The use of Fuller's earth to reduce acetoacetate interference in the measurement of serum creatinine by the Beckman Astra-4 analyser

Several authors have recently re-emphasized the positive interference of acetoacetate in the measurement of creatinine using the Jaff6 reaction [1-3]. On theoretical grounds, the measurement of creatinine in the presence of acetoacetate should be possible using a kinetic method [4]. In practice, however, measurements by some instruments employing a kinetic Jaff6 method suffer from marked interference by acetoacetate. Haekel described a procedure for the measurement of creatinine in the presence of acetoacetate, utilizing the adsorption of creatinine onto Fuller’s earth, and subsequent simultaneous elution and colour development with alkaline picrate [-5 and 6]. One of the variations of his procedure is that it omits a protein precipitation step [5]. In the present work, the procedure omitting a protein precipitation step has been adapted by eluting creatinine from the Fuller’s earth and measuring the creatinine in the eluate using a Beckman Astra-4 analyser (Beckman RIIC Ltd, High Wycombe, Buckinghamshire HP12 3NR, UK).


Introduction
Several authors have recently re-emphasized the positive interference of acetoacetate in the measurement of creatinine using the Jaff6 reaction [1][2][3]. On theoretical grounds, the measurement of creatinine in the presence of acetoacetate should be possible using a kinetic method [4]. In practice, however, measurements by some instruments employing a kinetic Jaff6 method suffer from marked interference by acetoacetate.
Haekel described a procedure for the measurement of creatinine in the presence of acetoacetate, utilizing the adsorption of creatinine onto Fuller's earth, and subsequent simultaneous elution and colour development with alkaline picrate [-5 and 6]. One of the variations of his procedure is that it omits a protein precipitation step [5]. In the present work, the procedure omitting a protein precipitation step has been adapted by eluting creatinine from the Fuller's earth and measuring the creatinine in the eluate using a Beckman Astra-4 analyser (Beckman RIIC Ltd, High Wycombe, Buckinghamshire HP12 3NR, UK).

Materials and methods
Apparatus Serum creatinine measurements were performed using a Beckman Astra-4 analyser. The instrument was equipped with modules for the measurement of sodium/potassium, bicarbonate, urea, and creatinine, but during this investigation only the sodium/potassium and creatinine modules were programmed for use. The instrument was calibrated using Beckman Astra standard solutions.

Reagents
Acetoacetate was obtained as the lithium salt (Sigma London Chemical Company Ltd, Poole, Dorset BH17 7NH, UK). All other reagents were obtained from B.D.H. Chemicals Ltd (Poole, Dorset BH 12 4NN, UK). The Fuller's earth reagent was prepared by adding 6 g of Fuller's earth to 500ml of deionized water, adding 10ml of 1.0ml/1 hydrochloric acid, and making the reagent volume up to 11 with deionized water. Whilst in use, the Fuller's earth was kept in suspension by continuous stirring with a magnetic stirrer. The eluting reagent consisted of 1.0tool/1 potassium hydroxide containing 100mol/1 sodium chloride.
Procedure 200 #1 of serum was added to 5.0ml of Fuller's earth reagent in a 10ml glass tube. The tube contents were mixed for min. by gentle inversion and the mixture was then centrifuged for min. at 3000 x g. Using a fine-tipped Pasteur pipette connected to a filter pump, as much as possible of the supernatant liquid was removed without disturbing the Fuller's earth pellet. 400/A of eluting reagent was added to the Fuller's earth pellet, and the Fuller's earth resuspended using a vortex mixer. The mixture was gently shaken for min. keeping the tube in a vertical position, then it was centrifuged for min. at 3000 x g and the sodium and creatinine concentrations in the resulting supernatant were measured on the Astra-4. The concentration of creatinine in the original serum sample was calculated as follows: Sodium levels were measured in the eluting reagent and final supernatant. This provides a correction factor to compensate for some liquid being retained by the Fuller's earth pellet after the first centrifugation step.

