examples.
females were assigned to one specific ‘female’ microbial The range of microorganisms recovered from urine
spectrum, different from that of males. In the mixed-gen- was quite wide and variable. In the female subjects, the
12
10
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
Dissimilarity level
0
f f f f f f f f f f f f mmmmmmmmmmm f f f f f f f f mmmmmmmmmmm f mmmmmm f f f
Fig. 1. Hierarchical clustering by urinary bacterial composition for males and females. m = Male; f = female.
60
50
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
Dissimilarity level
40
30
20
10
0
mmmmmmmmmmmmmmmmm f mmmm f mmmmmmm f f f f f f f f f f f f f f f f f f f f f f
Fig. 2. Hierarchical clustering by urinary bacterial concentration for males and females. m = Male; f = female.
catch samples, contaminants from the distal region of the differences concerning range and concentration of micro-
urethra might still be included but very minimized by di- organisms. As demonstrated in earlier studies, microbiota
lution. A similar approach of cultivation was chosen by found in urine of sexually active and inactive healthy
Hilt et al. [16], although urine specimens were obtained women were similar [18]. In contrast, the spectrum of
by transurethral catheter. Microbial communities identi- NCAB bacteria isolated from women with recurrent cys-
fied by Wolfe et al. [15] using 16S rRNA sequencing in titis was wider and the level of bacteriuria higher (p < 0.05)
parallel urine samples were similarly collected by trans- than in healthy virgin or sexually active women [19, 20].
urethral catheter and suprapubic aspirate. The microbial Thus, a wide range of FAB and NCAB of various con-
communities determined by 16S rRNA PCR and deep py- centrations found in urine is a natural condition in
rosequencing in male first voided urine are highly similar healthy humans, who are essentially asymptomatic. In-
to those in paired urethral swab specimens [17]. terestingly, the urinary bacterial spectra and concentra-
From our results, it can be concluded that urine of tions are distinguished by gender to a high degree as ana-
healthy sexually active men and women contain aerobic lyzed by dendograms.
and anaerobic microorganisms with the exception of rare Standard bacteriological examination of urine allows
cases (8.3%) when no aerobic microorganisms were found only to find microorganisms fast growing in the presence
in female urine. The number of species of microorganisms of oxygen. All female and male subjects investigated in this
varied widely (average number of species was 5.8 for study showed ‘no growth’ according to standard microbi-
women and 7.1 for men), with considerable sex-related ology, even when using a fairly low threshold of CFU ≥103/
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
Bacteroides sp. – – Bacteroides sp.
Lactobacillus sp. 103 Lactobacillus sp. 103 Lactobacillus sp. 103
Veillonella sp. 103 Veillonella sp. 101 Veillonella sp. 101
Peptococcus sp. 103 – – Peptococcus sp. 103
Prevotella sp. 102 Prevotella sp. 101 Prevotella sp. 102
Group II (male No. 15: 21 years)
Corynebacterium sp. 102 Corynebacterium sp. 102 Corynebacterium sp. 102
S. haemolyticus 102 S. haemolyticus 101 S. haemolyticus 102
S. lentus 102 – – S. xylosus 102
Propionubacterium sp. 102 Propionubacterium sp. 102 Propionubacterium sp. 103
Megasphaera 102 Megasphaera 102 Megasphaera 102
Peptostreptococcus sp. 102 Peptostreptococcus sp. 101 Peptostreptococcus sp. 102
Eubacterium sp. 102 Eubacterium sp. 102 Eubacterium sp. 103
– – Peptococcus sp. 101 – –
ml as ‘significant bacteriuria’. Only the use of an extended [7] also reported about the dominant role of Corynebac-
set of culture media, as in this study, has allowed the dis- terium sp. in the microbiome of healthy men using 16S
covery of aerobic and anaerobic microorganisms in urine, rRNA sequencing. They also found that Lactobacillus sp.,
information previously unknown concerning nonsterility Corynebacterium sp., Gardnerella sp., Prevotella sp., and
of urine in healthy individuals. Most of these media are not Enterococcus sp. determine the sex-related differences.
used for standard urine microbiology, but are used in oth- Our bacteriological data obtained from urine of healthy
er bacteriological examinations, e.g. of vaginal and intesti- sexually active men also revealed a dominance of coagu-
nal flora [21, 22]. This selection of culture media, which we lase-negative staphylococci (89.3%) and Corynebacteri-
have used in the past, might not be perceived as the final um sp. (78.6%) among FAB.
