![The World Congress on Controversies in Obstetrics, Gynecology and Infertility (COGI). Click here to learn more.]({"type":"image" , "src" : "/sda/112382/JournalAd-COGI2024-Leaderboard.jpg"})
https://orcid.org/0000-0002-8413-2064
Abstract
The aim of this retrospective study was to demonstrate the effectiveness of APRI, DNI, NLR, PLR, and PDW in predicting the severity of gestational hypertension (GHT) and PE and to determine whether these factors can be used as screening tools. Normotensive pregnant women (n = 792) served as the control group. 1,213 single pregnant women who met the following criteria for a GHT diagnosis were included in the study group. We found a significantly higher mean PLR and NLR value. The mean PDW value was significantly lower in the control group than in the other groups. The SPE group had a significantly higher mean APRI score. The groups did not differ by their DNI. We determined PDW and APRI as independent parameters that predicted SPE by multiple logistic regression analysis. In retrospective analysis of blood samples taken from these participants below week 20, we found that the APRI value differed significantly between the control and SPE groups. NLR, PLR, DNI, and PDW had no clinical significance. We further suggested that APRI may provide a clinical indication of progression from hypertensive pregnancy disorders to SPE, which seems to be a promising implication that should be verified by further studies.
What is already known on this subject? Hypertensive disorders in pregnancy are a major cause of maternal and perinatal morbidity and mortality. Screening pregnant women for risk factors for developing hypertensive disorders and identifying women at high risk in early pregnancy and initiating prophylactic treatment are important for pregnancy monitoring and planning in experienced centres. Because only 30% of women who will develop preeclampsia can be predicted by risk factors, the combined use of laboratory tests and imaging with risk factors to calculate a woman’s risk of developing preeclampsia is currently being investigated. However, no proven marker has yet been found.
What do the results of this study add? In our study, we found that NLR, PLR, DNI, and PDW have no clinical significance in assessing the risk of developing gestational hypertension and preeclampsia and in predicting the severity of preeclampsia. However, in our study, we found that APRI can provide a clinical indication of the progression of hypertensive pregnancy to SPE.
What are the implications of these findings for clinical practice and/or further research? This study represents an important contribution to the literature because it is the first study to examine the association between APRI and HT in pregnancy.
IMPACT STATEMENT
Introduction
Hypertensive disorders in pregnancy are a major cause of maternal and perinatal morbidity and mortality. Hypertensive disorders can occur at any stage of pregnancy and occur in approximately 10% of all pregnancies (Metz et al. Citation2020). Gestational hypertension is defined as systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg at two measurements 4 hours apart at 20 weeks of gestation in a previously normotensive woman (Metz et al. Citation2020). Preeclampsia is a multisystemic disorder occurring after 20 weeks of gestation and characterised by new-onset hypertension and concomitant disorders of various end organ disorders with or without new-onset proteinuria (Metz et al. Citation2020). Preeclampsia with severe features is diagnosed when SBP ≥160 mmHg and/or DBP ≥110 mmHg are present in pregnant women with diagnosed gestational hypertension and/or in the presence of signs or symptoms of end-organ damage (Metz et al. Citation2020). These patients are associated with a high risk of morbidity and mortality. Screening pregnant women for risk factors for preeclampsia (PE) and identifying women at high risk of developing the disease in early pregnancy are critical for initiating prophylactic treatment, monitoring pregnancy, and planning delivery in experienced centres. Because risk factors predict only 30% of women who would develop PE, the combined use of laboratory tests and imaging with risk factors to estimate a woman’s risk of developing PE is still under investigation (Leslie et al. Citation2011). However, an established marker for PE has not yet been found (Bibbins-Domingo et al. Citation2017, Engin-Üstün et al. Citation2019, Townsend et al. Citation2019).
