The Profile of Glycated Hemoglobin in Non-Diabetic Pregnant Women with Preeclampsia: A Case-Control Study in Kinshasa, the Democratic Republic of Congo

Background: A disturbed glucose metabolism is quite common during pregnancy. It is due to the diabetogenic potential of pregnancy and responsible for many obstetric complications. The glycated hemoglobin is one of the markers used to depict these disorders. Higher concentrations of this marker would be associated with unfavorable results of pregnancy. Objective: To describe the profile of HbA1c in non-diabetic preeclamptic pregnant women and to establish the association between the values of this marker and the maternal and fetal complications. Materials and Method: This is a case-control study of 142 pregnant women in their second and third trimester. They were followed in the maternity hospitals of University Clinics and the General Reference Hospital in Kinshasa, Democratic Republic of Congo between May and October 2019. The sampling of preeclamptic pregnant women was exhaustive. How to cite this paper: Luzolo, G.M., Beya, D.T., Numbi, D.K., Nyota, P.K., Ngandu, P.C., Manzambi, B.S.M., Natuhoyila, A.N., Zita, M.N., Ntita, G.I., Nkanga, M.N., Masidi, J.M., Nzongola-Nkasu, D.K., Mushengezi, D.S., Lobota, A.M., Buassa-Bu-Tsumbu, B., Risasi, C.A., Verdonck, F., Spitz, B. and Moyene, J.P.E. (2020) The Profile of Glycated Hemoglobin in Non-Diabetic Pregnant Women with Preeclampsia: A Case-Control Study in Kinshasa, the Democratic Republic of Congo. Open Journal of Obstetrics and Gynecology, 10, 365-389. https://doi.org/10.4236/ojog.2020.1030034 Received: February 17, 2020 Accepted: March 20, 2020 Published: March 23, 2020


Introduction
A disturbed glucose metabolism is quite common during pregnancy. It is linked to the diabetogenic potential of pregnancy and increase in the second half of gestation [1]- [6]. It may cause many complications including fetal malformations, prematurity, fetal death in utero, fetal macrosomia, preeclampsia and so on. In a large epidemiological study of 25,000 pregnant women, the risk of maternal and neonatal complications has been shown to increase linearly with maternal blood glycaemia [7] [8] [9] [10]. Furthermore, the International Diabetes Federation estimated the overall prevalence of hyperglycaemia during pregnancy at 16.9%, which more than justifies systematic screening [11] [12]. Different strategies can highlight these glycemic disorders, including the determination of glycated hemoglobin (HbA1c) [13] [14] [15]. Many studies revealed that the increase in the level of HbA1c during pregnancy is closely associated with the unfavorable results and that its dosage is significant importance in the monitoring of complications of pregnancy, particularly that of preeclampsia [16] [17] [18]. Preeclampsia is one of the major causes of maternal and perinatal morbidity and mains very high and variable, reaching up to almost 20% of cases [23]. Although the main cause of this condition remains even less well known, many studies agree on the role played by oxidative stress in the occurrence of this condition [24]- [28]. The results obtained from numerous studies note that the prevalence of the disease is quite high in pregnant women with hyperglycaemia [29] [30].
The oxidative stress generated at all stages of protein glycation and during the formation of HbA1c would be favorable for the occurrence of preeclampsia and its complications. In addition, some studies report an association between high levels of HbA1c and anti-angiogenic factors implicated in the occurrence of preeclampsia [31]. More recently, a very significant association has been established between elevated levels of protein glycation end products (AGE: Advanced glycation end-products) with preeclampsia [32] [33] [34].
Thus, the evaluation of HbA1c in pregnant women with pre-eclampsia is essential to detect the hyperglycaemia states that may be associated and to research the influence of this marker on the prognosis of the disease.
The present study aims to describe the profile of HbA1c in non-diabetic pregnant women and to establish the association between the values of this marker with maternal and fetal complications.

Materials and Method
This is a case-control study involving 71 preeclamptic women (cases) and 71 non-preeclamptic women (controls) matched for gestational age, between May and October 2019. The study population comprised black non-diabetic preeclamptic women from two hospitals (University Clinics and the Kinshasa General Reference Hospital), in the Democratic Republic of the Congo. The sampling of preeclamptic pregnant women was exhaustive. Controls were healthy pregnant women carrying pregnancies of the same type and of the same gestational age as the cases. Ultimately, 71 were preeclamptic and 71 healthy (controls).
Diabetics pregnant, women with sickle cell anemia, and those with pathologies characterized by polycythemia were excluded from this study. Gestational age was calculated from the date of the last menstrual period in pregnant women with a regular menstrual cycle and from an early ultrasound in those who did not know their last menstrual dates or who had a history of spaniomenorrhea.
Preeclampsia has been defined according to the criteria of the National High Blood Pressure Education Program of United States, namely the occurrence in the second part of the pregnancy of arterial hypertension (SAP ≥ 140 mmHg and DAP ≥ 90 mmHg) associated with proteinuria (quantitative ≥ 300 mg/24h or qualitative > with a cross evaluated by the strip method) in a previously normotensive woman. Preeclampsia was considered moderate when SAP was ≥ 140 mmHg but < 160 mmHg and the DAP was ≥ 90 mmHg but < 110 mmHg and in the absence of signs of severity. It was severe when the SAP was ≥ 160 mmHg and/or the DAP ≥ 110 mmHg or also in front of a hypertension associated with the following signs of severity: epigastric pain, nausea, vomiting, headache in helmet, osteotendinous hyperreflexia, headache, dizziness, blurred vision, ringing in the ears, convulsions, signs of acute lung edema, oliguria (<20 ml/H), or in front of serum creatinine > 100 μmol/L, of increase of liver transaminases > 3 times the normal value, of thrombocytopenia < 100,000/mm 3 [37] was applied to find sickle cell anemia among these women. Obstetric ultrasound and fetal monitoring (NST) were used to establish the fetal well-being score (Manning score) and to search for fetal complications such as fetal growth retardation, fetal distress and death in utero.
The determination of urea and that of creatinine were carried out by the enzymatic methods of urease and creatininase respectively. The GOT and GPT transaminases were assayed by the kinetic method. Obstetric ultrasound and fetal monitoring (NST) were used to establish the fetal well-being score (Manning score) and to search for fetal complications such as fetal growth retardation, fetal distress and death in utero.
To carry out this study, we received the approval of the ethical committees of the school of public health of the faculty of Medicine of the University of Kinshasa. Only those pregnant women who freely expressed their consent were selected for the study. The variables of interest were socio-demographic characteristics, the clinical data as well as the biological parameters were noted in all those pregnant women and mainly maternal age, medical history, obstetric history, gestational age, type of pregnancy, clinical form of preeclampsia, maternal and fetal complications.

Statistical Analyzes
The statistics used to describe the variables were means ± standard deviation for continuous quantitative variables with symmetric distribution. Qualitative variables have been described in terms of absolute (n) and/or relative frequency (%). For the analyses, the comparison of the means was carried out using the Student's t-test. Pearson's Chi-square or Fisher's exact test, as appropriate, was applied to compare the proportions. The linear regression test was applied to check the correlation between HbA1c and glycaemia. The linear regression coefficient in simple analysis was calculated to assess the association between HbA1c and glycaemia. Logistic regression was used to identify the factors associated with elevated HbA1c levels. Only the variables significantly associated with high HbA1c levels in univariate analysis were tested in multivariate analysis. The adjusted Odds-ratios (aOR) and their 95% confidence intervals (CI) were reported to estimate the association between the independent variables and the dependent variable. The value of p < 0.05, p was considered statistically significant.

