Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Risk of Pregnancy Related Hypertension Disorder Associated with Sickle Cell Trait in U.S. Service Women

The Risk of Pregnancy Related Hypertension Disorder Associated with Sickle Cell Trait in U.S.... Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 MILITARY MEDICINE, 185, 1/2:e183, 2020 The Risk of Pregnancy Related Hypertension Disorder Associated with Sickle Cell Trait in U.S. Service Women MAJ Christina O’Hara, MD, MPH, MC, USA*; CAPT Darrell E. Singer, MD, MPH, MC, USPHS*; COL David W. Niebuhr, MD, MPH, MSc, MC, USA (Ret.)* ABSTRACT Introduction The health impact of the heterozygous (Hb Ss) Sickle Cell Trait (SCT) among the estimated one million females of reproductive age in the United States on pregnancy related hypertensive disorders (PRHD) including preeclampsia is not as well understood or researched as Sickle Cell Disease (SCD). Materials and Methods This retrospective cohort study included 25,020 U.S. enlisted, active duty service women during 1992 to 2013. Race within the sample is primarily black (84%) given that blacks have the highest prevalence of SCT. All 5,004 SCT positive individuals and a matched sample of 20,016 from those considered SCT negative were followed while on active duty for PRHD outcomes including gestational hypertension (GHTN), preeclampsia and eclampsia. Results The adjusted incidence rate ratio (IRR) for any PRHD in SCT positive compared to negative individuals was 1.46 (95% CI 1.32 – 1.62). SCT positive versus negative subjects had higher health care utilization rates and utilization rate ratios (URR) for all PRHD diagnoses combined 2.03 (95%CI: 1.97 – 2.10) and for each specific PHRD diagnosis. The prevalence of preeclampsia or eclampsia did not vary by SCT status. The overall PRHD attributable risk due to SCT was 30.4% (95% CI 23.1–37.1%) and the overall number needed to screen to detect a case of PRHD was 156 (95% CI 117–220). Conclusions The results of this study indicate that being SCT positive is a risk factor for PRHD and provides evidence that SCT status may have an adverse effect on reproductive health. Future research needs to include other known risk factors for PRHD to include gravity and parity history, BMI, past history of PRHD prior to enlistment as well as to examine the association with maternal/child pregnancy outcomes. INTRODUCTION active duty enlisted Hispanic-Americans were SCT positive. Sickle cell trait (SCT) is a heterozygous medical condition The prevalence in females compared to males was 3% versus affecting an estimated three million people in the United States 1% respectively. and 8–10% of blacks. Data from state newborn screening Pregnancy related hypertension disorder (PRHD) is esti- programs estimated the 2010 incidence for sickle cell trait in mated to affect up to 10% of all pregnancies. It is one of the the United States at 15.5 cases per 1,000 births. The heaviest leading causes of maternal morbidity and mortality in the U.S. burden was among black Americans at 73.1 cases of SCT per and world-wide. Accepted classifications of PRHD includes 1,000 births compared to 3.0 cases per 1,000 white newborns, the following four categories: 1) preeclampsia/eclampsia, 2) and 6.9 cases per 1,000 Hispanic newborns. In the U.S. chronic hypertension (any cause), 3) chronic hypertension military, the prevalence of SCT among service members has with superimposed preeclampsia and 4) gestational hyperten- increased steadily between 1992 and 2012. Approximately sion. Over the past 20 years, incidence of preeclampsia has 4% black active duty enlisted service members and 0.5% in increased over 25% in the U.S., affecting 3–7% of all pregnan- cies, with an estimated 1/200 cases of these cases progressing 4,5 Department of Preventive Medicine and Biostatistics, Uniformed Ser- to eclampsia. Risk factors for the development of PRHD vices of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814. include age > 40 years, chronic hypertension, nulliparity, The opinions and assertions expressed in this article represent the private pre-gestational diabetes, black race, history of preeclampsia, views of the authors and do not reflect the official views of the Uniformed and BMI > 25 which showed the greatest attributable risk Services University of the Health Sciences, the U.S. Departments of the Army, 6–8 (64%). A 2014 study of 2,288 females receiving Veter- Navy, or Defense, National Heart, Lung, and Blood Institute, Food and Drug Administration or the U.S. Government. Nothing in the presentation implies ans Administration maternity benefits demonstrated a signifi- any Federal/Department of Defense endorsement. cantly higher standardized incidence rate (SIR) of developing doi:10.1093/milmed/usz143 gestational diabetes mellitus (SIR: 1.40; 95% [CI] 1.16, 1.68) Published by Oxford University Press on behalf of the Association of and hypertensive disorder (SIR: 1.32; 95% CI 1.15, 1.51) as Military Surgeons of the United States 2019. This work is written by (a) US compared to civilian women delivering in the United States. Government employee(s) and is in the public domain in the US. MILITARY MEDICINE, Vol. 185, January/February 2020 e183 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT The risks of PHRD to maternal-fetal health are high. The conditions and included 75,400 military subjects of which overall preeclampsia-eclampsia case fatality rate based on 25,020 were female. The primary purpose of the parent study Centers for Disease Control and Prevention data from 1988 to was to evaluate associations of SCT with exertional health 1992 was 6.4 per 10,000 cases at delivery. Maternal and fetal illness outcomes. However, the parent study also collected outcomes of PRHD can include intra-uterine growth restric- additional health outcome data to include PRHD conditions. tion (IUGR), pre-term labor and/or delivery, abruptio placen- The data sources for the parent study came from the Armed tae, stillbirth and maternal and/or fetal death. Prehyperten- Forces Health Surveillance Branch (AFHSB) Defense Med- sion carries a 70% increased stillbirth risk and 69% increased ical Surveillance System and the Navy and Marine Corps risk for small-for-gestational-age infants (SGA) when prehy- Public Health Center (NMCPHC). AFHSB data included per- 11,12 pertension occurs in late pregnancy (36 weeks). Without sonnel (i.e., enlistment, military occupation and discharge) immediate treatment, eclampsia is usually fatal. Long-term and medical encounter (i.e., hospitalization and ambulatory) effects in women with a history of all types of PRHD have not data. The NMCPHC provided SCT laboratory test result data been well studied, but a history of gestational hypertension, (qualitative or quantitative). Data elements were de-identified preeclampsia or eclampsia may be associated with hyperten- prior to transfer to the USUHS SCT study principal investiga- sion and other cardiovascular disease later in life. tor. Data elements for this study include the following: Controversy surrounds whether SCT increases risk to the Demographic data: race/ethnicity, gender, service, age cat- mother or fetus during pregnancy. Studies linking SCT with egory on enlistment (18–19, 20–24, 25–29, 30+), deploy- negative health outcomes suggest an increased risk of dis- ments (dates of deployment and redeployment), home of orders involving the cardiovascular, renal and musculoskele- record (first 3 digits of zip code), quarter and year of tal systems, while exertional related injuries are among the enlistment. 15,16 most common. However, in one systematic review of Medical encounter data: dates (in days relative to enlist- cohort and case-control studies, examining maternal com- ment) of all SCT ambulatory and hospitalization diagnostic plications including urinary tract infection (UTI), anemia, encounter codes of ICD-9 CM 282.5. (pre-)eclampsia, gestational diabetes, and premature labor, the The outcome of interest for this project includes all ICD- risk to the fetus was considered low. Another study showed a 9 encounter codes for PRHD (ICD-CM 642.X- Hypertension higher risk of pre-eclampsia among women with SCT and two 18–20 complicating pregnancy, childbirth, and the puerperium). Sup- studies with a small association with bacterial infection. plemental Table 1 provides the specific four digit ICD-9 CM Two recent studies observed a significantly shorter duration codes included. of pregnancy and increased fetal loss among SCT affected Descriptive statistics for the variables of interest were women as compare to controls and a third study concluded no 21–23 determined by SCT status and PRHD outcome. Descriptive increased perinatal risk to the SCT affected mother. The statistics of PRHD by ICD-9 codes of eclampsia, preeclampsia pathophysiologic basis for this association in the literature is and GHTN were tabulated and displayed by SCT status. hypothesized to be increased sickling and poorly developed Selected data from the parent retrospective cohort study uterine placental spiral