University of Groningen Acute Kidney Injury in critically ill patients Wiersema, Renske

(1)

University of Groningen

Acute Kidney Injury in critically ill patients

Wiersema, Renske

DOI:

10.33612/diss.133211862

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Wiersema, R. (2020). Acute Kidney Injury in critically ill patients: a seemingly simple syndrome. University of Groningen. https://doi.org/10.33612/diss.133211862

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

547452-L-bw-Wiersema 547452-L-bw-Wiersema 547452-L-bw-Wiersema 547452-L-bw-Wiersema Processed on: 25-8-2020 Processed on: 25-8-2020 Processed on: 25-8-2020

Processed on: 25-8-2020 PDF page: 177PDF page: 177PDF page: 177PDF page: 177 177

CHAPTER 10 Summary

General discussion

Future perspectives

(3)

(4)

Summary

Current evidence on Acute Kidney Injury and venous congestion

Up to only a decade ago, Acute Kidney Injury (AKI) in critically ill patients seemed to be a straightforward and low impact problem, often primarily caused by insufficient perfusion, or low cardiac index. In chapter 2 we showed that cardiac index was not associated with AKI, while variables of right ventricular function were. This led to the hypothesis that in contrast with cardiac afterload venous congestion might be one major contributing factor to AKI in the critically ill.

Designing an observational study to investigate AKI and venous congestion

Before evaluating the additional contribution of venous congestion as risk factor to existing known risk factors for the development of AKI, a baseline prognostic model for AKI needs to be established. In chapter 3 we identified a set of baseline (easily and immediately available) clinical signs which are independently associated with AKI. This retrospective analysis did not only identify the variables to be included and collected in the prospective study, but may also inform the identification of high-risk patients in settings where more sophisticated techniques are unavailable. In chapter 4 we described the protocol of a prospective observational study to assess the association between AKI and venous congestion; the Simple Intensive Care Studies-II (SICS-II).

Two complex clinical syndromes

At first glance AKI may seem, despite its complex pathophysiology, a rather straight forward binary outcome or complication of critical illness. When applying the Kidney Disease Improving Global Outcome (KDIGO) definition, different methods may lead to various incidences, which was elaborated in chapter 5. No definition exists to describe the concept of venous congestion, which is why we measured multiple proxies that could potentially approximate this complex state of system failure. In chapter 6, we studied arterial and venous renal Doppler ultrasound, and showed that both had limited diagnostic accuracy for AKI. Finally, all assessed signs of venous congestion were combined in chapter 7, in which we studied the association between all repeated collected signs of venous congestion and AKI. Fluid balance, the presence of peripheral edema and reduced right ventricular function were associated with AKI.

Heterogeneity of AKI and associated outcomes

AKI is a complex entity. First, variation according to differences in the applied definition complicates its research (chapter 5). Second, the syndrome itself develops due to many potential different pathophysiologic pathways, and it is thus plausible that specific pathways are more prominent in distinct subphenotypes of AKI. In chapter 8, a post-hoc analysis of the FINNAKI study, two different subphenotypes of septic AKI were identified which had different characteristics regarding renal recovery and 90-day mortality.

Finally, these different causes of heterogeneity likely contribute to varying associations between AKI and mortality. In chapter 9 we propose AKI burden, a variable which captures both the duration and the severity of AKI, and we show that AKI burden may improve the prediction of 90-day mortality compared to simplified variables like any AKI, severity of AKI or duration of AKI only.

(5)

(6)

General discussion

The underlying hypothesis of this thesis was that venous fluid overload could lead to increased renal ‘’afterload’’ or increased renal venous pressure, and in that way, contribute to the development or aggravate the severity of AKI. During the conduct of this thesis, multiple intervention studies have been initiated and conducted which investigate whether restrictive fluid therapy is associated with lower AKI occurrence. We attempted to study venous congestion and AKI, and their association using diagnostic and prognostic approaches in combined cross sectional and cohort study designs. The evaluation of venous congestion and AKI and their association was much more complex as anticipated. First, AKI is a very complex syndrome defined by criteria which are open to various interpretations. Second, the concept of venous congestion is still undefined and cannot be exactly measured or sufficiently estimated, while it is probably best defined by venous fluid overload approximated by various surrogate variables, such as elevated right atrial pressure or peripheral edema.

