Decrease Rate of Platelet Count as a Marker for Complications of Preeclampsia: A Cross-Sectional Study

Thrombocytopenia is considered a sign of maternal organ damage of preeclampsia. However, platelet count distribution during pregnancy shifts downward even in uncomplicated women, and platelet counts in some uncomplicated women falls below the cutoff value of thrombocytopenia. Thus, not only absolute platelet counts number but also platelet count reduction rate may relate with adverse outcomes of preeclampsia. This study aimed to evaluate whether the overall platelet count reduction rate was associated with adverse outcomes of preeclampsia. with outcomes of preeclampsia even in women without thrombocytopenia.


Introduction
Failure of trophoblast invasion and remodeling of spiral arteries followed by imbalance of angiogenic and antiangiogenic factors cause systemic endothelial damage, which results in preeclampsia [1,2].
Injured endothelium can activate platelets, increasing the consumption of platelets and causing thrombocytopenia in preeclampsia [3]. Therefore, thrombocytopenia is considered a sign of maternal organ damage of preeclampsia [4,5].
Platelet count (PC) in pregnant women decreases throughout pregnancy [6], and gestational thrombocytopenia (de ned as PC <150 × 10 9 /L without any other etiology) occurs in 4.4%-11% of pregnancies [7,8]. While gestational thrombocytopenia is not associated with maternal and neonatal complications, severe thrombocytopenia in preeclampsia including hemolysis, elevated liver enzymes, low platelet syndrome increases the risk of adverse maternal and fetal outcomes [9]. Expectant management can be chosen in cases with stable maternal and fetal condition [4,5]. However, thrombocytopenia can get worse during expectant management because preeclampsia is a progressive disease. PC distribution during pregnancy shifts downward even in uncomplicated women, and PC in some uncomplicated women falls below the cutoff value of thrombocytopenia (<150 × 10 9 /L). Therefore, the evaluation of not only the absolute PC value but also the overall reduction rate of PC from the baseline may give further information for predicting adverse outcomes of preeclampsia. This study aimed to examine whether the overall reduction rate of PC from the rst trimester is associated with adverse outcomes of preeclampsia.

Study Population
Women who were diagnosed with preeclampsia and delivered at Hokkaido University Hospital and Japan Community Health Care Organization Hokkaido Hospital between April 2010 and May 2019 were enrolled for this study. Women who were <18 years old at delivery or transferred to other hospitals before delivery were excluded. In addition, women were excluded if their thrombocytopenia was not associated with preeclampsia, the baby had severe congenital anomaly, and data of PC levels in the rst trimester were not available.

Data Collection
Diagnosis of preeclampsia was conducted based on According to the International Society for the Study of Hypertension in Pregnancy and Japan Society for the Study of Hypertension in Pregnancy guidelines, preeclampsia was de ned by the new onset of hypertension (systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg) accompanied by proteinuria (protein/creatinine ratio ≥0.3) and/or a sign of maternal organ damage and utero-placental dysfunction, as listed in Table 1 [4,10]. The results of blood tests conducted during 9-13 gestational weeks (GW) were used as the value of the PC levels at the rst trimester. The overall reduction rate of PC was de ned with the formula: (PC levels at the rst trimester -PC levels at delivery) / PC levels at the rst trimester. The women were divided into the three groups based on the PC levels at delivery: normal (PC > 150 × 10 9 /L), mild thrombocytopenia (≥100 × 10 9 /L and <150 × 10 9 /L), and severe thrombocytopenia (<100 × 10 9 /L).
The incidence of severe hypertension (systolic blood pressure, ≥160 mmHg; diastolic blood pressure, ≥110 mmHg), maternal organ damage, uteroplacental dysfunction, and preterm delivery <34 GW were analyzed. Maternal organ damage and uteroplacental dysfunction were de ned according to Table 1.