arterioles resulting in an inadequate was utilized to evaluate the association between SCT status in blood supply required for fetal development. active duty enlisted U.S. service women from 1992 through SCT and PRHD are both each relatively common condi- 2012 and followed for PRHD outcomes through 2013. To tions, particularly in black women. They are also associated accommodate incomplete years of service among the enlisted with significant morbidity which can impact readiness and personnel, incidence density rates and hazard ratios were cal- healthcare costs. The objective of this study was to study the culated. The unadjusted incidence density rate, rate ratio and association of SCT status with PRHD in U.S. enlisted service attributable risk of PRHD by SCT status and adjusted hazard women. ratio were calculated with Cox Proportional Hazard models over all enlistment cohort years. The number of enlistees needed to screen (NNS) for SCT to detect a case of PRHD METHODS The parent study (PMB-87–291, Version 3, 31 May 2015) for was calculated. The retrospective enlistment cohort data were this project is a retrospective, matched, active duty, enlisted further analyzed to estimate the risk of several related diseases cohort with SCT status as the exposure of interest. All active within PRHD (eclampsia, preeclampsia, and GHTN) in ser- duty enlisted military service members with evidence of vice women by SCT status. Health care utilization rates (UR) being positive for SCT from laboratory test report and/or and adjusted utilization rate ratios (URR) for PRHD overall a medical encounter (1 hospitalization and/or 2 or more and by each of the subcategories of PRHD were calculated ambulatory visits), were identified and matched with four with Poisson models. enlisted military service members without evidence of SCT The study population included a total of 25,020 female by branch of service, gender, race/ethnicity, and quarter and subjects 5,004 SCT positive and 20,016 SCT negative. Offi- year of enlistment. The parent study utilized data from 1992 cers were excluded from this study due to their generally through 2013 on SCT status and various SCT-related medical higher socioeconomic status (and associated better access to e184 MILITARY MEDICINE, Vol. 185, January/February 2020 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE I. Distribution of study subjects by SCT status Variable SCT(-) n % of SCT(-) SCT (+) n %ofSCT(+) Race/Ethnicity Non-Hispanic White 568 2.8 142 2.8 Non-Hispanic Black 16874 84.3 4219 84.3 Non-Hispanic Other 1290 6.4 322 6.4 Hispanic 1284 6.4 321 6.4 Age at Enlistment 18–19 11129 55.6 2731 54.6 20–24 7110 35.5 1826 36.5 25–29 1354 6.8 334 6.7 30+ 423 2.1 113 2.3 Service Army 5796 29.0 1448 28.9 Air Force 5012 25.0 1254 25.1 Marines 904 4.5 226 4.5 Navy 8304 41.5 2076 41.5 Enlistment Year 1992–1994 1208 6.0 302 6.0 1995–1999 3528 17.6 882 17.6 2000–2004 5780 28.9 1445 28.9 2005–2009 5936 29.7 1484 29.7 2010–2012 3564 17.8 891 17.8 ∗∗ Occupation Combat/Armor/Repair 4999 25.0 1270 25.4 Communications/Intelligence 7087 35.4 1915 38.3 Healthcare 2464 12.3 608 12.2 Other 5466 27.3 1211 24.2 Distribution of study subjects (n = 25,020) by sickle cell trait (SCT) status. Study participants possessing sickle cell trait (SCT(+)) (n = 5,004), and participants without sickle cell trait (SCT(-)) (n = 20,016). Matching factors in selection of cohort. ∗∗ Chi Square (p-value = 0.001). healthcare prior to military service), differing accession routes RESULTS to service (i.e., service academy, Reserve Officers’ Training Table I displays the distribution of race, age at enlistment, Corps) and potential overlap with enlisted who may later service and occupation at separation by SCT status. The attend Officer Candidate School, differences in the length, racial/ethnic distribution within the study was black (84%), rigor and attrition of initial entry training requirements and non-Hispanic other (6+%), Hispanic (6+%) and white (< average length of service. Reservist and National Guard enlis- 3%). Over half of the enlistees were less than 20 years old tees were excluded because longitudinal medical encounter (18–19) and a small percent (about 2%) were over 30 years data while not on active duty is not captured within the MHS old at enlistment. The majority of the study sample were in the databases. Navy (41.5%) and relatively few in the Marine Corps (4.5%). The comparison of the distribution of categorical variables The most common last recorded military occupation was in such as branch of service and race were assessed with the the communications/intelligence category (about 36%) and Chi-square test using the OpenEpi (Version 3.01) or STATA relatively few in healthcare (about 12%). (version 14) software. Incidence density rate, attributable risk PRHD diagnoses or primary outcome (ICD-9 CM 642.X) (AR) number needed to screen (NNS), and the associated con- are present in 561 SCT positive female subjects and 1,217 SCT fidence intervals were calculated with the Open Epi software. negative female subjects. Table II displays the distribution The calculations of incidence density rate and AR by SCT of subjects with PRHD medical encounters/PRHD outcomes status were intermediary steps towards calculating the NNS (eclampsia, preeclampsia and GHTN) by SCT status. There is in order to detect and potentially prevent one case of PRHD. a total of 1,778 PRHD outcomes among the entire cohort of Confidence intervals for AR% and NNS were calculated. 25,020 personnel (about 7%). Cases of eclampsia and mild- Multivariate modeling was conducted with Poisson regression severe preeclampsia were infrequent compared to GHTN and for PRHD health care utilization rates and Cox Proportional the distribution was similar (i.e., not statistically significant) Hazards modeling (CPH) for PRHD time to event. A p-value by SCT status. of <0.05 was considered statistically significant. The CPH Most (approximately 88%) of the overall PRHD outcomes analysis was conducted using STATA (version 14). were from the non-Hispanic black service women. The MILITARY MEDICINE, Vol. 185, January/February 2020 e185 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE II. Distribution of PRHD outcomes and subject versus SCT positive service women which is approximately characteristics by SCT status 12 years (results not shown). Fig. 1 displays the distribution of specific PRHD sub- Variable SCT(-) SCT(+) categories. GHTN diagnoses represent about 72% of the (n = 1,217) (n = 561) diagnoses that are considered PRHD outcomes compared to eclampsia which is about 2% of all PRHD outcomes. The n % n % distribution did not differ statistically by SCT status (p > PRHD 0.05). Fig. 2 presents the distribution of PRHD by age of Eclampsia 29 2.4 9 1.6 onset of incident PRHD medical encounter. The proportions Mild-severe Pre-eclampsia 315 25.9 137 24.4 Gestational Hypertension 873 71.8 415 74.0 by age of onset between SCT positive and negative peaked in Years to PRHD after enlistment the 20–24 age category and did not differ significantly (p > < 2 years 361 29.7 168 29.9 0.05). 2 < 7 years 670 55.0 296 52.8 Table III shows the overall crude incidence rate ratio (IRR) 7 < 12 years 158 13.0 80 14.3 by SCT status was 1.44 per 1,000 person years (95% CI of 1.30 > 12 years 28 2.3 17 3.0 Race/Ethnicity to 1.59). The IRRs were similar by branch of service except the Non-Hispanic White 20 1.6 13 2.3 Marine Corps which was not significant and were significantly Non-Hispanic Black 1074 88.2 500 89.1 higher in SCT positive versus negative service women of all Non-Hispanic Other 60 4.9 23 4.1 races/ethnicities combined and also when restricted to blacks Hispanic 63 5.2 25 4.5 only. The overall AR was 30.4% (95% CI 23.1–37.1%) and Age at Enlistment 18–19 756 62.1 336 59.9 the overall NNS to detect a case of PRHD was 156 (95% 20–24 387 31.8 188 33.5 CI 117–220). The NNS and AR% were similar by branch of 25–29 64 5.3 32 5.7 service for all races and restricted to blacks only, except for the 30+ 10 0.8 5 0.9 Marine Corps (not calculated due to rate difference less than PRHD Onset Age zero). 18–19 133 10.9 66 11.8 20–24 719 59.1 302 53.8 Table IV shows that the adjusted hazard ratio for PRHD 25–29 260 21.4 140 25.0 associated with SCT status was 1.43 (95% CI 1.30–1.58). 