|Year : 2022 | Volume
| Issue : 2 | Page : 119-122
Platelet count estimation on peripheral smear: What should be an acceptable 'multiplication factor'?
Arnav Tiwari, Shweta Pal, Neha Singh, Michael Leonard Anthony, Nilotpal Chowdhury, Shalinee Rao
Department of Pathology and Laboratory Medicine, All India Institute of Medical Sciences, Rishikesh, Uttarakhand, India
|Date of Submission||21-May-2020|
|Date of Decision||22-May-2021|
|Date of Acceptance||25-May-2021|
|Date of Web Publication||13-Jul-2022|
Dr. Neha Singh
Department of Pathology and Laboratory Medicine, All India Institute of Medical Sciences, Rishikesh - 249 203, Uttarakhand
Source of Support: None, Conflict of Interest: None
Context: Automated cell counters have revolutionised laboratory medicine and are being used worldwide. They give accurate results, can run large number of samples and generate accurate results within a short span of time. Quality check procedures in this equipment ensure accuracy of results. However, for platelet counts, the pathologist is still dependent on confirming the results by manual methods, especially in situations of low platelet count and flagging by the electronic counter. Different laboratories use different methods for manual estimation of platelet counts. Aims: This study was done to compare the platelet counts assessed on peripheral smear examination by different multiplication factors, and to identify the multiplication factor which derives the most accurate platelet count. Settings and Design: This was an observational study. Methods: Peripheral smears prepared from 100 ethylenediaminetetraacetic acid-anticoagulated blood samples were manually examined under oil immersion field (OIF) for platelets. Platelet count was further calculated by multiplying the number of platelets per OIF using multiplication factors of 10,000, 15,000 and 20,000. This was compared with platelet (PLT)/red blood cell (RBC) method and automated platelet counts. Accuracy of various manual methods was evaluated by calculating the per cent difference of the methods with automated analyser counts. Statistical Analysis Used: Descriptive statistics was used for statistical analysis. Results: Manual platelet count estimation on peripheral smear employing multiplication factor of 15,000 was found to have comparable accuracy to analyser counts, followed by the PLT/RBC ratio method. Conclusions: This study confirms that for manual estimation of platelet counts on peripheral smear, a standard multiplication factor of 15,000 on OIF gives the most accurate result, which is closest to counts obtained by automated analysers.
Keywords: Automated counts, manual counts, multiplication factor, platelets
|How to cite this article:|
Tiwari A, Pal S, Singh N, Anthony ML, Chowdhury N, Rao S. Platelet count estimation on peripheral smear: What should be an acceptable 'multiplication factor'?. J Med Evid 2022;3:119-22
|How to cite this URL:|
Tiwari A, Pal S, Singh N, Anthony ML, Chowdhury N, Rao S. Platelet count estimation on peripheral smear: What should be an acceptable 'multiplication factor'?. J Med Evid [serial online] 2022 [cited 2022 Oct 5];3:119-22. Available from: http://www.journaljme.org/text.asp?2022/3/2/119/355007
| Introduction|| |
In an era of evidence-based medicine, it is essential to explore and document standards required for accurate results in laboratories. Automated haematology analysers are routinely being used worldwide to determine blood counts with acceptable accuracy and precision. However, despite the availability of the most advanced automated analysers in haematology for determining complete blood counts, one needs to keep in mind that platelet count is a parameter which invariably needs manual verification. Therefore, a robust manual method of platelet estimation with good reliability and accuracy is essential. Frequently encountered situations in which a manual verification of platelet counts is done include flagging by automated analyser, in case of thrombocytopenia, platelet clumps and giant platelets.
Different manual methods of platelet count estimation have been proposed and are being used in different laboratories. One such method is the platelet (PLT)/red blood cell (RBC) ratio, estimated on the Peripheral Smear by counting the number of platelets per 1000 RBCs, and the platelet count is then back calculated from the RBC count provided by the electronic blood cell counter.,, The other commonly used method to obtain total platelet count is by counting the average number of platelets per oil immersion field (OIF) and then multiplying it with a multiplication factor, either 15,000 or 20,000.,, It has been found that there is no constancy in this multiplication factor, with different laboratories using a wide range of multiplication factors such as 10,000, 15,000 and 20,000. This leads to a wide variation in the platelet counts estimated on peripheral smear, resulting in high inter-observer variability. Multiplication factor required to obtain correct platelet count has been questioned in the past, and a focussed unbiased study may be helpful in resolving the dilemma. This study aims to validate the ideal multiplication factor for estimating manual platelet counts on peripheral smear, and to identify a manual method having comparable results with the haematology analyser.
| Methods|| |
This was an observational study conducted in a tertiary care hospital in Uttarakhand, India, after approval from the institutional ethics committee. AIIMS/IEC/15/103, dated 16-02-2015. Peripheral smears were prepared from ethylenediaminetetraacetic acid-anticoagulated blood samples of 100 patients. The peripheral smears were made by the standard 'wedge' procedure and stained with Leishman stain. The automated counts were performed by a calibrated and adequately quality-controlled Beckman Coulter LH750 analyser. The manual platelet counts were estimated using a magnus microscope (MLXi plus) having a field diameter of 0.2 mm. Since the automated analyser was being used as the reference method, any flagged sample which pointed towards an inaccurate platelet count (platelet clumps, giant platelet, etc.) was excluded.
