Introduction: Brazil is currently facing a dengue fever (DF) epidemic, with the highest number of cases and deaths ever reported in the city of São Paulo. Although DF is an increasing public health issue worldwide, there are few reports of clinical and laboratorial characteristics of this disease in patients with hematological diseases.

Objective: to report the characteristics and outcomes of DF in 34 adult patients followed at a Hematology Day Hospital facility in Brazil.

Methods: Retrospective, observational study conducted at the Instituto Central do Hospital das Clínicas, Faculdade de Medicina da Universidade de São Paulo (ICHCFMUSP). Patients seeking medical care at the Day Hospital were evaluated for dengue fever (DF). If DF was suspected, either an antigen test (NS1) or IgM serology was performed to confirm the diagnosis. During follow-up, patients were hospitalized if there was clinical or laboratory evidence of deterioration. Data were collected from medical records using REDCap electronic data capture tools hosted at ICHCFMUSP and analyzed on R v3.3.4. Laboratory exams are presented in median values with ranges. This study was approved by the local ethics committee.

Results: From March to June 2024, out of 48 suspected cases, 34 had either a positive NS1 or IgM test and were included in the analyses. 56% were female and median age was 38 years (range, 19-60). 11 patients had sickle cell disease (SCD) (8 SS, 2 Sb0, 1 SC), 7 had myelodysplastic syndrome (MDS), 4 had a myeloproliferative syndrome, 3 had immune thrombocytopenic purpura (ITP), 2 had hemophilia and 7 had another disease. 29% had previous underlying liver disease, mostly secondary hemochromatosis; 62% had other commorbidities. NS1 was positive in 94%; IgM was positive in 74%. At admission, 85% had myalgia, 53% had arthralgia, 88% fever, 65% headache and 9% exanthema. With respect to DF complications, 9% presented liver failure, 24% had a bleeding episode, 6% had hypovolemic shock and 3% had a cavitary effusion. 38% received at least one unit of red blood cell transfusion and 15% had platelet transfusion. 26% were hospitalized, of whom 67% were male, 44% had SCD and 56% had previous liver disease. 2 patients, both with SCD SS, were admitted in intensive care unit, of whom 1 developed bone marrow necrosis and died of multiorgan failure six days after the onset of symptoms; her autopsy showed pulmonary thromboembolism and bone marrow emboli in lung vessels. The other patient had ischemic hepatitis, secondary bacterial infection, needed mechanical ventilation, vasoactive drugs and kidney replacement therapy, but her conditions improved, and she was discharged after 31 days.

Regarding blood cell counts, at admission (n=31), Hb was 9 (5.3-16.6) g/dL, hematocrit was 26 (14-54) %, white blood cell (WBC) count was 2700 (0.1-17.7)/mm3 and median platelet count (PLT) was 95 (2-492)/mm3; aspartate aminotransferase (AST) was 81 (21-1,267) mg/dL and alanine aminotransferase (ALT) was 64 (8-832) mg/dL. At day 5 (n=13), Hb was 9 (5.6-14.2) g/dL, WBC was 3.1 (0.3-12.1)/mm3, PLT was 60 (7-217)/mm3, AST was 98 (48-148)mg/dL and ALT was 47 (15-85) mg/dL. At day 10 (n=6), Hb was 10.5 (5.8-12.9)g/dL, WBC 2.1 (0.2-6.1)/mm3, PLT 102 (14-252)/mm3, AST 27 (21-33) mg/dL and ALT 22 (21-23) mg/dL.

Discussion: to date, this is one of the largest cohorts of DF in patients with hematological diseases. Although DF warning signs that rely on blood cell counts are difficult to evaluate in such patients, they might present a higher risk of severe disease, especially those with cytopenias and/or subjacent liver disease. Indeed, the rate of hospitalization of 26% is higher than the rate of severe cases (2% out of symptomatic cases) in non-hematological settings (1), and several patients needed blood transfusions. In line with our findings, patients with SCD are known to have worse outcomes in DF, with several deaths reported, but to date this is the first report of bone marrow necrosis leading to death due to DF. The small sample size precluded univariate analyses; thus, larger cohorts are needed to establish risk factors of worse outcomes in this population.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution