key: cord-0869014-wnhp6hz9
authors: Kasman, Alex M.; Bhambhvani, Hriday P.; Li, Shufeng; Zhang, Chiyuan A.; Stevenson, David K.; Shaw, Gary M.; Simard, Julia F.; Eisenberg, Michael L.
title: Reproductive Sequelae of Parental Severe Illness Before the Pandemic: implications for the COVID-19 pandemic
date: 2020-09-23
journal: Fertil Steril
DOI: 10.1016/j.fertnstert.2020.09.153
sha: 3472869529ead85b5acb1ba39d2c0d5c22b7a4f4
doc_id: 869014
cord_uid: wnhp6hz9

Objective To investigate, with pre-COVID-19 data, whether parental exposure to severe systemic infections near the time of conception is associated with pregnancy outcomes. Design Retrospective cohort study. Setting Population based study covering births within the United States between 2009 and 2016. Participants IBM® MarketScan® Research database covers reimbursed healthcare claims data on inpatient and outpatient encounters who are privately insured through employment sponsored health insurance. Our analytic sample were pregnancies to paired fathers and mothers. Interventions Parental preconception exposure (0-6 months prior to conception) to severe systemic infection (e.g. sepsis, hypotension, respiratory failure, critical care evaluation). Main Outcome Measures Preterm birth (i.e. live birth before 37 weeks) and pregnancy loss. Results 999,866 pregnancies occurred with 214,057 (21.4%) pregnancy losses and 51,759 (5.2%) preterm births observed. Mothers receiving ICU care in the preconception period had increased risk of pregnancy loss (RR 1.99, 95% CI 1.69-2.34) as did fathers (RR 1.67, 95% CI 1.43-1.96). Mothers with preconception sepsis had higher risk of preterm birth (RR 1.61, 95% 1.02-2.54) and pregnancy loss (RR 1.38, 95% 1.07-1.78) while paternal sepsis exposure was associated with an increased risk of pregnancy loss (RR 1.55, 95% 1.22-1.98). Similar results were noted for hypotension. Additionally, a dose response was observed with preconception time in the ICU and the risk of preterm birth and pregnancy loss for both mothers and fathers. Conclusions In a pre-COVID-19 cohort, parental preconception severe systemic infection was associated with increased odds of preterm birth and pregnancy loss when conception was soon after the illness.

Since its emergence in December 2019, Severe Acute Respiratory Syndrome 2 (SARS-CoV-2) has caused at least four million infections with more than 250,000 deaths globally. Thus, aspects surrounding severe systemic infection and respiratory failure present unknown reproductive risks during the current pandemic. We therefore sought to examine the impact that severe systemic illness may have on pregnancy outcomes (e.g. preterm birth and pregnancy loss) when parents are exposed during the preconception period. We hypothesized that parents who suffered from recent severe systemic illness prior to conception may have adverse pregnancy outcomes.

J o u r n a l P r e -p r o o f

The IBM ® MarketScan ® Research database was utilized for our study cohort. This database provides reimbursed healthcare claims data on inpatient and outpatient encounters covering over 150 million individuals who are privately insured through employment sponsored health insurance and medicare coverage as supplement. We analyzed claims data from the years 2007 to 2016. Institutional Review Board approval was not required for the analysis as this dataset contains de-identified patient information.

Cohort assembly and outcome ascertainment was based on the previously described methodology of Ailes et al. and Wall-Weiler et al. (9, 10) Briefly, pregnant women aged 20-45 were identified from inpatient and outpatient files. Mothers, fathers, and infants were linked using family ID. Through member enrollment files, we verified baby's records using the estimated birth date and enrollment start date. To determine adjudicated gestational age we used ICD (International Classification of Diseases)/CPT(Current Procedure Terminology)/DRG (Diagnosis Related Group) codes using the aforementioned methodology of Ailes et al. and Wall-Weiler, et al. from inpatient and outpatient files from both the mother and newborn. (9, 11, 12) The relevant codes are listed in Supplemental   Table 1 . The medical records of mothers and fathers were obtained by inpatient and outpatient claims files. We included only those infants with one male and one female parent at birth. Mothers and fathers had to be enrolled in insurance plans associated with the database for at least one year prior to conception. Outcomes were identified via ICD-9/10 diagnosis and DRG codes from both in-and outpatient claims, as well as CPT codes from outpatient claims for the mother (see Supplemental Table 1 ).