Results
A total of 142 pregnant women took part in this study consisting of 71 preeclamptic women (cases) and 71 healthy pregnant women (controls).
The socio-demographic characteristics of all these women are shown in Table  1. No significant difference is noted when comparing the two groups according to age, marital status, type of marriage and place of residence. It appears however from the results of Table 2 that primigestity, primiparity, as well as the history of preeclampsia are mainly found in pregnant women with preeclampsia compared to controls (p < 0.05). Table 3 gives the characteristics of pregnancy in the two groups. There is a perfect pairing of two study groups in terms of gestational age. Table 4 shows the clinical data for pregnant women in the two groups. We also note in pre-eclamptic pregnant women the presence of certain serious clinical signs such as convulsions, coma, headache, dizziness, blurred vision not found in the controls. As illustrated in Figure 1, severe preeclampsia was the most common clinical form in pregnant women compared to moderate preeclampsia (62% vs. 38%, p < 0.001).
It appears from the results in Table 5 that among the many complications observed in pregnant women with preeclampsia, eclampsia was the most common Data are expressed as mean ± standard deviation or absolute and relative frequency in % between brackets.   Data are expressed as mean ± standard deviation or absolute and relative frequency in % between brackets.  Intrauterine growth retardation 7 9.9 Fetal death 4 5.6 Oligohydramnios 2 2.8 complication in the mother, while on the fetal side, it was acute fetal distress and prematurity.
A significant decrease in platelet count is observed in pre-eclamptic pregnant women compared to controls (167.8 ± 74.3 elements/mm 3  It emerges from the analysis of Figure 2 below that the proportion of preeclamptic pregnant women with pathological values of fasting glycaemia (≥92 mg/dl) was comparable to that of the controls 34% vs 24%, p = 0.133).
However, the proportion of pregnant women with pathological HbA1c values (>5.6%) was significantly higher in the group of preeclamptic pregnant women compared to controls (46.5% vs 9.9%, p < 0.001) (Figure 3).
A positive and significant linear correlation is noted between the HbA1c level    and the fasting glycaemia. This correlation is 88% (r = 0.881) as shown in Figure  4 below. Table 7 gives the clinical characteristics as a function of HbA1c values in preeclamptic pregnant women. It appears that a pathological values of HbA1c (>5.6%) is significantly (p < 0.00) more common in pregnant women with a history of preeclampsia or arterial hypertension, in those with complications such as eclampsia, eclamptic prodromes and prematurity; but also significantly in pregnant women with pregnancies complicated by acute fetal distress (p = 0.023) and in those carrying twin pregnancies (p = 0.035).
Univariate analysis noted an association between high levels of HbA1c with    The data are expressed as absolute and relative frequency in % between brackets. Open Journal of Obstetrics and Gynecology

Discussion
The objective of study was to establish an association between glycated hemoglobin (HbA1c) in the second and third trimester of pregnancy with the occurrence of preeclampsia and to investigate the influence of these glycated hemoglobin on the maternal and fetal prognosis. The reference value for the HbA1c used was 5.6% [35]. It emerges from this study that in the absence of a proven diabetes mellitus context, preeclamptic women had higher HbA1c values compared to non-preeclampsia pregnant women. Pathological values of HbA1c (>5.6%) are very significantly associated with preeclampsia and its complications. These results thus obtained underline the importance of regular glycemic controls to prevent these complications.

Glucose Metabolism and Preeclampsia
Preeclampsia remains one of the major causes of maternal and perinatal morbidity and mortality in developing countries where it kills nearly 50,000 women each year [20]. Although its main cause is largely unknown, numerous studies agree on the role played by oxidative stress [24]- [28]. Epidemiological studies show that hyperglycemic disorders occurring during pregnancy are a risk factor for the development of preeclampsia even outside of proven diabetes mellitus [29] [30]. These hyperglycemic disorders linked to the diabetogenic potential of pregnancy are the consequence of the anti-insulinogenic and lipolytic activity of the somatotropic chorionic hormone, prolactin, cortisol and glucagon whose concentrations increase in the second half of gestation [1]- [6]. In 2013, the International Diabetes Federation estimated the overall prevalence of hyperglycemia during pregnancy at 16.9% [12]. To date, several strategies have been used to detect these disorders, including oral provoked hyperglycemia test, fasting glycemia, as well as the determination of glycated hemoglobin HbA1c [38] [39].

