Disponible en ligne sur

www.sciencedirect.com
Mdecine et maladies infectieuses 43 (2013) 322330

General review

Broad-range PCR: Past, present, or future of bacteriology?

La PCR universelle : pass, prsent ou futur de la bactriologie ?
A. Renvois , F. Brossier , W. Sougakoff , V. Jarlier , A. Aubry
Laboratoire de bactriologie-hygine, facult de mdecine Pierre-et-Marie-Curie Paris-6, site Piti-Salptrire, 91, boulevard de lHpital, 75634 Paris cedex 13,
France
Received 6 February 2013; received in revised form 8 April 2013; accepted 17 June 2013
Available online 19 July 2013

Abstract
PCR targeting the gene encoding 16S ribosomal RNA (commonly named broad-range PCR or 16S PCR) has been used for 20 years as a
polyvalent tool to study prokaryotes. Broad-range PCR was first used as a taxonomic tool, then in clinical microbiology. We will describe the use
of broad-range PCR in clinical microbiology. The first application was identification of bacterial strains obtained by culture but whose phenotypic
or proteomic identification remained difficult or impossible. This changed bacterial taxonomy and allowed discovering many new species. The
second application of broad-range PCR in clinical microbiology is the detection of bacterial DNA from clinical samples; we will review the clinical
settings in which the technique proved useful (such as endocarditis) and those in which it did not (such as characterization of bacteria in ascites,
in cirrhotic patients). This technique allowed identifying the etiological agents for several diseases, such as Whipple disease. This review is a
synthesis of data concerning the applications, assets, and drawbacks of broad-range PCR in clinical microbiology.
2013 Elsevier Masson SAS. All rights reserved.
Keywords: 16S ribosomal RNA; Broad-range PCR; Molecular diagnosis

Rsum
La PCR amplifiant le gne codant pour lARN ribosomal 16S (plus communment appele PCR 16S ou PCR universelle) est utilise depuis
20 ans comme outil dtude polyvalent des procaryotes. La PCR 16S a dabord t utilise comme outil dtude taxonomique, puis son usage
sest rpandu en microbiologie clinique. Dans cette revue, nous dtaillons les applications de la PCR 16S en microbiologie clinique. La premire
application a t lidentification de souches bactriennes obtenues par culture, mais dont lidentification phnotypique ou protomique est difficile
ou impossible. Cette utilisation a modifi les perspectives de la taxonomie bactrienne et a permis la dcouverte de nombreuses nouvelles espces.
Lautre application de la PCR 16S en microbiologie clinique est la dtection dADN bactrien directement partir de prlvements cliniques ; nous
proposons une synthse des situations cliniques dans lesquelles cette technique est utile (par exemple les endocardites) et celles dans lesquelles
elle ne lest pas (par exemple pour les infections du liquide dascite chez les patients cirrhotiques). Cette technique a permis didentifier lagent
infectieux responsable de plusieurs pathologies (par exemple la maladie de Whipple). Cette revue fait la synthse des donnes sur les indications,
avantages et inconvnients de la PCR 16S en microbiologie clinique.
2013 Elsevier Masson SAS. Tous droits rservs.
Mots cls : ARN ribosomal 16S ; Diagnostic molculaire ; PCR universelle

1. Introduction

Corresponding author.
E-mail address: alexandra.aubry@upmc.fr (A. Aubry).

0399-077X/$ see front matter 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.medmal.2013.06.003

The debate on the contribution of 16S PCR can be heated

and opposes those who support its use and consider it as a
tool for broad-range identification to those who consider that its
indications are limited. This molecular technique often called
broad-range has rapidly become an essential element for

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

323

Fig. 1. Schematic representation of 16S ribosomal RNA gene. Non-hypervariable regions are those containing conserved regions used as target sequences for
universal primers. Amplification followed by sequencing of hypervariable regions can discriminate between bacterial species. Schematic representations of short
and long amplification of the gene are presented.
Reprsentation schmatique du gne codant pour lARN ribosomique 16S. Les rgions non hypervariables sont celles au sein desquelles on trouve les rgions
conserves utilises comme squences cibles pour les amorces universelles . Les rgions hypervariables sont celles dont lamplication et le squencage permettent
de diffrentier les espces bactriennes. Les reprsentations schmatiques dun 16S long et dun 16S court sont galement gures.

bacteriological diagnosis in hospitals. We reviewed the literature to determine the contribution of this technique to clinical
microbiology.
2. What is 16S? What is its contribution?
2.1. The bacterial ribosome
The ribosome is a ribonucleoprotein complex (made up of
proteins and RNA); it allows synthesizing proteins (also called
translation) by using mRNA as a source of information and
tRNA associated with amino acids as substrates. In bacteria,
ribosomes are composed of a large sub-unit (50S) and a small
sub-unit (30S). The functional ribosome (composed of the two
sub-units assembled around the mRNA) has a molecular mass
of 2.5-megadalton and a sedimentation coefficient of 70S. The
small sub-unit is composed of 16S ribosomal RNA (encoded
by a gene of 1500 nucleotides) and of 20 proteins; it allows
reading mRNA. The large sub-unit is composed of 23S ribosomal RNA (encoded by a gene of 2900 nucleotides), of 5S
ribosomal RNA (encoded by a gene of 120 nucleotides), and
of 30 proteins; it allows synthesizing the protein corresponding to the mRNA read by the small sub-unit. Furthermore,
various protein factors act on the ribosome at various stages of
translation.

