ROLE OF SERUM CYSTATIN-C IN ACUTE KIDNEY INJURY AMONG CHILDREN IN
PAEDIATRIC INTENSIVE CARE UNIT.
Dissertation submitted to
THE TAMILNADU DR.M.G.R.MEDICAL UNIVERSITY, CHENNAI-600 032 In partial fulfillment of the requirements for the degree of
M.D DEGREE (PEDIATRICS) BRANCH VII INSTITUTE OF SOCIAL PEDIATRICS GOVERNMENT STANLEY MEDICAL COLLEGE
CHENNAI – 600 001
APRIL 2020
CERTIFICATE BY THE GUIDE
This is to certify that the dissertation entitled “Role of Serum Cystatin-C in acute kidney injury among children in Paediatric intensive care unit” is a bonafide record of work carried out by DR.RINI SUSAN.G, in the Department of Paediatrics, Government Stanley Medical College, under my guidance and supervision during the period of her post graduate study for M.D. Paediatrics from May 2017 to March 2020.
Dr.T.RAVICHANDRAN M.D., DCH.
Place : Director
Institute of Social Pediatrics
Date : Stanley Medical College,
Chennai - 600001
i
CERTIFICATE BY THE INSTITUTION
This to certify that the dissertation titled “Role of Serum Cystatin- C in acute kidney injury among children in Paediatric intensive care unit” is a bonafide record of work carried out by Dr. RINI SUSAN.G, in the Department of Paediatrics under our direct supervision and guidance, during the academic year 2017 -2020 submitted to The Tamilnadu Dr.M.G.R Medical University, Chennai in partial fulfillment of the requirement of the award for the degree of M.D BRANCH VII (PAEDIATRICS).
Dr.T.RAVICHANDRAN M.D., DCH. Dr.R.SHANTHIMALAR M.D., DA.
Director Dean
Institute of Social Paediatrics Stanley Medical College
Stanley Medical college
i
ACKNOWLEDGEMENT
It is with immense pleasure and gratitude that I thank
Prof. Dr. R.SHANTHIMALAR, M.D., DA. THE DEAN, STANLEY MEDICAL COLLEGE for bestowing me the permission and privilege of presenting this study and for enabling me to avail the institutional facilities.
It is with great pleasure that I express a deep sense of gratitude to my teacher and Guide, Prof. Dr.T.Ravichandran. M.D., DCH, Director, Department of Paediatrics, for his valuable guidance and support during the preparation of this dissertation and also inspiring me at every step of this study.I express my gratitude to my Co- guide, Dr.Ekambaranath M.D, Assistant Professor of PICU for his valuable help and guidance throughout this study. I am very grateful to all my chiefs, Prof.Dr.J.Ganesh M.D, Prof.Dr.M.A.Aravind M.D. and Prof.Dr.P.Senthil Kumar, M.D. for their valuable guidance and motivation.
ii
I sincerely thank my Assistant Professors Dr.Sankara Narayanan M.D , Dr.P.Venkatesh M.D, Dr.Vinoth M.D, Dr.Parveen kumar M.D,
Dr.Senthil kumar M.D., Dr.Rajesh kumar M.D., Dr.Selvi M.D and Dr. Anandhi M.D.for their valuable support throughout the course of this study.
I thank all the post graduates in the Department of Paediatrics in Stanley Medical College who have helped me.
Finally I wish to express my whole-hearted thanks to the Heads of the Institution of the schools and the students and their parents for their cooperation and support.
Dr.RINI SUSAN.G
TABLE OF CONTENTS
CERTIFICATE i
ACKNOWLEDGEMENT ii
TABLE OF CONTENTS iii
LIST OF TABLES iv
LIST OF FIGURES v
ABBREVIATIONS AND ACRONYMS vi
1. INTRODUCTION 1
2. REVIEW OF LITERATURE 8
3. SUBJECTS AND METHODOLOGY 30
4. RESULTS 35
5. DISCUSSION 76
6. SUMMARY AND CONCLUSION 84
7. BIBLIOGRAPHY 93
8. ANNEXURE 114
iii
LIST OF TABLES
TABLE 1: DISTRIBUTION OF CASES AMONG THE TWO GROUPS
TABLE 2:AGE DISTRIBUTION
TABLE 3:GENDER DISTRIBUTION
TABLE 4:AGE ANDSEX DISTRIBUTION AMONG AKI GROUP
TABLE 5: AGE AND SEX DISTRIBUTION AMONG NON-AKI GRO UP
TABLE 6:TIME OF PRESENTATION WITH AKI AMONG THE CHILDREN
TABLE 7:HEIGHT OF THE CHILDREN IN THE TWO GROUPS
TABLE 8: WEIGHT OF THE CHILDREN IN THE TWO GROUPS
TABLE 9:BMI OF THE CHILDREN IN THE TWO GRO UPS
TABLE 10:BSA OF THE CHILDREN IN THE TWO GROUPS
iv
TABLE 11:IN-PATIENT CHARACTERISTICS OF THE TWO GROUPS
TABLE 12: DIAGNOSIS OF THE TWO GROUPS
TABLE 13:PRIFLE CRITERIA AND CATEGORIZATION OF THE PATIENTS
TABLE 14: SERUM CYSTATIN-C
TABLE 15:CYSTATIN VALUES IN AKI GROUP
TABLE 16:OUTCOME OF THE ILLNESS
TABLE 17:CORRELATION BETWEEN SERUM CYSTATIN-C AND PRIFLE CRITERIA
LIST OF FIGURES
FIGURE 1: ETIOLOGY OF ACUTE KIDNEY INJURY IN CHILDREN
FIGURE 2: SITE OF AKI
FIGURE 3:PRERENAL AKI
FIGURE 4: INTRINSIC AKI
FIGURE 5: OBSTRUCTIVE UROPATHY
FIGURE 6:CYSTATIN-C
FIGURE 7:SELECTION AND DISTRIBUTION OF CASES
FIGURE 8:DISTRIBUTION OF CASES AMONG THE TWO GROUPS
FIGURE 9:AGE DISTRIBUTION
FIGURE 10:GENDER DISTRIBUTION
FIGURE 11:TIME OF PRESENTATION WITH AKI AMONG THE CHILDREN
FIGURE 12:HEIGHT OF THE CHILDREN IN THE TWO GROUPS
FIGURE 13:WEIGHT OF THE CHILDREN IN THE TWO GROUPS
FIGURE 14:BMI OF THE CHILDREN IN THE TWO GROUPS
v
FIGURE 15:BSA OF THE CHILDREN IN THE TWO GROUPS
FIGURE 16:SERUM CREATININE IN THE AKI GROUP
FIGURE 17:SERUM CREATININE IN THE NON AKI GRO UP
FIGURE 18:GFR IN THE AKI GROUP
FIGURE 19:GFR IN THE NON AKI GROUP
FIGURE 20:IN PATIENT CHARACTERISTICS OF THE TWO GROUPS
FIGURE 21:PRIFLE CRITERIA AND CATEGORIZATION OF THE PATIENTS
FIGURE 22:SERUM CYSTATIN- C
FIGURE 23:CYSTATIN VALUES IN AKI GROUP
FIGURE 23b: CYSTATIN VALUES IN AKI GROUP
FIGURE 24:O UTCOME OF THE ILLNESS
FIGURE 25: SENSITIVITY AND SPECIFICITY ANALYSIS OF AKI AND SERUM CYSTATIN -C
FIGURE 26:SENSITIVITY AND SPECIFICITY ANALYSIS OF DEATH AND SERUM CYSTATIN -C
ABBREVIATIONS AKI- Acute kidney injury
AIN- Acute interstitial nephritis
BMI –Body mass index
BSA- Body surface area
GFR- Glomerular filtration rate
HUS- Hemolytic uremic syndrome
pRIFLE- paediatric Risk, Injury, Failure, Loss of kidney function, End stage renal disease
RPGN- Rapidly progressive glomerulonephritis
vi
1
INTRODUCTION
2 Introduction
Acute Kidney Injury-A perspective
