High Bleeding Events in SCD Warrant Consideration of In-Hospital Ulcer Prophylaxis

By the time they reach age 40, 21% of patients with sickle cell disease (SCD) will have experienced a bleeding event, with most events related to the upper gastrointestinal (GI) tract, according to a retrospective study published in Blood Advances. The findings, according to lead author Nisha Hariharan, MD, and colleagues from the University of California (UC) Davis School of Medicine, indicate the need to reduce nonsteroidal anti-inflammatory drug (NSAID) use whenever possible in these patients. In addition, the findings suggest clinicians should consider empiric ulcer prophylaxis in the management of hospitalized patients with SCD.

In the retrospective study, Dr. Hariharan and colleagues identified thousands of patients with SCD who were hospitalized between 1991 and 2014 to understand bleeding incidence, as well as bleeding types, risk factors for bleeding, and bleeding-associated mortality in the SCD population.

The researchers collected data from 6,423 patients using longitudinal records from the California Patient Discharge Dataset and the Emergency Department Utilization database. All patients were younger than age 65.

Of all identified patients, 1,345 (20.9%) experienced an incident bleeding event over a median follow-up duration of 215.7 months (range = 212.5-219.8). The most common type of bleeding event was GI bleeding (41.6%), followed by epistaxis (19.1%), menorrhagia (15.1%), intracranial hemorrhage (ICH; 10.2%), hemophthalmos (5.1%), and gross hematuria (2.6%). Most of the GI bleeds (n=401; 60%) occurred in the upper GI tract.

On average, age at first bleeding event was 32.9 years (standard deviation = ±14.6 years), and the cumulative incidence of a first bleeding event by the time patients reached age 40 was 21% (incidence rate = 1,603 per 100,000 person-years).

Corresponding study author Ted Wun, MD, director and principal investigator of the UC Davis Clinical and Translational Science Center, told ASH Clinical News that the high incidence of GI bleeds might be attributable to the frequent use of NSAIDs in patients with SCD. NSAIDs are commonly used as part of the management of pain due to osteonecrosis of the femoral head (ONFH) and vaso-occlusive crises in this population. In addition, the use of anticoagulants, particularly following a venous thromboembolism (VTE), may contribute to bleeding risk in these patients.

Frequent hospitalization, defined as 3 or more visits per year, occurred in 63.5% of patients with a bleeding event. Frequent hospitalization and less frequent hospitalization were associated with first bleed incidence rates of 30.5% and 12.6%, respectively, and less frequent hospitalization was associated with incident bleed by age 40. Approximately 12.8% of patients with frequent hospitalization experienced a GI bleed by age 40, whereas only 2.9% of patients with less frequent hospitalization had a GI bleed by this age.

Frequent hospitalization was associated with an increased risk of:

  • all bleeding events (hazard ratio [HR] = 2.16; 95% CI 1.93-2.42)
  • GI bleeding events (HR=2.73; 95% CI 2.29-3.26)

“One reason patients with SCD may have increased bleeding is stress ulceration from just being in the hospital,” said Dr. Wun. He also noted that patients who are extremely ill and those with frequent hospitalizations are, as suggested by the study, more likely to experience a stress ulceration and higher bleeding. Clinicians may want to consider giving patients a proton-pump inhibitor to prevent stress ulceration in these patients, he said.

Dr. Wun and researchers also found that occurrence of VTE within 180 days of the bleeding event was associated with an increased risk of:

  • all bleeding events (HR=4.24; 95% CI 2.86-6.28; p<0.0001)
  • ICH (HR=4.82; 95% CI 1.72-13.45; p=0.0027)
  • GI bleeding event (HR=3.19; 95% CI 1.80-5.66; p<0.0001)

A higher bleeding risk also was associated with ONFH (HR=1.25; 95% CI 1.08-1.46; p=0.0034) and ischemic stroke (HR=1.65; 95% CI 1.20-2.26; p=0.0022). The strongest associations between the occurrence of VTE within the 180-day period and the risk of bleeding were with epistaxis (HR=6.00; 95% CI 3.18-11.34) and menorrhagia (HR=5.10; 95% CI 2.25-11.57).

“One reason patients with SCD may have increased bleeding is stress ulceration from just being in the hospital.”

—Ted Wun, MD

Experiencing any bleeding event was independently associated with a twofold increase in mortality (HR=2.10; 95% CI 1.82-2.43), after adjustment for other complications of SCD, including acute coronary syndrome, VTE, organ failure, and ischemic stroke. Compared with all bleeding types evaluated in this study, ICH held the strongest association with mortality (HR=2.79; 95% CI 2.17-3.59), followed by GI bleeding (HR=1.74; 95% CI 1.48-2.05). The in-hospital case fatality rates for ICH and GI bleeding were 24.7% and 5.68%, respectively.

“This report, to our knowledge, is the first study of the incidence of bleeding events in patients with SCD,” the authors concluded. They noted, however, that the total number of bleeding events may be underestimated in their report, and they were unable to describe the overall natural history of all subsequent bleeding events because the study focused on incident bleeding. “An examination of the lifetime burden of bleeding events may be an area of interest for future studies,” the investigators added.

Other limitations of the study include the reliance on ICD-9 diagnostic coding, which is prone to inaccuracies, the lack of genetic data (e.g., HbSS vs. HbSC) and laboratory and medication data, the inclusion of only datasets of patients from California, and the lack of data on bleeding events that were managed outside of the hospital.

The authors report no relevant conflicts of interest.


Hariharan N, Brunson A, Mahajan A, Keegan THM, Wun T. Bleeding in patients with sickle cell disease: a population-based study. Blood Adv. 2020;4:793-802.