ORIGINAL ARTICLE
The Diagnostic Utility of Brain Natriuretic Peptide in Heart Failure Patients Presenting with Acute Dyspnea:
A Systematic Review and Meta-analysis Mohammad Amin Karimi1#.
Zahra Kazemi Ferezghi2#.
Reza Khademi3#.
Seyed Amirhossein Mazhari4.
Fatemeh Chichagi5.
Asma Rasouli6.
Reyhaneh Alikhani7.
Anis Sani8.
Shima Akhavan Rezayat9.
Nima Shakouri10.
Seyed Iraj Azimi11.
Faezeh Jadidian2.
Golnaz Nikeghbali2.
Mahfam Edrisian12.
Alaleh Alizadeh9.
Niloofar Deravi2.
Mohadeseh Poudineh13*.
Mahsa Asadi Anar2* School of Medicine.
Shahid Beheshti University of Medical Sciences.
Tehran.
Iran.
Student Research Committee.
School of Medicine.
Shahid Beheshti University of Medical Sciences.
Tehran.
Iran.
Student Research Committee.
Faculty of Medicine.
Mashhad University of Medical Sciences.
Mashhad.
Iran.
Student Research Committee.
Azerbaijan Medical University.
Baku.
Azerbaijan.
Iran.
Tehran Heart Center.
Tehran University of Medical Sciences.
Tehran.
Iran.
School of Medicine.
Zanjan University of Medical Sciences.
Zanjan.
Iran.
Tehran University of Medical Sciences.
Tehran.
Iran.
Student Research Committee.
Tabriz University of Medical Sciences.
Tabriz.
Iran.
Student Research Committee.
Faculty of Medicine.
Islamic Azad University of Mashhad.
Mashhad.
Iran.
School of Medicine.
Guilan University of Medical Sciences.
Rasht.
Iran.
Student Research Committee.
Iran University of Medical Sciences.
Tehran.
Iran.
Student Research Committee.
School of Medicine.
Iran University of Medical Sciences.
Tehran.
Iran.
Student Research Committee.
School of Medicine.
Zanjan University of Medical Sciences.
Zanjan.
Iran.
#The authors contributed equally to the article.
*Corresponding Authors:
Mahsa Asadi Anar.
MD.
Shahid Beheshti University of Medical Sciences.
Arabi Ave.
Daneshjoo Blvd.
Velenjak.
Tehran 19839-63113.
Iran.
Email: mahsa.
boz@gmail.
Mohadeseh Poudineh.
MD.
School of Medicine.
Zanjan University of Medical Sciences.
12th Street.
Shahrake Karmandan.
Zanjan 45139-56111.
Iran.
Email: mo.
poudineh@gmail.
ABSTRACT
Background: Patients with heart failure are often diagnosed based on clinical signs and serological markers.
Finding biomarkers with greater sensitivity and specificity for heart failure patients who also have episodic dyspnea is a challenge for researchers.
Thus, we conducted a systematic review and meta-analysis of previous research to determine the diagnostic value of B-type natriuretic peptide as a potential biomarker in heart failure patients experiencing acute dyspnea.
Methods: By searching PubMed/Medline.
Scopus, and Google Scholar up to March 2023, all cross-sectional and cohort studies were selected according to the PRISMA guidelines and assessed by the DeeksAo funnel plot asymmetry test for bias.
Results: A total of thirty-five qualifying studies had their data extracted.
In 26 investigations .
=16.
, the precision of B-type natriuretic peptide was evaluated.
There were significant differences in the reported sensitivity and specificity between trials.
One research study yielded the lowest sensitivity of 0.
68, 0.
, with a prevalence of 46% for heart failure and a BNP level of Ou500 pg/ml.
Specificity grew but stayed variable as the threshold rose, whereas sensitivity declined.
A diagnostic meta-analysis was carried out on 14 trials .