Formerly known as acute renal failure or acute renal insufficiency, acute kidney injury is a malady characterized by the presence of sudden progressive loss of renal function impacting the fluid and electrolyte homeostasis. The acute kidney injury syndrome is undergird by decreased urine output or by the accretion of urea and creatinine (end products of nitrogen metabolism) or sometimes by both.
The underlying cause for the acute kidney injury may be multifactorial and this is just a clinical presentation of an existing concealed problem. The most common presentation is seen in critically ill patients and in hospitalized cases, where the real cause is extrarenal, though the exact cause is debatable. This makes the management challenging. The real cause cannot be addressed and the treatment is mainly supportive. Renal biomarkers can be of help in making an early diagnosis. Renal replacement therapy is initiated in cases of severe injury to kidney presenting with complications like uremia and toxemia or severe biochemical and volume related changes. The prognosis is not good even when there are no premorbid conditions as the chances of progressing to chronic kidney disease is very high though in some cases, dialysis independence is seen1.
The incidence of AKI in children admitted to pediatric tertiary care center is 2-3% while the overall incidence is around 8% in children admitted in neonatal intensive care unit. For classification of patients in AKI, RIFLE criteria were used which was based on;
1) Risk 2) Injury
3 3) Failure
4) Loss
5) End-Stage Renal Disease
The criteria was modified for the pediatric age group and used as the pRIFLE criteria (which is discussed later). RIFLE focusses on the glomerular filtration rate, AKI network follow a categorization based on the levels of serum creatinine.
As evident from literature, the causes of Acute Kidney Injury are many and may manifest differently. For instance, RPGN or rapidly progressive glomerulonephritis that presents as AKI progresses swiftly to chronic kidney disease. On the other hand, few conditions like Henoch–
Schonleinpurpura, hemolytic–uremic syndrome (HUS) and obstructive uropathy show characteristics of acute kidney disease and present with improvement only to be further deteriorated into chronic kidney disease in the subsequent years.
When the acute kidney injury is due to ischemic-hypoxic insults, hemolytic–uremic syndrome, etc typically present with an urine output of <500ml in 24 hours, a condition known as oliguria or anuria. But certain conditions present with normal output like aminoglycoside nephrotoxicity, acute interstitial nephritis and contrast nephropathy. This is important in terms of the morbidity and mortality rates where patients with normal or near normal urine output present with better prognosis 13-17.
4
Acute kidney injury in Children and the role of biomarkers
Acute kidney injury is a commonly encountered problem in a paediatric intensive care unit which is often reversible. The recent reports have shown an increasing trend in the incidence of acute kidney injury. The etiology landscape too has shifted with more emphasise on multifactorial causes than the primary renal cause. The genetic predisposition has also been explored that states that some children may have hereditary inclination towards the disease. The acute renal injury is broadly classified as pre-renal failure, intrinsic renal disease and obstructive uropathies.
With no studies establishing the etiology of acute kidney injuries specifically, it is known that the incidence of AKI among hospitalized children is increasing18-22. The most commonly encountered scenarios of acute kidney injury in children is during stem cell transplantation and after a cardiac surgery. It is difficult to establish the causes in such cases when multiple factors dictate the initiation, development, progress and prognosis. The significant contributors to the development of the disease is nephrotoxic insults or ischemic/hypoxic injury. The real issue with AKI is there are no well defined studies for finding out the incidence of AKI in pediatric patients. This argues the case for relying on the incidence studies of adults. A large study showed that among a million population of adults, 209 people will suffer from the disease with 21% of them presenting with pre-renal causes and around 45% presenting with acute tubular necrosis23.
When the etiology is pre-renal, the kidney is structurally normal and when the condition is corrected and the perfusion is adequate, the function returns to normal. On the other hand, in acute tubular necrosis, the kidney is structurally damaged and reperfusion is not enough for the kidney to return to normal function.
5
In the last decade, lot of studies have been done to understand the role of hypoxic-ischemic injury in acute kidney injury with more focus on the cellular, molecular and biochemical events.
The probable causes of the injury can be arrived at through elaborate history taking, physical examination and using laboratory investigations and radiological evaluations. Experimental interventions using the renal-dose dopamine and diuretics have shown to be ineffective and the prognosis depends on the etiology of the injury and not the presentation. An initial insult predisposes the child for subsequent chronic disease or repeated similar injuries in the future.
One of the reasons for the diagnostic challenges is the use of laboratory parameters like serum creatinine which has a low sensitivity. In order to design a better strategy for the management of acute kidney injury, research is essential to understand the pathophysiology, role of biomarkers in AKI and an improved classification of the disease.
