ACCURACY OF Kt/V NATURAL LOGARITHM
FORMULA IN CALCULATING SINGLE POOL
Kt/V IN CHILDREN
O. Marsenic, G. Bigovic, A. Peco-Antic, O. Jovanovic
University Children's Hospital, Tirsova 10, 11000 Belgrade, Yugoslavia
INTRODUCTION AND OBJECTIVE
PAT I E N T S A N D M E T H O D S
RESULTS
2.2
Complex variable volume single pool (VVSP) urea
kinetic model (UKM) remains the recommended
method for evaluation of hemodialysis (HD) in children with higher target doses of delivered Kt/V.
Daugirdas’ natural logarithm formula (D) has been
accepted as its simple alternative for use in adults.
It has been found to be sufficiently accurate in clinical settings in adults. In order to enable a quicker
and more simple bedside evaluation of HD therapy
in children, we tested the accuracy of D when used
in children on chronic HD.
Patients: 15 stable patients (M:6,F:9) aged
14.5±3.28 years, BW: 31.40±7.94 kg, renal urea
clearance absent in
10/15 patients and from 0.292
1.59 ml/min/1.73m in remaining 5/15.
No of HD sessions: 43 (3 consecutive HD sessions in 13 patients and 2 consecutive HD sessions
in 2 patients due to a problem with AV fistula).
HD session characteristics: standard, bicarbonate,
duration 3.81±0.44 hours, single pool effective urea
clearance 4.84±1.25 ml/min/kg, UF rate 6.2±2.5%.
Urea kinetic modelling: Variable Volume Single
Pool UKM with postdialysis urea taken before and at
the end of HD. Dialyser clearances used were blood
side in vivo values for each HD session. Blood-water
flow was used instead of full blood flow.
Daugirdas’ natural logarithm formula:
Kt/V = - ln ( R - 0.008 x t) + ( 4 - 3.5 x R ) x UF/BW
Results are presented on Tables 1 and 2 and
Figures 1-4.
1.8
R- post-HD urea (C1) / pre-HD urea (C01), UF- ultrafiltration volume (l),
(i.e. DWt (kg) after HD), BW- body weight after HD (kg), ln-natural logarithm, t-HD session duration (hours)
Statistics: the results were compared using
Student’s t test for paired samples, Pearson correlation and linear regression analysis.
2
p>0.05
1.6
1.4
Kt/V
VVSP resulted in Kt/V of 1.68±0.42, while Kt/V calculated by D was 1.70±0.36. Generally, the results
were not significantly different (+1.19%, p>0.05)
and were highly correlated (r=0.916). Absolute
error of D was 7.34±6.81% and ranged from 0.49 to
28.33%, median 5.2%. In the range of inadequate
Kt/V (VVSP<1.2; n=6), D significantly overestimated VVSP by 13.4% (1.10±0.22 vs. 0.97±0.22;
p<0.001, r=0.978). In the range of adequate Kt/V
(VVSP=1.2-1.6; n=13), the difference was smaller
(+3.35%; 1.54±0.13 vs. 1.49±0.09), but also significant (p<0.05, r=0.800). In the range of more than
adequate Kt/V (VVSP>1.6; n=24) the difference
between D and VVSP was not significant
(1.94±0.22 vs. 1.96±0.26; -1.02%, p>0.05), but the
correlation coefficient was lowest (r=0.651).
1.2
1.68
1
1.7
0.8
0.6
0.4
0.2
0
VVSP
Natural log formula
VVSP
Mean Kt/V ± SD
Comparisons
P value (t-test)
R (Pearson)
Difference between the means
Absolute percent difference on
individual session basis
Minimum
Maximum
25th percentile
50th percentile
75th percentile
Natural Log
Formula
1.70±0.36
1.68±0.42
>0.05
0.916
+1.19%
7.34±6.81%
0.49%
28.33%
2.15%
5.2%
9.44%
30
2.7
25
2.5
20
2.3
15
2.1
10
1.9
5
1.4
1.7
1.2
2.2
1.8
1.5
-5
-10
-15
-20
-25
0.4
0.6
0.8
1
1.2
1.4
1.6 1.8
Kt/V
2
2.2
2.4
2.6
0.8
1.1
0.6
0.9
0.4
0.7
0.2
0.97
1.94
1.54
1.1
p<0.001
p<0.05
p>0.05
0
0.5
2.8
Figure 2: Error of Natural Logarithm Formula in calculating Kt/V on individual session basis
and in relation to dialysis dose.
1
1.3
0.5
-30
1.49
1.6
Kt/V
0
1.96
VVSP
Natural log formula
2
y = 1.0555x - 0.1206
2
R = 0.8399, p<0.0001
R = 0.916
Kt/V-VVSP
% error
Figure 1 and Table 1: Comparison of Kt/V values calculated by VVSP
and Natural Logarithm formula
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
2.3
2.5
2.7
<1.2
Kt/V-natural log formula
Figure 3: Correlation between Kt/V values calculated by VVSP and Natural Logarithm formula. The red line is the line of identity and the black line is the regression line
1.2-1.6
Kt/V range
>1.6
Figure 4: Comparison of Kt/V values calculated by VVSP and Natural Logarithm Formula in three different ranges of Kt/V values representing inadequate dialysis (<1.2), adequate dialysis (1.2-1.6) and
more than adequate dialysis (>1.6).
DISCUSSION:
Kt/V
range
< 1.2
(n=6)
1.2-1.6
(n=13)
> 1.6
(n=24)
VVSP
Natural
log
formula
% difference
between the
means
P value
(t-test)
R
(Pearson)
0.97±0.22
1.10±0.22
+13.4%
<0.001
0.978
1.49±0.09
1.54±0.13
+3.35%
<0.05
0.800
I
I
I
I
1.96±0.26
1.94±0.22
-1.02%
>0.05
0.651
I
Table 2: Comparison of Kt/V values calculated by VVSP and Natural Logarithm Formula in three different ranges of Kt/V values representing inadequate dialysis (<1.2), adequate dialysis (1.2-1.6) and more than adequate dialysis (>1.6).
Natural logarithm formula is recommended as a simple alternative
method for calculating Kt/V in adults
(DOQI 1997).
Its accuracy has not been investigated in children.
According to its author, the natural
logarithm formula should be accurate
(error within 2%) in the Kt/V range of
0.7-2.0 (Daugirdas JT, J Am Soc
Nephrol 1993;34:1205-13).
Our results show that the Natural
Logarithm Formula significantly overestimates Kt/V in the clinically most
sensitive range of Kt/V which denotes
inadequate HD, while very high Kt/V
(>2.1) values are underestimated.
The formula is sufficiently accurate
with the clinically acceptable error in
the range of Kt/V of 1.4-2.0.
CONCLUSION:
If the possible overestimation of
low values of Kt/V (which are
clinically most important) is
always thought of, then the natural logarithm formula can be
used as a simple bedside technique for Kt/V determination
instead of VVSP in children.