Estimate of the total platelet count by the first method (PLT/RBC ratio method) was made by counting the number of platelets/1000 RBCs on OIF examination. The number of RBCs observed in a quarter of OIF was multiplied by four instead of counting all the RBCs in the field. Then, all the platelets in the same field were counted. The total platelet counts were estimated by multiplying the ratio of platelets to the RBCs by the RBC counts derived from the analyser.
Estimate of the platelet count by the second method (PLT ×15,000 method) was made by counting the number of platelets in at least 10 OIFs in an appropriate area (junction of body and tail). The average number of platelets per OIF was calculated and the total platelet count was estimated by multiplying the average number of platelets per OIF by 15,000 to give an approximate count per mm3. Total platelet counts by the third method (PLT ×20,000 method) and fourth method (PLT ×10,000 method), respectively, were done in a similar manner.
The differences between the peripheral smear estimate of platelet count and analyser count were calculated. The per cent differences were also calculated. An acceptable difference in results between the peripheral smear-based estimates and the analyser counts was deemed to be 25% as suggested by Clinical Laboratory Improvement Amendments (CLIA). The number and proportion of peripheral smear estimates of platelet count having acceptable difference from the analyser count were calculated.
The same analysis was then repeated on samples showing platelet counts <100,000/mm3 (low platelet count), samples showing RBC counts <4 million/mm3 (low RBC count), samples showing both platelet counts <100,000/mm3 and RBC counts <4 million/mm3 and samples showing RBC counts between 4 and 6 million/mm3. The analysis was done using R statistical environment with R commander package.,
| Results|| |
The descriptive summary of the differences between the platelet counts estimated by the different methods and the analyser platelet count is depicted in [Table 1]. The percentage differences between the platelet counts estimated by the different methods and the analyser platelet count are given in [Table 2].
|Table 1: Descriptive summary of the differences of the peripheral smear-based estimates and the analyser counts for all cases and different groups|
Click here to view
|Table 2: Descriptive summary of the percentage differences of the peripheral smear-based platelet estimates and the analyser counts for all cases and different groups|
Click here to view
Comparative analysis showed that the PLT ×15,000 method has the highest percentage of overall cases falling within the acceptable error (86%), followed closely by the PLT/RBC method (81%). The percentage of cases falling within the acceptable error was markedly low by the PLT ×100,00 method (25%) and the lowest by the PLT ×20,000 method (18%). In the low platelet count group (<1,00,000/mm3), the accuracy of both PLT/RBC and PLT ×15,000 methods was equal (72.4%) followed by PLT ×10,000 method (58.6%) and least (6.9%) by PLT ×20,000 method. In the low RBC and low platelet counts along with low RBC groups, the accuracy obtained from PLT/RBC method was the highest (83.9% and 80%), respectively, closely followed by the PLT ×15,000 method (80.6% and 73.3%), followed by PLT ×10,000 method (29%, 53.3%), respectively, and lowest in PLT ×20,000 method (9.7%, 0%), respectively. In samples having normal RBC counts, the PLT ×15,000 method has the highest accuracy (88.1%), followed by PLT/RBC method (82.1%), while the PLT ×10,000 and PLT ×20,000 methods reported very low percentages of 23.9% and 20.9%, respectively.
| Discussion|| |
The importance of manual platelet count on peripheral blood smear is reflected by the fact that peripheral smear examination for platelet count estimation remains irreplaceable even in an era of automated haematology analysers. A result of thrombocytopenia based on automated analysers may need to be re-checked and confirmed by peripheral smear examination. This becomes necessary because there are many reasons of pseudo-thrombocytopenia which could be due to analytical or pre-analytical factors such as platelet clumping, satellitism, giant platelets, RBC microcytosis and bacterial contamination. However, due to the inherent subjectivity involved in platelet count in peripheral smear-based platelet estimation, it becomes essential to be assured as to which peripheral smear-based manual method has the potential to give the most accurate results. The current gold standard for platelet estimation is the flow cytometric assessment of platelets using CD41 and CD61. However, this facility may not be available in routine laboratory settings, and hence, the pathologist has to rely on peripheral smears for manual estimation of platelet counts.