One of the epidemiological challenges in studying AKI is the definition. There is no harmonization between the definitions of the adult and children AKI with multiple perspectives used to describe the same phenomenon. Serum creatinine is the most commonly used criteria for the diagnosis of AKI. This is plagued with insensitivity and the problem of delayed measure.
Need for study
The present study is to determine the prognostic ability of serum Cystatin C over serum creatinine among children with acute kidney injury in paediatric intensive care unit. The term AKI is used for defining an abrupt and potentially reversible form of structural or functional renal impairment. It seeks to include patients with early renal damage in an attempt to intervene early2. International society of nephrology 0 by 25 initiative estimates that out of 1.7 million
6
death per year caused by AKI globally, around 1.4 million occur in low and middle income Countries3. Hence there is a need to understand the incidence and determinants of AKI in our ICU thereby aiming to prevent AKI early.
There is also a need to evaluate a novel biomarker that can be easily done and will not only diagnose AKI but will also prognosticate AKI at once. Most of the studies on paediatric AKI is from western literature. Determinants of AKI and availability of laboratory tests for diagnosing and prognosingAKI in our part of country varies widely. The incidence of AKI in India according to studies done after 2010 varies widely between 14% and 42.9%.Using modified pRIFLE criteria is tedious work in an intensive care unit, as it encorporates changes in serum creatinine,fall in GFR, and changes in urine output4. Moreover, to calculate the fall in GFR one should know the baseline serum creatinine which is often unknown in most patients5. Additionally urine output guidelines are erroneous in detection of AKI in non-oliguric renal failure which is more common in children6,7.
Cystatin-c is a cysteine protease inhibitor synthesized by all nucleated cells at a constant rate8. It is freely filtered by glomerulus and reabsorbed completely by the proximal convoluted tubules and not secreted. Thus it is a promising tool for diagnosis of AKI. Age, gender, muscle mass has negligible effect on serum cystatin C unlike creatinine levels making it a potential substrate to be explored as a diagnostic marker for renal dysfunction.Here, in this study it is going to be explored as a prognostic marker, whether a single value of serum cystatin can predict the outcome of AKI in a child admitted in PICU.Studies showed that serum cystatin –c is a sensitive but not a specific marker.
A Study by Herget Rosenthal et al showed that cystatin c may detect AKI one or two days earlier than creatinine9. One more study by Volpon et al showed serum cystatin-c was increasing higher
7
according to AKI severity (Failure>injury>risk)10. They also showed that Serum cystatin-c levels of >0.70mg/L were associated with longer length of PICU stay and prolonged duration of mechanical ventilation. They showed that cystatin –C is an early and accurate marker for AKI and is associated with adverse clinical outcomes in a heterogenous population of critically ill children . While Annick et al showed that serum cystatin -C was poor biomarker for AKI.
Another study by BabakNakhjavan et al showed that cystatin-C has an acceptable prognostic value and diagnostic value for prediction of AKI in children11. Since the results are contradictory this study aims to find out the correlation between cystatin-C levels and pRIFLE stages whether increasing accordingly with AKI severity and also whether a single value of cystatin-c can predict patients outcome in terms of duration of hospital and PICU stay ,duration of mechanical ventilation, and survival12.
8
REVIEW OF
LITERATURE
9 Review of literature
An overview of acute kidney injury in children
The acute kidney injury syndrome is undergird by decreased urine output or by the accretion of urea and creatinine ( end products of nitrogen metabolism) or sometimes by both. The underlying cause for the acute kidney injury may be multifactorial and this is just a clinical presentation of an existing concealed problem. The incidence of AKI in children admitted to pediatric tertiary care center is 2-3% while the overall incidence is around 8% in children admitted in neonatal intensive care unit. For classification of patients in AKI, RIFLE criteria were used which was based on;
1)Risk 2)Injury 3)Failure 4)Loss
5)End-Stage Renal Disease
The criteria was modified for the pediatric age group and used as the pRIFLE criteria (which is discussed later). RIFLE focusses on the glomerular filtration rate, AKI network follow a categorization based on the levels of serum creatinine.
Acute Kidney Injury in children may be caused by many reasons that are listed below that are broadly classified as Pre-renal causes, intrinsic renal disease and obstructive neuropathy.
10
11 Figure 1: Etiology of Acute Kidney Injury in Children
As evident from literature, the causes of Acute Kidney Injury are many and may manifest differently. For instance, RPGN or rapidly progressive glomerulonephritis that presents as AKI progresses swiftly to chronic kidney disease. On the other hand, few conditions like Henoch–
Schönleinpurpura, hemolytic–uremic syndrome (HUS) and obstructive uropathy show characteristics of acute kidney disease and present with improvement only to be further deteriorated into chronic kidney disease in the subsequent years.
When the acute kidney injury is due to ischemic-hypoxic insults, hemolytic–uremic syndrome, etc typically present with an urine output of <500ml in 24 hours, a condition known as oliguria or anuria. But certain conditions present with normal output like aminoglycoside nephrotoxicity, acute interstitial nephritis and contrast nephropathy. This is important in terms of the morbidity and mortality rates where patients with normal or near normal urine output present with better prognosis 13-17. In this section, we have explored the following;
1) Epidemiology of Acute Kidney Injury 2) Causes of AKI
3) Pathogenesis and pathophysiology of AKI 4) Management
5) Potential future therapies
Epidemiology of acute kidney injury in children
With no studies establishing the etiology of acute kidney Injuries specifically, it is known that the incidence of AKI among hospitalized children is increasing18-22. The most commonly encountered scenarios of acute kidney injury in children is during stem cell transplantation and
12
after a cardiac surgery. It is difficult to establish the causes in such cases when multiple factors dictate the initiation, development, progress and prognosis. The significant contributors to the development of the disease is nephrotoxic insults or ischemic/hypoxic injury. The real issue with AKI is there are no well defined studies for finding out the incidence of AKI in pediatric patients. This argues the case for relying on the incidence studies of adults. A large study showed that among a million population of adults, 209 people will suffer from the disease with 21% of them presenting with pre-renal causes and around 45% presenting with acute tubular necrosis23.
