Академический Документы
Профессиональный Документы
Культура Документы
PII: S0883-9441(18)30399-X
DOI: doi:10.1016/j.jcrc.2018.08.040
Reference: YJCRC 53032
To appear in: Journal of Critical Care
Please cite this article as: Yu-long Bai, Bang-Li Hu, Han-Chun Wen, Yi-Ling Zhang, Ji-Jin
Zhu , Prognostic value of plasma brain natriuretic peptide value for patients with sepsis: A
meta-analysis. Yjcrc (2018), doi:10.1016/j.jcrc.2018.08.040
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
ACCEPTED MANUSCRIPT
Prognostic value of plasma brain natriuretic peptide value for patients with sepsis: a
meta-analysis
Yu-long Bai1, Bang-Li Hu2, Han-Chun Wen3, Yi-Ling Zhang3, Ji-Jin Zhu1*
PT
Shuangyong Road 6, Nanning 530021, Guangxi, PR China;
RI
2 Department of Research, Affiliated Tumor Hospital of Guangxi Medical University, Hedi
SC
Road 71, Nanning 530021, Guangxi, PR China;
Department of Emergency,
D
Email: 2637874438@qq.com
AC
Abstract
Purpose
The aim of this meta-analysis was to clarify the diagnostic role of plasma BNP and
Methods
A systematic review was conducted prior to January 2018. Summary sensitivity, specificity,
PT
positive likelihood ratio (PLR), negative likelihood ratio (NLR) and diagnostic odds ratio
RI
(DOR) of the prognostic value of plasma BNP and NT-proBNP for septic patients. The area
SC
under the receiver operating curves (AUROC) were used to summarize overall test
performance.
NU
Results
Twenty-two studies with 3417 septic patients were selected in the analysis. The summary
MA
sensitivity, specificity, PLR, NLR, DOR and the AUROC of the overall analysis of BNP
were: 0.84, 0.73, 3.1, 0.22, 14, 0.85; and these values of NT-proBNP were: 0.71, 0.73, 2.6,
D
0.39, 7 and 0.7 respectively; Subgroup analysis and meta-regression analyses showed that the
E
tested method and observation endpoint influenced the summary sensitivity, specificity of
PT
BNP, but the tested day, tested method or observation endpoint did not influence the
CE
Conclusions
AC
This meta-analysis indicates that both elevated plasma BNP and NT-proBNP have moderate
predicts value for the mortality of septic patients, and the tested method and observation
Introduction
Sepsis and septic shock have become increasingly important clinical problems. A total of
PT
40% of patients with sepsis have myocardial dysfunction, and the mortality rate of patients
RI
with sepsis with cardiac dysfunction is as high as 70% [1]. Echocardiography can accurately
SC
detect myocardial depression, and the bedside echocardiograms have become an increasing
prognosis of patients with sepsis, but these biomarkers have limitations and their prognostic
MA
value remains poor [3-5]. Therefore, finding a reliable biomarker to predict the prognosis of
pro-BNP (NT-proBNP) is an inactive metabolite of BNP that has a markedly longer plasma
PT
half-life and is more stable than BNP, making it more applicable for clinical use [6].
CE
inexpensive, and widely available. Increased plasma levels of both BNP and proBNP have
AC
been identified as early markers of myocardial dysfunction and increased mortality in patients
admitted to the intensive care unit (ICU) [7, 8]. Elevated levels of plasma BNP and
NT-proBNP have also been shown to be markers of unfavorable prognosis in patients with
Some studies reported the prognostic value of plasma BNP and NT-proBNP for the
patients with sepsis, but the results substantially varied generally because of difference in the
ACCEPTED MANUSCRIPT
sample size and testing methods for the markers. A meta-analysis uses a statistical approach
to combine the results from multiple studies and provides more reliable results than
individual studies. Previously, a meta-analysis was performed to explore the prognostic value
of BNP in patients with sepsis [12]; however, only a small number of studies were included
and the potential confounders that might affect the prognostic value of BNP were not
investigated. Thus, the generalizability of the results was limited. In this study, we aimed to
PT
examine the role of plasma BNP and NT-proBNP in predicting mortality in patients with
RI
sepsis.
