Академический Документы
Профессиональный Документы
Культура Документы
Summary
Lancet Infect Dis 2013;
13: 42635
Published Online
February 1, 2013
http://dx.doi.org/10.1016/
S1473-3099(12)70323-7
See Comment page 382
*Contributed equally
Department of Medical
Statistics, Computer Sciences
and Documentation
(C Wacker Cand Med,
A Prkno Cand Med,
Prof P Schlattmann PhD), and
Department of
Anaesthesiology and Intensive
Care Medicine
(Prof F M Brunkhorst MD),
Centre for Sepsis Control and
Care, Jena University Hospital,
Jena, Germany
Correspondence to:
Prof Peter Schlattmann,
Department of Medical Statistics,
Computer Sciences and
Documentation, Centre for
Sepsis Control and Care, Jena
University Hospital,
Bachstrae 18, 07743 Jena,
Germany
peter.schlattmann@mti.unijena.de
Background Procalcitonin is a promising marker for identication of bacterial infections. We assessed the accuracy
and clinical value of procalcitonin for diagnosis of sepsis in critically ill patients.
Methods We searched Medline, Embase, ISI Web of Knowledge, the Cochrane Library, Scopus, BioMed Central, and
Science Direct, from inception to Feb 21, 2012, and reference lists of identied primary studies. We included articles
written in English, German, or French that investigated procalcitonin for dierentiation of septic patientsthose
with sepsis, severe sepsis, or septic shockfrom those with a systemic inammatory response syndrome of noninfectious origin. Studies of healthy people, patients without probable infection, and children younger than 28 days
were excluded. Two independent investigators extracted patient and study characteristics; discrepancies were resolved
by consensus. We calculated individual and pooled sensitivities and specicities. We used I to test heterogeneity and
investigated the source of heterogeneity by metaregression.
Findings Our search returned 3487 reports, of which 30 fullled the inclusion criteria, accounting for 3244 patients.
Bivariate analysis yielded a mean sensitivity of 077 (95% CI 072081) and specicity of 079 (95% CI 074084).
The area under the receiver operating characteristic curve was 085 (95% CI 081088). The studies had
substantial heterogeneity (I=96%, 95% CI 9499). None of the subgroups investigatedpopulation, admission
category, assay used, severity of disease, and description and masking of the reference standardcould account
for the heterogeneity.
Interpretation Procalcitonin is a helpful biomarker for early diagnosis of sepsis in critically ill patients. Nevertheless,
the results of the test must be interpreted carefully in the context of medical history, physical examination, and
microbiological assessment.
Funding Ministry of Education and Research, the Deutsche Forschungsgemeinschaft, Thuringian Ministry for
Education, Science and Culture, the Thuringian Foundation for Technology, Innovation and Research, and the
German Sepsis Society.
Introduction
Worldwide, sepsis and its sequelae are still a common
cause of acute illness and death in patients with
community-acquired and nosocomial infections.1,2 The
American College of Chest Physicians and the Society
of Critical Care Medicine Consensus Conference
(Northbrook, IL, USA; August, 1991) dened sepsis as
systemic inammatory response caused by infection.3
However, no gold standard exists for proof of infection.
Bacteraemia is identied in only about 30% of patients
with sepsis, depending on previous antibiotic treatment.4,5 Furthermore, early clinical signs of sepsis, such
as fever, tachycardia, and leucocytosis, are non-specic
and overlap with signs of systemic inammatory response syndromes of non-infectious origin, especially
in patients who have undergone surgery. Other signs,
such as arterial hypotension, thrombocytopenia, or
increased lactate concentrations suggest, too late for
life-saving treatment, progression to organ dysfunction.
Thus, delay in diagnosis and treatment of sepsis increases mortality, prolongs length of hospital stay, and
increases costs,6,7 highlighting the need for early and
reliable diagnostic biomarkers for sepsis.
426
Articles
Methods
Search strategy and selection criteria
We systematically searched Medline (via PubMed),
Embase (via OvidSP), ISI Web of Knowledge, the
Cochrane Library, Scopus, BioMed Central, and Science
Direct for studies that assessed the accuracy of procalcitonin for the diagnosis of sepsis.
