^{2}

^{2}

^{1}

^{1}

^{3}

^{1}

^{4}

^{2}

^{1}

^{1}

^{2}

^{3}

^{4}

The performance of glomerular filtration rate- (GFR-) estimating equations was studied against creatinine clearance measured by 24-hour urine collection (CrCl_{24h-urine}) in critically ill patients. _{24h-urine} were included. We estimated GFR using Cockroft–Gault (CG), modification of diet in renal disease study (MDRD), chronic kidney disease epidemiology collaboration (CKD-EPI), and Jelliffe equations. For the CG equation, we entered the actual weight in one calculation (CG_{actual-wt}), and if BMI ≥30 kg/m^{2}, we entered the ideal body weight (CG_{ideal-wt}) and the adjusted body weight (CG_{adjusted-wt}) in two calculations. We calculated the MDRD equation based on 4 (MDRD-4) and 6 variables (MDRD-6). The performance of these equations was assessed by different ways including Spearman correlation, bias (difference between estimated GFR and CrCl_{24h-urine}), precision (standard deviation of bias), and Bland–Altman plot analysis. _{24h-urine} 108 ± 69 ml/min/1.73 m^{2}). The correlations between the different equations and CrCl_{24h-urine} were modest (_{actual-wt} (21 ml/min), CG_{adjusted-wt} (12 ml/min), and MDRD-6 (-10 ml/min) equations. Precision ranged from 46 to 54 ml/min. The sensitivity of equations to correctly classify CrCl_{24h-urine} 30–59.9 ml/min/1.73 m^{2} was 17.2% for CG_{actual-wt}, 30.0% for CG_{ideal-wt}, 31.0% for CG_{adjusted-wt}, 31.0% for MDRD-4, 39.1% for MDRD-6, 13.8% for CKD-EPI, and 34.5% for Jelliffe equation. _{24h-urine} and to correctly classify GFR into clinically relevant ranges that usually determine dosing of medications.

Appropriate dosing of medications is frequently dependent on renal function. The Kidney Disease Improving Global Outcomes (KDIGO) clinical practice guidelines consider GFR as the preferred measure of kidney function rather than serum creatinine (Cr) and recommend estimating GFR in most circumstances and measuring it when greater accuracy is required [

Multiple equations have been produced to estimate GFR, including Cockroft–Gault (CG) [

This is a substudy of the PermiT (Permissive Underfeeding versus Target Enteral Feeding in Adult Critically Ill Patients) trial (_{24h-urine}) measurement. Patients with end-stage renal disease requiring dialysis and those with anuria for any other reasons were excluded. Subjects with missing variables needed for calculations of the different equations were also excluded. The original trial was approved by the Institutional Review Board of Ministry of National Guard Health Affairs, Riyadh, Saudi Arabia.

At baseline, we collected data on patients’ demographics, chronic comorbid conditions, admission category (medical, surgical, and trauma), presence of traumatic brain injury, presence of sepsis on admission, Acute Physiology and Chronic Health Evaluation (APACHE II) score, Sequential Organ Failure Assessment (SOFA) score, use of mechanical ventilation, need for vasopressor therapy because of shock, daily caloric and protein intake, and laboratory results. We also obtained data about clinical outcomes, including mortality, duration of mechanical ventilation, and length of stay in the ICU and hospital.

In the study patients, urine was collected over 24 hours at baseline and then weekly as required by the trial when applicable. Measured CrCl_{24h-urine} was then calculated using the standard equation: (urine Cr × urinary flow in ml/min)/serum Cr, where urine and serum Cr were expressed in

We estimated GFR using CG [_{actual-wt}), and if body mass index (BMI) ≥30 kg/m^{2}, the ideal body weight (CG_{ideal-wt}) and the adjusted body weight (CG_{adjusted-wt}) were used in two calculations. We calculated the MDRD equation based on 4 (MDRD-4) and 6 variables (MDRD-6). Acute kidney injury in the enrollment day was assessed using the KDIGO classification [

Renal function estimating equations.