SC
Materials and Methods
NU
Search strategy
This meta-analysis was performed in accordance with the guidelines of the Preferred
MA
Reporting Items for Systematic Review and Meta-analyses. The electronic databases PubMed,
Cochrane Library, Web of Science, and Chinese National Knowledge Infrastructure were
D
comprehensively searched for eligible articles published before January 2018 using the
E
following search terms: “septic shock” or “Shock, Septic,” “Natriuretic Peptide, Brain” or
PT
strategy was a combination of at least three terms, for example, “septic shock” combined with
“BNP” and “prognostic.” The included articles were limited to human studies but not limited
AC
by language. Conference abstracts were not selected because they generally do not contain
sufficient data. Relevant articles were also searched using the related articles function in
PubMed. References within the identified articles were also searched manually.
The inclusion criteria were as follows: (1) the diagnosis of sepsis and septic shock was in
accordance with 2008 guidelines [13] or the updated 2013 [14] or 2016 version [15] for
sepsis and septic shock, (2) the patients were given standard treatment for sepsis; (3) the
study evaluated the prognostic value of plasma BNP or NT-proBNP for the patients with
sepsis; (4) data on the sensitivity and specificity of BNP or NT-proBNP for sepsis were
provided; (5) the patient outcome (survival and non-survival) and study endpoints were
PT
provided. Meanwhile, studies were excluded if they were review articles, letters, case reports,
RI
studies on cell lines, animal experiments, and human xenografts. If the same patient cohort
SC
was reported in several articles, we chose the most recent study.
NU
Data extraction and quality assessment
The following data were extracted: the name of the first author, year of publication, study
MA
location (country), cut-off of plasma BNP or NT-proBNP, number of patients who survived
and died, sensitivity and specificity value of BNP or NT-proBNP, study design, method of
D
evaluating plasma BNP or NT-proBNP, and endpoint of observation. The quality of the
E
studies was assessed using Quality Assessment of Diagnostic Accuracy Studies (QUADAS)
PT
[16], which is an evidence-based quality assessment tool developed for systematic reviews of
CE
studies of diagnostic accuracy. The maximum QUADAS score is 14, with a 6 score
indicating high quality. Two reviewers independently evaluated the eligibility of the studies.
AC
Statistical analysis
The following summary measures were calculated for each study: sensitivity and
specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic
odds ratio (DOR) with the corresponding 95% confidence intervals (CIs). The Cochran Q test
ACCEPTED MANUSCRIPT
and the inconsistency index (I2) were used to assess the threshold effect as an important
component of the source of variation of the studies. In general, I2 values <25% indicate mild
fixed-effects model (Mantel-Haenszel method) was used to calculate parameters when there
was low heterogeneity; otherwise, a random-effects model (DerSimonian and Laird) was
used. Subgroup analysis and meta-regression analyses were conducted to explore and explain
PT
the diversity (heterogeneity) among the results of different studies. Publication bias was
RI
assessed using Deek’s test [17]. All statistical tests were performed using STATA 11.2
SC
software (Stata Corp, College Station, TX) with 2-tailed P values. A P value <0.05 was
considered significant.
NU
Results
MA
Initially, 106 articles were retrieved. After screening the abstracts, 46 articles were
D
excluded because they were either reviews, animal studies, case reports, or irrelevant to the
E
current meta-analysis. Of the remaining 60 articles, 38 were further excluded after screening
PT
their full-text because they did not provide data on the sensitivity and specificity of BNP or
CE
NT-proBNP (n=18) or data on the survival rate of patients (n=15), and the prognostic hazard
ratio of BNP or NT-proBNP was provided instead of the sensitivity and specificity values
AC
(n=5). Finally, 22 studies [18-39] with 3417 patients with sepsis were included in our
The characteristics of the included studies are shown in Table 1. All the patients with
sepsis meet the criteria for severe sepsis or septic shock and were given standard treatment.
ACCEPTED MANUSCRIPT
Most of the studies were prospective in design and used 28-day mortality as the endpoint.