Our medical subject heading terms (for Medline),
EMTREE terms (for Embase), and text (for others) were
(procalcitonin OR PCT) AND (sepsis OR bacterial
infection OR systemic inammatory response syndrome OR SIRS). To reduce the number of results, for
searches in Science Direct, Embase, and Scopus, we also
used the search terms NOT (review OR letter OR
editorial OR animal experiment OR meeting abstract
OR proceeding paper OR poster presentation OR
meta-analysis OR case report). We searched the
databases between inception and Feb 21, 2012. We also
searched the reference list of each primary study
identied and of previous systematic reviews.
Studies were included if they assessed the accuracy of
procalcitonin for dierentiation between critically ill
patients with sepsis from those who have a systemic
inammatory response syndrome without infection.
To be eligible, studies had to have a well dened
reference standard for sepsis, which included the use of
denitions established by the American College of Chest
Physicians and Society of Critical Care Medicine
Consensus Conference3 or the German Sepsis Society.12
In accordance with these denitions, the presence of
infection had to be microbiologically conrmed or at
least clinically suspected because of one or more
characteristics: white blood cells in a normally sterile
body uid, perforated viscus, radiographic evidence of
pneumonia in association with production of purulent
sputum, and syndrome associated with a high risk of
infection (eg, ascending cholangitis).
Furthermore, the studies had to provide sucient
information to construct the 22 contingency tableie,
false and true positives and negatives were provided.
We only included publications written in English,
German, or French. Animal experiments, reviews,
correspondences, case reports, expert opinions, and
editorials were excluded. We also excluded all studies
that involved healthy people, patients without probable
infection, and children younger than 28 days.
Statistical analysis
We tabulated true positives, false negatives, false positives, and true negatives in patients with sepsis and
systemic inammatory response syndrome, stratied by
study. We used the numbers to calculate sensitivity and
specicity and a corresponding CI.
To synthesise data, we used an exact binomial
rendition16 of the bivariate mixed-eects regression
Procedures
Two investigators (CW, AP) independently extracted data,
including the quality assessment from the retrieved
studies. Discrepancies were resolved in a consensus
meeting or, if agreement could not be reached, they were
resolved by referral to a third investigator (FMB).
www.thelancet.com/infection Vol 13 May 2013
427
Articles
model developed by van Houwelingen17,18 for metaanalysis of treatment trials, modied for synthesis of
diagnostic test data.19,20 This model does not transform
pairs of sensitivity and specicity of individual studies
into a single indicator of diagnostic accuracy, but
preserves the two-dimensional nature of the data taking
into account any correlation between the two.
Based on this model, we estimated mean logit sensitivity
and specicity with their standard error and 95% CIs, the
between-study variability in logit sensitivity and specicity,
and the covariance between them. We back-transformed
these quantities to the original receiver operating curve
scale to obtain summary sensitivity, specicity, and
diagnostic odds ratios. We then used the derived logit
estimates of sensitivity, specicity, and respective variances
to construct a hierarchical summary receiver operating
Year
Population Admission
category
