†These authors contributed equally.
Acute kidney injury following cardiac surgery (CS-AKI) represents a severe postoperative complication, negatively impacting short-term and long-term mortality. Due to the lack of a specific treatment, effective prevention remains the most powerful tool to overcome the CS-AKI burden. Improving the preventive strategies is possible by establishing appropriate preoperative risk profiles. Various clinical models were proposed as a means to assist physicians in stratifying the risk of CS-AKI. However, these models are used for predicting severe forms of CS-AKI, while their predictive power for mild forms is insufficient. Our paper represents the first systematic approach to review all proposed preoperative risk factors and their predictive power. Our strategy is the starting point for selecting and comparing the predictive elements to be integrated into future risk models. Heart failure, chronic hyperglycemia, anemia, obesity, preoperative exposure to nephrotoxic drugs or contrast media, inflammation, proteinuria, and pre-existing kidney disease were systematically reviewed and were found to be associated with an increased risk of postoperative CS-AKI. As no externally validated and universally accepted risk models currently exist, the clinical judgment and a good knowledge of the preoperative risk factors in the light of new evidence may help personalize preoperative risk profiles as the cornerstone of prevention measures.
Acute kidney injury (AKI) following open-heart surgery represents a severe
postoperative complication, negatively impacting short-term and long-term
mortality. Given the large number of patients undergoing cardiovascular surgery
(CS), postoperative AKI incidence is high, ranging from 20% to 70% of cases
(Vives et al., 2020). Severe forms of CS-AKI necessitating renal replacement
therapy (RRT) have a lower prevalence (
Several important mechanisms are in charge for CS-AKI development: renal hypoperfusion (heart failure, cardiopulmonary bypass, hemorrhage events during surgery), inflammation and oxidative stress (tissue damage, blood passage through cardiopulmonary bypass circuit), use of nephrotoxic drugs and agents (angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers (ARBs), nonsteroidal anti-inflammatory drugs (NSAIDs), exposure to contrast agents), embolic factors and genetic predisposition (Mariscalco et al., 2011; Tecson et al., 2018; Wang and Bellomo, 2017).
Various clinical models were proposed as a means to assist physicians in stratifying the risk of CS-AKI: Cleveland and modified Cleveland scores, Mehta score, and Simplified Renal Index (SRI) score (Huen and Parikh, 2012; O’Neal et al., 2016; Vives et al., 2020). However, these scores are used for predicting severe forms of CS-AKI, while their predictive power for mild forms is poor. There is an imperious need for improved strategies regarding the prevention, diagnosis, and treatment of AKI related to open-heart surgery. Early recognition of AKI plays an important role, since the cornerstone of CS-AKI management is represented by preventive measures, as there is no specific treatment.
The objective of this updated systematic review is to weigh the impact of all reported risk factors in the development of CS-AKI and pave the way to more robust prevention strategies.
Our systematic review was conducted according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) checklist (Liberati et al., 2009).
We performed a systematic search in PubMed, ScienceDirect and Cochrane databases
for articles published from January 2005 until October 2020 using the following
query terms: “Preoperative risk factors”, “Cardiac surgery related acute
kidney injury”, “Heart failure”, “BNP”, “NT-proBNP”, “hs-cTnT”,
“cTnI”, “CK-MB”, “Type 1 diabetes mellitus”, “Type 2 diabetes mellitus”,
“Hemoglobin A
Studies involving
Data extracted from each included study were: authors’ last name, year of publication, type of study, the preoperative risk factors being evaluated, number of patients, age median/mean, the proportion of patients exposed to a particular risk factor (when available), type of surgical intervention, criteria used to define CS-AKI, number of CS-AKI events. When available, data are presented as percentages, mean/median values, risk ratio (RR), odds ratio (OR), and range of variation. Also, when reported, the ability of a particular preoperative risk factor to improve the prediction power of a clinical model was documented.
We assessed the risk of developing CS-AKI in relation to the various preoperative risk factors reported in the included trials.
