Mahapatro et al.

European Journal
European Journal of Medical Research (2024) 29:210
https://doi.org/10.1186/s40001-024-01782-y of Medical Research

REVIEW Open Access

Evaluating biomarkers for contrast‑induced

nephropathy following coronary interventions:
an umbrella review on meta‑analyses
Abinash Mahapatro1, Sara Nobakht2, Sindu Mukesh3, Amir Ali Daryagasht2, Aishwarya Reddy Korsapati4,
Shika M Jain5, Saman Soltani Moghadam2, Rozhin Moosavi6, Mona Javid2, Soheil Hassanipour9*,
Shrinidhi Vilas Prabhu7, Mohammad‑Hossein Keivanlou10, Ehsan Amini‑Salehi2* and Sandeep S. Nayak8

Abstract
Background Contrast-induced nephropathy (CIN) is a form of acute kidney injury (AKI) occurring in patients under‑
going cardiac catheterization, such as coronary angiography (CAG) or percutaneous coronary intervention (PCI).
Although the conventional criterion for CIN detection involves a rise in creatinine levels within 72 h after contrast
media injection, several limitations exist in this definition. Up to now, various meta-analyses have been undertaken
to assess the accuracy of different biomarkers of CIN prediction. However, the existing body of research lacks a cohe‑
sive overview. To address this gap, a comprehensive umbrella review was necessary to consolidate and summarize
the outcomes of prior meta-analyses. This umbrella study aimed to offer a current, evidence-based understanding
of the prognostic value of biomarkers in predicting CIN.
Methods A systematic search of international databases, including PubMed, Scopus, and Web of Science,
from inception to December 12, 2023, was conducted to identify meta-analyses assessing biomarkers for CIN predic‑
tion. Our own meta-analysis was performed by extracting data from the included studies. Sensitivity, specificity, posi‑
tive likelihood ratio, and negative likelihood ratio were assessed using Meta-Disc and CMA softwares.
Results Twelve studies were ultimately included in the umbrella review. The results revealed that neutrophil gelati‑
nase-associated lipocalin (NGAL) exhibited the highest area under the curve (AUC), followed by cystatin-C, urinary
kidney injury molecule-1 (uKIM-1), and brain natriuretic peptide (BNP) with AUCs of 0.91, 0.89, 0.85, and 0.80, respec‑
tively. NGAL also demonstrated the highest positive likelihood ratio [effect size (ES): 6.02, 95% CI 3.86–9.40], followed
by cystatin-C, uKIM-1, and BNP [ES: 4.35 (95% CI 2.85–6.65), 3.58 (95% CI 2.75–4.66), and 2.85 (95% CI 2.13–3.82),
respectively]. uKIM-1 and cystatin-C had the lowest negative likelihood ratio, followed by NGAL and BNP [ES: 0.25
(95% CI 0.17–0.37), ES: 0.25 (95% CI 0.13–0.50), ES: 0.26 (95% CI 0.17–0.41), and ES: 0.39 (0.28–0.53) respectively].
NGAL emerged as the biomarker with the highest diagnostic odds ratio for CIN, followed by cystatin-C, uKIM-1, BNP,
gamma-glutamyl transferase, hypoalbuminemia, contrast media volume to creatinine clearance ratio, preprocedural
hyperglycemia, red cell distribution width (RDW), hyperuricemia, neutrophil-to-lymphocyte ratio, C-reactive protein
(CRP), high-sensitivity CRP, and low hematocrit (P < 0.05).

*Correspondence:
Soheil Hassanipour
[email protected]
Ehsan Amini‑Salehi
[email protected]
Full list of author information is available at the end of the article

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​
mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 2 of 19

Conclusion NGAL demonstrated superior diagnostic performance, exhibiting the highest AUC, positive likelihood
ratio, and diagnostic odds ratio among biomarkers for CIN, followed by cystatin-C, and uKIM-1. These findings under‑
score the potential clinical utility of NGAL, cystatin-C and uKIM-1 in predicting and assessing CIN.
Keywords Contrast induced nephropathy, Biomarkers, Predicators, Cardiac catheterization, Meta-analysis, Umbrella
review, Coronary angiography, Percutaneous coronary intervention

Graphical Abstract

Introduction instances of hospital-acquired acute kidney injury (AKI)

Contrast-induced nephropathy (CIN), also referred to as [3, 4]. Its prevalence ranges from 1 to 2% in the general
contrast-induced acute kidney injury (CI-AKI), is a renal population and can escalate to as much as 50% in high-
injury caused by the administration of radio-opaque con- risk subgroups undergoing procedures such as coronary
trast media (CM) into the vasculature, particularly in angiography (CAG) or percutaneous coronary interven-
individuals who are susceptible to such adverse effects. tion (PCI) [5].
The origins of CIN can be traced back to the 1950s when Nowadays CIN is defined as a 25% relative increase,
case reports documented instances of fatal acute renal or a 0.5 mg/dL (44 µmol/L) absolute increase, in serum
failure subsequent to intravenous pyelography in patients creatinine (SCr) within 72 h of contrast exposure in the
with renal complications associated with multiple mye- absence of alternative conditions [6]. It is reported, that
loma [1, 2]. renal impairment occurring up to 7 days after contrast
Despite advancements in technology, CIN contin- delivery is classified as CIN if it cannot be attributed to
ues to be accountable for approximately one-third of all
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 3 of 19

any other potential cause of renal failure. Serum creati- search formulas for each database are presented in Addi-
nine levels peak 2–5 days after contrast exposure and tional file 1: Table S1.
typically revert to baseline in 14 days [7].
In addition to the rise in serum creatinine, recent stud- Study selection and eligibility criteria
ies propose alternative biomarkers that could predict Two independent researchers (AM and SN) selected the
CIN. Numerous original studies and meta-analyses have studies, with any disparities being resolved through cor-
delved into this area. For instance, a recent meta-analysis respondence (EAS). The eligible meta-analyses adhered
conducted by Javid et al. proposes that Gamma-glutamyl to specific criteria: the population included individuals
transferase (GGT) could serve as a reliable predictor of who had undergone cardiac catheterization, encompass-
CIN. Meanwhile, Wu et al. suggest that inflammatory ing procedures such as CAG or PCI. Furthermore, the
markers such as C-reactive protein (CRP) and neutro- studies were required to report CIN as an outcome of
phil to lymphocyte ratio (NLR) might also be indicative cardiac catheterization. Additionally, within this selected
of CIN [8, 9]. population, at least one biomarker had to be assessed as
Until now, several meta-analyses have introduced a a predictor of CIN. Narrative reviews, original studies,
variety of biomarkers for predicting CIN after cardiac commentaries, and editorials were excluded from the
catheterization. Nevertheless, there exists a need for an study.
umbrella review to comprehensively assess and summa-
rize the outcomes of prior meta-analyses and scrutinize Quality assessment
the quality of their findings. In this umbrella review, our The methodological quality of the included meta-anal-
objective was to appraise the biomarkers identified in yses was evaluated using the AMSTAR2 checklist [12].
previous meta-analyses as predictors of CIN. We intend Two reviewers (A.M, and S.N) provided answers to 16
to assess their robustness using power analysis to provide items on this checklist with the options "yes," "no," or
a thorough evaluation of their strength. "partial yes." Any disagreements were resolved by the
third researcher (E.AS). Based on the checklist score, the
Methods studies were then divided into four groups: high quality,
Our umbrella review, a systematic review encompassing moderate quality, low quality, and critically low quality.
various meta-analyses, adhered to the guidelines out-
lined in the Cochrane Handbook for Systematic Reviews Data extraction
[10]. The presentation of results followed the Preferred The information extracted from the meta-analyses
Reporting Items for Systematic Reviews and Meta-Anal- includes the first author’s name, the publication year, the
yses (PRISMA) guidelines [11]. The study protocol has journal of publication, the study’s country, the number of
been preregistered in PROSPERO under the registration incorporated original studies, the total sample size, pre-
identifier CRD42023493911. dictors of CIN, effect size and 95% confidence interval,
heterogeneity for each outcome, and the funding source,
Search strategy and the checklist for evaluating the quality of the original
To identify meta-analyses assessing predictors of CIN studies. This information was transcribed into an Excel
after cardiac catheterization, two independent research- spreadsheet format.
ers (A.M and S.N), designed a comprehensive search for- To ensure that the dataset was complete, we contacted
mula. This formula was applied across three international the corresponding and primary authors to address any
databases (PubMed, Scopus, and Web of Science) from missing information. Two researchers (A.M and S.N)
their inception up to December 12, 2023. The search worked together to carefully extract the data. If there
included keywords such as "Angiography," "Cardiac were any disagreements, a third researcher was consulted
Catheterization," "Acute Kidney Injury," "Nephropathy," to help resolve the issue.
"Acute Renal Injuries," "Meta-Analysis," and "Systematic
Review". Statistical analyses
To ensure the precision of the search strategy, the In the analysis of the current umbrella review, Compre-
expertise of two information specialists was enlisted. hensive Meta-Analysis (CMA) version 4 was utilized.
Additionally, a manual review of references from rel- For the construction of receiver operating characteristic
evant studies was conducted. There was no language (ROC) charts pertaining to reported biomarkers, Meta-
restriction. In instances of disagreement, resolution was disc software version 1.4 was employed. For assessing
achieved through the involvement of a third researcher summary receiver operating characteristics (SROC),
(E.AS). The organization and management of identified Moses’ constant of linear model was utilized. Der Simo-
studies were facilitated using EndNote X20. The detailed nian-Laird model was used to calculate the effect sizes
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 4 of 19

(ES) of sensitivity, specificity, positive like-hood ratio, Results

and negative like-hood ratio. In the initial phase of the search, a total of 514 stud-
The results of previous meta-analyses were summa- ies emerged, with 99 retrieved from PubMed, 254 from
rized using the pooled results and 95% confidence inter- Scopus, and 161 from Web of Science. Following the
vals of the biomarkers. In instances where two or more elimination of 115 duplicates and screening of titles and
meta-analyses evaluated a common biomarker, the selec- abstracts in the remaining articles, 33 studies progressed
tion criterion was based on the meta-analysis with the to a thorough full-text assessment. Ultimately, after this
largest sample size. To ascertain the adequacy of sample evaluation, 12 studies were eligible to be included in
sizes, power analyses were carried out for each outcome. the Umbrella review. The process of study selection is
Furthermore, the prediction intervals (PI) for each out- depicted in Fig. 1.
come were calculated using CMA software.

