Printer Friendly

Predictors and Outcomes of Acute Kidney Injury after Cardiac Surgery.

A cute kidney injury (AKI) is a common complication after cardiac surgery that represents a spectrum of minor changes in serum creatinine (SCr) or a brief period of oliguria that may progress to acute tubular necrosis requiring renal replacement therapy (RRT) (Shaw, 2012). The incidence of AKI after cardiac surgery (CS-AKI) varies from 7.7% to 40.0%, depending on patient selection, types of surgery, and criteria used for determining AKI (Chan et al., 2017; Corredor, Thomson, & Al-Subaie, 2016; Petaja et al., 2016; Thiele, Isbell, & Rosner, 2014). Over 90% of CS-AKI cases develop within four days postoperatively and may significantly increase intensive care unit (ICU) length of stay (LOS) and hospital LOS, as well as morbidity and mortality (Birnie et al., 2014; Corredor et al., 2016; Petaja et al., 2016; Thiele et al., 2014). Approximately 2% to 6% of patients with CS-AKI will require RRT (Petaja et al., 2016; Thiele et al., 2014), and those who develop severe CS-AKI have a 2.5-fold increase in mortality (Birnie et al. 2014; Corredor et al., 2016; Petaja et al., 2016; Thiele et al., 2014). Because advanced age has been a consistent risk factor for CS-AKI, the incidence of CS-AKI is anticipated to rise in the future due to the projected increase in the number of cardiac surgeries performed in an aging population (Birnie et al., 2014; Petaja et al., 2016; Shaw, 2012; Thiele et al., 2014).

Literature Review

AKI reflects structural injury preceding an abrupt loss of kidney function, leading to either an increase in SCr or a decrease in urinary output (Petaja et al., 2016; Shaw, 2012). The glomerular filtration rate (GFR) is the most accurate representation of renal function, and it may be estimated using SCr and timed urine collection (Petaja et al., 2016). However, using a 24-hour urine creatinine clearance is inconvenient and time-consuming, and both SCr and urinary output have been traditionally used as surrogate indicators of renal function (Machodo, Nakazone, & Maia, 2014; Petaja et al., 2016). Hourly urine output is frequently unreliable, and AKI based on SCr criteria has better validity than oliguria criteria (O'Neal, Shaw, & Billings, 2016). Several studies even focused on SCr alone to determine AKI (Birnie et al., 2014; Chan et al., 2017; Hansen et al., 2015; Machado et al., 2014). However, establishing the diagnosis of AKI using only SCr without urine output is not without issues, and may lead to both under-diagnosed and over-diagnosed AKI (Englberger et al., 2011; Petaja et al., 2016).

Various definitions of AKI have been used in the past (Petaja et al., 2017). The Acute Dialysis Quality Initiative (ADQI) was first created to develop evidence-based guidelines for the treatment and prevention of AKI. This was later named the RIFLE criteria, with RIFLE being an acronym for risk, injury, failure, loss, and end stage renal disease (ESRD) (Bellomo, Ronco, Kellum, Mehta, & Palevsky, 2004). This expert workgroup introduced the term 'AKI' to replace acute renal failure and proposed the diagnosis of AKI based on either increased SCr, decreased GFR, or decreased urinary output. Three stages of AKI severity (mild, moderate, and severe) and two outcome stages contingent on the required duration for RRT (loss of renal function and ESRD) were recommended (Bellomo RIFLE criteria have been extensively validated, but studies show they have limited application for some patient populations and with certain causes of AKI (Hocine et al., 2016; Shaw, 2012).

The Acute Kidney Injury Network (AKIN) was later formed to improve the understanding of AKI and classified AKI into three stages without retaining any outcome stages (Mehta et al., 2007; Shaw, 2012). AKIN criteria eliminated the use of GFR all together and added an absolute SCr in addition to a relative SCr increase in AKI Stage 1 (Mehta et al., 2007). These criteria defined changes within 48 hours instead of the seven days in RIFLE criteria. AKIN criteria improved the recognition of patients with AKI superimposed on underlying chronic kidney disease (CKD) (Park, 2017). In an attempt to resolve differences between the RIFLE and AKIN guidelines, the Kidney Disease: Improving Global Outcomes (KDIGO) clinical practice guideline for AKI was developed (KDIGO AKI Work Group, 2012; Stevens & Levin, 2013). KDIGO AKI staging is based on absolute SCr elevation within 48 hours, relative SCr elevation within seven days after surgery, and urinary output within the first 6 to 12 hours (KDIGO AKI Work Group, 2012). KDIGO staging criteria are believed to be suitable for the definition of CS-AKI, and to more accurately reflect the mortality and RRT risks associated with CS-AKI (Park, 2017).


The pathophysiology of CS-AKI is complex and may be attributed to kidney-related or surgery-related factors (O'Neal at al., 2016; Shaw, 2012). Kidney-related contributors are categorized as pre-renal, intrinsic, and post-renal factors. Inadequate renal perfusion, such as atherosclerosis, hypotension, and hypoxemia, are main pre-renal mechanisms (Park, 2017). Tissue ischemia, systemic inflammation, or direct toxin exposure are related to intrinsic renal factors (O'Neal et al., 2016; Thiele et al., 2015). Venous congestion after surgery is an example of a post-renal cause of CS-AKI (O'Neal et al., 2016).

Surgical-related factors may be divided into preoperative, intraoperative, and postoperative risk factors (O'Neal et al., 2016). Well-established preoperative risk factors include demographics, nephrotoxic medications or substances, preexisting CKD, underlying medical conditions, and cardiac disease (O'Neal et al., 2016). Specific risk factors include advanced age, female gender, ethnicity, obesity, hypertension (HT), diabetes mellitus (DM), diuretic/angiotensin converting enzyme inhibitor (ACEI) drugs, peripheral arterial disease (PAD), cerebrovascular accident (CVA), chronic obstructive pulmonary disease (COPD), CKD, anemia, cardiomegaly, decreased left ventricular function (LVF), congestive heart failure (CHF), valvular heart disease, and contrast media (O'Neal et al., 2016; Petaja et al., 2017; Shaw, 2012). The types of surgery and cardiopulmonary bypass (CPB) are operative etiologies, whereas intra-aortic balloon pump (IABP) and blood product administration are operative and postoperative risk factors (O'Neal et al., 2016; Shaw, 2012). Other postoperative risk factors include inotropic use, embolic events, other postoperative complications, and the use of antibiotics and non-steroidal anti-inflammatory drugs (Park, 2017).

