Acute kidney injury (AKI) is a challenging clinical syndrome with a high impact on the health and social care system.1,2 Its incidence is increasing progressively. Two population studies conducted in the last 22 years showed an increase in AKI cases of around 10% per annum.3,4 Postoperative AKI is a frequent complication after major surgery and is the second leading cause of hospital-acquired AKI.5 In the 1990s, an epidemiological study conducted in Madrid (Spain) established a connection between AKI and surgery.6 Although the incidence of AKI has been widely studied in cardiac surgery, reporting incidences between 0.3 and 29.7%,7 few studies have focused on AKI after major, open abdominal surgery. The use of the term ‘noncardiac surgery’ in the medical literature, referring to a wide variety of surgical procedures, including vascular surgery, makes it more difficult to identify the actual incidence of AKI after major, open abdominal surgery. The development of AKI lengthens hospital stay,8,9 has a deleterious effect on prognosis, and increases mortality rate.9,10 It is also known that a previous episode of AKI implies a higher risk of developing chronic kidney disease in the future.11 Therefore, as AKI is a common and severe syndrome, preventive measures should be found urgently.
In attempts to prevent or treat surgery-associated kidney dysfunction during the perioperative period many drugs have been used, but most have yielded little benefit.12–14 There is thus no effective preoperative prophylactic measure which can be recommended for the prevention of postoperative AKI.15 The best results in AKI prevention have been obtained in the context of contrast-induced nephropathy (CIN).16 In this scenario, hydration with either 0.9% normal saline or sodium bicarbonate, before administration of radiologic contrast medium, is effective. In relation to surgery, while identifying AKI risk factors and maintaining a normal blood volume to preserve adequate renal perfusion are still key elements for preventing AKI,17,18 hydration with intravenous fluids before surgery may help to avoid preoperative volume depletion and thus contribute to the maintenance of blood volume.
The primary aim of this trial was to explore whether an easy and inexpensive intervention such as the preoperative intravenous administration of 0.9% normal saline could lower the incidence of AKI during the first week after major abdominal surgery. Secondary endpoints were to establish the incidence of AKI in our population, to determine the needs for renal replacement therapy (RRT) and admission to intensive care, and to ascertain the mortality rate associated with elective major, open abdominal surgery.
Ethical approval for this study (Ethical Committee CI 196/05) was provided by the Institutional Ethics Committee of University Hospital Ramon y Cajal, Madrid, Spain (Chairperson M.A. Galvez on 15 December 2005). Written informed consent was obtained from all participants. This was a phase III, prospective, randomised, open-label, single-centre clinical trial. Patients were randomly assigned to either the treatment or control group (Co group).
Adult patients able to consent were included if they had been scheduled for major, open abdominal surgery and were classified in grades II to IV of the American Society of Anaesthesiologist Physical Status Scoring system. Exclusion criteria included pregnancy, laparoscopic or minor or urgent surgery, chronic kidney disease patients with a baseline serum creatinine (sCr) at least 3.0 mg dl−1, and patients having any disease that could result in volume overload with our trial protocol. Recipients of any preoperative intravenous hydration that had not been defined in our study protocol were also excluded. Patients could eat or drink up to 8 h before the surgery. Participants in the treatment group were administered 0.9% normal saline at a dose of 1.5 ml kg−1 h−1 (actual body weight) over the 12 h before surgery. All study participants were scheduled as the first case of the day. Patients assigned to the Co group did not receive any fluid intervention. Both groups received standard intraoperative management. This included active warming, basic cardiovascular monitoring, measures to preserve adequate oxygenation, a pre-specified transfusion trigger for packed red blood cells (8.5 g dl−1), and postoperative pain management. Haemodynamic stability (defined as a mean arterial pressure over 60 mmHg and kept within ±20% of the baseline value measured in the operating room before induction of anaesthesia) was accomplished with the use of vasodilators or vasoconstrictors as required. Owing to the pragmatic design of the trial, other aspects of the intraoperative and postoperative care of the patients, including anaesthetic technique, type and volume of fluids administered and additional monitoring were left to the discretion of the anaesthesiologist.