Identification of studies via databases and registers

Records identified from

Records removed before
Identification

databases (n = 514)
screening:
PubMed (n =99)
Duplicate records removed
Scopus (n =254)
(n = 115)
Web of Science (n =161)

Records screened Records excluded**

(n = 399) (n = 366)

Reports sought for retrieval Reports not retrieved

(n = 33) (n = 0)
Screening

Reports excluded (n=21)

Reports assessed for eligibility
(n = 33) . Meta-analyses without information of blood
biomarkers (n=10)

. Meta-analyses with information regarding

efficacy of an intervention in CIN (n=9)

. Systematic reviews without meta-analysis

(n=2)
Included

Studies included in review

(n = 12)

Fig. 1 Study selection process

Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 5 of 19

Study characteristic respectively. The sensitivity, specificity positive likelihood

Among the 12 studies incorporated in this analysis, ratio, and negative likelihood ratio were accompanied by
10 studies originated from China [9, 13–21], with the significant heterogeneity (88.7% and 96.4%, 93.7% and
remaining two studies from Iran [8] and the USA [20]. 90.3%, respectively). The area under curve (AUC) for
The publication timeline ranged from 2016 to 2023. The Cystatin C was stated as 0.90 (Table 2 and Fig. 2).
number of studies included and the total sample size Based on the results obtained from the meta-analy-
exhibited variability, ranging from 4 to 26 studies and 946 ses, the sensitivity, specificity, positive likelihood ratio,
to 29,454 participants, respectively. and negative likelihood ratio of BNP for predicting CIN
A comprehensive evaluation of 16 markers, including were 0.72, 0.73, 2.85, and 0.39, respectively. Notably, the
albumin, blood urea nitrogen (BUN), C-reactive pro- reported values for sensitivity, specificity, positive likeli-
tein (CRP), highly sensitive C-reactive protein (hs-CRP), hood ratio, and negative likelihood ratio were associated
neutrophil-to-lymphocyte ratio (NLR), red cell distribu- with varying degrees of heterogeneity, with I­ 2 statistics of
tion width (RDW), hematocrit (HCT), gamma-glutamyl 56.1%, 94.0%, 89.1%, and 70.6%, respectively. Addition-
transpeptidase (GGT), uric acid (UA), urinary kidney ally, the AUC for BNP, as determined was calculated at
injury molecule-1 (uKIM-1), blood glucose, brain natriu- 0.80 (Table 2 and Fig. 2).
retic peptide (BNP), cysteine C, neutrophil gelatinase- The outcomes from the meta-analyses revealed that
associated lipocalin (NGAL), contrast media volume to for uKIM-1, the sensitivity, specificity, positive likeli-
creatinine clearance ratio (V/cr), and platelet-to-lym- hood ratio, and negative likelihood ratio for predicting
phocyte ratio (PLR), was conducted by the meta-analyses CIN were 0.84, 0.76, 3.58, and 0.25, respectively. The het-
included in this review. erogeneity of sensitivity, specificity, positive likelihood
Quality assessments were executed using various ratio, and negative likelihood ratio was 42.4%, 64.0%,
checklists. Five meta-analyses employed the Newcas- 51.1%, and 20.3%, respectively. Additionally, the AUC for
tle Ottawa scale (NOS) checklist [9, 13, 15, 18, 21], four uKIM-1 was found to be 0.85 (Table 2 and Fig. 2).
utilized the quality assessment of diagnostic accuracy The outcomes of the meta-analyses revealed that for
studies QUADS 2 checklist [14, 17, 19, 22], one used NGAL in predicting CIN, the sensitivity, specificity, posi-
the QUADS checklist [16], one utilized Joanna Briggs tive likelihood ratio, and negative likelihood ratio were
Institute (JBI) [8], and one employed both the NOS and 0.78, 0.85, 6.02, and 0.26, respectively. The heterogeneity
Cochrane Collaboration tool [20]. Funding informa- for sensitivity, specificity, positive likelihood ratio, and
tion revealed that four studies received financial sup- negative likelihood ratio was 73.1%, 91.6%, 85.0%, and
port [16, 17, 19, 22], while the remainder did not have 65.5%, respectively. Furthermore, the AUC for NGAL
any financial backing [8, 9, 13–15, 18, 20, 21]. As per the was 0.91 (Table 2 and Fig. 2).
AMSTAR2 checklist, six studies were rated as critically
low quality [15–17, 19, 20, 22], four as low quality [8, 9, Diagnostic adds ratio of biomarkers
13, 21], and two as high quality [14, 18]. Only one study In the preceding meta-analyses, comprehensive evalu-
reported a previously registered protocol in PROSPERO ations of the diagnostic odds ratios for a total of 16
[14], with the remaining 11 studies lacking information biomarkers were conducted. The correlation of two bio-
on a registered protocol [8, 9, 13, 15–22]. Detailed infor- markers, including BUN and PLR, with CIN, was found
mation regarding the included studies is summarized to be statistically insignificant (P > 0.05) (Fig. 3 and
in the Table 1. Detailed information regarding the qual- Table 3). Conversely, the remaining biomarkers exhib-
ity of included studies is presented in Additional file 1: ited a significant association with the occurrence of CIN.
Table S2. Notably, NGAL emerged as the biomarker with the high-
est diagnostic odds ratio for predicting CIN (OR = 31.29,
95% CI 13.72–71.35, P < 0.01, PI: 1.61–605.16), followed
Diagnostic accuracy of biomarkers by Cystatin C and uKIM-1 (OR = 20.07, 95% CI 7.26–
Five meta-analyses assessed the diagnostic accuracy of 55.47, P < 0.01, PI: 0.51–776.18 and OR = 13.51, 95% CI
four biomarkers by reporting the number of true posi- 7.94–22.96, P < 0.01, P: 5.12–35.66, respectively) (Figs. 3
tive, false positive, true negative, and false negative and 4, Table 3). In contrast, hs-CRP (OR = 1.03, 95% CI
cases within original studies (Cystatin-C, BNP, uKIM-1, 1.01–1.05, P = 0.01, PI:0.98–1.08) demonstrated the low-
NGAL) [14, 16, 17, 19, 22]. est diagnostic odds ratio, followed by CRP and HCT
Based on the results of the included meta-analyses, (OR = 1.06, 95% CI 1.00 -1.12, P = 0.02, PI: 0.86–1.31
the pooled sensitivity, specificity, positive like-hood and OR = 0.94, 95% CI 0.91–0.96, P < 0.01, PI: 0.86–1.03,
ratio, and negative hood ratio of serum/plasma Cystatin- respectively) (Figs. 3 and 4, Table 3).
C for predicting CIN were 0.74 and 0.81, 4.35 and 0.25
Table 1 Characteristics of included studies
First author Journal Searched Date of Number of Total Reported Checklist Funding Previous Software Model of Amstar 2
name, databases search included sample size biomarkers used for status registered of analysis analysis score
year of studies quality protocol
publication assessment
of included
original
studies

Zuo [13] Interna‑ PubMed, Embase, April 18, 18 13,084 UA NOS No No STATA​ Random- Low quality
tional Cochrane Library, 2016 effects
Journal and CBM (Chinese model
of Cardiol‑ Biomedical Litera‑
ogy ture database)
Javid [8] Annals PubMed, Scopus, November 4 1346 GGT​ JBI No No CMA Random- Low quality
of medicine and Web of Sci‑ 2022 effects
Mahapatro et al. European Journal of Medical Research

and surgery ence model

Wu [9] Angiology PubMed, EMBASE, June 3, 26 29,454 CRP, hsCRP, NOS No No Revman Random Low quality
Google Scholar, 2020 NLR, PLR, effect model
Clinical Trials, HCT, RDW for hetero‑
and Science Direct genic and fix
effect model
(2024) 29:210

for non-
heterogenic
studies
Wu [14] Medicine PubMed, Embase, March 9, 12 7789 BNP QUADAS-2 No PROSPERO STATA, Random High quality
Cochrane 2022 (CRD42022318275) Review effects
Central Register Manager, model
of Controlled Trials and Meta‑
Library, and Web Disc
of Science
Zhang [15] Angiology PubMed, EMBASE, July 13, 18 16,171 Alb, Uric NOS No No Revman Random Critically Low-
Google Scholar, 2018 acid, BUN effect model quality
Clinical Trials, for hetero‑
and ScienceDirect genic and fix
effect model
for non-
heterogenic
studies
Wang [16] Canadian PubMed, MED‑ November 14 1520 NGAL QUADAS Yes No STATA​ Random Critically Low
Journal LINE, EMBASE, 2014 effect model quality
of Cardiol‑ Web of Science, for hetero‑
ogy ClinicalTrials.gov, genic and fix
Cochrane Library, effect model
and Google for non-
Scholar heterogenic
studies
Page 6 of 19
Table 1 (continued)
First author Journal Searched Date of Number of Total Reported Checklist Funding Previous Software Model of Amstar 2
name, databases search included sample size biomarkers used for status registered of analysis analysis score
year of studies quality protocol
publication assessment
of included
original
studies

