| 초록 |
Objectives : To evaluate the use of 1H NMR spectroscopy for diagnosis and prognosis of bacterial peritonitis associated with prolonged peritoneal dialysis (PD) therapy
Methods : NMR spectroscopy coupled with multivariate data analysis was applied to 24 PD effluent samples corresponding to 12 samples of noninfectious PD effluent and 12 samples of bacterial peritonitis Results and Discussion : As evident from Figure 1, a clear group separation and clustering of infected (bacterial peritonitis) and non-infected samples. The main metabolites responsible for the group separation are the cyclic fatty acids of microbial origin (peaks marked with asterisk in the loading plot) observed only in bacterial infected PD effluent. The other discriminatory metabolites are citric acid, lysine, creatinine, GPC and oxalacetate. The levels of citric acid and oxalacetate has been observed to be reduced, while the levels of creatinine, lysine and GPC has been elevated in the bacterial infected PD group as compared with the non-infected PD group (Fig. 1B).
Conclusions : The metabolic composition pofiles of PD effluent samples obtained from PD patients continuing normally on PD and those having bacterial peritonitis were found to be significatly different suggesting that 1H NMR of PD effluent has huge potential to predict the infections associated with PD in future clinical settings. We foresee that the identified metabolic signatures will provide rapid and non-invasive predictive measures along with the benefit of rapidity in early diagnosis of peritonitis and its recovery after the treatment. |