When the etiology is pre-renal, the kidney is structurally normal and when the condition is corrected and the perfusion is adequate, the function returns to normal. On the other hand, in acute tubular necrosis, the kidney is structurally damaged and reperfusion is not enough for the kidney to return to normal function.
A study in a tertiary care center among the pediatric population, the overall incidence was 0.8 for 100,000 children. Of these children, 227 of them received dialysis during the 8-year study period24. Another study among neonates showed the incidence between 8% and 24%. The incidence was positively correlated with asphyxia27,28. Other commonly associated causes of AKI were; birth weight <1500g, patent ductusarteriosus, NSAIDs and antibiotics in mothers and a low Apgar score29. In preterm infants, a low Apgar score along with intake of NSAIDs by mother is shown to decrease the renal function30. In a developing country, the incidence of AKI in newborn is 3.9 for every 1000 live births compared to 34.5 for every 1000 new born admitted to NICU31.
Studies show that multiple factors play a role in AKI including genetic and environmental factors. Certain gene polymorphisms are seen in AKI especially of the ACE (angiotensin-
13
converting enzyme) gene that leads to changes in the renin-angiotensin system32. Other studies to find out the relationship between AKI and polymorphisms in TNF-alpha, IL-1b, IL-6 and IL-10 have been done in new borns33. Though the results have been inconclusive, yet the presence of these polymorphisms may lead to higher inflammatory response and greater degree of AKI in infection of the newborn. This finding is important in the light of developing future therapies for the management of AKI in newborn as newer drugs may be designed to interrupt this inflammatory process. In few other studies, Patent ductusarteriosus was associated with ACE I/D allele genotypes or angiotensin I receptor gene that might indirectly predispose to AKI in children34. The heat shock protein 72 (1267) GG genetic variation was common in AKI among low birth weight babies35. This leads to the conclusion that some neonates are have higher predisposition to AKI36. Understanding the genetics behind the development of AKI is essential for planning future therapies.
Diagnosis and Etiology of Acute Kidney Injury
Some of the biomarkers that are being investigated are37;
Plasma NGAL- plasma neutrophil gelatinase-associated lipocalin (NGAL)
Serum Cystatin C
Urinary NGAL
IL-18 (Interleukin 18)
KIM-1 ( Kidney Injury Molecule-1)
14
The development of sensitive and specific biomarkers are essential to detect AKI early and plan interventions. A new classification system called the RIFLE criteria classifies patients based on the following;
R=Risk related to renal dysfunction I=Injury to kidney
F=Failure of renal function
L=Lose of renal function E=End stage renal disease
This is mainly used for adults and has been modified for children as pRIFLE38,39. The role of pRIFLE is also prognostic and is being used by the AKIN- Acute Kidney Injury Network comprising of critical care physicians, pediatric nephrologists, adult nephrologists and other societies that are involved in research on AKI. This AKIN is devoted to ensuring progress in diagnosing and treating AKI through collaborative contributions from international, intersociety and interdisciplinary groups40.
As discussed earlier, AKI is classified as pre-renal, intrinsic renal and obstructive uropathies.
The intrinsic renal failure includes vascular insults as well. Age predisposition among different etiologies are also observed like;
Neonates: Cortical Necrosis and Renal Vein Thrombosis
15 Young Children: HUS
Older children and adolescents: RPGN
One of the important features noted in neonates with AKI is the prenatal exposure to the drugs in mother that can affect nephrogenesis like NSAIDs, Angiotensin Receptor Blockers, ACE inhibitors, etc41-44. The causes of AKI is thus challenging to establish. A thorough history along with detailed physical examination should be wisely used along with laboratory tests and radiographic studies for arriving at a tentative diagnosis. Most of the time, children with AKI in ICU present with multiple etiologic factors.
16 Figure 2: Site of AKI
The pre-renal injury to the kidney occurs when the blood flow to the renals are compromised or reduced due to reduction in the volume of blood or because of intravascular volume contraction.
This state is reversible as the kidneys are structurally normal and when the prerenal cause is corrected, the kidney reverts back to normal. This requires restoring the blood volume and a stable hemodynamic conditions. When this condition is prolonged, the injury leads to hypoxic-
17
ischemic injury and acute tubular necrosis and the cause becomes intrinsic. This evolution from pre renal to intrinsic renal injury is a gradual event with a number of mechanisms failing to maintain the optimal blood flow.
Figure 3: Prerenal AKI
This happens in a number of ways. The compromise on renal perfusion leads to the relaxation of afferent arterioles to reduce the resistance to renal blood flow. This state of hypoperfusion releases vasodilatory chemicals namely vasodilatory prostaglandins and prostacyclin. When NSAIDs are used, this leads to the inhibition of renal perfusion and precipitate acute kidney injury45. On the other hand, when the renal perfusion is low, like in the case of stenosis of the
18
renal artery, the intraglomerular filtration is mediated by angiotensin II to elevate the efferent arteriolar resistance. This explains why administration of ACE II inhibitors will lead to acute kidney injury46,47. Therefore, the medications and therapies in such cases should be carefully designed to avoid AKI.
Dehydration secondary to gastrointestinal losses, renal and adrenal diseases causing salt-wasting, diabetes insipidus, burns, third space volume loss, etc lead to true volume contraction and thereby effects pre-renal injury. On the other hand, decreased effective blood volume is seen in cardiac tamponade, congestive cardiac failure and hepatorenal syndrome where blood volume is normal but the renal perfusion is decreased.
The true challenge lies in identifying the cause and the level of injury. There are several urinary parameters that are non-specific like urine sodium concentration, urine osmolality, renal failure index and fractional excretion of sodium that are utilized to differentiate between hypoxic/ischemic injury and pre-renal injury. The synonym for hypoxic/ischemic injury is vasomotor nephropathy or acute tubular necrosis. The initial state is vascular constrictions that is subsequently followed by tubular injury. This is the main differentiating feature where the acute tubular necrosis has injury to the intrinsic structure of the kidneys that is irreversible whereas in prerenal injury, the renal tubules function optimally to conserve salt and water48-54. In the pre- renal injury, the tubules conserve sodium and water leading to an elevated urine osmolality
>400-500 mosmol/l with a fractional excretion of sodium <1% and urine sodium <10-20 mEq/l.