In the 1980s, it was demonstrated that the phylogenic relationships among living beings could be determined by comparing
their nucleic sequences [1]. Indeed, since the 16S rRNA gene
encode for an rRNA with a constant function in evolution, it
could be used as a molecular timer to follow changes in bacterial evolution. This gene was used this way in the late 1980s, as
a study tool for bacterial evolution [3], and had a major part in
the study of bacterial phylogeny and taxonomy [4].

3. Limitations of broad-range 16S PCR. Global

problems
3.1. Cost and lack of automatization
The 16S molecular tool has some global limitations. Firstly,
the cost remains high, even though it was lowered since the technique was first described [5]. Some authors suggest performing a
short 16S (cf. addendum) to reduce the cost while maintaining
a good taxonomic value (Fig. 1) [1,6].
Furthermore, even though marketed systems such as
MicroSeqTM (Applied Biosystems) were developed [7], the
non-automatization of the technique was a limiting factor for
its global use. Nevertheless, the development of high output sequencing techniques could allow incorporating stages of
broad-range 16S PCR in a robotized system.

2.2. 16S rRNA

Ribosomal RNA (rRNA) 16S is the constituent RNA of the
small ribosomal sub-unit of prokaryote 30S (Fig. 1). The gene
encoding this rRNA is the 16S rRNA gene also called ribosomal 16S RNA or rrs [1], present in all bacterial species in a
variable number of copies [2]. It is composed of 1500 nucleotides
and includes nine hypervariable regions. The association of conserved regions and variable regions theoretically allows using
this gene to identify and detect all bacterial species.

3.2. Volume of the test sample

The detection threshold for end-point PCR (such as 16S) is
weak (theoretically 1 to 5 copies of DNA), but only 1 to 5 L of
the sample are used for PCR, whereas 100 to 5000 times greater
volumes are used for usual bacterial culture. The weak volume
of sample tested and PCR inhibitors may be associated to false
negative results (Fig. 2) [2].

324

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

Fig. 2. Contribution and limitations of broad-range PCR.

Intrts et limites de la PCR 16S.

3.3. False positive results

Contaminations during gene amplifications may occur and
make it difficult to interpret results (Fig. 2). The problem of
false positives may be limited by: using DNA free reagents,
dedicated pipettes, cotton tip pipettes, three separate rooms for
each step (1: preparation of reagent mix, 2: DNA extraction,
and 3: amplification), nucleotides containing dUTP (deoxyuridine triphosphate which is used instead of dTTP during PCR;
using uracil-DNA glycosylase in the reagent mix for the next
PCR allows degrading the DNA containing uracil, but has no
effect on natural DNA containing thymine) and with the expertise of a confirmed microbiologist. Using negative control for
each step of the protocol (extraction-amplification-sequencing)
is mandatory to detect these false positive results [8].

4. When should 16S PCR be used?

4.1. 16S PCR and bacterial identication
4.1.1. Using 16S PCR for bacterial identication of strains
isolated in culture
Bacterial identification from cultures usually relies on the
phenotypic characteristics of the bacterium: staining (Gram
for example), morphology, ability to grow on some culture
media, biochemical features detected by various techniques
on the market (APITM galleries [Biomrieux], VitekTM systems [Biomrieux], PhoenixTM [BD biosciences], BiologTM
[Biolog], etc.). Nevertheless, some bacteria are badly identified
phenotypically for various reasons:
small number of phenotypic characters expressed;
stress may have altered the phenotypic characters;

absence of rare bacteria in the databases of systems available

on the market;
phenotypic characters difficult to detect for some bacteria
difficult to cultivate.
In these cases, amplification and sequencing of the 16S rRNA
gene followed by the comparison of the obtained sequence with
databases have proved their value for bacterial identification
[9]. This is a broad-range method, accurate and reliable, the
inter-operator variability of which is limited compared to usual
techniques [2].
The authors of two important studies measured the performance of 16S PCR for the identification of clinical isolates not
identified by conventional methods. Kiratisin et al. reported that
isolates not identified by conventional techniques had been identified by a long 16S PCR (cf. addendum) at the species level for
74% of strains compared to 83.1% for Mignard et al. who had
used a short 16S (cf. addendum); at the genus level for 21% of
strains compared to 15.8% for Mignard et al.; and that 1 to 5% of
strains could definitely not be identified by Kiratisin et al. compared to 1% by Mignard et al. [5,10]. Furthermore, 16S PCR was
especially effective for the identification of Gram-positive aerobic bacilli. For 136 clinical strains of Gram-positive bacilli badly
identified by conventional techniques (only 52.2% of strains had
been identified at the genus level by usual techniques), Bosshard
et al. reported that 16S PCR had identified 65.4% of strains at
the species level (defined by similarity 99%) and 31.6% more
at the genus level (defined by similarity ranging from 95 to 99%)
[11].
Nevertheless, 16S PCR is weakly effective as an identification
tool for some species when the variations of sequences are too
small among species to allow discriminating. This is the case, for
example, for: some streptococci (Streptococcus mitis and Streptococcus pneumoniae), some enterobacteria (Escherichia coli