Leukocyte counts are more elevated in pregnant women with PE than in normal pregnancy, mainly due to an increased number of neutrophils as opposed to a decreased number of leukocytes (Canzoneri et al. Citation2009). As a result of these changes, the neutrophil-to-lymphocyte ratio (NLR) is expected to be increased in PE compared with normal pregnancies (Lurie et al. Citation1998, Canzoneri et al. Citation2009). The delta- neutrophil index (DNI) is a new index that indicates the proportion of circulating immature granulocytes (Pyo et al. Citation2013). It has been shown that the increase in immature granulocytes in peripheral blood is of prognostic and diagnostic importance in infections and sepsis (Pyo et al. Citation2013). This may suggest a significant link between the pathophysiology of PE and DNI (Laresgoiti-Servitje et al. Citation2013). Endothelial dysfunction in PE leads to increased platelet consumption in the maternal peripheral circulation and uncontrolled intravascular platelet activation. Several studies reported a significant decrease in platelet count and increase in mean platelet volume (MPV) and platelet distribution width (PDW) in pregnant women who developed PE compared with pregnant women who did not (Yang et al. Citation2014). Other studies suggest that platelet-to-lymphocyte ratio (PLR) and PDW can predict PE (Roberts and Bell Citation2013, Kanat-Pektas et al. Citation2014).
Aspartate aminotransferase (AST) is the predominant transaminase released into the peripheral circulation in liver dysfunction due to PE and is associated with periportal necrosis (Metz et al. Citation2020). AST Platelet ratio index (APRI) is a non-invasive indicator of inflammation in liver fibrosis. It has been reported that APRI level predicts HELLP syndrome better than AST alone (Wai et al. Citation2003, Loaeza-del-Castillo et al. Citation2008, Şaşmaz et al. Citation2020). DNI, NLR, PLR, PDW, and APRI parameters have recently become popular because of their low cost and ease of calculation by routine blood tests.
In this study, we aimed to demonstrate the efficacy of APRI, DNI, NLR, PLR, and PDW in predicting the severity of gestational hypertension (GHT) and PE and to determine whether these factors can be used as screening tools.
Materials and methods
This study followed the Declaration of Helsinki on Research Involving Human Subjects and was approved by the Ethics Committee of Etlik Zubeyde Hanim Women’s Health Training and Research Hospital, of the University of Health Sciences (date: 06/17/2020, approval number: 77).
In this retrospective study, we examined and classified pregnant women admitted to the delivery room or perinatology outpatient clinic of Etlik Zubeyde Hanim Women’s Health Training and Research Hospital of the University of Health Sciences between January 2017 and May 2020 who were normotensive (control) or had GHT, PE, or severe PE (SPE). The database and medical records included 2,702 patients (hypertensive, n = 1802; normotensive, n = 900) who became pregnant between the ages of 16 and 45 years and had a singleton pregnancy. A total of 697 patients with multiple pregnancies, systemic or chronic diseases (hypertension, diabetes, hepatic diseases, cardiac diseases, rheumatic disease, kidney disease, etc.), peripartum complications, or foetal anomalies who were taking corticosteroid therapy or were active smokers were not included in the study (hypertensive, n = 589; normotensive, n = 108). 1,213 single pregnant women who met the following criteria for a diagnosis of preeclampsia were included in the study group: Blood pressure ≥140/90 mm Hg on two occasions at least 4 hours apart or ≥160/110 mm Hg measured at a shorter interval; pregnant women who were normotensive before 20 weeks of gestation; and presence of proteinuria ≥300 mg in a 24–hour urine sample and/or end-organ dysfunction according to the ACOG 2020 Guidelines (Bateman et al. Citation2012). Normotensive pregnant women (n = 792) who met this criterion were included as a control group by randomising them according to use on odd-numbered days of the week. After the demographic and laboratory results were collected in this group, a retrospective analysis was performed and the results of participants who had blood sampling below the 20th week of gestation were analysed (). Blood parameters were studied in two different periods. In general, the blood parameters measured during the patients’ hospitalisation were evaluated as the parameters at the time of diagnosis. On the other hand, the patients were evaluated based on the blood parameters measured in the 20th week.
Figure 1. Flow diagram of the study.