The quality of the non-randomized included trials was assessed using the Newcastle-Ottawa scale for cohort studies concerning three criteria (study group selection, the comparability of the study groups, and the evaluation of the outcome of interest) (Stang, 2010). The risk of bias for randomized trials was appraised using the revised Cochrane Risk of Bias tool (RoB2) (Sterne et al., 2019).
Flow diagram representing the selection process of papers assessing the risk of developing acute kidney injury following cardiac surgery in relation to the various preoperative risk factors. The database search retrieved 841 papers. After removing titles and abstracts irrelevant to the topic, duplicates, meta-analysis, and studies that did not meet the inclusion criteria, 23 papers were included for the systematic review.
Known preoperative risk factors for CS-AKI grouped into six categories. (EF - ejection fraction; ACEIs - angiotensin-converting enzyme inhibitors; ARBs - angiotensin receptor blockers; hs-CRP - High-sensitivity C-reactive protein; CKD - chronic kidney disease; COPD - chronic obstructive pulmonary disease).
Our database search retrieved 841 results. The study selection process is illustrated in Fig. 1. After removing titles and abstracts irrelevant to the topic (n = 756), duplicates (n = 5), meta-analysis (n = 4), and studies that did not meet the inclusion criteria (n = 53), 23 papers were included for the systematic review.
Study design, patients’ characteristics, the number of CS-AKI events and the AKI definition for each included study are summarized in Table 1. Twenty non-randomized cohort studies (Barodka et al., 2011; Belley-Côté et al., 2016; Borde et al., 2019; Coca et al., 2012, 2013; De Santo et al., 2009; Duchnowski et al., 2020; Han et al., 2017; Hertzberg et al., 2015, 2019; Huang et al., 2011; Jiang et al., 2018; Karkouti et al., 2011; Kocogulları et al., 2017; Oezkur et al., 2015; O’Sullivan et al., 2015; Özkaynak et al., 2014; Patel et al., 2012; Ranucci et al., 2013; Shlipak et al., 2011; Wyler von Ballmoos et al., 2018; Zou et al., 2019) and one randomized trial (van Diepen et al., 2018) were included. Fourteen studies were single center: Sweden (Hertzberg et al., 2015, 2019), Turkey (Kocogulları et al., 2017; Özkaynak et al., 2014), Germany (Oezkur et al., 2015), Italy (De Santo et al., 2009; Ranucci et al., 2013), Canada (Karkouti et al., 2011; van Diepen et al., 2018), Ireland (O’Sullivan et al., 2015), USA (Barodka et al., 2011), China (Jiang et al., 2018), India (Borde et al., 2019), Taiwan (Huang et al., 2011) and seven studies were multi-centric (Belley-Côté et al., 2016; Coca et al., 2012, 2013; Han et al., 2017; Patel et al., 2012; Shlipak et al., 2011; Wyler von Ballmoos et al., 2018).
The evaluated preoperative risk factors (Fig. 2) and their relation with CS-AKI varied across studies: three studies assessed hemodynamic risk factors (Belley-Côté et al., 2016; Hertzberg et al., 2019; Patel et al., 2012), six papers evaluated metabolic risk factors (De Santo et al., 2009; Hertzberg et al., 2015; Karkouti et al., 2011; Kocogulları et al., 2017; Oezkur et al., 2015; O’Sullivan et al., 2015), seven studies analyzed nephrotoxic drugs and agents (Barodka et al., 2011; Borde et al., 2019; Coca et al., 2013; Jiang et al., 2018; Özkaynak et al., 2014; Ranucci et al., 2013; van Diepen et al., 2018), one paper studied inflammatory risk factors (Han et al., 2017), and four papers examined other comorbidities (Coca et al., 2012; Huang et al., 2011; Shlipak et al., 2011; Wyler von Ballmoos et al., 2018). The AKI definitions differed between studies, KDIGO and AKIN being the most used ones.
Table 2 summarizes the impact (expressed by OR, RR, and confidence intervals - CI) of the different risk factors on the development of CS-AKI as estimated for each study.
