AUGMENTED RENAL CLEARANCE IN EAST COAST MALAYSIA INTENSIVE CARE SETTING:

(1)

AUGMENTED RENAL CLEARANCE IN EAST COAST MALAYSIA INTENSIVE CARE SETTING:

ASSESSMENT OF PREVALENCE, RISK FACTORS AND OUTCOMES

BY

SHAHIR ASRAF BIN ABDUL RAHIM

A dissertation submitted in fulfilment of the requirement for the degree of Master of Medicine (Anaesthesiology)

Kulliyyah of Medicine

International Islamic University Malaysia

NOVEMBER 2018

(2)

ii

ABSTRACT

Introduction: Augmented renal clearance (ARC) is a phenomenon where there is elevated renal clearance and defined by creatinine clearance > 130ml/min/1.73m². ARC results in changes of the pharmacokinetic and pharmacodynamic of antimicrobial therapy being administered, which may result in its subtherapeutic dose. We evaluated the prevalence, risk factors and outcome of ARC in critically ill patients in two Intensive Care Unit in Kuantan. Methods: This is a multicentre, prospective observational study of critically ill patients. Inclusion criteria were ICU patients older than 15 years old with plasma creatinine <130 µmol/l and arterial and urinary catheter inserted. 4 hours serum and urinary creatinine and flow rate were measured for a day and creatinine clearance (CrCl) calculated. ARC is defined as CrCl of more 130 ml/min/1.73. Patient were follow up until hospital discharge. Results: Among 102 patients recruited, of which 57 (55.9%) had ARC. Those with younger age (39.9±19 years old, p=0.013) and lower SOFA score (2.8±2.6, p=0.012) is at more risk to develop ARC. No significant difference in other risk factors such as male and trauma. There was no difference in the ICU and hospital mortality (p=0.652 and p=0.128). Surprisingly, duration of ICU admission amongst survivors was shorter in patients with ARC but statistically not significant (4 ± 6 vs 6 ± 7 days, p=0.271). Measured creatinine clearance moderately correlated with the estimated glomerular filtration rate using 4 different formulas (r=0.436-0.552, p<0.0001). Conclusions: ARC occurs in almost half of critically ill patients and more common in younger age and lower SOFA score. However, there was no difference in the outcome in this study. Estimated Glomerular Filtration Rate may be used as surrogate for measure creatinine clearance in detecting ARC.

(3)

iii

APPROVAL PAGE

I certify that I have supervised and read this study and that in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Master of Medicine (Anaesthesiology).

……….

Mohd Basri bin Mat Nor Supervisor

……….

Azrina binti Md Ralib Co-Supervisor

……….

Rozilah @ Abdul Hadi bin Mohamed Co-Supervisor

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Master of Medicine (Anaesthesiology).

……….

Examiner

This dissertation was submitted to the Department of Anaesthesiology and Intensive Care and is accepted as a fulfilment of the requirement for the degree of Master of Medicine (Anaesthesiology).

……….

Rozilah @ Abdul Hadi bin Mohamed Head, Department of Anaesthesiology and Intensive Care

This dissertation was submitted to the Kulliyyah of Medicine and is accepted as a fulfilment of the requirement for the degree of Master of Medicine (Anaesthesiology).

………...

Azmi bin Md Nor

Dean, Kulliyyah of Medicine

(4)

iv

DECLARATION

I hereby declare that this dissertation is the result of my own investigations, except where otherwise stated. I also declare that it has not been previously or concurrently submitted as a whole for any other degrees at IIUM or other institutions.

Shahir Asraf bin Abdul Rahim

Signature ... Date ...

(5)

v

INTERNATIONAL ISLAMIC UNIVERSITY MALAYSIA

DECLARATION OF COPYRIGHT AND AFFIRMATION OF FAIR USE OF UNPUBLISHED RESEARCH

AUGMENTED RENAL CLEARANCE IN EAST COAST MALAYSIA INTENSIVE CARE SETTING: ASSESSMENT OF

PREVALENCE, RISK FACTORS AND OUTCOMES

I declare that the copyright holders of this dissertation are jointly owned by the student and IIUM.