Participants were assigned to their study group from a randomisation list with blocks of n = 6 (nQuery version 7.0, Statistical Solutions Ltd, 4500 Airport Business Park, Cork, Ireland). Group allocation was concealed by means of sealed opaque envelopes, numbered consecutively, and which were opened only after a patient had been enrolled into the study. This clinical trial design did not include any blinding of staff to the treatment group.
Sample size calculation was based on a 7% estimated incidence of AKI, whose severity was defined according to the RIFLE creatinine criteria [‘R’ (risk of kidney injury), ‘I’ (actual kidney injury), ‘F’ (some failure of renal function), ‘L’ (loss of renal function) and ‘E’ (end-stage renal disease)],19 as predicted from the results of a retrospective analysis of 100 patients who had undergone major, open abdominal surgery at our institution and who required admission to the ICU after surgery (unpublished data). We hypothesized that preoperative hydration could prevent low-grade (‘Risk’) AKI cases (such low-grade cases comprised over 70% in the retrospective sample), with an absolute risk reduction of 5%. To demonstrate this with a two-sided test, a 5% significance level, 80% power and assuming a 10% loss to follow-up, it was necessary to enrol 300 patients per group.
AKI was diagnosed using the RIFLE and the AKI network (AKIN) creatinine criteria.19,20 We did not use the urinary output criterion because collecting urine in patients admitted to the ward after surgery could not be performed accurately.
For every patient recruited, demographic and anthropometric features, Charlson comorbidity index, American Society of Anaesthesiologist Physical Status Scoring system score, type of surgery, use of bowel preparation, drugs taken the week before surgery and in the postoperative period, baseline haemoglobin and blood chemistry data, procedure duration, presence of haemodynamic instability and the need for vasoactive drugs and/or red blood cell transfusion together with the type of anaesthesia, the fluid balance and intraoperative urine output were recorded. Preoperative sCr was the value obtained nearest to the surgical date (usually measured the day before surgery) and was available for all the patients. Postoperative data were recorded for 24 h, 48 h, 72 h and 1 week after surgery. To analyse the safety of the intended intervention, the adverse events that occurred in both the groups were recorded and compared. For this purpose we amalgamated various adverse events into five categories: haemodynamic (hypotension, rhythm disturbances, heart failure and angina), respiratory (dyspnoea, respiratory insufficiency and respiratory infection), abdominal (anastomotic breakdown, intestinal obstruction, haematoma and abdominal bleeding), infectious (any fever or infection), miscellaneous (anaemia, hyponatraemia, hypokalaemia, hyperchloraemia, nausea and vomiting, pain hyperglycaemia, diarrhoea and acute confusional syndrome).
Absolute and relative frequencies were used to describe categorical variables. Mean and SD were used to describe continuous variables for normally distributed data and medians and interquartile ranges for non-normally distributed data. The analysis was performed on an intention to treat basis. To assess the primary endpoint, the number of AKI events in both groups was compared within the first week after surgery, using the univariate χ2 test or the Fisher's exact test as appropriate. For continuous variables, the Mann–Whitney test was used instead. In those cases where we did not record all the values regarding a variable, a data imputation strategy was applied, that is the worst value obtained for the variable during the study was used. Bonferroni[Combining Acute Accent]s correction was applied for multiple comparisons. P values less than 0.05 were regarded as statistically significant. All analyses were performed with Stata 13.1 (StataCorp.2013. Stata Statistical Software: Release 13, College Station, Texas, USA).