=6.
to determine the accuracy of N-terminal probrain natriuretic Acta Med Indones - Indones J Intern Med A Vol 57 A Number 2 A April 2025 Mohammad Amin Karimi Acta Med Indones-Indones J Intern Med When the threshold is raised, the pattern in NTproBNP is similar to that of B-type natriuretic peptides, with sensitivity falling and specificity increasing.
Following the final analysis, the confidence areas surrounding the pooled sensitivity and specificity for BNP vs NTproBNP showed a distinct overlap.
The overlap indicated that there was no statistically significant difference between the tests at the <100 pg/ml and O300 pg/ml ruleout levels, respectively (P>0.
Conclusion: The meta-analysis reveals a substantial degree of congruity in the sensitivity and specificity between the levels of BNP and NTproBNP as biomarkers.
Nevertheless, it's worth noting that, in the end, there exists a potential for overlooking heart failure diagnoses.
Larger future studies, overcoming past limitations, could likely establish a consensus.
Keywords: NTproBNP, heart failure.
B-type natriuretic peptide.
BNP, meta-analysis INTRODUCTION Breathlessness, often known as dyspnea, is one of the most prevalent reasons for visits to emergency departments.
Dyspnea is a subjective perception of breathing difficulty characterized by qualitatively diverse sensations of varying 1 During the initial hours, the primary diagnosis is occasionally ambiguous.
Some of the most prevalent underlying causes of acute dyspnea are cardiovascular disease (CVD), congestive heart failure (CHF), and chronic obstructive pulmonary disease (COPD).
For many patients with acute dyspnea, clinical evaluations based on symptoms, physical indicators, and chest radiography remain The most frequent presenting symptom in patients with heart failure (HF) is dyspnea.
specifically, dyspnea at rest was prevalent in 0% of patients in North America and 70.
1% of patients worldwide.
3 Dyspnea is commonplace in all phases of heart failure, and its underlying cause and severity must be evaluated to comprehend disease progression and treatment efficacy.
Even individuals with stable HF have many symptoms, with dyspnea being the most noticeable and In an HF outpatient clinic, up to 85 percent of patients judged shortness of breath to be the most significant symptom.
4 In the United States, the emergency department commences hospital-based care for more than 80% of patients with HF, and these Patients typically present with Given the broad differential for such a primary complaint, particularly in patients with numerous comorbidities, obtaining an accurate diagnosis is essential but can be challenging.
The discovery and creation of novel biomarkers that aim to improve physicians' ability to diagnose and predict their patients' prognosis has been a significant development in medicine over the past decade.
One of these biomarkers is brain natriuretic peptide (BNP).
Many studies have shown how crucial BNP is in the differential diagnosis of patients with acute dyspnea and how closely its level is correlated with the degree of heart failure.
Thirty-two amino acids make up BNP, which is created as the pre-prohormone protein proBNP.
ProBNP is secreted from cardiac myocytes in response to myocardial stretch, volume overload, and increased end-diastolic pressure.
It is then cleaved into an active BNP and an inactive N-terminal pro-B-type natriuretic peptide (NTproBNP) with 76 amino acids.
BNP and NTproBNP are typically used to evaluate cardiac function because they are biologically distinct from one another.
They also play a significant role in the control of natriuresis, diuresis, and vascular tone.
ACC/AHA guidelines recommend BNP and NT-proBNP for the sole diagnosis of heart failure.
Testing for BNP and NT-proBNP is an advantageous diagnostic and predictive method for heart failure.
6 To the best of our knowledge, this systematic review, for the first time, is going to critically elucidate the diagnostic utility of brain natriuretic peptide in heart failure patients presenting with acute dyspnea.
METHODS
This systematic review/meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.
Our design protocol was registered in the International Prospective Vol 57 A Number 2 A April 2025 The Diagnostic Utility of Brain Natriuretic Peptide in Heart Failure Register of Systematic Reviews, known as PROSPERO (CRD42023421.
ttps:// uk/prospero/display_record.
php?ID=CRD42023421.
Literature Search Firstly, a comprehensive search was done in PubMed (Medlin.
and Scopus databases on March 12th, 2023.