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

and Shigella spp.), or some Bacillus spp. (Bacillus cereus and

B. anthracis) [2,4,10,12]. In these cases, other genes can be
used to discriminate among these species, such as rpoB (which
encodes the sub-unit beta of RNA polymerase), or other target
genes including variable sequences surrounded by conserved
sequences such as: recA, tuf, gyrA, and gyrB [12].
Finally, some species have several copies of the 16S rRNA
gene. These copies may present with variations, which has for
consequence to generate sequences containing ambiguousness
[2]. Furthermore, some authors have reported a certain degree of
micro-heterogeneity among species; this corresponds to intraspecies variations inferior to 0.5% and to different genotypes of
sub-species [1]. But these small variations would have a weak
impact on routine hospital practice [10].
4.1.2. Contribution of 16S PCR for the description of new
species
Defining the concept of bacterial species is a delicate point;
the correspondence between a strain and a previously described
species is based on phenotypic and genetic similarity [13]. The
current methods used to define prokaryote species do not allow
covering the diversity found in nature and bacterial taxonomy
is influenced by breakthroughs in the genetics of populations,
ecology, genomics, and by the facility with which data may be
obtained [13]. Traditionally, a bacterial species is defined from
DNA-DNA hybridization, which is a complex and costly technique, less and less frequently used [14]. The genetic definition
of a species is quantifiable taking into account the kinetics of
DNA-DNA recombination. A species is then defined genetically as a group of strains, which have DNA-DNA relationships
resulting in:
a rate of DNA-DNA hybridization greater or equal to 70%;
thermic stability of hybrids less or equal to 5 C [4].
The Stackebrandt committee determined that any description of a new species should include a complete sequence of
the gene encoding for 16S rRNA [14]. Nevertheless, there is
no clearly determined threshold value for rates of similarity,
beyond which the scientific community agrees to define the rank
of species [4]. Indeed, if a close similarity ( 97%) between two
16S sequences does not allow systematically determining that
two strains belong to the same species, the contrary holds true:
similarity between two 16S sequences less than 97% allows
determining that the corresponding strains belong to different
species [4,13,15]. When 16S rRNA sequences present more than
97% of homology, the Stackebrandt committee recommends that
the study of DNA-DNA hybridization rate as well as thermic
stability of hybrids remain the reference to define a genomic
bacterial species [14].
The routine use of 16S PCR for bacterial identification in a
laboratory has allowed discovering new species or new bacterial genera. Numerous strains with a rate of genetic similarity
inferior to 97% were thus discovered thanks to the globalization
and the availability of the molecular tool. Broad-range PCR has
imposed itself as a means of discovering new taxa. For example,
Drancourt et al. reported discovering 11 new bacterial species

325

among 1404 isolates for which usual techniques had not been
contributive [15]. Rates of similarity greater than 99% had been
chosen to assign a strain to a previously described species, in
this study; 97 to 99% to assign it at the genus level; and less
than 97% to consider it as a new species. Thus, 16S PCR is tool
of great importance for the exploration of bacterial diversity
[4,15].
Finally, the main problem is the threshold from which a rate
of similarity would allow assigning the studied sequence to a
species, or to a genus, is not clearly determined. Several thresholds have been suggested. Most taxonomists accept thresholds of
97% for the genus and 99% for species [2]. Nevertheless, some
authors suggest using a threshold of 99.5% for species, whereas
for others, it seems impossible to determine a single threshold
for all the bacterial world [6]. Recommendations for the interpretation by bacterial categories were recently suggested by
the Clinical and Laboratory Standard Institute (CLSI), but their
prohibitive costs restrict their availability to routine laboratories
[12].
It should be noted that the rates of similarity obtained differ, whether a short 16S or a long 16S is used, but also
depending on the program used, and the parameters used for a
given program (cf. addendum) [1]. It should also be kept in mind
that the final interpretation of a bacterial identification result
based on 16S rRNA, should obviously also take into account the
phenotypic characteristics.

4.2. 16S PCR and bacterial detection from clinical samples

4.2.1. Global features
Even though 16S PCR has limitations, it is an alternative
method, independent of culture, used for bacterial detection
directly from clinical samples [2]. The advantage of 16S PCR
compared to usual bacterial culture was clearly demonstrated in
the following circumstances: (1) detection of bacteria difficult
to grow (mycobacteria and other bacteria containing mycolic
acids, intracellular bacteria such as Coxiella burnetii, Bartonella
spp., Ehrlichia spp., Anaplasma spp., Francisella tularensis,
Mycoplasma spp.) and (2) detection of bacteria the usual culture
of which is made impossible because of, among other reasons,
a previous antibiotic treatment (Fig. 2) [16]. But, as Rampini
et al. have very well demonstrated, the contribution of this technique depends almost exclusively on the value of criteria used
to select patients suspects of infection, as is illustrated in Fig. 3.
It should be specified that 16S PCR used for bacterial detection
never allows strain culture and thus prevents obtaining an antibiogram, which is a limitation for the therapeutic management of
patients. It should also be noted that detection of bacterial DNA
(by amplification of all or a part of the gene) in a sample, if it
is positive, should be followed by bacterial identification with
sequencing of all or a part of the 16S gene (which is similar to
identification of strains mentioned in Section 4.1).
Furthermore, the scope of 16S PCR is not theoretically
restricted to samples from normally sterile sites and sites of
mono-microbial infections, because cloning and pyrosequencing techniques allow identifying various microorganisms from