Patients with preoperative heart failure or elevated biomarkers of hemodynamic stress (B-type natriuretic peptide - BNP and its precursor, N-terminal pro-B-type natriuretic peptide - NT-proBNP) presented an increased risk of postoperative CS-AKI (Belley-Côté et al., 2016; Hertzberg et al., 2019; Patel et al., 2012). Belley-Côté et al. (2016) showed that other hemodynamic biomarkers (hs-cTnT, cTnI, and CK-MB) were not associated with an increased risk of CS-AKI after multivariate adjustments.
Regarding the metabolic risk factors, Hertzberg et al. (2015) found that
both type 1 and type 2 diabetes mellitus were associated with an increased risk
of postoperative CS-AKI independently of pre-existing kidney disease. CS-AKI risk
was higher in type 1 diabetes mellitus (OR 4.89, 95% CI 3.82-6.25) than type 2
diabetes mellitus (OR 1.27, 95% CI 1.16-1.40). A high preoperative level of
hemoglobin A
Preoperative anemia is significantly associated with CS-AKI, according to
De Santo et al. (2009) and Karkouti et al. (2011). Besides, Karkouti et al. (2011) showed that erythrocyte transfusion on the day of surgery also
correlated with higher postoperative AKI rates. Moreover, the correlative effect
had a greater magnitude in anemic patients than those without anemia
(P-value for trend
Data regarding preoperative treatment with ACEIs or ARBs is discordant. In a randomized trial, van Diepen et al. (2018) reported no difference in CS-AKI incidence between patients who continued the treatment with ACEI/ARB and those in whom treatment was discontinued two days before surgery (P = 0.991). Moreover, continued treatment with ACEI was not associated with higher intensive care unit length of stay (P = 0.420) or increased mechanical ventilation duration (P = 0.680). However, Coca et al. (2013) reported a higher risk of postoperative CS-AKI in patients who continued treatment with ACEI/ARB than those without a history of ACEI/ARB administration. Nonetheless, only patients exhibiting a high CS-AKI risk were enrolled, limiting the study’s relevance regarding lower-risk patients. Unexpectedly, Barodka et al. (2011) reported a lower risk of CS-AKI in patients treated with ACEI/ARB. However, the small number of CS-AKI events (n = 19, 5.5%) is an essential limitation of the study.
Two studies (Jiang et al., 2018; Ranucci et al., 2013) assessing the impact
of cardiac catheterization within seven days before surgery and coronary
angiography on the same day as surgery reported an increased risk of
postoperative CS-AKI. Moreover, Jiang et al. (2018) described an
increased CS-AKI risk for a dose of contrast media (CM)
High-sensitivity C-reactive protein (hs-CRP), a marker of inflammation, was associated with CS-AKI. Han et al. (2017) reported that patients with higher hs-CRP levels exhibited a greater risk of CS-AKI after multivariate adjustments.
Pre-existing kidney disease represents an inherent risk factor for postoperative CS-AKI. Shlipak et al. (2011) reported that cystatin C had a stronger correlation with the risk of CS-AKI (OR 4.8, 95% CI 2.9-7.7) compared to creatinine or estimated glomerular filtration rate (eGFR). Besides, only the worst kidney function quintile was independently linked to a greater risk of severe CS-AKI (OR 3.6, 95% CI 1.3-10.1 for cystatin C and OR 2.0, 95% CI 0.9-4.5 for creatinine, respectively). However, no significant correlation between eGFR quintile categories and severe CS-AKI was identified.
Galectin-3 is used as a cardiac biomarker in chronic kidney disease patients. Wyler von Ballmoos et al. (2018) reported that patients with higher levels of Galectin-3 had an approximately three-fold higher risk of stage 2 or 3 CS-AKI and a two-fold more significant risk of stage 1 CS-AKI. Preoperative urine albumin to creatinine ratio (UACR) and dipstick proteinuria was associated with an increased risk of postoperative CS-AKI, as reported by Coca et al. (2012) and Huang et al. (2011). However, most of the patients included in the former study experienced only mild CS-AKI, limiting the ability to assess a relationship between proteinuria and more severe CS-AKI. Notably, the latter study reported that more severe proteinuria had a higher risk of CS-AKI necessitating RRT (OR 7.29, 95% CI 3.00-17.73, P = 0.004).