No part of this unpublished research may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without prior written permission of the copyright holder except as provided below

1. Any material contained in or derived from this unpublished research may be used by others in their writing with due acknowledgement.

2. IIUM or its library will have the right to make and transmit copies (print or electronic) for institutional and academic purposes.

3. The IIUM library will have the right to make, store in a retrieved system and supply copies of this unpublished research if requested by other universities and research libraries.

By signing this form, I acknowledged that I have read and understand the IIUM Intellectual Property Right and Commercialization policy.

Affirmed by Shahir Asraf bin Abdul Rahim

……..……….. ………..

Signature Date

(6)

vi

ACKNOWLEDGEMENTS

Firstly, it is my utmost pleasure to dedicate this work to my dear parents and my family, who granted me the gift of their unwavering belief in my ability to accomplish this goal:

thank you for your support and patience.

I wish to express my appreciation and thanks to those who had dedicated their time, effort and support for this project. To the dissertation committee members, thank you for being with me.

Finally, a special thanks to Associate Professor Dr Mohd Basri bin Mat Nor and Associate Professor Dr Azrina binti Md Ralib for their continuous support, encouragement and leadership. And for that, I will be forever grateful.

(7)

vii

LIST OF TABLES

Table 2.1 Summary of Studies On ARC Prevalence and Significant Risk Factors

15

Table 4.1 Table 4.2

Demographic and Clinical Characteristics Multivariable Logistic Regression for ARC

38 39

Table 4.3 Patients’ Outcome 40

Table 4.4 Mean Creatinine Clearance by Measured 4-Hour Creatinine Clearance and Estimated Creatinine Clearance using CG, MDRD, CKD-EPI Creatinine and CKD-EPI Cystatin C Formula

42

Table 4.5 Correlation between 4h-Creatinine Clearance and Estimated GFR from eGFR CG, eGFR MDRD, eGFR CKD-EPI and eGFR Cystatin C

42

Table 4.6 Prevalence of ARC using Estimated GFR Based on Plasma Creatinine CG, MDRD, CKD-EPI Equation and Plasma Cystatin C CKD-EPI Equation and Its Sensitivity and Specificity.

48

(11)

xi

LIST OF FIGURES

Figure 1.1 Theoretical Framework 7

Figure 2.1 Cockroft-Gault formula for Estimation of Glomerular Filtration Rate (eGFRCG)

26

Figure 2.2 Modification of Diet in Renal Disease Formula for Estimation of Glomerular Filtration Rate (eGFRMDRD)

27

Figure 2.3

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Chronic Kidney Disease – Epidemiological Collaboration Formulas for Estimation of Glomerular Filtration Rate (eGFRCKD-EPI)

Scatter Plot Correlation Between Measured Creatinine Clearance and Estimated Creatinine Clearance by Cockcroft-Gault equation

Scatter Plot Correlation Between Measured Creatinine Clearance and Estimated Creatinine Clearance by MDRD equation

Scatter Plot Correlation Between Measured Creatinine Clearance and Estimated Creatinine Clearance by CKD- EPI equation

Scatter Plot Correlation Between Measured Creatinine Clearance and Estimated Creatinine Clearance by CKD- EPI equation with Cystatin C

28

43

44

45

46

(12)

xii

LIST OF ABBREVIATIONS

APACHE II Acute Physiology And Chronic Health Evaluation II ARC

CKD

Augmented Renal Clearance Chronic Kidney Disease

CKD-EPI Chronic Kidney Disease Epidemiology Collaboration CNS

CrCl CysC

Central Nervous System Creatinine Clearance Plasma Cystatin C

Egfr Estimate of Glomerular Filtration Rate

ICU Intensive Care Unit

MDRD Modification of Diet in Renal Disease RRT

SAPS II

Renal Replacement Therapy

Simplified Acute Physiology Score II SOFA Sequential Organ Failure Assessment

(13)