326

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

Fig. 3. It is essential to select patients for whom a diagnostic test is contributive (thus improving pre-test probability). Defining criteria for a biological test: sensitivity,
specificity, negative predictive value (NPV), and positive predictive value (PPV). This is a previously reported example, with a sensitivity of 93% and a specificity
of 98% for broad-range PCR applied in a population with a prevalence of meningitis at 44% [17]. Broad-range PCR is first performed on cerebrospinal fluid (CSF)
for any patient. This corresponds to a low prevalence of meningitis. Secondly, broad-range PCR is performed on selected samples for patients highly suspect of
meningitis. This corresponds to a high prevalence of the disease. We observe that negative and positive predictive values vary according to the incidence of the
disease. Thus, it is essential to select patients highly suspect of infection (whatever the type of infection) to adequately use broad-range PCR in a targeted population.
This allows improving the pre-test diagnostic probability and consequently the positive predictive value of broad-range PCR.
Illustration de lintrt de dterminer les patients pour lesquels un test diagnostique a un intrt (augmenter la probabilit pr-test). Dnitions des critres dun
test biologique : sensibilit, spcicit, valeur prdictive positive (VPP) et valeur prdictive ngative (VPN). On prend ici lexemple dune publication rapportant
une sensibilit de 93 % et une spcicit de 98 % pour une PCR 16S applique dans une population o la prvalence de la mningite tait de 44 % [17]. On applique
la PCR 16S dans une premire situation o la PCR 16S est effectue sur des prlvements de liquides cphalorachidiens (LCR) tout-venants. On se place donc dans
une situation o la prvalence des infections mninges est faible. On applique ensuite le test dans une seconde situation o seuls les prlvements de LCR suspects
de mningite sont soumis une PCR 16S. On se place alors dans une situation o la prvalence des mningites est leve. On observe que les valeurs prdictives
positives et ngatives du test varient selon que le test est appliqu une population faible (situation 1) ou haute incidence (situation 2) de la maladie. Il est donc
indispensable de slectionner les patients suspects dinfection (cela quel que soit le cadre nosologique) an dutiliser la PCR 16S bon escient dans une population
cible. Cela permet damliorer la probabilit diagnostique pr-test et par consquent la valeur prdictive positive de la PCR 16S.

the same sample [18]. Nevertheless, such a strategy, long and

costly, is difficult to apply in the routine activity of a laboratory.
4.2.2. Cardiovascular infections
The example of endocarditis with negative blood cultures
illustrates the contribution of 16S PCR for the diagnostic and
therapeutic management of patients. Indeed, the positivity of
blood cultures is part of Dukes criteria for the diagnosis of infectious endocarditis, but in 2.5 to 31% of cases, blood cultures
remain negative [19]. Furthermore, Greub et al. reported that
the culture of cardiac valves has a weak sensitivity (13%) and is
not more contributive to the diagnosis than 16S PCR [20], usual
culture of cardiac valves also being associated to numerous false
positives (soiled cultures). For more than 10 years, the contribution of 16S PCR on cardiac valves of patients undergoing
surgery for infectious endocarditis was largely described in the
literature [2025]. For example, Greub et al. reported that PCR
had allowed obtaining an etiological diagnosis for 23% of endocarditis with negative blood cultures [20]. The authors of these

studies showed that 16S PCR was especially contributive when:

(i) the patients had received previous antibiotic therapy, and (ii)
when infectious endocarditis was due to a fastidious bacterium or
to streptococci [19]. For example, Podglajen et al. had obtained
four diagnoses of endocarditis due to Bartonella sp. with PCR,
out of six cases of endocarditis with negative blood cultures [23].
This also allows improving the post-surgical therapeutic management of patients [2224]. It was also suggested to integrate
the molecular approach to Dukes criteria for the diagnosis of
infectious endocarditis, but this has not been taken into account
yet. In any case, 16S PCR is no longer an isolated diagnostic tool
and is currently part of a multimodal diagnostic strategy (serological, molecular, and histopathological) for endocarditis with
negative blood cultures [19]. Fournier et al. used this strategy
to identify a bacterium in 62.7% of cases of endocarditis with
negative blood cultures, with 57.3% of C. burnetii, 19.2% of Bartonella sp., 4% of Tropheryma whipplei, 0.4% of Legionella sp.,
0.4% of mycobacteria, 0.2% of Mycoplasma hominis, 0.2% of
Gemella morbillorum, and 0.2% of Abiotrophia defectiva [19].

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

But 16S PCR on cardiac valves may be associated to false

positives. The contaminations of PCR have already been mentioned (in 13 to 20% of cases [20]), and it was also demonstrated
that, for infectious endocarditis, bacterial DNA could persist
long after antibiotic treatment, especially for Bartonella spp.
and Streptococcus spp., making difficult the interpretation of
molecular results in case of a new episode of infectious endocarditis [26]. Conversely, false negatives may also occur in case
of previous antibiotic therapy, when inhibitors are present in
the sample or when a bad aliquot of the valve is selected for
the PCR (when the selected aliquot does not correspond to the
infectious micro-focus) [20,23,25].
But 16S PCR performed on whole blood in non-operated
patients and thus without any cardiac tissue for analysis is
not contributive for endocarditis with negative blood cultures,
because its sensitivity is too weak for the etiological diagnosis
of infectious endocarditis [19].
Finally, for endocarditis with positive blood cultures, 16S
PCR on cardiac valves allows confirming the diagnosis. There
may nevertheless be discordance between blood cultures and
16S PCR when the delay between the initiation of antibiotic therapy and surgery is too important; thus for 30 patients
with endocarditis with positive blood cultures, Podglajen et al.
obtained four cases of negative PCR with an average duration
of antibiotic therapy of 34.5 days whereas for the 26 cases with
a positive PCR, the average duration of antibiotic therapy was
24.6 days [23].
4.2.3. Neuromeningeal infections
The diagnosis of neuromeningeal infections, and more
especially of meningitis, is particularly important because management of these severe infections is an emergency. Broad-range
16S PCR has proved contributive in this context, in specific
situations: when patients have been given a previous antibiotic
therapy, and/or when the microscopic examination is negative,
and/or when usual culture remains negative [2729]. For other
authors, PCR allows ruling out the diagnosis of meningitis when
molecular technique is used on routine samples [30]. But for
the specific case of postoperative meningitis, PCR does not allow
performing any complementary microbiological diagnosis compared to culture and thus would not change the management of
aseptic postoperative meningitis [31].
The use of 16S PCR for the diagnosis of community-acquired
or postoperative meningitis is not clearly defined and according
to authors, there are important differences in sensitivity for 16S
PCR compared to culture; the great variability of protocols used
and of extraction techniques may account for these differences
[29]. Furthermore, the prevalence of the disease varies according
to study results, which implies that the test characteristics are
not comparable (Fig. 3).
16S PCR used for molecular detection on cerebro-spinal fluid
samples has limitations common to other uses of 16S PCR; furthermore, this is a biological sample containing PCR inhibitors
(proteins or other agents), making it a limiting element [29].
Finally, 16S PCR used on samples from cerebral abscesses
has proved contributive when amplification is followed by
cloning of PCR products; in this case, the technique allows

327

identifying more bacteria than culture [18]. But such an approach

is difficult to apply in the daily activity of a laboratory.
4.2.4. Bone and joint infections
The diagnosis of bone and joint infections (osteitis,
osteomyelitis, arthritis, or infections on material) is usually
made by bacterial culture. Nevertheless, culture results may
also be falsely positive (contamination of samples by cutaneous flora) but also falsely negative (because of a previous
antibiotic therapy or an infection due to fastidious bacteria)
[32]. Thus, Fenollar et al. analyzed 525 bone and joint samples and found 89 samples positive with PCR and culture,
nine samples positive with culture and negative with PCR, and
16 samples negative with culture and positive with PCR [32].
The falsely negative results with PCR could have been due a
bad quality of extraction while falsely negative results with
culture were associated, in this study, to previous antibiotic
therapy in seven out of 16 cases, to fastidious bacterium infections in two out of 16 cases, whereas in the seven remaining cases
16S PCR detected most frequently streptococci and enterococci
[32]. Broad-range 16S PCR could improve the diagnostic management but the results of studies are still too preliminary and
contradictory for 16S PCR to be used in the routine diagnosis of
bone and joint infections. The molecular tool allows detecting
new pathogens and allows improving the diagnosis of bone and
joint infections due to Kingella kingae, S. pneumoniae, Streptococcus agalactiae, Enterococcus faecalis, Mycoplasma spp. and
anaerobic bacteria [32,33]. Likewise, in case of spondylodiscitis,
16S PCR allowed making a diagnosis on 44% of the samples
that were negative in culture, the negativity being most often
related to fastidious bacteria [34]. Finally, 16S PCR associated
with cloning allowed improving the diagnosis of polymicrobial bone and joint infections [32]. Nevertheless, the diagnostic
value of 16S PCR for bone and joint infections varies according
to authors; this may be due to the variation of population panels
studied (Fig. 3) [32,35,36].
Furthermore, the absence of associated antibiogram may be a
limitation for therapeutic management. This limitation is especially a problem when dealing with bone and joint infections
on material, in which the causative germs have greatly variable
antibiotic susceptibility profiles, and for which microbiological
documentation is recommended. Even if the use of 16S PCR
remains to be defined for bone and joint infections, it seems
that it allows clarifying unreliable culture results but that it
should be kept for specific cases, such as infections with negative
culture but with clinical and biological arguments suggesting the
infection [32].
4.2.5. Other nosological settings
There are many studies reporting the use of 16S PCR in
various contexts, nevertheless, it has proved truly contributive
only in some indications (salpingitis, postoperative and posttraumatic uveitis, pericarditis), whereas for other indications it
should not be used (pacemaker infections or chronic pneumonia
in patients presenting with cystic fibrosis [37]).
Thus, for the salpingitis, 16S PCR allows identifying new
bacteria difficult to grow in culture [38]. The authors of a

328

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

Fig. 4. Strategic contribution of mass spectrometry compared to molecular biology (including broad-range PCR) in clinical bacteriology.
Place stratgique de la spectromtrie de masse par rapport la biologie molculaire (dont la PCR 16S) en bactriologie clinique.

study on uveitis analyzed 1520 samples according to a global

molecular strategy using 16S PCR and other molecular tools;
they concluded that 16S PCR lacked sensitivity and was not
contributive for the detection of intracellular bacteria, but was
recommended to improve the etiological diagnosis of postoperative and post-traumatic uveitis [39]. The authors of a
study on the molecular and non-molecular analysis of pericardial fluids, in a context of pericarditis, demonstrated that
16S PCR was useful in case of previous antibiotic therapy
[40].
But, in case of ascitic fluid infections in cirrhotic patients,
bacterial DNA may be detected in patients not meeting criteria
of ascitic fluid infections and for whom the diagnosis could be
episodes of bacterial translocation [41]. For some authors, there
could be a continuum between bacterial colonization and the
spontaneous primary peritonitis; the presence of bacterial DNA
in ascitic fluid could be associated to an increased risk of developing an infection of ascitic fluid [42]. In this case, 16S PCR
could allow defining a category of patients at risk in whom
the bacterial population would have been identified, thus allowing shortening the delay before therapeutic management [42];
nevertheless, the contribution of such a strategy remains to be
demonstrated.
16S PCR performed on blood, serum, or plasma samples
theoretically allows detecting bacteria that may be cultivated or
not. The technique performed on EDTA-treated whole blood
could allow targeting a broader bacterial spectrum, while serum
or plasma samples contain less inhibitors [43]. It was reported
as contributive in case of neonatal sepsis and for the detection of
non-cultivable bacteria such as Mycoplasma spp., Ureaplasma
spp., or Treponema pallidum [43].
16S PCR performed on pacemakers of symptomatic patients
shows that the presence of bacterial DNA is not a proof of