Li [17] Journal Medline, Embase, November 9 946 uKIM-1 QUADAS-2 Yes No STATA​ Bivariate Critically Low-
of Inter‑ ClinicalTrials. 31, 2019 mixed- quality
ventional gov, Cochrane effects
Cardiology Library database, regression
and the China
National Knowl‑
edge Infrastruc‑
Mahapatro et al. European Journal of Medical Research

ture (CNKI)
Nie [18] Angiology PubMed, Embase, Novem‑ 6 16,899 Contrast NOS No No STATA​ Random High Quality
and the Cochrane ber, 2020 media effect model
library volume
to creatinine
clearance
(2024) 29:210

ratio
Li [22] Journal PubMed, EMBASE, April 2020 9 2832 BNP QUADAS-2 Yes No STATA​ Bivariate Critically Low
of Inter‑ and the Cochrane random- quality
ventional Central Register effects
Cardiology of Controlled Trials regression
Library
Kewcharoen Cardiovas‑ MEDLINE January 8 10,991 Preproc‑ NOS No No STATA​ Random Critically Low
[20] cular Revas‑ and EMBASE 2020 edural blood and Cochrane effects quality
cularization glucose Collaboration model
Medicine tool
Chen [19] Toxicology PubMed, MED‑ March 10 2554 Cystatin C QUADAS-2 Yes No STATA​ Bivariate Critically Low
Letters LINE, and Embase 2018 meta-anal‑ quality
ysis
Jiang [21] Medicine PUBMED, EMBASE, Septem‑ 6 10,452 PLR NOS No No STATA Random Low quality
and Web of Sci‑ ber 15, and Rev‑ effect model
ence 2018 man for hetero‑
genic and fix
effect model
for non-
heterogenic
studies
Abbreviations: NOS: Newcastle Ottawa scale, JBI: Joanna Briggs Institute, QUADS: Quality Assessment of Diagnostic Accuracy Studies, UA: Uric Acid, GGT: Gamma-glutamyl transpeptidase, CRP: C-Reactive Protein, hsCRP:
Highly Sensitive C-Reactive Protein, NLR: Neutrophil-to-lymphocyte ratio, RDW: red cell distribution width, PLR: Platelet-to-Lymphocyte Ratio, HCT: Hematocrit, Alb: Albumin, BUN: Blood urea nitrogen, NGAL: Neutrophil
gelatinase-associated lipocalin, uKIM-1: Urinary kidney injury molecule-1, BNP: Brain natriuretic peptide
Page 7 of 19
 Mahapatro et al. European Journal of Medical Research

Table 2 Diagnostic accuracy of biomarkers

Biomarker Specificity Sensitivity Positive like hood ratio Negative like hood ratio AUC​ Heterogeneity Heterogeneity Heterogeneity of Heterogeneity of Q
of sensitivity of specificity Positive like hood Negative like hood
ratio ratio
(2024) 29:210

Cystatin-C 0.81 (0.79–0.82) 0.74 (0.69–0.79) 4.35 (2.85- 6.65) 0.25 (0.13–0.50) 0.89 88.7%, P < 0.01 96.4%, P < 0.01 93.7%, P < 0.01 90.3%, P < 0.01 0.82
BNP 0.73 (0.71–0.75) 0.72 (0.67–0.77) 2.85 (2.13–3.82) 0.39 (0.28–0.53) 0.80 56.1%, P = 0.01 94.0%, P < 0.01 89.1%, P < 0.01 70.6%, P < 0.01 0.73
uKIM-1 0.76 (0.73–0.79) 0.84 (0.78–0.88) 3.58 (2.75–4.66) 0.25 (0.17–0.37) 0.85 42.4%, P = 0.08 64.0%, P < 0.01 51.1% P = 0.03 20.3% P = 0.26 0.78
NGAL 0.85 (0.83–0.86) 0.78 (0.73–0.83) 6.02 (3.86–9.40) 0.26 (0.17–0.41) 0.91 73.1%, P < 0.01 91.6%, P < 0.01 85.0%, P < 0.01 65.5%, P < 0.01 0.84
NGAL: Neutrophil gelatinase-associated lipocalin, uKIM-1: Urinary kidney injury molecule-1, BNP: Brain natriuretic peptide
Page 8 of 19
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 9 of 19

Fig. 2 Summary receiver operating characteristics (SROC). A SROC for Cystatin-C, B SROC for BNP, C SROC for uKIM-1, D SROC for NGAL

Results power analysis Discussion

Based on the results of meta-analyses, the following Contrast-induced acute kidney injury (CI-AKI) fre-
biomarkers were associated with low statistical power: quently occurs following percutaneous coronary inter-
BUN (1-β = 0.05), CRP (1-β = 0.15), Hs-CRP (1-β = 0.12), vention (PCI) or coronary angiography (CAG). It leads
NLR (1-β = 0.46), Lower HCT (1-β = 0.33), and PLR to extended hospital stays, elevated healthcare costs,
(1-β = 0.43). In contrast, the prognostic value of the and, in certain instances, heightened cardiovascular
remaining biomarkers was associated with high statistical and renal morbidity and mortality [23]. Despite its sub-
power (Fig. 5 and Table 3). stantial impact, there are limited established strategies
for preventing CI-AKI [24]. Recognizing patients at an
GRADE assessment early stage who are at a high risk of developing CI-AKI
The epidemiological strength of outcomes was rigorously is crucial. While serum creatinine (SCr) concentration is
assessed using the GRADE criteria. Results indicated one widely acknowledged as the conventional clinical meas-
high-quality outcome (BNP), two with moderate qual- ure for identifying and characterizing CI-AKI, it has
ity (Cystatin-C and GGT), nine with low quality (NGAL, several limitations. Notably, it is influenced by factors
uKIM-1, albumin, RDW, PLR, NLR, CRP, BUN, HCT), such as muscle metabolism diet, gender, and hydration,
and five with very low quality (V/CR, preprocedural and its alteration rate following the initial injury is slow
hyperglycemia over 200, preprocedural hyperglycemia [25]. These characteristics render SCr an unreliable and
over 140, uric acid, HS-CRP). The detailed information temporally inadequate biomarker for diagnosing CI-AKI.
regarding GRADE criteria is presented in Table 4. Consequently, its utilization may compromise the effec-
tiveness of treatments and corrective interventions.
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 10 of 19

Fig. 3 Diagnostic odds ratio of biomarkers

Table 3 Results of meta-analysis, heterogeneity, publication bias, power analysis and prediction interval of biomarkers
Variable Odds ratio 95% CI, P value Heterogeneity P vale of publication Power analysis Prediction interval
bias based on eagers
regression test

Hypoalbuminemia 3.09 1.43–6.63, P < 0.01 90%, P < 0.01 P = 0.17 1-b = 1 0.17–53.56
Hyperuricemia 1.32 1.15–1.50, P < 0.01 88%, P < 0.01 P < 0.01 1-b = 0.99 0.86–2.00
BUN 1.06 0.97–1.15, P = 0.18 88%, P < 0.01 P = 0.52 1-b = 0.05 0.38–2.91
CRP 1.06 1.00–1.12, P = 0.04 98%, P < 0.01 P = 0.36 1-b = 0.15 0.86–1.31
HS-CRP 1.03 1.01–1.05, P = 0.01 50%, P = 0.04 P < 0.01 1-b = 0.12 0.98–1.08
NLR 1.10 1.01–1.20, P = 0.01 78%, P < 0.01 P = 0.26 1-b = 0.46 0.83–1.46
RDW 1.34 1.18–1.52, P < 0.01 0%, P = 0.73 P = 0.30 1-b = 0.70 1.18–1.52
Lower HCT 0.94 0.91–0.97, P < 0.01 69%, P < 0.01 P = 0.74 1-b = 0.33 0.86–1.03
PLR 1.11 0.99–1.24, P = 0.06 87%, P < 0.01 P = 0.09 1-b = 0.43 0.70–1.77
GGT​ 3.21 1.26–8.15, P = 0.01 91%, P < 0.01 P = 0.06 1-b = 1 0.03–266.17
BNP 8.93 5.99–13.30, P < 0.01 69%, P < 0.01 P = 0.06 1-b = 1 2.30–34.55
V/CR 2.61 1.93–3.53, P < 0.01 76%, P < 0.01 P < 0.01 1-b = 1 0.98–6.94
UKIM-1 13.51 7.95–22.97, P < 0.01 14%, P = 0.31 P = 0.01 1-b = 1 5.12–35.66
Preprocedural hyperglyce‑ 1.70 1.34–2.15, P < 0.01 26%, P = 0.25 P = 0.45 1-b = 1 0.78–3.68
mia over 140
Preprocedural hyperglyce‑ 2.06 1.87–2.28, P < 0.01 0%, P = 0.90 P = 0.30 1-b = 1 1.87–2.28
mia over 200
Cystatin-C 20.07 7.26–55.46, P < 0.01 88%, P < 0.01 P = 0.11 1-b = 1 0.51–776.18
NGAL 31.29 13.72–71. 34, P < 0.01 73%, P < 0.01 P < 0.01 1-b = 1 1.61–605.16
Gamma-glutamyl transpeptidase, CRP: C-Reactive Protein, hsCRP: Highly Sensitive C-Reactive Protein, NLR: Neutrophil-to-lymphocyte ratio, RDW: red cell distribution
width, PLR: Platelet-to-Lymphocyte Ratio, HCT: Hematocrit, BUN: Blood urea nitrogen, NGAL: Neutrophil gelatinase-associated lipocalin, uKIM-1: Urinary kidney injury
molecule-1, BNP: Brain natriuretic peptide, V/Cr: Contrast volume to creatinine clearance
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 11 of 19

