Abstract
Objectives
To estimate whether stillbirth ≥ 23 weeks of gestation is associated with increased risk of severe maternal morbidity compared to livebirth, when stratified by maternal comorbidities.
Methods
This retrospective cohort study used International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis and procedure codes within the Healthcare Cost and Utilization Project’s Florida State Inpatient Database. The first delivery of female Florida residents aged 13 to 54 years old from 2005 – 2014 was included. The exposure was an ICD-9-CM code of stillbirth ≥ 23 weeks of gestation; the control was an ICD-9-CM code of singleton livebirth. Deliveries were stratified by the presence of ≥ 1 condition within a well-validated maternal morbidity composite using ICD-9-CM codes during delivery hospitalization. The primary outcome was an ICD-9-CM diagnosis or procedure code during delivery hospitalization of any indices within the Centers for Disease Control and Prevention’s severe maternal morbidity composite. Multivariable analyses adjusted for maternal sociodemographic factors and delivery mode to compare outcomes after stillbirth versus livebirth delivery.
Results
Nine thousand five hundred twenty-three women who delivered stillborn fetuses and 1,353,044 with liveborn infants were included. Among 6590 stillbirths and 935,913 livebirths without maternal comorbidities, severe maternal morbidity was significantly more common during stillbirth delivery (n=345 (5.2%)), corresponding to a 7-fold increased risk compared to livebirth (n=8,318 (0.9%); adjusted odds ratio (aOR) 7.05 (95% Confidence Interval (CI) 6.27 – 7.93)). Among 2933 stillbirths and 417,131 livebirths with maternal comorbidities, severe maternal morbidity was significantly more common during stillbirth delivery (n=390 (13.3%)): the risk was more than 6-fold higher comparatively (n=11,122 (2.7%); aOR 6.21 (95% CI 5.54 – 6.96)). Most maternal comorbidities were individually associated with higher risk of severe maternal morbidity during stillbirth versus livebirth delivery.
Conclusions
Though severe maternal morbidity is overall uncommon, delivering a stillborn fetus ≥ 23 weeks is associated with increased likelihood of severe maternal morbidity, particularly among women with comorbidities, suggesting health care providers must be vigilant about severe maternal morbidity during stillbirth delivery.
Précis
Compared to delivering liveborn infants, delivering stillborn fetuses ≥23 weeks of gestation is associated with increased risk of severe maternal morbidity, particularly in the setting of maternal comorbidities.
Introduction
The Centers for Disease Control and Prevention (CDC) defines severe maternal morbidity as a composite including medical conditions occurring and procedures performed during delivery hospitalization (Box 1).1 Severe maternal morbidity has been identified as a risk factor for both postpartum morbidity2,3 and peripartum maternal mortality. 4 As such, multiple risk factors for severe maternal morbidity have been identified, including obstetric conditions like placental abruption, medical comorbidities like hypertensive disorders, and maternal demographic factors like black race.3,5–7
Box 1.
Conditions Included in the Centers for Disease Control and Prevention’s Severe Maternal Morbidity Composite
| Acute myocardial infarction | Pulmonary edema/Acute Heart Failure |
| Aneurysm | Severe anesthesia complications |
| Acute renal failure | Sepsis |
| Adult Respiratory Distress Syndrome | Shock |
| Amniotic Fluid Embolism | Sickle cell disease with crisis |
| Cardiac arrest/ventricular fibrillation | Air and thrombotic embolism |
| Conversion of cardiac rhythm | Blood transfusion |
| Disseminated intravascular coagulation | Hysterectomy |
| Eclampsia | Temporary tracheostomy |
| Heart failure/arrest during surgery of procedure | Ventilation |
| Puerperal cerebrovascular disorders |
Data from Centers for Disease Control and Prevention. Severe maternal morbidity indicators and corresponding ICD codes during delivery hospitalizations. Available at: https://www.cdc.gov/reproductivehealth/maternalinfanthealth/smm/severe-morbidity-ICD.htm. Retrieved May 10, 2018.
Though these same risk factors have also been associated with increased risk of stillbirth,8,9 the association between stillbirth and severe maternal morbidity is less clear. Prior studies are limited by a lack of control group10 or occur in resource-poor settings,10–13 and the effects of underlying comorbidities on severe maternal morbidity during stillbirth delivery are unknown. Given that stillbirth occurs in 1 in 160 deliveries in the United States,8 it is of crucial importance to determine whether there is an association between stillbirth and severe maternal morbidity and if this association differs for women with medical comorbidities. This insight could help providers appropriately triage their patients’ risk for severe maternal morbidity during stillbirth delivery.
Using a statewide database, we aimed to ascertain whether stillbirth ≥23 weeks of gestation was associated with increased risk of severe maternal morbidity compared to singleton livebirth among women who were and were not coded with medical comorbidities during their delivery hospitalization. We hypothesized that stillbirth was associated with increased risk of severe maternal morbidity compared to livebirth regardless of maternal medical comorbidities.
Methods
We conducted a retrospective cohort study using the Florida State Inpatient Database of the Agency for Healthcare Research and Quality’s Healthcare Cost and Utilization Project (HCUP) including all in-hospital deliveries from 2005 through 2014.14 Our methods have been published previously15; in brief, we identified deliveries in women aged 13–54 years using a validated algorithm of International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis and procedure codes.16 We defined the index delivery as the first delivery within the database during the study timeframe, regardless of subsequent deliveries or parity. We restricted deliveries to stillborn fetuses ≥23 weeks of gestation (ICD-9-CM codes 656.40, 656.41, and V271) and liveborn singletons (ICD-9-CM codes 650 and V270). We excluded women whose index deliveries were coded as both live-born singletons and stillborn fetuses ≥ 23 weeks of gestation or as both singleton and multiple gestation (ICD-9-CM codes 651.00, 651.01, 651.10, 651.11, 651.20, 651.21, V272, and V275). Females who did not reside in Florida and patients listed as “male” were also excluded. Sociodemographic data analyzed included age, race–ethnicity (non-Hispanic black, non-Hispanic white, Hispanic, and other), payer (private, public, or other), and income quartile by zip code.
Underlying maternal medical comorbidities coded during delivery hospitalization were identified using a maternal comorbidity composite validated for the HCUP databases7 (Box 2; Appendix 1). Women were stratified into subpopulations by whether they were or were not coded for at least one medical condition in the maternal comorbidity composite during their delivery hospitalization.
Box 2.
Conditions Included Within Validated Inpatient Maternal Co-morbidity Composite
| Pulmonary Hypertension | Systemic lupus erythematosus |
| Sickle cell disease | Human immunodeficiency virus |
| Placenta Previa | Cardiac valvular disease |
| Gestational Hypertension | Chronic congestive heart failure |
| Mild or unspecified preeclampsia | Asthma |
| Severe preeclampsia/eclampsia | Preexisting diabetes mellitus |
| Chronic renal disease | Gestational diabetes mellitus |
| Preexisting hypertension | Obesity |
| Chronic ischemic heart disease | Cystic fibrosis |
| Congenital heart disease | Previous cesarean delivery |
Modified from Bateman B, Myhre JM, Hernandez-Diaz S, Hubrechts KF. Fischer MA, Creanga AA, et al. Development of a comorbidity index for use in obstetric patients. Obstet Gynecol 2013;122:957–65.
The primary outcome was the CDC’s severe maternal morbidity composite; the individual indices were extracted using ICD-9-CM diagnosis and procedure codes as defined by the CDC1 (Appendix 2). There are concerns that some conditions within the severe maternal morbidity and maternal morbidity composites may be under-coded due to variations in coding practices between hospitals.17 Thus, our secondary outcomes included a severe maternal morbidity composite without blood transfusion as well as the individual indices within the primary severe maternal morbidity composite. Other secondary outcomes included whether specific conditions within the maternal comorbidity composite where associated with increased risk of severe maternal morbidity during stillbirth delivery. All outcomes and maternal comorbidities were extracted during the delivery hospitalization for the index pregnancy. Patients may have had more than one diagnosis within the severe maternal morbidity or comorbidity composites but were only counted once per composite.
Demographic and baseline clinical data were compared between women who delivered a stillborn fetus versus a liveborn singleton using the X2 test or Fisher’s exact test for categorical variables as appropriate. The primary multivariable regression models stratified women by the presence of comorbidities and type of delivery (stillbirth or livebirth) and analyzed both primary and secondary outcomes (i.e., severe maternal morbidity as a composite with and without blood transfusion and the individual conditions within the severe maternal morbidity composite). The secondary model limited the study population to women who delivered a stillborn fetus and stratified them by the presence of each condition within the comorbidity composite; again, the outcome was severe maternal morbidity. For composite outcomes, we used multivariable logistic regression models adjusted for age, race–ethnicity, payer, income quartile by zip code, and mode of delivery. Finally, we tested whether significant interactions existed between independent variables (age, race–ethnicity, payer, income quartile by zip code, and mode of delivery) within the primary models. Demographic data missing from the database was recoded as an indicator variable in order to ensure all patients were included in the multivariate analyses. All analyses considered a two-sided p <0.05 as statistically significant. Due to HCUP restrictions aimed to preserve patient privacy, counts <11 are reported as “n<11” for exposures and “--” for outcomes.
The Washington University in St. Louis Human Research Protection Office exempted this study from review given HCUP data does not contain personally identifiable information. SAS version 9.3 (SAS Institute, Cary NC) was used for all analyses.
Results
A total of 1,362,567 singleton deliveries were identified: 9523 (0.7%) were stillbirths ≥ 23 weeks of gestation and 1,353,044 (99.3%) were livebirths. A similar proportion of women delivering stillborn fetuses and liveborn singletons were coded for at least one condition within the maternal comorbidity composite during their delivery hospitalization (n=2933 (30.8%) for stillbirth; n=417,131 (30.8%) for livebirth). Sociodemographic and obstetric characteristics between women who delivered stillborn fetuses ≥ 23 weeks of gestation and liveborn singletons stratified by the presence of maternal comorbidities are shown in Table 1; missing data are presented in Appendices 3 and 4. Regardless of the presence of medical comorbidities, those with stillborn fetuses were significantly more likely to be older than 35 years old, be Black, have public insurance, have a vaginal delivery, and be in the lower two income quartiles by zip code—particularly in the lowest income quartile—when compared to women who delivered liveborn infants. In addition, compared to women who delivered liveborn infants, women without medical morbidities who had stillborn fetuses were significantly more likely to be <18 years old or Hispanic, while those with medical comorbidities who had stillborn fetuses were more likely to be in the second highest income quartile. Among women who delivered a stillborn fetus, 12% of those without any comorbidities and 37% of those with at least one comorbidities delivered by cesarean.
