| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Automation and Analytical Techniques |
1 Department of Chemistry, Tufts University, Medford, MA; 2 Department of Medicine, Boston University School of Medicine, Boston, MA; 3 Department of Periodontology and Oral Biology, Boston University Goldman School of Dental Medicine, Boston, MA; 4 Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA.
aAddress correspondence to this author at: Tufts University, 62 Talbot Ave., Medford, MA 02155. Fax 617.627.3443; e-mail david.walt{at}tufts.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: Solution-based colorimetric-detection chemistries for NO2– and UA were converted to a test strip format to produce a simple method for semiquantitatively measuring NO2– and UA concentrations in the clinic or at the patient’s home. We assessed the test strips with saliva samples collected from both ESRD patients undergoing dialysis and healthy control volunteers to qualitatively monitor the effect of dialysis on salivary NO2– and UA. We used computer software to analyze digital images of the resulting test strip color intensities.
Results: Test strip measurements showed that mean salivary concentrations of NO2– and UA were decreased in ESRD patients by 86% and 39%, respectively, compared with 15% and 9% for time-matched controls. Comparison of test strip results with calibrated solution-based assays suggests that the test strips can semiquantitatively measure salivary concentrations of NO2– and UA.
Conclusions: The colorimetric test strips monitored changes in salivary NO2– and UA concentrations that occurred in ESRD patients during dialysis. The test strips may prove useful for noninvasively evaluating dialysis progress and may also be useful for monitoring renal disease status.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With the goal of developing a diagnostic test for the simultaneous detection of multiple markers in saliva, we selected end-stage renal disease (ESRD)1 as a suitable target disease state, because the disorder is a well-defined phenotype and its effect on blood composition is well known. Owing to the contribution of serum-derived components to whole saliva, we hypothesized that changes in serum composition caused by hemodialysis would be reflected in saliva. Several studies have discovered that salivary concentrations of biomarkers are decreased during dialysis, suggesting that saliva tests could be used to evaluate dialysis efficacy (6)(7)(8)(9).
We describe the ability to monitor salivary nitrite (NO2–) and uric acid (UA) concentrations as a possible method for the surveillance of dialysis treatment efficacy. Once the trends associated with these analytes were confirmed, a simple method of multiplexed detection was required. The development of simple and robust testing methods for point-of-care (POC) use is of great interest, and the ideal method should be inexpensive and easy to interpret by both healthcare workers and patients, particularly in the home-testing setting. Colorimetric test strips provide an ideal format to fulfill the requirements of POC diagnostic tests (10)(11). We converted NO2–- and UA-detection chemistries into a test strip format, which we used to rapidly measure concentrations of these analytes in saliva samples collected from ESRD patients in a dialysis clinic.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
patients and sample collection
We recruited 43 ESRD patients undergoing regular hemodialysis for the study. Twelve patients participated in the initial phase to screen potentially informative analytes in saliva samples collected before and after dialysis, 12 participated in the second phase in which saliva samples were collected at hourly intervals throughout dialysis, and 19 patients participated in the POC test strip study. For the POC study, we also recruited 10 healthy control volunteers in the Clinical Research Center at Boston University School of Dental Medicine. Patient volunteers were recruited at a chronic dialysis center affiliated with Boston University Medical Center (BUMC). All study participants signed consent forms according to BUMC-approved institutional review board regulations. Participants were asked to fast for 2 h before providing saliva samples. For the 3 stages of the study, samples were collected in the following 3 manners: (a) once before and once after dialysis on a weekly basis for a 2-month period in the initial analyte-screening phase; (b) at hourly intervals throughout dialysis, with the final sample collected just before the conclusion of dialysis in the confirmatory analyte-screening phase; or (c) once before and once immediately after dialysis (approximately 4 h later) in the test strip study. Salivary flow was stimulated by mastication of a 1.5-g wax bolus (Parafilm) at a mastication rate of 30 strokes per minute. Participants were asked to expectorate every 30 s into a Falcon tube kept on ice until the collection volume totaled 5.0 mL. Samples were immediately processed by centrifugation at 27 000g for 20 min at 4 °C. Aliquots (1.0 mL) of saliva supernatant were placed in 1.5-mL polyethylene tubes and stored at –80 °C until use. For the test strip study, we collected approximately 1.0 mL of whole saliva for an initial test strip measurement in the clinic (see the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol54/issue9 ), collected an additional volume of approximately 3.0 mL, and processed the samples with the centrifugation and fractionation procedures described above.