Fig. 4 Results of prediction interval of the biomarkers. A Hypoalbuminemia, B Hyperuricemia, C BUN, D CRP, E hs-CRP, F NLR, G RDW, H Lower HCT,
I PLR, J GGT, K BNP, L V/Cr, M uKIM-1, N Preprocedural hyperglycemia over 140, O Preprocedural hyperglycemia over 200, P Cystatin-C, Q NGAL

Another risk assessment criterion for CIN is the 16, 17, 19, 22]. Specifically, Wu et al. [14] and Li et al.
Mehran risk score, designed to stratify the risk among [22] conducted meta-analyses on BNP, and due to the
patients [26]. While the Mehran risk score has demon- larger overall sample size in Wu et al. (7789 vs. 2832), we
strated efficacy in numerous studies with robust power, opted to report the results from their study in the current
certain variables included in the assessment, such as the umbrella review.
volume of contrast medium and the use of an intra-aortic In our examination, NGAL demonstrated the highest
balloon pump, remain uncertain before the procedure. AUC and positive likelihood ratio among the variables,
This uncertainty poses limitations on the practical clini- followed by Cystatin-C, uKIM-1, and BNP. Conversely,
cal application of the Mehran risk score. Similar chal- uKIM-1 exhibited the least negative likelihood ratio, fol-
lenges and limitations are observed in other predictive lowed by Cystatin-C, NGAL, and BNP.
models for CIN [27, 28]. Upon calculating and comparing the odds ratios of the
The principal aim of the present study was to assess the sixteen biomarkers, NGAL emerged as the biomarker
diagnostic value of biomarkers in predicting CIN by eval- with the highest diagnostic odds ratio for CIN, followed
uating and summarizing the results of previous meta- by Cystatin-C and uKIM-1. Conversely, hs-CRP exhib-
analyses. In addition, an assessment of the quality and ited the lowest diagnostic odds ratio, with CRP and HCT
power of these meta-analyses was conducted, enabling ranking just above.
the strength of the results to be appraised and gaps in the In the meta-analysis conducted by Wang et al. [16]
existing literature to be identified. additional investigations explored the predictive value of
Among the studies included in our analysis, five meta- NGAL. Their findings suggested that evaluating NGAL
analyses reported the raw figures for false positive, false within four hours after CM injection yielded superior
negative, true positive, and true negative cases related to results compared to assessments conducted beyond this
four biomarkers (Cystatin-C, BNP, uKIM-1, NGAL) [14, timeframe, though not statistically significant. Notably,
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 12 of 19

Fig. 5 Results of power analysis of the biomarkers. A Hypoalbuminemia, B Hyperuricemia, C BUN, D CRP, E hs-CRP, F NLR, G RDW, H Lower HCT, I
PLR, J GGT, K BNP, L V/Cr, M uKIM-1, N Preprocedural hyperglycemia over 140, O Preprocedural hyperglycemia over 200, P Cystatin-C, Q NGAL

the study observed no significant difference between (IGFBP7) and Tissue Inhibitor of Metalloproteinases
serum and urine NGAL levels in predicting CIN. Fur- (TIMP), as potential biomarkers for predicting AKI in
thermore, the prognostic efficacy of NGAL levels in pre- patients with sepsis [34]. Additionally, the accuracy of
dicting CI-AKI according to a non-traditional definition NGAL as a biomarker can be compromised by the pres-
was notably higher than its predictive accuracy for CI- ence of comorbid diseases. This further underscores the
AKI according to the traditional definition, although this complexity of relying on NGAL for predicting CIN [35].
was also not statistically significant. NGAL is a factor that serves multiple roles in renal epi-
It is noteworthy to mention that NGAL presents certain thelial cells, including promoting growth, differentiation,
limitations in predicting CIN. Specifically, NGAL may and structural organization. It also plays an important
not serve as a reliable biomarker in patients with chronic role in preventing cell death and preserving the structure
kidney disease (CKD). For example, a prospective, ran- of the renal tubules, which helps to ensure the integrity
domized controlled trial conducted by Ribitsch et al. of the kidneys [36, 37]. Studies have shown that exoge-
which included 617 patients with CKD, demonstrated nous NGAL can have a protective effect on the kidneys in
the limited efficacy of urinary NGAL in predicting CIN mouse models of renal ischemia–reperfusion injury [36].
after angiography [29]. Moreover, the presence of urinary However, CM can be harmful to the kidneys by causing
protein may interfere with the accuracy of NGAL meas- an increase in tubular osmolarity and impairing intracel-
urements, further complicating its use as a biomarker for lular transport and energy metabolism in the tubular epi-
predicting CIN [30]. Furthermore, NGAL measurements thelial cells [38]. As a result, NGAL level increases rapidly
can be influenced by various other factors. In a prospec- after exposure to CM and is considered a critical factor in
tive observational study conducted by Kumar et al. it was predicting the development of CI-AKI [16] (Fig. 6).
discovered that age, estimated Glomerular Filtration Rate In the meta-analysis conducted by Chen et al. [19],
(eGFR), hemoglobin levels, and the volume of contrast the predictive strength of Cystatin-C for CIN was sub-
used are significantly correlated with NGAL levels [31]. jected to further scrutiny. Their findings revealed that
Sepsis and inflammation represent other critical under- measuring Cystatin-C after 24 h following CM injection
lying conditions that can influence NGAL levels [32, demonstrated a higher diagnostic odds ratio compared
33]. Recent studies have introduced additional variables, to assessments conducted within the initial 24 h post-
such as Insulin-Like Growth Factor Binding Protein 7 CM injection. The results of our study align with these
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 13 of 19

Table 4 GRADE assessment of the biomarkers

Quality assessment Quality
No of studies Design Risk of bias Inconsistency Indirectness Imprecision Other
considerations

NGAL
14 Observational Serious No serious incon‑ No serious indi‑ No serious impre‑ Reporting bias ⊕⊕〇〇 low
studies sistency rectness cision very strong
association
Cystatin-C
10 Observational Serious No serious incon‑ No serious indi‑ No serious impre‑ Very strong ⊕⊕⊕〇
studies sistency rectness cision association moderate
uKIM-1
9 Observational Serious No serious incon‑ No serious indi‑ No serious impre‑ Reporting bias ⊕⊕〇〇 low
studies sistency rectness cision very strong
association
BNP
12 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ Very strong ⊕⊕⊕⊕ high
studies of bias sistency rectness cision association
GGT​
4 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ Strong associa‑ ⊕⊕⊕〇
studies of bias sistency rectness cision tion moderate
Albumin
5 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ Reporting bias ⊕⊕〇〇 low
studies of bias sistency rectness cision strong associa‑
tion
V/CR
6 Observational Serious No serious incon‑ No serious indi‑ No serious impre‑ Reporting bias ⊕〇〇〇 very
studies sistency rectness cision strong associa‑ low
tion
Preprocedural hyperglycemia over 200
7 Observational Serious Serious No serious indi‑ No serious impre‑ Strong associa‑ ⊕〇〇〇 very
studies rectness cision tion low
Preprocedural hyperglycemia over 140
4 Observational Serious Serious No serious indi‑ No serious impre‑ None ⊕〇〇〇 very
studies rectness cision low
RDW
5 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ None ⊕⊕〇〇 low
studies of bias sistency rectness cision
Uric acid
18 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ Reporting bias ⊕〇〇〇 very
studies of bias sistency rectness cision low
PLR
4 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ None ⊕⊕〇〇 low
studies of bias sistency rectness cision
NLR
5 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ None ⊕⊕〇〇 low
studies of bias sistency rectness cision
HS-CRP
9 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ Reporting bias ⊕〇〇〇 very
studies of bias sistency rectness cision low
CRP
5 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ None ⊕⊕〇〇 low
studies of bias sistency rectness cision
BUN
3 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ None ⊕⊕〇〇 low
studies of bias sistency rectness cision
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 14 of 19

Table 4 (continued)
Quality assessment Quality
No of studies Design Risk of bias Inconsistency Indirectness Imprecision Other
considerations

HCT
8 Observational No serious risk No serious incon‑ No serious indi‑ No serious impre‑ None ⊕⊕〇〇 low
studies of bias sistency rectness cision