Other endpoints included hospital mortality and ICU mortality. The quality of the studies was
PT
Prognostic value of plasma BNP for patients with sepsis
RI
Fourteen studies that included 2380 patients with sepsis evaluated the prognostic value of
SC
plasma BNP. The combined data showed that plasma BNP has moderate sensitivity and
moderate specificity, with pooled sensitivity and specificity of 0.84 (95% CI: 0.77-0.89) and
NU
0.73 (95% CI: 0.65-0.79), respectively. The PLR, NLR, and DOR was 3.1 (95% CI: 2.3-4.1),
0.22 (95% CI: 0.15-0.34), and 14 (95% CI: 7-27), respectively. However, significant
MA
heterogeneity was observed among the studies regarding their combined sensitivity
(I2=96.07%, P<0.01) and specificity (I2=74.21%, P<0.01) (Fig. 2). The area under the curve
D
(AUC) of the summary receiver operating characteristic (SROC) curve was 0.85 (0.81-0.88),
E
suggesting a moderate diagnostic accuracy (Fig. 3). No evidence of publication bias was
PT
The subgroup analysis and meta-regression analyses for the prognostic value of plasma
BNP was based on the testing day, testing method, or observation endpoint. The Biosite
Triage immunoassay has higher summary specificity than immunoradiometric assay, and the
sensitivity and specificity of NT-proBNP was lower in the 28-day mortality endpoint (Table
3). These results indicate that the testing method or observation endpoint may have caused
Nine studies with 691 patients with sepsis assessed the prognostic value of plasma
NT-proBNP for sepsis. The pooled results showed that NT-proBNP has a moderate
prognostic value, with the pooled sensitivity and specificity of 0.71 (95% CI: 0.60-0.80) and
0.73 (95% CI: 0.63-0.81), respectively. The PLR, NLR, and DOR was 2.6 (95% CI: 1.9-3.6),
PT
0.39 (95% CI: 0.28-0.55), and 7 (95% CI: 4-12), respectively. However, there was significant
RI
heterogeneity across the studies in the pooled sensitivity (I2=88.86%, P<0.01) and specificity
analysis (I2=67.25%, P<0.01) (Fig. 5.) The AUC of SROC was 0.78 (0.75-0.82), suggesting a
SC
moderate diagnostic accuracy (Fig. 6). The Deek’s test indicated no significant publication
NU
bias across the included studies (P=0.191; Fig. 7).
MA
Because there was significant heterogeneity in the summary sensitivity and specificity,
D
we further explored the potential causes of the heterogeneity in the meta-analysis using
E
subgroup analysis and meta-regression analyses based on the testing day, testing method, or
PT
observation endpoint. The results showed that the summary sensitivity and specificity of
CE
plasma NT-proBNP tested at day one is not significantly different compared with that tested
in other days. Similar results were also found in the subgroup analysis for testing the method
AC
Discussion
In this study, we evaluated the value BNP or NT-proBNP in predicting the mortality of
patients with sepsis. Our meta-analysis of twenty-two studies showed that both BNP and
NT-proBNP have moderate prognostic value for patients with sepsis. Because there was
ACCEPTED MANUSCRIPT
heterogeneity across the studies in terms of combined sensitivity and specificity, we analyzed
the prognostic value of these biomarkers using subgroup analysis of the different testing time,
testing method, or observation endpoint, which yielded more important information than
that may have been the source of heterogeneity, and the results were similar to that of the
PT
Sepsis and septic shock are severe systemic inflammatory response to infection.
RI
Myocardial dysfunction induces the endogenous release of various mediators with deleterious
SC
cardiovascular effects; thus, it is a crucial complication of septic shock [40]. BNP and
NT-proBNP are cardiac hormones with diuretic, natriuretic and vasorelaxation properties and
NU
play fundamental roles in regulating cardiac filling pressure and intravascular volume
homeostasis [41, 42]. As such, plasma BNP and NT-proBNP are widely used as a valuable
MA
biomarker of left ventricular (LV) dysfunction and increased LV filling pressure [43, 44].
However, increased plasma levels of BNP and NT-proBNP are not specific for heart failure,
D
and circulating levels may be influenced by several cardiac and non-cardiac conditions that
E
cause myocardial strain and affect the clearance of these prohormones, including myocardial
PT
Plasma BNP and NT-proBNP levels predict the prognosis in patients with severe sepsis
and septic shock. However, the association between these peptides and cardiac filling
AC
pressures in sepsis and septic shock is unclear. Previously, Wang et al [12] performed a
meta-analysis of the prognostic value of BNP or NT-proBNP for sepsis using 12 studies and
found that elevated BNP or NT-proBNP were significantly associated with increased risk of
mortality of patients with sepsis. The pooled sensitivity and specificity for BNP were 79%
(95% CI: 75-83) and 60% (95% CI: 57 -62). However, they did not analyze the prognostic
value of BNP or NT-proBNP separately; Thus, their individual values cannot be determined.
