| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Enterprise Advisory Services, Inc., Houston, TX2 Wyle Laboratories, Houston, TX3 Human Adaptation and Countermeasures Office, NASA Johnson Space Center, Houston, TX
aaddress correspondence to this author at: NASA Johnson Space Center, Nutritional Biochemistry Laboratory, Human Adaptation and Countermeasures Office, Mail Code SK3, 2101 NASA Pkwy, Houston, TX 77058; fax 281-483-2888, e-mail scott.m.smith{at}nasa.gov
Point-of-care testing is a valuable tool in many settings, including remote environments (1). The i-STAT Portable Clinical Blood Analyzer (i-STAT Corp.) has been tested extensively and has been used during spaceflight as both a research (2)(3) and a clinical(4) tool. One disadvantage of the Portable Clinical Blood Analyzer is the requirement for refrigerated storage for consumables (ambient storage is limited to 2 weeks). The effects of re-refrigeration after transient ambient storage are undocumented, but this is discouraged by the manufacturer.
We determined the effects of long-term ambient storage and the effects of re-refrigeration on the performance of i-STAT EC6+ cartridges. Cartridges were stored under one of three treatment conditions: controlled temperature (refrigerated; 2–8 °C), room temperature (18–25 °C), or re-refrigerated. The re-refrigerated cartridges started the study at room temperature and were placed in the refrigerator after a specified number of days. For example, cartridges in the re-refrigerated 16-day group were held at room temperature for 16 days and then were placed in the refrigerator. Subsets were analyzed on study day 23, and after 1 month, 2 months, and so forth.
We used two levels of aqueous controls (Bionostics Corp.) and two levels of blood controls (Hematronix). Each group consisted of 20 cartridges: 6 Bionostics level 1 controls, 6 Bionostics level 3, 4 Hematronix low controls, and 4 Hematronix high controls. Cartridges and aqueous controls were single lots; blood controls were from three lots (expiration constraints). The limit ranges (levels 1 and 3, respectively) of each lot of each control (from package inserts) were as follows: pH, 7.130–7.230 and 7.647–7.747; ionized calcium (iCa), 1.47–1.69 and 0.71–0.87 mmol/L; glucose, 2.2–3.2 and 13.4–19.8 mmol/L; sodium, 116–124 and 156–166 mmol/L; and potassium, 2.0–2.8 and 5.8–6.8 mmol/L. The overall hematocrit control limit range was 26–30% packed cell volume (PCV) for the low control and 43–49% PCV for the high control.
Reproducibility of the data (controlled temperature and room temperature) obtained with cartridges was calculated for each test period. A repeated-measures (RM) ANOVA was used to assess whether storage and time had an impact on the CV, which was also compared with the manufacturer’s published precision.
The mean difference was calculated, and differences between controlled temperature and room temperature cartridges were compared with the acceptable performance criteria (5). Regression analyses were used to determine whether the percentage of the difference changed over time.
To understand the magnitude and pattern of deterioration, we completed a two-way RM ANOVA on the controlled temperature and room temperature data. A "type by time" interaction would indicate deterioration as a function of time. The differences were calculated between the data and published means for each analyte. For dependent variables in which the pattern of means indicated that deterioration increased as a function of time, evidence was obtained by seeking significant linear trends using regression analysis.
Statistical analyses were performed with SigmaStat (SPSS) and STATView (SAS Institute). ANOVA post hoc comparisons were performed with the Tukey test.
Differences between controlled temperature and re-refrigerated data at 1 year were determined by one-way RM ANOVA with time as the repeated factor. To determine the clinical relevance of any changes or trends, we compared each significantly different mean with the published acceptable performance criterion (5).
At 1 year, mean values for pH (level 3) and iCa (level 3) controls measured with room temperature cartridges were outside the manufacturer’s acceptable ranges, whereas means for all other analytes were still within acceptable ranges (Fig. 1
). At 6 months, some individual room temperature data points for pH (n = 2 for level 1; n = 3 for level 3) and iCa (n = 1 for level 1) were outside the acceptable ranges. For the controlled temperature and re-refrigerated cartridges, mean values were not outside the manufacturer’s acceptable range for either control of any analyte.
|
There were no differences in CV between controlled temperature and room temperature cartridges over time through 12 months for any analyte (tested by RM ANOVA). The CV (and SD of CV) for the level 1 control (low control for hematocrit) for the controlled temperature and room temperature cartridges, and the manufacturer’s published values, respectively, were as follows: pH = 0.27 (0.11)%, 0.43 (0.34)%, and 0.08%; iCa = 2.7 (2.5)%, 1.6 (1.6)%, and 1.14%; glucose = 1.5 (1.0)%, 1.1 (0.6)%, and 3.1%; sodium = 0.6 (0.5)%, 0.3 (0.2)%, and 0.5%; potassium = 0.41 (0.76)%, 0.8 (2.3)%, and 1.14%; hematocrit = 2.7 (3.3)%, 1.3 (1.7)%, and 2.3%. The CV for the level 3 control (high control for hematocrit) for the controlled temperature and room temperature cartridges, and the manufacturer’s published values, respectively, were as follows: pH = 0.18 (0.08)%, 0.23 (0.12)%, and 0.04%; iCa = 3.0 (2.6)%, 2.4 (2.0)%, and 1.56%; glucose = 1.9 (1.4)%, 1.6 (0.8)%, and 2.6%; sodium = 0.4 (0.4)%, 0.5 (0.3)%, and 0.5%; potassium = 0.4 (0.6)%, 1.4 (2.0)%, and 1.70%; hematocrit = 2.3 (2.7)%, 1.8 (1.3)%, and 2.0%.