Fig. 6 Biomarkers for prediction of contrast-induced nephropathy

findings, positioning Cystatin-C as one of the top three the glomerulus. It is reabsorbed but not secreted by the
biomarkers for predicting CIN. Importantly, Cystatin-C renal tubules [42–44]. Due to such physiological attrib-
offers distinct advantages over serum creatinine, as it is utes, cystatin C has demonstrated superiority to SCr in
less influenced by patient characteristics such as mus- identifying slight reductions in glomerular filtration rate
cle mass or diet. This characteristic enhances its reli- especially in the setting of renal injuries [45–49] (Fig. 6).
ability as a biomarker compared to creatinine [39, 40]. The outcomes of the meta-analysis conducted by Li
Furthermore, Cystatin-C boasts a shorter half-life than et al. [17] underscored that uKIM-1 exhibits enhanced
creatinine and exhibits an earlier increase in AKI. Con- diagnostic accuracy when measured 24 h after CM injec-
sequently, measuring Cystatin-C enables the earlier diag- tion as opposed to assessments made within the initial
nosis of CIN compared to measuring creatinine [41]. 24 h post-CM injection. This aligns with findings from
These attributes collectively underscore the potential other studies, which have consistently reported the
clinical utility and superiority of Cystatin-C in the early favorable diagnostic accuracy of uKIM-1 in the context of
detection and prediction of CIN. Cystatin C is generated AKI. A comprehensive meta-analysis involving 11 stud-
by nucleated cells and undergoes free filtration through ies and 2979 subjects yielded a diagnostic odds ratio of
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 15 of 19

17.43, with sensitivity and specificity for uKIM-1 in AKI explanations have been suggested. Initially, BNP is pro-
reported at 74% and 86%, respectively [50]. Another duced in cardiomyocytes and is discharged into the
meta-analysis by Ho et al. highlighted the robust diag- bloodstream when there is stretching of the ventricu-
nostic accuracy of uKIM-1 in the context of cardiac lar wall due to elevated pressure or increased volume.
surgery-induced AKI, revealing AUC values of 0.68 for Elevated level of BNP is linked to a decrease in cardiac
intraoperative and 0.72 for postoperative AKI [51]. Func- output, potentially impacting the hemodynamics of the
tionally, KIM-1 serves as a phosphatidylserine receptor renal artery. Secondly, BNP is considered as an indica-
expressed in epithelial cells. In kidney stress conditions, tor of the renin-angiotensin aldosterone system and
uKIM-1 plays a crucial role in identifying and phagocyt- the sympathetic nervous system, suggesting a poten-
izing dead cells [52]. While uKIM-1 is not detectable in tial increase in renal vascular resistance and a reduc-
urine under normal circumstances, its levels significantly tion in renal blood flow. This, in turn, could result in
increase in AKI due to proximal tubule damage [53, 54] renal ischemia and hypoxia [68, 69]. Thirdly, patients
(Fig. 6). with chronic kidney disease, even in the absence of
The result of the current umbrella review underscored cardiovascular abnormalities, exhibit heightened lev-
the significance of hepatorenal status as a prognostic fac- els of BNP. Apart from a reduced extraction rate from
tor for CIN. Notably, individuals with hypoalbuminemia the blood, elevated levels of BNP may signify a decline
face an elevated risk of developing CIN. The connection in functional renal mass and degradation of clearance
between lower serum albumin levels upon admission and receptors. Consequently, individuals with higher lev-
the occurrence of angiographic no-reflow following PCI els of BNP may be more susceptible to the nephrotoxic
has been established. This association not only exacer- effects of contrast medium [70–72] (Fig. 6).
bates renal perfusion impairment but also increases the Based on the result of the current umbrella review,
likelihood of CIN onset [55, 56]. Several other studies high CRP, and hsCRP are significantly associated with
have identified hypoalbuminemia as a risk factor for AKI increased risk of CIN. CRP and hs-CRP are representa-
in diverse clinical settings, including liver transplanta- tive of systemic inflammation and systemic inflammation
tion, rhabdomyolysis, and cardiac surgery [56–60]. The heightens the susceptibility of the kidneys to the local
potential mechanisms linking CIN and serum albumin inflammatory responses triggered by the reabsorption
levels revolve around oxidative stress, inflammation, of contrast medium [73]. Elevated CRP levels are addi-
and endothelial dysfunction, collectively predisposing tionally linked to endothelial injury and compromised
individuals to CIN. Particularly, hypoalbuminemia may vasodilation, potentially resulting in acute renal dam-
compromise the antioxidant defenses against heightened age and the gradual decline of kidney function [74]. In
reactive oxygen species during renal ischemia. Conse- addition, Animal studies further indicate that the down-
quently, this could intensify the vasoconstrictor effects regulation of autophagy is linked to severe ischemia–
of contrast media, ultimately leading to the induction of reperfusion-induced AKI in mice that overexpress CRP,
CIN [61] (Fig. 6). suggesting that CRP renders the kidney more vulnerable
Serum uric acid, a byproduct of purine metabolism, to ischemic/oxidative injury by suppressing autophagy
has garnered significant recognition for its role in diverse flux [75]. Apart from PCI, the preoperative concentra-
pathophysiological processes in recent years. Compelling tion of CRP is a prognostic indicator for the occurrence
evidence indicates that elevated levels of serum uric acid of postoperative AKI in patients undergoing coronary
are linked to an increased risk of developing CKD [62, artery bypass grafting [76]. An increased level of hsCRP
63]. Beyond its direct impact on renal function, hyper- is associated with a heightened risk of AKI and the pro-
uricemia is implicated in the development of CIN [63, 64] gression of chronic kidney disease after myocardial
(Fig. 6). infarction, independent of the patient’s initial renal status
Our investigation revealed that BNP serves as [77] (Fig. 6).
another predictor of CIN. Notably, BNP has been NLR offers a straightforward yet promising assess-
well-established as a biomarker for heart failure, as ment of systemic inflammation and is employed as an
acknowledged in clinical guidelines [65, 66]. Moreover, indicator for cardiovascular diseases [78]. Its reliabil-
a recent retrospective cohort study focusing on elderly ity as an inflammatory prognostic marker persists when
individuals with chronic heart failure and a history of considering coronary heart disease mortality, whether
AKI found that a relative decline in serum BNP levels patients have ST-segment elevation myocardial infarc-
is associated with improved survival outcomes, poten- tion (STEMI), non-ST-segment elevation acute coronary
tially preventing the occurrence of AKI [67]. The com- syndrome, or peripheral artery disease [79–82]. In heart
plete understanding of the mechanisms through which failure patients with reduced left ventricular ejection
BNP predicts CIN remains unclear. Various potential fraction, the NLR level was linked to the progression of
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 16 of 19

renal dysfunction [83]. Additionally, the NLR was pre- drawn from such biomarkers, future investigations
dictive of the deterioration of renal function in diabetic should prioritize larger sample sizes.
patients [84] (Fig. 6). Furthermore, it is important to note that, with the
In our study, lower HCT was significantly associated exception of RDW, blood glucose, and uKIM-1, the
with CIN. Another study involving 6773 patients revealed remaining biomarkers exhibited significant heterogene-
that a lower baseline HCT was an independent predictor ity. This heterogeneity may stem from diverse variables,
of CIN, regardless of the presence of chronic kidney dis- including variations in the demographic characteristics
ease. The study observed a significantly elevated risk of of the study populations, discrepancies in the volume of
CIN with each 3% reduction in baseline HCT [85]. The contrast administered during CAG or PCI, variations in
correlation between HCT and inflammation may be a renal function among patients exposed to CM, differ-
contributing factor to these findings, as indicated by the ences in the types of CM employed across meta-anal-
previous study and our meta-analysis [9, 86] (Fig. 6). yses, and disparities in the definitions of CIN. These
The Cholesterol and Recurrent Events study high- factors underscore the complexity of the relationships
lighted that elevated RDW values may indicate an under- between biomarkers and clinical outcomes, necessitat-
lying inflammatory condition, playing a significant role ing a careful interpretation of the results. Future stud-
in the development of CIN [87]. Moreover, abnormal ies should consider addressing and controlling for these
RDW has been linked to preexisting impaired renal func- sources of heterogeneity to provide more accurate
tion and increased mortality in diverse clinical settings, insights into the clinical implications of the examined
including patients with acute myocardial infarction, biomarkers.
patients experiencing hemodialysis, and recipients of Furthermore, during our appraisal of the included
kidney transplants [88–90] (Fig. 6). studies’ quality, a significant observation was the lack
of study protocol registration in the majority of the
Strengths, limitations, and future suggestions
included meta-analyses. Moreover, a substantial por-
To our knowledge, this study represents a comprehensive tion of these meta-analyses did not adequately address
umbrella review focusing on meta-analyses that investi- the potential impact of low-quality studies on their
gate predictors of CIN in patients following CAG or PCI. reported results. To enhance the robustness of future
Our comprehensive examination involved the evalua- investigations, we recommend the incorporation of
tion of biomarkers identified in previous meta-analyses. subgroup analyses based on the quality of the original
Additionally, we meticulously assessed the quality of studies.
these meta-analyses, identifying gaps in the existing lit- Moreover, a noteworthy observation was the lack of
erature. Furthermore, we evaluated the sufficiency of information regarding the optimal timing for biomarker
the reported results through a rigorous power analysis, measurement in the majority of the included stud-
providing insights into the robustness of the evidence. ies. We strongly encourage future research endeavors
Recognizing the potential impact of heterogeneity on to assess the predictive value of biomarkers at various
the reported findings, we incorporated the calculation of time points following CM injection. This exploration
prediction intervals to offer a more nuanced interpreta- will contribute valuable insights into determining the
tion of the results. most opportune moments for biomarker measure-
Despite these methodological strengths, several limita- ment, enhancing the precision and clinical applicabil-
tions were encountered in the course of this study. Firstly, ity of these predictive markers for contrast-induced
only five meta-analyses provided detailed information nephropathy.
on the crude numbers of false positives, false nega- Abbreviations
tives, true positives, and true negatives among patients. AKI Acute kidney injury
BNP Brain natriuretic peptide
Consequently, our ability to assess metrics such as sen- BUN Blood urea nitrogen
sitivity, specificity, AUC, positive likelihood ratio, and CAG​ Coronary angiography
negative likelihood ratio for the remaining biomarkers CI-AKI Contrast-induced acute Kidney Injury
CIN Contrast-induced nephropathy
was restricted. Secondly, the power analysis underscored CM Contrast media
a limitation within the results, as evidenced by the lower CMA Comprehensive meta-analysis
power observed among the included biomarkers, BUN, CRP C-reactive protein
GGT​ Gamma-glutamyl transpeptidase
CRP, Hs-CRP, NLR, Lower HCT, and PLR. This outcome HCT Hematocrit
warrants caution in interpreting the findings of the afore- hs-CRP Highly sensitive C-reactive protein
mentioned biomarkers, given the associated low power JBI Joanna Briggs Institute
NLR Neutrophil-to-lymphocyte ratio
attributed to these biomarkers due to the relatively small NGAL Neutrophil gelatinase-associated lipocalin
sample size. To enhance the robustness of conclusions
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 17 of 19