Data was collected on age, gravida, body mass index, and blood pressure (systolic, diastolic, and mean arterial pressure) of the patients. The complete blood count and biochemical parameters of the patients were obtained from the records before the 20th week of gestation and on the day of hospitalisation. Complete blood count parameters were determined using the ADVIA 2120i (Siemens Healthcare) and biochemical parameters were determined using the Beckman Coulter AU680 and AU480 devices. APRI was calculated using the following equation: (AST (IU/L)/normal upper limit of AST (IU/L))/patient’s platelet count (109/L) × 100 [18]. NLR was calculated by dividing neutrophil count by lymphocyte count, as was PLR by dividing platelet count by lymphocyte count. DNI was calculated according to the following equation: Neutrophils (%) + Eosinophils (%) – polymorphonuclear neutrophils (%) [Cho et al. Citation2017, Mannaerts et al. Citation2019]. PDW was recorded as the value obtained from complete blood count analysis.
Statistical analysis
Data analysis was performed using SPSS (statistical package for social sciences, Chicago, IL, USA) 22.0 software. Descriptive statistics were presented as mean ± standard and median (smallest to largest value) for continuous variables and as numbers and percentages for categorical variables. The distribution of parameters was assessed by Shapiro-Wilk normality tests. The difference between repeated measurements in independent groups was evaluated by the Wilcoxon test. For normally distributed data, the independent-samples t test was used, and for non-normally distributed variables, the Mann Whitney U test was used. For comparison of independent categorical variables, the Chi-square test or Fisher’s exact test was used. For comparison of continuous data between more than two independent groups, the ANOVA test was used. Receiver operating characteristics (ROC) analyses were used to determine the appropriate cut-off point for the independent markers and to calculate sensitivity and specificity. The parameters studied (NLR, PLR, PDW, DNI, and APRI) were categorised according to their cut-off values, which were determined by ROC analysis. All statistical analyses were performed with two weights and a p-value of <0.05 was considered statistically significant.
The sample size was calculated by power analysis based on the previous study (Cho et al. Citation2017). With a power of 2.1491714 and an α-value of 0.05 in the independent samples t-test, the power (1-β) was calculated to be 0.95 with 56 participants. The number of participants was above the required sample size; we felt that the power of our analysis was not compromised.
Results
We identified 641 patients with GHT, 208 patients with PE, and 364 patients with SPE. Baseline characteristics of the study groups showed significantly lower mean age (27.0 ± 5.3 years) and BMI (28.4 ± 4.1 kg/m2) in the control group compared with patients in the two hypertensive groups. The mean BMI of the two hypertensive groups was similar. The hypertensive groups were hospitalised significantly earlier than the control group (p < 0.001), although the weeks of gestation at admission did not differ between the PE (34.8 ± 4.0 weeks) and SPE (35.1 ± 3.9 weeks) groups (p = 0.48). We found significantly higher SBP and DBP in the SPE group than in the other groups (p < 0.001) ().
Table 1. Comparison of the study groups by their general demographic and clinical characteristics.
Mean NLR at hospitalisation was significantly higher in the control (4.51 ± 1.89) and SPE (4.57 ± 3.03) groups than in the GHT (4.16 ± 1.63, p = 0.007 and p = 0.012, respectively) and PE (3.85 ± 1.56, p < 0.001 for each pairwise comparison) groups, with no difference between the GHT and PE groups or the control and SPE groups. We found a significantly lower mean PLR in the SPE group (121.6 ± 52.9) compared to that of the control (132.5 ± 59.5, p = 0.007) and GHT groups (136.6 ± 47.1, p < 0.001). Mean PDW was significantly lower in the control group (54.2 ± 7.4) than in the GHT (55.3 ± 8.3) and SPE (56.7 ± 8.8) groups (p = 0.046 and p < 0.001, respectively). The SPE group had a significantly higher mean APRI (1.41 ± 9.36) than the control group (0.23 ± 0.09, p < 0.001), the GHT group (0.23 ± 0.11, p < 0.001), and the PE group (0.24 ± 0.15, p = 0.004). The groups did not differ according to their DNI (p = 0.125), ().
Table 2. Comparison of the study groups in terms of tested laboratory parameters at diagnosis.