Glycemia and Glycated Hemoglobin Profile in Healthy and in Non-Diabetic Preeclamptic Women
In this study, we used fasting glycemia and HbA1c determination. Although the oral hyperglycemia test remains one of the most widely used dynamic tests for the detection of carbohydrate intolerance, the difficulties associated with the supply of anhydrous glucose, its conservation and its cost limited the realization of this test during the present study. Fasting glycemia was measured by the hexokinase method which is a reference method [40]. The threshold value used is that proposed by ADA and adopted by WHO, 92 mg/dl [41]. The results obtained from the HAPO study (Hyperglycaemia and Adverse Pregnancy Outcomes) and those of other studies have shown that this value is the threshold at which maternal glycemia levels are frequently associated with unfavorable pregnancy outcomes [7] [42] [43]. The determination of glycated hemoglobin was carried out by the immunoturbidimetry technique using a COBAS C111 brand analyzer [36] [44]. The choice of the HbA1c dosage is justified by the fact that it reflects the state of glycemia during the 2 to 3 months preceding the measurement [45], its dosage is reproducible and its level is not influenced by the time of blood collection or by fasting status or by any hypo or hyperglycemic treatment [46]. In addition, the oxidative stress generated during the formation of HbA1c would be favorable for the occurrence of preeclampsia and its complications [47]- [51]. In order to avoid errors of interpretation linked to the shortening of the duration of red blood cells, we excluded from this study the cases of sickle cell anemia, the hemoglobinopathy most encountered in our population [52] [53]. The results of the study by Arlène et al. [54] as well as those of many au-

Glycated Hemoglobin and Preeclampsia
It appears in the results obtained from the HAPO study and those of other studies than during pregnancy, the risk of maternal and neonatal complications increases linearly with maternal glycemia [7] [42] [43]. Several studies have established an association between the states of maternal hyperglycemia and the occurrence of preeclampsia and its complications [59] [60] [61]. However, the mechanism underlying this association remains less well known. Barden et al. [62] note in their study that insulin resistance is the essential factor in this association. In a large study aimed at identifying the determinants of the association between diabetes mellitus and preeclampsia, Gutaj et al. [63] rather note a very significant positive association between high values of HbA1c in the three trimesters of pregnancy with preeclampsia and its complications. The association between HbA1c levels and adverse pregnancy outcomes remains a large area to explore. Most of the data reported comes from first trimester HbA1c measure-  [71] showed that, in pregnant women suffering from carbohydrate intolerance during the second trimester, even if the glycemic values are below the threshold of proven diabetes, the high rates of HbA1c remained strong predictors of adverse obstetric outcomes. In particular, an HbA1c value greater than 5.3% is associated with a 2 times higher risk of occurrence of preeclampsia and neonatal morbidity. In a recent study carried out in New Zealand on 16,122 healthy pregnant women, Hughes et al. [72] showed that an HbA1c level ≥ 5.9% at the start of pregnancy was a clinically relevant marker of the occurrence of preeclampsia, shoulder dystocia and perinatal death. Our results corroborate those of the HAPO study as well as those obtained by Capula et al., then recently by Hugues et al. who note an association between high levels of HbA1c and the occurrence of preeclampsia in pregnant women with unknown diabetes.
It appears from the results of Nielsen et al. [66], that in pregnant women, each increase in the value of HbA1c by 1% above the normal threshold increases the risk of poor perinatal results by 5.5%. This risk is 79% when the HbA1c level is greater than 10.3%. Mané et al. [16] note in their study that in pregnant women without diabetes, an HbA1c level between 5.9% and 6.4% multiplies by 3 the risk of occurrence of preeclampsia and its complications. The results obtained from the study by Cavero-Redondo et al. [73] show that HbA1c is a very reliable indicator of the risk of cardiovascular mortality in diabetics and non-diabetics. In The results of numerous studies [81] [82] have shown very high concentrations of AGEs in preeclamptic pregnant women.