pacemaker infection [44]; 16S PCR should not be used in this

case.
4.3. 16S PCR and evidence of new diseases
Molecular detection in atypical cases without any prior
microbiological diagnostic orientation remains the prerogative
of broad-range 16S PCR. Thus, 16S PCR allowed linking Bartonella henselae or T. whipplei to bacillary angiomatosis and
Whipples disease, respectively [2,45]. The association between
the detection of a bacterial sequence and an infectious disease
was in this case guided by histopathological data. When this
is not the case, the causality between the detection of bacterial
DNA in a human sample and an infectious disease may be more
difficult to establish [2].
5. Is 16S PCR inferior to mass spectrometry?
For a few years, mass spectrometry, a new tool for
broad-range bacterial identification, has been available for
microbiologists (Fig. 4). It allows to identify rapidly and at a
low cost most strains cultivated in a medical bacteriology laboratory (Fig. 4) [46]. It has reduced the use of molecular biology,
but the latter remains the gold standard for rare bacteria or bacteria badly identified by mass spectrometry (check below), as
reported in a recent study, in which less than half of strains
identified by 16S PCR were identified by mass spectrometry
[47]. There are indeed some limitations to identification, for
example for streptococci of the viridans group, pneumococci,
anaerobic bacteria, but also for bacteria of the HACEK group,
Shigella spp., and some strictly aerobic bacteria [48]. It should
be noted that for some limitations of identification such as those
encountered with streptococci, pneumococci, or Shigella spp.,

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

using molecular biology is mandatory but it is better to choose

more adapted molecular targets than 16S PCR to discriminate
among these species. Nevertheless, as for 16S PCR (cf. addendum), the performances of mass spectrometry should improve
with the increased number and quality of databases [48].
It was also suggested to apply mass spectrometry directly
to samples (Fig. 4). This strategy was contributive for blood
culture vials detected as positive by an incubator automate: this
allowed reducing the delay for the diagnosis of mono-microbial
bacteremia, but without providing any information as to antibiotic susceptibility [48]. But this strategy was not contributive
for urine samples, for which using chromogenic agar for culture
allowed identifying the main urinary pathogenic bacteria at low
cost [48]. The main limitation of bacterial identification directly
on samples is the initial bacterial load of the sample and the
quantitative threshold of detection of current protocols [48].
Even if future technical breakthroughs allow detecting bacteria
directly on samples by mass spectrometry, this mass strategy would not be used on the same level as 16S PCR which
should be used only in well-defined settings, after a previous antibiotic therapy or if a fastidious bacterium is suspected
(Fig. 4).

6. What is the future for 16S PCR?

16S PCR is no longer an isolated diagnostic tool, but may
be integrated in a more global diagnostic strategy in which
the biologist-clinician dialogue is essential. Thus, in case of
endocarditis with negative blood cultures, a strategy combining
serological, histological, and molecular approaches has demonstrated its effectiveness [19]. It is necessary to fully understand
the patients clinical presentation to optimize the cost-benefit
ratio of 16S PCR. This presentation, as well as knowing about
a previous antibiotic therapy, will determine the contribution of
16S PCR, which should be integrated in a diagnostic armamentarium.
Furthermore, genomic data is the basis for determining new nuclear targets conserved between the species, to
design new probes and primers which would allow broadrange bacterial detection by real time PCR and no longer
by standard PCR [49]. This could allow gaining time and
sensitivity (independently of the secondary stage of bacterial
identification) for the initial nucleic detection stage on samples.
In conclusion, even if 16S PCR has limitations that should
restrain its use to well-defined cases, it is currently an essential
technique for bacteriological diagnosis of an infection.
More on how 16S PCR works.
Addendum is available upon request to the corresponding
author.

Disclosure of interest
The authors declare that they have no conflicts of interest
concerning this article.

