College of American Pathologists
Printable Version

  Q & A





cap today

December 2001

Q.  Recently a new pulmonologist at our hospital became convinced that the values we report for the pH of pleural fluid obtained by thoracentesis are inaccurate and "physiologically incompatible." For instance, several patients with effusions have had values ranging from 6.5 to 8.3. The pulmonologist is accustomed to values obtained from a blood gas analyzer, which we feel is invalid, rather than the manual pH probe technique that our laboratory uses. We are confident that the values we report are correct, but we find little information in the literature to support our method. What is the optimum technique for specimen handling, testing, and reporting?

A.  Several recent studies1-3 have reported on methods for measuring pleural fluid pH, and a fairly strong consensus exists on the following points:

  • The method of choice is a blood pH/gas analyzer.
  • pH paper is an unacceptable method. It is very imprecise and usually biased high.
  • A pH meter with an external electrode is not recommended. Results are biased high and are more imprecise than with a blood gas analyzer.

One survey of hospital laboratories showed that, of those performing the test in-house, 56 percent used pH paper and 12 percent used a pH meter. (In this survey, researchers also found that many hospitals sent the specimens to reference laboratories, one of which was sending the fluids overnight to a central facility, which then used pH paper for the determination.)

pH paper has repeatedly been found to be much too imprecise for clinical use. A pH meter is, in theory, capable of accurate and precise measurements. However, pleural fluid has a PCO2 similar to blood, and specimens must be collected and handled anaerobically. It is difficult to measure pH with an external electrode without significant exposure to room air, which is probably responsible for the poor agreement with blood pH/gas analyzers seen in some studies. More important, the pH data in the literature, which guides clinical decisions, were all determined with blood pH/gas analyzers at 37°C. Measurements with a pH meter will be at room temperature and therefore will be biased high.

The main objection to using a blood pH/gas analyzer for pleural fluid is that some specimens are viscous or prone to clot and can clog the sample path in the analyzer. In addition to being collected anaerobically, specimens should be heparinized and analyzed soon after collection. Grossly purulent fluid should not be analyzed. Some people recommend the use of a filtering device on the end of the syringe to remove small clots or particulate matter.

Finally, it has been reported that the i-Stat point-of-care analyzer gives pleural fluid pH results that are in excellent agreement with a blood pH/gas analyzer.4 The single-use cartridge avoids the issue of fouling the sample path of a larger instrument.

1.  Chandler TM, McCoskey EH, Byrd RP Jr, et al. Comparison of the use and accuracy of methods for determining pleural fluid pH. South Med J. 1999;92:214-217.
2.  Cheng DS, Rodriguez RM, Rogers J, et al. Comparison of pleural fluid pH values obtained using blood gas machine, pH meter, and pH indicator strip. Chest. 1998;114: 1368-1372.
3.  Lesko EP, Roth BJ. Is pH paper an acceptable low-cost alternative to the blood gas analyzer for determining pleural fluid pH? Chest. 1997;112:1291-1292.
4.  Kohn GL, Hardie WD. Measuring pleural fluid pH: high correlation of a handheld unit to a traditional blood gas analyzer. Chest. 2000;118:1626-1629.

Robert W. Burnett, PhD
Department of Pathology and Laboratory Medicine
Hartford (Conn.) Hospital
Consultant, CAPChemistry Resource Committee

Q.  Is there an accurate and consistent method for estimating the white blood cell count using the peripheral blood smear? We often encounter interferences with the automated WBC and need an accurate and consistent method for estimating the WBC using the peripheral smear.

A.  White blood cell counts generated on automated hematology analyzers sometimes need to be verified due to possible count interferences. These results are generally flagged by the instruments to indicate that the values may have questionable accuracy due to certain conditions, such as platelet clumps or presence of normal red blood cells. The flagged results require validation by a technologist prior to reporting.

One of the most common validation action methods is estimating the WBC count from a stained blood smear. The WBC estimate involves determining the average number of WBCs per high-power field in the proper area of a well-stained blood smear. This average is then multiplied by a predetermined factor (the WBC estimation factor) to derive the estimated WBC count.

A good procedure for determining the WBC estimating factor is described below:1

1.  Perform automated WBC counts on 30 consecutive specimens from fresh patient blood samples.
2.  Prepare and stain one peripheral blood smear for each sample.
3.  Count the number of WBCs in 10 consecutive fields under high magnification for each smear. (40x or oil magnification can be used, as long as the same magnification is used consistently throughout the procedure.)
4.  Divide the total number of WBCs counted by 10 to obtain the average number per high-power field.
5.  Divide the automated WBC by the average number of WBCs per high-power field to obtain the conversion factor for each sample.
6.  Add all the numbers obtained in step five and divide by 30 (the number of samples in this analysis) to obtain the average ratio of the automated WBC count to the WBC count per high-power field. Round this number off to the nearest whole number to obtain the WBC estimation factor.

The WBC estimation factor obtained by this method is equivalent to the number of peripheral blood WBCs represented by one WBC in a high-power microscopic field. The same procedure can be applied to determine the platelet estimation factor. Each laboratory should determine these estimation factors using its own instrumentation and microscopes.

Once the WBC estimation factor has been determined, the technologist calculates the average number of WBCs per high-power field on subsequent patient specimens and multiplies this value by the estimation factor to obtain the WBC estimate. For example, if the WBC estimation factor = 2 and there are 5.2 WBC/hpf, the WBC estimate is 10.4 x 109/L.

An accurate estimate depends on an acceptable anticoagulated specimen free from clots, a well-prepared blood smear, well-trained laboratory staff, and determination of a WBC estimation factor appropriate for that laboratory’s instrumentation. Estimates may also be affected by the hematocrit value of the specimen; the WBC estimate tends to be falsely low as the hematocrit increases and falsely high as the hematocrit decreases.

The greatest use of this method is to verify questionable automated WBC counts. In general, the instrument count is acceptable when it is within 20 percent of the estimated count. In the event that the instrument count cannot be verified, many laboratories choose to report the WBC estimate as low, normal, or high. Since it is difficult to perform leukocyte estimates accurately and consistently, this is probably a good practice.

1.  Stiene-Martin EA, Lotspeich-Steininger CA, Koepke JA. Clinical Hematology: Principles, Procedures, Correlations. Philadelphia, Pa: J.B. Lippincott Co. 1992.

Linda Sandhaus, MD
Department of Pathology
University Hospitals of Cleveland
Case Western Reserve University School of Medicine

CAP Hematology/Clinical
Microscopy Resource Committee