This invention relates to the field of hematology, and more particularly to the field of automated hematology analyzers.
It is common medical diagnostic practice to obtain a sample of a patient's blood and test the sample for various hematology parameters. For example, a patient's blood sample may be tested and analyzed to determine red blood cell count, platelet count, white blood cell count, white blood cell types, hematocrit and/or hemoglobin concentration. A number of other hematology parameters may also be determined and analyzed.
The parameters of the patient's blood revealed by the blood testing and analysis may be of significant assistance to a physician in making a diagnosis. For example, increased white blood cell count may indicate the existence of an infection in the body. Certain increased concentrations of white blood cells may indicate various conditions, such as leukemia. A high red blood cell count may indicate that the patient is not receiving enough oxygen and may suggest a condition such as lung disease or heart disease. A low red blood cell count may indicate that the patient is anemic.
Hemoglobin is the major substance in red blood cells. It carries oxygen and gives the blood its red color. Hemoglobin information is one parameter that the physician may use in making a diagnosis. For example, the amount of hemoglobin in the blood is a good indicator of the blood's ability to carry oxygen throughout a patient's body. A high hemoglobin value may be caused by a number of factors such as lung disease, heart disease or kidney disease. A low hemoglobin value may indicate anemia. Hemoglobin parameters may also be valuable in determining a patient's responsiveness to certain therapies, such as therapies directed toward diseases which affect hemoglobin. In addition to analyzing hemoglobin values, analysis may also be conducted on the various types of hemoglobin in the body. While there are only three types of normal hemoglobin, more than three hundred abnormal hemoglobin types have been discovered in patients with certain clinical symptoms. Abnormal hemoglobin types are often indicative of various conditions and/or diseases.
Automated hematology analyzers are currently used for measuring various hematology parameters of a patient's blood, including hemoglobin parameters such as hemoglobin concentration. These automated hematology analyzers are operable to analyze a number of hematology parameters, including white blood cell count, red blood cell count, platelet count and hemoglobin concentration.
When measuring hemoglobin concentration, the automated hematology analyzer takes a blood sample and first dilutes the sample with a diluent. A hemolytic reagent is then added to the diluted sample in order to lyse the red blood cells in the sample. Lysing the diluted sample converts the hemoglobin in the sample to methemoglobin. The methemoglobin is then complexed to form a relatively stable chromogen which is able to be detected and measured by UV spectroscopy at a given wavelength.
Following production of the chromogen in the lysed test sample, the test sample is passed through a hemoglobin absorption cuvette. A light source oriented on one side of the cuvette emits light through the cuvette. The light source emits light at a frequency at or near the peak absorption of the chromogen in the diluted sample (e.g., 540 nm). A detector positioned on the opposite side of the cuvette is used to detect the light that passes through the cuvette and sample. The detector and light may be provided as part of a spectrophotometer or other instrument operable to determine the absorption (or transmittance) of the light through the cuvette and sample. The absorption measurement obtained by the detector is then translated into a corresponding hemoglobin concentration for the sample. This translated hemoglobin concentration is multiplied by a calibration factor for the automated hematology analyzer to arrive at a final hemoglobin concentration measurement for the sample.
It has been noted that the temperature at which a hemoglobin measurement is taken for a blood sample has an effect on the hemoglobin measurement for such blood sample. One important reason for this is that the chromogen produced by the reaction of the diluted sample with the hemolytic reagent is not sufficiently stable to avoid sensitivity to its environment. The result is that the absorption of the chromogen varies with temperature. Because the absorption of the chromogen varies with temperature, different hemoglobin measurements may be obtained from a single sample depending upon the temperature of the sample when the measurement is taken. However, it should be noted that hemoglobin concentration is not the only hematology parameter that varies with temperature, as cellular size, counts and sub-population distribution may vary with temperature along with other hematology parameters.
In addition to variation with temperature, hemoglobin concentration and other hematology parameters may vary with time. In the case of hemoglobin concentration, the absorption of the chromogen produced from the hemolytic reaction decays with time. Accordingly, when obtaining a hematology measurement such as hemoglobin concentration, it is generally not acceptable to wait for the lysed sample to reach a steady state temperature. Instead, the absorption measurement must be taken relatively quickly following the reaction of the diluted sample with the hemolytic reagent. Since the measurement must be taken relatively quickly, some attempt must be made to deal with the temperature fluctuation of the lysed and diluted sample if an accurate hemoglobin measurement is desired.