Patient Outcomes

CS-AKI increases the risk of in-hospital mortality and morbidity, including sepsis, coagulopathy, mechanical ventilation, RRT, and ICU and hospital LOS based on the RIFLE and AKIN criteria (Corredor et al., 2016; Shaw, 2012). Additional studies have reported on incidence, morbidity, and mortality based on KDIGO criteria (Birnie et al., 2014; Engoren, Habib, Arslanian-Engoren, Kheterpal, & Schwann, 2014; Han et al., 2015; Hansen et al., 2015; Machado et al.,2014). For example, any stage of CS-AKI increased the postoperative stay, infection, and pulmonary complications, and result in an almost 5fold increase in the odds of in-hospital mortality (Birnie et al., 2014). Patients who do not require immediate postoperative RRT will have progression of AKI requiring RRT later, and they also have a higher intermediate and long-term mortality (Engoran et al., 2014; Han et al., 2015; Hansen et al., 2015; Machado et al., 2014). Engoran and colleagues (2014) reported increased 30-day and long-term mortality at 6 months to 6.3 years associated with all stages of AKI. Early-onset (within four days) and late-onset (within 30 days) CS-AKI have also been associated with increased 5-year risk of myocardial infarction (MI), CHF, CVA, and all-cause mortality (Hansen et al., 2015).

Several risk stratification scores were developed to identify those at risk for CS-AKI. Petaja and colleagues (2017) found that patients with CSAKI received more blood products and vasoactive agents, and that significant predictors for CS-AKI in multivariable logistic regression included age, body mass index (BMI), PAD, CKD, urgent surgery, low hematocrit (Hct), vasodilators, and postoperative diuretic treatment. Han and colleagues (2015) found that the EuroScore value, cardiopulmonary bypass (CPB) duration, and ICU LOS were predictors for any stage of CS-AKI in patients with elevated SCr. Birnie and colleagues (2014) reported predictors for any stage of CS-AKI based on KDIGO criteria, which included older age, male sex, BMI greater than 35 kg/[m.sup.2], current smokers, higher dyspnea categories, DM, PAD, HT, lower hemoglobin (Hb), lower estimated GFR, catheter to surgery within 24 hours, triple vessel disease, poor left ventricular ejection fraction (EF), emergency/salvage operations, and more complex surgery.

CS-AKI remains an important healthcare issue despite ongoing research over the past two decades. The lack of effective preventive measures, early recognition, and treatment for ongoing CS-AKI continue to be major contributors to the high incidence and poor outcomes of CS-AKI (Birnie et al., 2014; Thiele et al., 2014). Inconsistent definitions in the past have impeded the development of a consensus recommendations (Birnie et al., 2014). Most recent studies on CS-AKI adhere to KDIGO value criteria, but temporal identification remains heterogeneous, varying between 2 to 30 postoperative days (Birnie et al., 2014; Chan et al., 2017; Engoren et al., 2014; Hansen et al., 2014; O'Neal et al., 2016; Petaja et al., 2017). Consistent areas of investigation and findings on CS-AKI based on KDIGO criteria are still lacking, especially in predictors and prognosis of less severe CS-AKI (Birnie et al., 2012). The purpose of the present study was to explore predictors and outcomes of all CS-AKI stages based on the KDIGO value criteria throughout the length of hospital admission.

Methods and Sample

This study used a retrospective chart review of patients who had cardiothoracic surgery at a community hospital in a Midwestern city of a large healthcare system between June 2011 and December 2016. The study was approved by the relevant Institutional Review Boards and Human Rights Review Committees. During the study period, 1,053 patients underwent cardiac surgery. Only patients without known ESRD who were treated with HD were included in the study. The study excluded 38 patients with ESRD on HD, one patient with CKD Stage 5 not yet on HD, and five patients with operative mortality; the final study population included 1,009 patients.

Data collection was performed by a registered nurse who had been trained to collect data for the Society of Thoracic Surgeons (STS) Adult Cardiac Surgery National Database at the study facility (Shahian et al., 2013). The data collection process has undergone yearly independent auditing from the STS National Database throughout the years. Data on known predisposing risk factors for AKI in the general population were obtained, including age, gender, weight, height, smoking history, DM, HT, hyperlipidemia, PAD, SCr, ESRD on HD, decreased left-ventricular EF, COPD, and other related or contributing factors based on the STS database (see Table 1) (Shahian et al., 2013). Data collection also included possible operative risks (left internal mammary artery usage, CPB, and intraoperative transfusion) and postoperative risks for CS-AKI (reintubation, prolonged ventilation, highest postoperative SCr, pneumonia, intubation duration, and postoperative transfusion). Acute respiratory failure (ARF) was defined as prolonged intubation greater than 48 hours or reintubation. Preoperative SCr was that obtained within 24 hours before cardiac surgery, and peak postoperative SCr was the highest SCr at any time during hospital admission. No information was available regarding when peak postoperative SCr levels were obtained.

Patients' heights and weights were used to calculate BMIs. Pre-existing CKD was determined from estimated GFR (eGFR) calculated from preoperative SCr using the re-expressed 4variable Modification of Diet in Renal Disease (MDRD) Study equation for SCr (mg/dL): GFR = 175 x standardized [SCr.sub.-1.154] x ]age.sup.-0203] x 1.212 (if black) x 0.742 (if female) (Earley, Miskulin, Lamb, Levey, & Uhlig, 2012; Levey et al., 2007). CKD stages were defined from the eGFR as follows: CKD Stage 1 with GFR greater than 90 mL/ min/1.73[m.sup.2], CKD Stage 2 with GFR = 60 to 89 mL/min/1.73[m.sup.2], CKD Stage 3 with GFR = 30 to 59 mL/min/1.73 [m.sup.2], CKD Stage 4 with GFR = 15 to 29 mL/min/1.73 [m.sup.2], and CKD Stage 5 with GFR less than 15 mL/min/1.73[m.sup.2] (KDIGO, 2013). AKI for this study was calculated from the preoperative and highest postoperative SCr during the hospital stay using preoperative and highest postoperative SCr using KDIGO value criteria as follows: AKI Stage 1, including patients with absolute increased SCr equal to or greater than 0.3 mg/dL or relative increased SCr x 1.5 to 1.9 baseline; AKI Stage 2, including those with relative increased SCr x 2 to 2.9 baseline; and AKA Stage 3, including those with relative increased SCr 3x baseline or SCr equal to or greater than 4 mg/dL, or those with initiation of RRT (KDIGO AKI Work Group, 2012). In patients under 18 years of age, a decrease in eGFR less than 35 mL/min/1.73 [m.sup.2] is also considered AKI Stage 3 (KDIGO AKI Work Group, 2012).