329

References
[1] Clarridge III JE. Impact of 16S rRNA gene sequence analysis for identification of bacteria on clinical microbiology and infectious diseases. Clin
Microbiol Rev 2004;17:84062.
[2] Petti CA. Detection and identification of microorganisms by gene amplification and sequencing. Clin Infect Dis 2007;44:110814.
[3] Woese CR. Bacterial evolution. Microbiol Rev 1987;51:22171.
[4] Janda JM, Abbott SL. 16S rRNA gene sequencing for bacterial identification in the diagnostic laboratory: pluses, perils, and pitfalls. J Clin Microbiol
2007;45:27614.
[5] Kiratisin P, Li L, Murray PR, Fischer SH. Identification of bacteria recovered from clinical specimens by 16S rRNA gene sequencing. Eur J Clin
Microbiol Infect Dis 2003;22:62831.
[6] Schlaberg R, Simmon KE, Fisher MA. A systematic approach for discovering novel, clinically relevant bacteria. Emerg Infect Dis 2012;18:42230.
[7] Woo PCY, Ng KHL, Lau SKP, Yip KT, Fung AMY, Leung KW, et al.
Usefullness of the MicroSeq 500 16S ribosomal DNA-based bacterial
identification system for identification of clinically significant bacterial isolates with ambiguous biochemical profiles. J Clin Microbiol
2003;41:19962001.
[8] Sontakke S, Cadenas MB, Maggi RG, Diniz PP, Breitschwerdt EB. Use of
broad range16S rDNA PCR in clinical microbiology. J Microbiol Methods
2009;76:21725.
[9] Petti CA, Polage CR, Schreckenberger P. The role of 16S rRNA gene
sequencing in identification of microorganisms misidentified by conventional methods. J Clin Microbiol 2005;43:61235.
[10] Mignard S, Flandrois JP. 16S rRNA sequencing in routine bacterial identification: a 30-month experiment. J Microbiol Methods 2006;67:57481.
[11] Bosshard PP, Abels S, Zbinden R, Bottger EC, Altwegg M. Ribosomal DNA
sequencing for identification of aerobic gram-positive rods in the clinical
laboratory (an 18-month evaluation). J Clin Microbiol 2003;41:413440.
[12] Petti CA, Bosshard PP, Brandt ME, Clarridge III JE, Feldblyum TV, Foxall
P, et al. Interpretive criteria for identification of bacteria and fungi by DNA
target sequencing; approved guideline. CLSI; 2008.
[13] Gevers D, Cohan FM, Lawrence JG, Spratt BG, Coenye T, Feil EJ, et al.
Re-evaluating prokaryotic species. Nat Rev Microbiol 2005;3:7339.
[14] Stackebrandt E, Frederiksen W, Garrity GM, Grimont PAD, Kampfer P,
Maiden MCJ, et al. Report of the ad hoc committee for the re-evaluation
of the species definition in bacteriology. Int J Syst Evol Microbiol
2002;52:10437.
[15] Drancourt M, Berger P, Raoult D. Systematic 16S rRNA gene sequencing
of atypical clinical isolates identified 27 new bacterial species associated
with humans. J Clin Microbiol 2004;42:2197202.
[16] Rampini SK, Bloemberg GV, Keller PM, Buchler AC, Dollenmaier G,
Speck RF, et al. Broad-range 16S rRNA gene polymerase chain reaction for diagnosis of culture-negative bacterial infections. Clin Infect Dis
2011;53:124551.
[17] Rafi W, Chandramuki A, Mani R, Satishchandra P, Shankar SK. Rapid
diagnosis of acute bacterial meningitis: role of a broad range 16S rRNA
polymerase chain reaction. J Emerg Med 2010;38:22530.
[18] Al Masalma M, Armougom F, Scheld WM, Dufour H, Roche PH, Drancourt M, et al. The expansion of the microbiological spectrum of brain
abscesses with use of multiple 16S ribosomal DNA sequencing. Clin Infect
Dis 2009;48:116978.
[19] Fournier PE, Thuny F, Richet H, Lepidi H, Casalta JP, Arzouni JP, et al.
Comprehensive diagnostic strategy for blood culture-negative endocarditis:
a prospective study of 819 new cases. Clin Infect Dis 2010;51:13140.
[20] Greub G, Lepidi H, Rovery C, Casalta JP, Habib G, Collard F, et al. Diagnosis of infectious endocarditis in patients undergoing valve surgery. Am
J Med 2005;118:2308.
[21] Goldenberger D, Kunzli A, Vogt P, Zbinden R, Altwegg M. Molecular
diagnosis of bacterial endocarditis by broad-range PCR amplification and
direct sequencing. J Clin Microbiol 1997;35:27339.
[22] Gauduchon V, Chalabreysse L, Etienne J, Celard M, Benito Y, Lepidi H,
et al. Molecular diagnosis of infective endocarditis by PCR amplification and direct sequencing of DNA from valve tissue. J Clin Microbiol
2003;41:7636.

330

A. Renvois et al. / Mdecine et maladies infectieuses 43 (2013) 322330

[23] Podglajen I, Bellery F, Poyart C, Coudol P, Buu-Hoi A, Bruneval P, et al.