The search strategy was conducted based on the title, abstract, and suitable keywords and tags using advanced search features for each search engine.
limitation regarding time and language was The summary of our search strategy is available in Table 1.
After searching, removing duplicates, and retrieving initiated articles, two researchers independently reviewed the studiesAo titles and abstracts, and conflicts were resolved by discussion with a third member.
The next step was reviewing the full text of identified articles for exact inclusion criteria.
All studies met the inclusion criteria included in the study.
Inclusion and Exclusion Criteria In the present study, the authors included published papers discussing the diagnostic utility of BNP in patients presenting with dyspnea with a history of HF.
We included Randomized Controlled Trials (RCT.
, cross-sectional studies, and cohorts.
Other studies were excluded, including reviews, case reports, case series, letters to the editor, abstracts, and posters.
Methodological Quality Assessment We used the JBI criteria appraisal checklist to assess the included studies' quality, which is available at https://jbi.
global/criticalappraisal-tools.
Two team members evaluated all included studiesAo full texts independently, and disagreements were resolved by scientific Then, a data extraction form including author, country, study design, follow-up duration, sex, mean age, participants, index test, the reference standard for HF, adjustment, threshold, and outcomes was prepared, and two independent authors conducted this process.
Statistical Analysis From every study, we retrieved data on binary diagnostic accuracy and created 2x2 tables at every threshold.
To provide a general idea of test accuracy, we created a receiver operating characteristic plot of estimations for sensitivity and specificity for each threshold for each natriuretic peptide test.
These charts show the data divided by research design and index test.
One of two study designsAicasecontrol or cross-sectional/cohortAiwas specified.
Studies involving chosen healthy and unhealthy populations were included in the category of case-control studies.
We did not include casecontrol studies in our analysis since their results are prone to bias.
For every natriuretic peptide, we created matched forest plots along with matching 95 percent confidence intervals.
Based on these thresholds, we divided the data into two groups:
<100 and Ou500 for BNP, and O300 and Ou1800 for NT-proBNP.
For every natriuretic peptide, we generated average summary receiver operating characteristic .
ROC) curves and forest plots with estimates of the paired observed sensitivities and specificities at the predetermined threshold.
The charts illustrate how the accuracy of the studies varies.
We performed diagnostic meta-analyses when sufficient data were available.
We needed five or more studies for each criterion to pool data.
To pool test accuracy for the investigations, we employed the bivariate technique modeled in Winbugs (Medical Research Council Biostatistics Uni.
Logistic regression is employed in the bivariate approach to analyze the true positives, true negatives, false positives, and false negatives that were documented in the research.
constructed sROCs and plotted confidence regions .
sing methods outlined by Novielli and 7 From every study, we retrieved data on binary diagnostic accuracy and created 2x2 tables at every threshold.
To provide a general idea of test accuracy, we created a receiver operating characteristic plot of estimations for sensitivity and specificity for each threshold for each natriuretic peptide test.
These charts show the data divided by research design and index test.
One of two study designsAicasecontrol or cross-sectional/cohortAiwas specified.
Studies involving chosen healthy and unhealthy populations were included in the category of Mohammad Amin Karimi case-control studies.
We did not include casecontrol studies in our analysis since their results are prone to bias.
For every natriuretic peptide, we created matched forest plots along with matching 95 percent confidence intervals.
Based on these thresholds, we divided the data into two groups:
<100 and Ou500 for BNP, and O300 and Ou1800 for NT-proBNP.
For every natriuretic peptide, we generated average summary receiver operating characteristic .
ROC) curves and forest plots with estimates of the paired observed sensitivities Acta Med Indones-Indones J Intern Med and specificities at the predetermined threshold.
The charts illustrate how the accuracy of the studies varies.
We performed diagnostic metaanalyses when sufficient data were available.
We needed five or more studies for each criterion to pool data.
To pool test accuracy for the investigations, we employed the bivariate technique modeled in Winbugs (Medical Research Council Biostatistics Uni.
Logistic regression is employed in the bivariate approach to analyze the true positives, true negatives, false positives, and false negatives that were documented in the research.