Acute Kidney Injury, a seemingly simple syndrome

AKI is a prevalent complication of critical illness, and incidences range from 14% to 75% in cohorts depending on selection criteria and application of the criteria used. There is no specific treatment other than supportive care once AKI occurs, so the current aim is to prevent the occurrence of AKI. Hence, accurate prognostic models are needed to identify the patients at risk for AKI, as described throughout this thesis.

Classification of AKI

The Kidney Disease Improving Global Outcomes (KDIGO) criteria are employed to define AKI, but can be applied and interpreted in several ways. In randomized clinical trials, patients are selected based on the occurrence of AKI, or the outcome of AKI is reported based on these criteria, so it is extremely important to have a consensus about how AKI can be unambiguously defined and reported. In the Conservative vs. Liberal Approach to Fluid Therapy of Septic Shock in Intensive Care (CLASSIC) trial, which was designed to evaluate whether either restrictive or (more) liberal fluid administration is beneficial in patients with septic shock, one of the secondary aims is to assess the impact of various fluid strategies on AKI, where this outcome is defined as ”modified KDIGO-3”.1_{The researchers seem to have chosen this classification of AKI as it is described as a}

serious adverse outcome, indirectly acknowledging that not all types of AKI are equally severe. It is likely that urine output data, needed for AKI classification, is collected in different ways in such multicentre trials, which may also translate to different AKI incidences and thus different associations with outcomes. In various studies therefore the urine output criteria for AKI have been set to an average 12-hourly threshold rather than actual hourly assessments.2_{For trials}

investigating both AKI and venous congestion, it is important to appreciate AKI heterogeneity while also standardizing methods for AKI diagnosis throughout different study sites to improve the generalizability of their results.

(7)

The complexity of venous congestion

The concept of venous congestion is not new. In patients with heart failure, venous congestion has been shown to be associated with unfavourable outcomes.3_{In critically ill patients, venous}

congestion has only been investigated by single surrogate measures, such as either fluid overload or peripheral edema.4,5_{In these studies however, various definitions for venous congestion and}

various definitions for AKI have been used, which complicate comparison of data. The concept of venous congestion is not defined at all yet, but several surrogate measures can be used to assess its presence. Some surrogate measures may be more general and systemic, like the presence of pulmonary edema or fluid balance, while other surrogates may be more specific, such as renal ultrasound measures of arterial and venous flow. Based on previous literature we evaluated multiple surrogate measures to assess venous congestion in critically ill patients, even though all these surrogate measures have their limitations.

For example, the fluid balance reflects the total amount of fluids which has been given to a patient minus the total amount of fluids that the patient has lost. A very high positive cumulative fluid balance could potentially be associated with venous congestion and it has previously been suggested to be associated with increased incidences of AKI.4,6,7_{There are, however, several issues}

which complicate accurate registration and interpretation of fluid balance. First, at what point in time, should the data collection start? Most registration systems start data registration upon ICU admission, while illness frequently begins long before ICU admission and so the patient has already been given many fluids or has already lost many fluids before ICU admission or even hospital admission. Second, it is a challenging task to exactly track all fluid intake and output, as registration and estimation of all details may be too time-consuming for health care personnel, especially in regular wards or simply impossible (like insensible losses). Last, if these issues were to be taken for granted, the situation in the ICU is both highly complex and highly dynamic so that the large amount of registrations make the fluid balance prone for measurement error, and therefore, should be interpreted with caution.

Surrogate measurements obtained by ultrasound may be easy to measure but clinical applicability can be limited, and in this thesis, only few appeared to be associated with AKI. Even though ultrasound is an up and coming technique, which is patient-friendly and accessible for physicians, some pitfalls must be acknowledged. Most importantly, ultrasound is operator dependent which bears some consequences; first, adequate training for appropriate techniques and measurements is required, second, experts should independently validate images and measures, and third; as interrater variability by definition exists, only crude measures can reliably be used, reassessed and change evaluated over time. The latter, however, also poses a beneficial characteristic: as measures are best-used crude, they can be obtained rather quickly and with limited training.

