Utilization AKI risk stratification, termed renal angina index (RAI)

Utilization of RenalAngina Index for prediction of subsequent severe acute kidney injury Raina Kaur, Gurdeep Singh Dhooria, Puneet A.Pooni, Deepak Bhat, Sidharth Bhargava, Shruti Kakkar,  Department ofPediatrics, Dayanand Medical College and Hospital, Ludhiana, Punjab, IndiaCorrespondenceAddress:Dr Gurdeep SinghDhooriaDepartment ofPediatrics, Dayanand Medical College and Hospital, Ludhiana, Punjab, IndiaEmail:[email protected]  Abstract: Acute kidney injury (AKI) is independently associated with worsenedmorbidity and increased mortality in PICU. AKI risk stratification, termedrenal angina index (RAI) has been used in the west  to predict persistent severe AKI. Very fewstudies have been done on application of renal angina index in PICU of adeveloping country. Aim: To predictsevere subsequent AKI in children admitted in PICU using Renal Angina Index. Design: Prospective observationalstudy.

  Methods: Children admittedover one year in PICU between one month to 18 years of age with no previouskidney disease were included. RAI assessment was done from 8-12 hours ofadmission to PICU. RAI was calculated from product of Renal Risk  and Renal Injury score. Renal anginapositivity was defined as RAI  ? 8. Onday 4, serum creatinine is noted and GFR is calculated. RAI was correlated withpresence/absence of subsequent severe AKI.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

RAI positivity was also correlatedwith duration of PICU stay, need for dialysis, mechanical ventilation and mortality.Results: RAI positivity was seen in16.7 % cases. Of the RAI positive cases, 36.2 % cases developed AKI at 4 dayscompared to 2.3% in the RAI negative cases (p value <0.

001). Mean durationof PICU stay in RAI positive group was 7.19 ± 5.13 days vs 4.72 ± 2.

71 days in RAInegative group.  31%cases in RAI positive cases had poor outcome. Conclusions: Renal angina index could be used as a simple andimportant bedside tool to predict patients at risk of severe AKI.

Keywords:Acute kidney injury, children, critical care, risk stratification, renal anginaindex    Introduction: Acute kidneyinjury (AKI) is a significant problem in critical illness. Approximately 5–6%of all hospitalized adults and 10% of children suffer from varying degrees ofAKI 1. The presence of AKI in critical illness occurs at a rate of 10–15% andcarries a 50% mortality rate in children requiring dialysis 2,3,4. IncreasingAKI severity, characterized by serum creatinine (SCr)- and urine output(UOP)-based stratifications of AKI, is associated with increased mortality inadults5 and children.

6 Notably, small increases in SCr (0.3 mg/dl) mayreflect significant kidney damage and is associated with poor patient outcomes7,8.Consistently effective AKI therapy to prevent or limit the diseaseintensity is lacking, potentially due to delayed recognition of existing and/orongoing injury. AKI diagnosis is traditionally dependent on changes in serumcreatinine (SCr), a marker with limitations involving time, body habitus, sex,age, steady-state measurement, and patient condition. Primarily due to the lagin the rise of SCr, the diagnosis of AKI is often delayed, which creates asignificant barrier to effective early intervention. 9Treatment for acute myocardial infarction (MI) was transformed bythe use of troponin I measurements in patients with signs and symptoms of acardiac angina. Sensitivity and specificity of troponin elevations andelectrocardiographic changes for MI have allowed practitioners to instituteearly and life-saving therapy. However, whereas the novel AKI biomarkersrecently discovered may serve well as a renal troponin equivalent, AKI lacks animportant parallel to MI.

Simply put, AKI does not hurt.  In order to optimize the utility of AKIbiomarkers, screening systems are needed to identify patients who are at highrisk of developing AKI.9Goldstein SL recently proposed the empiric clinical model of renalangina to identify which critically ill patients would be at the greatest riskof AKI.

Using patient demographic factors and early signs of injury, renalangina is aimed to delineate patients at risk for subsequent severe AKI (AKIbeyond the period of functional injury) versus those at low risk. 10 In thisstudy, the concept of renal angina to improve prediction of subsequent severeAKI has been validated in Indian children admitted in PICU. Material AndMethods: Subjects: All children between 1 months-18 years age group admitted in PICUDuration: one year. Setting: PediatricIntensive Care Unit (PICU) of a tertiary care hospital in Ludhiana District,Punjab. Design: Prospectiveobservational studyInclusion Criteria: All critically ill children between 1 months to 18 years of age admittedto PICU were included in the study.Exclusion criteria: Children with previously known kidney disease and children with hospitalstay less than 72 hours were be excluded.The study was approved by the Institutional Ethics committee.

