EHA Library - The official digital education library of European Hematology Association (EHA)

Abstract: E1169

Type: Eposter Presentation

Background
Obese patients were reported to have poorer prognosis in pediatric acute lymphoblastic leukemia (ALL), however, the association with infection during chemotherapy was seldom reported.

Aims
To evaluate the difference of infectious characteristics and long-term outcome between obese and non-obese patients.

Methods
Total 252 newly diagnosed pediatric ALL patients from 1997 to 2012 at National Taiwan University Hospital treated by Taiwan Pediatric Oncology Group (TPOG) protocol were retrospectively reviewed. The fever event was classified into fever of unknown focus, clinically documented infection and microbiologically documented infection (MDI). Infectious sites were recorded by location: upper respiratory (EENT), lower respiratory (LRI), blood stream (BSI), genitourinary tract (GU), gastrointestinal tract, skin or soft tissue, central nervous system and others. Multivariate analysis for risk factors of infectious complications was evaluated by Poisson regression with incidence rate ratio (IRR) and 95% confidence interval (CI).

Results
Total 219 patients (86.9%) had fever with mean 2.74 episodes per person. Obese patients (n= 38) harbored more MDI, GU and LRI with pathogen infections than non-obese patients (IRR = 1.349, 2.228 and 2.014; 95% CI = 1.109-1.642, 1.558-3.187 and 1.22-3.326; and p = 0.003, < 0.001 and 0.006, respectively). Subgroup analysis revealed the IRR in obese patients was higher in female and younger age patients (Table). The pathogen characteristics in obese patients included more Gram negative pathogen both in blood stream (29% vs. 14%, p = 0.029) and GU tract (32% vs. 10%, p = 0.001). Klebsiella spp. bacteremia was higher in obese patients (11% vs. 3%, p = 0.047). Overall prognosis analysis showed obese patients had more relapsed rate (42.1 vs. 22.4%, p = 0.012) but the mortality rate was not different (36.8 vs. 23.4%, p = 0.063). The 10-year event-free survival was also lower in obese patients (54 ± 8% vs.72 ± 3%, p = 0.002, figure). Table. Subgroup Univariate Poisson Regression Analysis in Obese vs Non-obese Patients.

		MDI			GU			LRI-MDI
Variables	IRR	95% CI	p-value	IRR	95% CI	p-value	IRR	95% CI	p-value
Gender
Male	0.846	0.594-1.206	0.355	1.947	0.958-3.956	0.066	0.790	0.242-2.581	0.696
Female	1.683	1.321-2.145	<0.001	2.053	1.353-3.113	0.001	2.543	1.424-4.543	0.002
Age (years)
< 5	1.223	0.904-1.654	0.191	2.903	1.630-5.168	<0.001	3.728	2.073-6.705	<0.001
5-10	2.046	1.480-2.827	<0.001	2.764	1.474-5.183	0.002	1.102	0.323-3.759	0.877
> 10	0.948	0.613-1.467	0.811	1.213	0.619-2.376	0.573	-	-	-*
Risk Group
SR	1.365	0.908-2.054	0.135	4.013	1.894-8.503	<0.001	3.330	1.494-7.426	0.003
HR	1.563	1.174-2.080	0.002	2.462	1.383-4.384	0.002	0.990	0.415-2.361	0.982
VHR	1.016	0.706-1.462	0.932	1.401	0.774-2.533	0.265	3.537	1.223-10.225	0.020

*No case in this subgroup.LRI-MDI, lower respiratory infection with pathogen; SR, standard risk; HR, high risk; VHR, very high risk.

Conclusion
Obesity was a potential risk factor in infectious complications in pediatric acute lymphoblastic leukemia patients, especially in microbiologically documented infection and genitourinary tract infection.



Session topic: E-poster

Keyword(s): Acute lymphoblastic leukemia, Infection, Obesity, Pediatric