Comparative molecular and microbiologic diagnosis of bacterial endocarditis. Emerg Infect Dis 2003;9:15437.
[24] Lang S, Watkin RW, Lambert PA, Bonser RS, Littler WA, Elliott TS.
Evaluation of PCR in the molecular diagnosis of endocarditis. J Infect
2004;48:26975.
[25] Vondracek M, Sartipy U, Aufwerber E, Julander I, Lindblom D, Westling K. 16S rDNA sequencing of valve tissue improves microbiological
diagnosis in surgically treated patients with infective endocarditis. J Infect
2011;62:4728.
[26] Rovery C, Greub G, Lepidi H, Casalta JP, Habib G, Collart F, et al. PCR
detection of bacteria on cardiac valves of patients with treated bacterial
endocarditis. J Clin Microbiol 2005;43:1637.
[27] Schuurman T, de Boer RF, Kooistra-Smid AM, van Zwet AA. Prospective
study of use of PCR amplification and sequencing of 16S ribosomal DNA
from cerebrospinal fluid for diagnosis of bacterial meningitis in a clinical
setting. J Clin Microbiol 2004;42:73440.
[28] Welinder-Olsson C, Dotevall L, Hogevik H, Jungnelius R, Trollfors B, Wahl
M, et al. Comparison of broad-range bacterial PCR and culture of cerebrospinal fluid for diagnosis of community-acquired bacterial meningitis.
Clin Microbiol Infect 2007;13:87986.
[29] Srinivasan L, Pisapia JM, Shah SS, Halpern CH, Harris MC. Can Broadrange 16S ribosomal ribonucleic acid gene polymerase chain reactions
improve the diagnosis of bacterial meningitis? A systematic review and
meta-analysis. Ann Emerg Med 2012;60:60920.
[30] Saravolatz LD, Manzor O, VanderVelde N, Pawlak J, Belian B. Broadrange bacterial polymerase chain reaction for early detection of bacterial
meningitis. Clin Infect Dis 2003;36:405.
[31] Zarrouk V, Leflon-Guibout V, Robineaux S, Kalamarides M, NicolasChanoine MH, Sterkers O, et al. Broad-range 16S rRNA PCR with
cerebrospinal fluid may be unreliable for management of postoperative
aseptic meningitis. J Clin Microbiol 2010;48:33313.
[32] Fenollar F, Roux V, Stein A, Drancourt M, Raoult D. Analysis of 525
samples to determine the usefulness of PCR amplification and sequencing
of the 16S rRNA gene for diagnosis of bone and joint infections. J Clin
Microbiol 2006;44:101828.
[33] Fenollar F, Levy PY, Raoult D. Usefulness of broad-range PCR for the diagnosis of osteoarticular infections. Curr Opin Rheumatol 2008;20:46370.
[34] Fuursted K, Arpi M, Lindblad BE, Pedersen LN. Broad-range PCR as a
supplement to culture for detection of bacterial pathogens in patients with
a clinically diagnosed spinal infection. Scand J Infect Dis 2008;40:7727.
[35] Marin M, Garcia-Lechuz JM, Alonso P, Villanueva M, Alcala L, Gimeno
M, et al. Role of universal 16S rRNA gene PCR and sequencing in diagnosis
of prosthetic joint infection. J Clin Microbiol 2012;50:5839.
[36] Ince A, Rupp J, Frommelt L, Katzer A, Gille J, Lohr JF. Is aseptic loosening of the prosthetic cup after total hip replacement due to nonculturable

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

bacterial pathogens in patients with low-grade infection? Clin Infect Dis

2004;39:1599603.
Rudkjobing VB, Thomsen TR, Alhede M, Kragh KN, Nielsen PH,
Johansen UR, et al. True microbiota involved in chronic lung infection
of cystic fibrosis patients found by culturing and 16S rRNA gene analysis.
J Clin Microbiol 2011;49:43525.
Hebb JK, Cohen CR, Astete SG, Bukusi EA, Totten PA. Detection of novel
organisms associated with salpingitis, by use of 16S rDNA polymerase
chain reaction. J Infect Dis 2004;190:210920.
Drancourt M, Berger P, Terrada C, Bodaghi B, Conrath J, Raoult D, et al.
High prevalence of fastidious bacteria in 1520 cases of uveitis of unknown
etiology. Medicine (Baltimore) 2008;87:16776.
Levy PY, Fournier PE, Charrel R, Metras D, Habib G, Raoult D. Molecular analysis of pericardial fluid: a 7-year experience. Eur Heart J
2006;27:19426.
Such J, Frances R, Munoz C, Zapater P, Casellas JA, Cifuentes A, et al.
Detection and identification of bacterial DNA in patients with cirrhosis and
culture-negative, nonneutrocytic ascites. Hepatology 2002;36:13541.
Rogers GB, Russell LE, Preston PG, Marsh P, Collins JE, Saunders J,
et al. Characterisation of bacteria in ascitesreporting the potential of
culture-independent, molecular analysis. Eur J Clin Microbiol Infect Dis
2010;29:53341.
Mancini N, Carletti S, Ghidoli N, Cichero P, Burioni R, Clementi M. The era
of molecular and other non-culture-based methods in diagnosis of sepsis.
Clin Microbiol Rev 2010;23:23551.
Pichlmaier M, Marwitz V, Kuhn C, Niehaus M, Klein G, Bara C, et al. High
prevalence of asymptomatic bacterial colonization of rhythm management
devices. Europace 2008;10:106772.
Relman DA, Loutit JS, Schmidt TM, Falkow S, Tompkins LS. The agent
of bacillary angiomatosis. An approach to the identification of uncultured
pathogens. N Engl J Med 1990;323:157380.
Seng P, Drancourt M, Gouriet F, La SB, Fournier PE, Rolain JM, et al.
Ongoing revolution in bacteriology: routine identification of bacteria by
matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Clin Infect Dis 2009;49:54351.
Bizzini A, Jaton K, Romo D, Bille J, Prodhom G, Greub G. Matrix-assisted
laser desorption ionization-time of flight mass spectrometry as an alternative to 16S rRNA gene sequencing for identification of difficult-to-identify
bacterial strains. J Clin Microbiol 2011;49:6936.
Croxatto A, Prodhom G, Greub G. Applications of MALDI-TOF mass
spectrometry in clinical diagnostic microbiology. FEMS Microbiol Rev
2012;36:380407.
Clifford RJ, Milillo M, Prestwood J, Quintero R, Zurawski DV, Kwak
YI, et al. Detection of bacterial 16S rRNA and identification of four
clinically important bacteria by real-time PCR. PLoS One 2012;7:
e48558.