NOS Newcastle Ottawa scale References

PCI Percutaneous coronary intervention 1. Killmann SA, Gjorup S, Thaysen JH. Fatal acute renal failure following
PI Prediction interval intravenous pyelography in a patient with multiple myeloma. Acta Med
PLR Platelet-to-lymphocyte ratio Scand. 1957;158:43–6. https://​doi.​org/​10.​1111/j.​0954-​6820.​1957.​tb157​
QUADS Quality assessment of diagnostic accuracy studies 42.x.
RDW Red cell distribution width 2. Bartels ED, Brun GC, Gammeltoft A, GjØRup PA. Acute anuria following
ROC Receiver operating characteristic intravenous pyelography in a patient with myelomatosis. Acta Med
SCr Serum creatinine Scand. 1954;150:297–302. https://​doi.​org/​10.​1111/j.​0954-​6820.​1954.​tb186​
SROC Summary of receiver operating characteristic 32.x.
STEMI ST-segment elevation myocardial infarction 3. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am J
UA Uric Acid Kidney Dis. 2002;39:930–6. https://​doi.​org/​10.​1053/​ajkd.​2002.​32766.
uKIM-1 Urinary kidney injury molecule-1 4. Hou SH, Bushinsky DA, Wish JB, Cohen JJ, Harrington JT. Hospital-
V/cr Contrast media volume to creatinine clearance ratio acquired renal insufficiency: a prospective study. Am J Med. 1983;74:243–
8. https://​doi.​org/​10.​1016/​0002-​9343(83)​90618-6.
5. Mehran R, Nikolsky E. Contrast-induced nephropathy: definition, epide‑
Supplementary Information miology, and patients at risk. Kidney Int Suppl. 2006. https://​doi.​org/​10.​
The online version contains supplementary material available at https://​doi.​ 1038/​sj.​ki.​50003​68.
org/​10.​1186/​s40001-​024-​01782-y. 6. Fukushima Y, et al. Contrast-induced nephropathy (CIN) of patients
with renal dysfunction in CT examination. Jpn J Radiol. 2017;35:427–31.
Additional file 1. Table S1: Search strategy for each international data‑ https://​doi.​org/​10.​1007/​s11604-​017-​0649-4.
base. Table S2: Quality of included studies based on AMSTAR2 appraisal 7. Modi K, Padala SA, Gupta M. StatPearls (StatPearls Publishing Copyright ©
checklist. 2023. London: StatPearls Publishing LLC; 2023.
8. Javid M, et al. Gamma glutamyl transferase as a biomarker to predict
contrast-induced nephropathy among patients with acute coronary
Acknowledgements syndrome undergoing coronary interventions: a meta-analysis. Ann Med
Acknowledgement is given to BioRender for creating the illustrations. Surg (Lond). 2023;85:4033–40. https://​doi.​org/​10.​1097/​ms9.​00000​00000​
000967.
Author contributions 9. Wu X, et al. Inflammatory indicators and hematological indices in con‑
AM, SSN and EAS designed the study. SN and EAS planned the methods to trast-induced nephropathy among patients receiving coronary interven‑
generate the results. MJ, SM collected the data. SN and M-HK designed the tion: a systematic review and meta-analysis. Angiology. 2021;72:867–77.
illustrations. EAS, SH performed the literature search. All authors contributed https://​doi.​org/​10.​1177/​00033​19721​10004​92.
to drafting the manuscript. 10. Shuster JJ. Wiley Online Library; 2011.
11. Page MJ, et al. The PRISMA 2020 statement: an updated guideline for
Funding reporting systematic reviews. BMJ. 2021;372:n71. https://​doi.​org/​10.​1136/​
None. bmj.​n71.
12. Shea BJ, et al. AMSTAR 2: a critical appraisal tool for systematic reviews
Availability of data and materials that include randomised or non-randomised studies of healthcare inter‑
The datasets used and/or analyzed during the current study can be provided ventions, or both. BMJ. 2017;358:j4008. https://​doi.​org/​10.​1136/​bmj.​j4008.
from the corresponding author on reasonable request. 13. Zuo T, et al. Hyperuricemia and contrast-induced acute kidney injury:
a systematic review and meta-analysis. Int J Cardiol. 2016;224:286–94.
https://​doi.​org/​10.​1016/j.​ijcard.​2016.​09.​033.
Declarations 14. Wu X, Yang X, Xu Z, Li J. Brain natriuretic peptide as a biomarker for
predicting contrast-induced nephropathy in patients undergoing coro‑
Ethics approval and consent to participate nary angiography/intervention A systematic review and meta-analysis.
Not applicable. Medicine (United States). 2022. https://​doi.​org/​10.​1097/​MD.​00000​00000​
032432.
Consent for publication 15. Zhang E, et al. Predictive value of hepatorenal status in contrast-induced
Not applicable. nephropathy among patients receiving coronary angiography and/
or intervention: a systematic review and meta-analysis. Angiology.
Competing interests 2019;70:633–41. https://​doi.​org/​10.​1177/​00033​19718​816206.
All authors declare that they have no competing interests in any form. 16. Wang K, et al. Predictive value of neutrophil gelatinase-associated lipoca‑
lin for contrast-induced acute kidney injury after cardiac catheterization:
Author details a meta-analysis. Can J Cardiol. 2016;32(1033):e1019-1033.e1029. https://​
1
Hi-Tech Medical College and Hospital, Rourkela, Odisha, India. 2 Guilan Univer‑ doi.​org/​10.​1016/j.​cjca.​2015.​09.​011.
sity of Medical Sciences, Rasht, Iran. 3 Liaquat University of Medical and Health 17. Li Q, et al. The predictive value of urinary kidney injury molecular 1 for the
Sciences, Jamshoro, Pakistan. 4 University of Buckingham Medical School, diagnosis of contrast-induced acute kidney injury after cardiac catheteri‑
Buckingham, UK. 5 MVJ Medical College and Research Hospital, Bengaluru, zation: a meta-analysis. J Int Cardiol. 2020. https://​doi.​org/​10.​1155/​2020/​
India. 6 Tehran University of Medical Sciences, Tehran, Iran. 7 Kasturba Medical 49829​87.
College, Mangalore, India. 8 Department of Internal Medicine, Yale New Haven 18. Nie Y, Fan L, Song Q, Wu F. Contrast media volume to creatinine clearance
Health Bridgeport Hospital, Bridgeport CT, USA. 9 Gastrointestinal and Liver ratio in predicting nephropathy in patients undergoing percutaneous
Diseases Research Center, Guilan University of Medical Sciences, Rasht, Iran. coronary intervention: a systematic review and meta-analysis. Angiology.
10
Student Research Committee, School of Medicine, Guilan University of Medi‑ 2023;74:545–52. https://​doi.​org/​10.​1177/​00033​19722​11131​43.
cal Sciences, Rasht, Iran. 19. Chen CT, et al. Optimal measuring timing of cystatin C for early detection
of contrast-induced acute kidney injury: a systematic review and meta-
Received: 16 January 2024 Accepted: 10 March 2024 analysis. Toxicol Lett. 2020;318:65–73. https://​doi.​org/​10.​1016/j.​toxlet.​
2019.​10.​011.
20. Kewcharoen J, et al. Pre-procedural hyperglycemia increases the risk of
contrast-induced nephropathy in patients undergoing coronary angiog‑
raphy: a systematic review and meta-analysis. Cardiovasc Revasc Med.
2020;21:1377–85. https://​doi.​org/​10.​1016/j.​carrev.​2020.​04.​040.
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 18 of 19