Table 1.
Curated search strategies across databases Search Engine Search strategy Time Results Scopus TITLE-ABS-KEY ( heart ) AND ( TITLE-ABS-KEY ( dyspnea ) OR TITLE-ABS-KEY ( short AND of AND breath ) OR TITLEABS-KEY ( breathlessness ) OR TITLE-ABS-KEY ( breath AND shortness ) OR TITLE-ABS-KEY ( recumbent AND dyspnea ) OR TITLE-ABS-KEY ( rest AND dyspnea )) AND ( TITLE-ABS-KEY ( brain AND natriuretic AND peptide ) OR TITLE-ABS-KEY ( bnp-32 ) OR TITLE-ABS-KEY ( brain AND natriuretic AND peptide-32 ) OR TITLE-ABS-KEY ( natriuretic AND factor 32 ) OR TITLE-ABSKEY (BNP AND gene AND product ) OR TITLE-ABS-KEY ( type-b AND natriuretic AND peptide ) OR TITLE-ABS-KEY ( Nesiritide ) OR TITLE-ABS-KEY ( b-type AND natriuretic AND peptide ) OR TITLE-ABS-KEY (BNP) OR TITLE-ABS-KEY ( Natrecor )) March 12th, 2023 PubMed .
eart[Title/Abstrac.
) AND (.
yspnea[Title/Abstrac.
) OR (Shortness of Breath[Title/Abstrac.
) OR (Breath Shortness[Title/ Abstrac.
) OR (Breathlessness[Title/Abstrac.
) OR (Orthopnea[Title/ Abstrac.
) OR (Recumbent Dyspnea[Title/Abstrac.
) OR (Dyspnea.
Recumbent[Title/Abstrac.
) OR ( Platypnea [Title/Abstrac.
) OR (Trepopnea[Title/Abstrac.
) OR (Rest Dyspnea[Title/Abstrac.
) OR (Dyspnea.
Rest[Title/Abstrac.
) OR (Dyspneas.
Rest[Title/ Abstrac.
)) AND (.
rain natriuretic peptide[Title/Abstrac.
) OR (Peptide.
Brain Natriuretic[Title/Abstrac.
) OR (BNP-32[Title/ Abstrac.
) OR (BNP 32[Title/Abstrac.
) OR (Brain Natriuretic Peptide-32[Title/Abstrac.
) OR (Brain Natriuretic Peptide 32[Title/ Abstrac.
) OR (Natriuretic Peptide-32.
Brain[Title/Abstrac.
) OR (Peptide-32.
Brain Natriuretic[Title/Abstrac.
) OR (Natriuretic Factor-32[Title/Abstrac.
) OR (Natriuretic Factor 32[Title/Abstrac.
) OR (BNP Gene Product[Title/Abstrac.
) OR (Type-B Natriuretic Peptide[Title/Abstrac.
) OR (Natriuretic Peptide.
Type-B[Title/ Abstrac.
) OR (Type B Natriuretic Peptide[Title/Abstrac.
) OR (Natriuretic Peptide Type-B[Title/Abstrac.
) OR (Natriuretic Peptide Type B[Title/Abstrac.
) OR (Nesiritide[Title/Abstrac.
) OR (B-Type Natriuretic Peptide[Title/Abstrac.
) OR (Natriuretic Peptide.
B-Type[Title/Abstrac.
) OR (Natrecor[Title/Abstrac.
)) March 12th.
Vol 57 A Number 2 A April 2025 The Diagnostic Utility of Brain Natriuretic Peptide in Heart Failure RESULTS Study Selection A PRISMA flow diagram offers an overview of the exclusions and screening procedure (Figure .
Searches through PubMed.
Scopus, and Google Scholar produced 6228 papers that might be of interest.
498 of these 6228 documents were identical hits, so we decided not to give them any further thought.
After evaluating the abstracts and titles of 5730 papers, we eliminated 5410 as being unrelated to the review.
After obtaining and examining 320 full-text papers, we formally eliminated 285 of them .
hich had no relevant outcome.
