Home About us Current issue Back issues Submission Instructions Advertise Contact Login   

Search Article 
  
Advanced search 
 
Saudi Journal of Kidney Diseases and Transplantation
Users online: 118 Home Bookmark this page Print this page Email this page Small font sizeDefault font size Increase font size 
 

Table of Contents   
LETTER TO THE EDITOR  
Year : 2016  |  Volume : 27  |  Issue : 2  |  Page : 425-427
The need to automate urinary creatinine ratios adjusted for the predicted daily creatinine excretion


King Fahad Specialist Hospital, P. O. Box 15215, Dammam - 31444, Kingdom of Saudi Arabia

Click here for correspondence address and email

Date of Web Publication11-Mar-2016
 

How to cite this article:
Abutaleb N. The need to automate urinary creatinine ratios adjusted for the predicted daily creatinine excretion. Saudi J Kidney Dis Transpl 2016;27:425-7

How to cite this URL:
Abutaleb N. The need to automate urinary creatinine ratios adjusted for the predicted daily creatinine excretion. Saudi J Kidney Dis Transpl [serial online] 2016 [cited 2020 Jan 22];27:425-7. Available from: http://www.sjkdt.org/text.asp?2016/27/2/425/178595
To the Editor,

The frequently occurring troublesome collection errors on carrying out the 24 h urine collection based tests were behind the widespread use of random urinary substance/creatinine ratios to estimate daily urinary excretion of the measured substances. Values concluded from these ratios are obviously affected the values of the denominator of these ratios, that is the urinary creatinine (UCr) excretion rates. As usually interpreted, the absolute value of the substance/creatinine ratio (using same units of mg or g) is taken to equal the daily excretion of the tested substance in grams. Such interpretation is based on the assumptions that the daily UCr excretion is 1 g and that the excretion rates of both the urine substance and creatinine are constant throughout the whole day. In the case of using the international units (IU), the assumed value for normal UCr excretion, the denominator, is 10 mmol (0.884 g), and so the ratio will need to be multiplied by 10 to estimate the daily substance excretion. Though such assumption of a fixed value for UCr for all individuals regardless of their differences in muscle mass and actual UCr has made the interpretation of these UCr ratios easy, it has done so at the expense of significant loss in accuracy.

As the factors of age, gender, weight, and ethnicity correlate with creatinine values in serum and urine to the same degree, our choice of fixed UCr value for all individuals is like dropping all available serum creatinine (SCr) based estimated glomerular filtration rate (eGFR) equations, which takes into account all four above factors, and concluding eGFR values directly from SCr values without considering these four factors.

The reported ranges for UCr values for healthy adults are wide. Bittles et al, for example, reported ranges of 1.15-2. 86 g/day and 0.82-1.76 g/day, for the adult males and females, respectively. [1] It is commonly accepted that expected daily UCr is 15-25 mg/kg for males and 10-20 mg/kg for females. [2] This wide range in UCr, the denominator in the urinary ratios, points to the degree of errors that are potentially made on interpreting urinary substance/creatinine ratios of many individuals. For an individual with a UCr value of 1.8 g/day, UCr ratios would underestimate the daily excretion by 43%. For such individuals, high rate of false normal results are to be expected in the work up of clinical problems such as stone formation, proteinuria in kidney donors, albuminuria and proteinuria evaluations, work up of urinary excretions of electrolytes such as K, Mg, and PO [4] .

Another reason for the deviations of the urinary ratios, from equating the actual daily urinary excretion, is the changing rates of excretion of the substances (e.g., protein) and creatinine during the 24 h period. In this letter, the concern is on discussing the problem related to the variable UCr values.

Introducing an adjustment or correction factor (CF), to the values of urinary substance/ creatinine ratios is suggested. This CF is simply the predicted or actual UCr value itself. In the case of using IU system, the CF will be 0.1 of the measured or the predicted UCr value as the denominator here is originally 10 mmol.

A. Adjusting Urinary Creatinine Ratios for the Predicted Urinary Creatinine Value

Many reasonable UCr prediction equations [2],[3],[4] are already available in the literature. Some of these equations, as the Chronic Kidney Disease Epidemiology Collaboration (CKD-Epi) formula and the Cockcroft and Gault (CG) equation, were developed using the same data that were utilized to develop the parent eGFR or Cr clearance (Cl) prediction equations. The errors in such UCr equations would be similar to that of their parent eGFR equations. For example, the percentage of estimated GFR within 30% of measured GFR was reported by Levey et al [5] to be 84.1% for the CKD-Epi equation. The comparable figure for UCr prediction by two CKD-Epi equations, named D and E, were reported as 79% and 81%, respectively. [3]

A clarifying example (Actual) for using urinary creatinine as the c orrection factor

A 39-year-old male treated as nephrotic patient, weighing 89 kg with SCr of 70 umol/L (0.8 mg%) was reassured that his latest urinary protein creatinine ratio (uPCR) has decreased to 0.65 g/g. The patient rechecked his proteinuria in a private laboratory and returned to the clinic with following data:

  • Twenty-four hour urine data volume: 2300 mL, daily UCr: 1750 mg, and daily proteinuria: 1150 mg.


It was then explained to the patient that the significant deviation in our estimation of daily proteinuria happened in parallel for the distance of his UCr from the reference value of UCr, that is, 1.0 g/day.

I could have improved the accuracy of our uPCR if I adjusted its value for his predicted UCr. The value of predicted UCr can be calculated by any one of the equations available in the literature. For keeping the article short, I will mention here the example of using one of the equations to predict UCr value.