The only time-related trends that were discerned by regression analysis of the calculated difference for the room temperature cartridges were in the analysis of pH (level 3; P <0.001), glucose (level 1; P <0.05), and potassium (level 3; P <0.01; see the Data Supplement that accompanies the online version of this Technical Brief athttp://www.clinchem.org/content/vol50/issue4/). The mean difference for each analyte was also within the published acceptable performance criteria.
Means for each variable were obtained for all conditions and times of refrigeration and plotted (data not shown) to discern any patterns of degradation. Differences between study (controlled temperature and room temperature cartridges) data and the published mean values for each control were calculated and compared with the acceptable performance criteria (5). Acceptable performance criteria were calculated as 10% and 6% of the manufacturer’s means for glucose and hematocrit, respectively. The mean (SD) differences for the level 1 control (low control for hematocrit) for the controlled temperature and room temperature cartridges, and the acceptable performance criteria (5), respectively, were as follows: pH = 0.001 (0.012), 0.000 (0.028), and 0.04; iCa = 0.00 (0.02) mmol/L, 0.03 (0.04) mmol/L, and criteria unavailable; glucose = -0.21 (0.08) mmol/L, -0.09 (0.14) mmol/L, and 0.27 mmol/L; sodium = -1.0 (1.1) mmol/L, 1.5 (0.9) mmol/L, and 4 mmol/L; potassium = -0.01 (0.04) mmol/L, 0.10 (0.05) mmol/L, and 0.5 mmol/L; hematocrit PCV = 0.16 (0.82)% PCV, 0.43 (0.90)% PCV, and 1.7% PCV. The mean (SD) differences for the level 3 control (high control for hematocrit) for the controlled temperature and room temperature cartridges, and the acceptable performance criteria (5), respectively, were as follows: pH = 0.008 (0.014), -0.011 (0.041), and 0.04; iCa mmol/L = -0.01 (0.02) mmol/L, 0.00 (0.04) mmol/L, and criteria unavailable; glucose = 1.34 (0.20) mmol/L, 1.79 (0.26) mmol/L, and 1.66 mmol/L; sodium = -1.7 (0.7) mmol/L, 0.8 (0.6) mmol/L, and 4 mmol/L; potassium = -0.12 (0.05) mmol/L, -0.08 (0.12) mmol/L, and 0.5 mmol/L; hematocrit = -1.0 (1.3) % PCV, -0.28 (1.23) % PCV, and 2.8% PCV.
Trends in the data were significant (P <0.05) and linear for room temperature cartridges for pH (level 3), iCa (both levels), glucose (level 1), sodium (both levels), and potassium (level 3). Trends in the data were significant (P <0.05) and linear for controlled temperature cartridges for sodium (level 1) and potassium (both levels). Neither the significant interaction between storage condition and time nor the existence of any significant linear trends produced major differences between the mean differences obtained for controlled temperature and room temperature cartridges. With the exception of glucose (level 3 only), the mean differences consistently met the published acceptable performance criteria. Moreover, the mean difference for the level 3 glucose values was within 10% of the values obtained for controlled temperature cartridges at day 16.
Data from controlled temperature and re-refrigerated cartridges were compared by RM ANOVA (Table 1
). The concentrations of all analytes measured with the re-refrigerated cartridges at both control levels were within the acceptable range after 1 year (the maximum storage period at room temperature was 6 months). Values for pH, iCa, potassium, and hematocrit were similar for controlled temperature and re-refrigerated cartridges after 1 year of storage. The glucose concentration in level 1 controls was lower in cartridges re-refrigerated for 23 days, 1 month, 2 months, 4 months, and 6 months; the concentration in level 3 controls was lower in cartridges re-refrigerated for 16 and 23 days and for 6 months. Sodium values for the level 1 control were lower in cartridges re-refrigerated for 16 days, 23 days, 4 months, and 6 months. However, none of these differences exceeded the acceptable performance criterion.
|
Our data show that results for control material are acceptable with i-STAT EC6+ cartridges stored at room temperature for extended periods, specifically up to 4 months for all analytes and up to 1 year for glucose, sodium, potassium, and hematocrit. When measured with cartridges stored for 6 months at room temperature, some of the values obtained for pH and iCa concentration in control solutions were outside the acceptable ranges for those controls. However, although they were approaching the limit, the means were within the acceptable range. Because the next previous data collection point was 4 months, we speculate that stability lies between 4 and 6 months, perhaps closer to 5 or 6 months because of the lack of decay at 4 months. The data also show that re-refrigeration of cartridges after exposure to ambient temperature does not diminish cartridge reliability as assessed by results on control materials. This is relevant for field studies and for work in remote locations.
CLIA-certified laboratories may not use these data to modify their practices, but these results should encourage manufacturers to perform more extensive testing, including patient samples, to ensure maximum flexibility for the end-user.
Acknowledgments
This study was funded by NASA. We thank Jane Krauhs for technical editing of the manuscript.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  S. R. Zwart, G. E. Crawford, P. L. Gillman, G. Kala, A. S. Rodgers, A. Rogers, A. M. Inniss, B. L. Rice, K. Ericson, S. Coburn, et al. Effects of 21 days of bed rest, with or without artificial gravity, on nutritional status of humans J Appl Physiol, July 1, 2009; 107(1): 54 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Smith, S. A. Mathews Oliver, S. R. Zwart, G. Kala, P. A. Kelly, J. S. Goodwin, and C. B. Dyer Nutritional Status Is Altered in the Self-Neglecting Elderly J. Nutr., October 1, 2006; 136(10): 2534 - 2541. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and room temperature (
).