The incidence, risk factors and outcomes of acute kidney injury in critically ill patients undergoing emergency surgery: a prospective observational study

This prospective observational research was managed in the general ICUs from Guangdong Provincial People’s Hospital. From January 2014 to March 2018, patients who were admitted to ICU immediately after undergoing noncardiovascular emergency surgery were included. Some patients were already in the ICU prior to the surgery, and some were transferred to the ICU after surgery. Those excluded patients conformed to the criteria that were younger than 18 years, refusal of consent, preexisting ESRD, presence of AKI before emergency surgery, or missing admission data. The primary outcome was defined as the occurrence of AKI according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria within 1 week after noncardiovascular emergency surgery. And the secondary outcome comprised postoperative duration of mechanical ventilation, postoperative reintubation, postoperative RRT during ICU stay, ICU and hospital mortality, length of ICU and hospital stay, as well as ICU and hospital costs. Following Strengthening the Reporting of Observational Studies in Epidemiology guidelines [20], written informed consent was offered to patients or surrogates for patients’ inability to consent. This research was authorized by the Ethics Committee and executed complying with the Declaration of Helsinki.

Clinical and demographic characteristics and outcomes of these patients were collected once they were admitted to the ICU. Age, gender, body mass index (BMI), preexisting clinical conditions [hypertension, diabetes mellitus, CKD, cerebrovascular disease, and coronary artery disease (CAD)], American Society of Anesthesiologist (ASA) classification, classification of New York Heart Association (NYHA) heart function, preoperative medication including the preoperative use of nephrotoxic drugs [nonsteroidal anti-inflammatory drug (NSAID), angiotensin-converting enzyme inhibitor (ACEI), angiotensin receptor blocker (ARB), immunosuppressant, aminoglycoside, vancomycin, acyclovir, or amphotericin] and the preoperative administration of radiographic contrast, surgery group (neurosurgical surgery, noncardiovascular chest surgery, abdominal surgery, or others), and incision type were registered. Comprising the level of preoperative hemoglobin, baseline serum creatinine (sCr), baseline estimated glomerular filtration rate (eGFR), and concentration of postoperative sCr, hemoglobin, and the lactic acid (LAC) at ICU admission, laboratory data were recorded. Serum creatinine and hemoglobin were detected both preoperation and at least once a day as a part of routine clinical care during ICU hospitalization. The hourly urine output (U.O.) of each patient was also recorded from enrollment to ICU discharge. The postoperative Acute Physiology and Chronic Health Evaluation (APACHE II) score, which was utilized to estimate the patient’s overall condition, was evaluated instantly after anesthesia recovery. Postoperative reoperation within 1 week after the first noncardiovascular emergency surgery was taken notes. Surgical data containing general anesthesia, duration of surgery, intraoperative estimated blood loss, lowest mean arterial pressure (MAP; i.e., lowest MAP for at least five continuous minutes) during anesthesia, radiographic contrast, intraoperative U.O., amount and type of intraoperative fluids administered (crystalloid and artificial colloid), intraoperative transfusions [red blood cells (RBCs), platelets, and plasma] were recorded. Prognosis variables were also recorded, comprising duration of postoperative mechanical ventilation, the incidence of postoperative tracheal reintubation and RRT, ICU and in-hospital mortality, length of stay in hospital and ICU, and total ICU and in-hospital costs.

AKI was diagnosed according to the KDIGO criteria [21] within 1 week after surgery. However, because U.O. criteria could be affected by administrating diuretics or obesity, we adopted serum creatinine to diagnose AKI. The eGFR was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) creatinine equation [22]. CKD was defined as baseline eGFR < 60 ml/minute/1.73 m2. The baseline sCr was defined as following rules in sequence as described in previous study: (1) the most recent pre-ICU value (between 30 and 365 days before ICU admission); (2) for patients aged < 40 years, a stable pre-ICU value > 365 days before ICU admission (stable defined as within 15% of the lowest ICU measurement); (3) pre-ICU value (> 365 days before ICU admission) and less than the initial sCr at ICU admission; (4) a pre-ICU value (between 3 and 39 days before ICU admission) ≤ initial sCr at the time of admission to ICU and not distinctly in AKI; if patients did not have serum creatine value before ICU admission, (5) the lowest sCr upon initial admission value, the final ICU value, or the minimum value at follow-up unto 365 days [23‐25]. Surgical incision was classified into three types, including clean wound (type I), relative clean wounds (type II) and contaminated wounds (type III).

All samples were collected simultaneously within 1 h after ICU admission and analyzed at the central laboratory of the Guangdong Provincial People’s Hospital utilizing a standard protocol. The concentrations of samples were measured using commercially available multiplex assays and enzyme-linked immunosorbent assays following the manufacturer’s instructions.

To estimate the multivariable regression coefficients, events per variable (EPV) > 10 was a significant problem [26]. EPV = 10 should be obligatory in this outcome model to avoid bias. Therefore, to meet with a model with 5 covariates, we needed to involve nearly 50 outcome events. With an approximated postoperative AKI incidence of 15%, which was found by previous studies that the incidence of AKI fluctuated from 0.8 to 39% due to different surgical types, we beforehand computed the sample size. Thus, a sample size of 334 cases was essential. Given a possible dropout rate of 10%, we should require at least 368 patients.