ACCEPTED MANUSCRIPT
In addition, they did not separate the mortality endpoint, which included 28-day, ICU, and
in-hospital mortality. Thus, the conclusions are vague and could not reflect the true
prognostic value of BNP or NT-proBNP. A major strength of our study is that the sample size
was larger than the ideal one. Meanwhile, there were also some limitations that need to be
noted in the present study. First, because the optimal cut-off point of BNP or NT-proBNP
substantially varied across the studies, we only performed subgroup analysis by dividing the
PT
studies into two groups. Second, although the testing day and the endpoint differed in the
RI
studies, we only divided the studies into two groups in the meta-regression analysis. Further
SC
analysis should be performed on the specific variables in the future. Third, the data used in
the present study were not adjusted according to the potential confounders, such as the
NU
patients’ age and sex. It should be noted that the pooled estimates reflected unadjusted
associations between BNPs and all-cause mortality. Fourth, the cause of sepsis varied and
MA
included lung, intra-abdominal, and urinary tract infection, which might influence the
robustness of the results. Collectively, the differences in lead time bias, endpoint measures,
D
control for cardiac function, and sampling time might influence the prognostic value of BNP
E
or NT-proBNP in patients with sepsis. Our findings should be validated in future studies that
PT
Conclusions
AC
marker of outcome in sepsis.. However, large prospective studies that consider potential
confounders are needed to accurately determine the prognostic value of plasma BNP or
NT-proBNP in these patients. Considering the inconsistent results on the prognostic value of
BNP or NT-proBNP for patients with sepsis, our results provide clinicians with information
Conflict of Interest
Funding
This research did not receive any specific grant from funding agencies in the public,
PT
commercial, or not-for-profit sectors.
RI
SC
Ethical approval
This article does not contain any studies with human participants or animals performed by
NU
any of the authors.
MA
E D
PT
CE
AC
ACCEPTED MANUSCRIPT
Reference
PT
2016;351(6):555-62.
RI
[3] Hamed S, Behnes M, Pauly D, Lepiorz D, Barre M, Becher T, et al. Diagnostic value
SC
of Pentraxin-3 in patients with sepsis and septic shock in accordance with latest
[5] Wang B, Li J, Gao HM, Xing YH, Lin Z, Li HJ, et al. Necroptosis regulated proteins
peptide levels in a large noncardiac, general intensive care unit population. Shock
2009;31(4):342-7.
[9] Khoury J, Arow M, Elias A, Makhoul BF, Berger G, Kaplan M, et al. The prognostic
value of brain natriuretic peptide (BNP) in non-cardiac patients with sepsis, ultra-long
PT
Shock. Crit Care Med 2016;44(4):707-16.
RI
[11] Singh H, Ramai D, Patel H, Iskandir M, Sachdev S, Rai R, et al. B-Type Natriuretic
SC
Peptide: A Predictor for Mortality, Intensive Care Unit Length of Stay, and Hospital
[13] Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, et al. Surviving
D
Sepsis Campaign: international guidelines for management of severe sepsis and septic
E
[14] Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving
CE
Sepsis Campaign: international guidelines for management of severe sepsis and septic
[15] Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al.
The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3).
JAMA 2016;315(8):801-10.
[16] Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of
QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included
[17] Deeks JJ, Macaskill P, Irwig L. The performance of tests of publication bias and other
sample size effects in systematic reviews of diagnostic test accuracy was assessed. J
[18] Brueckmann M, Huhle G, Lang S, Haase KK, Bertsch T, Weiss C, et al. Prognostic
value of plasma N-terminal pro-brain natriuretic peptide in patients with severe sepsis.
Circulation 2005;112(4):527-34.
PT
[19] Charpentier J, Luyt CE, Fulla Y, Vinsonneau C, Cariou A, Grabar S, et al. Brain
RI
natriuretic peptide: A marker of myocardial dysfunction and prognosis during severe
SC
sepsis. Crit Care Med 2004;32(3):660-5.
[20] Mokart D, Sannini A, Brun JP, Faucher M, Blaise D, Blache JL, et al. N-terminal
NU
pro-brain natriuretic peptide as an early prognostic factor in cancer patients
insights into the mechanisms involved in B-type natriuretic peptide elevation and its
D
[22] Post F, Weilemann LS, Messow CM, Sinning C, Munzel T. B-type natriuretic peptide
PT
peptide helps to predict the outcome of patients with major trauma. Eur J Trauma
[24] Rivers EP, McCord J, Otero R, Jacobsen G, Loomba M. Clinical utility of B-type
natriuretic peptide in early severe sepsis and septic shock. J Intensive Care Med
2007;22(6):363-73.