Data Analysis

Descriptive and bivariate analyses were conducted using the Statistical Package for Social Sciences (SPSS) version 25 (International Business Machine Corporation, 2017), including the one-way analysis of valiance (ANOVA), Kruskal-Wallis test, and the [chi square] tests of independence. Data were examined for the contribution of possible causes for CS-AKI, such as predisposing, operative, and postoperative factors. Bivariate analyses were performed to initially address the relationship between hypothesized risk factors and CS-AKI. For continuous outcomes, a one-way ANOVA was conducted to compare variability in the outcome by AKI staging. However, when concerns over the assumptions of a one-way ANOVA arose, the Kruskal-Wallis test was typically conducted. For all discrete outcomes, the [chi square] tests of independence was conducted to identify whether observed frequencies vary by AKI staging.

The relationship between the risk factors and AKI staging were also explored and evaluated using ordinal logistic regression using Statistical Analysis System (SAS) 9.4 (Statistical Analysis System Institute, Inc., 2017). Ordinal logistic regression allows researchers to evaluate the influence of a set of predictors on an outcome variable measured on an ordinal scale (Tabachnick & Fidell, 2007). Due to the exploratory characteristic of this study, all variables were entered into an initial regression model and evaluated in turn for removal using significance tests, odd ratios for their substantial contribution, and amenability to regression analysis. For an example of the latter, some variables that exhibited too much collinearity (e.g., variance inflation factor [VIF] greater than 10) were subsequently eliminated from the model for their high level of redundancy. The Wald and likelihood ratio tests were used to evaluate the statistical contribution of each predictor with alpha set to 0.5, while an odds ratio of less than 1.50 was used as a guide to identify variables with a trivial contribution to predicting CS-AKI. Prior to interpretation, model diagnostics were inspected to ensure the validity of the result, including whether the variable selection method had an influence over the composition of the final mode.


Demographic and clinical variables were presented in three categories; preoperative (see Table 1), operative (see Table 2), and postoperative factors (see Table 3). The overall incidence of CS-AKI in this study was 30.3% for AKI Stage 1, 7.1% for AKI Stage 2, and 7.9% for AKI Stage 3. Forty-nine patients (4.9%) required RRT.

* Preoperative factors (see Table 1) statistically associated with CSAKI included age, BMI, Hct, albumin, race, smoking history, HT, DM, CKD, COPD, CHF, and previous percutaneous coronary intervention (PCI).

* Operative factors (see Table 2) statistically associated with CS-AKI included type of surgery, CPB, circulatory arrest, intraoperative blood transfusion, CPB time, and time in the OR.

* Postoperative factors (see Table 3) statistically associated with all-stage CS-AKI were acute respiratory failure (prolonged intubation greater than 48 hours/reintubation), reoperation for postoperative bleeding, postoperative transfusion, coma/encephalopathy, postoperative atrial fibrillation, and pneumonia.

Results of the final logistic regression model are displayed in Table 4. Among the preoperative, operative, and postoperative predictors, smoker status, hypertension, CPB, acute respiratory failure, postoperative blood product administration, and coma/ encephalopathy performed well as a set of predictors. The likelihood ratio test indicated that the set of predictors performed significantly better than the null model ([chi square] = 198.274, df=12, p=0.000). Akaike information criterion (AIC) also indicated that the final regression model performed better than the null ([AIC.sub.model] = 1357.346 and [AIC.sub.null] =1531.619).

Among the preoperative factors, smoking history and hypertension shared a salient association with AKI staging. Under the proposed model, smoking was associated with CS-AKI across all AKI stages. Patients who smoked were 2.058 times the odds of being classified by the model as Stage 0 than higher stages compared to nonsmokers. Smokers were less likely to suffer from CS-AKI, and smoking did not increase the odds of developing higher AKI stages. While holding all other variables constant, patients with HT were 3.071 times more likely to be classified into a higher AKI stage than patients without hypertension. The influence of hypertension was not consistent across all AKI stages because freeing the coefficient to vary across stages caused convergence failures.

CPB was the only operative variable that tenably contributed to higher AKI staging in the model. Patients having CPB were 2.358 times more likely to experience higher AKI stages than patients who had not undergone CPB. Among postoperative factors, ARF, postoperative blood transfusion, and coma/encephalopathy contributed to the risk of having CS-AKI. In general, postoperative factors tended to exhibit the largest associations with higher AKI staging. Participants who had ARF were likely to suffer all AKI stages. For example, the risk for AKI Stage 1 or higher was 2.183 higher for patients with ARF compared to those without ARF, and the risk continued to escalate for AKI Stage 2 compared to AKI Stage 1 (OR=3.441, p=0.000) and AKI Stage 3 compared to AKI Stage 2 (OR=4.840, p=0.000). Likewise, patients receiving postoperative blood transfusion were 1.927 times more likely to be classified into higher AKI stages by the regression model than those who did not receive post-operative blood transfusion, with the highest risk escalation for AKI Stage 3 (OR=8.850, p=0.000). Patients with a coma/encephalopathy were 5.051 times more likely to develop higher AKI stages than patients who had not experienced coma/encephalopathy.