21. Jiang J, et al. Could platelet-to-lymphocyte ratio be a predictor for con‑ 42. Zhang Z, Lu B, Sheng X, Jin N. Cystatin C in prediction of acute
trast-induced nephropathy in patients with acute coronary syndrome? kidney injury: a systemic review and meta-analysis. Am J Kidney Dis.
A systematic review and meta-analysis. Medicine (United States). 2019. 2011;58:356–65. https://​doi.​org/​10.​1053/j.​ajkd.​2011.​02.​389.
https://​doi.​org/​10.​12097/​MD.​00000​00000​016801. 43. Tan K, Sethi SK. Biomarkers in cardiorenal syndromes. Transl Res.
22. Li X, et al. Brain natriuretic peptide for predicting contrast-induced acute 2014;164:122–34. https://​doi.​org/​10.​1016/j.​trsl.​2014.​04.​011.
kidney injury in patients with acute coronary syndrome undergoing 44. Mårtensson J, et al. Plasma endostatin may improve acute kidney injury
coronary angiography: a systematic review and meta-analysis. J Int risk prediction in critically ill patients. Ann Intens Care. 2016;6:6. https://​
Cardiol. 2020. https://​doi.​org/​10.​1155/​2020/​10350​89. doi.​org/​10.​1186/​s13613-​016-​0108-x.
23. Tsai TT, et al. Contemporary incidence, predictors, and outcomes of acute 45. Li S, et al. Preprocedure and postprocedure predictive values of serum
kidney injury in patients undergoing percutaneous coronary interven‑ β2-microglobulin for contrast-induced nephropathy in patients undergo‑
tions: insights from the NCDR Cath-PCI registry. JACC Cardiovasc Interv. ing coronary computed tomography angiography: a comparison with
2014;7:1–9. https://​doi.​org/​10.​1016/j.​jcin.​2013.​06.​016. creatinine-based parameters and cystatin C. J Comput Assist Tomogr.
24. Sudarsky D, Nikolsky E. Contrast-induced nephropathy in interventional 2015;39:969–74. https://​doi.​org/​10.​1097/​rct.​00000​00000​000294.
cardiology. Int J Nephrol Renovasc Dis. 2011;4:85–99. https://​doi.​org/​10.​ 46. Alharazy SM, et al. Serum neutrophil gelatinase-associated lipocalin and
2147/​ijnrd.​S21393. cystatin C are early biomarkers of contrast-induced nephropathy after
25. Wasung ME, Chawla LS, Madero M. Biomarkers of renal function, which coronary angiography in patients with chronic kidney disease. Angiology.
and when? Clin Chim Acta. 2015;438:350–7. https://​doi.​org/​10.​1016/j.​cca.​ 2014;65:436–42. https://​doi.​org/​10.​1177/​00033​19713​483918.
2014.​08.​039. 47. Mårtensson J, et al. Plasma endostatin may improve acute kidney injury
26. Mehran R, et al. A simple risk score for prediction of contrast-induced risk prediction in critically ill patients. Ann Intens Care. 2016;6:1–9.
nephropathy after percutaneous coronary intervention: development 48. Dharnidharka VR, Kwon C, Stevens G. Serum cystatin C is superior to
and initial validation. J Am Coll Cardiol. 2004;44:1393–9. https://​doi.​org/​ serum creatinine as a marker of kidney function: a meta-analysis. Am J
10.​1016/j.​jacc.​2004.​06.​068. Kidney Dis. 2002;40:221–6.
27. Koowattanatianchai S, Chantadansuwan T, Kaladee A, Phinyo P, Patu‑ 49. Mussap M, et al. Cystatin C is a more sensitive marker than creati‑
manond J. Practical risk stratification score for prediction of contrast- nine for the estimation of GFR in type 2 diabetic patients. Kidney Int.
induced nephropathy after primary percutaneous coronary intervention 2002;61:1453–61.
in patients with acute ST-segment elevation myocardial infarction. 50. Shao X, et al. Diagnostic value of urinary kidney injury molecule 1 for
Cardiol Res. 2019;10:350–7. https://​doi.​org/​10.​14740/​cr939. acute kidney injury: a meta-analysis. PLoS ONE. 2014;9:e84131. https://​
28. Chen YL, et al. A simple preprocedural score for risk of contrast-induced doi.​org/​10.​1371/​journ​al.​pone.​00841​31.
acute kidney injury after percutaneous coronary intervention. Catheter 51. Ho J, et al. Urinary, plasma, and serum biomarkers’ utility for predicting
Cardiovasc Interv. 2014;83:E8-16. https://​doi.​org/​10.​1002/​ccd.​25109. acute kidney injury associated with cardiac surgery in adults: a meta-
29. Ribitsch W, et al. Neutrophil gelatinase-associated lipocalin (NGAL) fails analysis. Am J Kidney Dis. 2015;66:993–1005. https://​doi.​org/​10.​1053/j.​
as an early predictor of contrast induced nephropathy in chronic kidney ajkd.​2015.​06.​018.
disease (ANTI-CI-AKI study). Sci Rep. 2017;7:41300. https://​doi.​org/​10.​ 52. Ichimura T, et al. Kidney injury molecule-1 is a phosphatidylserine recep‑
1038/​srep4​1300. tor that confers a phagocytic phenotype on epithelial cells. J Clin Invest.
30. Wagener G, et al. Urinary neutrophil gelatinase-associated lipocalin and 2008;118:1657–68. https://​doi.​org/​10.​1172/​jci34​487.
acute kidney injury after cardiac surgery. Am J Kidney Dis. 2008;52:425– 53. Bonventre JV. Kidney injury molecule-1 (KIM-1): a urinary biomarker and
33. https://​doi.​org/​10.​1053/j.​ajkd.​2008.​05.​018. much more. Nephrol Dial Transplant. 2009;24:3265–8. https://​doi.​org/​10.​
31. Sahu AK, et al. Neutrophil gelatinase-associated lipocalin as a marker for 1093/​ndt/​gfp010.
contrast-induced nephropathy in patients undergoing percutaneous 54. Ichimura T, et al. Kidney injury molecule-1 (KIM-1), a putative epithelial
coronary intervention: a prospective observational analysis. Indian Jo cell adhesion molecule containing a novel immunoglobulin domain, is
Nephrol. 2022;32:1. up-regulated in renal cells after injury. J Biol Chem. 1998;273:4135–42.
32. Chen JJ, Lee TH, Lee CC, Chang CH. Using lipocalin as a prognostic 55. Kurtul A, et al. Serum albumin levels on admission are associated with
biomarker in acute kidney injury. Expert Rev Mol Diagn. 2021;21:455–64. angiographic no-reflow after primary percutaneous coronary interven‑
https://​doi.​org/​10.​1080/​14737​159.​2021.​19173​84. tion in patients with ST-segment elevation myocardial infarction. Angiol‑
33. Romejko K, Markowska M, Niemczyk S. The review of current knowledge ogy. 2015;66:278–85. https://​doi.​org/​10.​1177/​00033​19714​526035.
on neutrophil gelatinase-associated lipocalin (NGAL). Int J Mol Sci. 2023. 56. Oduncu V, et al. The prognostic value of serum albumin levels on admis‑
https://​doi.​org/​10.​3390/​ijms2​41310​470. sion in patients with acute ST-segment elevation myocardial infarction
34. Kellum JA, et al. Use of biomarkers to identify acute kidney injury to help undergoing a primary percutaneous coronary intervention. Coron Artery
detect sepsis in patients with infection. Crit Care Med. 2021;49:e360–8. Dis. 2013;24:88–94. https://​doi.​org/​10.​1097/​MCA.​0b013​e3283​5c46fd.
https://​doi.​org/​10.​1097/​ccm.​00000​00000​004845. 57. Hinson JS, et al. Risk of acute kidney injury after intravenous contrast
35. Kashani K, et al. Discovery and validation of cell cycle arrest biomarkers in media administration. Ann Emerg Med. 2017;69:577-586.e574. https://​
human acute kidney injury. Crit Care. 2013;17:1–12. doi.​org/​10.​1016/j.​annem​ergmed.​2016.​11.​021.
36. Mori K, et al. Endocytic delivery of lipocalin-siderophore-iron complex 58. Sang BH, Bang JY, Song JG, Hwang GS. Hypoalbuminemia within two
rescues the kidney from ischemia-reperfusion injury. J Clin Invest. postoperative days is an independent risk factor for acute kidney injury
2005;115:610–21. https://​doi.​org/​10.​1172/​jci23​056. following living donor liver transplantation: a propensity score analysis
37. Schmidt-Ott KM, et al. Dual action of neutrophil gelatinase-associated of 998 consecutive patients. Crit Care Med. 2015;43:2552–61. https://​doi.​
lipocalin. J Am Soc Nephrol. 2007;18:407–13. https://​doi.​org/​10.​1681/​asn.​ org/​10.​1097/​ccm.​00000​00000​001279.
20060​80882. 59. Rodríguez E, et al. Risk factors for acute kidney injury in severe rhabdo‑
38. Perrin T, Descombes E, Cook S. Contrast-induced nephropathy in invasive myolysis. PLoS ONE. 2013;8:e82992. https://​doi.​org/​10.​1371/​journ​al.​pone.​
cardiology. Swiss Med Wkly. 2012;142:w13608. https://​doi.​org/​10.​4414/​ 00829​92.
smw.​2012.​13608. 60. Lee EH, et al. Preoperative hypoalbuminemia is a major risk factor
39. Tangri N, et al. Changes in dietary protein intake has no effect on serum for acute kidney injury following off-pump coronary artery bypass
cystatin C levels independent of the glomerular filtration rate. Kidney Int. surgery. Intensive Care Med. 2012;38:1478–86. https://​doi.​org/​10.​1007/​
2011;79:471–7. https://​doi.​org/​10.​1038/​ki.​2010.​431. s00134-​012-​2599-8.
40. Vinge E, Lindergård B, Nilsson-Ehle P, Grubb A. Relationships among 61. Börekçi A, et al. Oxidative stress and paraoxonase 1 activity predict
serum cystatin C, serum creatinine, lean tissue mass and glomerular contrast-induced nephropathy in patients with ST-segment elevation
filtration rate in healthy adults. Scand J Clin Lab Invest. 1999;59:587–92. myocardial infarction undergoing primary percutaneous coronary inter‑
https://​doi.​org/​10.​1080/​00365​51995​01850​76. vention. Angiology. 2015;66:339–45.
41. Rickli H, et al. Time course of serial cystatin C levels in comparison with 62. Nough H, et al. Incidence and main determinants of contrast-induced
serum creatinine after application of radiocontrast media. Clin Nephrol. nephropathy following coronary angiography or subsequent balloon
2004;61:98–102. https://​doi.​org/​10.​5414/​cnp61​098.
Mahapatro et al. European Journal of Medical Research (2024) 29:210 Page 19 of 19

