Influence of HIV and its treatment on hypertensive disorders of pregnancy in women from a low‐ to middle‐income country

To establish a potential relationship between hypertensive disorders of pregnancy (HDP) and HIV infection.


| INTRODUC TI ON
Hypertensive disorders of pregnancy (HDP) are characterized by a new onset of elevated blood pressure (BP) with or without significant proteinuria after the 20th week of gestation in a patient who was previously normotensive. It affects approximately 5%-10% of all pregnancies and is a recognized major cause of maternal and perinatal morbidity and mortality worldwide. [1][2][3] A local study in Durban, South Africa, reported an incidence of 12%. 4 The cause of HDP remains largely unclear, although many studies have postulated that abnormal activation of the immune system may play a role in triggering the cascade of events leading to vascular maladaptation 5 and defective placentation resulting from inadequate cytotrophoblast invasion of the myometrium with subsequent abnormal remodeling of the spiral arteries. 5 Evidence from epidemiological studies also support the concept of maternal-fetal immune maladaptation and inflammatory immune responses. [5][6][7] The maladaptation results in decreased utero-placental blood circulation, tissue hypoxia, and production of free radicals with widespread endothelial damage, vasospasm, and multiple organ damage, which accounts for the clinical presentation of pre-eclampsia and other HDP. 5,6 HIV infection causes immune suppression and limits the degree to which the maternal immune system may respond to the presenting antigens and can potentially impact the development of HDP.
Further, highly active antiretroviral therapy (HAART) produces metabolic changes in the cardiovascular system and has been suggested as a possible mechanism for hypertensive disorders in women living with HIV (WLHIV) taking treatment. 8 Several retrospective studies have reported the possible protective role of immunosuppression secondary to HIV infection on HDP, 9 while others have noted no significance difference in the development of HDP. 10 However, some of these studies were performed prior to the widespread use of HAART for all pregnant women.
In a subanalysis of maternal deaths caused by hypertensive diseases of pregnancy, Sebitloane et al. showed a reduction in deaths attributable to HDP among HIV-infected women; however, the risk was elevated in a group of maternal deaths treated with HAART. 9 It is therefore postulated that the restoration of the immune system through the use of HAART could increase the risk of development of HDP and possibly worsen pregnancy outcomes in affected women. 7,9,11 Therefore, this study sought to determine the prevalence of HIV among women with HDP and establish whether the use of HAART initiated before or during pregnancy has any effect on the disease or its severity.

| Study design
This was cross-sectional observational study of patients with HDP, conducted over a 6-month period (September 2018 to February 2019) at King Edward VIII Hospital, a regional health facility in Durban, South Africa. The study was approved by the University of KwaZulu Natal's biomedical research ethics committee (reference: All pregnant women who presented with new onset of HDP after 20 weeks of gestation within the study period were enrolled into the study, regardless of their HIV status. HDP was defined as a diastolic BP >90 mm Hg and systolic BP >140 mm Hg on more than two occasions and managed per clinical protocols of HDP. Patients with chronic medical conditions were excluded, i.e. diabetes, epilepsy, asthma, chronic hypertension, cardiac disease, chronic kidney disease, connective tissue disease, and recreational drug use, as well as multiple pregnancy (risk factor for pre-eclampsia) and confirmed congenital fetal anomaly. We also excluded patients who were lost to follow-up or with incomplete data for analysis. Consent was obtained from individual patients and those who declined were excluded from the study. Data were entered from enrollment and patients were followed up until immediately postdelivery. This included demographic data, pregnancy details, highest BP recorded, and treatment given. In addition, we analyzed the type of hypertension, gestational age at onset of the elevated BP and at delivery, and HIV-specific bloods (CD 4 counts and viral loads).

| Exclusion criteria
Sample size and data handling A sample size of 300 patients was deemed necessary to give at least 120 WLHIV (based on local study, which shows a prevalence of 30% to 40% of the pregnancy population), in order to assess differences with an 80% power and 95% confidence intervals.
Data regarding patient demographics, obstetric complications, and HIV status were entered into a Microsoft Excel spreadsheet and exported to SPSS version 21 (IBM) for analysis. Continuous variables were summarized using means, standard deviations, and odds ratios.
For the numerical variables, differences were assessed using t test.
Differences in categorical variables were determined using Pearson χ 2 test or Fisher exact test, as appropriate. Statistical significance was assessed at P < 0.05.

