Reduced podocyte expression of α3β1 integrins and podocyte depletion in patients with primary focal segmental glomerulosclerosis and chronic PAN-treated rats

We have read the original article entitled “Reduced podocyte expression of α3β1 integrins and podocyte depletion in patients with primary focal segmental glomerulosclerosis and chronic PAN-treated rats” in The Journal of Laboratory and Clinical Medicine (Volume 147, Issue 2, February 2006). We would like to clarify the following points. Values were presented as mean ± standard error of the mean (SEM). The data were analyzed via unpaired t-test, one-way analysis of variance, simple linear regression, and correlation methods. In Table I, the sample size of patients with primary focal segmental glomerulesclerosis (FSGS) was 15 and the sample size of normal controls was 5. The mean ± SEM of “number of immuno-gold for β1” was presented as 0.12 ± 0.13. We can compute standard deviation (SD) of this variable by multiplying the SEM with the square root of the sample size. In this case, the mean ± SD will be equal to 0.12 ± 0.50. Thus, we can indicate that the SD (0.50) is greater than the mean value (0.12). Similar SD values can be computed for the other variables. These results show that data are not normally distributed. The normal distribution extends beyond two SDs on either side of the mean.1Altman D.G. Bland J.M. Statistics notes detecting skewness from summary information.BMJ. 1996; 312: 1200Crossref PubMed Scopus (316) Google Scholar In fact, the mean, the SD, and the SEM should be used only to describe approximately normal distributions. Therefore, we think that the use of a median value is more appropriate. Furthermore, we note that using the SD with the mean is more appropriate than using SEM when the data are normally distributed, because the SD indicates the extent of variability within a group of data associated with its mean value. The standard error does not quantify variability. This term is misleading because it has nothing to do with standards or errors. The SEM describes the accuracy or precision of the mean, in other words, how far away the sample mean is from the “true” mean.2Ettarh R. Common descriptive and analytical statistics in investigative studies.Radiography. 2004; 10: 299-302Abstract Full Text Full Text PDF Scopus (3) Google Scholar Many statistical tests rely on the assumption that analyzed data are derived from a population that has a normal (Gaussian) distribution. Regression, correlation, t-tests, and analysis of variance all depend on normally distributed data.3Livingston E.H. The mean and standard deviation what does it all mean?.J Surg Res. 2004; 119: 117-123Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar Nonparametric methods require no or very limited assumptions to be made about the format of the data, and they may, therefore, be preferable when the assumptions required for parametric methods are not valid.4Whitley E. Ball J. Statistics review 6 nonparametric methods.Critical Care. 2002; 6: 509-513Crossref PubMed Scopus (197) Google Scholar As a result, it is often more appropriate to use such statistical tools as nonparametric regression, Spearman correlation, Mann–Whitney U, and Kruskall–Wallis instead of simple linear regression, correlation, unpaired t-test, and analysis of variance. Finally, nonparametric tests are preferable to parametric tests. Reduced podocyte expression of α3β1 integrins and podocyte depletion in patients with primary focal segmental glomerulosclerosis and chronic PAN-treated ratsThe Journal of Laboratory and Clinical MedicineVol. 147Issue 2PreviewIntegrins attach cells to extracellular matrix (ECM) and mediate signals from ECM to cells or from cells to ECM. They regulate cell functions, including adhesion, migration, cell cycle regulation, and differentiation. Podocytes may detach from the glomerular basement membrane (GBM) and be excreted in the urine, and proteinuria is found in patients with primary focal segmental glomerulosclerosis (FSGS); both may be associated with loss of α3β1integrins. In this study, we have examined the podocyte number in patients with primary FSGS and normal controls, and the α3- and β1-integrin subunits expression of podocytes in patients with primary FSGS and chronic puromycin aminonucleoside (PAN)-treated rats by the morphometric, immunoperoxidase histochemical, and immunoelectron microscopic examination. Full-Text PDF