Umbilical Cord Blood Transplantation and Delay Engraftment Are Associated with Increased Risks of Diffuse Alveolar Hemorrhage

Diffuse alveolar hemorrhage (DAH) is a serious pulmonary complication of hematopoietic stem cell transplantation (HCT). The incidence of DAH in allograft recipients varies between 3%-7% with a reported mortality rate of 70% or higher. Although some risk factors are established, the effect of donor sources, especially umbilical cord blood (UCB), is unknown.

Methods: We studied patients receiving allogeneic HCT between 2008-2015 at the University of Minnesota. DAH defined by strict clinical criteria include acute onset of hypoxia in the present of diffuse pulmonary infiltrates on chest X- ray or CT of the scan, along with broncho-alveolar lavage (BAL) findings of progressively bloody alveolar washings.

Results: Of 1228 patients (658 bone marrow/peripheral blood stem cell (BM/PBSC) and 570 UCB) receiving allogeneic HCT, 59 patients (5%) were diagnosed with DAH. The median time to DAH from HCT was 30 days (range 3-168). Patients with DAH had the following characteristics: median age 32 years (range, 1-72), 66% were male, 42% had a smoking history, and 5% had history of prior lung disease. At the time of DAH, the median platelet count was 24 x 10⁹/L (range, 1.0-114), 34% had renal dysfunction (creatinine >1.3 mg/dL), 34% had liver dysfunction (ALT>35), 10 % had INR> 1.5, 46% had fever, 27% had documented systemic infection, and 19% had infectious organism isolated from BAL; Majority (92%) received high dose steroids for treatment of DAH and 75% required mechanical ventilator support.

In univariate analysis comparing DAH incidence, following factors were found associated with higher DAH incidence rate: patients with UCB transplants (7% vs. 3% in BM/PB, p=0.01), HCT for malignant disease (6% vs. 3% in nonmalignant disease, p<0.01),myeloablative conditioning regimens (6% vs 4% for NMA; p=0.04), total-body irradiation (TBI) in conditioning (6% vs. 3% for no TBI; p= 0.03) and patients with delayed engraftment (i.e., >21 days, 2% vs. 100%, p<0.01). There was no significant difference between these 2 groups regarding patient's age, sex or CMV serostatus.

When we compared UCB recipients with DAH (n=40) and without DAH (n=530), the DAH group received fewer total nucleated cells (TNC) x10⁷/kg recipient weight (0.39 (0.21-2.04) vs. 0.46 (0.03-99.01), p<0.01), more often had TBI (90.0% vs.76%, p=0.05), and double UCB grafts (90% vs. 63% single p=<0.01), and more neutrophil-recovery (70% vs. 91%, p=<0.01) and platelet-recovery (30% vs. 79%, p=<0.01).

In multivariate analysis, later neutrophil engraftment was the only risk factor associated with DAH for all patients and patients for the UCB cohort. Treatment-related mortality (TRM) at 6 months was 65% for the DAH group vs. only 14% for the no DAH group, p<0.01. Patients with DAH had significantly poorer 1 year overall survival (20% (95%CI 11-31%) vs. 70% (95% CI 68-73%) p <0.01). Poor survival in patients with DAH was not affected by donor type (13% (5-25%) in UCB vs. 37% (17-57%) in PB/BM p=0.14). Among UCB recipients, those with DAH had a higher 1 year mortality of 87% (95% CI 75% -95%) than those without DAH 31% (95% CI 27-35%), p < 0.01.

Conclusions: DAH after allogeneic HCT remains serious and most often fatal. MAC, TBI, malignant disease were risk factors for DAH. UCB HCT is associated with higher rates of DAH; particularly associated with late engraftment and with a smaller TNC dose infused (rather than unit number). Interestingly, prior lung disease or history of smoking did not have any effect on the incidence of DAH. Current therapy of DAH with high dose steroids is inadequate, and mortality rates in patients with DAH remain very high. Therefore, studies to expedite engraftment after UCB HCT and to further understand the pathophysiology of DAH are essential.