Predictive role of cerebrospinal fluid hydrogen sulfide in central nervous system leukemia.
- Author:
Shu-Xu DU
1
;
Jiang XIAO
;
Feng GUAN
;
Li-Ming SUN
;
Wan-Shui WU
;
Hong TANG
;
Jun-Bao DU
;
Chao-Shu TANG
;
Hong-Fang JIN
Author Information
- Publication Type:Journal Article
- MeSH: Adolescent; Central Nervous System Neoplasms; cerebrospinal fluid; metabolism; pathology; Child; Child, Preschool; Cystathionine beta-Synthase; genetics; Female; Humans; Hydrogen Sulfide; cerebrospinal fluid; Leukemia; cerebrospinal fluid; Lyases; genetics; Male
- From: Chinese Medical Journal 2011;124(21):3450-3454
- CountryChina
- Language:English
-
Abstract:
BACKGROUNDCentral nervous system leukemia (CNSL) is an important relapse in children with acute lymphoblastic leukemia (ALL). We investigated the possible role of endogenous hydrogen sulfide (H(2)S) of cerebrospinal fluid (CSF) in predicting CNSL.
METHODSFrom August 2008 to December 2010, 380 children were enrolled in this study at Shijitan Hospital, China. These children were from 2 to 16 years old, and the median age was 6.5 years. They were divided into a CNSL group (7 cases), a leukemia group (307 cases), a non-leukemia group (26 cases) and a healthy group (40 children). CSF specimens were obtained from conventional lumbar punctured, then centrifuged and supernatants preserved for H(2)S detection. Leukemic cells precipitates from CSF were found in three cases, the hCSE and hCBS mRNA expression was detected by reverse transcription polymerase chain reaction (RT-PCR), and H(2)S levels in serum were also measured. The receiver operating characteristic (ROC) curve and area under curve (AUC) were used to assess the predictive diagnosis role of CSF H(2)S in children with ALL and CNSL.
RESULTSThe serum H(2)S contents of the CNSL and leukemia groups were (96.98 ± 15.77) µmol/L and (93.35 ± 17.16) µmol/L respectively, much higher than those of healthy, (44.29 ± 2.15) µmol/L, and non-leukemia, (46.32 ± 6.54) µmol/L, groups (P < 0.01). Compared with the leukemia group, CSF H(2)S content of the CNSL group was significantly high (P < 0.01). Meanwhile, in contrast to the non-leukemia group, CSF H(2)S contents of the CNSL and leukemia groups were both significantly increased (P < 0.01). In addition, leukemic cells from CSF precipitations could express CBS and CSE mRNA. Furthermore, the ROC analysis showed the UAC was 0.929 (95%CI: 0.857 - 1.000), and the optimum cut-off value of CSF H(2)S was 12.08 µ mol/L, and the sensitivity and specificity were 83.3% and 97.2% respectively.
CONCLUSIONSCSF H(2)S contents were significantly increased in children with CNSL. After treatment, H(2)S contents were decreased subsequently. Therefore, we speculated that H(2)S levels of CSF would predict CNSL in ALL children.