Author, year | Type of study | RF evaluated | Patients, No | Patients with RF, No (%) | Type of surgical intervention | Age median/ mean | Acute kidney injury | |
Definition used | Events, No (%) | |||||||
Hemodynamic risk factors | ||||||||
(1) (Hertzberg et al., 2019) | Cohort study | Heart failure | 36403 | 3914 (11) | First isolated CABG | 67 | KDIGO | 4432 (14) no RF |
1000 (26) with RF | ||||||||
(2) (Patel et al., 2012) | Cohort study | BNP levels | 1139 | Stratified in quintiles | CABG only, valve only, or both | 72 | AKIN | 465 (41) |
(3) (Belley-Côté et al., 2016) | Cohort study | NT-proBNP hs-cTnT cTnI CK-MB | 960 | Stratified in tertiles | CABG only, valve only, or both | 71.53 (no AKI) | Severe AKI - doubling in SCr from baseline, or RRT | 37 (3.9) - severe AKI |
70.7 (severe AKI) | ||||||||
Metabolic risk factors | ||||||||
(4) (Hertzberg et al., 2015) | Cohort study | Diabetes mellitus | 36106 | 5581 (15.3) | Primary nonemergent isolated CABG | 67.4 | AKIN | 5199 (14) |
(5) (Kocogulları et al., 2017) | Cohort study | Hemoglobin A |
202 | 90 (44.5) - with HbA |
Isolated CABG | 63.0 (HbA |
KDIGO | 19 (9.4) |
60.0 (HbA | ||||||||
(6) (Oezkur et al., 2015) | Cohort study | Chronic hyperglycemia | 307 | 165 (53.7) | Isolated CABG or combination procedures | 69 | KDIGO | 148 (48.2) |
(7) (De Santo et al., 2009) | Cohort study | Anemia | 1047 | 320 (28) | Isolated CABG | 63.2 | RIFLE | 42 (4) |
(8) (Karkouti et al., 2011) | Cohort study | Anemia | 12388 | 2287 (18.4) | Isolated CABG, valve surgery or other procedures | 66 (anemia) | RIFLE | 256 (2) |
62 (no anemia) | ||||||||
(9) (O’Sullivan et al., 2015) | Cohort study | Obesity | 432 | 128 (29.6) | Isolated CABG, CABG and concomitant procedure, valve surgery, pericardiectomy, Redo procedure, subaortic membrane resection | 65.29 (obesity) | AKIN and KDIGO | 57 (13.2) |
66.37 (no obesity) | ||||||||
Use of nephrotoxic drugs and agents | ||||||||
(10) (Coca et al., 2013) | Cohort study | ACEi/ARB | 1594 | 231 (14.5) with continued ACEi/ARB | CABG and/or valve surgery | 73 (no ACEi/ ARB) | 543 (34.7) | |
70 (with ACEi/ ARB) | ||||||||
(11) (Barodka et al., 2011) | Cohort study | ACEi/ARB | 346 | 122 (35.26) | CABG or valve surgery | 74 | 19 (5.5) | |
(21) (van Diepen et al., 2018) | Randomized study | ACEi/ARB | 126 | 60 (47.6) with continued ACEi/ARB | CABG and/or valve surgery | 67 (with ACEi/ ARB) | 2 (1.6) | |
64 (no ACEi/ ARB) | ||||||||
(12) (Jiang et al., 2018) | Cohort study | Cardiac catheterization | 1069 | 888 (83.1) surgery |
Isolated CABG, valve surgery or both | 61.3 surgery |
KDIGO | 412 (38.5) |
181 (16.9) surgery |
63.9 surgery | |||||||
(13) (Ranucci et al., 2013) | Cohort study | Coronary angiography | 4440 | 552 (12.4) | Isolated CABG, valve surgery, CABG and other procedures | 65.8 (no AKI) | AKIN | 961 (21.7) |
69.8 (with AKI) | ||||||||
(14) (Borde et al., 2019) | Cohort study | Coronary angiography | 900 | 210 (23.3) surgery |
Isolated CABG | 60 | KDIGO | 214 (24) |
690 (76, 7) surgery | ||||||||
(15) (Özkaynak et al., 2014) | Cohort study | Cardiac catheterization | 573 | Stratified in groups | CABG and/or valve surgery or other procedures | 59.3 | AKIN | 233 (41) |
Inflammatory risk factors | ||||||||
(16) (Han et al., 2017) | Cohort study | High-sensitivity C-reactive protein | 1656 | Stratified in tertiles | Isolated CABG | 64.7 | KDIGO | 549 (33.2) |