Written consent was obtained from the attendants of thechildren. Data Collection: Day 0 was consideredfirst day of PICU admission. Day 3 consisted of the time period between 72 and96 hours after PICU admission. Baseline data at admission included demographicinformation including age, sex, primary diagnosis, system involved, PediatricRisk of Mortality (PRISM-II) scores 11 within 24 hours of hospital admission,renal dysfunction using KDIGO (Kidney Disease Improving Global Outcomes)staging 12, and baseline serum creatinine measurement (SCr). Calculation of Renal angina index (RAI):The index calculation for the fulfilment of renal angina is assessed 8-12 hafter a patient is admitted to an intensive care unit and used for predictionof severe acute kidney injury 72 hours (3 days) later. Risk factors aredetermined as shown and assigned a point value (1, 3, and 5, where 1 denotesthe lowest risk and 5 denotes the highest risk). Mechanical ventilation andvasoactive support should be used within the 12-h timepoint but are notrequired to be simultaneous for a patient to be scored 5 points.

Injury strata was assignedas depicted to a patient as appropriate.  Percentage fluid overload (FO %) wascalculated by {Fluid in(ml)-fluid out (ml) ÷ Patient Weight (gm)} x 100). GFRwas based on estimated creatinineclearance (eCrCl) calculated by the Schwartz equation 13, for determinationof the RAI. Calculation of RenalAngina Index (Range 1 to 40) is a multiplication of the risk and injury scoresassigned (Risk score x FO% score OR Risk score x GFR score), whichever is worseof the two is chosen. The index RAI ?8 was considered fulfillment of renalangina 14. Fulfillment or the absence of renal angina was denoted ‘RApositive’ or ‘RA negative’.    Outcomes: The primary outcome was the presence of severe AKI on Day3 (Day3-AKI). Severe AKI was defined by the KidneyDisease Improving Global Outcomes (KDIGO) AKI classification stage ?2: serumcreatinine of 200% baseline (a decrease in eCrCl of ?50% from baseline) (26).

 Day3 was chosen since most PICU patientsdevelop AKI within this timeframe and is a clinically relevant time frame forAKI management. Secondary outcomesincluded use of renal replacement therapy (RRT), need and duration ofmechanical ventilation, length of PICU stay (LOS), and incidence of mortality. Statistical AnalysesAll statistical analyses wereperformed using STATA version 12 (StataCorp, CollegeStation, TX), SAS version 9.3 (SAS Institute, Cary, NC), and R version 2.

14.1(R Development Core Team, Vienna, Austria). An a prioristudy sample size of 5250 patients was expected. The continuous data were summarized usingdescriptive statistics (mean ± standard deviation). Statistical differencesbetween the mean values were compared using Student’s t-test.

A differencebetween the two values was considered to be significant if the P < 0.05. Categoricalvariables were summarized using frequency and proportion and compared bychi-squared or Fisher's exact tests. An RAI cut-off of ?8 was used to definerenal angina fulfillment ANG(+) and this cut-off was used for operativecharacteristics (20). Simple and multivariablelogistic models were used to predict day 3 AKI using RAI.

Area under the curve(AUC) values were calculated for the prediction model (RAI) and compared usingDeLong’s method (27). In all analyses, a P value<0.05 was considered statistically significant.Results: Of the total ____________ patients admittedduring the one year period, 413 (    %)patients wereenrolled for study. Number______ could not enrolled for differentreasons CKD (n=   ), age less than 2months (n=______), hospital stay less than 72 hours(n=_______). Approximately one third ofpatients i.

e. 69/413(16.7 %) were RA + on day day 0 of PICU admission.

Meanage in RA +ve group was 5.92 ± 5.30 yearsas compared to RA -ve group 5.88 ± 5.32 years. Age, gender, admittingdiagnosis or primary system did not affect RA positivity.