For males: [(140—age) × actual BW]/sCr × 72 |

For females: ([(140—age) × actual BW]/sCr × 72) × 0.85 |

sCr in mg/dL |

The equation was calculated three times: |

(1) Using actual BW for all patients |

(2) Using actual BW for patients with BMI <30 kg/m^{2} and ideal BW for those with BMI >30 kg/m^{2} |

(3) Using actual BW for patients with BMI <30 kg/m^{2} and adjusted BW for those with BMI >30 kg/m^{2} |

Males: 50 kg + 2.3 kg for each inch above 60 inches of height |

Females: 45.5 kg + 2.3 kg for each inch above 60 inches of height |

Adjusted BW = ideal BW + [0.4 × (actual BW—ideal BW)] |

^{2}) |

Four-variable equation: 175 × sCr ^{−1.154} × age^{−0.203} × 0.742 (if female) |

Six-variable equation: 170 × sCr^{−0.999} × (Age)^{−0.176} × (0.762 if patient is female) × (BUN)^{−0.170}× (albumin)^{0.318} |

sCr in mg/dL, albumin in g/dL, BUN in mg/dL; to convert BUN from mmol/L to mg/dL, divide by 0.3571 |

^{2}) |

For females with sCr ≤ 0.7: GFR = 144 × (sCr/0.7^{)−0.329} × (0.993)^{age} |

For females with sCr > 0.7: GFR = 144 × (sCcr/0.7)^{−1.209} × (0.993)^{age} |

For males with sCr ≤ 0.9: eGFR = 141 × (sCr/0.9)^{−0.411} × (0.993)^{age} |

For males with sCr > 0.9: eGFR = 144 × (sCr/0.9)^{−1.209} × (0.993)^{age} |

Age in years and sCr in mg/dL |

^{2}) |

For males: (98–16) × (age—20/20)/sCr |

For females: [(98–16) × (age—20/20)/sCr] × 0.9 |

Age in years and sCr in mg/dL |

BUN: blood urea nitrogen, BW: body weight, sCr: serum creatinine

Continuous variables were reported as mean and standard deviation (SD). The coefficient of variation (SD/mean × 100) for CrCl_{24h-urine} and the estimated GFR were also calculated. Categorical data were presented as frequency with percentage. Chi square test was used to assess between-group differences in categorical variables. Student’s

The performance of the GFR-estimating equations compared with CrCl_{24h-urine} was assessed in several ways. Correlations were reported using Spearman correlation coefficient (_{24h-urine} and each of the equations estimating GFR [_{24h-urine}. An acceptable between-method error was defined as 30% or less [_{24h-urine} measurements. The 2002 Kidney Disease Outcomes Quality Initiative guidelines recommended that ≥90% of estimates be within 30% [_{24h-urine} on the

The predictive performance of the different equations was assessed when CrCl_{24h-urine} was <30, 30–59.9, 60–130, and > 130 ml/min. We also assessed the ability (sensitivity) of the different equations to correctly classify CrCl_{24h-urine} within clinically relevant ranges (<30, 30–59.9, 60–130, and >130 ml/min). Moreover, Spearman correlation was calculated in selected subgroups of patients: age < versus ≥ 65 years, BMI < versus ≥ 30 kg/m^{2}, APACHE II score < versus ≥ median value, which was 20, admission categories (medical, surgical, and nonoperative trauma), diagnosis of traumatic brain injury, presence of sepsis on ICU admission, baseline Cr < versus ≥ 110 _{24h-urine} >130 ml/min/173 m^{2}) [

Tests were two-sided and statistical significance was determined at

Two hundred and thirty-seven patients were included in this study. Table ^{2}), 31.7% had diabetes, 98.7% required mechanical ventilation, and 26.2% had traumatic brain injury.

Characteristics and outcomes of the 237 patients in the study cohort.