Other comorbidities | ||||||||
(17) (Shlipak et al., 2011) | Cohort study | Pre-existing kidney disease (creatinine, eGFR, cystatin C) | 1147 | Stratified in quintiles | Isolated CABG, valve surgery or both | 71 | AKIN | 407 (36) |
(18) (Wyler von Ballmoos et al., 2018) | Cohort study | Galectin-3 | 1498 | Stratified in tertiles | Isolated CABG | 64.02 (no AKI) | KDIGO | 568 (37.9) |
68.08 (with AKI) | ||||||||
(19) (Coca et al., 2012) | Cohort study | Urine albumin to creatinine ratio | 1159 | Stratified in groups | Isolated CABG, valve surgery or both | Stratified in groups | AKIN | 409 (35.2) |
(20) (Huang et al., 2011) | Cohort study | Proteinuria (dipstick) | 1052 | 315 (29.9) with mild proteinuria | Isolated CABG | 65.7 | AKIN | 183 (17.4) |
138 (13.1) with heavy proteinuria | ||||||||
(21) (Zou et al., 2019) | Cohort study | RDW | 13420 | 10274 (76) unmatched cohort | Isolated CABG | 53.3 | KDIGO | 32.80% |
3146 (24) matched cohort | ||||||||
(22) (Duchnowski et al., 2020) | Cohort study | RDW | 751 | Stratified in groups | Valve surgery | 63.5 | 46 (6%) |
Author, year | Risk Factor | RR/OR (95% CI) | P value |
(Hertzberg et al., 2019) | Heart failure | OR 1.12 (1.02-1.23) | − |
(Patel et al., 2012) | BNP (highest quintile) | RR 1.87 (1.40-2.49) - at least mild AKI | |
RR 3.17 (1.06-9.48) - severe AKI | 0.049 (for trend) | ||
(Belley-Côté et al., 2016) | NT-proBNP | OR 1.39 (1.03-1.88) | 0.03 |
hs-cTnT | OR 1.17 (0.84-1.61) | 0.4 | |
cTnI | OR 1.22 (0.92-1.63) | 0.2 | |
CK-MB | OR 1.55 (0.94-2.58) | 0.1 | |
(Hertzberg et al., 2015) | Type 1 diabetes mellitus | OR 4.89 (3.82-6.25) | − |
Type 2 diabetes mellitus | OR 1.27 (1.16-1.40) | ||
(Kocogulları et al., 2017) | Hemoglobin A |
OR 11.17 (2.21-56.33) | 0.003 |
(Oezkur et al., 2015) | Chronic hyperglycemia (HbA |
OR 1.65 (1.00-2.71) | 0.049 |
(De Santo et al., 2009) | Anemia | OR 2.06 (1.14-3.70) | 0.016 |
(Karkouti et al., 2011) | Anemia | − | 0.0007 |
(O’Sullivan et al., 2015) | Obesity | OR 2.12 (1.27-3.54) | 0.004 |
(Coca et al., 2013) | ACEi/ARB | RR 1.24 (1.02-1.5) | − |
(Barodka et al., 2011) | ACEi/ARB | OR 0.19 (0.04-0.84) | 0.029 |
(van Diepen et al., 2018) | ACEi/ARB | − | 0.991 |
(Jiang et al., 2018) | Cardiac catheterization within 7 days | OR 2.546 (1.548-4.189) | |
(Ranucci et al., 2013) | Coronary angiography on the same day | OR 1.58 (1.04-2.40) | 0.031 |
(Borde et al., 2019) | Coronary angiography within 7 days | OR 1.32 (0.93-1.9) | 0.13 |
(Özkaynak et al., 2014) | Cardiac catheterization | − | 0.09 |
(Han et al., 2017) | High-sensitivity C-reactive protein | OR 1.86 (1.387-2.489) | |
(Shlipak et al., 2011) | Cystatin C (highest quintile) | OR 4.8 (2.9-7.7) | − |
Creatinine (highest quintile) | OR 1.8 (1.2-2.6) | − | |
eGFR (highest quintile) | OR 1.7 (1.1-2.3) | − | |
(Wyler von Ballmoos et al., 2018) | Galectin-3 (highest tertile) | OR 2.95 (1.63-5.34) - AKI stage 2 or 3 | |
OR 1.71 (1.24-2.37) - AKI stage 1 | 0.001 | ||
(Coca et al., 2012) | Urine albumin to creatinine ratio 30-299 mg/g | RR 1.64 (1.33-1.97) | − |
Urine albumin to creatinine ratio |
RR 2.21 (1.66-2.73) | − | |
(Huang et al., 2011) | Mild proteinuria (dipstick) | OR 1.66 (1.09-2.52) | 0.018 |
Heavy proteinuria (dipstick) | OR 2.30 (1.35-3.90) | 0.002 | |
(Zou et al., 2019) | RDW | OR 1.3 (1.21-1.4) | |
(Duchnowski et al., 2020) | RDW | OR 1.59 (1.21-2.54) | 0.003 |