Postoperative outcomes of patients with CS-AKI are shown in Table 5. Patients who developed CS-AKI had significantly increased ICU LOS, postoperative LOS, and hospital LOS, and were less likely to be discharged home. Patients with CS-AKI also had a higher postoperative mortality. Although the LOS typically increases with CS-AKI, correlation values indicate there is very little change between adjacent stages, which could have been attributed to a large variation in LOS for all AKI stages. For example, while postoperative LOS is expected to increase by one day between patients without CS-AKI and patients with AKI Stage 1 (see Table 5), there is also a lot of variation within each stage (0 to 42 days for patients without CS-AKI and 0 to 44 days for those with AKI Stage 1). AKI Stage 1 did not exert any influence on postoperative mortality (OR=1.360). However, the percentage of mortality rose substantially as AKI increased from AKI Stage 1: the odds of dying were 12.600 times higher for those with AKI Stage 2 and much higher for those with AKI Stage 3 (OR=59.564) compared to those without CS-AKI. As for discharge status, the difference was small between patients without CS-AKI and those with AKI Stage 1 (OR=1.854), but more noticeable for AKI Stage 2 (OR=3.468) and AKI Stage 3 (OR=8.886).


The incidence of CS-AKI in this study was high and is consistent with other studies reporting a prevalence of CS-AKI, up to 40% using both the RIFLE and AKIN criteria (Corredor et al., 2016). However, the incidence in this study is different from others who used KDIGO criteria. (Birnie et al. 2014; Chan et al., 2017; Petaja et al., 2017). The reason for a higher incidence of CS-AKI in this study may be related to a lack of specific time when the peak SCr was obtained, and some values could have been sampled beyond the first 7th postoperative day. Patients with late AKI possibly secondary to other causes unrelated to cardiac surgery might have been included in the present study. Hansen and colleagues (2015) reported that this overestimation may be as high as 8% of the overall incidence. Fortunately, most cases of CS-AKI (70% to 80%) in the present study and others were AKI Stage 1 and did not impact patient outcomes as significantly as would be expected from higher AKI stages.

Similar to prior studies, CS-AKI was associated with several preoperative, operative, and postoperative factors; although, there were more associated factors in the present study than in others (Birnie et al., 2014; Han et al., 2015; Shaw, 2012). There were as many as 23 factors in this study that were associated with CS-AKI, including age, race, smoking history, diabetes, hypertension, COPD, Hct, albumin, CKD, previous PCI, previous CHF, time in OR, CPB, circulato ry arrest, intraoperative blood transfusion, postoperative ARF, postoperative transfusion, reoperation for postoperative bleeding, coma/encephalopathy, postoperative CVA, postoperative atrial fibrillation (POAF), and pneumonia. However, the sequential regression analysis in this study found that only smoking history, hypertension, CPB, reintubation, and coma/ encephalopathy were concurrent predictors for CS-AKI stages. Acute respiratory failure and postoperative blood transfusion were contributing factors to CS-AKI across all AKI stages, whereas HT, CPB, and coma/encephalopathy were predictors for higher AKI stages, and smoking history was predictive of only lower AKI stages. The present study is also different from other studies that stratified risk factors for all stages of CS-AKI (Birnie et al., 2014; Han et al., 2015; Petaja et al., 2017). In one study, the EuroScore value, CPB duration, and ICU LOS were predictors for any-stage of CS-AKI (Han et al., 2015). Another study developed and validated predictors for any-stage CSAKI; reported predictors for CS-AKI included older age, male sex, BMI greater than 35 kg/m2, current smokers, higher dyspnea categories, DM, PAD, HT, lower Hb, lower eGFR, cardiac catheterization within 24 hours before surgery, triple vessel disease, poor EF, and more complex or emergency surgery (Birnie et al., 2014). Several of these predictors were also found to have similar predictive values in a study by Petaja and associates (2017), including age, BMI, PAD, CKD, urgent surgery, low Hct, vasodilators, and postoperative diuretic.

In this study, CS-AKI adversely affected all postoperative outcomes, including increased ICU LOS, postoperative LOS, hospital LOS, and inhospital mortality, but reduced the likelihood of home discharge. The need for postoperative and the longterm need for RRT in this study is similar to others (Petaja et al., 2017; Thiele et al., 2014). The increased postoperative LOS and mortality in the present study are similar to findings reported by Birnie and associates (2014), but are different in that our study did not reveal increases in postoperative infection or other pulmonary complications. The increased postoperative mortality in the present study corresponds to one report that revealed higher mortality in patients with AKI Stages 2 and 3 (Engoren et al., 2014). However, this finding is different from the study by Birnie and associates (2014) who reported increased mortality across all stages of CS-AKI; the likelihood of dying was 4.6 times higher for patients with AKI Stage 1, 8.0 times for patients with AKI Stage 2, and 81.2 times higher for patients with AKI Stage 3.

Strengths and Limitations

The primary strength of this study is the ability to assess predictors and outcomes of all CS-AKI stages based on KDIGO criteria. The study was limited because it involved a crosssectional, retrospective chart review, and documentation errors are possible. The chart review was performed by only one nurse, which limits assessment of accuracy, reliability, and validity of the medical record review. However, the nurse was trained and experienced in this type of data collection. The study utilized only SCr without including urinary output criteria to stage AKI, and this may have under-diagnosed CS-AKI in some patients (Petaja et al., 2017). Using the highest postoperative SCr during the hospital stay in this study without having postoperative SCr at 48 hours or at Day 7 was also a limiting factor. Two hundred eighty-six (286) post-cardiac surgery patients with AKI Stages 1-3 (28.5%) were hospitalized longer than seven days after surgery, and several of them might potentially have had late AKI not meeting a strict KDIGO criteria. Future studies that include SCr samples at 48 hours and at Day 7, and hourly urine output during the first postoperative day should help clarify these limitations. Nevertheless, assessing postoperative outcomes of CS-AKI using only KDIGO criteria alone will still miss many late-occur ring CS-AKIs, which have as devastating prognosis as the early events (Hansen et al., 2015)

The diagnosis of AKI based on SCr may be too late to apply any preventive measures because SCr level does not begin to increase until one to three days after AKI has already been established (Shaw, 2012). Specific serum and urine protein biomarkers that are released as early as within 2 hours of AKI may be more superior in recognizing and managing CS-AKI (Shaw, 2012). These biomarkers include neutrophil gelatinase-associated lipocalin (NGAL), Cystatin C (CyC), kidney injury molecule-1 (KIM-1), tissue inhibitor of metalloproteinase-2 (TIMP-2), and insulin-like growth factor binding protein 7 (IGFBP7) (Shaw, 2012). Currently, specific treatments for AKI are lacking, but preventative measures may include maintenance of adequate hemodynamics, sufficient fluid resuscitation, avoiding nephrotoxic agents and optimizing preoperative hemoglobin levels, avoiding excessive transfusions, maintaining sufficient urine output and acid-base and electrolyte homeostasis, and early initiation of RRT (O'Neal et al., 2016; Shaw, 2012). More studies utilizing these serum and urine protein biomarkers are needed (Shaw, 2012).