A total of 35 studies were deemed eligible and incorporated into the review.
After searching in (PubMed/Medline.
Scopus, and Google Schola.
databases, a total of 6228 articles were obtained, and 498 duplicates were removed.
After reviewing the title & abstract screening, 320 studies remained.
The final review includes 35 articles of the final full-text results.
the rest of which had unrelated data were deleted.
Baseline Characteristics In total, 35 cohort studies were included in our systematic review and meta-analysis, which included 16102 patients.
The patients' average age was 68.
3, and 45.
5% of them were female.
The years of investigation were 2002 through The studies came from over ten different Records identiAed from: 6228 Pubmed/Medline .
Scopus .
=5.
Goole Scholar .
Records removed before Duplicate records removed .
Records screened .
=5.
Records excluded** .
=5.
Records assessed for eligibility .
Records excluded .
=285 Reasons:
Irrelevant outcom.
Studies included in review .
Figure 1.
Flowchart of literature inclusion following PRISMA guidelines.
Mohammad Amin Karimi countries in a variety of geographical situations.
The USA was the primary location for participant recruitment, with sample sizes varying from 41 to 1586 patients.
A thorough explanation of the data retrieved from the listed research is provided in Table 2.
B-type Natriuretic Peptide 26 studies evaluated the accuracy of B-type natriuretic peptide.
Data were reported at the <100 ng/L criterion in 7 studies .
ncluding 4434 peopl.
, the 100-500 ng/L threshold in 21 studies .
nvolving 9371 participant.
, and the Ou500 ng/L threshold in 6 studies .
nvolving 2297 participant.
The majority of research used a cross-sectional/cohort design to evaluate the accuracy of the Triage test in comparison to clinical assessment.
There were significant differences in the reported sensitivity and specificity between One study's findings, with a heart failure Acta Med Indones-Indones J Intern Med prevalence of 46%, had the lowest sensitivity.
The pooled sensitivity and specificity of B-type natriuretic peptide at a threshold of less than 100 were 0.
89, 0.
60, 0.
respectively, when diagnostic meta-analysis was performed.
The combined sensitivity and specificity at a B-type natriuretic peptide concentration of 100Ae500 pg/ml were 0.
70, 0.
The combined sensitivity and specificity at a B-type natriuretic peptide concentration of Ou500 pg/ml were 0.
68, 0.
72, 0.
Sensitivity dropped and specificity rose, but stayed unstable as the threshold rose (Figure .
More details are included in Figure 3, which shows the summary sensitivity and specificity points concerning the forest plots in Figure 1.
While the specificity was higher, the summary point for the threshold of Ou500 pg/ml is lower in the sROC space than the criterion of <100 pg/ml, indicating that heart failure diagnoses might be overlooked.
Figure 2.
Paired sensitivity and specificity plots for B-type natriuretic peptide at three threshold levels Vol 57 A Number 2 A April 2025 The Diagnostic Utility of Brain Natriuretic Peptide in Heart Failure Figure 3.
Results for B-type natriuretic peptide .
eparated by threshol.
displayed in summary receiver operating characteristic space.
The results of the diagnostic meta-analysis showed that, at a threshold of O300 ng/L, the pooled sensitivity, specificity, positive predictive value, and negative predictive value of NTproBNP were, respectively, 0.
87, 0.
57, 0.
The pooled sensitivity and specificity were 0.
82, 0.
80 at an NTproBNP level of 300Ae1800 pg/ml .
75, 0.
The pooled sensitivity and specificity were 0.
49, 0.
90 at an NTproBNP level of Ou1800 pg/ml .
85, 0.
Figures 4 and 5 show how B-type natriuretic peptides behave at various thresholds: as the threshold is raised, specificity increases and sensitivity decreases.
Figure 5 summary points unmistakably show a drop in mean sensitivity by threshold.
Comparisons Between Natriuretic Peptides N-terminal Pro-brain Natriuretic Peptide In 14 studies, the accuracy of B-type natriuretic peptide was evaluated.