Acute Kidney Injury and Venous congestion

Early 2019, Mullens et al. published a position statement in the European Journal of Heart Failure considering the use of diuretics in patients with heart failure and congestion.8_{Although this was}

a position statement specifically focused on patients with heart failure, it is mentioned that no

(8)

study has prospectively evaluated multiple surrogate measures including clinical, ultrasound and biomarker variables. Two examples that have integrated surrogate (ultrasound) measures to describe venous congestion, have done so in relatively small patient samples and predominantly during the perioperative period.9,10_{Recently, a book was published in which a chapter describes}

the assessment of venous congestion using ultrasound with a proposed VExUS score.11_The

VExUS score stratifies the severity of venous congestion into several grades based on multi organ ultrasound assessment.11_{In a prospective study including 145 patients after cardiac surgery they}

found that severe congestion was associated with AKI.12_{Even though such a score might seem}

promising and useful for clinical practice, it still necessitates a solid evidence base by validation in externally replicated cohort studies.

This thesis showed that signs of venous congestion may increase the risk of AKI. Liberal fluid therapy, which may result in venous congestion, has been associated with worse outcomes compared to conservative fluid therapy.13_{One of the biggest research questions in current}

critical care practice is to outline which patients benefit from timely administered fluid therapy and which patient suffer harm from fluid. The findings in this thesis suggest that fluid therapy might have adverse effects in patients at risk of AKI if signs of venous congestion are present. The observational nature of the data presented in this thesis provides hypotheses on venous congestion, but randomised trials need to provide causal evidence for recommendations and guidelines for use of fluid therapy, or alternatively fluid removal, in clinical practice.

(9)

(10)

Future perspectives

Substantial heterogeneity remains among critically ill patients and (the pathophysiology of ) their AKI episodes. Before new studies are designed to investigate AKI or various fluid strategies, it is important to reach a consensus on the method of definition and stratification of AKI.14

For venous congestion a definition is needed as well. This definition should include crude estimates of multiple surrogate measures or perhaps normal values for a given population may provide some sort of reference values for the degree of venous congestion. Future studies could further delineate which surrogate measures are most useful to consider when the diagnosis of venous congestion is at question and could equally serve as triggers or targets for interventions. Two patients suffering from a similar clinical syndrome can be immensely different and may have very divergent outcomes. Subphenotypes of complex clinical syndromes exist, and likely these require different treatment. Investigating subphenotypes of clinical syndromes may decrease heterogeneity in research cohorts and could stimulate future research towards precision medicine and personalized care for the critically ill patient. For example, a better outcome may be achieved if triggers of venous congestion were to be combined with baseline, clinical, and biomarker data for a more detailed patient characteristic for targeting interventions-tailored estimation of treatment requirements. However, the question whether these associations are based on correlation or causality remains.15_{Centres could collaborate in this matter and}

collect a sufficient amount of data following a standardized protocol structurally. A resulting multinational cohort could facilitate the assessment of subphenotypes for multiple complex and heterogeneous syndromes. Moreover, this provides the opportunity to create individual, preferentially multivariable, intervention triggers, adapted to the individual clinical situation, instead of treating patients based on population-based normal values.

The overarching goal is to develop easily implementable and interactive prognostic models for syndromes such as AKI upon ICU admission, to be able to start personalized prevention and treatment of underlying illness. Various recent initiatives are working towards such solutions, such as a nomogram to estimate the risk of septic AKI risk, or a tool to predict Major Adverse Kidney Events (MAKE).16,17

Future trials could assess whether use of vasopressors rather than fluid therapy decreases AKI incidence, or if certain interventions such as active fluid removal could prevent further risk of adverse outcomes in patients with severe venous congestion and AKI.