Table 1:
Comparison of characteristics among women who delivered a singleton stillborn fetus ≥ 23 weeks of gestation versus singleton liveborn infant, stratified by maternal medical comorbidities coded during delivery hospitalization
| Without Maternal Medical Comorbidities | With Maternal Medical Comorbidities | |||||
|---|---|---|---|---|---|---|
| Stillborn fetus ≥ 23 weeks (n=6590) | Liveborn infant (n=935,913) | Unadjusted Odds Ratio (95% Confidence Interval (CI)) | Stillborn fetus ≥ 23 weeks (n=2933) | Liveborn infant (n=417,131) | Unadjusted Odds Ratio (95% CI) | |
| Maternal age at delivery (n (%)) | ||||||
| < 18 years | 363 (5.5) | 41,247 (4.4) | 1.33 (1.19 – 1.47) | 61 (2.1) | 9242 (2.2) | 0.99 (0.77 – 1.28) |
| 18–34 years | 5194 (78.8) | 781,707 (83.5) | Reference | 2112 (72.0) | 316,682 (75.9) | Reference |
| 35–39 years | 760 (11.5) | 91,275 (9.8) | 1.25 (1.16 – 1.35) | 551 (18.8) | 69,419 (16.7) | 1.19 (1.08 – 1.31) |
| ≥ 40 years | 273 (4.1) | 21,684 (2.3) | 1.90 (1.68 – 2.14) | 209 (7.1) | 21,788 (5.2) | 1.44 (1.25 – 1.66) |
| Race/Ethnicity (n (%))1 | ||||||
| White | 2630 (40.5) | 483,949 (52.5) | Reference | 1098 (37.9) | 205,409 (49.9) | Reference |
| Black | 2381 (36.7) | 191,519 (20.8) | 2.29 (2.16 – 2.42) | 1210 (41.8) | 96,114 (23.3) | 2.36 (2.17 – 2.56) |
| Hispanic | 1114 (17.2) | 187,038 (20.3) | 1.10 (1.02 – 1.18) | 433 (14.9) | 84,578 (20.5) | 0.96 (0.86 – 1.07) |
| Other | 361 (5.6) | 60,085 (6.5) | 1.11 (0.99 – 1.24) | 157 (5.4) | 25,825 (6.3) | 1.14 (0.96 – 1.35) |
| Insurance Type (n (%)) | ||||||
| Private Insurance | 2911 (44.2) | 448,027 (47.9) | Reference | 1224 (41.7) | 206,509 (49.5) | Reference |
| Public Insurance | 3484 (52.9) | 458,757 (49.0) | 1.17 (1.11 – 1.23) | 1600 (54.6) | 197,944 (47.5) | 1.36 (1.27 – 1.47) |
| Other | 195 (2.9) | 29,129 (3.1) | 1.03 (0.89 – 1.19) | 109 (3.7) | 12,678 (3.0) | 1.45 (1.19 – 1.77) |
| Income Quartile, based on Zip code (n (%))2 | ||||||
| Quartile 1 (Poorest) | 2178 (38.7) | 253,488 (32.0) | 1.65 (1.49 – 1.82) | 1077 (42.7) | 114,597 (32.5) | 2.03 (1.73 – 2.37) |
| Quartile 2 | 1792 (31.9) | 246,393 (31.1) | 1.40 (1.26 – 1.55) | 754 (30.0) | 109,005 (30.9) | 1.49 (1.27 – 1.76) |
| Quartile 3 | 1195 (21.3) | 204,957 (25.9) | 1.12 (1.00 – 1.25) | 507 (20.1) | 90,330 (25.5) | 1.21 (1.02 – 1.43) |
| Quartile 4 (Wealthiest) | 456 (8.1) | 87,478 (11.0) | Reference | 181 (7.2) | 39,023 (11.1) | Reference |
| Mode of Delivery (n (%)) | ||||||
| Spontaneous Vaginal | 5310 (85.4) | 631,115 (67.4) | Reference | 1693 (60.1) | 140,061 (33.6) | Reference |
| Operative Vaginal | 159 (2.6) | 49,974 (5.3) | 0.35 (0.32 – 0.38) | 80 (2.8) | 11,263 (2.7) | 0.59 (0.47 – 0.74) |
| Cesarean | 745 (12.0) | 254,824 (27.2) | 0.38 (0.32 – 0.44) | 1044 (37.1) | 265,807 (63.7) | 0.33 (0.30 – 0.35) |
For women who had no comorbidities, missing data from 104 stillbirths and 13,322 livebirths. For comorbidities, missing data from 35 stillbirths and 5205 livebirths.
For women who had no comorbidities, missing data from 969 stillbirths and 143,597 livebirths. For comorbidities, missing data from 414 stillbirths and 64,176 livebirths
Primary and secondary outcomes among women who were not coded for any conditions within the medical comorbidity composite during delivery hospitalization are shown in Table 2, stratified by stillbirth versus livebirth and adjusted for maternal age, race–ethnicity, insurance type, income quartile by zip code, and mode of delivery. In Tables 2 – 4, outcomes with less than 11 women are marked as “—“ whereas counts of 0 are demarcated as such. During singleton livebirth deliveries without maternal medical comorbidities, 0.9% of women (n=8318) were coded for at least one condition within the severe maternal morbidity composite. However, the risk of severe maternal morbidity during delivery hospitalization was more than 7 times higher after delivery of a stillborn fetus ≥ 23 weeks of gestation without medical comorbidities (n=345 (5.2%); adjusted odd ratios (aOR) 7.05 (95% Confidence Interval (CI): 6.27 – 7.93)). The most common condition within the severe maternal morbidity composite coded for both stillbirth and livebirth delivery was blood transfusion, which occurred significantly more frequently among stillborn fetuses than liveborn infants (n=254 (3.9%) versus n=5972 (0.6%); aOR 7.07 (95% CI 6.17 – 8.10)). Nevertheless, the association between stillbirth and severe maternal morbidity among deliveries without maternal comorbidities remained robust when blood transfusion was excluded from the severe maternal morbidity composite (n=91 (1.4%) for stillbirth versus n=2346 (0.3%) for livebirth; aOR 6.50 (95% CI 5.19 – 8.12)). In fact, stillbirth ≥ 23 weeks of gestation among women without medical comorbidities was associated with increased risk of many conditions within the severe maternal morbidity composite, though six of the twenty severe maternal morbidity conditions did not occur in this population.
