Introduction: Hemophagocytic lymphohistiocytosis (HLH) is a rare clinical syndrome characterized by uncontrolled immune activation, resulting in multi-organ dysfunction and a high risk of death. Due to scarcity of data in adults, the diagnosis remains challenging and has been historically associated with a delayed time to diagnosis and heterogeneous practice patterns. We examined the processes of care measures and outcomes in a large cohort of adults with HLH.

Methods: This multicenter, retrospective cohort study examined the epidemiology, management, and outcomes of adult inpatients (≥18 years) diagnosed with HLH in the province of Alberta, Canada from January 1999 to December 2019. Patients were identified via International Classification of Diseases (ICD) codes, and those who met the HLH-2004 diagnostic criteria were included. Chart reviews were performed to collect demographics, clinical and laboratory characteristics, process of care measures, and outcomes. Overall survival (OS) was defined as the time from hospital presentation to death. Survival analysis was performed using the Kaplan-Meier method, and differences between groups were assessed using the log-rank test. Cox proportional hazards regression was used to assess the association between time to diagnosis and the hazard of death, adjusting for patient and disease characteristics.

Results: 83 patients met the HLH-2004 diagnostic criteria; 1 had a clear alternative diagnosis, hence 82 were included in the final data analysis. The median age of the cohort was 47 (interquartile range, IQR 32-62) years; 57 (70%) were male. The etiologies of HLH were as follows: malignancy (38%), infection (35%), autoimmune disease (16%), and idiopathic/others (11%). The most common clinical and laboratory findings were hyperferritinemia (100%), fever (99%), cytopenias (94%), hypofibrinogenemia and/or hypertriglycermidemia (91%), and hemophagocytosis on biopsy (89%). The median H-score was 244, with no significant difference across HLH etiologies. Interim data analysis showed that sCD25, natural killer (NK) cell activity, and molecular studies were ordered in 69%, 36%, and 29% of cases.

The median length of hospitalization was 33 days (IQR 16-62), with a 30-day mortality rate of 25.6%. A majority of patients (55%) required admission to the intensive care unit (ICU) (Table 1). The median OS was 8.5 months. Patients with malignancy-associated HLH had a significantly worse median OS compared with those with non-malignancy-associated HLH (2.1 vs 54.5 months, p=0.0004; Figure 1). There were no differences in the median OS between B-cell (n=11) and T/NK lymphoproliferative disorders (n=13), or between new cancer diagnosis (n=27) and active cancer receiving chemotherapy (n=4). Age (adjusted hazards ratio, aHR 1.06, 95% CI 1.01-1.10), Charlson comorbidity index (aHR 1.4, 95% CI 1.2-1.8), and ICU admission (aHR 4.1, 95% CI 1.4-12.2) were associated with increased hazards of death on multivariable Cox proportional hazards model. Malignancy was associated with increased hazard of death on univariate analysis (HR 2.8, 95% 1.5-4.9), but did not remain as an independent prognostic factor on multivariable analysis (aHR 1.0, 95% 0.4-2.6). Sex, delayed diagnosis, or peak ferritin level were not significant predictors of mortality.

The median times from presentation to diagnosis and from presentation to therapy were 10 days (IQR 2-19) and 14 days (IQR 6-25), respectively. Practice patterns varied significantly between centers without differences in survival outcome (log-rank p=0.8). Most patients (83%) received corticosteroids, whereas etoposide, intravenous gammaglobulin, and cyclosporine were administered in 38%, 34%, and 18%, respectively. Twelve (15%) patients did not receive any HLH-specific treatment, mostly in infection-associated HLH (n=8). Of the 31 patients receiving steroids and etoposide, 21 (68%), 12 (39%), and 20 (65%) receivedpneumocystis jirovecii, antiviral, and antifungal prophylaxis, respectively.

Conclusion: In our two-decade provincial study, we observed that delayed diagnosis and treatment heterogeneity were common. Older age, comorbidities, and ICU admission were independent predictors of poor survival, whereas delayed diagnosis and malignant HLH were not. Development of a standardized provincial diagnostic and treatment protocol has the potential to improve the quality of care in adult HLH.


Sun:Sanofi:Other: Advisory board;Octapharma:Other: Advisory board;Pfizer:Other: Advisory board;Octapharma:Research Funding;Novo Nordisk:Other: Advisory board.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract