Portal de Periódicos da CAPES

Third trimester and early postpartum period of pregnancy have the greatest risk for ACS in women with SCD

2019; Wiley; Volume: 94; Issue: 12 Linguagem: Inglês

10.1002/ajh.25643

1096-8652

Eugenia Vicky Asare, Edeghonghon Olayemi, Theodore Boafor, Yvonne Dei‐Adomakoh, Enoch Mensah Dip, Charles Hayfron‐Benjamin, Brittany Covert, Adetola A. Kassim, Allison James, Mark Rodeghier, Michael R. DeBaun, Samuel A. Oppong,

Cardiovascular Issues in Pregnancy

Pregnancy is a life-threatening occurrence in women with sickle cell disease (SCD), with increased odds of maternal and perinatal mortality compared to pregnant women without SCD.1 During pregnancy, women with SCD can also experience SCD related maternal morbidities, including acute vaso-occlusive pain episode, acute chest syndrome (ACS), and venous thromboembolism with expected increased incidence rates when compared to not being pregnant.2 We previously demonstrated in a case series that 87% of all maternal deaths were due to ACS, with almost 80% of episodes preceded by an acute pain event.3 In another prospective study, we demonstrated increased incidence rates of acute pain and ACS in pregnant women with SCD4 compared to historical non-pregnant women with SCD.5 As a planned follow-up to our prospective cohort study, we extended the outcome in our pre-existing cohort of pregnant women with SCD4 to determine the impact of pregnancy on acute pain events pre- and postpartum. We tested the hypothesis that pregnant women with SCD have higher incidence rates of acute pain requiring hospitalization and ACS during pregnancy compared to one-year postpartum. This study received approval from the Ethical and Protocol Review Committee, University of Ghana, and Vanderbilt University Medical Center Institutional Review Board. Permission was obtained from the Ghana Institute of Clinical Genetics (GICG) and the Korle-Bu Teaching hospital (KBTH) where the study was conducted. The obstetric unit at KBTH conducts about 10 000 deliveries/year including ∼250 women with SCD. The adult SCD clinic at GICG provides day-care services with ∼10 000 patient visits/year. The study period was from April 2015 to November 2017. Participants were assessed over two time periods: a) pregnancy period: period from enrollment (irrespective of gestational age) to 6 weeks after delivery; and b) postpartum period: period between 6 weeks plus 1 day to 1-year (52 weeks) post-delivery. The definition of acute pain episode as in our previous publications was maintained.3, 4, 6 During the postpartum period, "if the patient was able to judge whether the pain was of the type usually associated with crisis and reported such pain, this was considered appropriate evidence of an acute pain episode.5" The total number of acute pain episodes requiring hospitalization was confirmed with a chart review. The definition of ACS as in our previous publications was maintained.3, 4, 6 The research obstetricians and hematologists adjudicated all participants admitted for suspected ACS events. A consensus agreement was reached to determine the diagnosis of ACS. Table S1 highlights the adjudication criteria used for ACS. The observers and adjudicators were separate individuals. During the pregnancy period, outpatient and inpatient care were provided by our multidisciplinary care team as previously described.4 All participants in the pregnancy cohort were referred to the adult SCD clinic; however, some elected not to attend because they received their routine care outside of GICG. Sixty-six percent of the study participants were attendants at the adult sickle cell clinic during the postpartum period and were evaluated every 3 months by hematologists, general practitioners or both. Standard care protocols were used during acute pain episodes depending on pain rating.3, 4, 6 Using either the numeric or descriptive scales, participants who had reduction in pain intensity to mild pain (1/10-3/10) were discharged and asked to come for follow-up the next day. Participants with persistently moderate (4/10-6/10) or severe (>6/10) pain after 6 hours observation period were referred to the medical emergency for further management. None of the study participants was on hydroxyurea therapy pre-pregnancy and post-delivery. Data collected included demographics, gestational ages at enrolment and delivery, duration of follow-up during the pregnancy and postpartum periods, the number of acute pain and ACS episodes, the number of ACS episodes requiring mechanical ventilation, and blood transfusions. All data were entered and managed on REDCap. We evaluated outcomes in participants who had been followed-up at least 12 weeks during pregnancy and at least 6 months post-delivery. The primary outcomes assessed were incidence rates of acute pain and ACS, summarized as events per patient-years. Severe acute pain episodes and ACS requiring hospitalization that occurred within a period of 2 weeks were counted as one episode each. Comparisons of baseline and pregnancy characteristics of participants in the secondary analysis and those excluded was performed (Table S2). Data was analyzed using SPSS version 25 (Armonk, NY, IBM Corp.). Statistically significant difference was set at P value <.05. Of the original 149 cohort, we had data for 85.9% (128/149) till 1 year post-delivery. Per our exclusion criteria however, only 99 (66.4%; HbSC 58, HbSS 41) were included in the analysis. There was no difference in the baseline and pregnancy characteristics of participants in the secondary analysis and those excluded (Table S2). Of the 99, there was no difference in the mean gestational age at enrollment (P = .072) and the mean total follow-up time during the pregnancy (P = .292) and postpartum (P = .170) periods between HbSS and HbSC respectively. Table 1 summarizes the demographic