| RE SULTS AND ANALYS IS
A total of 300 patients with HDP were consecutively identified, 299 participants were available for final analysis and one patient was lost to follow-up. There were 209 (70%) HIV-uninfected participants and 90 (30%) WLHIV ( Table 1).
Most of the participants (60.7%, n = 182) were within the 21-to 34-year age bracket, and most developed HDP at gestations of 25-34 weeks at the time of admission. Table 1 shows all other demographic details, with HIV-infected women being significantly older (P < 0.0001) and of higher parity (P < 0.0001) and having a higher body mass index (BMI; P < 0.01).
HDP developed significantly later in gestation among WLHIV, i.e. 34 versus 32 weeks (P = 0.023); however, the BP measurements and the types of HDP were comparable in the two groups, with most of the women (58% in either group) having mainly pre-eclampsia ( Of the WLHIV, 32.2% (29 of 90) of participants were initiated on antiretroviral therapy during the index pregnancy, while 67.8% (61 of 90) had started treatment prior to pregnancy. Of the latter, 52 of 61 (85%) had viral load (VL) readings of <1000 copies/mL and had a higher mean CD4 count than those who initiated ART during pregnancy (448.7 cells/mm 3 compared with 281 cells/mm 3 , P < 0.0001).
Women with pre-delivery ART were also significantly older but had lower BMI ( Table 4).  The rate of placental abruption placental was also lower in the HIVnegative group (4.8%) compared with 15.6% in WLHIV (P = 0.020).

| PERINATAL OUTCOME S
The preterm delivery rate was 68% (204 of 299

| DISCUSS ION
In a setting of high HIV seroprevalence, HDP were found to occur less in WLHIV compared with uninfected individuals. The duration of HAART was not associated with the onset or severity of the disease.
The study further found that WLHIV developed HDP much later in gestation; however, the rate of adverse outcomes for both mother and infant were worse off in these women (higher maternal abnormal liver function tests, placental abruption, and admissions to the NICU).
Interestingly, there were fewer women who were primigravid with HDP among WLHIV compared with those not infected. The different categories of HDP were not statistically different according to HIV status. Higher admission rates to the NICU for HIV-exposed babies could have been for observation and administration of antiretroviral prophylaxis, which was beyond the scope of this study.
The above findings confirm most of the latest literature, which found that WLHIV were less likely to develop preeclampsia or other forms of HDP. 12  HIV is known to affect liver enzymes. It was beyond the scope of this study to look at the individual drugs used. The finding that placental abruption was higher among WLHIV has also been previously reported 16 ; however, this needs further follow-up. Other local studies found an increased rate of postpartum but not antepartum hemorrhage. 17 HIV-infected women were much older and of higher parity, mainly because of the larger proportion (i.e. 2/3) of women already taking HAART at the start of the pregnancy. The latter is in keeping with other hospital records and statistics. Notably, the duration of HAART did not make any significant difference among those with HIV infection. Some of the studies in the South African setting were mainly performed prior to the widespread use of HAART, and hence there were others that reported no difference. 12,18 A meta-analysis and a systematic review also found no association between HDP and HIV or the use of HAART. 19 The strength of this observational study was the ability to collect cross-sectional data in real time, which allowed for follow-up of clinical condition and outcomes. abruption and elevated liver enzymes in this group. Further, more robust and larger studies are needed.

AUTH O R CO NTR I B UTI O N S
Sunday Afemikhe Imogie-co-conceived the idea, wrote the proposal, collected, analyzed and interpreted the data, and wrote the manuscript. Hannah Motshedisi Sebitloane-conceived the idea, assisted with the proposal writing, assisted with data analysis and interpretation, wrote the manuscript. Both authors support their right to authorship.

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request from the authors.