angioplasty. Cardiorenal Med. 2013;3:128–35. https://​doi.​org/​10.​1159/​ coronary intervention. Heart Lung Circ. 2016;25:683–90. https://​doi.​org/​
00035​1981. 10.​1016/j.​hlc.​2016.​01.​007.
63. Takir M, et al. Lowering uric acid with allopurinol improves insulin 82. Vuruşkan E, Saraçoğlu E. Bilirubin levels are associated with contrast-
resistance and systemic inflammation in asymptomatic hyperuricemia. induced nephropathy in peripheral artery disease. Angiology.
J Investig Med. 2015;63:924–9. https://​doi.​org/​10.​1097/​jim.​00000​00000​ 2017;68:728–33. https://​doi.​org/​10.​1177/​00033​19716​679340.
000242. 83. Argan O, et al. Associations between neutrophil gelatinase associated
64. Ejaz AA, et al. The role of uric acid in acute kidney injury. Nephron. lipocalin, neutrophil-to-lymphocyte ratio, atrial fibrillation and renal dys‑
2019;142:275–83. https://​doi.​org/​10.​1159/​00049​9939. function in chronic heart failure. Med Sci Monit. 2016;22:4765–72. https://​
65. McDonagh TA, et al. 2021 ESC guidelines for the diagnosis and treatment doi.​org/​10.​12659/​msm.​898608.
of acute and chronic heart failure. Eur Heart J. 2021;42:3599–726. https://​ 84. Azab B, et al. Neutrophil-to-lymphocyte ratio as a predictor of worsen‑
doi.​org/​10.​1093/​eurhe​artj/​ehab3​68. ing renal function in diabetic patients (3-year follow-up study). Ren Fail.
66. Heidenreich PA, et al. 2022 AHA/ACC/HFSA guideline for the manage‑ 2012;34:571–6. https://​doi.​org/​10.​3109/​08860​22x.​2012.​668741.
ment of heart failure: executive summary: a report of the American 85. Nikolsky E, et al. Low hematocrit predicts contrast-induced nephropathy
college of cardiology/American Heart Association Joint Committee on after percutaneous coronary interventions. Kidney Int. 2005;67:706–13.
Clinical Practice Guidelines. Circulation. 2022;145:e876–94. https://​doi.​ https://​doi.​org/​10.​1111/j.​1523-​1755.​2005.​67131.x.
org/​10.​1161/​cir.​00000​00000​001062. 86. Yang M, et al. Hematocrit level could reflect inflammatory response and
67. Hou J, et al. Association of serum N-terminal pro-brain natriuretic disease activity in patients with systemic lupus erythematosus. Clin Lab.
peptide levels with survival and renal outcomes among elderly patients 2015;61:801–7. https://​doi.​org/​10.​7754/​clin.​lab.​2015.​141246.
with acute kidney injury in chronic heart failure. Front Cardiovasc Med. 87. Tonelli M, et al. Relation between red blood cell distribution width and
2023;10:1104787. cardiovascular event rate in people with coronary disease. Circulation.
68. Wu X, Yang X, Xu Z, Li J. Brain natriuretic peptide as a biomarker for 2008;117:163–8. https://​doi.​org/​10.​1161/​circu​latio​naha.​107.​727545.
predicting contrast-induced nephropathy in patients undergoing coro‑ 88. Akin F, et al. Relation between red cell distribution width and severity
nary angiography/intervention: a systematic review and meta-analysis. of coronary artery disease in patients with acute myocardial infarction.
Medicine (Baltimore). 2022;101:32432. https://​doi.​org/​10.​1097/​md.​00000​ Angiology. 2013;64:592–6. https://​doi.​org/​10.​1177/​00033​19712​461931.
00000​032432. 89. Ujszaszi A, Molnar MZ, Czira ME, Novak M, Mucsi I. Renal function is inde‑
69. Volpe M, Carnovali M, Mastromarino V. The natriuretic peptides system in pendently associated with red cell distribution width in kidney transplant
the pathophysiology of heart failure: from molecular basis to treatment. recipients: a potential new auxiliary parameter for the clinical evaluation
Clin Sci (Lond). 2016;130:57–77. https://​doi.​org/​10.​1042/​cs201​50469. of patients with chronic kidney disease. Br J Haematol. 2013;161:715–25.
70. Tagore R, et al. Natriuretic peptides in chronic kidney disease. Clin J Am https://​doi.​org/​10.​1111/​bjh.​12315.
Soc Nephrol. 2008;3:1644–51. https://​doi.​org/​10.​2215/​cjn.​00850​208. 90. Vashistha T, et al. Red cell distribution width and mortality in hemodialy‑
71. Anwaruddin S, et al. Renal function, congestive heart failure, and amino- sis patients. Am J Kidney Dis. 2016;68:110–21. https://​doi.​org/​10.​1053/j.​
terminal pro-brain natriuretic peptide measurement: results from the ajkd.​2015.​11.​020.
ProBNP Investigation of Dyspnea in the Emergency Department (PRIDE)
Study. J Am Coll Cardiol. 2006;47:91–7. https://​doi.​org/​10.​1016/j.​jacc.​2005.​
08.​051. Publisher’s Note
72. Santos-Araújo C, Leite-Moreira A, Pestana M. Clinical value of natriuretic Springer Nature remains neutral with regard to jurisdictional claims in pub‑
peptides in chronic kidney disease. Nefrologia. 2015;35:227–33. https://​ lished maps and institutional affiliations.
doi.​org/​10.​1016/j.​nefro.​2015.​03.​002.
73. Ridker PM. Clinical application of C-reactive protein for cardiovascular
disease detection and prevention. Circulation. 2003;107:363–9. https://​
doi.​org/​10.​1161/​01.​cir.​00000​53730.​47739.​3c.
74. Tonelli M, Sacks F, Pfeffer M, Jhangri GS, Curhan G. Biomarkers of
inflammation and progression of chronic kidney disease. Kidney Int.
2005;68:237–45. https://​doi.​org/​10.​1111/j.​1523-​1755.​2005.​00398.x.
75. Bian A, et al. Downregulation of autophagy is associated with severe
ischemia-reperfusion-induced acute kidney injury in overexpressing
C-reactive protein mice. PLoS ONE. 2017;12:e0181848. https://​doi.​org/​10.​
1371/​journ​al.​pone.​01818​48.
76. Han SS, et al. C-reactive protein predicts acute kidney injury and death
after coronary artery bypass grafting. Ann Thorac Surg. 2017;104:804–10.
https://​doi.​org/​10.​1016/j.​athor​acsur.​2017.​01.​075.
77. Fu EL, et al. High-sensitivity C-reactive protein and the risk of chronic
kidney disease progression or acute kidney injury in post-myocardial
infarction patients. Am Heart J. 2019;216:20–9. https://​doi.​org/​10.​1016/j.​
ahj.​2019.​06.​019.
78. Shah N, et al. Neutrophil lymphocyte ratio significantly improves the
Framingham risk score in prediction of coronary heart disease mortality:
insights from the National Health and Nutrition Examination Survey-III. Int
J Cardiol. 2014;171:390–7. https://​doi.​org/​10.​1016/j.​ijcard.​2013.​12.​019.
79. Kaya A, et al. Neutrophil-to-lymphocyte ratio predicts contrast-induced
nephropathy in patients undergoing primary percutaneous coronary
intervention. Angiology. 2014;65:51–6. https://​doi.​org/​10.​1177/​00033​
19713​484789.
80. Sun XP, et al. Platelet to lymphocyte ratio predicts contrast-induced
nephropathy in patients with ST-segment elevation myocardial infarction
undergoing primary percutaneous coronary intervention. Angiology.
2018;69:71–8. https://​doi.​org/​10.​1177/​00033​19717​707410.
81. Kurtul A, Yarlioglues M, Duran M, Murat SN. Association of neutrophil-to-
lymphocyte ratio with contrast-induced nephropathy in patients with
non-ST-elevation acute coronary syndrome treated with percutaneous