ACCEPTED MANUSCRIPT
[25] Roch A, Allardet-Servent J, Michelet P, Oddoze C, Forel JM, Barrau K, et al. NH2
and cardiac dysfunction in septic shock patients. Crit Care Med 2005;33(5):1001-7.
[26] Sturgess DJ, Marwick TH, Joyce C, Jenkins C, Jones M, Masci P, et al. Prediction of
PT
[27] Ueda S, Nishio K, Akai Y, Fukushima H, Ueyama T, Kawai Y, et al. Prognostic value
RI
of increased plasma levels of brain natriuretic peptide in patients with septic shock.
SC
Shock 2006;26(2):134-9.
Med 2007;35(5):1277-83.
MA
[29] Yucel T, Memis D, Karamanlioglu B, Sut N, Yuksel M. The prognostic value of atrial
and brain natriuretic peptides, troponin I and C-reactive protein in patients with sepsis.
D
[30] Zhao HY, An YZ, Liu F. [Prognostic values of B-type natriuretic peptide in severe
PT
sepsis and septic shock]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2009;21(5):293-5.
CE
[32] Hu BC, Wang YJ, Ge WD, Li FZ, Sun RH. [Combination of B-type brain natriuretic
peptide with left ventricular diastolic dysfunction for prediction of mortality in the
[33] Park BH, Park MS, Kim YS, Kim SK, Kang YA, Jung JY, et al. Prognostic utility of
failure assessment scores in patients with acute lung injury/acute respiratory distress
[34] Li SW, Sun FH, Tao ZX, Song J, Zhao YF. Distribution of pathogens isolated from
patients with acute abdomen patients complicated with septic shock and value of brain
2017;27(16):3707-310.
PT
[35] Li ZH, Xiao J, Li JZ. Prognostic value of N-terminal pro-B-type natriuretic peptide in
RI
critically ill patients. hinese Critical Care Medicine 2011;23(3):179-82.
SC
[36] Xing B, Wang XZ, Zeng Y, Tan SF, Huang S. Prognostic value of plasma B-type
natriuretic peptide in patients with septic shock. China Journal of Modern Medicine
NU
2013;23(32):75-9.
[37] Yang Y, Wang XP, Chen XJ, Ren H. Prognostic value of brain natriuretic peptides
MA
[38] Phua J, Koay ES, Lee KH. Lactate, procalcitonin, and amino-terminal pro-B-type
D
natriuretic peptide versus cytokine measurements and clinical severity scores for
E
[39] Perman SM, Chang AM, Hollander JE, Gaieski DF, Trzeciak S, Birkhahn R, et al.
CE
Relationship between B-type natriuretic peptide and adverse outcome in patients with
clinical evidence of sepsis presenting to the emergency department. Acad Emerg Med
AC
2011;18(2):219-22.
[40] Martin GS. Sepsis, severe sepsis and septic shock: changes in incidence, pathogens
[41] Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med
1998;339(5):321-8.
ACCEPTED MANUSCRIPT
[42] Tschope C, Kasner M, Westermann D, Walther T, Gaub R, Poller WC, et al. Elevated
NT-ProBNP levels in patients with increased left ventricular filling pressure during
[43] Park SJ, Cho EJ, Ahn J, Carriere K, Kim EK, Lee GY, et al. Additive prognostic
with degenerative mitral regurgitation and preserved left ventricular ejection fraction.
PT
Int J Cardiol 2017;236:387-92.
RI
[44] Siva Sankara C, Rajasekhar D, Vanajakshamma V, Praveen Kumar BS, Vamsidhar A.
SC
Prognostic significance of NT-proBNP, 3D LA volume and LV dyssynchrony in
Figure legend
Figure 2 Summary sensitivity, specificity plotted on forest graphs for BNP in predicting the
Figure 3 SROC curve graph for BNP in predicting the mortality of septic patients.
Figure 4 Deek’s funnel plot asymmetry test of BNP in predicting the mortality of septic
PT
patients.
RI
Figure 5 Summary sensitivity, specificity plotted on forest graphs for NT-proBNP in
SC
predicting the mortality of septic patients.
Figure 6 SROC curve graph for NT-proBNP in predicting the mortality of septic patients.
NU
Figure 7 Deek’s funnel plot asymmetry test of NT-proBNP in predicting the mortality of
septic patients.