SPSS version 16.0 software program (SPSS Inc., Chicago, Illinois, USA) was used in the statistical analyses. A two-sided P-value of less than 0.05 was deemed as statistically significant. Mean ± standard deviation (S.D.), median and interquartile range (IQR) performed in continuous variables, while percentages were utilized to present categorical variables. In terms of continuous variables, a t-test was used to compare normally distributed variables. At the same time, the Wilcoxon rank-sum test was utilized in the comparison of non-normally distributed variables. Meanwhile, the chi-square test or Fisher’s exact test were used in the comparison of categorical variables. Univariate logistic regression analysis was performed to examine the relationship between each indicator and postoperative AKI separately. We also conducted multivariate logistic regression to evaluate the variables which were independently related to postoperative AKI. A criterion of P <  0.10 in the univariate analysis entered into multivariate analysis. Multivariate logistic forward stepwise regression was subsequently utilized to evaluate the most competent predictors of postoperative AKI. OR with 95% confidence intervals (C.I.s) was used to describe the results.

Figure 1 presented the protocol and flow diagram of the screening process. Among 412 patients enrolled for the study, 29 were excluded due to the subsequent reasons: younger than 18 years (n = 2), refusal of consent (n = 4), end-stage renal disease (n = 2), presence of AKI before emergency procedure (n = 14), and missing data (n = 7). Finally, a total of 383 patients were involved, among whom 67 (17.5%) were already in the ICU before surgery, and 316 (82.5) were admitted to the ICU after surgery. Of them, 151 (39.4%) patients occurred in postoperative AKI basing on the KDIGO criteria. Of those patients who evolved into postoperative AKI, 92 patients (60.9%) developed to stage 1, 40 patients (26.5%) were progressed to stage 2, and 19 patients were evolved into stage 3 (12.6%). Among the 151 postoperative AKI patients, 110 (72.8%) developed postoperative AKI on the first day after the operation, 25 (16.6%) on the second day, 6 (4.0%) on the third day, and 10 (6.6%) patients beyond 3 days. Thus, 93.4% of the patients reached postoperative AKI within 3 days after emergency operations.

As presented in Table 1, compared with non-AKI patients, patients with AKI had a significantly higher rate of preexisting clinical conditions of hypertension, and a larger proportion of patients in the AKI group underwent abdominal surgery than in the non AKI group. Moreover, both ASA classification and Classification of NYHA heart function were significantly higher in patients involving in postoperative AKI. Whereas, compared with postoperative AKI, there were no significant differences in age, sex, BMI, the preexisting clinical conditions (including diabetes mellitus, cerebrovascular disease, CAD), preoperative hemoglobin concentration, baseline sCr, baseline eGFR, preoperative medication of radiographic contrast, some varieties of surgery group (containing noncardiovascular chest surgery, others) and incision type in patients without postoperative AKI. Contrary to expectations, there was no marked difference between postoperative AKI patients and non-postoperative-AKI patients in the preoperative medication of nephrotoxic drugs.

Table 2 demonstrated the intraoperative parameters of patients in this cohort. During the operation, patients who developed postoperative AKI were accompanied by a longer duration of surgery, more estimated blood loss, and lower minimum MAP. Meanwhile, postoperative AKI patients had a higher rate of radiographic contrast. Additionally, the patients who accepted total artificial colloid or blood transfusion (RBCs, plasma, platelets) during the operation were more likely to develop postoperative AKI. However, the general anesthesia, intraoperative U.O., and total crystalloid of intraoperative fluids were inconspicuous for postoperative AKI.

Revealing by Table 3, patients with more potential to develop postoperative AKI had the following features: the higher APACHE II score reflected the severity of the disease and the patient’s overall situation, the higher concentration of the postoperative sCr and postoperative LAC concentration. In addition, patients who went through postoperative AKI had a greater likelihood of developing oliguria and need for undergoing reoperation after the first emergency surgery.

Univariate analysis showed that hypertension, ASA classification, classification of NYHA heart function, preoperative hemoglobin, neurosurgical surgery, abdominal surgery, duration of surgery, estimated blood loss, minimum MAP, radiographic contrast, intraoperative U.O., total infused artificial colloid, RBCs or plasma, postoperative APACHE II score, postoperative sCr, postoperative hemoglobin, postoperative U.O., postoperative lactic acid, and postoperative reoperation were risk factors that are associated with postoperative AKI (Table S1). Since postoperative sCr and U.O. are a manifestation and a diagnostic criterion for AKI, postoperative sCr and U.O. were not included in the multivariate study to analyze whether it was a risk factor for AKI. As for postoperative AKI, postoperative reoperation, postoperative APACHE II score, and postoperative LAC concentration were the independent risk factors after multivariable adjustment, which was shown in Table 4. Postoperative reoperation was an independent risk factor of postoperative AKI with the adjusted OR (ORadj) of 1.854 (95% CI, 1.091–3.152). At the same time, we discovered that higher postoperative APACHE II score and postoperative LAC could be used to predict postoperative AKI incidence with [ORadj 1.059 (95% CI, 1.018–1.102)] and [ORadj 1.239 (95% CI, 1.047–1.467)], respectively.

Elucidated in Table 5, the occurrence of postoperative AKI would lead to higher rates of postoperative RRT, ICU mortality, and in-hospital mortality. Moreover, patients with postoperative AKI were more likely to go through the long duration of postoperative mechanical ventilation, prolonging ICU and hospital length of stay, higher total ICU costs, and higher hospital ICU costs. But there was no marked difference between postoperative AKI and reintubation.