All patients | |
---|---|

45.0 ± 20.2 | |

60 (25.3) | |

166.3 ± 9.7 | |

78.2 ± 19.6 | |

^{2}), mean ± SD | 28.3 ± 7.2 |

Diabetes | 75 (31.7) |

Chronic respiratory disease | 27 (11.3) |

Chronic cardiac disease | 23 (9.7) |

Immunocompromised disorder | 6 (2.5) |

Chronic renal disease | 6 (2.5) |

Chronic liver disease | 11 (4.6) |

Medical | 112 47.3) |

Surgical | 11 (4.6) |

Nonoperative trauma | 114 (48.1) |

62 (26.2) | |

51 (21.5) | |

20.4 ± 8.1 | |

10.0 ± 2.8 | |

135 (57.0) | |

234 (98.7) | |

Standard feeding | 120 (50.6) |

Permissive underfeeding | 117 (49.4) |

Total caloric intake (kcal/day)—mean ± SD | 1143.6 ± 466.1 |

Total protein intake—(g/day) mean ± SD | 55.9 ± 21.0 |

Inclusion blood glucose—(mmol/L), mean ± SD | 9.0 4.1 |

Creatinine—( | 100.8 ± 93.9 |

Bilirubin—( | 25.4 ± 39.2 |

Platelets—(10^{9}/L), mean ± SD | 214 ± 128 |

Albumin—(g/L), mean ± SD | 28.8 ± 5.6 |

Mechanical ventilation duration—(days), mean ± SD | 13.0 ± 25.0 |

ICU LOS—(days), mean ± SD | 15.9 ± 10.5 |

Hospital LOS—(days), mean ± SD | 59.3 ± 83.1 |

90-day mortality | 61 (25.7) |

ICU mortality—no. (%) | 38 (16.0) |

Hospital mortality—no. (%) | 56 (23.6) |

ICU-acquired infections—no. (%) | 96 (40.5) |

SD: standard deviation; APACHE: acute physiology and chronic health evaluation; SOFA: sequential organ failure assessment; ICU: intensive care unit; LOS: length of stay.

The baseline serum Cr was 100.8 ± 93.9 _{24h-urine} >130 ml/min at baseline. The measured CrCl_{24h-urine} and estimated GFRs based on the various equations are presented in Figure _{24h-urine} was 108.4 ± 68.9 ml/min in the first 237 measurements. The estimated GFR by the different equations were 129.6 ± 65.6 ml/min for CG _{actual-wt} (_{ideal-wt} (_{adjusted-wt} (

Mean values of the measured creatinine clearance by 24-hour urine collection (CrCl_{24h-urine}) and estimated glomerular filtration rate by different equations. The first 237 24-hour urine samples were used in this analysis. Error bars represent 95% confidence interval. The difference between the different methods was significant (

The performance of the different GFR-estimating equations against CrCl_{24h-urine} is described in Table _{24h-urine} were significant (_{actual-wt} (21.1 ml/min), CG_{adjusted-wt} (11.5 ml/min), and MDRD-6 (-10.3 ml/min) equations. When using all 453 urine measurements, the bias was large and statistically significant for CG_{actual-wt} (27.4 ml/min), CG_{ideal-wt} (12.3 ml/min) CG_{adjusted-wt} (18.3 ml/min), MDRD-4 (7.1 ml/min), and MDRD-6 (−5.7 ml/min) equations. In both calculations, CKD-EPI and Jelliffe equations had no significant bias. The error was >70% for all equations.

Predictive performance of equations estimating glomerular filtration rate compared with creatinine clearance (CrCl_{24h-urine}) measured by 24-hour urine collection.

CrCl_{24h-urine} | CG_{actual-wt} | CG_{ideal-wt} | CG_{adjusted-wt} | MDRD-4 | MDRD-6 | CKD-EPI | Jelliffe | ||
---|---|---|---|---|---|---|---|---|---|

GFR estimate (mlH/min) ±SD | 108.4 ± 68.9 | 129.6 ± 65.6 | 113.5 ± 59.2 | 119.9 ± 59.9 | 108.9 ± 52.5 | 102.2 ± 48.7 | 102.1 ± 40.4 | 102.0 ± 49.3 | |

Coefficient of variation (%) | 63.6 | 50.6 | 52.2 | 50.0 | 48.2 | 47.7 | 39.6 | 48.3 | |