Pregnancy outcomes analyzed in the study included live birth, still birth, ectopic pregnancy, induced abortion, spontaneous abortion, and preterm birth (e.g. less than 37 weeks).

Pregnancy loss included ectopic, abortions (induced and spontaneous), and stillbirths.

We initially identified parental (mother or father) exposure to severe illness related to infection (e.g. sepsis, hypotension, respiratory failure) in the three months prior to J o u r n a l P r e -p r o o f F&S 30521 style revision estimated conception. This time period was chosen as spermatogenesis takes approximately three months and therefore outcomes related to insults that occur during this time may be captured. A sensitivity analysis of up to six months prior to conception was also performed. Exposures related to severe systemic infection and respiratory failure were chosen based on those that have been reported for COVID-19 or influenza. Inpatient variables were examined with regard to these outcomes utilizing ICD9/10, CPT, and DRG codes from 0-6 months prior to conception. The relevant exposure codes are listed in Supplemental table 1 and include illness associated with sepsis/systemic inflammatory response syndrome, respiratory failure/acute respiratory distress syndrome, hypotension/shock, influenza, and critical care evaluation and management. Reference groups for relative risk were those individuals with no exposures.

Descriptive statistics were presented as mean ± standard deviation. Categorical variables were expressed in frequencies with percentages. Differences in illness in both parents were examined using Chi-Squared or Fisher's Exact, as appropriate. Generalized estimating equation and generalized logit models estimated the relative risks and corresponding 95% confidence intervals of each outcome to allow for some families contributing subsequent births for binary and multinomial outcomes, respectively. All models were adjusted for birth year, region of care, and maternal factors including age, obesity, diabetes mellitus, hypertension, hyperlipidemia, and smoking. To evaluate unmeasured confounding effects, we calculated E-values, which estimates the minimum strength of association on the risk ratio scale that an unmeasured confounder would need to have with both the exposure and the outcome to fully explain away a specific exposure-outcome association (https://www.evalue-calculator.com).(13) All tests were two-sided and p <0.05 was considered statistically significant. Analyses were done in SAS software version 9.4 (SAS Institute Inc, Cary, NC).

J o u r n a l P r e -p r o o f

In total, 999,866 pregnancies were observed during the study period with 214,057 (21.4%) pregnancy losses and 51,759 (5.2%) preterm birth observed (Table 1 ). The mean paternal age was 35.4 years (SD 5.4) and mean maternal age was 33.2 years (SD 4.4).

Preconception respiratory/severe systemic infection in fathers and mothers was associated with preterm birth and pregnancy loss ( Table 2 ). Any ICU admission in the three months prior to conception was associated with an increased risk of pregnancy loss for both mothers (RR 1.99, 95% CI 1.69-2.34) and fathers (RR 1.67, 95% CI 1.43-1.96). A sensitivity analysis for unmeasured confounding (i.e. E-value) determined that the minimum strength of association for an unmeasured confounder to explain away the identified associations between severe systemic infection and adverse pregnancy outcomes varied between 1.39 to 2.6 for men and 1.51 to 3.39 for women. We next examined the relative risks according to abortion types, and did not identify differences between the two (Supplementary Table 2 ). In addition, a longer stay in the ICU was associated with a higher risk of preterm birth and pregnancy loss for both mothers and fathers (Table 3) .

Mothers and fathers with preconception sepsis were at higher risk of having a child with preterm birth (RR 1.61, 95% 1.02-2.54; RR 1.38, 95% CI 1.07-1.78, respectively) while fathers with preconception sepsis had a higher risk of pregnancy loss (RR 1.55, 95% CI 1.22-1.98). Those mothers with respiratory failure during the preconception period had higher risk of pregnancy loss (RR 1.31, 95% CI 1.13-1.53). Mothers diagnosed with hypotension or shock in the preconception period had higher risk of pregnancy loss (RR 1.99, 95% CI 1.69-2.34) as did fathers (RR 1.67, 95% CI 1.43-1.96). Furthermore, parents with multiple diagnoses of severe systemic illness (i.e. sepsis, critical respiratory failure, hypotension/shock, and/or critical care evaluation) had higher risk of pregnancy loss though the sample size was small (Supplemental Table 2 ). Expanding the exposure interval up to six months did not meaningfully alter the results (Supplementary Table 3 ). The diagnosis of influenza during the preconception period was not associated with a higher risk of preterm birth or pregnancy loss for both mothers and fathers. 