The quality of non-randomized studies was assessed using the Newcastle-Ottawa scale for cohort studies, and the results are summarized in Supplemental (Table 1). RoB2 assessed the risk of bias of the randomized trial (van Diepen et al., 2018), and the overall risk of bias was estimated as low.
Our paper represents the first systematic approach to review all proposed preoperative risk factors and their predictive power.
Our strategy is the starting point for selecting and comparing the predictive elements to be integrated into future risk models. Systematically reviewing all risk factors is useful for both physicians and patients as the timely identification and correction (when possible) of preoperative risk profiles may result in safer surgical strategies and better postoperative outcomes.
CS-AKI represents a frequent complication without a specific treatment. Effective prevention, thus, remains the most powerful tool to overcome this problem. Improving the preventive strategies is possible by establishing appropriate preoperative risk profiles. Current clinical models display a low predictive value for milder forms of CS-AKI, although less severe forms are more common and have a negative impact on short and long-term mortality. Therefore, new and integrative models are needed to encompass risk predictions for all severity categories accurately. Before the construction of new models, the first essential step is to comprehensively review all updated original work studying preoperative risk factors for CS-AKI.
Management of pre-existing heart failure is vital as it is associated with a
higher incidence of postoperative CS-AKI. Hertzberg et al. (2019)
stratified the CS-AKI risk according to the left ventricular ejection fraction
(EF). Patients with an EF lower than 30% had a greater risk of CS-AKI than those
with a higher EF (OR 1.32, 95% CI 1.06-1.65). In addition, heart failure was
associated with increased 30-day mortality. However, an important limitation of
this study is that left ventricular filling pressures and relaxation were not
considered. Also, patients with high BNP and NT-proBNP levels had a greater
incidence of postoperative CS-AKI, as reported by Patel et al. (2012),
and Belley-Côté et al. (2016). BNP levels enhanced the
clinical model’s predictive performance proposed by Patel et al. (2012).
In addition, Belley-Côté et al. (2016) reported a higher 1-year
mortality in patients with elevated NT-proBNP (OR 1.70, 95% CI 1.35-2.14,
P
A significant number of patients undergoing cardiac surgery have diabetes
mellitus, another critical risk factor for CS-AKI. Hertzberg et al. (2015) showed that the therapeutic regimen influences CS-AKI risk in
diabetic patients. A greater risk was reported in patients treated with insulin
whether combined with oral antidiabetic drugs or not (OR 1.82, 95% CI 1.61-2.06)
compared to those treated with oral antidiabetic medication only (OR 1.23, 95%
CI 1.08-1.41) or diet only (OR 1.13, 95% CI 0.93-1.38). Moreover,
Kocogulları et al. (2017) and Oezkur et al. (2015) reported
that chronic hyperglycemia (defined by a high preoperative level of hemoglobin
A
De Santo et al. (2009) and Karkouti et al. (2011) reported that patients with anemia or erythrocyte transfusions had an increased incidence of postoperative CS-AKI. These results indicate that early correction of anemia may play an essential role in reducing the postoperative risk.