30+ 105 8.6 53 9.5 The risk of PRHD in SCT positive versus negative subjects Service decreases as enlistment age increases, is lower in the Air Army 334 27.4 204 36.4 Force compared to the Army and lower in Hispanic and Air Force 322 26.4 143 25.5 Marines 45 3.7 14 2.5 non-Hispanic other races compared to black service women. Navy 516 42.4 200 35.7 Adjusted PRHD IRRs based on Poisson regression models Occupation were calculated (results not shown). The overall adjusted IRR Combat/Armor/Repair 257 21.1 129 23.0 for PRHD (all covariates included in the analysis) was 1.46 Communications/Intelligence 563 46.3 279 49.7 (95% CI of 1.32–1.62). Healthcare 166 13.6 75 13.4 Other 231 19.0 78 13.9 The adjusted health care utilization rate (UR) and (URR) for PRHD by SCT status is presented in Table V. SCT positive Distribution of PRHD Outcomes by sickle cell trait (SCT) status. PRHD= versus negative subjects had higher utilization rates and rate Pregnancy Related Hypertensive Disorder ratios for all PRHD diagnoses combined and for each specific Matching factors in selection of cohort. ∗∗ Chi Square (p-value = 0.001). PHRD diagnosis. The URR by SCT status was highest for eclampsia and similar for GHTN and preeclampsia. distribution of PRHD outcomes, years to PRHD after enlistment, race/ethnicity, age at enlistment, age at PRHD DISCUSSION onset and occupation were similar by SCT status. More than This retrospective cohort analysis was based on a large, 21- 93% of all female enlistees with PRHD were less than 25 years year open cohort of 25,020 enlisted service women. The pop- old and about 60% were in the 18 to 19 years of age. About ulation was predominantly black, high school educated, and 90.5% of first PRHD outcomes occurred between the ages of a young age (90% were less than 25 years old). We found of 18 and 29 with more than 50% occurring between 20 and an approximately 43% increase in the adjusted risk of PRHD 24. The Communications/Intelligence occupation category associated with a positive SCT status. Approximately 30% of had the highest percentage of PRHD cases in both the SCT the risk of PRHD in this study sample was attributable to SCT negative and SCT positive groups. status. The NNS to detect a case of PRHD was approximately The distribution of PRHD by SCT status and branch of 160. The utilization rate for PRHD overall was approximately service showed that the Army was over-represented in SCT twice as high in SCT positive versus negative service women positive service women. The average length of military service and was higher in those with eclampsia compared to GHTN for SCT negative service women is approximately six years and preeclampsia. e186 MILITARY MEDICINE, Vol. 185, January/February 2020 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT ∗ ∗ FIGURE 1. Distribution of Specific PRHD Diagnosis by SCT Status . No significant difference in distribution between SCT positive and SCT negative within each category (p > 0.05). ∗ ∗ FIGURE 2. Distribution of PRHD Diagnosis by Age Groups at Onset and SCT Status . No significant difference between SCT positive and SCT negative within each category (p > 0.05). Our results are consistent with some prior research in the first to study the number needed to screen and health care this field. Sickle cell disease has been shown previously to utilization patterns for PRHD by SCT status. increase risk of preeclampsia (preeclampsia (RR = 2.43, 95% The United States Preventive Services Task Force (USP- CI (1.75–3.39), maternal mortality (RR = 5.98, 95% CI (1.94– STF) provides a class B recommendation (indicating with 18.44); pre-term delivery (RR = 2.21, 95% CI (1.47–3.31), moderate certainty that there is a substantial net benefit versus stillbirth (RR = 3.94, 95% CI (2.60–5.96) and small for harm for this action) for preeclampsia screening in pregnant gestational age infants (RR = 3.72, 95% CI (2.32–5.9) [16]. women with blood pressure readings throughout pregnancy. Studies have been performed to assess the association between The USPSTF also made a class B recommendation of low SCT and PRHD (GHTN, preeclampsia and eclampsia), but dose aspirin (81 mg) use beginning at 12 weeks gestation in with conflicting results [15–21]. To our knowledge, we are women at high risk of developing preeclampsia. High risk, in MILITARY MEDICINE, Vol. 185, January/February 2020 e187 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE III. PRHD Incidence Density (ID) per 1,000 PY, Incidence Rate Ratio (IRR), Attributable Risk (AR) Percentage and the Number ∗∗ Needed to Screen (NNS) by SCT Status SCT(-) I.D. (per 1,000 PY) 95% CI SCT(+) I.D. (per 1,000 PY) 95% CI IRR 95% CI AR % (95% CI) NNS 95% CI All Subjects 14.8 14.0–15.7 21.3 19.6–23.1 1.44 1.3–1.6 30.4 23.1–37.1 156 117–220 Army 15.9 14.3–17.7 24.2 21.0–27.7 1.52 1.3–1.8 34.1 21.5–44.6 122 83–222 Navy 15.4 14.1–16.8 21.5 18.6–24.7 1.40 1.2–1.6 28.4 15.6–39.1 164 107–353 Air Force 12.7 11.4–14.2 18.6 15.7–21.9 1.46 1.2–1.8 31.5 16.5–43.7 171 108–400 Marines 14.8 10.8–19.7 12.4 6.8–20.8 0.84 0.5–1.5 N/S Blacks Only All 15.2 14.3–16.1 21.9 20.0–23.9 1.44 1.3–1.6 30.6 22.8–37.6 150 113–218 Army 16.5 14.7–18.4 24.8 21.4–28.6 1.51 1.3–1.8 33.7 20.6–44.7 120 81–228 Navy 16.1 14.6–17.6 22.0 18.9–25.6 1.37 1.2–1.6 27.1 13.2–38.9 168 104–420 Air Force 12.8 11.4–14.4 19.3 16.1–22.9 1.50 1.2–1.9 33.3 17.8–45.9 156 99–362 Marines 16.7 11.6–23.2 14.0 7.0–25.1 0.84 0.4–1.7 N/C N/C Study subjects’ Incidence Density (I.D.) per 1,000 person years (PY), distributed by sickle cell trait (SCT) status. Study participants possessing sickle cell trait (SCT(+)), and participants without sickle cell trait (SCT(-)). PRHD = Pregnancy Related Hypertensive Disorder, IRR= Incidence Rate Ratio, NNS= Number Needed to Screen, AR= Attributable Risk, CI= Confidence Interval. N/C not calculated due to AR less than zero. TABLE IV. Risk (Hazard Ratio) of PRHD Associated with It is estimated that aspirin use will lower the risk of ∗∗ SCT Status preeclampsia, IUGR and preterm birth by approximately 24%, 20% and 14%, respectively. Aspirin use should be started in Variable Hazard Ratio 95 % CI pregnant women at high risk after 12 weeks of gestation and before the 16th week of gestation for optimal fetal maternal Status SCT(-) Reference protection. The anti-platelet and anti-inflammatory effects of ∗∗ SCT(+) 1.43 (1.30, 1.58) low dose aspirin aid in prevention of PRHD in pregnancies of Age Category elevated risk. The results of this study showed SCT as a risk 18–19 Reference factor for PRHD and suggests that heightened surveillance for ∗∗ 20–24 0.46 (0.41, 0.51) ∗∗ PRHD in pregnant SCT positive service members is indicated 25–29 0.20 (0.16, 0.24) ∗∗ 30 + 0.06 (0.03, 0.10) and that aspirin prophylaxis should be considered for SCT Race/Ethnicity positive pregnant service women. The association of a positive Black Reference Category SCT status with significantly higher rates of PRHD and health ∗∗∗ White 0.81 (0.57, 1.14) care utilization rates in this enlisted service women sample ∗∗∗ Hispanic 0.76 (0.61, 0.95) ∗∗∗ may be an impetus to institute universal SCT testing of at least Other 0.77 (0.62, 0.95) Branch of Service female personnel entering the service (a change in policy for Army Reference Category the Army only). Further research is needed to confirm this Navy 0.91 (0.81, 1.01) association in other SCT female populations of reproductive ∗∗ Air Force 0.75 (0.66, 0.85) age as well as to demonstrate the effectiveness of aspirin Marines 0.79 (0.61, 1.04) prophylaxis to prevent PRHD in SCT women. Year of Enlistment ∗∗ Enlistment 1.03 (1.02, 1.04) The most significant strength of this study was the ability Year to assemble a relatively large cohort of over 25,020 young service women into a longitudinal dataset to examine the asso- Risk (Hazard Ratio) of PRHD Associated with SCT Status ciation of SCT status with PRHD. This analytic dataset from Pregnancy Related Hypertensive Disorder associated with sickle cell ∗∗ trait status. multiple data sources allowed us to adjust for covariates possi- ∗∗ p < 0.001. bly associated with risk of PRHD including age and race/eth- ∗∗∗ p < 0.02. nicity. A relatively large sample size and the longitudinal MHS health record databases allowed for non-differential capture of PRHD outcomes and analysis of enlistment populations with the USPSTF recommendation statement includes a history of minimal loss to follow-up. Another strength is that there is no preeclampsia, multifetal gestation, chronic hypertension, type reason to expect a differential misclassification of pregnancy 1 or 2 diabetes, renal disease or autoimmune disease. Other status by SCT status nor a differential ascertainment of PRHD factors of consideration include nulliparity, black race, low by SCT status, given that all service women have equal access socioeconomic category, advanced maternal age and obesity. to healthcare. Universal access to health care for the study This study provides evidence that SCT may be an additional population was available and therefore limited access was not risk factor for PRHD. considered a potential confounder. e188 MILITARY MEDICINE, Vol. 185, January/February 2020 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE V. PRHD Utilization Rate (UR) and Rate Ratio (URR) by SCT Status ∗ ∗ SCT(-) UR (n = 20,016) SCT(+)UR (n = 5,004) Adjusted Poisson Model URR 95%CI PRHD 0.66 1.33 2.03 1.97–2.10 GHTN 0.53 1.13 2.03 1.97–2.10 Pre-Eclampsia 0.09 0.18 1.95 1.81–2.11 Eclampsia 0.01 0.03 2.62 2.11–3.24 ∗ ∗ Utilization rate (UR) and rate ratio (URR) of healthcare by sickle cell trait. (-) compared to sickle cell trait (+) patients (Poisson model). PRHD = Pregnancy-related. Hypertensive Disorder (PRHD); GHTN = Gestational Hypertension; CI = Confidence Interval. ∗∗ Adjusted for age, service, race and service begin year. The results of