Statistical Analyses
Statistical analyses were conducted using Stata/SE version 15.1 (StataCorp, College Station, TX, USA). The normality of the data was analyzed using histograms in terms of skewness and kurtosis. Continuous data are reported as mean ± standard deviation. Categorical variables are expressed as frequency and percentage. Statistical signi cance was calculated using independent t-test and chi-square test for continuous data and categorical variables, respectively. PC decrease is part of the natural course of pregnancy and 10%-20% reduction is observed in noncomplicated pregnancy [6,11]. Therefore, a 30.0% reduction was used as the cutoff value of overall reduction of PC decrease from the rst trimester. Odds ratio calculated with logistic regression analysis often overestimates relative risk (RR) for common outcomes (incidence >10.0 %) [12]. Therefore, the Poisson regression analysis with a robust error variance was used to estimate the RR and 95% con dence interval (CI).

Ethical Approval
The institutional review board of Hokkaido University Hospital (019-0070) and Japan Community Health Care Organization Hokkaido Hospital (2020-4) approved this study. Hokkaido University Hospital Clinical Research Administration Center and Japan Community Health Care Organization Hokkaido Hospital Ethics Review Bord did not require informed consent for this retrospective study. Information about this study was placed on the home pages of Hokkaido University Hospital and Japan Community Health Care Organization Hokkaido Hospital with the opportunity to opt out. All methods were carried out in accordance with the Declaration of Helsinki.

Results
During the study period, 329 women were diagnosed with preeclampsia. Of these women, 89 ful lled the exclusion criteria and were not included in the analysis. Therefore, 240 women were evaluated in this study. Among them, 44 (18.3%) of women developed mild thrombocytopenia and 22 (9.2%) of women developed severe thrombocytopenia (Fig 1). In addition, 95/240 (39.6%) of women had ≥30.0% of overall platelet count reduction from the rst trimester.
Clinical characteristics based on the overall reduction rate of PC from the rst trimester in women with preeclampsia are shown in Table 2. PC in the rst trimester was similar between the two group (<30.0% of reduction, ≥30.0% of reduction; 249 ± 53 × 10 9 /L, 254 ± 59 × 10 9 /L, p = 0.442). At delivery, 95/240 (39.6%) of women had a ≥30.0% of overall PC reduction from the rst trimester and developed mild and severe thrombocytopenia more frequently than those with <30.0% of overall PC reduction (p <0.001). The incidences of maternal organ damage except for thrombocytopenia (12.4%, 22.1%; p = 0.047), and preterm delivery <34 GW (30.3%, 48.4%; p = 0.005) were also higher in women with ≥30.0% of overall PC reduction than those otherwise. The RR of ≥30.0% of overall PC reduction for the adverse outcomes in adjusting maternal age are presented in Table 3. A ≥30.0% of overall PC reduction was signi cantly associated with the increased risk of preterm delivery <34 GW (RR, 1.58; 95% CI, 1.14-2.18).