analysis of saliva composition
A SpectraMax Gemini microplate spectrofluorometer (Molecular Devices) and untreated, flat-bottom, 96-well black microtiter plates (Costar; Corning) were used for solution-based fluorescence assays. SoftMax Pro software (Molecular Devices) was used for data acquisition and analysis.
uric acid
Salivary UA concentrations were measured with an Amplex Red Uric Acid/Uricase Assay Kit according to the manufacturer’s instructions. In brief, 5 µL of saliva was diluted with 45 µL of deionized water and combined with a solution of uricase, horseradish peroxidase, and Amplex Red indicator in each well of the microtiter plate. Fluorescence was measured with excitation and emission wavelengths of 550 nm and 590 nm, respectively.
sodium
Sodium Green indicator was used to measure the Na+ concentration. Six aqueous Na+ calibrators (0.2– 6.25 mmol/L) were prepared from a 1-mol/L NaCl stock solution. We pipetted 50 µL of either undiluted saliva or NA+ calibrators into microtiter plate wells to react with 50 µL of 10 µmol/L Sodium Green in DMSO. The excitation and emission wavelengths used for fluorescence measurements were 507 nm and 532 nm, respectively.
chloride
Cl– concentrations were measured with the fluorescent indicator SPQ. Seven aqueous Cl– calibrators (1.56–100 mmol/L) were prepared from a 1-mol/L NaCl stock solution. We allowed 100 µL of either undiluted saliva or Cl– calibrators to react with 50 µL of 2 mmol/L SPQ in deionized water. Fluorescence was measured with excitation and emission wavelengths of 344 nm and 443 nm, respectively.
nitrite
NO2– concentrations were measured fluorometrically with DAN (12)(13). Eight aqueous NO2– calibrators (0.39–50 µmol/L) were prepared from a 0.1-mol/L NO2– stock solution. Saliva samples were diluted with 3 volumes of deionized water. We then pipetted 100 µL of diluted saliva or the NO2– calibrator into microtiter plate wells to react for 15 min with 10 µL of 0.1 g/L DAN in 0.6 mol/L HCl. Before measuring the fluorescence, we added 10 µL of 2.8 mol/L NaOH to each well. Fluorescence was measured with excitation and emission wavelengths of 365 nm and 450 nm, respectively.
test strip study and confirmatory analysis
Chromatography paper was impregnated with solution-based NO2–- and UA-detection chemistries (14)(15)(16)(17). The 2 test papers were then immobilized onto vinyl backing material, which was then hand-cut into strips. The colorimetric test paper for salivary NO2– measurement uses the principle of the Griess reaction, a common method for NO2– quantification (17). When a sample containing NO2– comes into contact with the paper, the citric acid impregnated in the paper converts NO2– to nitrous acid, which then diazotizes sulfanilamide. The diazotized sulfanilamide then couples with N-(1-napthyl)-ethylenediamine to produce a red-violet azo compound. The intensity of the resulting color is proportional to the amount of NO2– present. The colorimetric test paper for UA is based on a bicinchoninate chelate method (14). When the UA test paper is exposed to a sample containing UA, Cu(II) is reduced to Cu(I), which then forms a chelate with sodium bicinchoninate. The intensity of the resulting deep-violet precipitate is proportional to the amount of UA present in the sample. Detailed information on test strip fabrication and characterization is available in the online Data Supplement.
Saliva samples were collected from 19 ESRD patients in the BUMC dialysis clinic immediately before and immediately after treatment. Ten control individuals also donated 2 samples each. In parallel with the approximate time required for ESRD patients to undergo a complete dialysis treatment, the first sample was collected at time zero, and the second sample was collected 4 h later. Samples were processed as described above, frozen, and sent to Tufts University for storage at –80 °C. Aliquots of saliva supernatant were thawed and analyzed by immersing test strips in the samples and imaging the strips on a desktop digital scanner. Test strip color intensities were measured by means of the histogram function in Adobe Photoshop and converted to concentrations by comparing the intensities with a calibration curve of digitally analyzed test strip color intensities produced by calibrator solutions of known concentration (see the online Data Supplement). We also tested saliva samples with solution-based quantitative assays for NO2– and UA, as described above (the DAN method and the Amplex Red Uric Acid/Uricase Assay Kit, respectively).