Table 2:
Comparison of severe maternal morbidity among women who did not have medical co-morbidities coded during delivery hospitalization for stillborn fetus ≥ 23 weeks of gestation compared to liveborn infant1
| Without Maternal Medical Comorbidities | ||||
|---|---|---|---|---|
| Stillborn fetus ≥ 23 weeks (n=6590) | Liveborn infant (n=935,913) | Unadjusted Odds Ratio (95% Confidence Interval (CI)) | Adjusted Odds Ratio (95% CI)2 | |
| Overall Morbidity Composite (n=8,663) | 345 (5.2) | 8318 (0.9) | 6.16 (5.52 – 6.88) | 7.05 (6.27 – 7.93) |
| Transfusion (n=6,226) | 254 (3.9) | 5972 (0.6) | 6.24 (5.49 – 7.10) | 7.07 (6.17 – 8.10) |
| Composite Morbidity Without Transfusion (n=2,437) | 91 (1.4) | 2346 (0.3) | 5.57 (4.51 – 6.88) | 6.50 (5.19 – 8.12) |
| Acute Myocardial Infarction (n<11)3 | 0 (0.0) | -- | -- | -- |
| Aneurysm (n<11)3 | 0 (0.0) | -- | -- | -- |
| Acute renal failure (n=119) | 19 (0.3) | 100 (0.01) | 27.06 (16.56 – 44.20) | -- |
| Adult Respiratory Distress Syndrome (n=359) | 39 (0.6) | 320 (0.03) | 17.41 (12.48 – 24.30) | -- |
| Amniotic Fluid Embolism (n=28) | 0 (0.0) | 28 (0.00) | -- | -- |
| Cardiac Arrest / Ventricular Fibrillation (n=50)3 | -- | -- | 9.07 (2.82 – 29.14) | -- |
| Conversion of Cardiac Arrhythmia (n=118)3 | -- | -- | 10.34 (5.05 – 21.21) | -- |
| Disseminated Intravascular Coagulation (n=1563) | 91 (1.4) | 1472 (0.2) | 8.89 (7.18 – 11.00) | -- |
| Eclampsia (n=0) | 0 (0.0) | 0 (0.0) | -- | -- |
| Heart Failure/Arrest During Surgery or Procedure (n=77) | 0 (0.0) | 77 (0.01) | -- | -- |
| Puerperal Cerebrovascular Disorder (n=169)3 | -- | -- | 0.85 (0.12 – 6.04) | -- |
| Pulmonary Edema / Acute heart failure (n=137)3 | -- | -- | 2.11 (0.52 – 8.51) | -- |
| Severe Anesthesia Complication (n=120)3 | -- | -- | 2.41 (0.60 – 9.74) | -- |
| Sepsis (n=335) | 53 (0.8) | 282 (0.03) | 26.90 (20.04 – 36.11) | -- |
| Shock (n=202) | 17 (0.3) | 185 (0.02) | 13.08 (7.80 – 21.51) | -- |
| Sickle Cell Disease (n=145)3 | -- | -- | 4.03 (1.49 – 10.89) | -- |
| Air and Thrombotic Embolism (n=103)3 | -- | -- | 5.74 (2.11 – 15.60) | -- |
| Hysterectomy (n=385) | 13 (0.2) | 372 (0.04) | 4.97 (2.86 – 8.65) | -- |
| Temporary tracheostomy (n=16)3 | 0 (0.0) | 16 (0.00) | -- | -- |
| Ventilation (n=26)3 | -- | -- | 8.90 (1.19 – 66.73) | -- |
Data presented as (n (%)) unless otherwise noted
Among morbidities with counts ≥ 50 women, adjusted for mode of delivery and maternal age, race–ethnicity, insurance type, and income quartile by zip code
Unable to report (less than 11 women)
Table 4:
Comparison of severe maternal morbidity among women who had a stillborn fetus ≥ 23 weeks of gestation and who were versus were not coded for maternal co-morbidities during delivery hospitalization1
| During delivery of stillborn fetus ≥ 23 weeks (n=9523) | ||||
|---|---|---|---|---|
| SMM with maternal co-morbidity | SMM without maternal co-morbidity2 | Unadjusted Odds Ratio (95% Confidence Interval (CI)) | Adjusted Odds Ratio (95% CI)3 | |
| Overall Maternal Co-Morbidity Composite (n=2933) | 390 (13.3) | 345 (5.2) | 2.78 (2.39 – 3.23) | 1.82 (1.53 – 2.16) |
| Pulmonary hypertension (n<11)4 | -- | -- | 84.49 (10.38 – 687.61) | -- |
| Placenta previa (n=126) | 25 (19.8) | 710 (7.6) | 3.03 (1.94 – 4.72) | -- |
| Preeclampsia or gestational hypertension without severe features (n=596) | 101 (17.0) | 634 (7.1) | 2.67 (2.12 – 3.36) | -- |
| Preeclampsia with severe features (n=401) | 131 (32.7) | 604 (6.6) | 6.84 (5.47 – 8.56) | -- |
| Chronic kidney disease (n=63) | 39 (61.9) | 696 (7.3) | 20.46 (12.23 – 34.22) | -- |
| Chronic hypertension (n=492) | 75 (15.2) | 660 (7.3) | 2.28 (1.76 – 2.95) | -- |
| Ischemic heart disease (n<11)4 | -- | -- | 11.97 (0.75 – 191.59) | -- |
| Congenital heart disease (n=11)4 | -- | -- | 1.09 (0.14 – 8.43) | -- |
| Sickle Cell Disease (n=126) | 25 (19.8) | 710 (7.6) | 3.03 (1.94 – 4.72) | -- |
| Systemic Lupus Erythematosus (n=36)4 | -- | -- | 4.02 (1.89 – 8.59) | -- |
| Human immunodeficiency virus (n=59)4 | -- | -- | 1.89 (0.89 – 3.99) | -- |
| Cardiac valvular disease (n=41)4 | -- | -- | 3.90 (1.90 – 7.98) | -- |
| Chronic heart failure (n=0) | 0 (0.0) | 735 (7.7) | -- | -- |
| Asthma (n=250) | 30 (12.0) | 705 (7.6) | 1.66 (1.12 – 2.45) | -- |
| Pre-gestational diabetes (n=357) | 37 (10.4) | 698 (7.6) | 1.40 (0.99 – 1.99) | -- |
| Gestational diabetes (n=372) | 20 (5.4) | 715 (7.8) | 0.67 (0.42 – 1.06) | -- |
| Obesity (n=530) | 53 (10.0) | 682 (7.6) | 1.36 (1.01 – 1.82) | -- |
| Cystic fibrosis (n<11) | 0 (0.0) | 735 (7.7) | -- | -- |
| History of cesarean delivery (n=799) | 83 (11.3) | 652 (7.5) | 1.44 (1.13 – 1.83) | -- |
Data presented as (n (%)) unless otherwise noted.