Sepsis was thediagnosis in 21 % cases and did not affect RA status. Patients in RA+ve group had higher mean PRISM-IIscores (18.62 ± 6.49) compared to RA -ve group (12.74 ± 6.49) (p value < 0.001).

Additionally,patients in RA+ grouphad longer duration of mechanical ventilation (mean 4.94 ± 4.10 days vs mean 1.08 ± 2.

68 days) (pvalue < 0.001), PICU stay (mean  7.19 ± 5.13 days vs mean  4.

72 ± 2.71) (p value < 0.001), need fordialysis ( 23.2 % vs 0.6% %, p value < 0.001)  and higher mortality( 31.09 % vs 2 % , p value < 0.001).

Predictionof subsequent, severe AKI by Renal Angina Index on admission.  Of thetotal 413 patients enrolled, 33 patients developed Day3-AKI (8%).  Theincidence of Day 3 AKI was significantly higher in patients with RAI ?8 (RA +group) 25 of 69 (36.2 %) versus 8 of 344 (2.3 %); P= <0.001. Day 0 RAI positivity(RAI > 8) predicted Day-3 AKI with anAUC of 0.

883 95% confidence interval (CI) = 0.823-0.943. RAI > 8positivity had ahigh negative predictive value (NPV)of 97.

67% % (95% CI = 95.84-98.7%), with sensitivity and specificity of  75% and 88.42 % , respectively, and positivepredictive value (PPV) of 35.29% (95% CI = 27.92-43.

44 %). RAIprediction by GFR criteria and Fluid overload (FO %) criteriaThepredictive value of  RAI was broken down by composite factorsof kidney injury. The predictive value for Day-3 AKI by GFR score alone by AUC values was consistentlysuperior when compared tofluid overload score (FO %) AUC 0.877 (95% CI = 0.817-0.936) vs0.774 (95% CI = 0.

685- 0.864).The AUC forRAI for Day-3 AKI improved whenRAI incorporated worse of the two scores (GFR score/FO score). (AUC0.883(95% CI= 0.823-0.943).  RAIversus KDIGO stage and PRISM score Prediction of RA for Day-3 AKI was superior to KDIGOstage 1 injury at admission;fulfillment of renal angina demonstrated higher sensitivity (27.

27%), PPV of 25%, NPV of 93.63%and a higherYouden’s index ( ____________) than KDIGO stage 1,although specificity was found to be higher with KDIGO stage 1 (92.89%). Similar results were seen when RA was compared to KDIGO stages 2–3 (Youden’s index=_____). When compared directly, RAI outperformed PRISM-II for theprediction of day 3 AKI. (AUC=0.

764)(95% confidence interval (CI) = 0.672-0.856).Discussion:  Renal angina index was developedby Goldstein to identify critically ill patients at greatest risk of AKI.10  In the current study, we looked at theapplicability of renal angina index in a tertiary care hospital of a developingcountry and show that RAI can be used as bedside tool to prediction severe AKI followingPICU admission and also outperforms stage 1 KDIGO renal injury and PRISM IIscore in this context. Fora clinician, the ability to predict the presence of severe AKI 3 days inadvance carries obvious benefit. RAI was derived as a compositeof risk factors and clinical signs of AKI. The logic behind the equationdictates that as a patient achieves higher risk they require less “clinicalsign of AKI” early on to fulfill renal angina.

Similarly, if a patient has lessrisk but shows more overt signs of clinical AKI signs, renal angina would alsobe fulfilled.15 RAIderivation was based on available AKI epidemiology reported in select pediatricpopulations: children admitted to the ICU carry increased risk over the generalpopulation (4.5–10%),16,17children receiving bone marrow transplantation have ~3× risk (11–21%)18, and those who are intubatedand on vasopressor support carry nearly 5× risk versus the general ICUpopulation (51%). 3The ‘signs of injury’ (i.e.

, kidney pain) in the RAI include GFR and fluidoverload.Fluids are the second mostcommon intervention in acutely ill patients (after oxygen). The benefits ofearly fluid resuscitation in patients with shock and acute kidney injury (AKI)are already accepted.