1

CHAPTER ONE INTRODUCTION

1.1 BACKGROUND OF THE STUDY

Multiple factors are required to ensure success in treatment especially for critically ill patients. Correct, adequate and timely dose of therapeutic drugs are especially important, and these depends on factors affecting the absorption, distribution and excretion of the drugs. The main method of excretion of drugs from the body is via renal clearance. How kidney handle drugs are crucial in ensuring adequate dosing which are important to ensure successful therapy. This is important especially for hydrophilic drugs where its clearance is dependent on renal clearance. Renal clearance may increase or decrease in the critically ill which may result in under or overdosing. Dose adjustment in patients with reduced renal clearance due to renal impairment are commonly applied.

However, dose adjustments in patients with enhanced renal clearance are still not commonly applied due to the difficulty in ascertaining such condition.

Augmented renal clearance (ARC) is a phenomenon where there is elevated renal clearance and defined by measured creatinine clearance greater than 130ml/min/1.73m².This cut off point is used by most literature because a subtherapeutic concentration level of Vancomycin were seen when the measured creatinine clearance exceed 130ml/min/1.73m²(Udy, Roberts and Lipman, 2011). It has been described in multiple literatures regarding its common but variable prevalence (17.9% to 51.6%) in developed countries, depending on method of diagnosis and the case-mix studied.

Patients who are at risk of developing ARC are young, post-trauma and postoperative

(14)

2

patients and has low severity of illness scores such as low Sequential Organ Failure Assessment (SOFA) score.

ARC results in increase clearance of therapeutic drugs from the body. Several studies showed that ARC results in significant subtherapeutic dosing of critical drugs such as antibiotics (e.g. vancomycin and beta lactam antibiotics) and anti-epileptic (e.g.

levetiracetam) (Carlier et al., 2013; Udy et al., 2012; Cook, Arora, Davis, & Pittman, 2013). In treatment of bacterial infection, it is strongly associated with serum concentration of vancomycin below therapeutic range on the first three days of therapy.

This may expose patient to treatment failure if dose up-titration is not instituted (Carlier et al., 2013).

It appears that ARC commonly present in trauma patients and may persist throughout the first week of ICU stay (Udy et al., 2014). Combination of disease burden and suboptimal therapeutic intervention may promote prolonged period of subtherapeutic concentrations of antimicrobial agents leading to therapeutic failure and the development of resistance organism (Udy, Putt, Shanmugathasan, Roberts, &

Lipman, 2010). Therefore, there is a need to promptly identify such condition in ICU in order to individualize the drug dosing especially for high risk patients (Adnan et al., 2014)

Surprisingly, the link between ARC and clinical failure was not well established despite widespread association of ARC and subthreshold antimicrobial concentrations.

Therefore, further studies to determine whether standard dose regimes of antimicrobials in the presence of ARC have a negative impact on clinical outcome and emergence of antibiotic resistance organism are strongly suggested (Huttner et al., 2015).

The only study regarding ARC in Malaysia by Adnan et al. (2014) who studied ARC during the first 24 hours of ICU admission. They demonstrated a prevalence of

(15)

3

ARC of 39% with emergency admission as the only significant independent risk factor.

They also described the poor correlation of creatinine clearance by 24-hours urine collection with the estimated creatinine clearance by two commonly used formulas by Cockcroft-Gault (eGFR CG) and Modified Diet of Renal Disease (eGFR MDRD).

There was no significant outcome with ICU mortality, however the impact on antibiotic therapy was not monitored.