this study are limited by potentially con- CONCLUSION founding variables such as BMI and previous histories of Due to lack of consistent results, there has been mainly con- hypertension and PRHD that were not available from existing jecture surrounding the effect of SCT trait on pregnancy data sources for this analysis. Several other limitations in our risk. A well-designed study with adequate power is required analysis could have influenced our effect estimates and gen- to address this issue. The United States military healthcare eralizability of the findings. Our limited study dataset did not system (MHS) is an ideal setting to conduct such a study given include universal subjects laboratory confirmation, laboratory that longitudinal databases of personnel and medical records or medical encounter diagnosis of pregnancies, gestational age exist including laboratory confirmed SCT status and medical associated with PRHD encounters, gravity and parity history, encounters for all active duty personnel. The Army continues BMI, past history of PRHD prior to enlistment or maternal to be the only branch of the US military not employing univer- and child pregnancy outcomes. PRHD outcomes were based sal SCT laboratory testing at enlistment. However, soldiers are on International Classification Diagnosis codes and were not eventually screened as a readiness requirement either during validated by record review. We are not aware of literature pre-deployment evaluation or as part of their annual Periodic suggesting these other PRHD risk factors or diagnostic codes Health Assessment (PHA). The MHS provides an opportunity are differentially distributed by SCT status, suggesting that to effectively investigate SCT and its reproductive and other these limitations likely may not have introduced a differential health consequences in a relatively large enlisted population bias. It is assumed that all service women were not pregnant at within a military health care system that provides equal access enlistment because of universal enlistment pregnancy testing. to healthcare. Another limitation is that this study included women who voluntarily enlisted in the U.S. military. These women self- SUPPLEMENTARY MATERIAL selected and met the required personnel and medical standards Supplementary material is available at Military Medicine online. and so may not be representative of the overall general popu- lation of reproductive age women in terms of risk of PRHD. The results of our study suggest a significant increase CONFLICTS OF INTEREST The authors declare that they have no conflicts of interest. in the risk of developing PRHD among SCT positive ser- vice women. The health care utilization of SCT positive enlisted women was significantly higher than for SCT neg- ative enlisted women. However, there is a need for additional ACKNOWLEDGMENTS The authors thank Lee Hurt DrPH, MPH of the Armed Forces Health study on this topic, in particular, inclusion of other known risk Surveillance Branch and Gosia Nowak MSc, MPH of the Navy and Marine factors and potential confounders such as BMI, gravity/parity Corps Public Health Center for creation of the study analytic data set. The and associated maternal and fetal outcomes. If the results of authors wish to thank Ligong Chen M.D., Stephanie Shao M.P.H., and Cara this study are reproduced in other study populations, preg- Olsen Dr PH. for assistance in creating the analytic dataset and providing nant woman with SCT positive status should be considered statistical consultation. We have obtained written permission from all persons named in the Acknowledgment. at increased risk for PRHD and potentially offered aspirin prophylaxis for preeclampsia. Further studies of the effective- ness of aspirin chemoprophylaxis to prevent PRHD in SCT FUNDING positive pregnant service women are indicated, as well as cost- The study was funded by an Interagency Agreement with the National effectiveness analysis to determine whether universal versus Heart, Lung and Blood Institute (NHLBI). Data from the Defense Medical selective (i.e., based on military occupation per Army policy) Surveillance System, The Armed Forces Health Surveillance Branch, U.S. enlistment SCT screening of service women is indicated to Department of Defense, Silver Spring, Maryland; data period 1992–2012; prevent associated PRHD morbidity and potentially mortality. data released Oct 27, 2014. MILITARY MEDICINE, Vol. 185, January/February 2020 e189 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT REFERENCES of delivering a small-for-gestational-age neonate. Hypertens Pregnancy 2016; 35(3): 350–60. 1. American Society of Hematology: Education For Patients: Sickle Cell 13. Dulay AT: Gynecology and obstetrics: preeclampsia and eclampsia. In: Trait. Available at http://www.hematology.org/Patients/Anemia/Sickle- The Merck Manuals. Edited by Porter RS, Kaplan JL Kenilworth, NJ, Cell-Trait.aspx; Accessed Sep 29, 2018. USA, Merck Sharp & Dohme Corp, 2018. https://www.merckmanuals. 2. Ojodu J, Hulihan MM, Pope SN, Grant AM, Centers for Disease Control com/professional/gynecology-and-obstetrics/abnormalities-of- and Prevention (CDC): Incidence of sickle cell trait–United States, 2010. pregnancy/preeclampsia-and-eclampsia#. Accessed October 2, 2018. MMWR Morb Mortal Wkly Rep 2014; 63(49): 1155–8. 14. Wilson BJ, Watson MS, Prescott GJ, et al: Hypertensive diseases of 3. Niebuhr DW, Chen L, Shao S, Goldsmith J, Byrne C, Singer DE: Sickle pregnancy and risk of hypertension and stroke in later life: results from cell trait prevalence among U.S. military service members: 1992–2012. cohort study. BMJ 2003; 326(7394): 845. Mil Med 2017; 182(3): e1819–e1824. 15. Kotila TR: Sickle cell trait: a benign state? Acta Haematol 2016; 136(3): 4. American College of Obstetricians and Gynecologists: Task Force on 147–51. Epub 2016 Jul 16. Review. Hypertension in Pregnancy. Hypertension in pregnancy. Report of the 16. Tsaras G, Owusu-Ansah A, Boateng FO, Amoateng-Adjepong Y: Com- American College of Obstetricians and Gynecologists’ Task Force on plications associated with sickle cell trait: a brief narrative review. Am J Hypertension in Pregnancy. Obstet Gynecol 2013; 122(5): 1122–31. Med 2009; 122(6): 507–12. 5. Ananth CV, Keyes KM, Wapner RJ: Pre-eclampsia rates in the United 17. Jans SM, de Jonge A, Lagro-Janssen AL: Maternal and perinatal out- States, 1980–2010: age-period-cohort analysis. BMJ 2013; 347: f6564. comes amongst haemoglobinopathy carriers: a systematic review. Int J 6. Leeman L, Fontaine P: Hypertensive disorders of pregnancy. Am Fam Clin Pract 2010; 64(12): 1688–98. Review. Physician 2008; 78(1): 93–100. Review. 18. Larrabee KD, Monga M: Women with sickle cell trait are at increased 7. Paré E, Parry S, McElrath TF, Pucci D, Newton A, Lim KH: Clinical risk for preeclampsia. Am J Obstet Gynecol 1997; 177: 425–8. risk factors for preeclampsia in the 21st century. Obstet Gynecol 2014; 19. Tita AT, Biggio JR, Chapman V, Neely C, Rouse DJ: Perinatal and 124(4): 763–70. maternal outcomes in women with sickle or hemoglobin C trait. Obstet 8. Bartsch E, Medcalf KE, Park AL, Ray JG, High Risk of Pre-eclampsia Gynecol 2007; 110(5): 1113–9. Identification Group: Clinical risk factors for pre-eclampsia determined 20. Whalley PJ, Pritchard JA, Richards JR Jr: Sickle cell trait and pregnancy. in early pregnancy: systematic review and meta-analysis of large cohort JAMA 1963; 28(186): 1132–5. studies. BMJ 2016; 353: i1753. Review. 21. Taylor MY, Wyatt-Ashmead J, Gray J, Bofill JA, Martin R, Morrison 9. Irwin DE, Savitz DA, Hertz-Picciotto I St, André KA: The risk of JC: Pregnancy loss after first-trimester viability in women with sickle pregnancy-induced hypertension: black and white differences in a mili- cell trait: time for a reappraisal? Am J Obstet Gynecol 2006; 194(6): tary population. Am J Public Health 1994; 84(9): 1508–10. 1604–8. Epub 2006 Apr 25. 10. MacKay AP, Berg CJ, Atrash HK: Pregnancy-related mortality 22. Taylor MY, Wyatt-Ashmead J, Gray J, Bofill JA, Martin RW, Mor- from preeclampsia and eclampsia. Obstet Gynecol 2001; 97(4): rison JC: Pregnancy loss after first trimester viability in women with 533–8. sickle cell trait: a preliminary report. South Med J 2008; 101(2): 11. Wikström AK, Gunnarsdottir J, Nelander M, Simic M, Stephansson 150–1. O, Cnattingius S: Prehypertension in pregnancy and risks of small for gestational age infant and stillbirth. Hypertension 2016; 67(3): 640–6. 23. Stamilio DM, Sehdev HM, Macones GA: Pregnant women with the 12. Block-Abraham DM, Adamovich D, Turan OM, Doyle LE, Blitzer MG, sickle cell trait are not at increased risk for developing preeclampsia. Baschat AA: Maternal blood pressures during pregnancy and the risk Am J Perinatol 2003; 20(1): 41–8. e190 MILITARY MEDICINE, Vol. 185, January/February 2020 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Military Medicine Oxford University Press

The Risk of Pregnancy Related Hypertension Disorder Associated with Sickle Cell Trait in U.S. Service Women

Download PDF
8 pages

Loading next page...
