Diabetic Ketoacidosis in Pregnancy

Chloe Zera, MD, MPHAs many as 2% of all pregnancies in the United States are complicated by pregestational diabetes, with the prevalence increasing as rates of type 2 diabetes mellitus increase among young people.1 The well-characterized inequities in adverse maternal and fetal outcomes in the United States are heightened among individuals with diabetes, who are more likely to be members of racial and ethnic minority groups and to have socioeconomic deprivation than those without diabetes.2,3 Though the diagnosis is rare, diabetic ketoacidosis (DKA) results in significant health care expenditures and is still the most significant driver of morbidity and mortality in young people with both type 1 and type 2 diabetes.4 When care is delivered well, DKA is a preventable complication. The two studies on DKA in pregnancy in this month's issue of Obstetrics & Gynecology (see pages 579 and 590) describe the current failures of health care delivery for pregnant people with diabetes and should be seen together as an urgent call for quality improvement.5,6 Over the past 10 years, there have been rapid advances in the management of type 1 and type 2 diabetes. Despite this progress, Wen and colleagues found that, among pregnant people with diabetes, the weighted prevalence of DKA during delivery hospitalization in a nationally representative database increased from 0.7% in 2010 to 2.0% by 2020.5 The majority of these cases occurred in people with type 1 diabetes, although more than 10% were in those with type 2 diabetes. Similar to what is seen outside of pregnancy, younger age, public insurance, and low income were all associated with DKA. Due to the inherent limitations of administrative data, the authors were unable to connect these risk factors to inequitable care delivery; however, their findings are consistent with data that demonstrate persistent disparities in both access to technology and glycemic control in youth with type 1 diabetes.7 Grasch et al6 provide us with more details, representing the first case series reporting detailed outcomes in the contemporary era of diabetes care. This case series describes 129 admissions for laboratory-confirmed DKA at a single Midwestern academic center between 2012 and 2021. Though most (77%) patients had type 1 diabetes, uptake of diabetes technology was low—only 20% of the patients used a continuous glucose monitor, and fewer than 30% used an insulin pump. Taken together, these results suggest that pregnancy diabetes programs should emulate successful quality-improvement initiatives that focus on access to technology8 while people have (often newly granted) access to care during pregnancy. Because diabetes is a common chronic condition, diabetes care is part of nearly every obstetrician's practice. The data in both studies highlight that diabetes, including type 2 diabetes, remains a serious and complicated illness. A sizeable minority of the cases of DKA in both studies occurred among people with type 2 diabetes, who, in the case series, required insulin infusions for twice as long as those with type 1 diabetes. Grasch et al also found significant obstetric morbidity among the affected pregnancies, including six pregnancy losses (three during hospitalization) and a high rate of indicated preterm delivery, with a median gestational age of 34 6/7 weeks (interquartile range 33 2/7–36 2/7 weeks). In Wen et al's study, nearly one in three of the pregnancies affected by DKA ended in preterm birth and nearly one in five patients experienced nontransfusion severe maternal morbidity. As facilities consider their maternal care capabilities, appropriate teams for people with diabetes should be a key consideration; in the Consensus Statement on Levels of Maternal Care, diabetes is highlighted among conditions that may require care at a level II (specialty) center.9 The data presented in this issue are an argument that anyone who develops DKA should have access to at least level III (subspecialty) maternal care due to the significant morbidity associated with the diagnosis. Ultimately, policy solutions are critical to improve health care delivery for people with diabetes. Pregnancy is considered a window of opportunity for chronic disease prevention and management, in large part because access to insurance coverage is nearly universal. Nonetheless, nearly half of pregnant people with low incomes in the 10 states that have not yet expanded Medicaid lack insurance before pregnancy, when care for diabetes is most likely to improve outcomes.10 Indeed, the patients in the study by Grasch and colleagues had generally suboptimal glycemic control, with a median hemoglobin A1C value of 8.1%. They authors also found that insulin nonadherence precipitated more than 20% of the cases of DKA. There are many reasons for insulin nonadherence, but cost is widely recognized to contribute. Although most state Medicaid programs provide access to insulin with minimal cost sharing, currently only 25 states and Washington, DC, enforce insulin cost caps for private payers.11 Finally, any improvements in diabetes-related health during pregnancy must be partnered with ongoing care after birth. Nearly all states have extended Medicaid coverage for 12 months postpartum, but time-bound insurance coverage is an insufficient solution to the postpartum care gap.12 If we want to avoid preventable morbidity and achieve the best possible pregnancy outcomes, we must advocate to ensure that all people with diabetes have continuous access to the medications, technology, and systems of care that improve outcomes before, during, and after pregnancy.