All these parameters are slightly different in the pediatric age group as the renal tubules are immature. The urine osmolality is >350 mosmol/l with a fractional excretion of sodium <2.5%
and urine sodium <20-30 mEq/l55,56. In the case of injury, the urine osmolality is <350 mosmol/l
19
with a fractional excretion of sodium >2% and urine sodium >30-40 mEq/l. These values are relative and also requires a comparison with the value before the injury which is not available most of the cases.
These diagnostic confusion is very challenging in establishing the level of injury as the treatment protocols are quite different in both the scenarios.
Intrinsic renal disease
Hypoxic/ischemic acute kidney injury
Vasoconstriction followed by tubular necrosis is characteristic of Hypoxic/ischemic AKI. The endothelial cells of the vasculature is seen as the target in the intrinsic kidney injury emphasizing the role of the vasculature of the kidney in the pathogenesis of the acute and chronic injury of the kidney. A number of factors are involved in this including abnormality of peritubular capillary blood flow, dysfunction of the normal endothelial cells and distortion of the morphology and function of peritubularpericapillary structures57,58. The cell injury may be mediated by a number of mechanisms like generation of free radicals, , heat shock proteins, ATP depletion, alterations in cytoskeletons, endothelin, nitric oxide, etc59-74. An imbalance in the vasostimulatory- vasoconstrictive stimuli is known to initiate and progress the pathogenesis of AKI in hypoxia- ischemic injuries. Earlier it was thought that the injury is reversible but present studies show that the injury is long term with increased morbidity and precipitation of further acute events will establish a chronic illness75.
20 Figure 4: Intrinsic AKI
21 Nephrotoxic acute kidney injury
The kidney injury associated with medications and drugs either therapeutically or non- therapeutically leads to a complex mechanism of vasoconstriction, oxidant stress associated with heme and deposition of pigments in the lumen of the tubules76. The drugs include;
1. aminoglycoside antibiotics 2. intravascular contrast media 3. amphotericin B
4. ifosfamide 5. cisplatin 6. acyclovir 7. acetaminophen
All these drugs affect multiple functions of the kidney leading to tubular injury and acute kidney injury.
Uric acid nephropathy and tumor lysis syndrome
The highest risk for acute kidney injury is seen in ALL and B-cell lymphoma leading to tumor lysis syndrome or uric acid nephropathy77,78. It is due to the precipitation of crystals in the tubules of the kidney that lead to the obstruction of the urine flow. This is commonly seen in chemotherapy79. Precipitation of hypoxanthine and xanthine is seen in chemotherapy with allopurinol80. Rasburicase is therefore used in pediatric patients because of better solubility than allopurinol81. In tumor lysis syndrome, AKI may also result from the precipitation of calcium phosphate crystals from breakdown of tumor cells82.
22 Acute interstitial nephritis
The Acute interstitial nephritis (AIN) may be due to idiopathic causes or drug reactions. Children with AIN present with;
1. Rash 2. Arthralgias 3. Fever 4. Eosinophilia 5. Pyuria
Drugs that might cause AIN include83; 1. Methicillin
2. Penicillin analogs 3. Cimetidine 4. Sulfonamides 5. Rifampin 6. NSAIDs
7. Proton pump inhibitors
Rapidly progressive glomerulonephritis
Glomerulonephritis in any form may present with acute kidney injury and rapidly progressive glomerulonephritis usually presents with;
- hypertension - edema
23 - hematuria
- Increased BUN - Creatinine
RPGN is characterized by the presence of crescent formation which is very important pathologically as treatment depends on the diagnosis.
Vascular insults
One of the most common cause of Acute Kidney Injury in Children is due to HUS which leads to morbidity and mortality and may present with long term complications84.
Obstructive uropathy
Figure 5: Obstructive Uropathy
Acute kidney injury can be due to the obstruction of the urinary tract (obstruction in solitary kidney, bilateral involvement of ureters or obstruction of urethra). Congenital malformations like bilateral ureteropelvic obstruction, posterior urethral valve and bilateral obstructive
24
ureterocelescan lead to obstruction. Obstruction can also be due to acquired causes like kidney stones or rarely from neoplasms.
Management of acute kidney injury in children
Preventive measures
Studies from Nigeria on Acute Kidney Injury showed that AKI was due to volume depletion and other preventable causes85,86. Improved fluid balance, clearance of creatinine, lowering of serum creatinine levels were found in neonates with respiratory distress who were given theophylline intravenous infusion in the first hour of their birth87-89. Other studies show that this method improved the function of the kidneys and the reduction of the beta-2 microglobulin excretion.
Diuretics and dopamine receptor agonist
For prevention and limiting AKI, renal dose of dopamine and diuretics are used. Studies show the use of renal dose of dopamine, mannitol and diuretics for AKI90-97. Fenoldopam is a dopamine-1 receptor agonist which is short-acting, selective and potent98. The incidence of AKI with a reduced need for renal replacement therapy and reduced morbidity and mortality has been reported with meta-analysis of Fenoldopam from 16 trials99. Studies show that using Fenoldopam is effective in children who received therapy using ventricular-assist device100.
Reduction in injury and promotion of recovery
Several potential futuristic therapies are being explored for preventing and decreasing injury in children with AKI and also promote recovery. The therapies might include the presence of the following;
25
While there is no current specific therapy to prevent renal injury or promote recovery in human ATN, several potential therapies are being studied, and future management of AKI may also include101-105;
Antioxidant
anti-adhesion molecule therapy
vascular mediators
Multipotentmesenchymal stem cells have been useful in animal models106. But human studies have been disappointing which may be due to lack of validation of animal models in clinical practice107-108. The reason may be because intervention was done after the AKI set in109-110.
Prognosis of acute kidney injury
The etiology decides the prognosis of the acute kidney injury in children. When it is present as a part of anmultiorgan or multi system failure, then the morbidity and mortality rates are higher. A number of studies show that the evolution of acute kidney injury to a chronic nature is highly probable111-116. When there is a loss of nephrons, then the chances are higher especially in cases of HUS or Henoch–Schönleinpurpura where there is cortical necrosis. Even when there are reversible insults like hypoxic and ischemic injuries and nephrotoxicity, there might be subtle long term physiological and morphological changes in the long term. This is attributed to the injury to the kidney before the nephrons reach the adult stature. AKI in a neonate is correlated with chronic kidney disease116.Two studies are worth mentioning in this regard;
26
A study among six children showed that all of them had AKI that did not mandate a dialysis in the neonatal period. Out of the six children, three of them progressed to chronic renal failure, one requiring dialysis while only two were uneventful113.
