| Objective:This study attempts to investigate the changes of urine Raman spectroscopy of different bone mass conditions among patients with type 2 diabetes and analyze the relationship between differences and abnormal bone metabolism.Methods:The patients with type 2 diabetes mellitus were selected as the objects.We divided these patients into three groups on the basis of the results of dual-energy X-ray absorptiometry.The osteoporosis group?n=31?,the osteopenia group?n=30?and the normal bone mass group?n=30?.Then we collected the samples both urine and serum from all these objects.The diagnosis of the urine is supported by the technology of surface enhanced Raman spectroscopy?SERS?.At the same time,we also measured the biochemical markers of bone turnover markers and the level of serum lipid.Finally,we used principal component analysis and linear discriminant analysis besides single-factor analysis of variance,to compare both the differences and similarities between biochemical parameters and Raman spectra.Results:1.Characteristics of ObjectsWe found that age,BMI,biochemical markers of bone turnover,blood lipids and sex hormones levels of all the objects were conformed to a Gaussian distribution.The result of single-factor analysis of variance indicated no difference of both statistics in TP1NP,total vitamin D,osteocalcin,TG,HDL-C,HbA1c,fasting plasma C-peptide,PRL,E2,T,P,SHBG?P>0.05?.The age of the normal bone mass group,the osteopenia group and the osteoporosis group were?50.13±11.71?,?62.07±7.84?and?65.71±10.05?years old.Compared with the normal bone mass group,there was a statistically significant difference?P<0.01?.However,there was no difference of age between the osteopenia group and the osteoporosis group?P>0.05?.The BMI of the osteopenia group and the osteoporosis group were?22.85±3.15?kg/m2 and?23.10±3.70?kg/m2,there was no difference between this two groups?P>0.05?.While comparing to the normal bone mass group which BMI was?21.45±3.08?kg/m2,there was a statistically significant difference?P<0.01?.The level of CTX which is one of the most important bone turnover markers in osteoporosis were?0.33±0.12?ng/mL,?0.45±0.06?ng/mL and?0.55±0.25?ng/mL of the normal bone mass group,the osteopenia group and the osteoporosis group respectively.The latter two groups had statistical difference?P<0.05?to the normal bone mass group.By comparison,there was no difference between each other?P>0.05?.Lipid index TC of the normal bone mass group,the osteopenia group and the osteoporosis group were?4.65±0.83?mmol/L,?5.13±0.94?mmol/L and?5.18±1.17?mmol/L.The two groups behind showed statistically significant differences?P<0.05?,comparison was made with the normal group.When it comes to the osteopenia group and the osteoporosis group,no statistical difference was found?P>0.05?.Lipid index LDL-C,FSH and LH of the normal bone mass group were?2.66±0.54?mmol/L,?22.63±25.17?mIU/ml and?12.72±14.51?mIU/ml,the osteopenia group were?2.97±0.62?mmol/L,?40.25±36.22?mIU/ml and?19.09±17.3?mIU/ml,the osteoporosis group were?2.97±0.65?mmol/L,?40.11±30.27?mIU/ml and?20.03±15.84?mIU/ml.These three kinds of biochemical parameters showed the same result——the latter two groups had statistical difference?P<0.05?to the normal bone mass group when the osteopenia group and the osteoporosis group existed no statistical difference?P>0.05?.2.Comparison of Normalized Averaging Urine Raman SpectroscopyAfter normalizing and smoothing there average Raman spectrums,we compare them together.The results showed that the spectral peaks of the osteopenia group and the osteoporosis group had no significant misregistration comparing with the normal bone mass group.Taking the displacement at 1000 cm-1 as an example among the three groups,there was no significant difference.We totally found 10 characteristic peaks in the range of 200 cm-1-2000 cm-1 wavelength,locating at 253 cm-1,300 cm-1,451 cm-1,572 cm-1,785 cm-1,890 cm-1,1000 cm-1,1377 cm-1,1434 cm-1,1525 cm-1.Before 253 cm-1,the normalized average peak intensity of the normal bone mass group was obviously stronger