We directly compared the test strip assays and the solution-based assays by analyzing the same set of saliva samples with both methods. The net changes in NO2– and UA concentrations (initial at t = 0 minus final at t = 4 h) measured with the test strips were compared with those obtained with the microtiter plate methods for the same set of patients and control individuals. The degree of agreement of the net differences for the 2 test methods was examined with Bland–Altman plots. For each participant, the difference between the net change in concentration measured with the test strips and that obtained with the microtiter plate–based assay was plotted against the mean concentration change obtained by both methods.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, the UA and NO2– results correlated best with dialysis progression, because a net decrease in their concentrations was observed in all 12 patients (Fig. 1
, A and D). NA+ and Cl– did not consistently show the same trends (Fig. 1
, B and C). Owing to sample-volume limitations for several time points, we could not acquire complete time traces for every patient. Despite this lack of information for several patients, we collected pre- and postdialysis samples for every patient and measured all variables in these samples. Thus, we were able to evaluate the overall effect of dialysis on analyte concentrations. Although several patients exhibited fluctuations in UA and NO2– concentrations throughout dialysis before the final measurement, we consistently observed net decreases in the concentrations of these analytes from predialysis samples to postdialysis samples. Normalization and averaging of all patient data provide a simplified representation of the trends we observed for both UA and NO2– across all 12 patients (Fig. 2
). Furthermore, the normalized predialysis and postdialysis values were significantly different with no overlapping SDs at the 95% confidence interval (P < 0.0001 for both UA and NO2–). The net decreases in concentration observed for both analytes from predialysis samples to postdialysis samples support their potential utility as markers of dialysis efficacy.
|
|
test strip study
We observed a wide range of initial/predialysis NO2– concentrations among ESRD patient samples with the test strips (Fig. 3
). Salivary NO2– concentrations measured with the test strips decreased after dialysis in all but 1 ESRD patient. Compared with the ESRD patients, the control samples showed no consistent trend in salivary NO2– concentration between the time zero and the 4-h time points. The concentrations for 3 control individuals increased over the 4-h time period, and the range of initial concentrations was less variable. Salivary UA concentrations measured with the test strips decreased in all but 2 ESRD patients, whereas, again, we observed no consistent trend in salivary UA concentrations in the time-matched control samples. The UA concentration increased from the zero time point to the 4-h time point in 5 of the 10 control volunteers. Although the mean initial NO2– and UA values for the ESRD patients (161 µmol/L and 202 µmol/L, respectively) and the healthy controls (75.2 µmol/L and 160 µmol/L) were not significantly different (P = 0.15 and 0.12, respectively), the mean net changes in concentration observed for the 2 populations were significantly different (P = 0.018 and 0.0077 for NO2– and UA, respectively).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Nitric oxide (NO), a strong vasorelaxant, is continually produced by the kidneys via the metabolism of L-arginine (21). Owing to the complicated functions of NO in the human body, its role in ESRD has been debated (21)(22)(23). Rapid oxidation of NO to nitrate (NO3–) and NO2– occurs in biological fluids; therefore, NO3– and NO2– have been examined as qualitative markers of NO production (22). Of note, however, is that NO constitutes only a small fraction to the source of NO3– and NO2– in serum and saliva. A major determinant of serum NO3– and NO2– is the catabolism of proteins and amino acids, and dietary sources are another (24). Nonetheless, research has suggested a possible association of increased NO metabolite concentrations in plasma with enhanced NO production and decreased renal function (25). The concentration of NO2– plus NO3– in serum has been observed to decrease more than either blood urea nitrogen (BUN) or creatinine following dialysis, and the concentrations of these 2 analytes have been suggested as a potential indicator of renal impairment (26). Similar associations of NO3– concentrations have been observed in pre- and postdialysis plasma (27). Because salivary NO3– is reduced to NO2– by bacteria in the oral cavity (28) and because both NO3– and NO2– are removed from the blood by dialysis (26), salivary NO2– has also been selected as a suitable candidate marker for monitoring dialysis efficacy.
The gold standard test for evaluating dialysis efficacy is to measure pre- and posttreatment BUN concentrations. The reduction ratios for BUN, as well as for salivary NO2– and UA, were calculated as: [(predialysis concentration) – (postdialysis concentration)]/(predialysis concentration). Comparison of the salivary UA, NO2–, and BUN reduction ratios revealed some correlation (r = 0.24 and 0.43 for UA and NO2–, respectively; Table 1
). At least one of the 2 salivary analytes yielded similar reduction ratios in 9 of the 12 patients. For patients 2, 7, and 12, although the UA and NO2– reduction ratios did not correlate with the BUN reduction ratio (r = –0.81 and –0.55 for UA and NO2–, respectively), salivary UA and NO2– concentrations were both substantially decreased after dialysis. Of note, however, is that there is no a priori reason why saliva measurements for these analytes should necessarily correlate with BUN concentration. Given the conflicting observations regarding the correlation of salivary urea with BUN (6)(29), we have sought to offer evidence for alternative salivary markers that may be useful for noninvasively monitoring dialysis efficacy and renal function. The screening of a larger patient cohort—including comparisons of salivary NO2– and UA concentrations with salivary carbonyl, serum NO2–, and serum UA concentrations, as well as with other common serum (creatinine and/or BUN) or urine tests—could encourage the acceptance of saliva testing for evaluating dialysis efficacy. The results of the preliminary screening tests conducted in microtiter plates suggest that 2 salivary analytes, UA and NO2–, exhibit net decreases in concentrations in all patients throughout the dialysis process. Thus, we developed colorimetric test strips to rapidly and easily measure the concentrations of these 2 analytes in saliva.