standard. It can and should be modified and supplemented Accepting the paradigm shift that human urine is not
in accordance with research and clinical purposes. For re- sterile, a careful analysis of the urinary microbiota and the
liable recovery of FAB and NCAB, however, multiple me- role of the difficult-to-cultivate and thus neglected patho-
dia are absolutely necessary. gens by standard microbiology becomes of utmost prior-
Our results correlate well with studies using 16S rRNA ity for a variety of urogenital diseases related to infection
metagenomic sequencing of urine. Siddiqui et al. [14] and inflammation. Even for clinically distinct infection,
found in their studies 45 different genera of microorgan- like acute cystitis, very often the standard microbiology
isms with domination of Lactobacillus, Prevotella, and does not reveal the expected results. Often women suffer-
Gardnerella referred to the three most frequent phyla: ing from symptoms of UTI have a negative urine culture
Firmicutes, Bacteroides, and Actinobacteria. 16S rRNA when conventional methods are used and their condition
sequences of Firmicutes and Bacteroides were found in all is described as ‘urethral (or ‘dysuria/frequency’) syn-
urine samples. The authors concluded that in urine from drome’. It appears, however, that no attempt has been
healthy females there is a noticeable intraindividual vari- made in any recent study to use urine culture techniques
ation even at the phylum level. Our data also indicate that capable of detecting bacteria other than the recognized
in female urine Lactobacillus spp. are dominant and that aerobic pathogens [23]. In a large study including 4,400
urine microbiota are normally polymicrobial. Fouts et al. adult female patients between 18 and 65 years with acute
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
urogenital infections [24, 26–28]. communities in urine, and should alter the standard ap-
Therefore, with such an extended set of culture media proach to the diagnostics of infectious and inflammatory
as used in our institution, we have started to investigate diseases of the urogenital tract.
the role of difficult-to-cultivate bacteria in the develop-
ment of infectious and inflammatory diseases of different
urogenital organs, e.g. development of acute obstructive Acknowledgments
pyelonephritis due to NCAB and bacterial spectrum of
urine and bladder tissue samples in women with recur- The study was supported by a research grant of the Rostov State
rent UTI, etc. [19, 20, 29–31]. Medical University. The authors are thankful to I.N. Leus and M.L.
Chernizkaya (Department of Microbiology, Rostov State Medical
The gene pyrosequencing method, an important tool University, Rostov on Don, Russia) for technical assistance, and to
for research, will remain available in scientific laborato- S.A. Zarutskiy (Department of Economic Cybernetics, Southern
ries only for specific clinical situations, e.g. diagnostics of Federal University, Rostov State Medical University, Rostov on
atypical microorganisms in patients with chronic prosta- Don, Russia) for performing the dendrograms.
titis/chronic pelvic pain syndrome [32]. Sequencing
methods do not allow (1) to differentiate between alive
and dead bacteria, (2) to detect the level of bacteriuria, or Disclosure Statement
(3) to determine the bacterial antibiotic susceptibility. All
of these issues are of high importance for applied medi- The authors declare no conflict of interest in regard to this
cine. Therefore, careful bacterial examination of the spec- study. The authors M.I. Kogan, Y.L. Naboka, K.S. Ibishev, I.A.
Gudima are members of Rostov State Medical University,
imens should still be considered the gold standard for di- which sponsored the study. K.G. Naber contributed to the study
agnostics and should be improved further by appropriate as Visiting Professor supported by the Rostov State Medical
extension of the culture media. University.
References
1 Zasloff M: Antimicrobial peptides, innate im- 2010, pp 314–318. http://www.icud.info/ ic bladder associated with spinal cord injury.
munity, and the normally sterile urinary tract. urogenitalinfections.html (accessed August J Transl Med 2012;10:174.
J Am Soc Nephrol 2007;18:2810–2816. 2014). 8 Trautner BW, Bhimani RD, Amspoker AB,
2 Kass EH: Bacteriuria and pyelonephritis of 5 Cai T, Mazzoli S, Mondaini N, Meacci F, Nesi Hysong SJ, Garza A, Kelly PA, Payne VL, Nalk
pregnancy. Arch Intern Med 1960; 44: 194– G, D’Elia C, Malossini G, Boddi V, Bartoletti AD: Development and validation of an algo-
198. R: The role of asymptomatic bacteriuria in rithm to recalibrate mental models and re-
3 Nicolle LE, Bradley S, Colgan R, Rice JC, young women with recurrent urinary tract in- duce diagnostic errors associated with cathe-
Schaeffer A, Hooton TM: Infectious Diseases fections: to treat or not to treat? Clin Infect ter-associated bacteriuria. BMC Med Inform
Society of America guidelines for the diagno- Dis 2012;55:771–777. Decis Mak 2013;13:48–56.