Measuring exact creatinine clearance is less practical in daily practice because there is not much information gained in knowing a small difference of creatinine clearance. It is also inconvenient and time consuming. Therefore, methods to estimate glomerular filtration rate based on serum creatinine are developed. Several creatinine- based equations have been developed to estimate the GFR in the chronic kidney disease population with stable creatinine level. These include Cockroft-Gault (CG), Modification of Diet in Renal Disease (MDRD) and CKD-EPI equation.

Other alternative biomarker to assess renal clearance is plasma Cystatin-C (Cys C). Cystatin C is produced by all nucleated cells in the body. It is completely filtered by the glomeruli and reabsorbed and catabolized in the proximal tubules (Charlton, Portilla,

& Okusa, 2014). Muscle mass, age less than 50 years old, sex, race and hydration status does not alter its production (Guillouet et al., 2011). Several studies showed that plasma Cystatin C may predict change of renal function earlier than serum creatinine (Bennett

& Devarajan, 2011; Volpon, Sugo, & Carlotti, 2015; Dai et al., 2015; Aydogdu et al., 2013). Equations on estimation of eGFR based on CysC has been developed, and were shown to be able to provide GFR estimates that are nearly as accurate as serum creatinine level adjusted for age, sex, and race, thus providing an alternative GFR estimate that is not affected by difference in muscle mass (Stevens et al., 2008). The correlation of eGFR CysC was slightly higher than the estimates by serum creatinine

(16)

4

equations such as CG and MDRD (Rule, Bergstralh, Slezak, Bergert, & Larson, 2006).

The CKD-EPI equations using serum cystatin C were also found to be useful for risk stratification and show better prognostic performance than creatinine-only based eGFR equations (Zamora et al., 2014).

A good correlation of estimated creatinine clearance with the measured creatinine clearance would greatly ease the identification of ARC among at risk patient.

However, these equations are frequently used in acute kidney injury (AKI) and chronic kidney disease (CKD) patient as it is derived from stable CKD population. Therefore, a lot of bias and imprecision when these equations are used in non-CKD patient (Cirillo, 2009). There are poor correlation with the measured creatinine clearance when using these equations in ARC patient (Grootaert, Willems, Debaveye, Meyfroidt, & Spriet, 2012; Adnan et al., 2014).

At the moment, to the best of our knowledge regarding ARC in our region, there is only one study regarding ARC by Adnan et al. (2014). It is a study with small sample size with lower ARC prevalence compared to other study. This study would provide more local data regarding ARC.

1.2 STATEMENT OF THE PROBLEM

Augmented Renal Clearance is common but is an underdiagnosed phenomenon among critically ill patient in the Intensive Care Unit with varying prevalence from various study.

1.3 PURPOSE OF THE STUDY

The purpose of this study is to determine the prevalence of ARC in Intensive Care Unit, assess the risk factors of developing ARC and evaluate the effect of ARC on patients’

(17)

5

outcome. In addition, this study also assessed the correlation of measured creatinine clearance and estimated creatinine clearance from four different formulas, whereby three are plasma creatinine-based formula and one is Cystatin C-based formula. The purpose of this assessment is to determine which formula has better correlation and hence can be used as a tool to detect ARC in daily clinical practice.

1.4 RESEARCH OBJECTIVES

The study aimed to achieve the following objectives:

i. To determine the prevalence of ARC in critically ill patients in our setting ii. To identify the risk factors for ARC in critically ill patients

iii. To evaluate the association between ARC and patient’s morbidity and mortality

iv. To investigate the correlation between creatinine clearance and estimated GFR using the eGFR CG, eGFR MDRD, eGFR CKD-EPI and eGFR Cystatin C equations.

v. To determine prevalence of ARC using eGFR CG, eGFR MDRD, eGFR CKD–EPI and eGFR Cystatin C equations and its sensitivity and specificity.

1.5 RESEARCH QUESTIONS

This study was conducted to search for answers of the following questions:

i. What is the prevalence of ARC in critically ill patient in our setting?

ii. What are the risk factors for developing ARC?

iii. What are the impacts of ARC to patient’s morbidity and mortality?