Discussion
This study revealed that the overall reduction rate of PC was associated with the adverse outcomes of preeclampsia, and it was also observed in women without thrombocytopenia. Close monitoring of both mother and baby is essential especially after developing preeclampsia [4,5]. Although serum biomarkers, such as the ratio of soluble fms-like tyrosine kinase-1 (sFlt-1) to placental growth factor (PlGF) have a predictive value for the adverse outcomes of preeclampsia [13], these biomarkers are not widely available especially in developing countries. On the other hand, a complete blood count is a more globally accessible test. Therefore, assessing the risk of adverse outcomes of preeclampsia with PC has clinical bene ts.
Physiological changes during pregnancy decrease PC. Hemodilution caused by increased plasma volume contributes to lower PC during pregnancy [14]. In nonpregnant women, one-third of circulating platelets are trapped within the splenic sinusoids [15]. The size of the spleen increased by 50% during pregnancy [17], and the placental circulation is similar to the splenic circulation [18]. Because of these changes, PC gradually decrease by 17% at the time of delivery from a nonpregnant situation in uncomplicated pregnancy [6]. This study used PC in the rst trimester as a baseline because PC in nonpregnant situation is often unavailable in clinical practice. The PC level at the rst trimester is below the level of nonpregnant women [6]. Therefore, the overall reduction of PC ≥30.0% from the rst trimester exceeds the physiological change. Thus, ≥30.0% of overall PC reduction may be a sign of elevated consumption of the platelets caused by endothelial damage with preeclampsia. In addition, 7.6% of women with <30.0% of overall PC reduction developed mild thrombocytopenia in this study, and this rate is compatible with the incidence of gestational thrombocytopenia (4.4%-11%) in all pregnancy [7,8]. This nding implies that thrombocytopenia in women with <30.0% of overall PC reduction is not associated with preeclampsia.
A previous study reported that little variation of the PC reduction existed in women with uncomplicated pregnancy. Therefore, women whose PC are at the lower end of the normal range before pregnancy tend to have thrombocytopenia during pregnancy [6]. Actually, the recurrent risk of gestational thrombocytopenia was 14.2 times higher in women with a history of gestational thrombocytopenia than those who have no history of gestational thrombocytopenia [6]. In this study, a ≥30.0% of overall PC reduction was associated with severe hypertension and preterm delivery <34GW even in the women whose PC levels were within normal limits. This may be because the PC levels of these women before pregnancy were at the higher end of the normal range. Thus, their PC did not reach the cutoff value for thrombocytopenia although the PC reduction was a result of elevated consumption of the platelets caused by endothelial damage with preeclampsia.
Uteroplacental dysfunction in this study was negatively correlated with the overall reduction of PC. Inadequate trophoblast invasion and following uteroplacental malperfusion cause both preeclampsia and fetal growth restriction [18,19]. However, not all cases of fetal growth restriction are associated with preeclampsia. It is di cult to assess the effect of preeclampsia on fetal growth restriction because the etiology of fetal growth restriction varies [20]. In this study, hypertensive women with fetal growth restriction were classi ed as preeclampsia even if they did not have proteinuria or sign of maternal organ damage. The symptoms of preeclampsia may not be severe in some of those women because fetal growth restriction was not a result of preeclampsia; therefore, those women less likely to develop a ≥30.0% of overall PC reduction. This might be a reason why uteroplacental dysfunction was observed more frequently in women with < 30.0% of overall PC reduction in this study.
This study has several limitations. First, this is a retrospective study and thrombocytopenia is a sign of maternal organ damage. Thus, the incidence of preterm delivery can be biased. However, the association between adverse outcomes of preeclampsia and overall reduction of PC was also detected in women without thrombocytopenia. Second, PC in the rst trimester may be affected by hemoconcentration because of hyperemesis gravidarum, and the severity of hyperemesis gravidarum could not be assessed from the medical records in this study. However, PC in the rst trimester was not different based on the overall reduction of PC. Thus, hyperemesis gravidarum would not affect the results of this study. Finally, sFlt-1 and PlGF were not widely available in Japan during the study period. Therefore, the relationship between the overall reduction of PC and these biomarkers could not be analyzed.

Conclusion
This study implies that a decrease rate of PC was an independent risk factor of the complications of preeclampsia irrespective of PC levels despite these limitations. Women who develop a ≥30.0% PC decrease need to be considered high risk for developing complications of preeclampsia. Abbreviations CI: con dence interval Organization Hokkaido Hospital Ethics Review Bord did not require informed consent for this retrospective study. Information about this study was placed on the home pages of Hokkaido University Hospital and Japan Community Health Care Organization Hokkaido Hospital with the opportunity to opt out. All methods were carried out in accordance with the Declaration of Helsinki.

Consent for publication
Not required

Competing interests
The authors declare that they have no competing interests.

Funding
This research did not receive any speci c grant from funding agencies in the public, commercial, or notfor-pro t sectors.

Availability of data and materials
All data generated or analysed during this study are included in this published article as a supplementary information le.

Author contributions
All listed authors meet criteria for authorship. MM1 made substantial contributions to conceptualization, design of the work, and drafting the work. MM2 made substantial contributions to conceptualization, the acquisition, and the analysis. TY and TU made substantial contributions to interpretation of data. KN1 and KN2 made substantial contributions to the acquisition, and the analysis. YS and KC made substantial contributions to design of work. HW made substantial contributions to drafting and revising work. All authors approved the submitted version and agreed to be personally accountable for their contributions to the work.  PC levels, Normal, >150 × 10 9 /L; mild thrombocytopenia, < 150 × 10 9 /L and ≥100 × 10 9 /L; severe thrombocytopenia, < 100 × 10 9 /L   Figure 1 Schematic illustration of the patient selection criteria

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. dataBMC.csv