Prevalence Severity
(%)
TP
FP
TN
FN
Sensitivity
(95% CI)
Specicity
(95% CI)
Ahmadinejad24 2009
Adult
Medical and ED
surgical
PCT-Q
05
120
59%
No information
63
11
38
089
(079095)
078
(063088)
Al-Nawas25
1996
Adult
Medical
PCT-LIA
05
337
36%
73
45
170
49
060
(051069)
079
(073084)
Arkader26
2006
Paediatric
PCT-LIA
28
50%
No information
12
14
086
(057098)
100
(077100)
Bell27
2003
Adult
PCT-LIA
1575
83
75%
No information
47
19
15
076 (063
086)
090
(070099)
Castelli28
2004
Adult
PCT-LIA
12
49
69%
21
13
13
062
(044078)
087
(060098)
Clech29
2006
Adult
Medical
ICU
PCT-Kryptor
76
47%
Septic shock
29
38
081
(064092)
095
(083099)
Clech29
2006
Adult
Surgical
ICU
PCT-Kryptor
97
67
46%
Septic shock
28
27
090
(074098)
075
(058088)
Dorizzi30
2006
Adult
PCT-LIA
83
61%
42
26
082
(069092)
081
(064093)
Du31
2003
Adult
PCT-LIA
16
51
39%
16
23
080
(056094)
074
(055088)
Gaini32
2006
Adult
Medical
HW
PCT-Kryptor
93
80%
56
10
18
076
(064085)
053
(029076)
Gibot33
2004
Adult
Medical
ICU
PCT-LIA
06
76
62%
39
20
083
(069092)
069
(049085)
Groselj-Grenc34 2009
Paediatric
PCT-LIA
028
36
67%
20
083
(063095)
075
(043095)
Harbarth35
2001
Adult
PCT-LIA
11
78
77%
58
14
097
(088100)
078
(052094)
Hsu36
2011
Adult
Medical
ICU
PCT-Kryptor
22
66
83%
31
11
24
056
(042070)
100
(072100)
Ivancevic37
2008
Adult
Surgical
PCT-LIA
11
63
65%
No information
34
17
083
(068093)
077
(055092)
Jimeno38
2004
Adult
Medical
PCT-LIA
05
104
39%
No information
17
58
24
041
(026058)
092
(082097)
Kofoed39
2007
Adult
Medical
HW and PCT-Kryptor
ED
025
151
64%
No information
77
23
32
19
080
(071088)
058
(044071)
Latour-Perez40
2010
Adult
PCT-Q
05
114
63%
53
37
19
074
(062083)
088
(074096)
Meynaar41
2011
Adult
PCT-Kryptor
76
42%
31
35
097
(084100)
080
(065090)
428
Articles
Year
Population Admission
category
Prevalence Severity
(%)
TP
FP
TN
FN
Sensitivity
(95% CI)
Specicity
(95% CI)
2006
Adult
Medical and
surgical
PCT-Q
05
Oshita43
2010
Adult
PCT-Q
05
Pavcnik-Arnol44 2007
Paediatric
PCT-Kryptor
579
49
Ruiz-Alvarez45
2009
Adult
PCT-Kryptor
032
Sakr46
2008
Adult
Surgical
ICU
PCT-LIA
Selberg47
2000
Adult
Medical
ICU
PCT-LIA
Simon48
2008
Paediatric
Suprin49
2000
Adult
Medical
Tsalik50
2011
Adult
Tsangaris51
2009
Tugrul52
Wanner53
22
24
088
(069097)
096
(080100)
No information
76
11
45
36
068
(058076)
080
(068090)
61%
17
17
13
057
(037075)
089
(067099)
103
76%
65
16
13
083
(073091)
064
(043082)
327
36%
82
92
116
37
069
(060077)
056
(049063)
33
33
67%
19
086
(065097)
055
(023083)
PCT-LIA
25
64
39%
No information
17
10
29
068
(046085)
074
(058087)
ICU
PCT-LIA
95
79%
49
14
26
065
(053076)
070
(046088)
ED
PCT-Kryptor
01
336
74%
168
33
56
79
068
(062074)
063
(052073)
Adult
PCT-Kryptor
50
54%
19
21
070
(050086)
091
(072099)
2002
Adult
PCT-LIA
131
85
88%
55
20
073
(062083)
080
(044097)
2000
Adult
Surgical
PCT-LIA
15
133
34%
No information
34
20
68
11
076
(060087)
077
(067086)
ED and
ICU
50
50%
168 67%
All assays made by BRAHMSGmbH (Hennigsdorf, Germany). TP=true positive. FP=false positive. TN=true negative. FN=false negative. ED=emergency department. ICU=intensive care unit. PICU=paediatric
intensive care unit. HW=hospital ward.
Results
Our database search retrieved 3487 articles. After
reviewing the titles and abstracts, we excluded 3321.
After a full text review we excluded a further 136, leaving
30 studies for inclusion (gure 1). Because in one study
investigators reported diagnostic accuracy separately for
medical and surgical patients, the study was divided
into two parts, thus we analysed 31 datasets. Search of
the reference lists of the identied articles and previous
systematic reviews10,11,23 did not identify any more
relevant articles.