Data at the O300 pg/ml level were reported in 5 investigations, including 1671 people.
At the 300Ae1800 pg/ ml barrier in 11 investigations, including 4056 people, and the Ou1800 pg/ml threshold in two studies involving 586 participants.
Following diagnostic meta-analysis, the confidence areas encompassing the pooled sensitivity and specificity for B-type natriuretic peptide and NTproBNP overlapped.
The overlap indicated that there was no statistically significant difference between the tests conducted at the <00 ng/L and O300 ng/L rule-out thresholds, respectively (P>0.
(Figure .
Figure 4.
Paired sensitivity and specificity plots for N terminal probrain natriuretic peptide at three threshold levels.
Mohammad Amin Karimi Acta Med Indones-Indones J Intern Med Publication Bias and Heterogeneity No indication of publication bias was found for BNP<100, 100OBNP<500, 300<NTproBNP<1800, and NTproBNPOu1800 according to Deeks' funnel plot asymmetry test .
ee supplementary fil.
The I2 data for every peptide at every threshold are shown in Table 3.
I2 statistics were consistently greater than 50%, as anticipated in diagnostic meta-analyses, due to variations in the underlying diagnoses and comorbidities of the patients.
DISCUSSION
Figure 5.
Results for N terminal pro-brain natriuretic peptide .
eparated by threshol.
displayed in summary receiver operating characteristic space.
Figure 6.
Comparison of pooled B type natriuretic peptide and N terminal pro-brain natriuretic peptide diagnostic accuracy results at lowest threshold This systematic study investigated the effectiveness of Brain Natriuretic Peptide (BNP) and N-terminal pro-B-type brain natriuretic peptide (NT-proBNP) as diagnostic tools for heart failure patients experiencing acute dyspnea.
The research included a total of 16102 Participants from 35 separate studies.
The findings revealed an intriguing pattern, as the threshold values for both BNP and NT-proBNP increase, there is a noticeable decrease in sensitivity, while specificity tends to rise.
The meta-analysis unveiled that at a threshold of less than 100 pg/ml.
BNP exhibited a sensitivity of 0.
ith a confidence interval 89 to 0.
and a specificity of 0.
ith a confidence interval of 0.
60 to 0.
Notably.
Koulouri et al.
9 suggested that BNP thresholds below 100 pg/mL were optimal for maximizing sensitivity to rule out congestive heart failure (CHF).
However, 100 pg/mL provided higher specificity for diagnosing CHF, especially in pediatric patients with respiratory distress.
BNP
cut-offs of 40 pg/mL and 60 pg/mL displayed even higher sensitivity at 91% and 83%, respectively, compared to 78% for 100 pg/mL.
Table 3.
Sensitivity and Specificity for All Peptides Natriuretic peptide .
BNP
NTproBNP
<100
Ou500
O300
Ou1800 Number of Total patients Sensitivity (%) Specificity (%) 92 .
89, 0.
83, 0.
68, 0.
87, 0.
82, 0.
49, 0.
60, 0.
70, 0.
72, 0.
57, 0.
75, 0.
85, 0.
Vol 57 A Number 2 A April 2025 The Diagnostic Utility of Brain Natriuretic Peptide in Heart Failure Deek's Funnel Plot Asymmetric Test P-value = 0.
(A)
Deek's Funnel Plot Asymmetric Test P-value = 0.
(B)
(C)
Mohammad Amin Karimi Acta Med Indones-Indones J Intern Med
(D)
Deek's Funnel Plot Asymmetric Test Pvalue = 0.
(E)
(F)
Figure 7.
Thoroughly evaluates publication bias across distinct subsets of studies.
Funnel plots are depicted for specific B-type natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NTproBNP) concentration ranges, including: (A) BNP < 100 pg/ml (B) 100 O BNP < 500 pg/ml (C) BNP Ou 500 pg/ml (D) NTproBNP O 300 pg/ml (E) 300 < NTproBNP < 1800 pg/ ml (F) NTproBNP Ou 1800 pg/ml.
