Jinhong Yu1, Yanwei Miao1, Lizhi Xie2, Bingbing Gao1, Yu Bing1, Li Yang1, and Ailian Liu1
1The First Affiliated Hospital of Dalian Medical University, Dalian, China, 2GE Healthcare, Beijing, China
Synopsis
Exposure
to high levels of bilirubin can cause severe motor
symptoms and cerebral palsy. The aim of this study
was to analyze the diagnostic value for early brain damage caused by hyperbilirubinemia
(HB) in neonates using T1-weighted imaging (T1WI)
and T2-weighted imaging (T2WI).
Introduction
Bilirubin encephalopathy (BE) is the most
serious complication of hyperbilirubinemia. Currently, there is no specific diagnostic indicators for early
monitoring of its occurrence. The purpose of this study was to analyze the
diagnostic value of quantitative values for early brain injury caused by
hyperbilirubinemia (HB) in neonates using
T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI).Methods
Seventy-five
cases of neonates with HB were retrospectively collected from October 2014 to
March 2019 by PACS system, including 46 males and 29
females, median age of 9 days, mean weight (3.38±0.47) kg. All children were
divided into two groups based on their globus pallidus(GP) T1WI signal intensity: including normal
signal intensity group (HBn group ( Figure 1 ): 46 patients, 29 males, 17
females, median age of 9 days, mean weight (3.38±0.47) kg) and high signal
intensity group (HBh group ( Figure 1 ) : 29 patients, 17 males and 12 females,
median age 8 days, mean weight (3.36± 0.46) kg). In addition, forty-four healthy
subjects (CON group) who matched the HB group by gender, age and weight were
recruited, including 32 males and 12 females, with median age of 8 days and
mean weight of (3.40±0.55) kg. The maximum levels of GP and frontal white matter (FWM )on T1WI and
T2WI were selected and the ROI was
manually selected. The signal intensity values of GP and FWM on both sides
were measured three times to obtain the
average value ( Figure 2 ); and the GP/FWM average signal intensity
ratios of T1 and T2 were
calculated (referred to as T1 ratio and T2 ratio).
The differences of all parameters were assessed using multiple independent
samples Krukal- Wallis H test among the three groups, while the ROC curve was used
to determine the diagnostic performance of each parameter.Results
There was significant
difference in T1 ratios among three groups (median ratios were 2.135, 2.405,
3.075, P < 0.05 after adjustment). In addition, the ROC curve analysis
showed that when the T1 ratio was 2.675 as the threshold, the area under curve
(AUC) between HBh group and CON group was the largest (0.973), and the
sensitivity and specificity were 0.948 and 0.898, respectively( Figure 3 C). When the T1
ratio between HBn group and HBh group was 2.655, the AUC
was the largest (0.926), and the sensitivity and specificity were 0.926 and 0.983,
respectively ( Figure3 B). When the T1 ratio between HBn group and CON group was
2.155, the AUC value was the highest (0.633) with the sensitivity and
specificity of 0.717 and 0.528, respectively ( Figure3 A).
Furthermore, there were significant differences in T2
ratios among three groups (median ratios were 0.700, 0.690, 0.655, P < 0.05
after adjustment). ROC curve analysis showed that when T2 ratio was 0.595 as
the threshold, the AUC of HBn group and HBh group was the
largest (0.274), with sensitivity of 1.000 and specificity of 0.011 ( Figure3 B). When the
T2 ratio between CON group and HBh group was 0.595, the AUC was 0.202,
the sensitivity was 1.000, and the specificity was 0.034( Figure3 C). There was no obvious difference
between CON group and HBn group (P < 0.05 after adjustment).Discussion and conlclusion
GP
involvement is a well-known MRI finding of BE 1, 2. Some scholars used
the signal intensity threshold of the GP to determine the occurrence of BE3,but
there is lack of accuracy due to many factors. Therefore, in this study, the
GP/FWM signal intensity ratio was used as a reference value. Our data indicated
significant differences in T1 and T2 ratios between HBn group, HBh
group and control subjects. In addition, when the T1 ratio was >2.155,
although no lesions were visually observed, quantitative indicators showed
early brain damage, with sensitivity and specificity up to 0.717 and 0.528,
respectively. The difference in T2
quantized value is a new finding, which is different from previous research
results3, 4.
This quantitative index not only
improves the detection of early lesions, but also provides more information for
clinical evaluation of brain damage in the future.
To sum up, the relative ratio of T1 and
T2 in magnetic resonance imaging can be used to quantitatively diagnose brain
damage in neonatal hyperbilirubinemia. This method is especially useful for the
detection of early lesions that are invisible to the naked eye.Acknowledgements
No acknowledgement found.References
1.
Katar S, Akay HO, Taskesen M, et
al.Clinical and cranial magnetic resonance imaging (MRI) findings of 21
patients with serious hyperbilirubinemia.J Child Neurol, 2008, 23 (4) :415-417.
2.
Wu WL, Zhang P, Lou MW, et al.The
application of DWI and MRIin bilirubin encephalopathy and hyperbilirubinemia of
the newborn.JPract Radiol, 2011, 27 (8) :1242-1244.
3.
Ruifang Yan,Dongming Han,Jipeng Ren, et
al. Diagnostic value of conventional MRI combined with DTI for neonatal
hyperbilirubinemia[J]. Pediatrics & Neonatology,2018,59(2):161-167
4.
Xiaoyi Wang,Wulin Wu,Bob L. et al.Studying neonatal bilirubin encephalopathy with
conventional MRI, MRS, and DWI[J]. Neuroradiology,2008,50(10) :885-893