than that of the osteopenia group and the osteoporosis group which the Raman lines were shifted downwards.In the range of 300 cm-11 to 1377 cm-1,the normal bone mass group and the osteoporosis group were stronger than the osteopenia group which the Raman line was moved downwards.There was no significant difference after 1377 cm-1 among the three groups.At253 cm-1,there was significantly different?P<0.01?of the peak intensity between the osteoporosis group and the normal bone mass group.When it comes to the osteopenia group and the osteoporosis group,the conclusion was the same.At 1377 cm-1 and 1525 cm-1,the osteopenia group and the osteoporosis group respectively compared with the normal bone mass group,the differences were significant?P<0.01?.The osteopenia group and the osteoporosis group also had significantly different?P<0.01?at 1377 cm-11 and 1525 cm-1.There was a difference between the normal bone mass group and the osteopenia group?P<0.05?at 1430 cm-1.3.Comparison of Relative Intensity of I1377We calculated the relative peak intensity?crest/trough?of 61 objects at the characteristic peak1377 cm-11 from the osteoporosis group and the normal bone mass group,and drew the scatter plot depending on the results.The scatter plot showed that distribution of these two groups had obvious difference by drawing a horizontal line at 1.4 as a dividing line.Most coordinates of the osteoporosis group were behind the dividing line while those of the normal bone mass group were above the dividing line.In the same way,we calculated the relative peak intensity?crest/trough?of60 objects at the characteristic peak 1377 cm-11 from the osteopenia group and the normal bone mass group,and drew the scatter plot depending on the results.The scatter plot showed that distribution of these two groups had obvious difference by drawing a horizontal line at 1.4 as a dividing line.Most coordinates of the osteopenia group were behind the dividing line while those of the normal bone mass group were above the dividing line.It was the opposite situation when we drew the the scatter plot of the osteopenia group and the osteoporosis group.It was difficult to distinguish these two abnormal bone mass groups.We used discriminant analysis to confirm this method.The results showed that 25 cases of osteoporosis patients were correctly diagnosed,6 cases were misdiagnosed;21 cases of normal patients were correctly diagnosed,9 cases were misdiagnosed as osteoporosis.The sensitivity and specificity were 80.6%and 70.0%,the accuracy rate was only 75.4%.Besides,23 cases of osteopenia patients were correctly diagnosed,7 cases were misdiagnosed;21 cases of normal patients were correctly diagnosed,9 cases were misdiagnosed as osteoporosis.The sensitivity and specificity were 76.7%and 70.0%,the accuracy rate was only 73.3%.4.PCA–LDA ResultsWe chose 1377 cm-1 Raman peak as the reference peak to normalize the other 9 characteristic peaks of 91 objects.Then calculated the PCA score of each case.KMO?Kaiser-Meyer-Olkin?measure of sampling adequacy was 0.853>0.6 and Bartlett's test showed P<0.05 which means data is suitable for factor analysis.Finally,we obtained two principal components which eigenvalue was greater than 1 and the accumulated contribution rate was 85.977%.On the basis of the PCA score after discriminant analyzing and examining,the sensitivity of diagnosis of osteoporosis was 67.7%,the specificity was 86.7%,the general accuracy was 77.0%;the sensitivity of the osteopenia group was 63.3%,the specificity was 93.3%,the general accuracy was 78.3%.Conclusion:1.There was a significant difference to urine Raman spectroscopy of the abnormal bone mass groups and the normal bone mass group in the change of intensity among the patients with type 2 diabetes mellitus.2.The biological molecular information of urine Raman spectroscopy matched the physiological and biochemical changes of disease among the osteoporosis group.3.Urine Raman spectroscopy can be used to diagnose the abnormal bone mass among the patients with type 2 diabetes mellitus. |
PDF Full Text Request