|
Although the mean predialysis NO2– and UA concentrations obtained with test strips (Fig. 3
) do not agree with those obtained with the microtiter plate assays in preliminary screening experiments (Fig. 1
), it is important to point out that these 2 studies were performed with different individuals. Although the results of the 2 methods do not agree precisely, they give the same qualitative result for the 2 different patient groups when the net predialysis to postdialysis change in concentration is taken into account. When we used the 2 methods to test the same population, an examination of the Bland–Altman plot shows that the microtiter plate and test strip data were in good agreement for NO2– (Fig. 4
). The one exception was when the net concentration change was >700 µmol/L. In this instance, the results obtained with the 2 methods differed by nearly 300 µmol/L, but such a change is extreme, and both methods readily identified it as a large reduction. In this case, the test strip assay underestimated the net concentration change, compared with the solution-based assay. For the UA test strip assay, the Bland–Altman plot reveals less agreement between the net concentration changes obtained with the UA portion of the test strip and the solution-based assay. For UA, the 2 methods were in good agreement for nearly 75% of the samples. Although the UA test provides a lower agreement rate than the NO2– test, the combination of the 2 tests may provide useful information regarding the effectiveness of a dialysis treatment. These results demonstrate that the net changes in salivary NO2– and UA concentrations obtained with the 2 methods were in agreement and suggest that the test strips can be used for semiquantitative measurements.
|
In comparison, the test strip method yields instantaneous results for NO2– and UA simultaneously, whereas the solution-based assays require longer assay times and expensive analytical instrumentation. Although the bicinchoninate-chelate UA-detection method is not as selective as UA detection based on uricase (30), the presence of interfering substances in saliva is believed to be minimal; therefore, changes in UA levels should not be masked in the test strip assay (16). A comparison of test strip results with those obtained with the uricase-based detection method showed that the qualitative changes in UA concentrations obtained with the 2 testing methods were similar (Fig. 4
). Furthermore, nonenzymatic detection methods are preferable to enzymatic detection methods for test strip preparation because the shelf lives are longer and the reagents do not require any additional special storage conditions. After several months of storage in an airtight container with desiccant, the NO2–/UA test strips produced measurements comparable to those obtained with a freshly prepared batch (data not shown).
Several studies have examined the compositional changes in saliva produced by dialysis (8)(9), but the present investigation is the first to offer a method for simultaneously monitoring 2 salivary analytes for qualitatively assessing dialysis efficacy. It is important to note that in some cases a single analyte does not change while the second analyte shows a decrease consistent with the clinical status. Such a possibility underscores the potential value of performing multianalyte tests when making clinical decisions. Clearly, these analytes and this method require further study with saliva samples to conclusively evaluate their clinical utility. With the acceptance of salivary NO2– and UA as qualitative biomarkers, the test strip method we have presented may also become useful for patients with progressive renal failure or who are undergoing peritoneal dialysis, because baseline analyte concentrations can be established for individual patients and the concentrations can be monitored routinely by the patients in the convenience of their homes. Furthermore, for ESRD patients undergoing dialysis, technicians can quickly monitor salivary analyte concentrations throughout treatment to determine if a patient has achieved a successful dialysis before the standard completion time. We hypothesize that salivary NO2– and UA could be used in addition to the current gold standard blood tests (BUN and creatinine) used to evaluate renal function.
|
Acknowledgments |
|---|
Financial Disclosures: None declared.
Acknowledgments: The authors thank all study participants for donating saliva samples. The also authors acknowledge the staff at the BUMC dialysis clinic for their understanding during sample collection and J. Très Brazell for reviewing the manuscript.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. M. Nagler Saliva Analysis for Monitoring Dialysis and Renal Function Clin. Chem., September 1, 2008; 54(9): 1415 - 1417. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