Correlation | 0.77 | 0.71 | 0.75 | 0.63 | 0.62 | 0.67 | 0.66 | ||

Bias (ml/min) | 21.1 | 5.0 | 11.5 | 0.5 | −10.5 | −6.3 | −6.4 | ||

Precision (ml/min) | ±46.0 | ±49.9 | ±45.9 | ±54.2 | ±53.9 | ±51.6 | ±51.9 | ||

Error (%) | 77.3 | 90.0 | 80.4 | 99.7 | 100.4 | 98.0 | 98.6 | ||

±15% | 12.7 | 27.4 | 30.0 | 25.7 | 22.8 | 28.3 | 25.7 | ||

±30% | 49.4 | 48.1 | 51.2 | 48.9 | 47.4 | 48.9 | 49.4 | ||

±50% | 66.7 | 72.2 | 71.3 | 69.2 | 75.1 | 70.5 | 57.4 | ||

GFR estimate (ml/min) ±SD | 102.7 ± 65.4 | 130.1 ± 65.9 | 114.9 ± 60.2 | 121.0 ± 60.6 | 109.8 ± 52.1 | 99.8 ± 47.1 | 103.2 ± 40.0 | 103.2 ± 49.3 | |

Coefficient of variation (%) | 63.7 | 50.7 | 52.4 | 50.1 | 47.4 | 47.2 | 38.8 | 47.8 | |

Correlation | 0.79 | 0.75 | 0.79 | 0.67 | 0.66 | 0.67 | 0.70 | ||

Bias (ml/min) | 27.4 | 12.3 | 18.3 | 7.1 | −5.7 | 0.53 | 0.53 | ||

Precision (ml/min) | ±43.0 | ±44.3 | ±40.9 | ±49.1 | ±49.3 | ±47.2 | ±46.8 | ||

Error (%) | 73.9 | 81.5 | 73.2 | 92.5 | 96.1 | 91.8 | 90.9 | ||

±15% | 24.1 | 27.8 | 29.1 | 26.7 | 27.5 | 28.0 | 26.7 | ||

±30% | 42.8 | 47.5 | 49.2 | 49.9 | 50.5 | 50.3 | 50.8 | ||

±50% | 60.0 | 69.3 | 66.2 | 69.3 | 73.5 | 69.8 | 72.2 | ||

_{24h-urine} | |||||||||

GFR estimate (ml/min) ±SD | 13.3 ± 8.4 | 39.7 ± 25.7 | 30.8 ± 25.7 | 34.4 ± 25.6 | 34.4 ± 33.3 | 32.1 ± 28.4 | 38.2 ± 30.6 | 33.7 ± 27.9 | |

Coefficient of variation (%) | 63.2 | 64.7 | 83.4 | 74.4 | 96.8 | 88.5 | 80.1 | 82.8 | |

Correlation | 0.58 | 0.50 | 0.54 | 0.55 | 0.58 | 0.56 | 0.54 | ||

Bias (ml/min) | 26.4 | 17.6 | 21.1 | 21.1 | 18.7 | 25.0 | 20.4 | ||

Precision (ml/min) | ±21.9 | ±22.7 | ±21.8 | ±29.5 | ±24.4 | ±26.9 | ±24.4 | ||

±15% | 1.4 | 16.7 | 5.6 | 12.5 | 13.1 | 11.1 | 8.3 | ||

±30% | 6.9 | 23.6 | 15.3 | 26.4 | 27.9 | 23.6 | 19.4 | ||

±50% | 11.1 | 36.1 | 23.6 | 36.1 | 39.3 | 30.6 | 31.9 | ||

_{24h-urine} | |||||||||

GFR estimate (ml/min) ±SD | 45.3 ± 8.6 | 85.6 ± 34.5 | 76.9 ± 37.3 | 80.4 ± 35.4 | 84.8 ± 40.6 | 73.8 ± 36.5 | 86.3 ± 29.5 | 76.0 ± 35.3 | |

Correlation | 0.44 | 0.46 | 0.46 | 0.30 | 0.46 | 0.35 | 0.33 | ||

Coefficient of variation (%) | 19.0 | 40.3 | 48.5 | 44.0 | 47.9 | 49.5 | 34.2 | 46.4 | |