Unfortunately, it is not easy to produce chromogen from the hemolytic reaction at a single stable temperature immediately following the reaction. For example, hemolytic reagents used to lyse hemoglobin often result in different reaction temperatures, and these reaction temperatures vary over time. Additionally, the environmental temperature of a laboratory may have an effect on reaction temperature.
Several prior art systems and methods have been proposed and used in an attempt to avoid fluctuating hemoglobin measurements because of temperature variations. However, these prior systems and methods have not produced satisfactory results, as significant temperature variations continue to produce different hemoglobin measurements when using these systems methods.
One proposed method for reducing the affects of temperature in hemoglobin measurements involves selecting ligands for the hemolytic reagent with high affinity to provide more stable chromogens that do not significantly vary with temperature, such as that disclosed in U.S. Pat. No. 5,763,280. Another method for reducing the effects of temperature on hemoglobin measurement involves using a hemoglobin stabilizer, such as that disclosed in U.S. Pat. No. 5,968,832. However both of these methods are unsatisfactory in their results as well as their additional costs.
The calibration method is another example of a prior art method for reducing the effects of temperature variation on hemoglobin measurement. The calibration method is used by many current automated hematology analyzers. This method acknowledges that the initial uncalibrated hemoglobin measurement taken by the automated hematology analyzer is not always accurate because of various factors such as engineering tolerances and environmental factors, and unique instrument characteristics. Using this method, an initial uncalibrated hemoglobin measurement is first obtained using the automated hematology analyzer. This uncalibrated measurement is then multiplied by a calibration factor to arrive at the calibrated hemoglobin measurement (e.g., HgbFinal=CalibrationFactor * HgbUncalibrated). The calibration factor is generally determined by empirical testing and programmed into the instrument before it is sold. The same calibration factor is applied to all hemoglobin measurements made with the instrument or to hemoglobin measurements made within a certain temperature operating range. While the calibration method provides for scaling of the measured temperature, these same changes are generally applied to all measurements or a whole range of measurements, and are not exact changes that account for temperature variations over a range of temperatures. Accordingly, the temperature calibration method provides generally unsatisfactory results when attempting to accurately measure hemoglobin.
Yet another example of a prior art method for reducing the effects of temperature variation on hemoglobin measurement is the temperature control method. The temperature control method involves the use of an automated hematology analyzer having a built-in temperature control unit. The temperature control unit in such an automated hematology analyzer generally warms the hemoglobin reaction temperature to a predetermined temperature such that all hemoglobin measurements using the automated hematology analyzer are taken at nearly the same temperature. Unfortunately, inclusion of a temperature control unit within the automated hematology has several problems. For example, the inclusion of the temperature control unit adds significant cost to the instrument which is then passed on to the purchaser of the instrument in the form of an increased price. Furthermore, the temperature control unit adds size to the instrument, and space is often a valuable resource in the laboratory environment. In addition, when a temperature control unit is added, additional parts are included in the machine that make the machine more susceptible to failure and need of repair. Moreover, even with a temperature control unit, measurement results are not always accurate, as the hemoglobin reaction temperature may change frequently or may be higher than expected (e.g., higher than the predetermined temperature), resulting in a measurement being taken before the temperature control unit stabilizes the temperature to the predetermined temperature.
Accordingly, it would be desirable to provide an automated hematology analyzer that is capable of accurately measuring various hematology parameters of a sample, such as hemoglobin concentration, at various sample temperatures, and does not require a temperature control unit.
A method of measuring a hemoglobin parameter of a test sample of blood, such as hemoglobin concentration, is described herein. The method comprises providing the test sample to be measured in the loading deck of an automated hematology analyzer. The automated hematology analyzer is operable to dilute and lyse the test sample in a reaction vessel. A temperature corresponding to the test sample is then obtained. The temperature corresponding to the test sample may be the temperature of the reaction vessel immediately after the test sample is diluted and lysed. However, numerous other temperatures corresponding to the temperature of the test sample may be obtained. Thereafter, the diluted and lysed test sample is delivered to a cuvette, and a spectrophotometer determines the absorbance and/or transmittance of the sample in the cuvette. With the absorbance and/or transmittance of the test sample, a first measurement of the hemoglobin parameter of the test sample is obtained. After a first measurement of the hemoglobin parameter is obtained, a processor determines a corrected measurement of the hemoglobin parameter of the test sample. The corrected measurement is a function of the measured temperature that corresponds to the test sample and the first measurement of the hemoglobin parameter. The method of measuring a hemoglobin parameter is valid over a range of test sample temperatures.