sis and treatment of asymptomatic bacteriuria 6 Wagenlehner FME, Naber KG: Asymptom- 9 Meijer-Severs GJ, Aarnoudse JG, Mensink
in adults. Clin Infect Dis 2005;40:643–654. atic bacteriuria – shift of paradigm. Clin In- WF, Dankert J: The presence of antibody-
4 Wullt B, Sundén F: Asymptomatic bacteriuria fect Dis 2012;55:778–780. coated anaerobic bacteria in asymptomatic
with the model strain Escherichia coli 83972 7 Fouts DE, Pieper R, Szpakowski S, Pohl H, bacteriuria during pregnancy. J Infect Dis
protects against symptomatic urinary tract in- Knoblach S, Suh M-J, Huang S-T, Ljungberg 1979:140:653–658.
fections; in Naber KG, Schaeffer AJ, Heyns C, I, Sprague BM, Lucas SK, Torralba M, Nelson 10 Allen-Vercoe E: Antimicrobials. Genomics
Matsumoto T, Shoskes D, Bjerklund-Johan- KE, Groah SL: Integrated next-generation se- bringing the gut microbiota into focus
sen TE (eds): Urogenital Infections. Arnhem, quencing of 16S rDNA and metaproteomics through microbial culture: recent progress
European Association of Urology – Interna- differentiate the healthy urine microbiome and future perspective. Curr Opin Microbiol
tional Consultation on Urological Diseases, from asymptomatic bacteriuria in neuropath- 2013;16:625–629.
Downloaded from http://karger.com/uin/article-pdf/94/4/445/3600347/000369631.pdf by Universidad Nacional Autonóma de México user on 25 November 2024
the female urine microbiota by high through- lower urinary tract infections in Russian). women with chronic recurrent cystitis (in
put sequencing of 16S RNA amplicons. BMC Urologiia 2011;6:12–15. Russian). Urologiia 2013;4:16–18.
Microbiol 2011;11:244–251. 21 Bondarenko VM, Matsulevich TV: Intestine 30 Kogan MI, Pasechnik DG, Naboka IuL, Ibi-
15 Wolfe AJ, Toh E, Shibata N, Rang R, Kenton K, Dysbacteriosis as Clinical and Laboratory shev KhS, Gazaev ZI, Gudima IA: Non-clos-
FritzGerald MP, Mueller ER, Schreckenberger Syndrome: Current State of the Problem. tridial anaerobic bacteria can cause acute
P, Dong Q, Nelson DE, Brubaker L: Evidence Moscow, GEOTAR-Media, 2007, pp 114– pyelonephritis (an experimental trial) (in
of uncultivated bacteria in the adult female 128. Russian). Urologiia 2012;2:8–13.
bladder. J Clin Microbiol 2012;50:1376–1383. 22 Menshikov VV: Methodology of Clinical Lab- 31 Naboka IuL, Kogan MI, Gudima IA, Tcher-
16 Hilt E, McKinley K, Pearce MM, Rosenfeld oratory Studies. Moscow, Labora, 2009, pp nitskaia ML, Ibishev KhS, Khasigov AV, Mi-
AB, Zilliox MJ, Mueller ER, Brubaker L, Gai 51–83. tusova EV: Role of nonclostridial anaerobes in
X, Wolfe AJ, Schreckenberger P: Urine is not 23 Maskell RM: The natural history of urinary the development of infectious and inflamma-
sterile: use of enhanced urine culture tech- tract infection women. Med Hypotheses tory diseases of the urinary and reproductive
niques to detect resident bacterial flora in the 2010;74:802–806. systems (in Russian). Urologiia 2013; 6: 118–
adult female bladder. J Clin Microbiol 2014; 24 Naber KG, Schito G, Botto H, Palou J, Mazzei 121.
52:871–876. T: Surveillance study in Europe and Brazil on 32 Xiao J, Ren L, Lv H, Ding Q, Lou S, Zhang W,
17 Dong Q, Nelson DE, Toh E, Diao L, Gao X, clinical aspects and antimicrobial resistance Dong Z: Atypical microorganisms in ex-
Fortenberry JD, Van Der Pol B: The micro- epidemiology in females with cystitis pressed prostatic secretion from patients with
bial communities in male first urine are high- (ARESC): implications for empiric therapy. chronic prostatitis/chronic pelvic pain syn-
ly similar to those in paired urethral swab Eur Urol 2008;54:1164–1178. drome: microbiological results from a case-
specimens. PLoS One 2011;6:e19709. 25 Hooton TM, Roberts PL, Cox ME, Stapleton control study. Urol Int 2013;91:410–416.
AE: Voided midstream urine culture and
acute cystitis in premenopausal women. New
Engl J Med 2013;369:1883–1891.
26 Mohanty NK, Jolly BB: Incidence of anaero-
bic bacterial infection following transurethral
instrumentation. Indian J Pathol Microbiol
1996;39:33–36.