In summary, the present study reaffirms the common prevalence of CS-AKI and demonstrates its independent devastating contribution to LOS, morbidity, mortality, and patient discharge home after cardiac surgery. Improved recognition of early CS-AKI, as well as effective therapies and preventative measures for this complication, requires ongoing research.


Bellomo, R., Ronco, C., Kellum, J.A., Mehta, R.L., & Palevsky, P. (2004). Acute renal failure--Definition, outcome measures, animal models, fluid therapy and information technology needs: The Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Critical Care, 8(4), R204-R212.

Birnie, K., Verheyden, V., Pagano, D., Bhabra, M., Tilling, K., Sterne, J.A., & Murphy, GJ. (2014). Predictive models for Kidney Disease: Improving Global Outcomes (KDIGO) defined acute kidney injury in UK cardiac surgery. Critical Care, 18(6), 606-625.

Chan, M.J., Lee, C.C., Chen, S.W., Tsai, F.C., Lin, PJ., Fan, P.C., ... Chang, C.H. (2017). Effect of different surgical type of coronary artery bypass grafting on kidney injury: A propensity score analysis. Medicine, 96(45), 46-51.

Corredor, C., Thomson, R. & Al-Subaie, N. (2016). Long-term consequences of acute kidney injury after cardiac surgery: A systematic review and meta-analysis. Journal of Cardiothoracic and Vascular Anesthesia, 30(1), 69-75.

Earley, A., Miskulin, D., Lamb, E.J., Levey, A.S., & Uhlig, K. (2012). Estimating equations for glomerular filtration rate in the era of creatinine standardization: A systematic review. Annals of Internal Medicine, 156(11), 785-795.

Englberger, L., Suri, R.M., Li, Z., Casey, E.T., Daly, R.C., Dearani, J.A., & Schaff, H.V. (2011). Clinical accuracy of RIFLE and Acute Kidney Injury Network (AKIN) criteria for acute kidney injury in patients undergoing cardiac surgery. Critical Care, 15(1), R16-R24.

Engoren, M., Habib, R.H., Arslanian-Engoren, C., Kheterpal, S., & Schwann, T.A. (2014). The effect of acute kidney injury and discharge creatinine level on mortality following cardiac surgery. Critical Care Medicine, 42(9), 2069-2074.

Han, S.S., Shin, N., Baek, S.H., Ahn, S.Y., Kim, D.K., Kim, S., ... Na, K.Y. (2015). Effects of acute kidney injury and chronic kidney disease on longterm mortality after coronary artery bypass grafting. American Heart Journal, 169(3), 419-425.

Hansen, M.K., Gammelager, H., Jacobsen, CJ., Hjortdal, V.E., Layton, J.B., Rasmussen, B.S., ... Christiansen, C.F. (2015) Acute kidney injury and long-term risk of cardiovascular events after cardiac surgery: A population-based cohort study. Journal of Cardiothoracic and Vascular Anesthesia,, 29(3), 617-625.

Hocine, A., Defrance, P., Lalmand, J., Delcour, C., Biston, P., & Piagnerelli, M. (2016). Predictive value of the RIFLE urine output criteria on contrast-induced nephropathy in critically ill patients. BioMed Central Nephrology, 17, 36-44.

International Business Machine Corporation. (2017). IBM SPSS Statistics for Window, Version 25.0. Armonk, NY: IBM Corp.

Kidney Disease: Improving Global Outcomes (KDIGO). (2013). KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney International, 3(1), 136-150.

Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury (AKI) Work Group. (2012). Clinical practice guideline for acute kidney injury. Kidney International, 2(Suppl. 1), 1-138. Retrieved from

Levey, A.S., Coresh, J., Greene, T., Marsh, J., Stevens, L.A., Kusek, J.W., & Van Lente, F. (2007). Expressing the modification of diet in renal disease study equation for estimating glomerular filtration rate with standardized serum creatinine values. Clinical Chemistry, 53(4), 766-772.

Machado, M.N., Nakazone, M.A., & Maia, L.N. (2014). Acute kidney injury based on KDIGO (Kidney Disease Improving Global Outcomes) criteria in patients with elevated baseline serum creatinine undergoing cardiac surgery. The Brazilian Journal of Cardiovascular Surgery, 29(3), 299-307.

Mehta, R.L., Kellum, J.A., Shah, S.V., Molitoris, B.A., Ronco, C., Warnock, D.G., & Levin, A. (2007). Acute Kidney Injury Network: Report of an initiative to improve outcomes in acute kidney injury. Critical Care, 11(2), R31-R38.

O'Neal, J.B., Shaw, A.D., & Billings, F.T., 4th. (2016). Acute kidney injury following cardiac surgery: current understanding and future directions. Critical Care, 20, 187-195.

Park, J.T. (2017). Postoperative acute kidney injury. Korean Journal of Anesthesiology, 70(3), 258-266.

Petaja, L., Vaara, S., Liuhanen, S., Suojaranta-Ylinen, R., Mildh, L., Nisula, S., ... Pettila, V. (2017). Acute kidney injury after cardiac surgery by complete KDIGO criteria predicts increased mortality. Journal of Cardiothoracic and Vascular Anesthesia, 31(3), 827-836.

Shahian, D.M.,Jacobs, J.P., Edwards, F.H., Brennan, J.M., Dokholya, R.S., Prager, R.L., . Glover, F.L. (2013). The Society of Thoracic Surgeons National Database. Heart, 99, 1494-1501.

Shaw, A. (2012). Update on acute kidney injury after cardiac surgery. The Journal of Thoracic and Cardiovascular Surgery, 143(3), 676-681.