(18)

6

iv. Is there correlation between measured creatinine clearance and estimated glomerular filtration rate (GFR) to detect ARC?

v. Is estimated GFR using plasma creatinine CG, MDRD, CKD-EPI equation and plasma Cystatin C CKD-EPI equation are sensitive and specific to detect ARC?

(19)

7 1.6 THEORETICAL FRAMEWORK

Figure 1.1 Theoretical Framework

(20)

8 1.7 RESEARCH HYPOTHESES

It is hypothesized that Augmented Renal Clearance has high prevalence in Intensive Care Unit with identifiable risk factors. It also hypothesized that measured creatinine clearance based on four hours sampling time correlates well with estimated creatinine clearance and estimated creatinine clearance can be used as a surrogate to detect ARC.

1.8 SIGNIFICANCE OF THE STUDY

This study may provide more information regarding the prevalence of ARC in Malaysian Intensive Care Unit. This could emphasise the importance of considering the presence of ARC in critical drug dosing in the critically ill, especially when using antibiotics.

1.9 LIMITATIONS OF THE STUDY

There are few limitations to this study, which are:

i. Unequal sample size from two centres.

ii. Two centres within same locality but very different case-mix and case turn- over rate.

iii. Small sample size to determine correlation between measured creatinine clearance and estimated creatinine clearance.

iv. A single sampling done during initial day of ICU stay, which may detect ARC during the initial days only.

v. Measured creatinine clearance using 24 hours urine collection was not done to determine the accuracy of 4 hours creatinine clearance

(21)

9 1.10 DEFINITIONS OF TERMS

Augmented Renal Clearance

A phenomenon where there is elevated renal clearance and defined by measured creatinine clearance >130ml/min/1.73m²(Udy et al., 2011).

Measured Creatinine Clearance

Method of assessment Glomerular Filtration Rate by urinary creatinine clearance which computed from a timed averaged urine collection and blood sampling for serum (Stevens, Coresh, Greene, & Levey, 2006).

Estimated Creatinine Clearance

Method of assessment Glomerular Filtration Rate by Estimating equations include variables such as age, sex, race, and body size, in addition to serum creatinine and serum cystatin C, as surrogates for muscle mass (Stevens et al., 2006).

1.11 CHAPTER SUMMARY

This chapter introduced to the reader regarding problems that this research intended to investigate and solve. It also outlines the theoretical framework for better understanding of issues addressed by this study. Apart from that, it outlines study objectives, research questions and hypotheses.

(22)

10

CHAPTER TWO LITERATURE REVIEW

2.1 INTRODUCTION

This study is mainly to assess the prevalence of ARC, its risk factors and the effect on patients’ morbidity and mortality. Therefore, the literature review will cover the mechanism of ARC, its prevalence from other studies, method of diagnosing ARC, the risk factors to develop ARC, the impact of ARC to patients, what is known regarding ARC in Malaysia and the limitations of previous study on ARC. Other than that, it will also cover regarding estimates creatinine clearance using different equations as a surrogate to measured creatinine clearance.

2.2 MECHANISM OF AUGMENTED RENAL CLEARANCE

The definite mechanism of ARC is difficult to be determined but the development of a systemic inflammatory response syndrome (SIRS) appears closely associated with ARC (Shimamoto et al., 2013) (Figure 1). SIRS is common in the critically ill patients with severe infections, trauma, burns injury, pancreatitis, major surgery, ischemia and hematological malignancy. Inflammatory mediators released with SIRS can markedly increase cardiac output and decrease the systemic vascular resistance via activation of the sympathetic response. Therefore, increasing the renal blood flow and hence GFR in normal kidneys. This can be further aggravated by the use of vasoactive drugs and high volume fluid therapy, which are common in the critically ill. Severe inflammatory conditions in non-ICU settings may also predispose patients to ARC (Minkute et al., 2013). A recent study demonstrated an increase in GFR in a small cohort of SIRS

(23)

11

patients supporting this key mechanism (Udy et al., 2014). Interestingly, this study also demonstrated increased tubular secretion of anions, which could possibly contribute to increased clearance of anionic antibiotics like some beta-lactams (Udy et al., 2014).