Using the ROC analysis curves, the cut-off values with the highest specificity and sensitivity were calculated, and the factors predicting GHT compared to the control group were assessed univariately (Table SI (Supplementary Material)). When multiple logistic regression analysis was performed with statistically significant parameters, PDW was found to be the only independent factor associated with GHT (OR 1.28, 95% CI: 1.02–1.61; p = 0.032) (Table SII, ). When comparing the control and PE groups using ROC analysis, NLR was the only parameter that predicted PE (OR, 0.56, 95% CI: 0.41–0.76; p < 0.001), (Table SIII (Supplementary Material), ).
Figure 2. ROC analysis curve for PDW between the control and GHT groups.
Figure 3. ROC analysis curve for NLR between the control and GHT groups.
ROC analysis was performed between the control and SPE groups, and factors that predicted SPE were evaluated univariately (Table SIV (Supplementary Material)). We determined PDW and APRI as independent parameters that predicted SPE by multiple logistic regression analysis (Table SV, Figure S1 (Supplementary Material)).
Using laboratory values before 20 weeks of gestation, we found that mean PDW and APRI were statistically significant between groups (p = 0.003 and p = 0.034, respectively), whereas there were no differences in NLR, PLR, and DNI ().
Table 3. Comparison of the study groups in terms of available laboratory parameters by 20th weeks of gestation.
Discussion
Our results showed that NLR, PLR, DNI, and PDW did not appear to have clinical relevance in predicting the risk of developing GHT and PE or the severity of PE. In our study, blood samples collected during hospitalisation showed that the mean APRI score of the SPE group was significantly higher than that of the control, GHT, and PE groups.
To evaluate our results, we conducted an extensive literature search. A recent retrospective study comparing 2050 pregnant women (n = 164 in the PE and n = 1886 in the control group) reported that NLR and PLR were similar in blood samples before 20 weeks of gestation, in contrast to a higher NLR and lower PLR in the PE group in blood samples just before delivery (Mannaerts et al. Citation2019). Gezer et al. (Citation2016) reported that NLR and PLR values were significantly higher in the PE group than in the control group, and identified NLR as the strongest predictive variable in multiple regression analysis. In the study by Wang et al. (Citation2019), the ROC analysis showed that NLR had diagnostic accuracy in distinguishing PE from the control group and a predictive role in estimating disease severity (cut-off: 4.198, sensitivity 53%, specificity 83%). One study reported higher levels of inflammatory markers in pregnant women with PE and concluded that measurement of NLR and PLR is probably useful for predicting preeclampsia (Zheng et al. Citation2019). However, another study reported that NLR and PLR parameters fail in predicting progression from PE to SPE (Yavuzcan et al. Citation2014). In a study of healthy subjects, it was reported that NLR and PLR cannot be used to determine SPE (Targońska-Stępniak et al. Citation2020). In the study by Kurtoglu et al. (Citation2015), the cut-off value for NLR was 4.48 with a sensitivity of 58% and specificity of 63%, although there was no significant relationship with disease severity. When we compared the control group and the PE group in terms of NLR in our study, it was found that the NLR was lower in the PE group, contrary to expectations. Moreover, the analysis of ROC, showed that the cut-off value between the control group and PE was 3.55, with a sensitivity of 65% and a specificity of 51%. In contrast to the existing studies, our cut-off value, as well as the mean NLR value, was lower than other studies. However, it was observed that this relationship did not continue between the control group and the PE group. Therefore, it was interpreted as not clinically useful.
One study found no effect of PDW and NLR values in predicting PE (Örgül et al. Citation2019). Two other studies also reported that PDW value had no role in predicting the severity of PE (Al-Sheeha et al. Citation2016, Duan et al. Citation2020). In contrast, Yang et al. (Citation2014) reported a positive association between PDW and the severity of PE, suggesting that the parameter is a potential tool for predicting the severity of PE. The study by Cho et al. (Citation2017) reported that DNI was elevated in SPE cases compared with PE and normotensive subjects. In view of the results of our study, it was suggested that the DNI value was not sufficient to evaluate the risk of developing gestational hypertension and preeclampsia and to predict the severity of preeclampsia. The fact that our study included a higher number of cases increased the value of our results. It was concluded that prospective randomised controlled trials with larger numbers of cases are needed on this topic.