With reference to
The loading platform 12 is designed to receive a test vial/test tube containing a test sample of a patient's blood. The loading platform takes the vial and delivers its contents to the sample divider 14. The sample divider splits the sample into at least a first aliquot and a second aliquot. The first aliquot is delivered to the reaction chamber 16 in one portion of the automated hematology analyzer operable to determine white blood cell count and hemoglobin concentration. The second aliquot is delivered to another portion (not shown) of the automated hematology analyzer operable to determine red blood cell count and platelet count. In
After the blood sample is divided into separate aliquots, the first aliquot is delivered to a reaction chamber 16. The reaction chamber is generally a relatively large chamber compared to the size of the first aliquot. For example, in one embodiment, aliquots of about 28 μl are delivered to a reaction chamber that is between 7,000 μl and 10,000 μl in size. The reaction chamber is relatively large, because a large amount of diluent and/or hemolytic reagent is added to the test sample in the reaction chamber 16. For example, a 28 ml test sample may be combined with about 6,000 μl of diluent and about 1,000 μl of lysing agent in the reaction chamber 16. A reagent reservoir 18 is connected to the reaction chamber 16 and is operable to deliver the diluent and/or hemolytic reagent to the reaction chamber 16 when needed.
A temperature sensor 17 is connected to the reaction chamber and is operable to take a temperature of the reaction chamber. In one embodiment, the temperature sensor is a thermistor. However, the temperature sensor may take the form of any number of other types of temperature sensors operable to provide accurate and precise temperature readings, such as resistance temperature devices (RTDs), thermometers, IR thermometers, and thermocouples. The temperature taken by the temperature sensor 17 corresponds to the test sample in some manner. In the embodiment of
The reaction chamber is designed to feed its contents to a spectrophotometer 20 or other instrument operable to measure a hematology parameter, such as other commercially available photometers. To this end, the spectrophotometer 20 is operable to measure the absorption, transmittance, and/or other characteristic of the diluted and lysed test sample. The measured characteristic is then converted into a corresponding measurement for the hematology parameter.
The spectrophotometer includes a light source 21a, a cuvette 21b, and a detector 21c. To arrive at an absorption or transmittance reading, the test sample is passed through the cuvette and the light source emits light through the cuvette and passing test sample. The detector positioned on the opposite side of the cuvette obtains an absorption and/or transmittance reading for the test sample. To convert the absorbance and/or transmittance reading for the test sample into a hematology measurement, a look up table may be used to correlate the reading to the hematology measurement. This is accomplished by a processor 24 and memory 25.
In the embodiment of
The processor 24 is connected to an input/output device 26, such as an LED display screen and a keyboard. In the embodiment of
With reference to
Once the test sample is diluted and lysed, it is passed through a cuvette in step 108. As explained above, in one embodiment, the cuvette is part of a spectrophotometer or other measurement instrument. The spectrophotometer obtains an absorption and/or transmittance measurement for the test sample in step 110. This measurement is then passed on to the processor in step 112, where an uncorrected hemoglobin measurement is determined by the processor based on the absorption/transmittance measurement. In one embodiment, the processor determines the uncorrected hemoglobin measurement using look up tables stored in the memory. In particular, to arrive at the uncorrected hemoglobin measurement, the processor simply uses the absorption measurement obtained for the test sample to arrive at a corresponding hemoglobin measurement in the look-up table.
After the uncorrected hemoglobin measurement is determined, the processor obtains a corrected hemoglobin measurement in step 114. The corrected hemoglobin measurement is a function of the uncorrected hemoglobin measurement obtained in step 112, and the temperature measured in step 106.
In order to arrive at an equation for corrected hemoglobin that is a function of the uncorrected hemoglobin measurement and the measured temperature, extensive measurement of hemoglobin variations at different temperatures were recorded for multiple patients. With the hemoglobin measurements for different patients at different temperatures in hand, several mathematical functions were regressed from the data. These functions included both linear and non-linear functions.