2.3 PREVALENCE OF AUGMENTED RENAL CLEARANCE

The prevalence of ARC in the intensive care patients ranging from 30 to 65% despite normal serum creatinine observed in these patients (Hobbs et al. 2015). A prospective study on 89 critically ill patients in Spain showed that 17.9% of patients presented with glomerular hyperfiltration or ARC on admission, the prevalence increased up to 30%

during the first week of admission (Fuster-Lluch et al., 2008). Another multicentre observational study of four, tertiary-level, university-affiliated, ICUs in Australia, Singapore, Hong Kong, and Portugal where the cohort manifested ARC prevalence was as high as 65.1% (183 out of 281 patients) in on at least one occasion during the first seven study days (Udy et al., 2014). A small size Malaysian data from 49 patients in a single centre prospective observational study in one of the main trauma centre in Malaysia showed ARC prevalence of 39% in the initial 24 hours of ICU stay.

The incidence is up to 30-85% in certain ICU population with sepsis (Claus, Hoste, Colpaert, Robays, Decruyenaere, & De Waele, 2013), trauma (Udy, Roberts, Boots, Paterson, & Lipmanl, 2010) ,traumatic brain injury (TBI) (Udy et al., 2013), subarachnoid haemorrhage (SAH) (Fuster-Lluch et al., 2008) and central nervous system (CNS) infection (Lautrette et al., 2012) (Hobbs et al., 2015). This may be explained by differences in case mix, patient characteristics and also by the cut off value used to diagnose ARC. Fuster- Lluch et al. (2008) defined ARC measured creatinine clearance more than 120ml/min/1.73m²and found that the prevalence was 30% in severe sepsis or septic shock cohort. Baptista, Sousa, Martin & Pimentel (2012) who

(24)

12

conducted the study in a single centre adult ICU in Portugal who were ventilated for severe sepsis or septic shock and received Vancomycin defined ARC as measured creatinine clearance more than 130ml/min/1.73m² found that 40% of study population has ARC. Udy et al. (2010) who studied ARC in small cohort of 20 traumatic brain injury patients who underwent neuroprotection measures using cut-off value greater than 160ml/min/1.73m² for men and greater than 150ml/min/1.73m²for women found that 85% of the patient with ARC. In another study by Conil et al. (2007) found that 54% of patients with creatinine clearance more than 120ml/min/1.73m² in a burn unit.

This shows that ARC may be a common physiologic changes seen in critically ill patient but there is still lack of reporting and remained unappreciated (Udy, T. Putt, J. Boots, & Lipman, 2011). For this study a cutoff value of 130ml/min/1.73m² was used to define ARC based on the evidence of increase vancomycin clearance for those with ARC as defined by 130ml/min/1.73m² threshold (Baptista et al. 2012).

2.4 METHOD OF DIAGNOSIS FOR AUGMENTED RENAL CLEARANCE Clinician routinely monitor the renal function to guide and optimize drug therapy and as means to recognize deterioration of renal function. Regular monitoring urine output to detect oliguria (urine output < 0.5ml/kg/hour) and biochemical parameter such as plasma creatinine level are commonly used. However, this is best to detect deteriorating renal function compared to phenomenon such as Augmented Renal Clearance (ARC).

Common method of assessment such as using serum creatinine level alone or estimated glomerular filtration rate (eGFR) formulas i.e. Cockcroft-Gault (CG), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), Modification of Diet in Renal Disease (MDRD) may not accurately identify patients with ARC. Serum creatinine level may be affected by age, gender, sex, race, metabolic state, diet and