Stability of creatinine in the elutin9 reagent
Since creatinine is unstable under alkaline conditions [7], its stability in the eluting reagent was studied. No change in creatinine concentratiion was found after h incubation in the eluting reagent, but after 2h the creatinine had fallen to approximately 90o ofthe original value. All measurements were therefore performed within h of adding the eluting reagent.
Comparison of Fuller's earth procedure with the standard Astra procedure Creatinine values obtained using the procedure described were compared with values obtained using the standard Astra procedure. The comparison was performed on serum samples referred to the laboratory for electrolyte measurements. Serum samples from patients with ketoacidosis were excluded. The 69 samples used in the comparison consisted of 22 samples with creatinine in the range 50-140#mol/1 as measured by the standard Astra procedure, 27 samples in the range 140-350 #mol/1, and 20 samples with serum creatinine in excess of 350 gmol/1. Statistical analysis of this data, taking the standard Astra procedure as the x axis and the Fuller's earth procedure as the y axis yielded a correlation coefficient of 0.997, a slope of 0"938, and an intercept on the y axis of -10/mol/1.

Recovery
Creatinine result using Fuller's earth procedure Creatinine result using standard Astra procedure The mean recovery was 87 (S.D.-8).
procedure was determined at three levels of serum creatinine. The results of this study are presented in table 2. standard Astra procedure before and after addition of 4 mmol/l acetoacetate; l(b): Fuller's earth procedure before and after addition of 4 mmol/l aceoacetate; 2(a): standard Astra procedure before and after addition of 10mmol/1 acetoacetate; 2(b): Fuller's earth procedure before and after addition of 10 mmol/l acetoacemte.

Discussion
The Fuller's earth procedure produced somewhat lower values for creatinine than the standard Astra procedure. The lower values could have been due to incomplete recovery of creatinine either at the stage of adsorption of creatinine onto the Fuller's earth, or at the stage of alkaline elution from the Fuller's earth. Alternatively, the lower values could have been due to the removal of some non-creatinine chromogen from serum by the Fuller's earth procedure. The similarity between the observed recovery of endogenous and added creatinine (87o and 91) suggests that the lower values obtained using the Fuller's earth procedure were mainly due to loss of creatinine at some stage rather than being due to the removal of a positively interfering substance. However, the difference in mean recoveries of endogenous and added creatinine is statistically significant (p < 0.05), so more than one effect may be operating. Attempts to increase the recovery ofcreatinine by varying the mixing times of serum and Fuller's earth, and eluting reagent and Fuller's earth were unsuccessful. Varying potassium hydroxide concentration in the eluting reagent also failed to improve the recovery. Although quantitatively acetoacetate is not the major 'ketone body' in ketoacidosis, it is by far the most important compound in terms of interference in the Jaff6 reaction on the Astra-4. Serum acetoacetate levels of 4 mmol/1 are not uncommon in ketoacidosis, and it can be seen from figure that the Fuller's earth procedure virtually eliminates the positive interference of acetoacetate at this level. Serum acetoacetate levels of 10mmol/1 represent the extreme and figure shows that acetoacetate interference at this level is not abolished in all the serum samples studied, though the interference is considerably reduced.
The practical effect of acetoacetate interference in creatinine measurement on the Astra-4 is demonstrated in table 3, which shows the results obtained during the treatment of a patient admitted to hospital in diabetic ketoacidosis.
The rapidity with which the Fuller's earth procedure can be performed makes in suitable to emergency investigations. For convenience, the Fuller's earth reagents may be stored in 5 ml aliquots as suggested by Haeckel [5]. It is important to note that the eluate which is analysed on the Astra contains potassium at a concentration of 1.0 mol/1. This causes an increase in potassium in the following sample of l-2mmol/l due to carry-over. To combat this it is suggested that the Fuller's earth procedure samples are followed by a sample ofdeionized water on the Astra sample tray.