 
/lp/oxford-university-press/the-risk-of-pregnancy-related-hypertension-disorder-associated-with-XP0AS6k2Ye

References (26)

Publisher
Oxford University Press
Copyright
Copyright © 2022 The Society of Federal Health Professionals
ISSN
0026-4075
eISSN
1930-613X
DOI
10.1093/milmed/usz143
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 MILITARY MEDICINE, 185, 1/2:e183, 2020 The Risk of Pregnancy Related Hypertension Disorder Associated with Sickle Cell Trait in U.S. Service Women MAJ Christina O’Hara, MD, MPH, MC, USA*; CAPT Darrell E. Singer, MD, MPH, MC, USPHS*; COL David W. Niebuhr, MD, MPH, MSc, MC, USA (Ret.)* ABSTRACT Introduction The health impact of the heterozygous (Hb Ss) Sickle Cell Trait (SCT) among the estimated one million females of reproductive age in the United States on pregnancy related hypertensive disorders (PRHD) including preeclampsia is not as well understood or researched as Sickle Cell Disease (SCD). Materials and Methods This retrospective cohort study included 25,020 U.S. enlisted, active duty service women during 1992 to 2013. Race within the sample is primarily black (84%) given that blacks have the highest prevalence of SCT. All 5,004 SCT positive individuals and a matched sample of 20,016 from those considered SCT negative were followed while on active duty for PRHD outcomes including gestational hypertension (GHTN), preeclampsia and eclampsia. Results The adjusted incidence rate ratio (IRR) for any PRHD in SCT positive compared to negative individuals was 1.46 (95% CI 1.32 – 1.62). SCT positive versus negative subjects had higher health care utilization rates and utilization rate ratios (URR) for all PRHD diagnoses combined 2.03 (95%CI: 1.97 – 2.10) and for each specific PHRD diagnosis. The prevalence of preeclampsia or eclampsia did not vary by SCT status. The overall PRHD attributable risk due to SCT was 30.4% (95% CI 23.1–37.1%) and the overall number needed to screen to detect a case of PRHD was 156 (95% CI 117–220). Conclusions The results of this study indicate that being SCT positive is a risk factor for PRHD and provides evidence that SCT status may have an adverse effect on reproductive health. Future research needs to include other known risk factors for PRHD to include gravity and parity history, BMI, past history of PRHD prior to enlistment as well as to examine the association with maternal/child pregnancy outcomes. INTRODUCTION active duty enlisted Hispanic-Americans were SCT positive. Sickle cell trait (SCT) is a heterozygous medical condition The prevalence in females compared to males was 3% versus affecting an estimated three million people in the United States 1% respectively. and 8–10% of blacks. Data from state newborn screening Pregnancy related hypertension disorder (PRHD) is esti- programs estimated the 2010 incidence for sickle cell trait in mated to affect up to 10% of all pregnancies. It is one of the the United States at 15.5 cases per 1,000 births. The heaviest leading causes of maternal morbidity and mortality in the U.S. burden was among black Americans at 73.1 cases of SCT per and world-wide. Accepted classifications of PRHD includes 1,000 births compared to 3.0 cases per 1,000 white newborns, the following four categories: 1) preeclampsia/eclampsia, 2) and 6.9 cases per 1,000 Hispanic newborns. In the U.S. chronic hypertension (any cause), 3) chronic hypertension military, the prevalence of SCT among service members has with superimposed preeclampsia and 4) gestational hyperten- increased steadily between 1992 and 2012. Approximately sion. Over the past 20 years, incidence of preeclampsia has 4% black active duty enlisted service members and 0.5% in increased over 25% in the U.S., affecting 3–7% of all pregnan- cies, with an estimated 1/200 cases of these cases progressing 4,5 Department of Preventive Medicine and Biostatistics, Uniformed Ser- to eclampsia. Risk factors for the development of PRHD vices of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814. include age > 40 years, chronic hypertension, nulliparity, The opinions and assertions expressed in this article represent the private pre-gestational diabetes, black race, history of preeclampsia, views of the authors and do not reflect the official views of the Uniformed and BMI > 25 which showed the greatest attributable risk Services University of the Health Sciences, the U.S. Departments of the Army, 6–8 (64%). A 2014 study of 2,288 females receiving Veter- Navy, or Defense, National Heart, Lung, and Blood Institute, Food and Drug Administration or the U.S. Government. Nothing in the presentation implies ans Administration maternity benefits demonstrated a signifi- any Federal/Department of Defense endorsement. cantly higher standardized incidence rate (SIR) of developing doi:10.1093/milmed/usz143 gestational diabetes mellitus (SIR: 1.40; 95% [CI] 1.16, 1.68) Published by Oxford University Press on behalf of the Association of and hypertensive disorder (SIR: 1.32; 95% CI 1.15, 1.51) as Military Surgeons of the United States 2019. This work is written by (a) US compared to civilian women delivering in the United States. Government employee(s) and is in the public domain in the US. MILITARY MEDICINE, Vol. 185, January/February 2020 e183 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT The risks of PHRD to maternal-fetal health are high. The conditions and included 75,400 military subjects of which overall preeclampsia-eclampsia case fatality rate based on 25,020 were female. The primary purpose of the parent study Centers for Disease Control and Prevention data from 1988 to was to evaluate associations of SCT with exertional health 1992 was 6.4 per 10,000 cases at delivery. Maternal and fetal illness outcomes. However, the parent study also collected outcomes of PRHD can include intra-uterine growth restric- additional health outcome data to include PRHD conditions. tion (IUGR), pre-term labor and/or delivery, abruptio placen- The data sources for the parent study came from the Armed tae, stillbirth and maternal and/or fetal death. Prehyperten- Forces Health Surveillance Branch (AFHSB) Defense Med- sion carries a 70% increased stillbirth risk and 69% increased ical Surveillance System and the Navy and Marine Corps risk for small-for-gestational-age infants (SGA) when prehy- Public Health Center (NMCPHC). AFHSB data included per- 11,12 pertension occurs in late pregnancy (36 weeks). Without sonnel (i.e., enlistment, military occupation and discharge) immediate treatment, eclampsia is usually fatal. Long-term and medical encounter (i.e., hospitalization and ambulatory) effects in women with a history of all types of PRHD have not data. The NMCPHC provided SCT laboratory test result data been well studied, but a history of gestational hypertension, (qualitative or quantitative). Data elements were de-identified preeclampsia or eclampsia may be associated with hyperten- prior to transfer to the USUHS SCT study principal investiga- sion and other cardiovascular disease later in life. tor. Data elements for this study include the following: Controversy surrounds whether SCT increases risk to the Demographic data: race/ethnicity, gender, service, age cat- mother or fetus during pregnancy. Studies linking SCT with egory on enlistment (18–19, 20–24, 25–29, 30+), deploy- negative health outcomes suggest an increased risk of dis- ments (dates of deployment and redeployment), home of orders involving the cardiovascular, renal and musculoskele- record (first 3 digits of zip code), quarter and year of tal systems, while exertional related injuries are among the enlistment. 15,16 most common. However, in one systematic review of Medical encounter data: dates (in days relative to enlist- cohort and case-control studies, examining maternal com- ment) of all SCT ambulatory and hospitalization diagnostic plications including urinary tract infection (UTI), anemia, encounter codes of ICD-9 CM 282.5. (pre-)eclampsia, gestational diabetes, and premature labor, the The outcome of interest for this project includes all ICD- risk to the fetus was considered low. Another study showed a 9 encounter codes for PRHD (ICD-CM 642.X- Hypertension higher risk of pre-eclampsia among women with SCT and two 18–20 complicating pregnancy, childbirth, and the puerperium). Sup- studies with a small association with bacterial infection. plemental Table 1 provides the specific four digit ICD-9 CM Two recent studies observed a significantly shorter duration codes included. of pregnancy and increased fetal loss among SCT affected Descriptive statistics for the variables of interest were women as compare to controls and a third study concluded no 21–23 determined by SCT status and PRHD outcome. Descriptive increased perinatal risk to the SCT affected mother. The statistics of PRHD by ICD-9 codes of eclampsia, preeclampsia pathophysiologic basis for this association in the literature is and GHTN were tabulated and displayed by SCT status. hypothesized to be increased sickling and poorly developed Selected data from the parent retrospective cohort study uterine placental spiral arterioles resulting in an inadequate