Bias (ml/min) | 40.3 | 31.6 | 35.1 | 39.5 | 28.4 | 41.0 | 30.6 | ||

Precision (ml/min) | ±31.7 | ±34.3 | ±32.4 | ±38.9 | ±33.5 | ±27.6 | ±33.5 | ||

±15% | 5.9 | 11.8 | 14.7 | 7.4 | 15.5 | 7.4 | 11.8 | ||

±30% | 16.2 | 26.5 | 25.0 | 20.6 | 32.8 | 14.7 | 35.3 | ||

±50% | 27.9 | 44.1 | 39.7 | 38.2 | 53.4 | 23.5 | 51.5 | ||

_{24h-urine} 60–129.9 ml/min | |||||||||

97.8 ± 20.5 | 134.2 ± 44.5 | 121.6 ± 43.8 | 126.6 ± 41.9 | 120.8 ± 37.6 | 107.7 ± 35.5 | 114.2 ± 26.4 | 111.7 ± 36.2 | ||

Coefficient of variation (%) | 21.0 | 33.2 | 36.0 | 33.1 | 31.1 | 33.0 | 23.1 | 32.4 | |

Correlation | 0.37 | 0.38 | 0.40 | 0.12 | 0.18 | 0.32 | 0.23 | ||

Bias (ml/min) | 36.4 | 23.8 | 28.8 | 23.0 | 9.6 | 16.4 | 13.9 | ||

Precision (ml/min) | ±41.6 | ±40.6 | ±38.7 | ±40.7 | ±37.6 | ±20.7 | ±37.2 | ||

±15% | 24.4 | 27.6 | 29.5 | 27.6 | 32.6 | 42.3 | 28.8 | ||

±30% | 44.2 | 46.2 | 49.4 | 53.8 | 54.5 | 61.5 | 55.8 | ||

±50% | 66.7 | 71.8 | 69.2 | 72.4 | 73.1 | 84.0 | 78.8 | ||

_{24h-urine} | |||||||||

GFR estimate (ml/min) ±SD | 173.3 ± 41.6 | 186.6 ± 44.2 | 163.3 ± 37.6 | 172.6 ± 34.3 | 144.1 ± 32.3 | 130.3 ± 30.5 | 129.4 ± 16.3 | 138.4 ± 31.3 | |

Coefficient of variation (%) | 24.0 | 23.7 | 23.0 | 19.9 | 22.4 | 23.4 | 12.6 | 22.6 | |

Correlation | 0.29 | 0.20 | 0.28 | 0.22 | 0.14 | 0.14 | 0.20 | ||

Bias (ml/min) | 13.3 | −10.0 | −0.7 | −29.2 | −43.7 | −44.0 | −34.9 | ||

Precision (ml/min) | ±51.3 | ±50.0 | ±45.8 | ±46.7 | ±48.2 | ±42.5 | ±46.8 | ||

±15% | 42.0 | 40.1 | 45.9 | 40.8 | 33.1 | 30.6 | 39.5 | ||

±30% | 69.4 | 68.8 | 75.2 | 69.4 | 57.4 | 66.9 | 66.9 | ||

±50% | 89.8 | 93.0 | 94.3 | 94.9 | 88.5 | 93.6 | 93.0 |

CKD-EPI: chronic kidney disease epidemiology collaboration; CG: Cockroft-Gault; CrCl_{24h-urine}: creatinine clearance measured by 24-hour urine collection; GFR: glomerular filtration rate; MDRD: modification of diet in renal disease study; SD: standard deviation. CG_{actual-wt}: the CG equation was calculated using actual body weight. CG_{ideal-wt}: the CG equation was calculated using ideal body weight. CG_{adjusted-wt}: the CG equation was calculated using adjusted body weight. MDRD-4: the MDRD equation was calculated using four variables. MDRD-6: the MDRD equation was calculated using six variables.