We assessed 769 patients scheduled for elective major, open abdominal surgery, from June 2006 to February 2011 (Fig. 1). At the planned interim analysis an independent data monitoring committee reviewed the results and reported that the actual incidence of AKI assessed by the RIFLE creatinine criteria was lower than expected. As the data of the two groups were considered to be well balanced, it was deemed that the trial would be futile in achieving its goals and the principal investigator was advised to stop the recruitment, close the trial and communicate the results. The trial was terminated after 336 patients had been recruited. Of all such patients, 170 were assigned to the Co group and 166 were assigned to the 0.9% normal saline treatment group. For the statistical analysis, only 328 patients were considered: seven participants were excluded after randomisation as their surgery was cancelled, and one randomisation envelope was lost (Fig. 1). For calculating the incidence of AKI only data from the 289 patients with at least one postoperative sCr value available during the first postoperative week were considered.
At baseline, the two groups were well balanced in terms of sex, anthropometric data, co-morbidities, drugs taken before surgery and bowel preparation (Table 1). As regards the type of surgery, there were no differences between the two groups (Table 1). Other baseline data were similar in both the groups (Table 2).
The mean surgery duration was 203 min (SD 101 min) without a significant difference between groups. Furthermore, there were no differences between the groups regarding haemodynamic stability, intraoperative urine output, need for vasoactive drugs, type of anaesthesia, use of furosemide, use of NSAIDs or red blood cell transfusion requirements during surgery. Estimated intraoperative fluid loss (urine output, nasogastric tube output, blood loss and insensible losses) was significantly higher in the Co group. Global fluid balance, measured as the difference between fluid replacement and estimated fluid loss, was lower in the Co group (Table 3). Drugs administered to patients during the postoperative period were similar in both the groups (Table 4).
During the first postoperative week AKI, as measured with RIFLE creatinine criteria, had an overall incidence of 4.8% (4.7% in the 0.9% normal saline group and 5.0% in the Co group) and, as measured with the AKIN criteria, 9.7% (11.4% in the 0.9% normal saline and 7.9% in the Co group). There was no statistically significant difference between the groups. Absolute risk reductions [95% confidence interval (CI)] were −0.3% (−5.3 to 4.7%) for RIFLE and 3.5% (−10.2 to 3.6%) for AKIN creatinine criteria. All 14 patients were diagnosed as having AKI using the RIFLE creatinine criteria and 28 patients were classified with the AKIN creatinine criteria. There were no statistically significant differences when comparing stages of AKI between the 0.9% normal saline and Co groups (Table 5). After adjusting for the overall perioperative fluid balance, the difference remained statistically non-significant. In the sub-group of patients who had undergone bowel preparation, again we found no statistically significant differences between the control and 0.9% normal saline groups. Among our patients, 44.5% required postoperative ICU admission, with no significant difference between the groups (48.1% for the 0.9% normal saline group and 40.9% for the Co group). Two study participants (0.7%), one in each group, required RRT during the first week after surgery. There were no differences in the complications between the groups (Table 6). After Bonferroni[Combining Acute Accent]s correction, only pain was found to be more frequent in the Co group (3.1 vs. 13.8%; P = 0.01) (see Table 7, supplementary appendix, http://links.lww.com/EJA/A84). There was no difference in length of hospital stay between groups and no difference in the mortality rates during hospital admission: 2.4%, (3.7% for normal saline group, 1.2% for Co group).
In this clinical trial of 328 cases of major, open abdominal surgery, randomised to receive either no fluid or intravenous 0.9% normal saline over a 12 h period before surgery, we were unable to find any difference in the incidence of postoperative AKI over the first postoperative week.
In our study population, the AKI incidence, as measured with RIFLE creatinine during the first week after surgery, was 4.8%, much lower than the 7% estimate used in our study design. Using AKIN criteria, the AKI incidence reached 9.7%. Thus, in our trial, AKIN sCr criteria were more sensitive than RIFLE sCr criteria in detecting AKI, (especially in the mildest stages of this syndrome). However, whether the RIFLE or AKIN sCr criteria was used we were unable to demonstrate any difference in outcomes and our hypothesis that 0.9% normal saline would be protective could not be supported.
