Obstetrical deliveries associated with maternal malignancy in California, 1992 through 1997

OBJECTIVE: This study aims to characterize the rate of occurrence and nature of outcomes associated with obstetrical deliveries in women with malignant neoplasms among 3,168,911 women who delivered in California in 1992 through 1997. DESIGN: The study is a population-based retrospective review of infant birth and death certificates and maternal and neonatal discharge records. Cases of malignant neoplasms associated with obstetrical delivery were attributed to 1 of 3 categories, depending on the earliest documented hospital discharge diagnosis, as follows: "prenatal" if the diagnosis was first documented by hospitalization within 9 months preceding delivery, "at delivery" if the diagnosis was established from the delivery hospitalization, or "postpartum" if the diagnosis was first documented by hospitalization within 12 months after delivery. METHODS: Computer-linked infant birth and death certificates and maternal and neonatal discharge records were used to identify cases and outcomes. Cases of malignant neoplasms were identified by using International Classification of Diseases, Ninth Revision codes (140-208). Noninvasive neoplasms and carcinoma in situ neoplasms were excluded. In analysis of outcomes, the Mantel-Haenszel estimate for adjusted odds ratios was used. RESULTS: Among 3,168,911 obstetrical deliveries over the 6-year span, a total of 2247 cases of primary malignancy were identified. The observed rate of occurrence for primary malignant neoplasms was 0.71 per 1000 live singleton births. Most cases (53.3%) were first documented in the postpartum period as follows: prenatal, 587 cases (0.18 per 1000); at delivery, 462 cases (0.15 per 1000); and postpartum, 1198 cases (0.38 per 1000). The most frequently documented primary malignant neoplasms associated with obstetrical delivery were breast cancer (423 cases, 0.13 per 1000), thyroid cancer (389 cases, 0.12 per 1000), cervical cancer (266 cases, 0.08 per 1000), Hodgkin's disease (172 cases, 0.05 per 1000), and ovarian cancer (123 cases, 0.04 per 1000). Odds ratios for a variety of demographic factors identified maternal age as the most significant risk factor for development of malignant neoplasms (age greater than 40 vs 20-25, odds ratio 5.7, CI 4.6-6.9). Age-adjusted odds ratios for maternal cancer of any type suggested significantly elevated risks for cesarean delivery (odds ratio 1.4, CI 1.3-1.6), blood transfusion (odds ratio 6.2, CI 4.5-8.5), hysterectomy (odds ratio 27.4, CI 20.8-36.1), and maternal postpartum hospital stay greater than 5 days (odds ratio 30.6, Cl 27.9-33.6), but not for postpartum maternal death (odds ratio 0.8, Cl 0.6-1.0). Odds ratios also suggested significantly elevated risks for premature newborn (odds ratio 2.0, Cl 1.8-2.2), very low birth weight (odds ratio 2.9, Cl 2.2-3.8), and newborn hospital slay longer than 5 days (odds ratio 2.6, Cl 2.4-3.0), but not for neonatal death (odds ratio 1.6, Cl 0.8-3.1) or infant death (odds ratio 1.2, Cl 0.5-3.3). However, several types of malignant neoplasms did confer significant elevations in risk for neonatal death. Hospital charges for both maternal and neonatal care were significantly elevated in the maternal malignant neoplasm group. CONCLUSION: A lower than expected occurrence rate of obstetrical delivery associated with maternal malignancy was seen when compared with previously published hospital-based reports. Malignant neoplasms associated with obstetrical delivery were most frequently first documented in the postpartum period. Maternal and neonatal morbidity were significantly increased, yet the risk of in-hospital maternal death was not significantly elevated. A significant increase in risk of neonatal death for infants of mothers with cervical cancer was found.