Mean platelet volume vs. optical platelet volume measurement: which is a better indicator for neonatal sepsis?

Introduction

Sepsis is a systemic inflammatory response syndrome caused by pathogenic microorganisms such as bacteria. Critically ill patients often experience insufficient organ perfusion in addition to systemic inflammatory response and primary infectious lesions [1]. With over 18 million severe sepsis cases globally each year, and an annual rise of 1.5–8.0 %, the condition remains highly prevalent [2]. In China, prospective multicenter research reported a sepsis incidence of 77.6 % (843/1,087 cases), and a septic shock rate of 38.7 % (423/1,087 cases) [3]. Due to undeveloped immune systems of newborns, neonatal sepsis (NS) poses higher risks, particularly among very low birth weight infants and premature infants, ranging from 1 to 10 in every 1,000 neonates and 1–70 %, respectively [4].

Confirmation of NS typically relies on body fluid cultivation, which is time-consuming and prone to false negatives, limiting its clinical utility [5]. Other diagnostic methods include white blood cell counting and inflammatory markers such as C-reactive protein (CRP). While CRP is effective for ruling out infections, it has limited sensitivity in the early stages of infection [6].

Platelet-related parameters in complete blood count, such as platelet (PLT) count, mean platelet volume (MPV), and plateletcrit (PCT), are measured using the Sysmex XN9000 analyzer [7]. Previous studies have reported a close relationship between MPV and NS, with MPV potentially indicating the occurrence, progression, and recovery of the condition [8]. As NS worsens, platelet volume increases. However, MPV is not an ideal marker due to limitations with traditional impedance method. On blood analyzer, traditional impedance method may misclassify small red blood cells (RBCs), RBC debris, and varying-sized platelets, leading to inaccurate MPV measurement [9]. In neonates, Capillary blood and the high production of macroplatelets during infections further complicate accurate MPV measurement. Our study of 1,070 infectious newborns found that MPV could not be measured in 24.67 % of cases, with alarm information in PLT parameters indicating unavoidable issues. Of 479 cases with undetermined MPV values, 76 had “PLT aggregation”, 118 had “thrombocytopenia” and 285 had “unequal size of RBC” (Table 1), with some histograms showing alarms despite MPV values, which were unreliable. These findings highlight the limitations of current MPV detection methods and suggest a need for more reliable approaches for measuring neonatal platelet volume.

Since 2000s, a novel method for measuring PLT count, known as Fluorescence PLT (PLT-F), has been invented and applied on Sysmex XN9000 blood analyzer. This technique utilizes a new fluorescent dye with high specificity, staining only platelets and minimizing interference from reticulocytes and erythrocytes [10]. The platelet count is still determined using flow cytometry, where the X-axis of PLT-F histogram represents side scatter (SSC), and the Y-axis represents forward scattered light (FSC). Since FSC reflects particle volume, we are wondering whether the Y-axis value of PLT-F histogram (PLT-F-Y) can serve as an indicator of platelet volume and aid in predicting the onset, progression, and recovery of NS.

Materials and methods

Results

There was a correlation between PLT-F-Y and neonatal sepsis

In cases of neonatal sepsis, different stages of the disease were monitored during hospitalization. Weekly measurements of MPV, PLT-F-Y, and CRP were taken. A strong correlation was observed between PLT-F-Y values and the severity of neonatal sepsis. PLT-F-Y values increased with worsening infection severity and decreases as recovery progressed, aligning with previous studies (Figure 1) [8]. These results indicate that PLT-F-Y closely correlates with the stages of NS.

Figure 1:

Four cases of neonatal sepsis were collected. PLT-F-Y and CRP showed the same trend of change. As the value of CRP increased, PLT-F-Y increased. As the condition gradually improved, both CRP and PLT-F-Y decreased simultaneously.

In addition, blood samples were collected from neonates without sepsis to serve as a control group, while neonates with NS were designated as experimental group. Measurements of MPV and PLT-F-Y were taken for both groups. Compared to experimental group, values of PLT-F-Y were lower in control group, which was consistent with MPV changing trend (Figure 2). Both PLT-F-Y and MPV were effective indicators for predicting and monitoring neonatal sepsis. However, there was a significant issue with MPV detection, with a 40 % failure rate (Table 4). This suggests that PLT-F-Y may provide a more stable and reliable measurement.

Figure 2:

Comparison of patients and control group in values of PLT-F-Y and MPV. Values of PLT-F-Y and MPV were significantly higher in patient group.

Table 4:

Comparison of the probability that two indicators cannot be detected in the four cases.

Indicator	Having results	No result	Proportion (no result)
MPV	23	16	41.02 %
PlT-F-Y	39	0	0.00 %

1. The proportion of the former was much higher than the latter. PLT-F-Y was always detectable on Sysmex XN-9000.

We investigated whether PLT-F-Y could predict the prognosis in patients with NS by comparing values in discharge and during follow-up visits. Our analysis revealed that patients with higher PLT-F-Y values at discharge had more frequent follow-up visits compared to those with lower PLT-F-Y values (Figure 3, p<0.05). Unfortunately, a patient with persistently high PLT-F-Y levels ultimately passed away, suggesting that elevated PLT-F-Y level is associated with poor prognosis. These findings indicate that PLT-F-Y is a valuable indicator for predicting the prognosis of NS.