Life after AKI and critical illness

Finally, while this thesis focused mainly on the interaction between venous congestion, AKI, and mortality, the most relevant question for patients concerns life after their ICU stay. Fortunately, many patients do survive ICU but we do not yet comprehend the impact of AKI on long term patient quality of life. Future studies investigating AKI and subsequent treatment should thus

(11)

also focus on how these affect the development of comorbidities and consequently, quality of life. The use of collaborative research infrastructures and the identification of subphenotypes of complex syndromes should be combined with a team that focuses on gathering outcome data that are most relevant to patients. After all, surviving ICU admission with high AKI burden and a long recovery is worth nothing if the joys of life can no longer be enjoyed.

(12)

References

Meyhoff, T. S. et al. Conservative vs liberal fluid therapy in septic shock (CLASSIC) trial— Protocol and statistical analysis plan. Acta Anaesthesiol. Scand. 63, 1262–1271 (2019).

REVERSE-AKI Randomized Controlled Pilot Trial - Full Text View - ClinicalTrials.gov. Available at: https:// clinicaltrials.gov/ct2/show/NCT03251131. (Accessed: 5th December 2019)

Rubio-Gracia, J. et al. Prevalence, predictors and clinical outcome of residual congestion in acute decompensated heart failure. Int. J. Cardiol. (2018). doi:10.1016/j.ijcard.2018.01.067.

Salahuddin, N. et al. Fluid overload is an independent risk factor for acute kidney injury in critically Ill patients: results of a cohort study. BMC Nephrol. 18, 45 (2017).

Chen, K. P. et al. Peripheral Edema, Central Venous Pressure, and Risk of AKI in Critical Illness. Clin. J. Am. Soc. Nephrol. 11, 602–8 (2016).

Wang, N., Jiang, L., Zhu, B., Wen, Y. & Xi, X.-M. Fluid balance and mortality in critically ill patients with acute kidney injury: a multicenter prospective epidemiological study. Crit. Care 19, 371 (2015).

Bagshaw, S. M., Brophy, P. D., Cruz, D. & Ronco, C. Fluid balance as a biomarker: impact of fluid overload on outcome in critically ill patients with acute kidney injury. Crit. Care 12, 169 (2008).

Mullens, W. et al. The use of diuretics in heart failure with congestion — a position statement from the Heart Failure Association of the European Society of Cardiology. Eur. J. Heart Fail. (2019). doi:10.1002/ ejhf.1369.

Beaubien-Souligny, W. et al. Alterations in portal vein flow and intrarenal venous flow are associated with acute kidney injury after cardiac surgery: A prospective observational cohort study. J. Am. Heart Assoc. 7, (2018).

Eljaiek, R. et al. High postoperative portal venous flow pulsatility indicates right ventricular dysfunction and predicts complications in cardiac surgery patients. Br.J.Anaesth.122, 206–214 (2019).

Rola, P., Haycock, K. & Spiegel, R. Special Skills: Venous Congestion. in Bedside Ultrasound: A Primer for Clinical Integration, 2nd edition (2019).

Beaubien-Souligny, W. et al. Quantifying systemic congestion with Point-Of-Care ultrasound: development of the venous excess ultrasound grading system. Ultrasound J 12, 16 (2020).

Silversides, J. A., Perner, A. & Malbrain, M. L. N. G. Liberal versus restrictive fluid therapy in critically ill patients. Intensive Care Medicine 45, 1440–1442 (2019).

Levey, A. S. et al. Nomenclature for kidney function and disease: report of a Kidney Disease: Improving Global Outcomes (KDIGO) Consensus Conference. Kidney Int. (2020) doi:10.1016/j.kint.2020.02.010. Girling, B. J., Channon, S. W., Haines, R. W. & Prowle, J. R. Acute kidney injury and adverse outcomes of critical illness: correlation or causation? doi:10.1093/ckj/sfz158.

Deng, F. et al. Nomogram to predict the risk of septic acute kidney injury in the first 24 h of admission: an analysis of intensive care unit data. Ren. Fail. 42, 428–436 (2020).

Sukmark, T. et al. SEA-MAKE score as a tool for predicting major adverse kidney events in critically ill patients with acute kidney injury: results from the SEA-AKI study. Ann. Intensive Care 10, (2020). 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Future perspectives

(13)