Portal de Periódicos da CAPES

Response to ‘Quantitative analysis of the dysnatremias’

2006; Elsevier BV; Volume: 70; Issue: 7 Linguagem: Inglês

10.1038/sj.ki.5001716

1523-1755

Troels Ring,

Thermoregulation and physiological responses

I wrote a letter to ask Kurtz and Nguyen1.Kurtz I. Nguyen M.K. Evolving concepts in the quantitative analysis of the determinants of the plasma water sodium concentration and the pathophysiology and treatment of the dysnatremias.Kidney Int. 2005; 68: 1982-1993Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar to specify how accurate their assessment of the equation [Na+]pw=1.11(Nae+Ke)/TBW-25.6 and its proposed augmentations could possibly be: intervals of confidence for slope and intercept. If these are undefined, I would assume that the many equations could not be distinguished in clinical practice, including the one with intercept=0 and slope=1. The authors do not seem to know an answer to that question. Boling and Lipkind provided an analysis of the intercept measurement in the Edelman paper similar to what I did, based on the original values and wrote: ‘…the degree of certainty regarding the magnitude of this constant term is not great; the origin of regression lies within the 99% confidence limits for the regression line, and the 95% confidence limits (at X=0) are −5.2 and −45.8’.2.Boling E.A. Lipkind J.B. Body composition and serum electrolyte concentrations.J Appl Physiol. 1963; 18: 943-949PubMed Google Scholar I had −5.2 and −46.0 with present-day software. As Edelman and co-workers described, this equation was obtained by careful assessment of a great number of patients by isotopic dilutions. These dilutions are not simple, requiring 40 h for K and 24 h of observation for Na (probably explaining that they were not frequently performed in recent times) and the problem of knowing when equilibrium is reached certainly is not straightforward and necessarily mixes with subtle definitions of osmotically active and inactive moieties. The suggestion that osmotic balance is rapidly attained is not contested, but steady state must be of interest when assessing the exchangeability of electrolytes during 40 h of equilibrium. Recent experimental studies arguing about the status of osmotic inactivation and also observations from sports medicine show that quite large amounts of Na and K can behave unpredictably.3.Heer M. Baisch F. Kropp J. et al.High dietary sodium chloride consumption may not induce body fluid retention in humans.Am J Physiol. 2000; 278: F585-F595PubMed Google Scholar, 4.Seeliger E. Ladwig M. Reinhardt H.W. Are large amounts of sodium stored in an osmotically inactive form during sodium retention? Balance studies in freely moving dogs.Am J Physiol. 2006; 290: R1429-R1435Google Scholar, 5.Noakes T.D. Sharwood K. Speedy D. et al.Three independent biological mechanisms cause exercise-associated hyponatremia: evidence from 2135 weighed competitive athletic performances.Proc Natl Acad Sci USA. 2005; 102: 18550-18555Crossref PubMed Scopus (292) Google Scholar Hence, even though the authors cite a number of situations wherein they have difficulties in understanding [Na], I cannot see how they conclude that a and b in [Na+]pw=a(Nae+Ke)/TBW-b are primarily to fault when (Nae+Ke)/TBW are not measured in any of these studies. I certainly agree with the authors that the work of Edelman et al. was second to none, and really wrote my letter to ask for more of that kind instead of math work. Edelman et al. carefully specified the uncertainty of the estimates, as I wrote, and even gave the individual data (from which, by the way, it can be verified that minor miscalculations occurred in patient numbers 25 and 37 (heart disease) – but, rest assured, removing them changes nothing). However, a complicated thing such as measuring deuterium enrichment has changed during the years, and the falling drop method as used by Edelman et al. has since been surpassed.6.Ellis K.J. Wong W.W. Human hydrometry: comparison of multifrequency bioelectrical impedance with 2H2O and bromine dilution.J Appl Physiol. 1998; 85: 1056-1062PubMed Google Scholar It is true that Edelman et al. made a correction for imprecision in measurements to augment the correlation coefficient, but it is wrong as stated by Kurtz et al. in their original paper that this bears directly over to the regression – which was carried out simply by least squares. The authors are right, however, that uncertainty attenuates the regression, but I doubt it is easy to say by how much. One of the most important determinants will be the spread in independent values – which is quite large in the original Edelman data, as mentioned by Boling et al.1.Kurtz I. Nguyen M.K. Evolving concepts in the quantitative analysis of the determinants of the plasma water sodium concentration and the pathophysiology and treatment of the dysnatremias.Kidney Int. 2005; 68: 1982-1993Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar The Passing–Bablok regression, now introduced by Kurtz et al. was described in 1983 and not mentioned by Edelman in 1958 (which was the question). However, Kurtz et al. may well be right that the true slope in [Na+]pw=a(Nae+Ke)/TBW-b often is more than 1 and the intercept often different from 0. The issue was if 11 new equations were separable, however. It is true that to all this must be added problems of individual patient frailty and autocorrelations between repeated measurements in each patient and more. I did not miss the point, that Edelman made a brilliant cross-sectional study. This only further strengthens the demand for data rather than speculation when trying to untangle this difficult subject. Measurements, at least of whole-body balances and exchangeable cations, are needed to understand when and why and how the famous Edelman equation needs augmentation. Until then, it might as well be kept simple, yes, that was my point. Like others, I have found good use of the simple form,7.Mallié J-P. Ait-Djaffer Z. Larouche F. et al.Variations in plasma sodium concentration in post-operative patients depend on an electrolyte-free water balance, part of a tonicity balance.Clin Nephrol. 1998; 49: 287-292PubMed Google Scholar but I am prepared that it will not always suffice.8.Guglielminotti J. Tao S. Maury E. et al.Hyponatremia after hip arthroplasty may be related to a translocational rather than to a dilutional mechanism.Crit Care Med. 2003; 31: 442-448Crossref PubMed Scopus (16) Google Scholar