Utilizing a chronic kidney disease-Epi data derived equation

The CKD epidemiology collaboration group derived two equations to estimate UCr. The two UCr equations have been already validated on large number of subjects. [3] They are awaiting for their incorporation into our practice. I do very much suggest incorporating one of these CKD-Epi equations into the hospitals automated laboratory reporting systems; exactly as we did, earlier with eGFR equations. For discussing above example, I have chosen, the simpler of the two reported equations: [3]

Expected UCr (mg/day) = 879.89 + 12.51 × weight (kg) − 6.19 × age + (34.51 if black) − (379.42 if female).

Applying this equation to the real example above, an estimated UCr (i.e., CF) value of 1.74 g was obtained. This is surprisingly very adjacent to the measured UCr value by the private lab. In fact, the same UCr value was obtained on using the alternative CKD-Epi equation.

The uPCR value of 0.65 could have been adjusted by multiplying its value by the CF that is the above predicted UCr; so:

Adjusted uPCR value = 1.74 × 0.65 = 1.131 g/day which very near to the measured UCr value of 1150 mg.

B. Adjusting urinary creatinine ratios when measured urinary c reatinine is available

It is still helpful to carry out occasional 24 h urine collection to directly measure UCr (and Cr Cl). Such measured UCr would serve for reasonable period during the follow-up of the individual patient not only to help adjusting UCr ratios but also as a prognostic factor in monitor muscle mass and as an aid to improve the accuracy of interpreting eGFR values. The latter is mainly on reference to the eGFR values for the extremely wasted patients.

In the example above, the measured UCr value of 1.75 will be utilized to adjust future uPCR and uACR values for this same patient during his follow-up over coming few years.

C. Incorporating timed urine samples into practice as an alternative approach to measure urinary creatinine based ratios

For a number of our significantly wasted patients, the adjustment of UCr ratios by using their equation based predicted UCr value is still not sufficiently accurate. Such scenario is similar to the failure of eGFR to reflect true renal function in such wasted patients. It is often essential for such patients, to directly measure Cr Cl and daily proteinuria (or other substance) from a 24-h urine collection sample. However, as the 24 h urine collection is not practical for many patients including children, estimation of Cr Cl and daily substance excretion can be made for such patients using a simple approach to timed urine sample.

  1. We need our laboratory technicians to respond to the requests of urine substance concentrations or creatinine ratios by giving the patients larger containers (about 0.5 L) and asking them to collect fresh and full urine samples into these containers accompanied writing the time of the last earlier voiding. The accuracy of the time interval between giving the test sample and the earlier micturition is likely to improve with time. This approach will maintain the simplicity of spot urine samples while achieving higher accuracy, in addition to allow Cr Cl calculation, especially for the wasted individuals and the young, but continent children. It could prove itself as a practical alternative to measuring UCr based ratios.


The need to automate the adjustments for the urinary creatinine ratios

As stated above, the logic of adjusting reported urinary substance/creatinine ratios for the predicted UCr values is the same logic for introducing eGFR after SCr values in the lab reports. As I discussed above, such suggested adjustment is also intended for the same factors of age, gender, weight, and ethnicity and were often derived from the same data as is the case on using CKD-Epi and CG equations. Automating the reports of the suggested adjusted urinary substance/creatinine ratios is likely to be of great help in many clinical scenarios as those related to the work up of kidney donors, recurrent stone formers, nephrotic syndrome, and CKD patients.

In summary, there is a very strong basis and justifying logic for adjusting urinary ratios for the common anthropometric characters of age, weight, gender, and race in order to improve the accuracy of the estimations made for daily substances excretions. Automating the adjustment of UCr ratios into hospitals reporting systems seems very easy, practical and needy step. Adopting one of the two CKD-Epi equations to predict UCr seems most ideal for carrying out this automated adjustment.

Conflict of interest: None declared.

 
   References Top

1.
Bittles AH, Bell JF, Neill DW. Simple automated method for urinary creatinine estimation. J Clin Pathol 1965;18:377-8.  Back to cited text no. 1
[PUBMED]    
2.
Ix JH, Wassel CL, Stevens LA, et al. Equations to estimate creatinine excretion rate: The CKD epidemiology collaboration. Clin J Am Soc Nephrol 2011;6:184-91.  Back to cited text no. 2
    
3.
Abutaleb N. Narrowing the margins of errors intrinsic to the estimations made from uACR and uPCR by introducing simple correction factors. Saudi J Kidney Dis Transpl 2013;24:805-6.  Back to cited text no. 3
[PUBMED]  Medknow Journal  
4.
De Keyzer W, Huybrechts I, Dekkers AL, et al. Predicting urinary creatinine excretion and its usefulness to identify incomplete 24 h urine collections. Br J Nutr 2012;108:1118-25.  Back to cited text no. 4
    
5.
Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604-12.  Back to cited text no. 5
    

Top
Correspondence Address:
Dr. Nasrulla Abutaleb
King Fahad Specialist Hospital, P. O. Box 15215, Dammam - 31444
Kingdom of Saudi Arabia
Login to access the Email id


DOI: 10.4103/1319-2442.178595

PMID: 26997407

Rights and Permissions




 

Top
   
 
 
    Similar in PUBMED
    Search Pubmed for
    Search in Google Scholar for
    Email Alert *
    Add to My List *
* Registration required (free)  
 


 
    References
 

 Article Access Statistics
    Viewed1054    
    Printed9    
    Emailed0    
    PDF Downloaded144    
    Comments [Add]    

Recommend this journal