Each plot illustrates the distribution of effect sizes relative to their standard errors, providing insights into potential publication bias within individual study subsets.
Vol 57 A Number 2 A April 2025 The Diagnostic Utility of Brain Natriuretic Peptide in Heart Failure In contrast, the 100 pg/mL cut-off showed a better specificity of 85% compared to 77% for both 40 pg/mL and 60 pg/mL.
These findings indicate that BNP thresholds below 100 pg/mL are more effective at excluding cases of CHF due to their high sensitivity.
However, given its higher specificity, a 100 pg/mL threshold appears more appropriate for confirming CHF diagnosis.
Focusing on investigations with BNP values in the 100-500 pg/ml range, the sensitivity and specificity were 0.
ith a confidence interval of 83 to 0.
ith a confidence interval 70 to 0.
, respectively.
Furthermore, our data demonstrated that at a BNP level of 500 pg/ ml or higher, sensitivity and specificity were 0.
ith a confidence interval of 0.
68 to 0.
and 79 .
ith a confidence interval of 0.
72 to 0.
In the study conducted by Strunk et al.
10 A diagnostic performance evaluation of brain natriuretic peptide (BNP) for congestive heart failure (CHF) was carried out across various cut-off levels, ranging from 80 pg/mL to 500 pg/ The researchers observed a trade-off between sensitivity and specificity as the BNP cut-off value increased.
Lower BNP thresholds within the 80-100 pg/mL range exhibited very high sensitivity, exceeding 90%, although specificity was somewhat suboptimal, falling below 75%.
On the contrary, higher BNP cut-offs nearing 500 pg/mL demonstrated excellent specificity, exceeding 90%, but at the expense of considerably lower sensitivity, dropping below 60%.
The researchers suggested that an optimal balance between sensitivity and specificity is achieved by utilizing mid-range BNP levels, specifically between 200-400 pg/mL.
These thresholds seem to strike a reasonable compromise between maintaining reasonably high sensitivity while enhancing specificity, as opposed to the extremes of the tested range.
This study highlights an interesting and potentially unfavorable association between sensitivity and specificity, which appears to depend on the chosen diagnostic BNP cut-off.
Parab et al.
11 conducted a study focusing on different brain natriuretic peptide (BNP) cut-offs for diagnosing congestive heart failure (CHF) in elderly individuals, with cut-offs ranging from 100 pg/mL to 700 pg/mL.
The study revealed a gradual reduction in sensitivity with increasing BNP thresholds, ranging from 95.
7% at the 100 pg/mL threshold to 76.
5% at the 700 pg/mL In contrast, specificity exhibited an opposite trend, rising as the cut-offs increased, from a mere 26% at 100 pg/mL to 78.
2% within the 600-700 pg/mL range.
The commonly used 100 pg/mL threshold showcased outstanding sensitivity but poor specificity.
Overall, an inverse relationship between sensitivity and specificity was observed, strongly influenced by the chosen BNP threshold, particularly in the context of diagnosing CHF in elderly patients.
Choi et al.
12 conducted a study involving BNP cut-offs ranging from 12.
5 pg/mL to 983.
pg/mL for diagnosing congestive heart failure in Korean patients experiencing dyspnea.
Their findings indicated a gradual decrease in sensitivity from 100% at 12.
5 pg/mL to 39.
5 pg/mL.
at the same time, specificity exhibited a converse trend, climbing from 27.
8% across the same range of thresholds.
The researchers achieved an optimal balance between sensitivity .
5%) and specificity .
4%) at a BNP cut-off of 296.
5 pg/mL.
Below this threshold, sensitivity exceeded specificity, whereas specificity surpassed sensitivity beyond 5 pg/mL.
This pattern illustrates the typical negative relationship between sensitivity and specificity as dictated by the chosen diagnostic BNP threshold.
Furthermore, this observation, suggesting a decline in sensitivity and an increase in specificity with higher cardiac BNP thresholds, resonates with several other studies that have explored the utility of BNP in detecting heart failure and acute dyspnea.
Regarding the assessment of NT-proBNP, a total of 14 articles were considered.