www.thelancet.com/infection Vol 13 May 2013
429
Articles
076 (060087)
077 (067086)
073 (062083)
080 (044097)
070 (050086)
091 (072099)
068 (062074)
063 (052073)
065 (053076)
070 (046088)
068 (046085)
074 (058087)
086 (065097)
055 (023083)
069 (060077)
056 (049063)
083 (073091)
064 (043082)
057 (037075)
089 (067099)
068 (058076)
080 (068090)
088 (069097)
096 (080100)
097 (084100)
080 (065090)
074 (062083)
088 (074096)
080 (071088)
058 (044071)
041 (026058)
092 (082097)
083 (068093)
077 (055092)
056 (042070)
100 (072100)
097 (088100)
078 (052094)
083 (063095)
075 (043095)
083 (069092)
069 (049085)
076 (064085)
053 (029076)
080 (056094)
074 (055088)
082 (096092)
081 (064093)
090 (074098)
075 (058088)
081 (064092)
095 (083099)
062 (044078)
087 (060098)
076 (063086)
090 (070099)
086 (057098)
100 (077100)
060 (051069)
079 (073084)
089 (079095)
078 (063088)
Combined
077 (072081)
079 (074084)
Q=13671, degrees of
Q=13521, degrees of
03
04
05
06
07
08
09
Sensitivity
10
freedom=3000, p=000
I2=7781 (70348528)
02
03
04
05
06
07
Specificity
08
09
10
freedom=3000, p=000
I2=7806 (70698542)
430
Articles
10
Sensitivity
Discussion
05
Individual studies
Summary operating point
Summary receiver operating characteristic curve
95% confidence contour
95% prediction contour
0
10
05
Specificity
Articles
998
05
07
1
995
993
99
2
3
5
7
10
20
30
40
50
60
70
80
90
93
95
97
98
997
Likelihood ratio
1000
500
200
100
50
98
97
95
93
90
20
10
5
80
2
1
05
60
02
01
005
30
002
001
0005
0002
0001
70
50
40
20
999
10
7
5
3
2
99
993
995
1
07
05
997
03
998
02
999
01
432
Articles
useful information is lost.81 Because procalcitonin concentrations rise as disease severity advances,56 patients
with a high procalcitonin concentration are more likely
to have sepsis than are patients with a low procalcitonin
concentration. To provide more precise information
about the reliability of the test, we suggest calculating
likelihood ratios based on multiple cutos.
As our results show, procalcitonin is not a perfect
marker for diagnosis of sepsis, but an ideal marker
does not exist. Sepsis is a pathophysiological process
rather than a specic syndrome and is too complex
to be described by a single measure. Nevertheless,
procalcitonin is one of the most promising parameters.
Several other mediators and molecules of the host
response to infectionC-reactive protein, soluble TREM1,
interleukin 6, interleukin 8, and soluble PLAURhave
been investigated, but with no outstanding result.23,35,40,82
In conclusion, procalcitonin is a helpful marker for
diagnosis of sepsis in critically ill patients. However, it
cannot be recommended as the single denitive test for
sepsis diagnosis, but rather it must be interpreted in
context with information from careful medical history,
physical examination, and when feasible, microbiological
assessment. Moreover, continuing re-evaluation during
the course of disease is advisable.
Contributors
CW had the idea for and designed the study, searched the scientic
literature, collected, analysed, and interpreted data, and wrote and
critically revised the report. AP searched the scientic literature,
collected data, and drafted and critically revised the report. FMB had the
idea for and designed the study, interpreted data, drafted and critically
revised the report, supervised the study, and gave administrative,
technical, and material support. PS had the idea for and designed the
study, statistically analysed and interpreted the data, drafted and critically
revised the report, supervised the study, and gave administrative,
technical, and material support.
8
9
10
11
12
13
14
15
16
17
18
19
Conicts of interest
We declare that we have no conicts of interest.
Acknowledgments
The work was supported by the German Centre for Sepsis Control & Care,
funded by the Ministry of Education and Research (grant number 01 E0
1002), the Deutsche Forschungsgemeinschaft (Schl 3-1), Thuringian
Ministry for Education, Science and Culture (ProExcellence; PE 108-2), the
Thuringian Foundation for Technology, Innovation and Research, and the
German Sepsis Society.
References
1
Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J,
Pinsky MR. Epidemiology of severe sepsis in the United States:
analysis of incidence, outcome, and associated costs of care.
Crit Care Med 2001; 29: 130310.
2
Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of
sepsis in the United States from 1979 through 2000. N Engl J Med
2003; 348: 154654.
3
Bone RC, Balk RA, Cerra FB, et al. Denitions for sepsis and organ
failure and guidelines for the use of innovative therapies in sepsis.
The ACCP/SCCM Consensus Conference Committee. American
College of Chest Physicians/Society of Critical Care Medicine. 1992.
Chest 2009; 136 (suppl): e28.
4
Bates DW, Cook EF, Goldman L, Lee TH. Predicting bacteremia in
hospitalized patients. A prospectively validated model.
Ann Intern Med 1990; 113: 495500.
5
Bates DW, Sands K, Miller E, et al. Predicting bacteremia in patients
with sepsis syndrome. Academic Medical Center Consortium
Sepsis Project Working Group. J Infect Dis 1997; 176: 153851.
20
21
22
23
24
25
26
27
433
Articles
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
434
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
Articles
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
435