| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTER TO THE EDITOR
The letter states that although we conclude that "an anomalous tubule Ca reabsorption explains hypercalciuria," at least 2 of 10 idiopathic hypercalciuria calcium stone formers (IHSF) behave exactly like normal patients (N). Although Worcester et al. (2) repeatedly state that we made an inference about mean fractional calcium reabsorption (FRCa), Gambaro and Abaterusso (1) have interpreted our conclusions as being at the patient level. The goal of most statistical analyses is to generalize to a population of interest, typically by focusing on the comparison of a summary measure between groups. For many reasons (both scientific and statistical), one often chooses to compare the mean or average response between groups. In this case, conclusions should not be interpreted at the individual level. For example, we know that on average aspirin cures headaches. However, this does not guarantee that every person who has a headache and takes aspirin will be relieved of their symptoms. Worcester et al. (2) have presented inferences on the mean and concluded that, on average, FRCa does statistically differ between IHSF and N during both the fasting and fed states and also found that these groups differ with respect to average change between fasting and fed states. Nowhere did we claim such a difference would be prevalent within every individual, and our results should not be interpreted as such.
Gambaro and Abaterusso (1) state that although there is large variability, there seems to be "a real difference" between N and IHSF with respect to higher natriuretic response after breakfast. As they have correctly noted, one must take into account the variability when comparing such measurements. The statistical analysis presented in Worcester et al. (2) found that mean urine Ca differed significantly between IHSF and N during the fed but not the fasting state, while FRCa differed significantly between IHSF and N in both states. However, with respect to urine sodium and fractional sodium reabsorption, one can see from Fig. 4 that the mean response for IHSF during the fasting state (where the largest difference in the mean response between N and IHSF occurred) was well within the confidence interval for the estimate corresponding to N. Thus while one is guaranteed to show a statistical difference between two parameters with a large enough sample size, our data do not come close to allowing us to draw this conclusion.
The letter states that "for the same FRCa (IHSF and N) are apart of 
15 pg/ml; furthermore, the average excursion fed/fast in IHSF is much higher than in N (10 pg/ml vs. 4 pg/ml)." As such, Gambaro and Abaterusso (2) conclude that this suggests a different parathyroid hormone (PTH) sensitivity in the two groups and in this sense does not exclude a role of PTH in postmeal calciuria. They further speculate that a study on a larger number of subjects may not reach the same conclusion that serum PTH does not account for the lower calcium absorption of IHSF vs. N as was found in Worcester et al. (2). First, we note that the difference of 15 pg/ml in serum PTH is likely quoted from Fig. 7 but is not comparing IHSF to N in the same feeding state. During the same feeding state and controlling for FRCa, the maximum distance between mean serum PTH comparing IHSF to N was 8 pg/ml (during the fasting state). With only adjustments for age and gender, this difference was estimated to be 7.01 pg/ml (Table 4). Next, simply because one may find a significant difference in PTH between IHSF and N with a larger study, this would not necessarily imply that PTH plays a role in the postmeal calciuria. Worcester et al. (1) found that after adjustment for PTH, mean FRCa remained significantly different between IHSF and N in both feeding states. This result is consistent with our stated conclusion that "serum PTH (did) not account for the lower reabsorption in IHSF vs. N."
Finally, Gambaro and Abaterusso (1) note that the meal composition with respect to amino acids should be investigated as a possible mechanism for linking eating and reduction of calcium reabsorption. While we agree that further investigation of the role of amino acids in this mechanism is of interest, it was not the focus of our study and does not impact our results, since IHSF and N were both given a controlled diet.
FOOTNOTES
Address for reprint requests and other correspondence: F. Coe, Dept. of Medicine, Univ. of Chicago, 5841 S. Maryland Ave., Chicago, IL 60637 (e-mail: fcoe{at}chicago.edu)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