MA
E D
PT
CE
AC
ACCEPTED MANUSCRIPT
PT
28d Roche,
Park
Korea d2 NA 18/31 82/81 P morta Elecsys 2010 12
/2011
lity analyzer
RI
28d
Li 74.3/69
China d1 1418 53/35 P morta ELISA 13
/2011 .8
SC
lity
hospit
Roche,
Sturgess Austra al
d2 400 15/6 83/40 P Elecsys 2010 13
NU
/2010 lia morta
analyzer
lity
28d Roche,
Phua Singa 70.8
d2 6624 42/30 P morta Elecsys 2010 13
MA
hospit
Roche,
Varpula Finlan al
E
lity
ICU Roche,
Roch/20 1360
France d1 22/17 73/83 P morta Elecsys 2010 13
05 0
CE
lity analyzer
Brueckm 28d
Germa 1190
ann d2 41/16 50/90 P morta ELISA 13
ny 0
/2005 lity
AC
BNP
Biosite
28d
Li 86.21/8 Triage
China d1 789 38/29 P morta 13
/2017 9.47 immunoassa
lity
y
Biosite
60d
Hu Triage
China d1 38.8 73/54 86/72.9 P morta 13
/2016 immunoassa
lity
y
28d Biosite
Yang 95.5/76
China d1 1085 39/26 P morta Triage 13
/2016 .4
lity immunoassa
ACCEPTED MANUSCRIPT
y
Biosite
Papanik 28d
Greec Triage
olaou d1 800 20/22 65/64 P morta 13
e immunoassa
/2014 lity
y
Biosite
28d
Xing Triage
China d3 255 82/46 76.1/78 R morta 12
/2013 immunoassa
lity
y
Biosite
30d
Perman Triage
PT
USA d2 49 767/58 63/69 P morta 13
/2011 immunoassa
lity
y
hospit Biosite
RI
Sturgess Austra al Triage
d3 254 15/6 83/60 P 13
/2010 lia morta immunoassa
SC
lity y
Biosite
28d
Chen Triage
China d1 113 707/293 86/55 P morta 13
NU
/2009 immunoassa
lity
y
Biosite
28d
Zhao 681. 91.4/80 Triage
MA
ICU Immunoradi
Yucel Turke
PT
Rivers Triage
USA d1 210 214/38 79/59 P morta 13
/2007 immunoassa
lity
y
AC
28d Immunoradi
Ueda
Japan d2 650 22/12 92/80 P morta ometric 13
/2006
lity assay
Charpent 28d Immunoradi
ier France d1 190 24/10 70/67 P morta ometric 13
/2004 lity assay,
P: prospective; R: Retrospective; ICU: intensive care unit
Table 2 Subgroup analysis and meta-regression analyses of BNP
Subgroup nstudies Sensitivity P1 Specificity P2
Tested 0.86(95%CI: 0.73(95%CI:
Day one 9 0.22 0.19
day 0.79-0.92) 0.64-0.82)
ACCEPTED MANUSCRIPT
0.79(95%CI: 0.72(95%CI:
Non-day one 5
0.68-0.91) 0.59-0.86)
Tested Immunoradiometric 0.82(95%CI: 0.70(95%CI:
11 0.05 0.02
method assay 0.76-0.89) 0.62-0.77)
Biosite 0.89(95%CI: 0.86(95%CI:
3
Triageimmuno assay 0.78-1.00) 0.73-0.98)
0.82(95%CI: 0.71(95%CI:
Endpoint 28d mortality 11 0.02 0.04
PT
0.75-0.88) 0.63-0.79)
0.90(95%CI: 0.80(95%CI:
Non-28d mortality 3 .
RI
0.81-0.99) 0.66-0.94)
SC
Table 3 Subgroup analysis and meta-regression analyses of NT-proBNP
Subgroup nstudies Sensitivity
NU P1 Specificity P2
Tested Day one 4 0.69(95%CI: 0.17 0.77(95%CI: 0.52
day 0.54-0.83) 0.65-0.88)
Non-day one 5 0.74(95%CI: 0.69(95%CI:
MA
0.61-0.87) 0.57-0.82)
Tested Elecsys2010analyzer 7 0.74(95%CI: 0.93 0.72(95%CI: 0.20
D
0.41-0.84) 0.61-0.94)
PT
Highlights
Plasma BNP and NT-proBNP were used to predict the mortality for septic patients.
The current results of BNP and NT-proBNP for the predictive value remain
controversial.
3417 septic patients were selected for the large meta-analysis.
Both BNP and NT-proBNP have moderate predicts value for the mortality of septic
PT
patients.
The tested method of BNP and NT-proBNP observation endpoint influence the results
RI
of BNP.
SC
NU
MA
E D
PT
CE
AC
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7