Rishabh Singha, b · Joanna Dodkinsa, c · James F. Doylec, d · Lui G. Fornia, c, e
aRoyal Surrey County Hospital NHS Foundation Trust, Guildford, Surrey, bMinimal Access Therapy Training Unit, Post Graduate Medical School, University of Surrey, Manor Park, Guildford, Surrey, cDepartment of Intensive Care Medicine and Surrey Peri-Operative Anaesthesia and Critical Care Collaborative Research Group, Guildford, Surrey, dDepartment of Intensive Care Medicine, Royal Brompton and Harefield NHS Foundation Trust, London, and eDepartment of Clinical and Experimental Medicine, Faculty of Health and Medical Sciences, University of Surrey, Guildford, UK
______________________
Abstract
The definition of acute kidney injury (AKI) is now well established and encompasses changes in both the urine output and the serum creatinine (SCr) over time. Many studies to date have concentrated solely on the SCr criteria, as this is relatively easy to monitor, given that accurate urine output is rarely measured outside critical care areas. However, many studies have emphasised the inadequacies of SCr in highlighting potential renal injury in a timely fashion. These limitations reflect not only acute changes in creatinine metabolism in the critically ill but also the kinetics of creatinine generation that may hinder early recognition of AKI. In turn, this may prevent judicious intervention promoting the misconception that little can be done for patients with this devastating condition except treating the consequences. Such observations have led to much research focused on identifying early indicators of AKI that may enable early treatment and hopefully lead to improved outcomes. This explains in part the reasoning behind the interest in biomarkers of AKI and which may see them develop a role as part of established clinical tool(s) in both the assessment of severity of AKI and the potential to assess recovery. However, much of the effort behind biomarker research has focused on the ability of such candidate molecules to predict AKI as defined by the imperfect gold standards used currently. It may be that the presence of renal biomarkers associated with renal stress or injury in isolation dictates poor outcomes and as such may provide diagnostic certainty in their own right.
© 2018 S. Karger AG, Basel
Introduction
The introduction of more robust definitions of acute renal failure including the Risk, Injury, Failure, Loss, End-stage kidney disease criteria and, more recently, the Kidney Disease: Improving Global Outcomes (KDIGO) criteria have revolutionized the field of critical care nephrology [1–3]. The introduction of the concept of acute kidney injury (AKI) has enabled identification and classification of critically ill patients with AKI who previously may have been labeled as having acute renal failure or whose renal function may have been dismissed as a relatively benign consequence of the underlying pathological process [4, 5]. The success of this approach is borne by the explosion of new data describing this syndrome and the consequences thereof [1, 6]. This has proved particularly enlightening with regard to the long-term complications of this syndrome [7–9]. The main drivers behind this “rebranding” of acute renal failure stemmed from the desire to introduce robust criteria to enable research questions to be answered coupled with the realization that even relatively small changes in serum creatinine (SCr) translate into a significantly poorer outcome for patients [10]. However, the classification of AKI has not been without criticism [11–13]. This is driven, in part, by the use of the readily available classical measures of renal function (namely SCr and/or urine output) as the diagnostic criteria by which AKI is defined both of which are associated with significant inadequacies [14, 15].
SCr and Urine Output: The Problems
Both the urine output and the SCr have inherent flaws when applied to the acute situation. A fall in urine output may be either a normal physiological response to stressors or a pathological response to renal injury [16–18]. Where the urine volume is such that the normal solute load cannot be cleared, this will be reflected in classical changes in serum electrolytes, acid base parameters and traditional “markers” of renal function including SCr and urea. As such, urine output may be of limited sensitivity and specificity in defining AKI as is reflected in the poor performance of urinary biochemistry particularly in the intensive care unit (ICU) setting [19]. Moreover, the clinical response to isolated oliguria may result in unnecessary intervention such as inappropriate fluid therapy and therefore, the presence of oliguria should always be interpreted within the clinical context as a whole [20]. This is not to advocate ignoring the urine output: the transition from physiological to pathological oliguria is unlikely to be stepwise. Therefore, detection of renal dysfunction may be made earlier if oliguria is noted. For example, in a study on critically ill patients, the diagnosis of AKI based solely on the presence of oliguria found that approximately 50% of these patients proceeded to develop AKI based on SCr criteria 24 h later [21]. Similarly, in a mixed ICU cohort, 69% of oliguric patients subsequently developed creatinine criteria for AKI diagnosis [22]. These results imply that not all patients with oliguria develop changes in SCr and as such using urine output criteria will increase the number of patients classified as having AKI. Furthermore, in some cases, this may identify a different cohort of patients. What is clear is that where the AKI criteria are satisfied using both SCr and urine output, this portends a worse outcome [23].
The SCr and the derived glomerular filtration rate provide the mainstay for outpatient management of chronic kidney disease. Creatinine is neither reabsorbed nor metabolised by the kidney (although tubular secretion does occur) and is freely filtered across the glomerulus and as such serves as a marker of renal function. However, the use of the SCr as a marker for AKI is far from ideal. Although the formation of creatinine is relatively constant under most conditions outside the ICU, this is not the case in the critically ill or where there is considerable metabolic disturbance as is found in many causes of AKI. Changes in SCr may overestimate renal function as a result of secretion in the proximal tubule and increases can be seen following the administration of medications that inhibit tubular secretion despite no change in renal function. SCr metabolism may also be affected by conditions common to many ICU patients including sepsis that may reduce production rates considerably [24, 25]. In addition, creatinine is distributed in the total body water and may, therefore, be affected by variations in volume status although this is unlikely to be of any major practical significance [26].