Among these, five studies .
nvolving 1671 participant.
presented results using the threshold of O300 pg/ml, while 11 studies .
ncompassing 4056 participant.
utilized the 300-1800 pg/ml threshold.
Additionally, two studies .
omprising 586 participant.
applied theOu1800 pg/ml threshold.
Our findings indicated that at the O300 ng/L level, the sensitivity and specificity for NT-proBNP were 0.
ith a confidence interval of 0.
87 to 0.
ith a confidence interval of 0.
57 to 0.
, respectively.
For the NT-proBNP cut-off range of 300-1800 pg/ ml, sensitivity and specificity were 0.
ith a Mohammad Amin Karimi confidence interval of 0.
82 to 0.
ith a confidence interval of 0.
75 to 0.
, respectively.
Furthermore, at an NT-proBNP level of Ou1800 pg/ml, the pooled sensitivity and specificity 62 .
ith a confidence interval of 0.
ith a confidence interval of 85 to 0.
, respectively.
In a study conducted by Kozhuharov et al.
17 An NT-proBNP threshold of 300 pg/mL exhibited a high sensitivity of 98%, albeit at the cost of low specificity at 22%, making it suitable for ruling out acute heart failure (AHF).
In contrast, a cut-off of 2200 pg/ mL demonstrated a specificity of 70% but a lower sensitivity of 83%, making it more suitable for ruling in AHF.
Another study by Mueller et al.
established an NT-proBNP threshold of 825 pg/ mL, ideally balancing sensitivity at 87% with specificity at 81% for AHF diagnosis.
Nazerian et al.
18 identified an NT-proBNP cut-off below 450 pg/mL for patients under 50 years old and below 900 pg/mL for patients over 50.
These thresholds achieved good sensitivity, exceeding 90%, but featured lower specificity at around 50%, allowing for the exclusion of AHF.
Also.
Klemen et al.
19 reported that an NT-proBNP threshold of 2000 pg/mL exhibited a sensitivity of 84% and specificity of 83% for AHF diagnosis, while a threshold of 300 pg/mL yielded a high sensitivity of 99% but lower specificity at 54%, making it suitable for ruling out AHF.
In conclusion, the collective findings from these studies suggest that lower NT-proBNP cut-offs, ranging from 450-300 pg/mL, optimize sensitivity for excluding AHF, whereas higher thresholds in the range of 2000900 pg/mL provide greater specificity for ruling in AHF.
The best compromise between sensitivity and specificity is observed within the 825-2000 pg/mL range.
The relationship between sensitivity and specificity based on thresholds for both BNP and NT-proBNP has never been thoroughly studied before the publication of this systematic review and meta-analysis.
Our study shed light on the intricate relationship between sensitivity and specificity based on the chosen diagnostic thresholds for both BNP and NT-proBNP, providing valuable insights into the diagnostic utility of these biomarkers in detecting heart failure and acute dyspnea.
High heterogeneity in Acta Med Indones-Indones J Intern Med the results of studies included in our analysis is the main drawback of this study, and it is advised to carry on further studies to support our findings.
CONCLUSION
In summary, our findings suggest that elevating the B-type natriuretic peptide (BNP) and N-terminal pro-B-type natriuretic peptide (NTproBNP) cut-off values, utilized as diagnostic tests for heart failure and for distinguishing the underlying causes of acute dyspnea, exhibits a direct correlation with heightened specificity and an inverse correlation with sensitivity.
This implies that adopting a higher threshold may result in an increased occurrence of overlooked heart failure diagnoses, notwithstanding the enhancement in specificity.
Determining the optimal threshold hinges upon whether the clinical priority pertains to optimizing sensitivity or specificity.
The principal limitations of this study stem from the substantial heterogeneity observed in the outcomes of the studies subjected to our analysis.
Consequently, further research is advisable to substantiate our outlined conclusions.
CONFLICT OF INTEREST
The authors declare there is no conflict of interest for this study.
FUNDING
None
ACKNOWLEDGMENTS