The accuracy values for the different equations were generally modest. When using the first 237 urine measurements, accuracy within ±10% of CrCl_{24h-urine} ranged between 12.7% (CG_{actual-wt} equation) and 30.0% (CG_{adjusted-wt} equation). Accuracy within ±30% ranged between 47.4% (MDRD-6 equation) and 51.2% (CG_{adjusted-wt} equation). Accuracy within ±57.4% ranged between 12.7% (Jelliffe equation) and 75.1% (MDRD-6 equation). The accuracy values were similar when all 453 measurements were used in calculation (Table

Bland–Altman plots are depicted in Figure _{actual-wt}, 103.0 and −92.9 for CG_{ideal-wt}, 101.4 and −78.4 ml/min for CG_{adjusted-wt}, 106.7 and −105.7 ml/min for MDRD-4, 95.2 and −116.2 ml/min for MDRD-6, 94.8 and −107.4 ml/min for CKD-EPI, and 95.2 and −108.1 ml/min for Jelliffe equations. Multiple points were outside the limits of agreement, which were wide for all equations.

Bland–Altman plot of the creatinine clearance measured by 24-hour urine collection (CrCl_{24h-urine}) versus the equations estimating glomerular filtration rate. (a) Cockroft–Gault equation using actual body weight; (b) Cockroft–Gault equation using ideal body weight; (c) Cockroft–Gault equation using adjusted body weight; (d) 4-variable modification of diet in renal disease equation; (e) 6-variable modification of diet in renal disease equation; (f) chronic kidney disease epidemiology collaboration equation; and (g) Jelliffe equation. The _{24h-urine} and the equation estimating glomerular filtration rate. The _{24h-urine} and the equation estimating glomerular filtration rate. The solid line represents the bias (mean difference obtained across the range of values), whereas the dashed lines are the limits of agreement (±1.96 × standard deviation).

Correlation, bias, precision, and accuracy for the different equations are reported in Table _{24h-urine} < 30, 30–59.9, 60–130, and >130 ml/min using the 453 measurements, which were considered to be independent observations. Bias was significant for all equations except for CG_{adjusted-wt} equation when CrCl_{24h-urine} > 130 ml/min.

The sensitivity of GFR equations to correctly classify CrCl_{24h-urine} <30 ml/min was 44.7% for CG_{actual-wt}, 71.1% for CG_{ideal-wt}, 57.9% for CG_{adjusted-wt}, 60.5% for MDRD-4, 64.5% for MDRD-6, 59.5% for CKD-EPI, and 60.5% for Jelliffe equation. The sensitivity to correctly classify CrCl_{24h-urine} 30–59.9 ml/min was 17.2% for CG_{actual-wt}, 30.0% for CG_{ideal-wt}, 31.0% for CG_{adjusted-wt}, 31.0% for MDRD-4, 39.1% for MDRD-6, 13.8% for CKD-EPI, and 34.5% for Jelliffe equation. The sensitivity to correctly classify CrCl_{24h-urine} 60–129.9 ml/min was 59.5% for CG_{actual-wt}, 60.8% for CG_{ideal-wt}, 63.3% for CG_{adjusted-wt}, 58.2% for MDRD-4, 59.7% for MDRD-6, 79.7% for CKD-EPI, and 63.3% for Jelliffe equation. The sensitivity to correctly classify CrCl_{24h-urine} ≥130 ml/min was 87.9% for CG_{actual-wt}, 70.3% for CG_{ideal-wt}, 79.1% for CG_{adjusted-wt}, 60.4% for MDRD-4, 49.4% for MDRD-6, 45.1% for CKD-EPI, and 53.3% for Jelliffe equation.

Table _{24h-urine}. The values of _{24h-urine} >130 ml/min.

Correlation between equations estimating glomerular filtration rate and creatinine clearance measured by 24-hour urine collection.

Spearman correlation ( | |||||||
---|---|---|---|---|---|---|---|

CG_{actual-wt} | CG_{ideal-wt} | CG_{adjusted-wt} | MDRD-4 | MDRD-6 | CKD-EPI | Jelliffe | |

