Adult Hemophagocytic Lymphohistiocytosis(HLH): Experience of an Urban, Public Hospital over Two Decades

Introduction:HLH is a rare, life-threatening disorder, characterized by hyperstimulation of immune system leading to systemic inflammation and multi-organ failure. It is categorized as primary and secondary HLH. Secondary HLH usually affects adolescents and adults. It results from acquired immune dysregulation secondary to a number of etiologies, including infections, malignancy, and autoimmune diseases. Owing to less epidemiological data, adult HLH is thought to be underdiagnosed, making a true assessment difficult, however, some observational data suggest 40% of HLH cases occurs in adults. Disease presentation includes fever, cytopenias, organomegaly, liver function anomalies, elevated ferritin levels, and/or demonstration of macrophage activation in hematopoietic organs. In 2014, Fardet et al proposed the H-Score, a novel diagnostic score derived from 162 adult patients with HLH.We aim to report a retrospective review of Adult HLH in an urban safety-net hospital over the course of two decades along with predictive value of H-score in our patient population.

Methods:We conducted a retrospective review of patients diagnosed with HLH at Cook County Health, Chicago between January 2000 and January 2019 after approval by the Institutional Review Board. Patients were identified from electronic records using ICD-10 codes D76.1, D76.2, and ICD-9 code 288.4. Patients under 18 years were excluded. MS excel was used for data collection and further descriptive statistics were calculated with frequencies and percentage.

Results:After initial review, 12 confirmed and eligible cases were included in the study. Mean age at diagnosis of adult HLH at our center was 37, with male predominance(7 males, 4 females, and 1 female transgender). 5 were African-American, 6 were Hispanic, and 1 was Asian. Most common presentation was fever, seen in 10 out of 12 cases, along with variety of symptoms like fatigue, sore throat and jaundice.4 out of 12 patients (33%) had HIV/AIDS, with CD4 counts between 79 to 180. 3 were already receiving anti-retroviral therapy at the time of HLH diagnosis, while 1 was diagnosed with HIV/AIDS at the time of HLH diagnosis. Etiologic spectrum mainly included infectious (4 HIV and 3 EBV) and autoimmune (2 systemic lupus erythematous, 1 cold immune hemolytic anemia with immune thrombocytopenic purpura) causes. 1 patient had an underlying malignancy (diffuse large B-cell lymphoma). Etiology was not established in 1patient with no familial associations found in subsequent genetic evaluation.

All patients had elevated liver enzymes. The mean ferritin level in our cohort was 19,198 ng/ml. Leucopenia was seen among most cases, 11 out of 12. The 1 patient noted to have a high white cell count was actually receiving corticosteroid therapy for cold immune hemolytic anemia. Most common bone marrow findings were hemophagocytosis (9 patients) and hypocellularity (7 patients). 2 had hypercelullar marrow and 1 had normal marrow. Genetic testing was performed in 4 patients; chromosomal abnormalities were not observed in any.

Specific lab parameters in our cohort as included in HLH-2004 criteria is shown in Table 1.

Calculated H scores in our cohort is shown in table 2. 11 patients fall under high probability for HLH.

Conclusion:The most widely used diagnostic criteria for HLH is the HLH-2004 diagnostic criteria, derived from pediatric HLH study. It is often extrapolated for use in adults. There are several limitations to the HLH-2004 diagnostic criteria. sIL2r and NK function testing is not available in all centers, and many of the manifestations of HLH in adults are not included in the criteria. When using H score, a cutoff value of 169, corresponds to sensitivity of 93% and specificity of 86% in diagnosing HLH. In our study, 11 out of 12 patients (91.66) scored higher than 169, which is highly suggestive of HLH. The remaining 1 patient with H score of only 118, however, met the diagnosis of HLH by HLH-2004 criteria (score: 5/8). Therefore, although our study population was small, results of our study were in favor of using H-score as an appropriate diagnostic tool in adult-onset HLH, which also helps mitigate the restrictions of HLH-2004 criteria in adult population.

View largeDownload slide

View largeDownload slide

Close modal