In the OPTIMISE trial21 of high-risk patients undergoing gastrointestinal surgery, the incidence of AKI defined as AKIN grade 2 or higher was 4.7%. In the current study, AKIN grade 2 or higher was observed in only 1.7% (Table 4). This is probably as a result of our patients being healthier and at lower risk of AKI. Among patients who developed AKI, two (0.7%) required RRT. We were unable to find any published information on the need for RRT in patients undergoing elective major, open abdominal surgery. The mortality rate in our study was 2.4%, again with no statistically significant difference between the groups. Published mortality rates depend on both the AKI severity and the actual procedure being performed within the abdomen.9,22–24 Among all the adverse events recorded in our study, only pain was statistically different between the groups but this was without clinical relevance.
Several strategies have been investigated for prophylaxis of AKI in the postoperative period, but none have been found to be clearly effective. As regards AKI prophylaxis after abdominal surgery, few studies have compared one fluid with another or with a placebo25 but it seems that target-oriented fluid therapy, with or without dobutamine, lowers the incidence of postoperative AKI.14,21,26,27 To minimise AKI, the literature offers only non-specific recommendations such as adequate blood volume expansion with intravenous fluids, avoiding both under-hydration and over-hydration.14–16,27 For CIN prevention, studies have shown that intravenous hydration, before exposure to the contrast agent, is the most effective strategy, and intravenous volume expansion with either 0.9% normal saline or a sodium bicarbonate solution is recommended (recommendation grade 1A).28
Our study was designed to test a similar easy intervention in patients undergoing what could be described as fairly typical/routine major, open abdominal surgery. When considering the lack of effect of 0.9% normal saline in our sample there are some points to keep in mind.
First, hydration may not have been effective because volume depletion after an 8-h fasting period is minimal. However, even in the sub-group of patients with bowel preparation, who presumably had a significant volume deficit, we observed no benefit from the 0.9% normal saline. In this latter sub-group during the first week after surgery, AKI measured with the RIFLE sCr criteria had an overall incidence of 4.8% (4.3% in the 0.9% normal saline group and 5.4% in the Co group) and, as measured with the AKIN sCr criteria, 7.0% (4.3% in the 0.9% normal saline and 9.8% in the Co group), with no statistically significant differences between the groups. Whether the assessment of preoperative volume status by some of the dynamic parameters currently available would have altered the outcome is unknown but, as these were not available locally at the inception of our trial, we investigated an inexpensive and easy to implement intervention.
Second, perioperative fluid therapy may have neutralised any fluid deficit differences between the groups. Although the duration of surgery was similar in the two groups we did not record fluid therapy during the postoperative period. Thus it is not known whether uncontrolled fluid administration during the intraoperative and postoperative periods resulted in a greater volume being given to patients in the Co group. However, it seems to us rather unlikely because postoperative hydration was guided pragmatically by urine output and haemodynamic parameters, and not with reference to the study.
A third reason for the lack of efficacy observed may be the type of fluid selected for hydration. When we chose 0.9% normal saline as the hydration solution for our study, clinical evidence regarding the potential adverse effects of this chloride-rich solution compared with other balanced solutions were scarce or lacking.29,30 The use of 0.9% normal saline as a fluid replacement therapy could be considered a confounding factor in the development of AKI but, although hyperchloraemia and acidosis after normal saline resuscitation is well documented, these confounders usually do not lead to clinical consequences.31 In our study population, the pH values, 24 h after surgery, were within the normal range in 81.4% of the patients, and these values were almost identical between the two groups (82.7%, 0.9% normal saline group and 80.1%, Co group). Considering hyperchloraemia, we did not find a statistically significant difference between the groups (Table 6). Although 0.9% normal saline may not be considered the ideal fluid for resuscitation, it must also be appreciated that there are specific problems associated with the alternatives.31–33
Fourth, the population selected comprised mostly patients without high-baseline comorbidity, and the AKI incidence was lower than expected. It is possible that the patients chosen for the study were not the patients who may have gained the largest benefits from 0.9% normal saline prophylaxis. While restricting our study to patients with a higher preoperative risk of AKI may have been more informative, we designed this trial looking for a simple but effective preventive strategy that could be beneficial for most inpatients.
