Cerebrospinal fluid white blood cell counts: technical observations

S. Gunti and colleagues have recently published an article entitled “Reference range for cerebrospinal fluid values in neonates: 5-year retrospective study” in The Journal of Maternal-Fetal & Neonatal Medicine [1]. I congratulate the authors on their study, which adds our understanding of the cytologic and biochemical features of this body fluid. I would like to add some technical details about the challenges inherent to the analysis of white blood cells from the cerebrospinal fluid (CSF), which I hope will add some useful context to this discussion. My comments would be of interest to the readership of The Journal of Maternal-Fetal & Neonatal Medicine who are considering the extent to which the findings of the study by Gunti et al. apply to their clinical practice. Studies that use clinical laboratory data are often lacking in detail regarding the preanalytical and analytical phases of their testing environments, which make the findings of studies potentially difficult to generalize. This is not surprising, since checklists such as the Standards for Reporting Diagnostic accuracy studies (STARD) do not mandate the reporting of preanalytical factors, despite our knowledge that preanalytical and analytical phase variables are responsible for a majority of laboratory errors [2]. Potential preanalytical sources of bias include delays between specimen collection and analysis, ambient temperature, transport (including pneumatic tube transportation), and the decision about whether to use a preservative agent. This last factor may be of particular importance for CSF specimens, since these specimens are generally collected in tubes that do not contain Roswell Park Memorial Institute (RPMI) medium or other agents in order to facilitate cell counting. This practice makes delays between specimen collection and analysis critical, since delays as brief as 60min have been shown to result in significant decreases in CSF white blood cell (WBC) count [3]. Another common problem encountered in clinical specimens is hemodilution, which can alter cell counts and make it difficult or impossible to determine whether abnormal cells originate from the peripheral blood or CSF [4]. In addition, the choice of analytical platform has major implications for the sensitivity of detection and imprecision of rare event detection in low cell count body fluids, including the counting of WBCs in the CSF [5]. The gold standard method for CSF cell counting, manual microscopy with cell enumeration using a Fuchs-Rosenthal or other type of counting chamber, is labor-intensive and is impractical for high-throughput laboratories [6]. For this reason, semi-automated and automated approaches have been developed, including 1) dedicated body fluid cell counting instruments, 2) complete blood cell count analyzers operating in “body fluid” mode 3) flow cytometry, and 4) specialized instruments for low cell count fluids (Table 1) [4,6]. Each of these technologies has associated advantages and disadvantages related to 1) cost of equipment; 2) volume of specimen necessary to perform in low cell count environments; and 3) technologists with sufficient expertise to execute these complex tasks. Based on the extant literature, dedicated CSF cell counting instruments appear to have the best performance (lowest degree of imprecision) [7]. Although not routinely used for CSF cell counting, flow cytometry methods including next-generation flow cytometry and measurable residual disease testing may be a promising approach in low cell count environments such as the CSF [8]. In addition to the aforementioned preanalytical and analytical phase variables, another obvious problem with CSF analysis is the limited sensitivity of manual microscopy, the “gold standard” for detection of abnormal cells [4,9]. In closing, I thank the authors for their valuable contribution to the CSF literature. For the reasons I have mentioned, I would welcome a response from them regarding the preanalytical and analytical phase variables (including the instrumentation platform) from their study. These additions would add useful context to their work and will allow clinicians and researchers to more easily evaluate their important efforts.

S. Gunti and colleagues have recently published an article entitled "Reference range for cerebrospinal fluid values in neonates: 5-year retrospective study" in The Journal of Maternal-Fetal & Neonatal Medicine [1]. I congratulate the authors on their study, which adds our understanding of the cytologic and biochemical features of this body fluid. I would like to add some technical details about the challenges inherent to the analysis of white blood cells from the cerebrospinal fluid (CSF), which I hope will add some useful context to this discussion. My comments would be of interest to the readership of The Journal of Maternal-Fetal & Neonatal Medicine who are considering the extent to which the findings of the study by Gunti et al. apply to their clinical practice.
Studies that use clinical laboratory data are often lacking in detail regarding the preanalytical and analytical phases of their testing environments, which make the findings of studies potentially difficult to generalize. This is not surprising, since checklists such as the Standards for Reporting Diagnostic accuracy studies (STARD) do not mandate the reporting of preanalytical factors, despite our knowledge that preanalytical and analytical phase variables are responsible for a majority of laboratory errors [2]. Potential preanalytical sources of bias include delays between specimen collection and analysis, ambient temperature, transport (including pneumatic tube transportation), and the decision about whether to use a preservative agent. This last factor may be of particular importance for CSF specimens, since these specimens are generally collected in tubes that do not contain Roswell Park Memorial Institute (RPMI) medium or other agents in order to facilitate cell counting. This practice makes delays between specimen collection and analysis critical, since delays as brief as 60 min have been shown to result in significant decreases in CSF white blood cell (WBC) count [3]. Another common problem encountered in clinical specimens is hemodilution, which can alter cell counts and make it difficult or impossible to determine whether abnormal cells originate from the peripheral blood or CSF [4].
In addition, the choice of analytical platform has major implications for the sensitivity of detection and imprecision of rare event detection in low cell count body fluids, including the counting of WBCs in the CSF [5]. The gold standard method for CSF cell counting, manual microscopy with cell enumeration using a Fuchs-Rosenthal or other type of counting chamber, is labor-intensive and is impractical for high-throughput laboratories [6]. For this reason, semi-automated and automated approaches have been developed, including 1) dedicated body fluid cell counting instruments, 2) complete blood cell count analyzers operating in "body fluid" mode 3) flow cytometry, and 4) specialized instruments for low cell count fluids (Table 1) [4,6]. Each of these technologies has associated advantages and disadvantages related to 1) cost of equipment; 2) volume of specimen necessary to perform in low cell count environments; and 3) technologists with sufficient expertise to execute these complex tasks. Based on the extant literature, dedicated CSF cell counting instruments appear to have the best performance (lowest degree of imprecision) [7]. Although not routinely used for CSF cell counting, flow cytometry methods including next-generation flow cytometry and measurable residual disease testing may be a promising approach in low cell count environments such as the CSF [8].
In addition to the aforementioned preanalytical and analytical phase variables, another obvious problem with CSF analysis is the limited sensitivity of manual microscopy, the "gold standard" for detection of abnormal cells [4,9].
In closing, I thank the authors for their valuable contribution to the CSF literature. For the reasons I have mentioned, I would welcome a response from them regarding the preanalytical and analytical phase variables (including the instrumentation platform) from their study. These additions would add useful context to their work and will allow clinicians and researchers to more easily evaluate their important efforts.

Funding
The author(s) reported there is no funding associated with the work featured in this article.