characteristics and the clinical differences in all participants. Acute pain episodes occurred in 49.5% (74/149) and 18.2% (18/99) of the study participants during the pregnancy and postpartum periods, respectively. The pain rate decreased from 1.8 to 0.3 events/ patient-years (P < .001; Wilcoxon rank test), during pregnancy to the postpartum period, respectively. Within the major phenotypes HbSC and HbSS, the pain rates decreased in the two periods from 1.5 to 0.2 (P < .001) and 2.2 to 0.5 events/patient-years, respectively (P < .001; Wilcoxon rank test), Table 1. Similarly, a significant reduction in ACS rates occurred in the postpartum period. About 15.2% (15/99) of the participants developed ACS during the pregnancy period, with each having one episode of ACS and none in the postpartum period. The ACS incidence rate decreased from 0.4 to 0.0 events/patient-years (P = .001; Wilcoxon rank test). In participants with HbSC and HbSS, the ACS incidence rates decreased from 0.4 to 0.0 (P = .012; Wilcoxon rank test) and 0.4 to 0.0 events/patient-years, respectively (P = .018), Table 1. An almost equal number of the ACS episodes during the pregnancy period occurred in the third trimester [53.3% (8/15)] and within the first 4 days of delivery [46.7% (7/15)]. No ACS events occurred in the first or second trimester. Out of the 15 participants who had ACS, three (20.0%) of the study participants met three criteria, six (40.0%) met four criteria, five (33.3%) met five criteria, and one (6.7%) met all the six criteria for diagnosing ACS. Almost all (93.3%) of those who had ACS during the pregnancy period had a preceding acute pain episode. Out of the 15 episodes of ACS, three (20.0%) required mechanical ventilation and intensive care management. All the cases required prolonged hospitalization ranging from 5-14 days. Incidence rates of pain and ACS among HbSS and HbSC were similar during pregnancy and up to one-year postpartum, Table 1. Simple blood transfusion was used in the management of acute pain episodes and ACS during the pregnancy period based on evidence of cardiopulmonary compromise. Approximately a third (33.6%) of the participants received simple blood transfusion during their treatment and none received regular prophylactic transfusions. Twenty-eight percent (14/50) of these were HbSC while 72% (36/50) were HbSS. All 15 (HbSC:8, HbSS:7) participants who had ACS received simple blood transfusion as part of the protocol for management of ACS. There were no blood transfusions during the postpartum period. In our prospective cohort, there was a 6-fold and 4-fold reduction in acute pain and ACS incidence rates postpartum when compared to the pregnancy period. We also demonstrate that there was no difference in the incidence rates of pain and ACS between the two SCD phenotypes (HbSS and HbSC). Our study has few limitations. The increased frequency of acute events is not compared to each woman's pre-pregnancy rate, because this data is difficult to ascertain. We recognize the challenge that nursing mothers may not seek treatment for mild to moderate pain since this may interfere with breastfeeding, resulting in a reduction in the acute pain incidence rates postpartum. However, ACS episodes requiring hospitalization are unlikely to be treated at home when breast feeding. Thus, most mothers will likely seek medical care for ACS (one of the two outcome measures of the study). To prevent potential bias in assessment of pain, in a new prospective cohort study, electronic pain diaries will be used. Other than acute pain in the third trimester or early postpartum period, preceding ACS, we were unable to identify predictive clinical or laboratory factors for morbidity. In conclusion, pain and ACS occur more frequently during pregnancy in women with SCD when compared to the same women when not pregnant. Importantly ACS occurs most frequently in the third trimester and early postpartum period; and is most often preceded by acute vaso-occlusive pain requiring hospitalization. Understanding the mechanism and targeting primary prevention of acute pain episodes in pregnant women in the third trimester and early postpartum period may lead to a better understanding of ACS in pregnant women with SCD. We acknowledge the leadership of Department of Obstetrics and Gynecology, including Professor Obed who provided the support for this ambitious project, KBTH for administrative support, funds from Office of Research and Innovation and Development (ORID), University of Ghana, Doris Duke Charitable Foundation, Phillips Family Donation, Aaron Ardoin Foundation for Sickle Cell Anemia, and J.C. Peterson, M.D. endowed chair funds from School of Medicine, Vanderbilt University. We also acknowledge the donation of pulse oximetry machines by the team from Vanderbilt University Medical Center; as well as the sickle cell team from the Ghana Institute of Clinical Genetics. SAO, MD, AK, and EO conceived the idea; EVA, EM, BC, CHB, YDA, EO, TKB, and SAO performed the data search, and EVA, MD, and MR analyzed the data. All authors participated in writing the article and reviewed and approved the final version before submission. Full disclosure of interests available to view online as supporting information. This project was partly funded by Office of Research and Innovation and Development (ORID) Research Fund/8/LMG-008, University of Ghana (awarded to EO and SAO), Doris Duke Charitable Foundation; Burroughs Wellcome Foundation; Phillips Family Donation; Aaron Ardoin Foundation for Sickle Cell Anemia; Vanderbilt University School of Medicine (J.C. Peterson, M.D. endowed chair funds) and Vanderbilt University Medical Center Gift Funds. We also acknowledge the donation of pulse oximetry machines by the team from Vanderbilt University Medical Center. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.