The reproductive sequelae of severe systemic illness are unknown and understanding these are of particular importance during the COVID-19 pandemic. Using a US claims cohort, the current report found that profound systemic infection preconceptionally, in both fathers and mothers who were able to conceive soon after severe illness, was associated with higher relative risks of pregnancy loss and preterm birth. Moreover, the higher number and more severe the illnesses (e.g. respiratory failure, sepsis, ICU care), the higher the risk of these two adverse pregnancy outcomes.

As SARS-CoV-2 can infect those of reproductive age with most recovering, understanding the reproductive sequelae effects of the disease is important for counseling and aftercare of these patients as well as others with severe systemic infections. An estimated 5-10% of COVID-19 positive patients will require intensive care unit admission and mechanical ventilation which includes reproductive aged men and women. (14) (15) (16) Severe systemic infection and its sequelae can put tremendous strain on an individual's body which may impact health long after discharge and has been observed with deconditioning, muscle atrophy, cognitive decline, and increased mortality.(5-7) Additionally, those who survive sepsis may have an increased risk of early mortality, rehospitalization, emotional distress leading to anxiety and depression, and cardiovascular disease.(8) However, the reproductive sequelae of recent preconception exposure to severe illness are unknown and may also apply to a subset of individuals who are able to conceive so soon after such ailment.

We found that recent exposure to severe systemic infection in mothers or fathers is associated with adverse pregnancy outcomes such as preterm birth and pregnancy loss.

The potential mechanisms that underlie these outcomes are unknown, but may include a direct effect of the infection and its consequences on reproductive organs or gametes (e.g. toxic or ischemic) or the side effects of treatments during the illness. Additionally, a disruption in endocrine signaling may also play a role as this is critical for conception. The With regard to mothers, the underlying mechanisms driving adverse pregnancy outcomes from a preconception exposure to severe illness are likely complex as the exposures may extend into pregnancy itself. A direct toxic effect on gametes may also drive these effects in mothers, similar to fathers. Preconception stress in forms other than an acute insult have been documented to adversely impact pregnancy outcomes such as from prenatal underweight mothers and psychosocial stress. (24, 25) Stressful environments preconception may be independent of traditional social (e.g. alcohol, drugs) and medical (e.g. placental abnormalities, gestational hypertension) as suggested by these studies. As the mechanisms through which maternal stress impacts pregnancy are unknown one can only postulate that these events may be driven by epigenetic changes induced by the event. (26) (27) (28) Additionally, preconception micronutrient deficiencies that may be induced by the event can lead to adverse birth outcomes. (29) J o u r n a l P r e -p r o o f

A few additional limitations warrant mention. As with any database that relies on diagnosis and procedural codes, errors in coding may influence the results and, as such, specific details underlying a patient's comorbidities and treatments could not be ascertained. Further, under sampling of diagnoses such as influenza may occur due to rule out diagnoses in uninfected and true cases never seen in the healthcare setting, which would likely bias findings to the null. However, others have used similar techniques to identify influenza cases within the MarketScan ® database. (30) (31) (32) (33) (34) The database is composed of individuals with commercial employer based health care and thus may not be generalizable to other populations. (35) For example, the frequency of preterm birth (preterm deliveries/all deliveries) was only 5.2% which is significantly lower than the reported rate for the general population.(36) Additionally, early pregnancy losses from undetected pregnancies are not captured and therefore may alter observed results. Thus, the analysis only applies to couples who were able to have a recognized pregnancy following severe systemic illness. Next, we are only examining couples who achieved pregnancy and some exposed men and women may have been unable to conceive at all. In addition, many social determinants of health, which may represent confounders, were not available in the database (e.g. education, race/ethnicity, income, parity) which may influence the pregnancy related outcomes. Similarly, several lifestyle factors (e.g. substance abuse) that also may influence outcomes were not available. Furthermore, conception dates were estimated so that the exact timing of conception may not be precise. Finally, unmeasured confounding may persist which can influence reproductive outcomes (e.g.