was utilized to evaluate the association between SCT status in blood supply required for fetal development. active duty enlisted U.S. service women from 1992 through SCT and PRHD are both each relatively common condi- 2012 and followed for PRHD outcomes through 2013. To tions, particularly in black women. They are also associated accommodate incomplete years of service among the enlisted with significant morbidity which can impact readiness and personnel, incidence density rates and hazard ratios were cal- healthcare costs. The objective of this study was to study the culated. The unadjusted incidence density rate, rate ratio and association of SCT status with PRHD in U.S. enlisted service attributable risk of PRHD by SCT status and adjusted hazard women. ratio were calculated with Cox Proportional Hazard models over all enlistment cohort years. The number of enlistees needed to screen (NNS) for SCT to detect a case of PRHD METHODS The parent study (PMB-87–291, Version 3, 31 May 2015) for was calculated. The retrospective enlistment cohort data were this project is a retrospective, matched, active duty, enlisted further analyzed to estimate the risk of several related diseases cohort with SCT status as the exposure of interest. All active within PRHD (eclampsia, preeclampsia, and GHTN) in ser- duty enlisted military service members with evidence of vice women by SCT status. Health care utilization rates (UR) being positive for SCT from laboratory test report and/or and adjusted utilization rate ratios (URR) for PRHD overall a medical encounter (1 hospitalization and/or 2 or more and by each of the subcategories of PRHD were calculated ambulatory visits), were identified and matched with four with Poisson models. enlisted military service members without evidence of SCT The study population included a total of 25,020 female by branch of service, gender, race/ethnicity, and quarter and subjects 5,004 SCT positive and 20,016 SCT negative. Offi- year of enlistment. The parent study utilized data from 1992 cers were excluded from this study due to their generally through 2013 on SCT status and various SCT-related medical higher socioeconomic status (and associated better access to e184 MILITARY MEDICINE, Vol. 185, January/February 2020 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE I. Distribution of study subjects by SCT status Variable SCT(-) n % of SCT(-) SCT (+) n %ofSCT(+) Race/Ethnicity Non-Hispanic White 568 2.8 142 2.8 Non-Hispanic Black 16874 84.3 4219 84.3 Non-Hispanic Other 1290 6.4 322 6.4 Hispanic 1284 6.4 321 6.4 Age at Enlistment 18–19 11129 55.6 2731 54.6 20–24 7110 35.5 1826 36.5 25–29 1354 6.8 334 6.7 30+ 423 2.1 113 2.3 Service Army 5796 29.0 1448 28.9 Air Force 5012 25.0 1254 25.1 Marines 904 4.5 226 4.5 Navy 8304 41.5 2076 41.5 Enlistment Year 1992–1994 1208 6.0 302 6.0 1995–1999 3528 17.6 882 17.6 2000–2004 5780 28.9 1445 28.9 2005–2009 5936 29.7 1484 29.7 2010–2012 3564 17.8 891 17.8 ∗∗ Occupation Combat/Armor/Repair 4999 25.0 1270 25.4 Communications/Intelligence 7087 35.4 1915 38.3 Healthcare 2464 12.3 608 12.2 Other 5466 27.3 1211 24.2 Distribution of study subjects (n = 25,020) by sickle cell trait (SCT) status. Study participants possessing sickle cell trait (SCT(+)) (n = 5,004), and participants without sickle cell trait (SCT(-)) (n = 20,016). Matching factors in selection of cohort. ∗∗ Chi Square (p-value = 0.001). healthcare prior to military service), differing accession routes RESULTS to service (i.e., service academy, Reserve Officers’ Training Table I displays the distribution of race, age at enlistment, Corps) and potential overlap with enlisted who may later service and occupation at separation by SCT status. The attend Officer Candidate School, differences in the length, racial/ethnic distribution within the study was black (84%), rigor and attrition of initial entry training requirements and non-Hispanic other (6+%), Hispanic (6+%) and white (< average length of service. Reservist and National Guard enlis- 3%). Over half of the enlistees were less than 20 years old tees were excluded because longitudinal medical encounter (18–19) and a small percent (about 2%) were over 30 years data while not on active duty is not captured within the MHS old at enlistment. The majority of the study sample were in the databases. Navy (41.5%) and relatively few in the Marine Corps (4.5%). The comparison of the distribution of categorical variables The most common last recorded military occupation was in such as branch of service and race were assessed with the the communications/intelligence category (about 36%) and Chi-square test using the OpenEpi (Version 3.01) or STATA relatively few in healthcare (about 12%). (version 14) software. Incidence density rate, attributable risk PRHD diagnoses or primary outcome (ICD-9 CM 642.X) (AR) number needed to screen (NNS), and the associated con- are present in 561 SCT positive female subjects and 1,217 SCT fidence intervals were calculated with the Open Epi software. negative female subjects. Table II displays the distribution The calculations of incidence density rate and AR by SCT of subjects with PRHD medical encounters/PRHD outcomes status were intermediary steps towards calculating the NNS (eclampsia, preeclampsia and GHTN) by SCT status. There is in order to detect and potentially prevent one case of PRHD. a total of 1,778 PRHD outcomes among the entire cohort of Confidence intervals for AR% and NNS were calculated. 25,020 personnel (about 7%). Cases of eclampsia and mild- Multivariate modeling was conducted with Poisson regression severe preeclampsia were infrequent compared to GHTN and for PRHD health care utilization rates and Cox Proportional the distribution was similar (i.e., not statistically significant) Hazards modeling (CPH) for PRHD time to event. A p-value by SCT status. of <0.05 was considered statistically significant. The CPH Most (approximately 88%) of the overall PRHD outcomes analysis was conducted using STATA (version 14). were from the non-Hispanic black service women. The MILITARY MEDICINE, Vol. 185, January/February 2020 e185 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE II. Distribution of PRHD outcomes and subject versus SCT positive service women which is approximately characteristics by SCT status 12 years (results not shown). Fig. 1 displays the distribution of specific PRHD sub- Variable SCT(-) SCT(+) categories. GHTN diagnoses represent about 72% of the (n = 1,217) (n = 561) diagnoses that are considered PRHD outcomes compared to eclampsia which is about 2% of all PRHD outcomes. The n % n % distribution did not differ statistically by SCT status (p > PRHD 0.05). Fig. 2 presents the distribution of PRHD by age of Eclampsia 29 2.4 9 1.6 onset of incident PRHD medical encounter. The proportions Mild-severe Pre-eclampsia 315 25.9 137 24.4 Gestational Hypertension 873 71.8 415 74.0 by age of onset between SCT positive and negative peaked in Years to PRHD after enlistment the 20–24 age category and did not differ significantly (p > < 2 years 361 29.7 168 29.9 0.05). 2 < 7 years 670 55.0 296 52.8 Table III shows the overall crude incidence rate ratio (IRR) 7 < 12 years 158 13.0 80 14.3 by SCT status was 1.44 per 1,000 person years (95% CI of 1.30 > 12 years 28 2.3 17 3.0 Race/Ethnicity to 1.59). The IRRs were similar by branch of service except the Non-Hispanic White 20 1.6 13 2.3 Marine Corps which was not significant and were significantly Non-Hispanic Black 1074 88.2 500 89.1 higher in SCT positive versus negative service women of all Non-Hispanic Other 60 4.9 23 4.1 races/ethnicities combined and also when restricted to blacks Hispanic 63 5.2 25 4.5 only. The overall AR was 30.4% (95% CI 23.1–37.1%) and Age at Enlistment 18–19 756 62.1 336 59.9 the overall NNS to detect a case of PRHD was 156 (95% 20–24 387 31.8 188 33.5 CI 117–220). The NNS and AR% were similar by branch of 25–29 64 5.3 32 5.7 service for all races and restricted to blacks only, except for the 30+ 10 0.8 5 0.9 Marine Corps (not calculated due to rate difference less than PRHD Onset Age zero). 18–19 133 10.9 66 11.8 20–24 719 59.1 302 53.8 Table IV shows that the adjusted hazard ratio for PRHD 25–29 260 21.4 140 25.0 associated with SCT status was 1.43 (95% CI 1.30–1.58). 