Sasmaz et al. (2020) reported that the APRI score with a cut-off value of 0.339 (sensitivity of 82.6% and specificity of 87.6%) predicted HELLP syndrome better than AST alone (sensitivity of 71.1% and specificity of 91.2%), suggesting that the APRI score is an effective and useful method for estimating HELLP syndrome. Searching the literature, no study was found that investigated the relationship between PE and APRI. Our study is the first to investigate this association.
The results of our study need to be interpreted considering its weaknesses and strengths. The main limitation of our study is its retrospective design. On the other hand, the fact that we conducted it in a tertiary care centre with a comparatively larger sample could be considered a strength. In addition, the analysis of data from a single central laboratory avoided instrument-related variability. A significant contribution to the literature was made by evaluating all parameters with a large number of cases. In addition, the literature was illuminated with respect to further investigation of APRI values in future PE studies.
Conclusion
Considering the available data in our study, we concluded that NLR, PLR, DNI, and PDW have no clinical significance in assessing the risk of developing GHT and PE and predicting the severity of preeclampsia. We further hypothesise that the APRI might provide clinical evidence of progression from hypertensive pregnancy disorders to SPE, which seems to be a promising implication that should be tested by further studies. To our knowledge, this was the first study to present such data and thus made an important contribution to the literature. Our results also highlight the difficulty in predicting hypertensive disorders in pregnancy. Therefore, the most important step to be taken within the available evidence is to correct modifiable risk factors such as BMI. Prospective randomised controlled trials with large numbers of cases are needed to provide more accurate information on this topic.
Author contributions
All of the authors have contributed to project development, data collection, data analysis, and writing of the manuscript.
Contributions: Conception and design of the study, acquisition of data, analysis and/or interpretation of data, drafting the manuscript.
Mujde Can Ibanoglu Contributions: Analysis and/or interpretation of data, drafting the manuscript.
Kevser Adar Contributions: Analysis and/or interpretation of data, drafting the manuscript.
Merve Ozkan Contributions: Analysis and/or interpretation of data, drafting the manuscript.
Omer Lutfi Tapisiz Contributions: Conception and design of the study, acquisition of data.
Yaprak Engin-Ustun Contributions: Conception and design of the study, analysis and/or interpretation of data, drafting the manuscript.
Additional author: Can Tekin Iskender.
Ethics statement
Approval was obtained by the institutional review board from Ankara Etlik Zubeyde Hanım Women’s Health Training and Research Hospital on 17.06.2020 # 2020/77.
Supplemental Material
Download MS Word (15.9 KB)Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Data are openly available in a public repository that issues datasets with DOIs.
References
- Al-Sheeha, M.A., et al., 2016. Platelet count and platelet indices in women with preeclampsia. Vascular Health and Risk Management, 12, 477–480.
- Bateman, B.T., et al., 2012. Hypertension in women of reproductive age in the United States: NHANES 1999–2008. PLoS One, 7 (4), e36171.
- Bibbins-Domingo, K., et al., 2017. Screening for preeclampsia: US preventive services task force recommendation statement. JAMA, 317 (16), 1661–1667.
- Canzoneri, B.J., et al., 2009. Increased neutrophil numbers account for leukocytosis in women with preeclampsia. American Journal of Perinatology, 26 (10), 729–732.
- Cho, H.Y., et al., 2017. Increased delta neutrophil index in women with severe preeclampsia. American Journal of Reproductive Immunology, 78 (3), 12705.
- Duan, Z., et al., 2020. Alterations of several serum parameters are associated with preeclampsia and may be potential markers for the assessment of PE severity. Disease Markers, 2020, 1–7.
- Engin-Üstün, Y., et al., 2019. Changing trends in the Turkish maternal deaths, with a focus on direct and indirect causes. European Journal of Obstetrics, Gynecology, and Reproductive Biology, 234, 21–25.
- Gezer, C., et al., 2016. High first-trimester neutrophil-to-lymphocyte and platelet-to-lymphocyte ratios are indicators for early diagnosis of preeclampsia. Ginekologia Polska, 87 (6), 431–435.