Each row sums to 735 women, the overall number of women with stillbirth who had severe maternal morbidity. In this column, the “n” presented is the number of women with stillbirth who did not have that specific comorbidity but did have severe maternal morbidity. The percentage presented is calculated using the denominator of total number of women with stillbirth (n=9523).
Among morbidities with counts ≥ 50 women, adjusted for mode of delivery and maternal age, race–ethnicity, insurance type, and income quartile by zip code
Unable to report (less than 11 women)
Primary and secondary outcomes among women who were coded for at least one condition within the maternal medical comorbidity composite during delivery hospitalization are shown in Table 3. Outcomes were again stratified by stillbirth versus livebirth and adjusted for maternal age, race–ethnicity, insurance type, income quartile by zip code, and mode of delivery. Three hundred ninety (13.3%) of women who had a stillborn fetus and at least one maternal comorbidity experienced severe maternal morbidity during their delivery, corresponding to more than six-fold increased risk compared to those who had a liveborn infant and at least one maternal medical comorbidity (n=11,122 (2.7%); aOR 6.21 (95% CI 5.54 – 6.96)). Blood transfusion was again the most common condition within the severe maternal morbidity composite for both stillbirth and livebirth delivery; however, the association between stillbirth and severe maternal morbidity among deliveries with maternal comorbidities remained strong when blood transfusion was excluded from the severe maternal morbidity composite (n=145 (4.9%) for stillbirth versus n=4645 (1.1%) for livebirth; aOR 5.04 (95% CI 4.22 – 6.02)). Among women coded for any medical comorbidity during delivery hospitalization, stillbirth ≥ 23 weeks of gestation was associated with increased risk of nearly all conditions within the severe maternal morbidity composite, including both common conditions like acute renal failure, adult respiratory distress syndrome, disseminated intravascular coagulation, or sepsis and rare conditions like cardiac arrest.
Table 3:
Comparison of severe maternal morbidity among women coded for at least one medical co-morbidities coded during delivery hospitalization for a stillborn fetus ≥ 23 weeks of gestation compared to a liveborn infant1
| With Maternal Medical Comorbidities | ||||
|---|---|---|---|---|
| Stillborn fetus ≥ 23 weeks (n=2933) | Liveborn infant (n=417,131) | Unadjusted Odds Ratio (95% Confidence Interval (CI)) | Adjusted Odds Ratio (95% CI)2 | |
| Overall Morbidity Composite (n=11,512) | 390 (13.3) | 11,122 (2.7) | 5.60 (5.03 – 6.24) | 6.21 (5.54 – 6.96) |
| Transfusion (n=6,772) | 245 (8.4) | 6477 (1. 6) | 5.78 (5.06 – 6.60) | 6.33 (5.51 – 7.27) |
| Composite Morbidity Without Transfusion (n=4,790) | 145 (4.9) | 4645 (1.1) | 4.62 (3.89 – 5.47) | 5.04 (4.22 – 6.02) |
| Acute Myocardial Infarction (n=28)3 | -- | -- | 23.75 (8.24 – 68.48) | -- |
| Aneurysm (n=14)3 | 0 (0.0) | 14 (0.00) | -- | -- |
| Acute renal failure (n=772) | 85 (2.9) | 637 (0.2) | 19.52 (15.52 – 24.54) | -- |
| Adult Respiratory Distress Syndrome (n=998) | 59 (2.0) | 939 (0.2) | 9.10 (6.98 – 11.87) | -- |
| Amniotic Fluid Embolism (n=43)3 | -- | -- | 6.94 (1.68 – 28.71) | -- |
| Cardiac Arrest / Ventricular Fibrillation (n=89)3 | -- | -- | 10.30 (4.50 – 23.61) | -- |
| Conversion of Cardiac Arrhythmia (n=154)3 | -- | -- | 4.78 (1.96 – 11.67) | -- |
| Disseminated Intravascular Coagulation (n=1754) | 132 (4.5) | 1622 (0.4) | 12.07 (10.07 – 14.47) | -- |
| Eclampsia (n=1385) | 21 (0.7) | 1364 (0.3) | 2.20 (1.43 – 3.39) | -- |
| Heart Failure/Arrest During Surgery or Procedure (n=121)3 | -- | -- | 1.19 (0.17 – 8.49) | -- |
| Puerperal Cerebrovascular Disorder (n=169)3 | -- | -- | 3.76 (1.78 – 7.98) | -- |
| Pulmonary Edema / Acute heart failure (n=716) | 19 (0.7) | 697 (0.2) | 3.90 (2.47 – 6.15) | -- |
| Severe Anesthesia Complication (n=126)3 | -- | -- | 1.14 (0.16 – 8.14) | -- |
| Sepsis (n=337) | 36 (1.2) | 301 (0.07) | 17.21 (12.16 – 24.37) | -- |
| Shock(n=247) | 26 (0.9) | 221 (0.05) | 16.88 (11.22 – 25.38) | -- |
| Sickle Cell Disease (n=133)3 | -- | -- | 3.29 (1.05 – 10.33) | -- |
| Air and Thrombotic Embolism (n=142)3 | -- | -- | 3.07 (0.98 – 9.65) | -- |
| Hysterectomy (n=700) | 18 (0.6) | 682 (0.2) | 3.77 (2.36 – 6.03) | -- |
| Temporary tracheostomy (n=39)3 | -- | -- | 3.74 (0.51 – 27.28) | -- |
| Ventilation (n=117)3 | -- | -- | 2.47 (0.61 – 10.12) | -- |
Data presented as (n (%)) unless otherwise noted
Among morbidities with counts ≥ 50 women, adjusted for mode of delivery and maternal age, race-ethnicity, insurance type, and income quartile by zip code
Unable to report (less than 11 women)
Additional multivariable analyses tested whether significant interactions between independent variables existed in the primary model. For women who were not coded for medical comorbidities during delivery hospitalization, the p values for interaction were not significant for race–ethnicity (p=0.28), payer type (p=0.29), or income quartile by zip code (p = 0.17). However, the p values for interaction were statistically significant for mode of delivery (p<0.001). Controlling for other independent variables, women who had cesarean or operative vaginal delivery were significantly more likely to have severe maternal morbidity during stillbirth versus livebirth delivery (aOR 11.90 (95% CI 9.72 – 14.56) for cesarean; aOR 5.13 (95% CI 2.38 – 11.04) for operative vaginal delivery).
Conversely, for women who were coded for at least one medical comorbidity during delivery hospitalization, the p values for interaction were not significant for race–ethnicity (p=0.28) or payer type (p=0.29) but were statistically significant for income quartile by zip code (p = 0.040) and mode of delivery (0.02). Controlling for other independent variables, women with at least one medical comorbidity who had cesarean or operative vaginal delivery were significantly more likely to have severe maternal morbidity during stillbirth versus livebirth delivery (aOR 7.07 (95% CI 6.02 – 8.30) for cesarean; aOR 3.28 (95% CI 1.40 – 7.69) for operative vaginal delivery). In terms of income quartile by zip code, women in the lowest two quartiles had similar risk of severe maternal morbidity during stillbirth versus livebirth delivery (aOR 5.99 (95% CI 4.92 – 7.30) for lowest quartile; aOR 5.90 (95% CI 4.66 – 7.48) for second lowest quartile), but the risk was higher among women in the second highest quartile (aOR 8.75 (95% CI 6.71 – 11.41)).
Table 4 describes the prevalence and risk of severe maternal morbidity for each condition within the maternal medical comorbidity composite among women who had a stillborn fetus ≥ 23 weeks of gestation, stratified by severe maternal morbidity, and adjusted for mode of delivery and maternal age, race–ethnicity, insurance type, and income quartile by zip code. Overall, the risk of severe maternal morbidity was significantly higher for women who had a stillborn fetus and at least one medical comorbidity compared to stillbirth and no medical comorbidities (n=390 (13.3%) for comorbidities versus n=345 (5.2%) for no comorbidities; aOR 1.82 (95% CI 1.53 – 2.16). The condition associated with the highest risk of severe maternal morbidity during delivery of a stillborn fetus was chronic kidney disease (CKD): nearly two thirds of the 63 women with CKD who delivered stillborn fetuses had severe maternal morbidity (n=39 (61.9%); Odds Ratio 20.46 (95% CI 12.23 – 34.22)). Many more common comorbidities including preeclampsia with or without severe features, chronic hypertension, placenta previa, sickle cell disease, asthma, and history of cesarean delivery were also associated with increased risk of severe maternal morbidity during delivery of a stillborn fetus. Conversely, gestational and pregestational diabetes and human immunodeficiency virus were not associated with increased risk of severe maternal morbidity during stillbirth delivery.
Discussion
In this large retrospective cohort study, we provide insight into the incidence of severe maternal morbidity during delivery of stillborn fetuses at ≥ 23 weeks of gestation versus liveborn singletons. Specifically, the risk of severe maternal morbidity was markedly higher for women who had a stillbirth delivery compared to a livebirth delivery regardless of maternal medical comorbidities (aOR 7.05 (95% CI 6.27 – 7.93) for no comorbidities; aOR 6.21 (95% CI 5.54 – 6.96) with at least one comorbidity). In addition, we identified specific medical conditions including chronic kidney disease, hypertensive disease, sickle cell disease, and placenta previa that significantly increased the risk of severe maternal morbidity among women delivering stillborn fetuses. Thus, though the majority of women who have a stillborn fetus ≥ 23 weeks will have delivery unaffected by severe maternal morbidity, our findings suggest providers must be vigilant about the increased risk of severe maternal morbidity during stillbirth delivery, particularly for women with the medical comorbidities placing them at highest risk for severe maternal morbidity.
Our findings supporting prior studies identifying non-Hispanic black race and advanced maternal age as risk factors for stillbirth8 while identifying additional risk factors for stillbirth: public insurance compared to private insurance and lower socioeconomic status. However, though ACOG states maternal comorbidities are associated with increased risk of stillbirth,8 our study population had similar rates of maternal medical comorbidities coded during hospitalization for stillbirth delivery versus livebirth delivery (30.8%). This similarity may reflect undercoding of medical comorbidities within the HCUP database,17 actual demographic similarities between the two groups, or the fact that our inpatient database does not include any outpatient comorbidities associated with stillbirth like maternal infection.8 Additional prospective research is needed to confirm this finding to better clarify the association between comorbidities and stillbirth. Of note, though ACOG recommends reserving cesarean delivery for stillbirth delivery for unusual circumstances,8 more than one third of women with comorbidities and more than one tenth of women without comorbidities delivered their stillborn fetuses by cesarean. Clinical practice regarding stillbirth mode of delivery does not appear to align with ACOG recommendations.
Our study offers several strengths. First, we provide granular insight into the association between specific medical conditions and severe maternal morbidity during stillbirth delivery, which may help providers triage their patient’s individual risk for severe maternal morbidity based on her specific comorbidities. Second, our data derive from a large, comprehensive, all-payer database that allowed us to analyze outcomes for all deliveries in Florida for a decade, increasing the generalizability of our findings. Third, our ICD-9-CM coding has been validated for all critical variables included in this analysis, including deliveries,16 medical comorbidities,7 and severe maternal morbidity.1,3 This validation strengthens our findings.
Nevertheless, limitations should be considered. First, our analyses cannot determine causality. Because severe maternal morbidity and delivery type were coded in the same inpatient hospitalization, we cannot determine whether severe maternal morbidity occurred after the stillbirth or caused the stillbirth. In addition, as in any retrospective study, there is a residual risk of confounding. For example, ICD-9-CM coding used in our analyses did not account for disease severity, though factors like uncontrolled diabetes or hypertension are associated with stillbirth18 and prolonged duration of stillbirth prior to delivery may be associated with increased risk of severe maternal morbidity.8 In particular, we identified a strong association between chronic kidney disease and severe maternal morbidity during delivery of a stillborn fetus; however, the ICD-9-CM code used for this condition does not delineate between the type and severity of the chronic renal disease, which likely confounds the association. The lack of causality and potential for confounding require our findings to be confirmed with prospective data. Second, though stillbirth is defined in the United States as pregnancy loss at or after 20 weeks of gestation,8 our study defined stillbirth as ≥ 23 weeks of gestation due to ICD-9-CM diagnosis code definitions and because the HCUP database does not include gestational age at delivery. The lack of inclusion of stillborn fetuses between 20 and 22 weeks and specific gestational age at delivery may have affected our results. Third, our database was not comprehensive: though tobacco use is associated with stillbirth, comorbidities, and severe intrapartum morbidity, it was not included in our study population as this variable has been shown to be under-coded in the Florida HCUP database.19 This omission may confound our results. Fourth some conditions within the comorbidity or severe maternal morbidity composites may have been under-coded in the HCUP dataset. For example, in our study population, 530 of the 9523 women who had a stillborn fetus (5.6%) were coded with obesity, but, per the CDC Pregnancy Risk Assessment Monitoring System, almost 20% of mothers in Florida were obese prior to pregnancy in 2009–2011.20 This risk of under-coding affects all studies using administrative dataset and may affect the association between stillbirth and severe maternal morbidity. Lastly, it is possible that providers were more likely to code for medical comorbidities during stillbirth versus livebirth delivery, resulting in selection bias. The effects of selection bias on our results is likely not significant given the association between severe maternal morbidity and stillbirth versus livebirth was similar among women with and without medical comorbidities but warrants additional prospective research.
In conclusion, women who have a stillborn fetus ≥ 23 weeks of gestation have higher risk of severe maternal morbidity during their delivery hospitalizations compared to those who have a liveborn singleton, particularly in the setting of maternal comorbidities. In addition, the risk of nearly all conditions within the severe maternal morbidity composite—not just blood transfusion—was increased during delivery of stillbirth versus livebirth. Finally, specific medical comorbidities were identified that were associated with particularly high risk of severe maternal morbidity during stillbirth delivery. Taken together, these findings could help providers triage their patients’ risk of severe maternal morbidity during stillbirth delivery while increasing their vigilance for all severe maternal morbidity, not just blood transfusion.
Supplementary Material
Acknowledgments
Financial Disclosure
Margaret A. Olsen disclosed that money has been paid to her institution from the NIH and the AHRQ. She has received funds from Pfizer, Merck, and Sanofi Pasteur (money paid to institution [grant funding, relationship is ongoing]). The other authors did not report any potential conflicts of interest.
Dr. Lewkowitz is supported in part by a National Institutes of Health training grant T32-HD-55172-9. The Center for Administrative Data is supported in part by the Washington University Institute of Clinical and Translational Sciences grant UL1 TR002345 from the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH) and Grant Number R24 HS19455 through the Agency for Healthcare Research and Quality (AHRQ). The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official view of the NIH.
The authors thank Dustin Stwalley for his statistical support.
Footnotes
Each author has confirmed compliance with the journal’s requirements for authorship.
References
- 1.Severe Maternal Morbidity Indicators and Corresponding ICD Codes during Delivery Hospitalizations. Centers for Disease Control and Prevention; https://www.cdc.gov/reproductivehealth/maternalinfanthealth/smm/severe-morbidity-ICD.htm. Published 2018. Accessed May 10, 2018. [Google Scholar]
- 2.Sheen JJ, Smith HA, Tu B, Liu Y, Sutton D, Bernstein PS. Risk Factors for Postpartum Emergency Department Visits in an Urban Population. Maternal and Child Health Journal. 2019. [DOI] [PubMed] [Google Scholar]
- 3. Aziz A, Gyamfi-bannerman C, Siddiq Z, Wright JD, Goffman D, Sheen J, Alton MED Mph AMF. Maternal outcomes by race during postpartum readmissions. Am J Obstet Gynecol. 2019;(February):E pub ahead of print. 10.1016/j.ajog.2019.02.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Ray JG, Park AL, Dzakpasu S, Dayan N, Deb-Rinker P, Luo W, Joseph KS. Prevalence of Severe Maternal Morbidity and Factors Associated With Maternal Mortality in Ontario, Canada. JAMA Netw Open. 2018;1:e184571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kilpatrick SJ, Abreo A, Gould J, Greene N, Main EK. Confirmed severe maternal morbidity is associated with high rate of preterm delivery. Am J Obstet Gynecol. 2016;215:233.e1–7. [DOI] [PubMed] [Google Scholar]
- 6.Admon LK, Winkelman TNA, Zivin K, Terplan M, Mhyre JM, Dalton VK. Racial and Ethnic Disparities in the Incidence of Severe Maternal Morbidity in the United States, 2012–2015. Obstet Gynecol. 2018;132(5):1158–1166. doi: 10.1097/AOG.0000000000002937 [DOI] [PubMed] [Google Scholar]
- 7.Bateman B, JM M, Hernandez-Diaz S, KF H, Fischer M, Creanga A, Callaghan W, Gagne J. Development of a comorbidity index for use in obstetric patients. Obs Gynecol. 2014;122:957–965. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.ACOG. Practice Bulletin No. 102: management of stillbirth. Obstet Gynecol. 2009. March;113(3):748–61. doi: 10.1097/AOG.0b013e31819e9ee2. [DOI] [PubMed] [Google Scholar]
- 9.Bukowski R, Carpenter M, Conway D, Coustan D, Dudley DJ, Goldenberg RL, Rowland Hogue CJ, Koch MA, Parker CB, Pinar H, Reddy UM, Saade GR, Silver RM, Stoll BJ, Varner MW, Willinger M. Association between stillbirth and risk factors known at pregnancy confirmation. JAMA - J Am Med Assoc. 2011;306:2469–2479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Magann EF, Chauhan SP, Bofill JA, Waddell D, Rust OA, Morrison JC. Maternal morbidity and mortality associated with intrauterine fetal demise: five-year experience in a tertiary referral hospital. South Med J. 2001;94:493–495. [PubMed] [Google Scholar]
- 11.Ifnan F, Jameel M. Maternal morbidity and mortality associated with delivery after intrauterine fetal death. J Coll Physicians Surg Pak. 2006;16:648–651. [DOI] [PubMed] [Google Scholar]
- 12.Høj L, Da Silva D, Hedegaard K, Sandström A, Aaby P. Factors associated with maternal mortality in rural Guinea-Bissau. A longitudinal population-based study. BJOG An Int J Obstet Gynaecol. 2002;109:792–799. [PubMed] [Google Scholar]
- 13.Olagbuji BN, Ezeanochie MC, Igbaruma S, Okoigi SO, Ande AB. Stillbirth in cases of severe acute maternal morbidity. Int J Gynecol Obstet. 2012;119:53–56. [DOI] [PubMed] [Google Scholar]
- 14.Agency for Healthcare Research and Quality. Agency for Healthcare Research AndHCUP Database. Healthcare Cost and Utilization Project (HCUP) Rockville, MD: http://www.hcup-us.ahrq.gov/databases.jsp. [PubMed] [Google Scholar]
- 15.Lewkowitz AK, Rosenbloom JI, Keller M, Lopez JD, Macones GA, Olsen MA, Cahill AG. Association between stillbirth at or after 23 weeks gestation and acute psychiatric illness within one year of delivery. Am J Obs Gynecol. 2019;pii: S0002:Epub ahead of print. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kuklina E V, Whiteman MK, Hillis SD, Jamieson DJ, Meikle SF, Posner SF, Marchbanks PA. An enhanced method for identifying obstetric deliveries: Implications for estimating maternal morbidity. Matern Child Health J. 2008;12:469–477. [DOI] [PubMed] [Google Scholar]
- 17.Berthelsen C Evaluation of coding data quality of the HCUP National Inpatient Sample. Top Heal Inf Manag. 2000;21:10–23. [PubMed] [Google Scholar]
- 18.Fingar KR, Hambrick MM, Heslin KC, Moore JE. Trends and Disparities in Delivery Hospitalizations Involving Severe Maternal Morbidity, 2006–2015: Statistical Brief #243. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. [PubMed] [Google Scholar]
- 19.Al Kazzi ES, Lau B, Li T, Schneider EB, Makary MA, Hutfless S. Differences in the prevalence of obesity, smoking and alcohol in the United States Nationwide Inpatient Sample and the Behavioral Risk Factor Surveillance System. PLoS One. 2015;10:e0140165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.National Center for Chronic Disease Prevention and Health Promotion, Division of Nutrition, Physical Activity & O. State of Florida: Physical Activity, and Obesity Profile; https://www.cdc.gov/obesity/stateprograms/fundedstates/pdf/florida-state-profile.pdf. Published 2012. Accessed March 6, 2019. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.