30+ 105 8.6 53 9.5 The risk of PRHD in SCT positive versus negative subjects Service decreases as enlistment age increases, is lower in the Air Army 334 27.4 204 36.4 Force compared to the Army and lower in Hispanic and Air Force 322 26.4 143 25.5 Marines 45 3.7 14 2.5 non-Hispanic other races compared to black service women. Navy 516 42.4 200 35.7 Adjusted PRHD IRRs based on Poisson regression models Occupation were calculated (results not shown). The overall adjusted IRR Combat/Armor/Repair 257 21.1 129 23.0 for PRHD (all covariates included in the analysis) was 1.46 Communications/Intelligence 563 46.3 279 49.7 (95% CI of 1.32–1.62). Healthcare 166 13.6 75 13.4 Other 231 19.0 78 13.9 The adjusted health care utilization rate (UR) and (URR) for PRHD by SCT status is presented in Table V. SCT positive Distribution of PRHD Outcomes by sickle cell trait (SCT) status. PRHD= versus negative subjects had higher utilization rates and rate Pregnancy Related Hypertensive Disorder ratios for all PRHD diagnoses combined and for each specific Matching factors in selection of cohort. ∗∗ Chi Square (p-value = 0.001). PHRD diagnosis. The URR by SCT status was highest for eclampsia and similar for GHTN and preeclampsia. distribution of PRHD outcomes, years to PRHD after enlistment, race/ethnicity, age at enlistment, age at PRHD DISCUSSION onset and occupation were similar by SCT status. More than This retrospective cohort analysis was based on a large, 21- 93% of all female enlistees with PRHD were less than 25 years year open cohort of 25,020 enlisted service women. The pop- old and about 60% were in the 18 to 19 years of age. About ulation was predominantly black, high school educated, and 90.5% of first PRHD outcomes occurred between the ages of a young age (90% were less than 25 years old). We found of 18 and 29 with more than 50% occurring between 20 and an approximately 43% increase in the adjusted risk of PRHD 24. The Communications/Intelligence occupation category associated with a positive SCT status. Approximately 30% of had the highest percentage of PRHD cases in both the SCT the risk of PRHD in this study sample was attributable to SCT negative and SCT positive groups. status. The NNS to detect a case of PRHD was approximately The distribution of PRHD by SCT status and branch of 160. The utilization rate for PRHD overall was approximately service showed that the Army was over-represented in SCT twice as high in SCT positive versus negative service women positive service women. The average length of military service and was higher in those with eclampsia compared to GHTN for SCT negative service women is approximately six years and preeclampsia. e186 MILITARY MEDICINE, Vol. 185, January/February 2020 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT ∗ ∗ FIGURE 1. Distribution of Specific PRHD Diagnosis by SCT Status . No significant difference in distribution between SCT positive and SCT negative within each category (p > 0.05). ∗ ∗ FIGURE 2. Distribution of PRHD Diagnosis by Age Groups at Onset and SCT Status . No significant difference between SCT positive and SCT negative within each category (p > 0.05). Our results are consistent with some prior research in the first to study the number needed to screen and health care this field. Sickle cell disease has been shown previously to utilization patterns for PRHD by SCT status. increase risk of preeclampsia (preeclampsia (RR = 2.43, 95% The United States Preventive Services Task Force (USP- CI (1.75–3.39), maternal mortality (RR = 5.98, 95% CI (1.94– STF) provides a class B recommendation (indicating with 18.44); pre-term delivery (RR = 2.21, 95% CI (1.47–3.31), moderate certainty that there is a substantial net benefit versus stillbirth (RR = 3.94, 95% CI (2.60–5.96) and small for harm for this action) for preeclampsia screening in pregnant gestational age infants (RR = 3.72, 95% CI (2.32–5.9) [16]. women with blood pressure readings throughout pregnancy. Studies have been performed to assess the association between The USPSTF also made a class B recommendation of low SCT and PRHD (GHTN, preeclampsia and eclampsia), but dose aspirin (81 mg) use beginning at 12 weeks gestation in with conflicting results [15–21]. To our knowledge, we are women at high risk of developing preeclampsia. High risk, in MILITARY MEDICINE, Vol. 185, January/February 2020 e187 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE III. PRHD Incidence Density (ID) per 1,000 PY, Incidence Rate Ratio (IRR), Attributable Risk (AR) Percentage and the Number ∗∗ Needed to Screen (NNS) by SCT Status SCT(-) I.D. (per 1,000 PY) 95% CI SCT(+) I.D. (per 1,000 PY) 95% CI IRR 95% CI AR % (95% CI) NNS 95% CI All Subjects 14.8 14.0–15.7 21.3 19.6–23.1 1.44 1.3–1.6 30.4 23.1–37.1 156 117–220 Army 15.9 14.3–17.7 24.2 21.0–27.7 1.52 1.3–1.8 34.1 21.5–44.6 122 83–222 Navy 15.4 14.1–16.8 21.5 18.6–24.7 1.40 1.2–1.6 28.4 15.6–39.1 164 107–353 Air Force 12.7 11.4–14.2 18.6 15.7–21.9 1.46 1.2–1.8 31.5 16.5–43.7 171 108–400 Marines 14.8 10.8–19.7 12.4 6.8–20.8 0.84 0.5–1.5 N/S Blacks Only All 15.2 14.3–16.1 21.9 20.0–23.9 1.44 1.3–1.6 30.6 22.8–37.6 150 113–218 Army 16.5 14.7–18.4 24.8 21.4–28.6 1.51 1.3–1.8 33.7 20.6–44.7 120 81–228 Navy 16.1 14.6–17.6 22.0 18.9–25.6 1.37 1.2–1.6 27.1 13.2–38.9 168 104–420 Air Force 12.8 11.4–14.4 19.3 16.1–22.9 1.50 1.2–1.9 33.3 17.8–45.9 156 99–362 Marines 16.7 11.6–23.2 14.0 7.0–25.1 0.84 0.4–1.7 N/C N/C Study subjects’ Incidence Density (I.D.) per 1,000 person years (PY), distributed by sickle cell trait (SCT) status. Study participants possessing sickle cell trait (SCT(+)), and participants without sickle cell trait (SCT(-)). PRHD = Pregnancy Related Hypertensive Disorder, IRR= Incidence Rate Ratio, NNS= Number Needed to Screen, AR= Attributable Risk, CI= Confidence Interval. N/C not calculated due to AR less than zero. TABLE IV. Risk (Hazard Ratio) of PRHD Associated with It is estimated that aspirin use will lower the risk of ∗∗ SCT Status preeclampsia, IUGR and preterm birth by approximately 24%, 20% and 14%, respectively. Aspirin use should be started in Variable Hazard Ratio 95 % CI pregnant women at high risk after 12 weeks of gestation and before the 16th week of gestation for optimal fetal maternal Status SCT(-) Reference protection. The anti-platelet and anti-inflammatory effects of ∗∗ SCT(+) 1.43 (1.30, 1.58) low dose aspirin aid in prevention of PRHD in pregnancies of Age Category elevated risk. The results of this study showed SCT as a risk 18–19 Reference factor for PRHD and suggests that heightened surveillance for ∗∗ 20–24 0.46 (0.41, 0.51) ∗∗ PRHD in pregnant SCT positive service members is indicated 25–29 0.20 (0.16, 0.24) ∗∗ 30 + 0.06 (0.03, 0.10) and that aspirin prophylaxis should be considered for SCT Race/Ethnicity positive pregnant service women. The association of a positive Black Reference Category SCT status with significantly higher rates of PRHD and health ∗∗∗ White 0.81 (0.57, 1.14) care utilization rates in this enlisted service women sample ∗∗∗ Hispanic 0.76 (0.61, 0.95) ∗∗∗ may be an impetus to institute universal SCT testing of at least Other 0.77 (0.62, 0.95) Branch of Service female personnel entering the service (a change in policy for Army Reference Category the Army only). Further research is needed to confirm this Navy 0.91 (0.81, 1.01) association in other SCT female populations of reproductive ∗∗ Air Force 0.75 (0.66, 0.85) age as well as to demonstrate the effectiveness of aspirin Marines 0.79 (0.61, 1.04) prophylaxis to prevent PRHD in SCT women. Year of Enlistment ∗∗ Enlistment 1.03 (1.02, 1.04) The most significant strength of this study was the ability Year to assemble a relatively large cohort of over 25,020 young service women into a longitudinal dataset to examine the asso- Risk (Hazard Ratio) of PRHD Associated with SCT Status ciation of SCT status with PRHD. This analytic dataset from Pregnancy Related Hypertensive Disorder associated with sickle cell ∗∗ trait status. multiple data sources allowed us to adjust for covariates possi- ∗∗ p < 0.001. bly associated with risk of PRHD including age and race/eth- ∗∗∗ p < 0.02. nicity. A relatively large sample size and the longitudinal MHS health record databases allowed for non-differential capture of PRHD outcomes and analysis of enlistment populations with the USPSTF recommendation statement includes a history of minimal loss to follow-up. Another strength is that there is no preeclampsia, multifetal gestation, chronic hypertension, type reason to expect a differential misclassification of pregnancy 1 or 2 diabetes, renal disease or autoimmune disease. Other status by SCT status nor a differential ascertainment of PRHD factors of consideration include nulliparity, black race, low by SCT status, given that all service women have equal access socioeconomic category, advanced maternal age and obesity. to healthcare. Universal access to health care for the study This study provides evidence that SCT may be an additional population was available and therefore limited access was not risk factor for PRHD. considered a potential confounder. e188 MILITARY MEDICINE, Vol. 185, January/February 2020 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT TABLE V. PRHD Utilization Rate (UR) and Rate Ratio (URR) by SCT Status ∗ ∗ SCT(-) UR (n = 20,016) SCT(+)UR (n = 5,004) Adjusted Poisson Model URR 95%CI PRHD 0.66 1.33 2.03 1.97–2.10 GHTN 0.53 1.13 2.03 1.97–2.10 Pre-Eclampsia 0.09 0.18 1.95 1.81–2.11 Eclampsia 0.01 0.03 2.62 2.11–3.24 ∗ ∗ Utilization rate (UR) and rate ratio (URR) of healthcare by sickle cell trait. (-) compared to sickle cell trait (+) patients (Poisson model). PRHD = Pregnancy-related. Hypertensive Disorder (PRHD); GHTN = Gestational Hypertension; CI = Confidence Interval. ∗∗ Adjusted for age, service, race and service begin year. The results of this study are limited by potentially con- CONCLUSION founding variables such as BMI and previous histories of Due to lack of consistent results, there has been mainly