- Kanat‐Pektas, M., et al., 2014. Could mean platelet volume in late first trimester of pregnancy predict intrauterine growth restriction and pre‐eclampsia? Journal of Obstetrics and Gynaecology Research, 40 (7), 1840–1845.
- Kurtoglu, E., et al., 2015. May ratio of neutrophil to lymphocyte be useful in predicting the risk of developing preeclampsia? A pilot study. The Journal of Maternal-Fetal & Neonatal Medicine, 28 (1), 97–99.
- Laresgoiti‐Servitje, E., 2013. A leading role for the immune system in the pathophysiology of preeclampsia. Journal of Leukocyte Biology, 94 (2), 247–257.
- Leslie, K., Thilaganathan, B. and Papageorghiou, A., 2011. Early prediction and prevention of pre-eclampsia. Best practice and research. Best Practice & Research. Clinical Obstetrics & Gynaecology, 25 (3), 343–354.
- Loaeza-del-Castillo, A., et al., 2008. AST to platelet ratio index (APRI) for the noninvasive evaluation of liver fibrosis. Annals of Hepatology, 7 (4), 350–357.
- Lurie, S., Frenkel, E. and Tuvbin, Y., 1998. Comparison of the differential distribution of leukocytes in preeclampsia versus uncomplicated pregnancy. Gynecologic and Obstetric İnvestigation, 45 (4), 229–231.
- Mannaerts, D., et al., 2019. Are neutrophil/lymphocyte ratio (NLR), platelet/lymphocyte ratio (PLR), and/or mean platelet volume (MPV) clinically useful as predictive parameters for preeclampsia? The Journal of Maternal-Fetal & Neonatal Medicine, 32 (9), 1412–1419.
- Metz, T.D., et al., 2020. Obstetric care consensus# 10: management of stillbirth:(replaces practice bulletin number 102, March 2009). American Journal of Obstetrics and Gynecology, 222 (3), 2–20.
- Örgül, G., et al., 2019. First trimester complete blood cell indices in early and late onset preeclampsia. Turkish Journal of Obstetrics and Gynecology, 16 (2), 112–117.
- Pyo, J., et al., 2013. Delta neutrophil index as a marker for differential diagnosis between flare and infection in febrile systemic lupus erythematosus patients. Lupus, 22 (11), 1102–1109.
- Roberts, J.M. and Bell, M.J., 2013. If we know so much about preeclampsia, why haven’t we cured the disease? Journal of Reproductive Immunology, 99 (1–2), 1–9.
- Şaşmaz, M.İ., et al., 2020. Aspartate-aminotransferase to platelet ratio index score for predicting HELLP syndrome. The American Journal of Emergency Medicine, 38 (3), 459–462.
- Targońska-Stępniak, B., et al., 2020. The relationship between hematological markers of systemic inflammation (Neutrophil-To-Lymphocyte, Platelet-To-Lymphocyte, Lymphocyte-To-Monocyte Ratios) and ultrasound disease activity parameters in patients with rheumatoid arthritis. Journal of Clinical Medicine, 9 (9), 2760.
- Townsend, R., et al., 2019. Prediction of pre‐eclampsia: review of reviews. Ultrasound in Obstetrics & Gynecology, 54 (1), 16–27.
- Wai, C.-T., et al., 2003. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology, 38 (2), 518–526.
- Wang, J., et al., 2019. Assessment efficacy of neutrophil-lymphocyte ratio and monocyte-lymphocyte ratio in preeclampsia. Journal of Reproductive İmmunology, 132, 29–34.
- Yang, S.W., et al., 2014. Significance of the platelet distribution width as a severity marker for the development of preeclampsiaa. European Journal of Obstetrics & Gynecology and Reproductive Biology, 175, 107–111.
- Yavuzcan, A., et al., 2014. Mean platelet volume, neutrophil-lymphocyte ratio and platelet-lymphocyte ratio in severe preeclampsia. Ginekologia Polska, 85 (3), 197–203.
- Zheng, W.-F., et al., 2019. Diagnostic value of neutrophil-lymphocyte ratio in preeclampsia: a PRISMA-compliant systematic review and meta-analysis. Medicine, 98 (51), e18496.