Finally, we consider that the pathophysiology of contrast-induced AKI is, to some extent, different to postoperative AKI. Prophylactic measures which are effective against CIN, such as hydration with either 0.9% normal saline or sodium bicarbonate, do not work in surgical patients. Indeed, in cardiac surgery, attempts to use sodium bicarbonate solutions as prophylaxis for postoperative AKI has been associated with an increase in mortality.33
Some strengths and limitations to this study must be considered. One of the strengths is that it is a prospective randomised controlled trial of moderate size, investigating an easily applied intervention that has been found beneficial in other clinical setting (CIN). A significant limitation is that we did not include urine output criteria. But it is well known that assessing urine output is impractical outside the ICU, where few patients have a urinary catheter. In the absence of accurate urine output we could not record data on postoperative fluid balance. We accept that this may have had some influence on the AKI incidence but we found this task difficult to achieve, and not in line with the pragmatic nature of our trial. Another limitation is that our study was open label and single-centred. Although we considered the possibility of blinding the anaesthesiologist to the preoperative orders, in the end we did not do so, and blinding of the patients was unachievable. Finally, the clinical management of the study participants included in our trial could be viewed as outdated. Certainly, during the last few years many institutions, including our own, have implemented ‘enhanced recovery after surgery’ protocols for patients undergoing abdominal surgery. The rates of AKI we observed seem atypical, but these were common surgical procedures, generalisable to other institutions, and because of this our outcomes should be investigated in similar studies in other institutions.
In conclusion, we have shown that in our population, the incidence of AKI after elective major, open abdominal surgery is low and generally mild. Preoperative intravenous hydration with 0.9% normal saline was not effective in preventing renal dysfunction in patients with moderate morbidity and normal baseline sCr. Considering the lack of data related to AKI incidence after major elective open abdominal surgery, our observation may be helpful in the design of future trials.
Acknowledgements relating to this article
Assistance with this study: the authors thank Ana Castro, Data Manager at the Ramon y Cajal University Hospital in Madrid, for her continuous support during the study, and Dr Alfonso Sanjuanbenito, MD, for proof reading and linguistic advice. They would also acknowledge the comments and suggestions of the anonymous reviewers, who contributed to improve this article.
Financial support and sponsorship: this work was funded by the ‘Instituto Carlos III’, Spain (FIS: ECO007/90461) and has also received a grant from the ‘Ministerio de Economía y competitividad’, Spain (MINECO – Grant EC10-103).
Conflicts of interest: none.
Presentation: part of these data was presented as a poster at the ‘ASN Kidney Week’, November 2011, Philadelphia.
1. Hsu C-Y, McCulloch CE, Fan D, et al. Community-based incidence of acute renal failure. Kidney Int
2. Liangos O, Wald R, O[Combining Acute Accent]Bell JW, et al. Epidemiology and outcomes of acute renal failure in hospitalized patients: a national survey. Clin J Am Soc Nephrol
3. Hsu RK, McCulloch CE, Dudley RA, et al. Temporal changes in incidence of dyalisis-requiring AKI. J Am Soc Nephrol
4. Xue JL, Daniels F, Star RA, et al. Incidence and mortality of acute renal failure in Medicare beneficiaries, 1992-2001. J Am Soc Nephrol
5. Carmichael P, Carmichael AR. Acute renal failure in the surgical setting. Aust N Z J Surg
6. Liaño F, Pascual J. The Madrid Acute Renal Failure Study Group. Epidemiology of acute renal failure: a prospective, multicenter community-based study. Kidney Int
7. Burns KE, Chu MW, Novick RJ. Perioperative N-acetylcisteine to prevent renal dysfunction in high-risk patients undergoing CABG surgery: a randomized controlled trial. JAMA
8. Chertow GM, Burdick E, Honour M, et al. Acute kidney injury, mortality, length of stay, and costs in hospitalized patients. J Am Soc Nephrol
9. Hoste EA, Schurgers M. Epidemiology of acute kidney injury: how big is the problem? Crit Care Med
2008; 36 (4 suppl):s146–s151.