There is evidence that fluid administration beyond thecorrection of hypovolaemia is associated with increased morbidity, a longerhospital stay and mortality. In a recent article in Critical Care, Wang et al.analysed the data of 2526 patients admitted to 30 intensive care units (ICUs)in China and showed that even relatively small degrees of fluid overload wereindependently associated with an increased risk of AKI and mortality 19.In the RajitBasu et al study,based on the most optimal Youden’s index (0.49) and highest negative predictivevalue (to safely rule out development of subsequent AKI), an RAI > 8 wastaken as cutoff to label Renal Angina positivity.15 Only day 3 AKI was chosen to define outcome as most PICUpatients develop AKI within this time frame and it surpasses the time frame offunctional AKI (prerenal AKI). Also, time frame of 8 h was kept to assess fluidoverload as it was beyond the generally accepted window of ‘early goal-directedtherapy’ (EGDT) of resuscitation. 20    In our study atotal of 413 patients were included.

Day 0 Renal Angina positive was seen in16.7% patients. Of  renal angina positivepatients 36.

2 %  developed subsequentsevere AKI compared to 2.3 % of the other group, which was highly significant(p<0.001). Performance of the test was also calculated.

Sensitivity andspecificity came tobe 75% and 88.42% respectively. Positive predictive value waslow (35.

29%) whereas a high negative predictive value of 97.67% was present.AUC for the same came to be 0.877.When the RAI was derived basedon either GFR score or Fluid overload alone, it was still able to predict Day-3AKI. Although FO score did not perform as well as GFR score, FO score wassimilar to PRISM-II scores for prediction of Day-3 AKI. More importantly, theprediction for Day-3 AKI improved when the worse value of FO or GFR score wasused compared to using  GFR score alone.

Also, Renal angina outperformed early signs of injury i.e. KDIGO stages I. Ourresults were similar to those seen in the Rajit Basu et al. study.

15RAI is a clinical tool developedto identify critically ill children with evidence of early injury who are atrisk for developing severe AKI. While effectively ruling out patients at lowrisk for severe AKI ( high NPV), the PPV of renal angina for predicting severe AKI was less in the population studied,highlighting the need for additional biomarker beyond established clinicalpredictors. RAI might also help guide biomarker measurement and improveperformance by focusing on the at risk population and thereby reducing false positive results, a problem observed by lessselective measurement of biomarkers, such as troponin. 21 Shina Menon et al in their study showed, renal angina positivity at admission in 32.6% patients with significantly longerduration of mechanical ventilation, organ failure days,and increased mortality than with absence of renal angina.

Day3-AKI waspresent in 9.6% patients, 80% fulfilling Day0renal angina. Also,incorporation of urinary biomarkers (urinaryneutrophil gelatinase-associated lipocalin) into the RAImodel increased the specificity and positive likelihood, and demonstrated improvementfor the prediction of Day3-AKI with increased AUC of RAI for Day3-AKI from 0.80 to 0.97. 22Limitation of the study was thatbaseline creatinine was calculated from admission serum creatinine and patientheight using the Schwartz correction.

This was done, as most patients did nothave their lowest creatinine value (up to 3 months before PICU admission) to establisha reference value. In our study, to define severe subsequent AKI, estimatedcreatinine clearance criteria of KDIGO (SCr of 200% baseline or  decrease in eCCl of ?50% from baseline) wastake to define primary outcome and urine output criteria was not taken.  Thus, renal angina index couldbe used as a simple and important bedside tool without the need of anyexpensive equipment to detect patients at risk of severe AKI. This can allow usto use novel AKI biomarker or therapy trial, which could ultimately guidetreatment strategy in critically ill children. Future studies can be done in ourcountry incorporating biomarkers to RAI to further enhance the predictability ofAKI in critically ill children.             Financial support and sponsorship: Nil.Conflicts of interest: Thereare no conflicts of interest.

 References: 1. Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morgera S,Schetz M, Tan I, Bouman C, Macedo E, Gibney N, Tolwani A, Ronco C, Beginningand Ending Supportive Therapy for the Kidney (BEST Kidney) Investigators. Acuterenal failure in critically ill patients: a multinational, multicenter study. JAMA.2005 Aug 17; 294(7):813-8.2.  Hui-Stickle S, Brewer ED, Goldstein SL.Pediatric ARF epidemiology at a tertiary care center from 1999 to 2001.

Am JKidney Dis. 2005;45:96–101. 3. Akcan-Arikan A,Zappitelli M, Loftis LL, Washburn KK, Jefferson LS, Goldstein SL. ModifiedRIFLE criteria in critically ill children with acute kidney injury. Kidney Int.2007;71:1028–1035.

