The reference standard based upon clinical diagnosis of heart failure (as described in Section 2.1).Percentage recovery ?? NT-proBNP concentration in spiked saliva NT-proBNP concentration in un-spiked saliva =amount of spiked NT-proBNP ?1002.5.2 Intra-and inter-assay coefficient of variation. To Title Loaded From File determine intra- and inter- assay variation, triplicates of saliva samples from 45 HF patients and 40 healthy controls were run in!Results 3.1 ParticipantsIn total 45 symptomatic HF patients (with left ventricular ejection fraction ,40 ) and 40 healthy volunteers (young, and old) were enrolled in the study. The group of HF patients consisted of 23 males and 22 females, with a median age of 73 years (ages from 53 to 88), body mass index (BMI) of 29.14, and systolic and diastolic blood pressure of 125 mm Hg and 71 mm Hg, respectively. The group of healthy controls consisted of 20 males and 20 females, with a median age of 56 years (ages from 40 to 71) and BMI of 25. Gender, age, BMI, blood pressure, heart rate, New York Heart Association (NYHA) classification of the 85 volunteers were summarised in Table 1.Figure 1. Comparison of the NT-proBNP immunoassay when compared with a commercially available diagnostic assay (Roche Diagnostics, USA). r2 = 0.78 and p,0.001. doi:10.1371/journal.pone.0048452.gRelevance of Salivary NT-ProBNP and Heart FailureFigure 2. The comparison of the plasma and salivary NT-proBNP concentrations for HF patients (n = 45). (A), NT-proBNP levels in saliva of HF patients (n = 45) and healthy participants (n = 40). (B), The correlation of NT-proBNP concentrations in plasma and saliva of HF patients measured by our NT-proBNP immunoassay. doi:10.1371/journal.pone.0048452.g3.2 Assay Performance for the NT-proBNP AlphaLISA(R) ImmunoassayThe performance characteristics of the NT-proBNP immunoassay is summarised in Table 2. Intra- and inter-assay coefficients of variation (CVs) for the NT-proBNP immunoassays were below 10 . The LOD for the salivary immunoassay was approximately 16 pg/mL.The clinical sensitivity and specificity of the salivary Title Loaded From File NTproBNP immunoassay was 82.2 and 100 respectively, with an overall diagnostic accuracy of 90.6 . The positive predictive value for the salivary immunoassay was 100 , and negative predictive value was 83.3 .3.6 Salivary vs. Plasma NT-proBNP Concentrations in the HF PopulationThe NT-proBNP concentration in the plasma samples ranged from 486 pg/mL to 97,319 pg/mL, with a median of 22731 pg/ mL (IQR, 5386 pg/mL to 36833 pg/mL). There was no correlation between salivary NT-proBNP and plasma NT-proBNP concentrations in the HF patients (Figure 2B). The correlation of NT-proBNP concentration in plasma and saliva are as follow: n = 45; r2 = 0.006, p = 0.66.3.3 Comparative Analysis of NT-proBNP Immunoassay with a Commercially Available AssayIn total, 37 plasma samples that have been previously measured for NT-proBNP levels (concentration ranges between 5 pg/mL to 42,150 pg/mL) were analysed using the two methods. The results are shown in Figure 1.3.4 The Effect of Concentrating Saliva SamplesSalivary NT-proBNP concentrations from 18 HF patients were initially below the LOD of our immunoassay (16 pg/mL). Upon concentrating saliva samples, we detected NT-proBNP above the LOD (10 saliva sample from HF patients with 27.1 pg/mL to 243.8 pg/mL) and no NT-proBNP levels were detected in the filtrates. Furthermore, NT-proBNP was not detected in both concentrated saliva samples (concentrates and filtrates) coll.The reference standard based upon clinical diagnosis of heart failure (as described in Section 2.1).Percentage recovery ?? NT-proBNP concentration in spiked saliva NT-proBNP concentration in un-spiked saliva =amount of spiked NT-proBNP ?1002.5.2 Intra-and inter-assay coefficient of variation. To determine intra- and inter- assay variation, triplicates of saliva samples from 45 HF patients and 40 healthy controls were run in!Results 3.1 ParticipantsIn total 45 symptomatic HF patients (with left ventricular ejection fraction ,40 ) and 40 healthy volunteers (young, and old) were enrolled in the study. The group of HF patients consisted of 23 males and 22 females, with a median age of 73 years (ages from 53 to 88), body mass index (BMI) of 29.14, and systolic and diastolic blood pressure of 125 mm Hg and 71 mm Hg, respectively. The group of healthy controls consisted of 20 males and 20 females, with a median age of 56 years (ages from 40 to 71) and BMI of 25. Gender, age, BMI, blood pressure, heart rate, New York Heart Association (NYHA) classification of the 85 volunteers were summarised in Table 1.Figure 1. Comparison of the NT-proBNP immunoassay when compared with a commercially available diagnostic assay (Roche Diagnostics, USA). r2 = 0.78 and p,0.001. doi:10.1371/journal.pone.0048452.gRelevance of Salivary NT-ProBNP and Heart FailureFigure 2. The comparison of the plasma and salivary NT-proBNP concentrations for HF patients (n = 45). (A), NT-proBNP levels in saliva of HF patients (n = 45) and healthy participants (n = 40). (B), The correlation of NT-proBNP concentrations in plasma and saliva of HF patients measured by our NT-proBNP immunoassay. doi:10.1371/journal.pone.0048452.g3.2 Assay Performance for the NT-proBNP AlphaLISA(R) ImmunoassayThe performance characteristics of the NT-proBNP immunoassay is summarised in Table 2. Intra- and inter-assay coefficients of variation (CVs) for the NT-proBNP immunoassays were below 10 . The LOD for the salivary immunoassay was approximately 16 pg/mL.The clinical sensitivity and specificity of the salivary NTproBNP immunoassay was 82.2 and 100 respectively, with an overall diagnostic accuracy of 90.6 . The positive predictive value for the salivary immunoassay was 100 , and negative predictive value was 83.3 .3.6 Salivary vs. Plasma NT-proBNP Concentrations in the HF PopulationThe NT-proBNP concentration in the plasma samples ranged from 486 pg/mL to 97,319 pg/mL, with a median of 22731 pg/ mL (IQR, 5386 pg/mL to 36833 pg/mL). There was no correlation between salivary NT-proBNP and plasma NT-proBNP concentrations in the HF patients (Figure 2B). The correlation of NT-proBNP concentration in plasma and saliva are as follow: n = 45; r2 = 0.006, p = 0.66.3.3 Comparative Analysis of NT-proBNP Immunoassay with a Commercially Available AssayIn total, 37 plasma samples that have been previously measured for NT-proBNP levels (concentration ranges between 5 pg/mL to 42,150 pg/mL) were analysed using the two methods. The results are shown in Figure 1.3.4 The Effect of Concentrating Saliva SamplesSalivary NT-proBNP concentrations from 18 HF patients were initially below the LOD of our immunoassay (16 pg/mL). Upon concentrating saliva samples, we detected NT-proBNP above the LOD (10 saliva sample from HF patients with 27.1 pg/mL to 243.8 pg/mL) and no NT-proBNP levels were detected in the filtrates. Furthermore, NT-proBNP was not detected in both concentrated saliva samples (concentrates and filtrates) coll.