International consensus guidelines advocate for preoperative discontinuation of ACEIs and ARBs (Nadim et al., 2018), although data regarding their CS-AKI development risk is discordant. Coca et al. (2013) reported an increased risk of CS-AKI for patients treated with ACEIs or ARBs. Conversely, a randomized trial (van Diepen et al., 2018) revealed similar risks after medication discontinuation. Surprisingly, Barodka et al. (2011) revealed a protective effect of the preoperative treatment with ACEIs or ARBs against CS-AKI development. However, the latter study only included patients over 65 years, limiting the results’ interpretation for the younger population.
The international consensus recommends no exposure to contrast agents 24 to 72 h
before cardiac surgery (2C level of evidence) (Nadim et al., 2018), although
studies are inconsistent. A single-center study from China (Jiang et al., 2018) reported that the dose of the contrast agent is also essential,
suggesting that patients receiving a dose higher than 240 mg/kg should postpone
the surgery (
A South Korean observational study (Han et al., 2017) revealed that patients
with high hs-CRP levels had an increased risk of postoperative CS-AKI after
excluding septic patients (OR 1.88, 95% CI 1.392-2.525, P
The preoperative kidney function is crucial as it indicates the baseline functional reserve and may identify the patients with a high risk of CS-AKI. Preoperative serum creatinine, cystatin C, and eGFR are used for evaluating kidney function. Mainly, cystatin C displays a stronger association with postoperative CS-AKI compared to the others (Shlipak et al., 2011). However, only serum creatinine and eGFR are routinely used due to lower costs.
Galectin-3 is a novel biomarker linked to chronic kidney disease, heart failure, and inflammation. Wyler von Ballmoos et al. (2018) reported that patients with a high concentration of Galectin-3 had a greater incidence of CS-AKI, and the addition of Galectin-3 improved the predictive power of the existing clinical models.
Coca et al. (2012) and Huang et al. (2011) reported that UACR
and dipstick proteinuria was associated with a higher risk of CS-AKI. However,
there was no association between proteinuria and CS-AKI in patients with eGFR
Regarding preoperative intra-aortic balloon pump (IABP) placement, the international consensus position (Nadim et al., 2018) is that IABP could have a beneficial effect on postoperative CS-AKI incidence. This is based on the results of a meta-analysis of 17 studies (Wang et al., 2016) reporting that the preoperative placement of IABP in patients at high risk was associated with a lower incidence of AKI after CABG (OR 0.54, 95% CI 0.36-0.79, P = 0.002).
Our study may be regarded as a source of analysis of recent studies on risk factors of AKI after cardiac surgery. AKI following cardiac surgery represents a frequent and severe complication. As there is no specific treatment, prevention measures that identify and minimize preoperative risk factors play a crucial role. Heart failure, chronic hyperglycemia, anemia, obesity, preoperative exposure to nephrotoxic drugs or contrast media, inflammation, proteinuria, and pre-existing kidney disease are associated with an increased risk of postoperative CS-AKI. To date, there are no externally validated or universally accepted clinical models to accurately predict postoperative CS-AKI occurrence, especially in non-severe clinical forms. For this reason, the clinical judgment and a good knowledge of the preoperative risk factors in the light of new evidence may help to personalize preoperative risk profiles as the cornerstone of prevention measures.
Thanks to all the peer reviewers and editors for their opinions and suggestions.
Concept/design: CB, GT;
Data collection: CB, GT;
Data analysis/interpretation: all authors;
Drafting article: all authors;
Critical revision of article: AB, IVP;
Approval of article: all authors.
The authors declare no conflict of interest.