4. Symons JM, ChuaAN, Somers MJ, Baum MA, Bunchman TE, Benfield MR, et al. Demographiccharacteristics of pediatric continuous renal replacement therapy: a report ofthe prospective pediatric continuous renal replacement therapy registry. Clin JAm Soc Nephrol.

2007;2:732–738. 5. Srisawat N,Hoste EE, Kellum JA. Modern classification of acute kidney injury. Blood Purif.2010;29:300–307.

6. Slater MB,Anand V, Uleryk EM, et al. A systematic review of RIFLE criteria in children,and its application and association with measures of mortality and morbidity.Kidney Int. 2012;81:791–798.

7. Hoste EA,Kellum JA. Incidence, classification, and outcomes of acute kidney injury.Contrib Nephrol. 2007;156:32–38. 8. Zappitelli M,Bernier PL, Saczkowski RS, Tchervenkov CI, Gottesman R, Dancea A, Hyder A,Alkandari O.

A small post-operative rise in serum creatinine predicts acutekidney injury in children undergoing cardiac surgery. Kidney Int.2009;76:885–892. 9. Basu RK, ChawlaLS, Wheeler DS, Goldstein SL. Renal angina: an emerging paradigm to identifychildren at risk for acute kidney injury. Pediatric Nephrology2012;27:1067-1078. 10.

  Goldstein SL, Chawla LS. Renal angina. Clin JAm Soc Nephrol. 2010;5:943-9.11.

 Pollack MM, Ruttimann UE, Getson PR: Pediatric risk of mortality (PRISM) score. Crit Care Med 16: 1110–1116, 198812.       Kellum JA, Lameire N,  Aspelin P, et al.KDIGO Clinical Practice Guideline for Acute Kidney Injury 2012, Kidney IntSuppl , 2012, vol. 2 (pg.

1-138)13.       Schwartz GJ, Work DF: Measurement andestimation of GFR in children and adolescents. Clin J Am Soc Nephrol 2009;4:1832–1843.

14.       Goldstein SL, CurrierH, Graf C et al. Outcome in children receiving continuous venovenoushemofiltration. Pediatrics 2001; 107: 1309–1312.15.

       Basu RK , ZappitelliM, Brunner L et al.  . Derivation andvalidation of the renal angina index to improve the prediction of acute kidneyinjury in critically ill children.

Kidney Int 2014; 85: 659–667.16. Schneider J, Khemani R, Grushkin C, Bart R.

Serum creatinine asstratified in the RIFLE score for acute kidney injury is associated withmortality and length of stay for children in the pediatric intensive care unit.Crit Care Med. 2010;38(3):933-9.

17. Bailey D, Phan V, Litalien C, Ducruet T, Mérouani A, Lacroix J,Gauvin F. Risk factors of acute renal failure in critically ill children: Aprospective descriptive epidemiological study.

Pediatr Crit Care Med. 2007 Jan;8(1):29-35.18. Michael M, Kuehnle I,Goldstein SL. Fluid overload and acute renal failure in pediatric stem celltransplant patients. Pediatr Nephrol.

2004 Jan; 19(1):91-5.19. Ostermann M, Straaten HM, Forni LG. Fluid overload and acutekidney injury: cause or consequence? Crit Care.

2015;19:443.20. Dellinger RP, Levy MM, Rhodes A, et al.

Surviving sepsis campaign: international guidelines for management of severe sepsisand septic shock: 2012. Crit Care Med. 2013;41:580–637.21. Malhotra R, Siew ED.Biomarkers for the Early Detection and Prognosis of Acute Kidney Injury.

Clin JAm Soc Nephrol. 2017;12:149-173.22.  Menon S, Goldstein SL, Mottes T, Fei L,Kaddourah A, Terrell T, et al.  Urinarybiomarker incorporation into the renal angina index early in intensive careunit admission optimizes acute kidney injury prediction in critically ill children:a prospective cohort study.

Nephrol Dial Transplant. 2016;31:586-94.  Several AKI biomarkershave demonstrated promising results for the identification and prediction ofAKI in children.21 Identifying patients at risk for severe andlong-lasting AKI in the PICU is crucial as risk stratification could allow morejudicious AKI biomarker assessment to drive therapeutic intervention, therebyincreasing their predictive performance and cost-effectiveness. 22