Statistical Analysis System Institute, Inc. (2017). SAS 9.4 [Computer Software]. Cary, NC: SAS Institute Inc.

Stevens, P.E., & Levin, A. (2013). Evaluation and management of chronic kidney disease: Synopsis of the Kidney Disease: Improving Global Outcomes 2012 clinical practice guideline. Annals of Internal Medicine, 158 (11), 825-830.

Tabachnick, B.G., & Fidell, L.S. (2007). Using multivariate statistics (5th ed.). Boston, MA: Allyn & Bacon/Pearson Education.

Thiele, R.H., Isbell. J.M., & Rosner, M.H. (2015). AKI associated with cardiac surgery. Clinical Journal of the American Society of Nephrology, 10(3), 500-514.

Joanne Thanavaro, DNP, RN, AGACNPBC, AGPCNP-BC, DCC, FAANP, is the Associate Dean of Graduate Nursing Education, the Director of Advanced Practice Nursing, and an Associate Professor of Nursing, St. Louis University School of Nursing. St. Louis, MO.

John Taylor, PhD, is an Assistant Professor, St. Louis University School of Nursing. St. Louis, MO.

Linda Vitt, BSN, RN, is the Clinical Data Coordinator, SSMHealth St. Louis Network, St. Louis, MO.

Mary S. Guignon, BA, RN, MHA, is the Team Leader--Clinical Outcomes Department, SSM Health St. Louis Network, St. Louis, MO.

Statement of Disclosure: The authors reported no actual or potential conflict of interest in relation to this continuing nursing education activity.

Note: The Learning Outcome, additional statements of disclosure, and instructions for CNE evaluation can be found on page 41.
Table 1 Demographics, Preoperative Factors, and Acute Kidney Injury
After Cardiac Surgery (CS-AKI)

CS-AKI                          No CS-AKI             AKI Stage 1
Incidence                     54.8% (n=550)          30.3% (n=304)

Preoperative Factors

Age (Years)                62.29 [+ or -] 12.31   66.40 [+ or -] 12.11

Body mass index            29.26 [+ or -] 6.16    30.80 [+ or -] 6.91

Preoperative hematocrit    38.85 [+ or -] 4.68    38.13 [+ or -] 4.84

Preoperative albumin        3.63 [+ or -] 0.47     3.59 [+ or -] 0.46

Race (White)                   457 (83.1%)            241 (79.5%)

Smoking                        131 (23.8%)             41 (13.5%)

Diabetes mellitus              159 (28.9%)            110 (36.2%)

Hyperlipidemia                 442 (80.4%)            262 (86.2%)

Hypertension                   468 (85.1%)            286 (94.1%)


Mild                           138 (25.1%)             70 (2.0%)

Moderate                        39 (7.1%)              18 (5.9%)

Severe                          33 (6.0%)              26 (8.6%)

Obstructive sleep apnea         78 (14.2%)             54 (17.8%)

Peripheral arterial             66 (12.0%)             36 (11.8%)

Cerebrovascular accident        87 (15.8%)             49 (16.1%)

Preoperative CKD

Stage 1                        138 (25.1%)             14 (4.6%)

Stage 2                        290 (52.7%)             52 (17.1%)

Stage 3                        122 (22.2%)            238 (78.3%)

Beta blocker treatment         508 (92.4%)            273 (89.8%)

ACE I treatment                224 (40.7%)            143 (47.0%)

Lipid treatment                244 (44.4%)            159 (52.3%)

Preoperative AF                104 (18.9%)             60 (19.7%)


Previous PCI                   152 (27.6%)             99 (32.6%)

Present PCI                    113 (20.5%)             66 (21.7%)

Previous CABS                   13 (2.4%)              11 (3.4%)

Ml after catheterization       173 (31.5%)             96 (31.6%)

Previous CHF                   232 (42.2%)            146 (48.0%)

Recent CHF                     154 (28.0%)             85 (28.0%)

Number of vessels

1-vessel                        59 (10.7%)             42 (13.8%)

2-vessel                        88 (16.0%)             40 (13.2%)

3-vessel                       167 (30.4%)            108 (35.5%)

[greater than or equal          50 (9.1%)              26 (8.6%)
to] 50% Left Main artery

IABP                            30 (5.5%)              17 (5.6%)

CS-AKI                         AKI Stage 2            AKI Stage 3
Incidence                      7.1% (at=71)           7.9% (n=79)

Preoperative Factors

Age (Years)                69.38 [+ or -] 10.89   65.38 [+ or -] 10.89

Body mass index            32.78 [+ or -] 7.59    30.95 [+ or -] 6.50

Preoperative hematocrit    36.00 [+ or -] 4.32    35.30 [+ or -] 4.43

Preoperative albumin        3.50 [+ or -] 0.44     3.36 [+ or -] 0.56

Race (White)                    52 (73.2%)             55 (69.6%)

Smoking                         9 (12.7%)              13 (16.5%)

Diabetes mellitus               34 (47.9%)             37 (46.8%)

Hyperlipidemia                  61 (85.9%)             64 (81.0%)

Hypertension                    70 (98.6%)             75 (94.9%)


Mild                            17 (23.9%)             22 (27.8%)

Moderate                        12 (16.9%)             8 (10.1%)

Severe                           6 (8.5%)              10 (12.7%)

Obstructive sleep apnea         9 (12.7%)              11 (13.9%)

Peripheral arterial             8 (11.3%)              14 (17.7%)

Cerebrovascular accident        18 (25.4%)             18 (22.8%)

Preoperative CKD

Stage 1                          1 (1.4%)               0 (0.0%)

Stage 2                          3 (4.2%)               0 (0.0%)

Stage 3                         67 (94.4%)            79 (100.0%)

Beta blocker treatment          65 (91.5%)             72 (91.1%)

ACE I treatment                 37 (52.1%)             31 (39.2%)

Lipid treatment                 38 (53.5%)            37 (46.80%)

Preoperative AF                 19 (26.8%)             20 (25.3%)


Previous PCI                    23 (32.4%)             35 (44.3%)

Present PCI                     23 (23.9%)             21 (26.6%)

Previous CABS                    3 (3.6%)               5 (8.8%)

Ml after catheterization        21 (29.6%)             31 (39.2%)