In children who underwent bone marrow transplantation, incidence of AKI was high and those who survived, around 11% developed the CKD. AKI was the only regressor in this case117.
Cystatin-C
Cystatin-c is a cysteine protease inhibitor synthesized by all nucleated cells at a constant rate8. It is freely filtered by glomerulus and reabsorbed completely by the proximal convoluted tubules and not secreted. Thus it is a promising tool for diagnosis of AKI. Age, gender, muscle mass has negligible effect on serum cystatin C unlike creatinine levels making it a potential substrate to be explored as a diagnostic marker for renal dysfunction.
Serum creatinine is used more frequently in clinical practice for the assessment of renal function.
But serum creatinine is not reliable as their levels are affected by age, sex, tubular secretion, diet and physical activity. Presence of many confounding factors reduces the sensitivity and specificity of serum creatinine for measuring glomerular filtration rate (GFR)118. These limitations are overcome partly by using Cockcroft-Gault119 and MDRD (Modification of Diet in Renal Disease)120, but these equations are not completely valid or reliable120-122.
The recent advancements in the field of biochemistry has brought us to the door of a more reliable marker serum Cystatin-C than serum creatinine123-125. It belongs to the family of inhibitors of lysosomal cysteine protease that are present in all the cells that are nucleated and is a 13-kDa protein.
27 Figure 6: Cystatin C
The excretion of cystatin C is completely dependent on GFR because of the free filtration of cystatin-C in glomerulus along with other features like126-129;
complete reabsorption and catabolism in the proximal tubule
lack of tubular secretion
The limitations of using serum creatinine and the advantages of using Cystatin-C
28
Serum creatinine is not the accurate measure of the GFR during the non-steady state of AKI130. This has led to the search for laboratory markers that are more specific than serum creatinine131,132.
Early and easily measurable marker is cystatin-C with a better diagnostic and prognostic value133,134. It is also superior to serum creatinine in terms of confounding factors135-145.
Previous relevant studies
Studies showed that cystatin c may detect AKI one or two days earlier than creatinine9. One more study showed serum cystatin-c was increasing higher according to AKI severity (Failure>injury>risk)10. They also showed that Serum cystatin-c levels of >0.70mg/L were associated with longer length of PICU stay and prolonged duration of mechanical ventilation.
While one more study showed that serum cystatin -C was poor biomarker for AKI.
Another study showed that cystatin-C has an acceptable prognostic value and diagnostic value for prediction of AKI in children11. Since the results are contradictory this study aims to find out the correlation between cystatin-C levels and pRIFLE stages whether increasing accordingly with AKI severity and also whether a single value of cystatin-c can predict patients survival12. Gap in the literature
The present study is to determine the prognostic ability of serum Cystatin C over serum creatinine among children with acute kidney injury in paediatric intensive care unit. The term AKI is used for defining an abrupt and potentially reversible form of structural or functional renal impairment. It seeks to include patients with early renal damage in an attempt to intervene early2. International society of nephrology 0 by 25 initiative estimates that out of 1.7 million death per year caused by AKI globally, around 1.4 million occur in low and middle income
29
Countries3. Hence there is a need to understand the incidence and determinants of AKI in our ICU thereby aiming to prevent AKI early.
There is also a need to evaluate a novel biomarker that can be easily done and will not only diagnose AKI but will also prognosticate AKI at once. Most of the studies on paediatric AKI is from western literature. Determinants of AKI and availability of laboratory tests for diagnosing and prognosingAKI in our part of country varies widely. The incidence of AKI in India according to studies done after 2010 varies widely between 14% and 42.9%.Using modified pRIFLE criteria is tedious work in an intensive care unit, as it encorporates changes in serum creatinine,fall in GFR, and changes in urine output4. Moreover, to calculate the fall in GFR one should know the baseline serum creatinine which is often unknown in most patients5. Additionally urine output guidelines are erroneous in detection of AKI in non-oliguric renal failure which is more common in children6,7.
Age, gender, muscle mass has negligible effect on serum cystatin C unlike creatinine levels making it a potential substrate to be explored as a diagnostic marker for renal dysfunction. Here, in this study it is going to be explored as a prognostic marker, whether a single value of serum cystatin can predict the childs survival in PICU.
30
MATERIALS AND
METHODS
31 MATERIALS AND METHODS
AIM OF THE STUDY
To determine the role of Cystatin C over serum creatinine for Acute Kidney Injury (AKI) in children admitted to Paediatric Intensive Care Unit (PICU).
OBJECTIVES Primary Objective
1. To determine the prognostic ability of cystatin C for Acute Kidney Injury (AKI) in patients admitted to Paediatric Intensive Care Unit (PICU).
Secondary Objective
1. To assess the role of cystatin –c levels on various acute kidney injury stages (as defined by pRIFLE criteria)
2. To determine the underlying factors which predispose patients to acute kidney injury.
STUDY DESIGN
Cross sectional study STUDY POPULATION
Patients admitted to PICU, Stanley Government Medical College.
STUDY PERIOD
April 2018- April 2019
32 SAMPLE SIZE
369 INCLUSION CRITERIA
All consecutive patients admitted to paediatric intensive care unit (PICU) whose care givers express willingness to participate in the study by signing a written informed consent.
EXCLUSION CRITERIA
1. Patients with H/O chronic kidney disease (CKD) or End Stage Renal Disease (ESRD).
2. Patients whose stay in PICU is less than a period of 24 hours.
3. Patients treated elsewhere for AKI and referred to Stanley Government Medical college METHODOLOGY
Sample Collection
Blood was withdrawn by venipuncture as per the unit protocol for measuring serum creatinine. Samples will also be subjected for serum cystatin-c analysis if children are diagnosed with AKI as per pRIFLE criteria. According to studies 50% had AKI on admission , almost half of the patients with AKI had their maximum RIFLE score within 24 hours , and almost nearly all developed AKI within 72 hours and 90.2 % had AKI on admission .