Determination of a standard multiplication factor for manual platelet count estimation is essential to obtain an accurate platelet count and to ensure uniformity and objectivity in assessing platelet counts by manual smear-based methods. In the present study, we found the PLT ×15,000 method followed by PLT/RBC ratio method to be the best amongst the various peripheral smear-based methods of assessing manual platelet counts. Different researchers have tried to estimate the platelet count by applying different multiplying factors to the platelet count per OIF. In a study for deriving platelet count, Abbey and Nosanchuk suggested a factor of approximately 20,000 for platelet counts,, while in another study, Webb et al. suggested a multiplication factor of 15,000. In contrast, Moreno and Menko suggested a factor between 15,000 and 20,000, without committing to either.
The PLT/RBC ratio method devised by Brahimi et al. also stands validated by this study; however, this method does require a haematology analyser for determining RBC count. The PLT ×15,000 and PLT/RBC ratio method gave reasonably accurate estimates for all subgroups, thus confirming their utility in laboratory practice. The PLT ×15,000 method may, therefore, prove useful in validating the peripheral smear review of flagged samples. In comparison to the PLT/RBC ratio method, an additional advantage of the PLT ×15,000 method is that it does not require an analyser and can independently assess platelet counts. Therefore, this method can be further utilised in resource-poor settings to accurately assess the platelet count on peripheral smear, without the need of analyser and with reliable and dependable results.
Another manual method of platelet estimation on peripheral smear has been described, which depends on the determination of a conversion factor derived from examination of multiple smears as per a standard protocol which is influenced by the field diameter of the oil immersion lens of the microscope in use. Hence, this factor may be indigenous to that particular microscope on which the calculation was done. Each microscope in use would require calculation of its conversion factor for manual platelet count estimation, which may be a cumbersome process. In addition, the agreeable rate for this method to an analyser is ± 40%, a value which is much higher/lower than the actual platelet count. In view of this, a more feasible and better alternative would be to derive a standard, uniform multiplication factor, applicable to any microscope being used in haematology laboratories worldwide. The PLT × 15,000 method used in this study has an agreeable rate of ± 25% which is as per the CLIA recommendations.
It needs to re-emphasised that in spite of automation ruling the haematology laboratories, importance of manual method cannot be undermined, especially for determining platelet count in many instances. Hence, one needs to pay attention as to which method to adopt to ensure accurate results.
| Conclusions|| |
This study concludes that manual platelet count estimation under OIF, on peripheral smear using a multiplication factor of 15,000, has comparable accuracy to the analyser counts, followed by the PLT/RBC ratio method. This method is simple and easy to apply in any laboratory setting and provides a rapid and accurate estimation of platelet counts on peripheral smear.
The authors would like to thank Mr. Manish Mehta, Technical Assistant, Department of Pathology and Laboratory Medicine, AIIMS, Rishikesh, for his technical help in running the samples in the analyser and expertise in preparing the smears for this study.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Brahimi M, Osmani S, Arabi A, Enta-Soltane B, Taghezout Z, Elkahili BS, et al.
The estimation of platelet count from a blood smear on the basis of the red cell: Platelet ratio. Turk J Haematol 2009;26:21-4.
College of Physicians and Surgeons of Saskatchewan, Laboratory Quality Assurance Program. Laboratory Guidelines; 2014. p. 59-64.
Bain BJ. Detecting erroneous blood couts. In: Blood Cells a Practical Guide. 5th
ed. Chichester, UK: John Wiley and Sons; 2015. p. 186-210.
Abbey AP, Belliveau RR. Enumeration of platelets. Am J Clin Pathol 1978;69:55-6.
Nosanchuk JS, Chang J, Bennett JM. The analytic basis for the use of platelet estimates from peripheral blood smears. Laboratory and clinical applications. Am J Clin Pathol 1978;69:383-7.
Webb DI, Parker L, Webb K. Platelet count assessment from peripheral blood smear (PBS). Alaska Med 2004;46:92-5.
Abid BF. Estimation of platelet count on the basis of red cell: Platelet ratio. Iraqi J Med Sci 2009;7:40-5.
Madison WI. Westgard QC 2019. CLIA Requirements for Analytical Quality. [Online]. 2019 [cited 2021 Jul 16]; [about 5 screens]. Available from: https://www.westgard.com/clia.htm
R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2014.
Fox J. The R commander: A basic statistics graphical user interface to R. J Stat Softw 2005;14:1-42.
Adewoyin AS, Nwogoh B. Peripheral blood film – A review. Ann Ib Postgrad Med 2014;12:71-9.
Pewarchuk W, VanderBoom J, Blajchman MA. Pseudopolycythemia, pseudothrombocytopenia, and pseudoleukopenia due to overfilling of blood collection vacuum tubes. Arch Pathol Lab Med 1992;116:90-2.
Harrison P, Ault KA, Chapman S, Charie L, Davis B, Fujimoto K, et al.
An interlaboratory study of a candidate reference method for platelet counting. Am J Clin Pathol 2001;115:448-59.
Moreno A, Menke D. Assessment of platelet numbers and morphology in the peripheral blood smear. Clin Lab Med 2002;22:193-213.
[Table 1], [Table 2]