Previous CHF                    34 (47.9%)             51 (64.6%)

Recent CHF                      19 (26.8%)             42 (53.2%)

Number of vessels

1-vessel                        13 (18.3%)             13 (16.5%)

2-vessel                        8 (11.3%)              8 (10.1%)

3-vessel                        24 (33.8%)             24 (30.4%)

[greater than or equal           1 (8.5%)               7 (7.9%)
to] 50% Left Main artery

IABP                             2 (1.4%)               4 (5.1%)


Preoperative Factors          Statistic (df)         P

Age (Years)                  13.34 (3, 1000)       0.000

Body mass index               8.649 (3, 998)       0.000

Preoperative hematocrit      18.79 (3, 1000)       0.000

Preoperative albumin          8.77 (3, 972)        0.000

Race (White)                    10.74 (3)          0.013

Smoking                         16.20 (3)          0.001

Diabetes mellitus               18.93 (3)          0.000

Hyperlipidemia                   5.31 (3)          0.151

Hypertension                    26.98 (3)          0.000


Mild                            35.96 (12)         0.000



Obstructive sleep apnea          2.45 (3)          0.485

Peripheral arterial              2.32 (3)          0.509

Cerebrovascular accident         6.05 (3)          0.109

Preoperative CKD

Stage 1                         44.32 (9)          0.000

Stage 2

Stage 3

Beta blocker treatment           1.65 (3)          0.648

ACE I treatment                  5.99 (3)          0.112

Lipid treatment                  6.02 (3)          0.111

Preoperative AF                  3.79 (3)          0.286


Previous PCI                     9.86 (3)          0.020

Present PCI                      1.75 (3)          0.626

Previous CABS                    7.00 (3)          0.072

Ml after catheterization         2.20 (3)          0.533

Previous CHF                    14.77 (3)          0.002

Recent CHF                      22.24 (3)          0.000

Number of vessels

1-vessel                        11.01 (9)          0.275



[greater than or equal           0.09 (3)          0.993
to] 50% Left Main artery

IABP                             2.25 (3)          0.523

Notes: Counts and relative frequencies are displayed for
categorical preoperative factors, and p-values are based upon the
[chi square] test of independence. Means and standard deviations
are displayed for continuous factors, and p-values are based upon a
one-way analysis of variance F-test.

ACEI = angiotensin converting enzyme inhibitor, AF = atrial
fibrillation, CABS = coronary artery bypass surgery, CHF =
congestive heart failure, CKD = chronic kidney disease, COPD =
chronic obstructive pulmonary disease, CS AKI = acute kidney injury
after cardiac surgery, df= degrees of freedom, IABP = intra-aortic
balloon pump, PCI = percutaneous coronary intervention.

Table 2 Operative Factors and Acute Kidney Injury After Cardiac
Surgery (CS-AKI)

                               No CS-AKI             AKI Stage 1
Operative Factors            54.8% (n=550)          30.3% (n=304)

Type of surgery

CABS                          198 (36.0%)            106 (34.9%)

Valve                         206 (37.5%)            117 (38.5%)

Combined                      132 (24.0%)             66 (21.7%)

Others                         14 (2.5%)              15 (4.9%)

Cardiopulmonary bypass        482 (88.6%)            300 (93.8%)

Circulatory arrest              1 (0.2%)               6 (2.0%)

LIMA graft                    229 (41.6%)            121 (39.8%)

Intraoperative                149 (26.6%)            128 (42.1%)

Cardiopulmonary bypass    96.0 (0.0 to 247.0)    98.0 (0.0 to 316.0)
time ([dagger])

Cross-clamp time          81.0 (0.0 to 199.0)    79.0 (0.0 to 212.0)

Time in operating room     5.61 [+ or -] 1.35     5.79 [+ or -] 1.44

                              AKI Stage 2            AKI Stage 3
Operative Factors             7.1% (at=71)           7.9% (n=79)

Type of surgery

CABS                           18 (25.4%)             20 (25.3%)

Valve                          37 (52.1%)             31 (39.2%)

Combined                       16 (22.5%)             22 (27.8%)

Others                          0 (0.0%)               6 (7.6%)

Cardiopulmonary bypass         82 (98.8%)             54 (96.4%)

Circulatory arrest              0 (0.0%)               5 (6.3%)

LIMA graft                     26 (36.6%)             25 (31.6%)

Intraoperative                 30 (42.3%)             56 (70.9%)

Cardiopulmonary bypass      102.0 (0 to 349)     114.0 (0.0 to 361.0)
time ([dagger])

Cross-clamp time          81.0 (30.0 to 204.0)   88.0 (35.0 to 305.0)

Time in operating room     6.03 [+ or -] 1.54     6.34 [+ or -] 1.80

Operative Factors            Statistic (df)         P

Type of surgery

CABS                           18.17 (9)          0.033




Cardiopulmonary bypass         16.32 (3)          0.001

Circulatory arrest             24.83 (3)          0.000

LIMA graft                      3.25 (3)          0.354

Intraoperative                 66.39 (3)          0.000

Cardiopulmonary bypass         15.83 (3)          0.001
time ([dagger])

Cross-clamp time                7.06 (3)          0.070

Time in operating room       7.26 (3, 1000)       0.000

Notes: Counts and relative frequencies are displayed for
categorical intra-operative factors, and p-values are based upon
the [chi square] test of independence. Mea and standard deviations
are displayed for continuous factors, and p-values are based upon a
one-way analysis of variance F-test unless noted otherwise.

CABS = coronary artery bypass surgery, LIMA = left internal mammary
artery, df= degrees of freedom.

([dagger]) Medians and ranges are displayed and significance
testing was conducted using the Kruskal-Wallis Htest.