10. Coca SG, King TK, Rosenthal RA, et al. The duration of postoperative acute kidney injury is an additional parameter predicting long-term survival in diabetic veterans. Kidney Int
11. Bucaloiu ID, Kirchner HL, Norfolk ER, et al. Increased risk of death and de novo chronic kidney disease following reversible acute kidney injury. Kidney Int
12. Patel NN, Rogers CA, Angelini GD, et al. Pharmacological therapies for the prevention of acute kidney injury following cardiac surgery: a systematic review. Heart Fail Rev
13. Venkataraman R. Can we prevent acute kidney injury? Crit Care Med
14. Reddy VG. Prevention of postoperative acute renal failure. J Postgrad Med
15. Zacharias M, Muqawar M, Herbison GP, et al. Interventions for protecting renal function in the perioperative period. Cochrane Database Syst Rev
2013; doi: 10.1002/14651858.CD003590.pub4 http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD003590.pub4/abstract
16. Weisbord S, Palevsky PM. Prevention of contrast-induced nephropathy with volume expansion. Clin J Am Soc Nephrol
17. Jones DR, Lee HT. Perioperative renal protection. Best Pract Res Clin Anaesthesiol
18. Brienza N, Giglio MT, Marucci M. Preventing acute kidney injury after noncardiac surgery. Curr Opin Crit Care
19. Bellomo R, Ronco C, Kellum JA, et al. the ADQI workgroup. 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. Crit Care
20. Metha RL, Kellum JA, Shah SV, et al. Acute Kidney Injury Network (AKIN): report of an initiative to improve outcomes in acute kidney injury. Crit Care
21. Pearse RM, Harrison DA, McDonald N, et al. Effect of a perioperative, Cardiac Output-guided hemodynamic therapy algorithm on outcomes following major gastrointestinal surgery a randomized clinical trial and systematic review. JAMA
22. Abosaif NY, Tolba YA, Russell J, et al. The outcome of acute Renal Failure in the Intensive Care Unit according to RIFLE: model application, sensitivity, and predictability. Am J Kidney Dis
23. Teixeira C, Rosa R, Rodrigues N, et al. Acute kidney injury after major abdominal surgery: a retrospective cohort analysis. Crit Care Res Pract
24. Minjae K, Joanne EB, Guohua L. Variations in the risk of acute injury across intra-abdominal surgery procedures. Anesth Analg
25. Kellum JA, Cerda J, Kaplan LJ, et al. Fluids for prevention and management of acute kidney injury. Int J Art Org
26. Gang TJ, Soppitt A, Maroof M, et al. Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology
27. Brienza N, Giglio MT, Marucci M, et al. Does perioperative hemodynamic optimization protect renal function in surgical patients? A meta-analytic study. Crit Care Med
28. Lameire N, Kellum JA. for the KDIGO AKI Guideline Work Group. Contrast-induced acute kidney injury and renal support for acute kidney injury and renal support for acute kidney injury: a KDIGO summary (Part 2). Crit Care
29. Shaw AD, Bagshaw SM, Goldstein SL, et al. Major complications, mortality, and resource utilization after open abdominal surgery. 0.9% saline compared to Plasma-Lyte. Ann Surg
30. Lobo DN, Awad S. Should chloride-rich crystalloids remain the mainstay of fluid resuscitation to prevent ‘prerenal’ acute kidney injury? Con Kidney Int
31. Ince C, Groeneveld J. The case for 0.9% NaCl: is the undefendable, defensible? Kidney Int
32. Morgan TJ. The ideal crystalloid: what is balanced? Curr Opin Crit Care
33. Haase M, Haase-Fielitz A, Plass M, et al. Prophylactic perioperative sodium bicarbonate to prevent acute kidney injury following open heart surgery: a multicenter double-blinded randomized controlled trial. PLoS Med
2000; 10:e1001426[accessed 21/01/2016].