Age ≥ 65 years ( | 0.70 | 0.62 | 0.67 | 0.60 | 0.61 | 0.67 | 0.61 |

Age < 65 years ( | 0.71 | 0.62 | 0.69 | 0.57 | 0.56 | 0.57 | 0.58 |

BMI ≥ 30 kg/m^{2} ( | 0.82 | 0.82 | 0.83 | 0.76 | 0.73 | 0.81 | 0.78 |

BMI < 30 kg/m^{2} ( | 0.73 | 0.73 | 0.73 | 0.61 | 0.64 | 0.63 | 0.64 |

Admission category: medical ( | 0.79 | 0.72 | 0.78 | 0.63 | 0.66 | 0.70 | 0.67 |

Admission category: surgical ( | 0.74 | 0.58 | 0.68 | 0.70 | 0.48 | 0.66 | 0.67 |

Admission category: nonoperative trauma ( | 0.54 | 0.37 | 0.46 | 0.34 | 0.30 | 0.22 | 0.32 |

APACHE II ≥ 20 ( | 0.77 | 0.74 | 0.77 | 0.68 | 0.67 | 0.71 | 0.71 |

APACHE II < 20 ( | 0.71 | 0.59 | 0.67 | 0.48 | 0.46 | 0.53 | 0.55 |

Diabetes ( | 0.78 | 0.70 | 0.77 | 0.65 | 0.62 | 0.67 | 0.68 |

No diabetes ( | 0.70 | 0.62 | 0.68 | 0.52 | 0.52 | 0.57 | 0.55 |

Sepsis admission ( | 0.76 | 0.74 | 0.77 | 0.62 | 0.64 | 0.69 | 0.65 |

No sepsis on admission ( | 0.74 | 0.66 | 0.72 | 0.59 | 0.57 | 0.61 | 0.62 |

Traumatic brain injury ( | 0.56 | 0.39 | 0.49 | 0.41 | 0.43 | 0.21 | 0.32 |

No traumatic brain injury ( | 0.78 | 0.72 | 0.77 | 0.64 | 0.63 | 0.70 | 0.67 |

SOFA renal > 0 (Cr ≥ 110 | 0.90 | 0.86 | 0.90 | 0.82 | 0.82 | 0.81 | 0.85 |

SOFA renal = 0 (Cr < 110 | 0.64 | 0.54 | 0.62 | 0.41 | 0.43 | 0.45 | 0.46 |

Acute kidney injury on admission ( | 0.79 | 0.68 | 0.75 | 0.68 | 0.63 | 0.68 | 0.70 |

No acute kidney injury on admission ( | 0.65 | 0.56 | 0.63 | 0.41 | 0.45 | 0.46 | 0.47 |

Baseline CrCl_{24h-urine} ≥ 130 ml/min ( | 0.38 | 0.26 | 0.36 | 0.27 | 0.17 | 0.22 | 0.23 |

Baseline CrCl_{24h-urine}< 130 ml/min ( | 0.70 | 0.68 | 0.70 | 0.62 | 0.62 | 0.71 | 0.64 |

APACHE: acute physiology and chronic health evaluation; BMI: body mass index; CKD-EPI: chronic kidney disease epidemiology collaboration; CG: Cockroft-Gault; CrCl_{24h-urine}: creatinine clearance measured by 24-hour urine collection; GFR: glomerular filtration rate; MDRD: modification of diet in renal disease study. CG_{actual-wt}: the CG equation was calculated using actual body weight. CG_{ideal-wt}: the CG equation was calculated using ideal body weight. CG_{adjusted-wt}: the CG equation was calculated using adjusted body weight. MDRD-4: the MDRD equation was calculated using four variables. MDRD-6: the MDRD equation was calculated using six variables.

In this study, we found that the commonly used equations to estimate GFR performed modestly against the measured urinary CrCl with high bias and accuracy within 30% present in approximately 50%. The equations with the highest sensitivity to correctly classify CrCl_{24h-urine} 30–59 and ≥130 ml/min, ranges where medication dose adjustment is frequently needed, were MDRD-6 and CG_{actual-wt}.