1) Glomerular filtration rate will be computed using Schwartz Equation as follows:
=
33 Where, l=length in cms
Pcr=plasma creatinine in mg/dl
K= constant of proportionality that reflects the relationship between urinary creatinine excretion and units of body size
( k= 0.33 in preterm infants, 0.45 in full term infants,0.55 in children and adolescent girls, 0.70 in adolescent boys).
(Baseline serum creatinine will be obtained from patient records. If unavailable, the median level of serum creatinine for the age group will be assumed.)
2) Acute Kidney Injury (AKI) will be staged using pRIFLE criteria. pRIFLE criteria is as follows.
3) Concentration of serum creatinine will be estimated by Jaffes method
4) Cystatin C concentration will be determined by particle enhanced turbidimetricimmuno assay.
Case selection and Recruitment
34
All cases admitted in PICU were part of the study. The children who fulfilled pRIFLE criteria for Acute Kidney Injury were classified into the AKI group. A total of 111 cases were initially suspected with AKI and serum cystatin-C was measured. Out of 111, 40 of them later did not fit p-RIFLE criteria and were classified as non-Kidney Injury. Remaining 71 cases were classified into AKI group.
PRIVACY/CONFIDENTIALITY OF STUDY SUBJECTS:
Privacy of the subjects shall be maintained.
EXPECTED OUTCOME
1. Prognostic potential of Cystatin C for Acute Kidney Injury will be assessed.
2. Prevalence of Acute Kidney Injury among patients admitted in PICU.
3. Effect of pRIFLE stage on serum cystatin C level.
4. Determination of factors that predispose patients to Acute Kidney Injury in PICU.
STATISTICAL ANALYSIS
All data were recorded in structured questionnaires, coded and entered in Microsoft Excel. The data was then cleaned, checked for inconsistencies, missing values and prepared for analysis using SPSS v23. The data was then analysed for descriptive statistics and inferential statistics.
The tests for significance were run to statistically validate the data. The results were then tabulated and visualised in Microsoft word.
35
RESULTS
36 Results
During the study period, all cases admitted in PICU were recruited and classified into two groups based on the diagnosis of Acute Kidney Injury. Cases that did not fit into the inclusion criteria and also patients who left against medical advice or referred to higher center were excluded. The following flow chart shows the selection of cases. Out of total408 cases admitted in the PICU, 39 cases were excluded from the study and finally only 310 cases whose laboratory parameters could be measured were selected as a part of the study. The following flow chart shows the selection of cases. Out of 310 cases, 71 cases were children in acute kidney injury group while 239 cases were in the non-acute injury group.
In the AKI group, majority of them were in the <1 year group while in the non-AKI group majority of them were between 1 and 5 years. The chi-square analysis for comparison of groups show that the two groups differ significantly with p<0.005. The distribution of gender between the two groups show that in both the groups, females were in majority with 53.5% (n=38) in AKI group and 54.8% (n=131) in non-AKI group. The two groups were comparable gender wise.
Both the groups are comparable in terms of age and gender with not much difference. Out of 71 children under this group, 41 (57.7%) of them presented with AKI on admission, 22 (31.0%) of them presented with AKI at 24 hours, 5 (7.0%) of them presented with AKI at 48 hours and 3 (4.2%) of them presented with AKI at 72 hours. Visualization of the serum creatinine values between the two groups show that they are similar. Mann-Whitney U test shows that there is no significant difference (p>0.05).
37
GFR in two groups significantly differ with repeated measures ANOVA showing a F value of 2411.93 with p-value <0.005 in non-AKI group while in the AKI group F is 411.09 with p-value
<0.005.
Out of 71 patients in the AKI group, 63 (88.7%) of them were serum cystatin positive while in non-AKI group, only 7 (2.9%) were cystatin positive. Chi-square analysis shows that these values differ significantly between two groups with p<0.005.
The mean cystatin values in the AKI group was with a mean of 1.31 with a standard deviation of 0.53 ranging between 0.64 to 3.22. Out of the 71 patients in the AKI group, 29.6% (n=21) expired compared to the 10.5% (n=25) in the non-AKI group with a significant chi-square analysis.
Tests show that Serum Cystatin-C and pRIFLE are positively correlated with a correlation of 0.897 with a p-value highly significant (p<0.005). The logistic regression of Serum Cystatin-C on death and AKI shows that serum cystatin-C is a significant predictor with significance
<0.005.
Serum Cystatin-C predicts AKI with 88.73% sensitivity and 97.07% specificity.
Serum Cystatin-C predicts Death with 43.75% sensitivity and 81.30% specificity.
38 Findings
Selection and Distribution of Cases
During the study period, all cases admitted in PICU were recruited and classified into two groups based on the diagnosis of Acute Kidney Injury. Cases that did not fit into the inclusion criteria and also patients who left against medical advice or referred to higher center were excluded. The following flow chart shows the selection of cases. Out of total 408 cases admitted in the PICU, 39 cases left the study and finally only 310 cases whose laboratory parameters could be measured were selected as a part of the study. The following flow chart shows the selection of cases. Out of 310 cases, 71 cases were children in acute kidney injury group while 239 cases were in the non-acute injury group.
Figure 7: Selection and Distribution of Cases
Total number of cases in PICU =408 Children
Selection of cases after referral to higher center and left against medical advice=369 Children (39 Excluded)
Selection of cases based on the parameters measured
(excluding cases where adequate laboratory parameters
could not be measured)=310 cases
39
Algorithm for choosing patients for Serum Cystatin-C
Out of total 310 cases selected for the study, serum creatinine values were serially measured as per the unit protocol . Then based on clinical signs and symptoms, a diagnostic decision was made to measure serum cystatin values only in selected cases.
The following chart shows the selection of cases for measuring serum cystatin-C.
Based on the follow up and final diagnosis, the following groups were arrived at
Serum Cystatin-C positive was found in 70 cases
Serum Cystatin-C was negative in 41 cases
Out of 310 cases, based on clinical signs and diagnostic decision based on clinician's discretion
111 cases selected for measuring Serum Cystatin-C
199 cases were classifed as Non-AKI and considered to be Serum Cystatin-C
negative
Total of 310 cases selected for the study
40
Group Acute Injury Group Non-acute Injury Group
Number of cases 71 239
Table 1: Distribution of cases among the two groups
Figure 8: Distribution of cases among the two groups
71 239
N U M B E R O F C A S E S
Acute Injury Group Non-acute Injury Group
41 Age distribution
The following tables and figures show the age distribution of the children. In the AKI group, majority of them were in the <1 year group while in the non-AKI majority of them were between 1 and 5 years. The chi-square analysis for comparison of groups show that the two groups differ significantly with p<0.005.