Figure 3:

Relationship between PLT-F-Y and prognostic visits. The higher the PLT-F-Y value at discharge was the more prognostic visit times patient suffered.

Observation of platelet sizes in different stages

Blood samples from the “before treatment”, “in treatment” and “post-treatment” stages were collected. Smears were prepared, stained by Wright-Giemsa method, and examined with an automatic smear. Results showed that PLT sizes were larger before treatment. As treatment progressed, platelet volumes decreased, confirming that PLT volume reduced with the improvement of NS (Figure 4).

Figure 4:

PLT size observation at different stages of neonatal sepsis. As the treatment progressed, the volume of PLT decreased from large to normal.

Discussion

Neonatal sepsis is a critical condition that poses a severe threat to newborns, making early assessment essential. Although there are several diagnostic methods for neonatal sepsis, many are time-consuming [2]. Mean platelet volume is fundamental indicator in blood routine test and has been shown to correlate well with neonatal sepsis. Previous research indicates a significant increase in MPV values in infants with sepsis [12], suggesting that MPV may be a potential indicator for predicting the disease. However, MPV measurements can be unreliable due to factors such as uneven red blood cell sizes, the presence of immature red blood cells, or severe infection, which may interfere with accurate readings by blood analyzers. This underscores the need for alternative assessment method. The PLT-F-Y channel on the Sysmex XN9000 represents a novel approach for PLT determination, utilizing nucleic acid fluorescence staining to specifically differentiate platelets from other blood components. This method provides a scatter diagram on the analyzer, where the Y-axis is the set of forward scattered light, reflecting the average size of PLT [13]. Given this, PLT-F-Y values could potentially serve as an indicator for measuring MPV. To determine whether the value of PLT-F-Y was increasing or decreasing, a normal range of PLT-F-Y was established.

Our study demonstrated that as the severity of infection worsened, the PLT-F-Y value increased. This change can be attributed to the alterations in the coagulation cascade characteristic of sepsis. During sepsis, multiple pro- and anti-inflammatory cytokines are released, which can lead to thrombus formation and the subsequent depletion of fibrinolytic factors and fibrinogen, resulting in further platelet destruction [14]. Additionally, P-selectin, which is expressed on the surface of platelets during systemic inflammation, enhances platelet adhesion to white blood cells and promotes platelet aggregation [15]. In neonatal sepsis, platelet depletion and thrombus formation due to active endothelium contribute to thrombocytopenia [16]. Meanwhile, CRP, which peaks around 50 h, aids in complement binding to foreign or damaged cells, playing a role in the inflammatory response. Although a non-specific marker, CRP can be a help in diagnosing sepsis, which is consistent with our findings [17]. Furthermore, Platelet activation is a significant physiological feature of NS. The increased production of young, larger platelets in the bone marrow leads to elevated PLT-F-Y values on the Sysmex XN9000, indicating its potential to predict the onset and progression of NS.

As mentioned above, MPV measurements in neonates can sometimes be challenging in our laboratory. One reason is the sample type; neonatal samples are often collected from capillary blood, which can be contaminated with tissue fluid due to compression. This contamination may lead the analyzer to misidentify tissue fluid components as platelets, thus affecting the accuracy of platelet-related indicators like MPV [18]. Additionally, anemia in neonates can further impact MPV readings. Overall, multiple factors can influence the accuracy of MPV detection.

The prognosis of NS is a critical concern for clinicians. Our study found that patients with higher PLT-F-Y values at discharge required more frequent follow-up visits, and a patient with persistently high PLT-F-Y levels unfortunately passed away. This observation aligns with previous research indicating that more severe NS is associated with higher MPV values. Therefore, PLT-F-Y may be able to provide a valuable prognostic indicator. If a patient’s PLT-F-Y value exceeds the reference upper limit at discharge, clinician should exercise heighted vigilance. Overall, platelet volume appears to have a strong correlation with the prognosis of NS.

In our study, we noted an unusual case where, despite the progression of sepsis and consistently low platelet counts on different days, values of MPV remained within the normal range. Interestingly, the PLT-F-Y values were consistent with the MPV values. As the CRP values decreased, PLT-F-Y and MPV values elevated. We hypothesized that after the resolution of infection, the platelet production function in the bone marrow recovered well, leading to the recovery of PLT count, resulting in the increase of MPV.

The study still has some limitations. Firstly, observational studies are influenced by confounding variables, which may affect true overall effect. Secondly, there is an incomplete data on some patients’ treatment processes, thus limiting our ability to further track the mortality and recurrence rates. Lastly, relationship between PLT-F-Y and other infectious diseases may be discussed in the future.