Table 3
Postoperative Factors and Acute Kidney Injury After Cardiac
Surgery (CS-AKI)

                            No CS-AKI         AKI Stage 1
Postoperative Factors       54.8% (n=550)     30.3% (n=304)

Prolonged ventilation/      42 (7.6%)         39 (12.8%)

Reoperation for             9 (1.6%)          5 (1.6%)
postoperative bleeding

Coma/encephalopathy         9 (1.6%)          17 (5.6%)

Infection                   6 (1.1%)          8 (2.6%)

Postoperative atrial        126 (22.9%)       105 (34.5%)

Pneumonia                   16 (2.9%)         18 (5.9%)

                            AKI Stage 2       AKI Stage 3
Postoperative Factors       7.1% (n=71)       7.9% (n=79)

Prolonged ventilation/      15 (21.1%)        51 (64.6%)

Reoperation for             0 (0.0%)          5 (6.3%)
postoperative bleeding

Coma/encephalopathy         10 (14.1%)        27 (34.2%)

Infection                   1 (1.4%)          3 (3.8%)

Postoperative atrial        30 (42.3%)        34 (43.0%)

Pneumonia                   9 (12.7%)         25 (31.6%)

Postoperative Factors       Statistic (df)    P

Prolonged ventilation/      182.28 (3)        0.000

Reoperation for             10.04 (3)         0.018
postoperative bleeding

Coma/encephalopathy         132.31 (3)        0.000

Infection                   4.62 (3)          0.202

Postoperative atrial        27.77 (3)         0.000

Pneumonia                   94.67 (3)         0.000

Notes: Counts and relative frequencies are displayed for
categorical preoperative factors, and p-values are based upon the
%2 test of independence. Medians and ranges are displayed for the
continuous factor, and the associated p-value based upon the
Kruskal-Wallis H test.

Table 4 Sequential Logistic Regression of Predictors for Acute Kidney
Injury after Cardiac Surgery (CS-AKI)

Predictors                  CS-AKI Stage                Slopes

Preoperative risk factors

Smoker                      0 vs. 1 [less than or       0.721
                            equal to]

Smoker                      1 [greater than or equal    0.390
                            to] vs. 2 [less than or
                            equal to]

Smoker                      2 [greater than or equal    0.304
                            to] vs. 3

Hypertension                --                          -1.122

Operative risk factors

Cardiopulmonary bypass      --                          -0.857

Postoperative risk

Prolonged ventilation/      0 vs. 1 [less than or       -0.781
reintubation                equal to]

Prolonged ventilation/      1 [greater than or equal    -1.236
reintubation                to] vs. 2 [less than or
                            equal to]

Prolonged ventilation/      2 [greater than or equal    -1.577
reintubation                to] vs. 3

Postoperative blood         0 vs. 1 [less than or       0.656
product                     equal to]

Postoperative blood         1 [greater than or equal    1.122
product                     to] vs. 2 [less than or
                            equal to]

Postoperative blood         2 [greater than or equal    2.182
product                     to] vs. 3

Coma/encephalopathy         --                          -1.621

Predictors                  S.E.      Wald      P         OR

Preoperative risk factors

Smoker                      0.215     11.258    0.001     0.486

Smoker                      0.317     1.506     0.220     0.677

Smoker                      0.393     0.596     0.440     0.738

Hypertension                0.298     14.214    0.000     3.071

Operative risk factors

Cardiopulmonary bypass      0.313     7.512     0.006     2.358

Postoperative risk

Prolonged ventilation/      0.258     9.139     0.003     2.183

Prolonged ventilation/      0.277     19.845    0.000     3.441

Prolonged ventilation/      0.346     20.784    0.000     4.840

Postoperative blood         0.179     13.390    0.000     1.927

Postoperative blood         0.251     19.993    0.000     3.067

Postoperative blood         0.445     24.046    0.000     8.850

Coma/encephalopathy         0.324     25.048    0.000     5.051

Table 5 Postoperative Outcomes for Patients Developing Acute
Kidney Injury After Cardiac Surgery (CS-AKI)

                              No CS-AKI           AKI Stage 1
                            54.8% (n=550)        30.3% (n=304)

ICU LOS (hour)            90.2 (0 to 673.1)   102.5 (0 to 1013.5)
Postoperative LOS (day)    7.0 (0 to 42.0)      8.0 (0 to 44.0)
Hospital LOS (day)         7.0 (0 to 70.0)      9.0 (0 to 46.0)
Postoperative mortality       4 (0.7%)             3 (1.0%)
Home                         468 (85.6%)          230 (76.2%)
Extended care                 51 (9.3%)           39 (12.9%)
Nursing home                  21 (3.8%)            25 (8.3%)
Others                        7 (1.3%)             8 (2.6%)

                               AKI Stage 2
                              7.1% (at=71)

ICU LOS (hour)            145.6 (48.2 to 960.0)
Postoperative LOS (day)    10.0 (5.0 to 40.0)
Hospital LOS (day)         11.0 (5.0 to 40.0)
Postoperative mortality         6 (8.5%)
Home                           41 (63.1%)
Extended care                  17 (26.2%)
Nursing home                    7 (10.8%)
Others                          0 (0.0%)

                               AKI Stage 3
                               7.9% (n=79)         Statistic (df)

ICU LOS (hour)            256.5 (44.1 to 1198.6)     186.25 (3)
Postoperative LOS (day)     16.0 (2.0 to 76.0)       164.08 (3)
Hospital LOS (day)          19.0 (2.0 to 77.0)       140.04 (3)
Postoperative mortality         24 (30.4%)           182.94 (3)
Home                            22 (40.0%)            87.93 (9)
Extended care                   18 (32.7%)
Nursing home                    9 (16.4%)
Others                          6 (10.9%)


ICU LOS (hour)             0.000
Postoperative LOS (day)    0.000
Hospital LOS (day)         0.000
Postoperative mortality    0.000
Home                       0.000
Extended care
Nursing home

Notes: Counts and relative frequencies are displayed for
categorical preoperative factors, and p-values are based upon the
%2 test of independence. Medians and ranges are displayed for the
continuous factors, and the associated p-value based upon the
Kruskal-Wallis H test.

ICU = intensive care unit, df= degrees of freedom, LOS = length of
COPYRIGHT 2019 Jannetti Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Thanavaro, Joanne; Taylor, John; Vitt, Linda; Guignon, Mary S.
Publication:Nephrology Nursing Journal
Article Type:Report
Date:Jan 1, 2019
Previous Article:Quality of Life after Dietary Self-Management Intervention for Persons with Early Stage CKD.
Next Article:Use of an Educational Intervention to Improve Fluid Restriction Adherence in Patients on Hemodialysis.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters