Severe maternal morbidity is increasing in the US.1 The rising rate of maternal morbidity is likely driven by a multitude of factors, including increasing maternal age, higher rates of obesity, increased medical comorbidity among patients, and a rising cesarean delivery rate.1,2
To date, little data have examined the association between cancer and maternal morbidity and mortality.2 Not only is the incidence of cancer in women of reproductive age increasing, but as treatment improves, women with cancer are living longer.3 In addition, there has been an increased focus on fertility preservation in both men and women of reproductive age and with cancer. The objective of this cohort study was to examine nationwide trends and characteristics of severe maternal morbidity and mortality among pregnant women with cancer in the US.
We performed a population-based retrospective cohort study using data from the National (Nationwide) Inpatient Sample.4 We analyzed 14 648 135 deliveries for national estimates from January 2016 through December 2019. We examined patients with an International Statistical Classification of Diseases and Related Health Problems, Tenth Revision code for cancer, as defined by the National Cancer Institute5 and American Cancer Society classification6 for case identification (40 types in 16 systems). Patients with cancer were compared with those without. This study was deemed exempt by the institutional review board at the University of Southern California owing to use of publicly available, deidentified data.
The first step of analysis was to examine patient, hospital, and pregnancy characteristics associated with cancer, which we assessed using a multivariable binary logistic regression model (backward selection). The second step of the analysis was to examine severe maternal morbidity, as defined by the Centers for Disease Control and Prevention (21 indicators),1 and mortality during the index admission for delivery among patients with cancer.
To account for the effect of clinical and demographic characteristics on cancer, we performed a propensity score analysis with inverse probability of treatment weighting (IPTW). A log Poisson regression generalized linear model including clinical and demographic characteristics associated with cancer was developed. The propensity score weights were then estimated to create IPTW cohorts. To account for possible confounder and mediator effects on severe maternal morbidity, the exposure-outcome association in the IPTW cohort was adjusted for obstetric and delivery factors.
The prevalence rate of pregnancy with cancer was 69.3 per 100 000 deliveries and increased from 64.5 to 73.4 between 2016 and 2019 (relative increase, 13.8%; P < .001). The most common cancer type was breast cancer, followed by lymphoma, leukemia, and gynecologic cancers. Over time, the prevalence of skin, soft tissue, breast, and oral cavity/pharyngeal cancers increased. Factors associated with cancer included (1) patient characteristics with advanced age, more recent year of delivery, White race, obesity, smoking, preexisting hypertension, and chemotherapy exposure, and (2) pregnancy characteristics with early preterm delivery and cesarean delivery (Table 1).
The cancer and noncancer groups were well balanced in the IPTW model (standardized difference, ≤0.04). After accounting for potential mediators between cancer and adverse outcomes, including obstetric factors and delivery factors, patients in the cancer group vs noncancer group were more likely to experience severe maternal morbidity (97.4 vs 17.3 per 1000 deliveries; adjusted odds ratio, 3.63; 95% CI, 3.42-3.84) and death during the index hospital stay for delivery (2.2 vs <0.1 per 1000 deliveries; adjusted odds ratio, 13.34, 95% CI, 8.72-20.39) (Table 21). Similar associations were observed for the majority of the individual morbidities, including sepsis, ventilation, and hysterectomy.
The contemporaneous analysis of this cohort study found that among pregnancies complicated by cancer, the risks of severe maternal morbidity and mortality were increased. Key limitations included lack of information on cancer stage and anticancer treatment, disease status at the time of delivery, inability to perform cancer type–specific analyses owing to limited sample size, and ascertainment bias. Cause of death, oncologic outcomes, neonatal outcomes, and information after discharge were also not available in the database used.
Accepted for Publication: April 11, 2022.
Published Online: June 16, 2022. doi:10.1001/jamaoncol.2022.1795
Corresponding Author: Koji Matsuo, MD, PhD, Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Keck School of Medicine, University of Southern California, 2020 Zonal Ave, IRD 520, Los Angeles, CA 90033 (koji.matsuo@med.usc.edu).
Author Contributions: Drs Matsuo and Mandelbaum had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Matsuo, Klar, Youssefzadeh, Mandelbaum, Ouzounian.
Acquisition, analysis, or interpretation of data: Matsuo, Klar, Mandelbaum, Roman, Wright.
Drafting of the manuscript: Matsuo, Klar, Ouzounian.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Matsuo, Klar.
Obtained funding: Matsuo.
Administrative, technical, or material support: Matsuo, Mandelbaum, Ouzounian.
Supervision: Klar, Roman, Ouzounian, Wright.
Conflict of Interest Disclosures: Prof Klar reported personal fees from CooperSurgical, KLS Martin, and GlaxoSmithKline outside the submitted work. Dr Roman reported personal fees from Quantgene outside the submitted work. Dr Wright reported grants from Merck and personal fees from UpToDate outside the submitted work. No other disclosures were reported.
Funding/Support: This work was supported by an Ensign Endowment for Gynecologic Cancer Research to Dr Matsuo.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Information: The data on which this study is based are publicly available on request from the Healthcare Cost and Utilization Project of the Agency for Healthcare Research and Quality (https://www.hcup-us.ahrq.gov/nisoverview.jsp).