Complications during Delivery Hospitalizations in Women with Thalassemia: An Analysis of the National Inpatient Sample Database (2002-2019)

Background: Women with thalassemia generally have some degree of anemia that is exacerbated during pregnancy. A few studies have shown an association of thalassemia with adverse pregnancy outcomes like intrauterine growth restriction and low birth weight. However, there is a paucity of data regarding the burden of maternal complications during hospitalization for delivery.

Methods: The National Inpatient Sample Database (NIS) was queried with the International Classification of Disease, Ninth Revision, and Tenth Revision (ICD-9, ICD-10) codes to identify delivery hospitalizations and thalassemia diagnosis. To strengthen the statistical analysis, case-control matching was performed for age, race, and general co-morbidities like diabetes, hypertension, hypothyroidism, liver disease, obesity, smoking, drug abuse, alcohol use disorder, and electrolyte abnormalities to match delivery hospitalizations with thalassemia to delivery hospitalizations without thalassemia. With the matched data, thalassemia hospitalizations were further subclassified into 'alpha-thalassemia', 'beta-thalassemia’ and 'thalassemia-other'. A secondary multinominal logistic regression was performed to account for median income, admission type, insurance type, preterm and/ or c-section deliveries in addition to the previously mentioned comorbidities to generate the odds ratio for the preselected outcomes. We also compared the cost-of-care for delivery hospitalization with among the three groups of thalassemia patients compared to those without thalassemia.

Results: A data sample that constituted of 14,514,439 (14.5 million) delivery hospitalizations without thalassemia and 19,764 hospitalizations with thalassemia (0.1% of the sample) between 2002-2019 was taken. Unadjusted analysis revealed patients with thalassemia were older than those without thalassemia (30.1 years vs 28.3 years, p<0.001), had higher prevalence of obesity (4.7% vs 1.9%, p<0.001), hypothyroidism (4.6% vs 2.6%, p<0.001), liver disease (0.7% vs 0.2%, p<0.001), diabetes mellitus (1.2% vs 1.0%, p<0.001), chronic hypertension (0.4% vs 0.2%), and multiple gestation (2.7% vs 1.9%, p<0.001). Patients with thalassemia were also more likely to be black (24.4% vs 14.1%, p<0.001), or Asian/ Pacific Islanders (27.4% vs 5.5%, p<0.001) (Figure 1). A total of 17944 hospitalizations with thalassemia were matched with their 17944 counterparts by case-control matching. A multinominal logistic regression assessing for major outcomes revealed that patients with thalassemia of any type had higher odds of requiring a blood transfusion during the hospitalization- alpha thalassemia: odds ratio (OR)= 3.52, (95% CI: 2.57-4.83; p<0.001), beta-thalassemia: OR= 4.77, (95% CI: 3.62-6.28; p<0.001), thalassemia-unspecified: OR= 3.03, (95% CI: 2.58-3.55; p<0.001). Patients with thalassemia of any type also had a statistically higher risk of coagulopathy, preeclampsia, post-partum hemorrhage, and cardiac arrhythmias. An independent-samples Mann-Whitney U test revealed patients with thalassemia had a statistically higher median cost of hospitalization ($15,565 v $12,400; p<0.001). There was no statistical difference in length of stay or all cause mortality between patients with thalassemia compared to those without thalassemia.

Conclusion: Despite an overall low prevalence of thalassemia in the US, these results suggest that pregnant women with thalassemia are at a higher risk of complications during hospitalization for delivery. This is especially important when considering that thalassemia is disproportionally more prevalent in minority groups. Though anemia is strongly associated with poor pregnancy outcomes, whether the genetic abnormalities with thalassemia confers additional risk merits further study.

Jaglal:Sanofi: Consultancy; Novartis: Consultancy.