con- hypertension and PRHD that were not available from existing jecture surrounding the effect of SCT trait on pregnancy data sources for this analysis. Several other limitations in our risk. A well-designed study with adequate power is required analysis could have influenced our effect estimates and gen- to address this issue. The United States military healthcare eralizability of the findings. Our limited study dataset did not system (MHS) is an ideal setting to conduct such a study given include universal subjects laboratory confirmation, laboratory that longitudinal databases of personnel and medical records or medical encounter diagnosis of pregnancies, gestational age exist including laboratory confirmed SCT status and medical associated with PRHD encounters, gravity and parity history, encounters for all active duty personnel. The Army continues BMI, past history of PRHD prior to enlistment or maternal to be the only branch of the US military not employing univer- and child pregnancy outcomes. PRHD outcomes were based sal SCT laboratory testing at enlistment. However, soldiers are on International Classification Diagnosis codes and were not eventually screened as a readiness requirement either during validated by record review. We are not aware of literature pre-deployment evaluation or as part of their annual Periodic suggesting these other PRHD risk factors or diagnostic codes Health Assessment (PHA). The MHS provides an opportunity are differentially distributed by SCT status, suggesting that to effectively investigate SCT and its reproductive and other these limitations likely may not have introduced a differential health consequences in a relatively large enlisted population bias. It is assumed that all service women were not pregnant at within a military health care system that provides equal access enlistment because of universal enlistment pregnancy testing. to healthcare. Another limitation is that this study included women who voluntarily enlisted in the U.S. military. These women self- SUPPLEMENTARY MATERIAL selected and met the required personnel and medical standards Supplementary material is available at Military Medicine online. and so may not be representative of the overall general popu- lation of reproductive age women in terms of risk of PRHD. The results of our study suggest a significant increase CONFLICTS OF INTEREST The authors declare that they have no conflicts of interest. in the risk of developing PRHD among SCT positive ser- vice women. The health care utilization of SCT positive enlisted women was significantly higher than for SCT neg- ative enlisted women. However, there is a need for additional ACKNOWLEDGMENTS The authors thank Lee Hurt DrPH, MPH of the Armed Forces Health study on this topic, in particular, inclusion of other known risk Surveillance Branch and Gosia Nowak MSc, MPH of the Navy and Marine factors and potential confounders such as BMI, gravity/parity Corps Public Health Center for creation of the study analytic data set. The and associated maternal and fetal outcomes. If the results of authors wish to thank Ligong Chen M.D., Stephanie Shao M.P.H., and Cara this study are reproduced in other study populations, preg- Olsen Dr PH. for assistance in creating the analytic dataset and providing nant woman with SCT positive status should be considered statistical consultation. We have obtained written permission from all persons named in the Acknowledgment. at increased risk for PRHD and potentially offered aspirin prophylaxis for preeclampsia. Further studies of the effective- ness of aspirin chemoprophylaxis to prevent PRHD in SCT FUNDING positive pregnant service women are indicated, as well as cost- The study was funded by an Interagency Agreement with the National effectiveness analysis to determine whether universal versus Heart, Lung and Blood Institute (NHLBI). Data from the Defense Medical selective (i.e., based on military occupation per Army policy) Surveillance System, The Armed Forces Health Surveillance Branch, U.S. enlistment SCT screening of service women is indicated to Department of Defense, Silver Spring, Maryland; data period 1992–2012; prevent associated PRHD morbidity and potentially mortality. data released Oct 27, 2014. MILITARY MEDICINE, Vol. 185, January/February 2020 e189 Downloaded from https://academic.oup.com/milmed/article/185/1-2/e183/5524310 by DeepDyve user on 13 July 2022 Risk of Pregnancy Related Hypertension with SCT REFERENCES of delivering a small-for-gestational-age neonate. Hypertens Pregnancy 2016; 35(3): 350–60. 1. American Society of Hematology: Education For Patients: Sickle Cell 13. Dulay AT: Gynecology and obstetrics: preeclampsia and eclampsia. In: Trait. Available at http://www.hematology.org/Patients/Anemia/Sickle- The Merck Manuals. Edited by Porter RS, Kaplan JL Kenilworth, NJ, Cell-Trait.aspx; Accessed Sep 29, 2018. USA, Merck Sharp & Dohme Corp, 2018. https://www.merckmanuals. 2. Ojodu J, Hulihan MM, Pope SN, Grant AM, Centers for Disease Control com/professional/gynecology-and-obstetrics/abnormalities-of- and Prevention (CDC): Incidence of sickle cell trait–United States, 2010. pregnancy/preeclampsia-and-eclampsia#. Accessed October 2, 2018. MMWR Morb Mortal Wkly Rep 2014; 63(49): 1155–8. 14. Wilson BJ, Watson MS, Prescott GJ, et al: Hypertensive diseases of 3. Niebuhr DW, Chen L, Shao S, Goldsmith J, Byrne C, Singer DE: Sickle pregnancy and risk of hypertension and stroke in later life: results from cell trait prevalence among U.S. military service members: 1992–2012. cohort study. BMJ 2003; 326(7394): 845. Mil Med 2017; 182(3): e1819–e1824. 15. Kotila TR: Sickle cell trait: a benign state? Acta Haematol 2016; 136(3): 4. American College of Obstetricians and Gynecologists: Task Force on 147–51. Epub 2016 Jul 16. Review. Hypertension in Pregnancy. Hypertension in pregnancy. Report of the 16. Tsaras G, Owusu-Ansah A, Boateng FO, Amoateng-Adjepong Y: Com- American College of Obstetricians and Gynecologists’ Task Force on plications associated with sickle cell trait: a brief narrative review. Am J Hypertension in Pregnancy. Obstet Gynecol 2013; 122(5): 1122–31. Med 2009; 122(6): 507–12. 5. Ananth CV, Keyes KM, Wapner RJ: Pre-eclampsia rates in the United 17. Jans SM, de Jonge A, Lagro-Janssen AL: Maternal and perinatal out- States, 1980–2010: age-period-cohort analysis. BMJ 2013; 347: f6564. comes amongst haemoglobinopathy carriers: a systematic review. Int J 6. Leeman L, Fontaine P: Hypertensive disorders of pregnancy. Am Fam Clin Pract 2010; 64(12): 1688–98. Review. Physician 2008; 78(1): 93–100. Review. 18. Larrabee KD, Monga M: Women with sickle cell trait are at increased 7. Paré E, Parry S, McElrath TF, Pucci D, Newton A, Lim KH: Clinical risk for preeclampsia. Am J Obstet Gynecol 1997; 177: 425–8. risk factors for preeclampsia in the 21st century. Obstet Gynecol 2014; 19. Tita AT, Biggio JR, Chapman V, Neely C, Rouse DJ: Perinatal and 124(4): 763–70. maternal outcomes in women with sickle or hemoglobin C trait. Obstet 8. Bartsch E, Medcalf KE, Park AL, Ray JG, High Risk of Pre-eclampsia Gynecol 2007; 110(5): 1113–9. Identification Group: Clinical risk factors for pre-eclampsia determined 20. Whalley PJ, Pritchard JA, Richards JR Jr: Sickle cell trait and pregnancy. in early pregnancy: systematic review and meta-analysis of large cohort JAMA 1963; 28(186): 1132–5. studies. BMJ 2016; 353: i1753. Review. 21. Taylor MY, Wyatt-Ashmead J, Gray J, Bofill JA, Martin R, Morrison 9. Irwin DE, Savitz DA, Hertz-Picciotto I St, André KA: The risk of JC: Pregnancy loss after first-trimester viability in women with sickle pregnancy-induced hypertension: black and white differences in a mili- cell trait: time for a reappraisal? Am J Obstet Gynecol 2006; 194(6): tary population. Am J Public Health 1994; 84(9): 1508–10. 1604–8. Epub 2006 Apr 25. 10. MacKay AP, Berg CJ, Atrash HK: Pregnancy-related mortality 22. Taylor MY, Wyatt-Ashmead J, Gray J, Bofill JA, Martin RW, Mor- from preeclampsia and eclampsia. Obstet Gynecol 2001; 97(4): rison JC: Pregnancy loss after first trimester viability in women with 533–8. sickle cell trait: a preliminary report. South Med J 2008; 101(2): 11. Wikström AK, Gunnarsdottir J, Nelander M, Simic M, Stephansson 150–1. O, Cnattingius S: Prehypertension in pregnancy and risks of small for gestational age infant and stillbirth. Hypertension 2016; 67(3): 640–6. 23. Stamilio DM, Sehdev HM, Macones GA: Pregnant women with the 12. Block-Abraham DM, Adamovich D, Turan OM, Doyle LE, Blitzer MG, sickle cell trait are not at increased risk for developing preeclampsia. Baschat AA: Maternal blood pressures during pregnancy and the risk Am J Perinatol 2003; 20(1): 41–8. e190 MILITARY MEDICINE, Vol. 185, January/February 2020

Journal

Military Medicine – Oxford University Press

Published: Feb 13, 2020

Keywords: pre-eclampsia; pregnancy; hypertension; sickle cell trait; eclampsia; military personnel; mothers; health care use; diagnosis; african continental ancestry group; risk, attributable; hypertension, pregnancy-induced; reproductive physiological process

There are no references for this article.