Radiology: MRI-detected pediatric microhemorrhages predict prognosis
The presence of intraparenchymal brain microhemorrhages in children with nonaccidental trauma as detected on susceptibility-weighted (SW) MR images correlates with significantly poor long-term neurologic outcome, according to a study published in the September issue of Radiology.
Cherie A. Colbert, MD, from the department of radiology at Loma Linda University Children’s Hospital in California, and colleagues conducted the study to determine the prevalence of parenchymal brain microhemorrhages, as identified by SW MRI and compared with other radiologic findings, in infants with nonaccidental, or abuse-related trauma. The authors sought to verify if identification of microhemorrhages can better predict long-term neurologic outcome.
The researchers retrospectively reviewed the data for 101 children between the ages of one to 32 months (62 males, average age of 8.4 months; and 39 females, average age, 7.4 months) with forensic pediatric specialist–confirmed nonaccidental trauma at the children’s hospital from 2001 to 2007. According to the authors, finalized reports from head CT and head MRI studies were reviewed by a radiology resident with five years of experience and a neuroradiologist with 33 years of experience.
Colbert and colleagues excluded any children with pre-existing cognitive delays, central nervous system malformations, previous brain injuries and/or birth before 30 weeks gestation.
Normal, mild disability, or moderate disability neurological outcome data versus severe disability, vegetative state, or death at greater than or equal to six months were available for 53 patients (36 males, average age, 7.3 months; and 17 females, average age, 7.4 months). Logistic regression was used to determine whether the presence of SW imaging–depicted microhemorrhages, as compared with other radiologic findings, resulted in improved prediction of long-term neurologic outcome, the researchers explained.
Of the 101 patients, SW MRI showed that 29 percent of the patients had microhemorrhages, 65 percent had extraaxial hemorrhages, 51 percent had retinal hemorrhages and 35 percent had evidence of acute ischemic injury. “We found the presence of ischemic injury, as determined with SW MRI, to be a strong predictive variable, however, this was not a hypothesis of this study,” the authors wrote.
A significantly larger number of children with poor outcomes than children with good outcomes presented with brain microhemorrhages, (nine of 14, compared to seven of 39) and ischemic injury (13 of 14, compared to 17 of 39), determined Colbert and colleagues. Logistic regression analysis revealed presence of microhemorrhages via SW MR imaging—followed by acute ischemic injury, initial Glasgow Coma Scale score and age—to be the most significant single variable in the final model, with an overall predictive accuracy of 92.5 percent, explained the authors.
“Our findings suggest that the presence of microhemorrhage lesions and acute ischemic injury combined reflects brain injury that contributes more to long-term disability than do other radiologic or clinical findings commonly seen after nonaccidental trauma,” offered Colbert.
The researchers believe the presence of microhemorrhage lesions, particularly in the presence of ischemic injury, correlated with poor long-term neurologic outcomes because microhemorrhages represent foci of axonal injury due to a combination of traumatic and vascular injuries. Ischemic injury occurred in 57 percent of patients with nonaccidental trauma.
“By itself, this finding contributed significantly to outcome prediction, and when combined with the presence of microhemorrhages, it yielded the best overall predictive accuracy,” the authors said. “The patients with both acute ischemic injury and microhemorrhage were 33 times more likely to have a poor outcome.”
Cherie A. Colbert, MD, from the department of radiology at Loma Linda University Children’s Hospital in California, and colleagues conducted the study to determine the prevalence of parenchymal brain microhemorrhages, as identified by SW MRI and compared with other radiologic findings, in infants with nonaccidental, or abuse-related trauma. The authors sought to verify if identification of microhemorrhages can better predict long-term neurologic outcome.
The researchers retrospectively reviewed the data for 101 children between the ages of one to 32 months (62 males, average age of 8.4 months; and 39 females, average age, 7.4 months) with forensic pediatric specialist–confirmed nonaccidental trauma at the children’s hospital from 2001 to 2007. According to the authors, finalized reports from head CT and head MRI studies were reviewed by a radiology resident with five years of experience and a neuroradiologist with 33 years of experience.
Colbert and colleagues excluded any children with pre-existing cognitive delays, central nervous system malformations, previous brain injuries and/or birth before 30 weeks gestation.
Normal, mild disability, or moderate disability neurological outcome data versus severe disability, vegetative state, or death at greater than or equal to six months were available for 53 patients (36 males, average age, 7.3 months; and 17 females, average age, 7.4 months). Logistic regression was used to determine whether the presence of SW imaging–depicted microhemorrhages, as compared with other radiologic findings, resulted in improved prediction of long-term neurologic outcome, the researchers explained.
Of the 101 patients, SW MRI showed that 29 percent of the patients had microhemorrhages, 65 percent had extraaxial hemorrhages, 51 percent had retinal hemorrhages and 35 percent had evidence of acute ischemic injury. “We found the presence of ischemic injury, as determined with SW MRI, to be a strong predictive variable, however, this was not a hypothesis of this study,” the authors wrote.
A significantly larger number of children with poor outcomes than children with good outcomes presented with brain microhemorrhages, (nine of 14, compared to seven of 39) and ischemic injury (13 of 14, compared to 17 of 39), determined Colbert and colleagues. Logistic regression analysis revealed presence of microhemorrhages via SW MR imaging—followed by acute ischemic injury, initial Glasgow Coma Scale score and age—to be the most significant single variable in the final model, with an overall predictive accuracy of 92.5 percent, explained the authors.
“Our findings suggest that the presence of microhemorrhage lesions and acute ischemic injury combined reflects brain injury that contributes more to long-term disability than do other radiologic or clinical findings commonly seen after nonaccidental trauma,” offered Colbert.
The researchers believe the presence of microhemorrhage lesions, particularly in the presence of ischemic injury, correlated with poor long-term neurologic outcomes because microhemorrhages represent foci of axonal injury due to a combination of traumatic and vascular injuries. Ischemic injury occurred in 57 percent of patients with nonaccidental trauma.
“By itself, this finding contributed significantly to outcome prediction, and when combined with the presence of microhemorrhages, it yielded the best overall predictive accuracy,” the authors said. “The patients with both acute ischemic injury and microhemorrhage were 33 times more likely to have a poor outcome.”