Measuring GFR cannot be done routinely. Measured urinary CrCl is more widely available, but it may overestimate GFR because of Cr filtration and secretion; the latter can be affected by medications known to compete with active tubular secretion of Cr [_{24h-urine} with measured GFR in the ICU are limited. One study found that urinary CrCl with short collection times (1–2 h) had the highest correlation with measured GFR using inulin clearance (^{2} for GFR <60ml/min, 24 mL/min/1.73 m^{2} for GFR 60–90ml/min, and 44 mL/min/1.73 m^{2} for GFR >90ml/min [_{24h-urine} was compared with GFR measured by the infusion clearance of chromium-ethylenediaminetetraacetic acid [_{24h-urine.} The study found that the different equations tended to overestimate the CrCl for low eGFR values and to underestimate the CrCl for normal and high values [_{actual-wt} corrected for body surface area had the lowest bias (-3.2 ml/min for indigenous and 8.2 ml/min for nonindigenous patients) [^{2}, and the 95% limits of agreement were −128.9 to 36.7 mL/min/1.73 m^{2} [^{2}, respectively) [^{2} [_{24h-urine}. All had significant bias, inadequate precision, high error, low accuracy, and wide agreement limits on the Bland-Altman plots. The correlations were moderate to strong nevertheless. Importantly, the sensitivity to correctly identify CrCl_{24h-urine} in the clinically important ranges (such as 30–59 and >130 ml/min) was low in general for all equations.

Studies on the performance of GFR-estimating equations in critically ill patients with AKI are scarce. One study evaluated 30 ICU patients with early AKI. GFR-estimating equations, CG, MDRD-4, and CKD-EPI equations, performed poorly when compared with measured GFR. The biases ranged from 7.4 ml/min for CG_{actual-wt} to 11.6 ml/min for MDRD-4 [_{24h-urine} were fair. Moreover, MDRD-6 had the highest sensitivity (39.1%) to correctly classify CrCl_{24h-urine} 30–59.9 ml/min. This was mostly due to overestimation of GFR.

Other studies evaluated GFR-estimating equations in patents with ARC. A study of 390 patients with ARC in a surgical ICU in Belgium showed fair correlation between measured and estimated clearances (Spearman _{24h-urine}, with CG displaying the smallest bias [_{24h-urine} and CG (_{24h-urine} ≥ 130 ml/min than lower ranges. CG_{adjusted-wt} had low bias (−0.7 ml/min), the highest accuracy± 30% (75.2%), and sensitivity to correctly classify CrCl_{24h-urine} ≥130 ml/min (79.1%). It should be noted that failure to correctly identify ARC may lead to subtherapeutic dosing of medications increasing the risk of treatment failure, emerging microbial resistance, prolonged ICU stay, and increased mortality [

GFR-estimating equations may not perform well in certain populations, such as the very elderly [_{24h-urine} and the different GFR-estimating equations was weak in patients with polytrauma, who commonly have ARC [

The findings of this study should be interpreted taking into consideration its strengths and limitations. The strength includes the prospective data collection, relatively large sample size, the study of seven GFR-estimating equations, and the evaluation of their performance using several methods. The limitations include being a single-center study and the use of CrCl_{24h-urine} instead of more accurate GFR measures (e.g., inulin, ^{125}I-sodium iothalamate clearance or cystatin C-based equations). Serum cystatin C-based equations have been found to outperform serum creatine-based equations in estimating GFR in critically ill patients [_{24h-urine} is less accurate when kidney function is not steady and dysfunction is evolving [

In conclusion, GFR-estimating equations that are commonly used in clinical practice had limited ability to properly estimate CrCl_{24h-urine} and likely true GFR. They had limited ability to correctly classify GFR into clinically relevant ranges that are usually needed to determine dosing of medications. The clinical significance of these findings needs to be studied further.

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

The PermiT trial was approved by the Institution Review Board of the Ministry of National Guard Health Affairs, Riyadh, Saudi Arabia.

Informed consent was obtained from enrolled patients/next of kin.

All authors declare no conflicts of interest.

Conception and design of the work was performed by HMD and YMA. Acquisition of data was done by AAA, ASA, AMA, and MS. Analysis and interpretation of data were carried out by HMD, HT, AAA, ASA, AMA, MS, EE, and YMA. Manuscript was drafted by HMD. Manuscript was revised by HMD, HT, AAA, ASA, AMA, MS, EE, and YMA. All authors read and approved the final manuscript.

The PermiT trial was funded by King Abdullah International Medical Research Center, Riyadh, Saudi Arabia.

^{2}