presence of an underlying chronic condition). Table 2 . Risk of maternal and paternal preconception exposure (up to 3 months prior to conception) on preterm birth and pregnancy loss. Adjusted for year of birth, maternal age, and region of birth, and maternal factors including obesity, diabetes mellitus, hypertension, hyperlipidemia, and smoking. Percentages may not add to 100% due to rounding. PTB = preterm birth, SAB = spontaneous abortion, TAB = therapeutic abortion. * = acute respiratory failure, ARDS, acute on chronic respiratory failure. J o u r n a l P r e -p r o o f § -E-values estimate the minimum strength of association on the risk ratio scale that an unmeasured confounder would need to have with both the exposure and the outcome to fully explain away a specific exposure-outcome association. Lowest possible E value is 1 meaning no unmeasured confounding would explain the association.

J o u r n a l P r e -p r o o f Table 3 . Risk of maternal and paternal preconception exposure (up to 3 months prior to conception) on spontaneous and therapeutic abortion. Adjusted for year of birth, maternal age, and region of birth, and maternal factors including obesity, diabetes mellitus, hypertension, hyperlipidemia, and smoking. Percentages may not add to 100% due to rounding. SAB = spontaneous abortion, TAB = therapeutic abortion. * = acute respiratory failure, ARDS, acute on chronic respiratory failure. J o u r n a l P r e -p r o o f

World Health Organization

Second-Trimester Miscarriage in a Pregnant Woman With SARS-CoV-2 Infection

Prior and Novel Coronaviruses, COVID-19, and Human Reproduction: What is Known?

High risk of temporary alteration of semen parameters after recent acute febrile illness

Recovery after critical illness: putting the puzzle together -a consensus of 29

Cognitive decline after sepsis

Blood glucose dysregulation and cognitive outcome in ARDS survivors

Post-sepsis syndromean evolving entity that afflicts survivors of sepsis

Using insurance claims data to identify and estimate critical periods in pregnancy: An application to antidepressants

Benzodiazepine use before conception and risk of ectopic pregnancy

Severe Maternal Morbidity Among Stillbirth and Live Birth Deliveries in California

Association of preconception paternal health on perinatal outcomes: Analysis of US claims data

Sensitivity Analysis in Observational Research: Introducing the E-Value

Management of Critically Ill Adults With COVID-19

Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area

Coronavirus Disease 2019 (COVID-19) in Italy

Sexual dysfunction in uraemic patients undergoing haemodialysis: predisposing and related conditions

Both Hyper-and Hypogonadotropic Occur Transiently in Acute Illness: Bio-and Immunoactive Gonadotropins

Systemic lupus erythematosus and risk of preterm birth: a systematic review and meta-analysis of observational studies

Influence of paternal preconception exposures on their offspring : through epigenetics to phenotype

Age-Associated Sperm DNA Methylation Alterations: Possible Implications in Offspring Disease Susceptibility

Cigarette smoking significantly alters sperm DNA methylation patterns

Association of perinatal outcomes: analysis of U.S. claims data

Women's prepregnancy underweight as a risk factor for preterm birth: a retrospective study

Preconceptional stress and racial disparities in preterm birth: an overview

Fetal programming: link between early nutrition, DNA methylation, and complex diseases

Reproductive epigenetics

Preconception Care

Periconceptional multivitamin use and risk of preterm or small-for-gestational-age births in the Danish National Birth

Healthcare resource use and economic burden attributable to respiratory syncytial virus in the United States: a claims database analysis

Rheumatoid arthritis and the incidence of influenza and influenza-related complications: a retrospective cohort study

Employer-incurred health care costs and productivity losses associated with influenza

The burden of influenza-like illness in the US workforce

Model estimates of the burden of outpatient visits attributable to influenza in the United States

CDC grand rounds: Public health strategies to prevent preterm birth

Effects of maternal age and agespecific preterm birth rates on overall preterm birth rates -United States

Ectopic pregnancy 777

Spontaneous abortion 59820

Risk of maternal and paternal preconception exposure (up to 6 months prior to conception) on preterm birth and pregnancy loss. Adjusted for year of birth, maternal age, and region of birth, and maternal factors including obesity, diabetes mellitus, hypertension, hyperlipidemia, and smoking. Percentages may not add to 100% due to rounding. PTB = preterm birth, SAB = spontaneous abortion, TAB = therapeutic abortion. * = acute respiratory failure, ARDS, acute on chronic respiratory failure.