AKI Group Non-AKI Group Chi-square
Analysis & p- value
Age
Distribution
Frequency Percent Frequency Percent
12.36
P=0.0047
Significant
<1 year 42 59.2 81 33.9
1-5 years 16 22.5 94 39.3
5-12 years 13 18.3 64 26.8
Total 71 100.0 239 100.0
Table 2: Age Distribution
42 Figure 9: Age Distribution
0 10 20 30 40 50 60 70 80 90 100
<1 year 1-5 years 5-12 years
Age Distribution
AKI Group Non-AKI Group
43 Gender Distribution
The distribution of gender between the two groups show that in both the groups, females were in majority with 53.5% (n=38) in AKI group and 54.8% (n=131) in non-AKI group. The two groups were comparable gender wise.
AKI Group Non-AKI Group Chi-square
Analysis &
p-value
Gender Frequency Percent Frequency Percent
5.09
P=0.02
Significant
Male 33 46.5 108 45.2
Female 38 53.5 131 54.8
Total 71 100.0 239 100.0
Table 3: Gender Distribution
44 Figure 10: Gender Distribution
0 20 40 60 80 100 120 140
Male Female
Gender Distribution
AKI Group Non-AKI Group
45 Age and Gender Distribution among the two groups
The following tables shows the comparison of age and sex between the two groups. Both the groups are comparable in terms of age and gender with not much difference.
Age and Sex Distribution among AKI group Sex Total
Female Male
Age <1 year Frequency 25 17 42
% within Age 59.5% 40.5% 100.0%
1-5 years Frequency 7 9 16
% within Age 43.8% 56.3% 100.0%
5-12 years Frequency 6 7 13
% within Age 46.2% 53.8% 100.0%
Total Frequency 38 33 71
% within Age 53.5% 46.5% 100.0%
Table 4: Age and Sex Distribution among AKI group
46
Age and Sex Distribution among non-AKI group Sex Total
Male Female
Age < 1 year Count 38 43 81
% within Age 46.9% 53.1% 100.0%
1-5 Years Count 38 56 94
% within Age 40.4% 59.6% 100.0%
5-12 Years Count 32 32 64
% within Age 50.0% 50.0% 100.0%
Total Count 108 131 239
% within Age 45.2% 54.8% 100.0%
Table 5: Age and Sex Distribution among non-AKI group
47 Characteristics of the AKI group
Out of 71 children under this group, 41 (57.7%) of them presented with AKI on admission, 22 (31.0%) of them presented with AKI at 24 hours, 5 (7.0%) of them presented with AKI at 48 hours and 3 (4.2%) of them presented with AKI at 72 hours. The following table and figure shows the time of presentation with AKI among the children.
Time of presentation Frequency Percentage
On Admission 41 57.7
At 24 Hours 22 31.0
At 48 Hours 5 7.0
At 72 hours 3 4.2
Table 6:Time of presentation with AKI among the children
48
Figure 11: Time of presentation with AKI among the children
41 22 5 3
57.7 31 7 4.2
O N A D M I S S I O N A T 2 4 H O U R S A T 4 8 H O U R S A T 7 2 H O U R S
TIME OF PRESENTATION WITH AKI AMONG THE CHILDREN
Frequency Percentage
49 Physical characteristics of the children of the study
Height of the children in the two groups
The following tables and figures show that the height of the children between two groups are comparable. There is no significant difference between the two groups.
Height (m) AKI Group Non-AKI Group Mann-Whitney U test
p-value
Mean 0.76 0.89
46.17 P=0.65
Not Significant
Median 0.71 0.84
Mode 0.55 0.76
Standard Deviation 0.247 0.28
Minimum 0.42 0.42
Maximum 1.29 1.5
Table 7: Height of the children in the two groups
50 Figure 12: Height of the children in the two groups
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281
Height Distribution of Two Groups
AKI Non AKI
51 Weight of the children in the two groups
The following tables and figures show that the weight of the children between two groups are comparable. There is no significant difference between the two groups.
Weight (Kg) AKI Group Non-AKI Group Mann-Whitney U test
p-value
Mean 9.43 13.5
28.19 P=0.89
Not Significant
Median 7.3 11
Mode 12 10
Standard Deviation 6.4 9.56
Minimum 2.6 2.1
Maximum 30 50
Table 8: Weight of the children in the two groups
52 Figure 13: Weight of the children in the two groups
0 0.5 1 1.5 2 2.5 3 3.5
1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281
Weight Distribution of the Children
AKI Non AKI
53 BMI of the children in the two groups
The following tables and figures show that the BMI of the children between two groups are comparable. There is no significant difference between the two groups.
BMI AKI Group Non-AKI Group Mann-Whitney U
test p-value
Mean 14.52 14.92
69.01 P=0.71
Not Significant
Median 14.41 14.58
Mode 13.0 14.0
Standard Deviation 2.55 2.95
Minimum 6.03 7.08
Maximum 23.31 25.51
Table 9: BMI of the children in the two groups
54 Figure 14: BMI of the children in the two groups
0 0.5 1 1.5 2 2.5 3 3.5
1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281
BMI Distribution of the Children
AKI Non AKI
55 Body Surface Area of the children in the two groups
The following tables and figures show that the BSA of the children between two groups are comparable. There is no significant difference between the two groups.
BSA (square-meter) AKI Group Non-AKI Group Mann-Whitney U test
p-value
Mean 0.12 0.204
122.01 P=0.539 Not Significant
Median 0.07 0.133
Mode 0.16 0.105
Standard Deviation 0.198 0.201
Minimum 0.02 0.01
Maximum 0.54 1.00
Table 10: BSA of the children in the two groups
56 Figure 15: BSA of the children in the two groups
0 0.5 1 1.5 2 2.5
1 9 17 25 33 41 49 57 65 73 81 89 97 105 113 121 129 137 145 153 161 169 177 185 193 201 209 217 225 233 241 249 257 265 273 281
BSA
AKI Non AKI