One exemplary third order mathematical function that yielded accurate results over a wide temperature range was the following equation (1):
HgbCorrected=HgbUncorrected+a3(TMeasured−TRef)3+a2(TMeasured−TRef)2+a1(TMeasured−TRef) (1)
where,
HgbCorrected equals the corrected measurement of the hemoglobin concentration,
HgbUncorrected equals the uncorrected measurement of the hemoglobin concentration,
TMeasured equals the operating temperature,
TRef equals a reference temperature, and
a3, a2, a1 equal a third, a second, and a first order constants.
In the above equation (1), the reference temperature was 75° F. In addition, the first order constant, second order constant, and third order constant were all determined by the empirical method to regress the equation.
As mentioned previously, the above equation provided an accurate corrected hemoglobin reading over a wide temperature range. However, it is well known that a nonlinear system may be approximated by a linear one given a small enough range. In the present case, it was discovered that a linear function could provide accurate results for measured temperatures between 60° F. and 90° F. In that case, the linear function used to determine the corrected hemoglobin was determined to be as follows:
HgbCorrected=(Cal_Factor)*[HgbUncorrected+(Corr_Factor1)*(Tmeasured−Tstandard)] (2)
where
HgbCorrected is the corrected hemoglobin measurement;
Cal_Factor is a predetermined calibration factor;
HgbUncorrected is the uncorrected hemoglobin measurement;
Corr_Factor1 is a predetermined correction factor;
Tmeasured is the measured temperature corresponding to the test sample; and
Tstandard is a “standard” or reference temperature from which the measured temperature may deviate.
According to the above equation (2), the Cal_Factor, Corr_Factor1, and Tstandard are all predetermined values (predetermined constants in the equation). To determine the above mathematical function, as well as the predetermined constants, extensive measurement of hemoglobin variations at different temperatures were recorded for multiple patients. With the hemoglobin measurements for different patients at different temperatures in hand, the mathematical function was regressed from the data, and the correction factor (Corr_Factor1), reference temperature (Tstandard) and calibration factor (Cal_Factor) were determined empirically. When determining the calibration factor (Cal_Factor), the temperature reading for normal temperature calibration must factored into the value, as well as the reference temperature and the correction factor, as explained in the following paragraph.
The Cal_Factor value in the above embodiment is determined using a known hemoglobin assay for calibration. Cal_Factor may be dependent upon calibration temperature, the reference temperature, and/or a correction factor. Thus, in one embodiment, the calibration factor is determined according to the following equation (3):
Cal_Factor=Hgbknown
where
Hgbknown
Hgbmeasured
Corr_Factor2 is a predetermined correction factor;
Tcalibration is the measured temperature during the calibration; and
Tstandard is a “standard” or reference temperature from which the calibration temperature may deviate.
The nonlinear and linear equations (1) and (2) provided above are but two representative equations that may be used to arrive at a corrected hemoglobin measurement. As explained above, in both cases a corrected hemoglobin measurement is provided which is a function of an uncorrected hemoglobin measurement and a measured temperature. The above example equations for corrected hemoglobin measurement are not intended as limiting, but are provided as example functions that have been regressed from empirical data. Numerous other functions of uncorrected hemoglobin and measured temperature may be possible to arrive at corrected hemoglobin results.
The following example is illustrative of the method of hemoglobin correction due to temperature correction described herein, and should in no way be interpreted as limiting the invention, as defined in the claims.
Hemoglobin concentration was measured using a Beckman Coulter automated hematology analyzer for thirty blood samples. The thirty blood samples were diluted and lysed using Lyse S™ III. Hemoglobin concentration measurements were obtained for each of the thirty blood samples at several different temperatures ranging from 70° F. to 88° F. The temperatures were measured at the reaction chamber of the automated hematology analyzer immediately before the hemoglobin concentration measurements were taken. After uncorrected hemoglobin measurements were obtained, corrected hemoglobin measurements were obtained using the method described herein. Both the uncorrected and corrected hemoglobin results were logged as shown in the tables below. Table 1 below provides the uncorrected hemoglobin concentration measurements for the thirty samples measured. Table 2 below provides the corrected hemoglobin concentration for the thirty samples measured.
The data from the above tables are provided in graphical form in
Although the present invention has been described with respect to certain preferred embodiments, it will be appreciated by those of skill in the art that other implementations and adaptations are possible. For example, different hemolytic reagents could be used, with each producing different equation constants, depending upon the ligands used and the dilution ration. Moreover, there are advantages to individual advancements described herein that may be obtained without incorporating other aspects described above. Therefore, the spirit and scope of the appended claims should not be limited to the description of the preferred embodiments contained herein.