Rodby RA, Firanek CA
A re-evaluation of PET determined patient solute transport group classifications
Am Soc Nephrol
J Am Soc Nephrol (abstract) (Sep) 7:1433 1996

The authors first recall, that using the classification of PET according to Twardowski, and assuming a Gaussian patient distribution, we would expect that approx. 33% of patients should be low average (LA), 33% High average (HA), 17% high (H) and 17% Low (L).

To determine if their patients were similar to those that defined the PET groupings, they reviewed the 153 initial Standard PET tests done in their PD program. All patients were adults and PET tests were not done within 2 months of an episode of peritonitis. The mean 4-hour D/P creatinine ratio in their patients was 0.71 +/- 0.10 (median = 0.69, mode = 0.72). This compared to a mean of 0.65 +/- 0.15 as determined by Twardowski et al, and indicates that their patients had higher mean solute transport rates and tighter ranges within transport groups than previously reported.

Only 1.3% of their patients fell into the previously described L range, with 24.8% LA, 50.3% HA and 17.0% H. Using their data they would redefine the groups by the 4-hour D/P creatinine ratio of: L < 0.61, LA 0.61 to 0.71, HA 0.71 to 0.81, and HA > 0.81. Doing this they found 16.3% of patients were in the low transporter group L, 34% were LA, 30.1% were HA, and 19.6% were H (high transporters), a Gaussian distribution.

They conclude that the 4-hour D/P creatinine ratio ranges that presently determine PET transport groupings grossly underestimate low transporters in their patient population. This could lead to an inappropriate dialysis prescription in some patients. They suggest that these ranges need to be reevaluated.

Comment: This study emphazises the fact that peritoneal tests may vary in their distribution from one center to another, probably depending on patient selection. Duration of dialysis before the test, policy concerning the number of hypertonic solutions used, age of patients, previous peritonitis rates and methodology of testing may all affect the distribution of results, sometimes with unpredictable consequences. As a result, comparison from one center to another is often difficult. However the argument of real Gaussian distribution in the patients of the authors (not in those of Twardowski and all) deserves strong attention and should be considered to redefine normal ranges.

In our experience, the PET test should be used as a general guide to the prescription, rather than as a precise mean to predict adequacy of dialysis. In practice every one should define the distribution of PET tests in their own center, and should be critical of use of this test to classify patients in terms of permeability when the mean differs from the median. (Christian Verger MD, Pontoise, France)

To go back use the BACK button on your browser.
Otherwise click on the desired link to this article below:
Am Soc Nephrol
Basic peritoneal dialysis : PET testing