Glycated Hemoglobin and Risk Factors of Preeclampsia in Non-Diabetic Women
We noted during this study that primigestity, primiparity and a history of preeclampsia were risk factors for the onset of the disease. Our results corroborate those published by other authors. Merviel et al. [83] note in their study that primigestity, primiparity, primipaternity as well as personal history of preeclampsia are risk factors for the occurrence of preeclampsia. Similar results are published by English et al. [84] as well as by Cissé et al. [85]. In the study by Gutaj et al. [63], it was noted that the history of nulliparity associated with the high level of HbA1c between 12 -24 weeks of pregnancy constituted a major risk factor for the occurrence of preeclampsia.
The same is true for the history of primiparity or hypertension associated with high levels of HbA 1c in the three trimesters of pregnancy which constitutes a major risk factor for the occurrence of preeclampsia. It thus emerges from our study that primigestity, primiparity, history of preeclampsia or hypertension are risk factors for the occurrence of preeclampsia. This risk is increased and multiplied by 7 in the presence of high levels of HbA1C measured in the second and third trimester in an unknown diabetic pregnant. In the study by Mohamed et al. [86], it was noted that multigestity was one of the risk factors for the occurrence of preeclampsia. We noted in this study an association between high levels of

Glycated Hemoglobin and Preeclampsia Prognosis
We noted in this study that severe preeclampsia was the most common form, unlike moderate preeclampsia (62% vs 32%). Our observation is similar to that of Cisse C.T et al. [85] who find in developing countries the predominance of severe forms of preeclampsia whose frequency can reach up to 88 It occurs as a result of endothelial damage associated with the obstruction of intrahepatic micro vessels by microthrombi [101]. It is characterized by hemolysis, elevated liver transaminases and thrombocytopenia. It was noted in the present study that compared to the controls, the preeclamptic pregnant women presented a moderate but very significant elevation of hepatic transaminase levels (GOT 48.3 ± 5.9 IU/L vs 24.3 ± 9.1 IU/L, p = 0.001; GPT 38.6 ± 4.7 IU/L vs 13.8 ± 6.9 IU/L, p = 0.000) as well as a low platelet count (167.8 ± 74.3. 103 elements vs 228.8 ± 56.4.103 vs elements, p < 0.001). Although this study did not establish a relationship between pathological values of HbA1c and the occurrence of this complication, it can be understood that the endothelial dysfunction caused by high levels of HbA1c and by Advanced glycation end-products would be one of the pathophysiological mechanisms. In the study by Asadifar et al. [77], it was shown that high levels of HbA1c also favored aggregability and the platelet activation which would be responsible for the formation of microthrombi and multi visceral attack which can lead HELLP syndrome.
This study is the first to be carried out in Congolese pregnant women. Its weakness lies first, to the small sampling used and the failure to carry out the oral provoked hyperglycemia test in women with pathological values of HbA1c, which would contribute to determining the prevalence of diabetes mellitus, and secondly, the absence of data relating to neonatal parameters on which high levels of HbA1c could have an influence. This prompts us to recommend carrying out a study on a large sampling taking into account all these other aspects.
The strength of this study is to have established in non-diabetic pregnant women, a relationship between the high levels of HbA1c in the second and third trimester of pregnancy with the occurrence of preeclampsia and its complications, which allows to consider this parameter as a bio-marker of oxidative stress and of the appreciation for the maternal and fetal prognosis during preeclampsia. Our results corroborate those published by other authors, and underline the importance of glycemic control during the second and third trimester of pregnancy among the strategies for preventing preeclampsia and its complications.

Conclusion
We searched to determine the profile of glycated hemoglobin in non-diabetic preeclamptic pregnant women and know if elevated HbA1c levels were associated with maternal and fetal complications during this affection. It appears from the results obtained that 46.5% of those women have high levels of HbA1C. These high levels of HbA1c noted in the second and third trimester of pregnancy are very significantly associated with preeclampsia and are observed mainly in those with maternal or fetal complications. Thus, the integration of the HbA1c level of the second and third trimester among the other risk factors such as primiparity, primigestity, multigestity, and the history of hypertension could help to better assess the risk of preeclampsia and to assess the maternal and fetal prognosis during this condition. These results corroborate those published by other authors. They emphasize the importance of the systematic screening of hyperglycemic disorders during pregnancy. Some limitations of the study being noted, we recommend a study on a large sample in order to validate these results to make HbA1c a marker of risk of occurrence of preeclampsia and for appreciation of the maternal and fetal prognosis in our environment faced with the difficulties of assaying markers of oxidative stress.