Adapted barkovich scoring system (adbarko) in asphyxiated full-term infants

Aims: To develop an extended asphyxia-score based on ultrasound and MR-scan in order to gain further insight into the pathophysiology and to refine individual prognostication.

Methods: Full-term infants born between January 2001 and February 2005, fullfilling criteria of Levene for asphyxia. Serial sonograms and MR-scan (T1, T2 and DWI) on day 4 or 5 post partum were used for designing a score of injury concerning the basal ganglia/thalamus (Bg/Thal) and (sub)cortex/white matter (Sc/Wm). Involvement of caudate and equivocal inversion of signal intensity in posterior limb of the internal capsule (PLIC) were added. This score was compared with the Barkovich score (Barko) from which ours was derived. Bad outcome (cerebral palsy (CP) or death) was related to the scores.

Results: 60 MR/ultrasound scans were scored. Whether or not neonates developed CP was examined at the minimum age of 18 months. 4 scans were normal as were the neonates. In 29 (48%) the scores differed from Barko. Bg/Thal were bilaterally affected in 39: 3 developed CP, 27 died . In 13, only Bg/Thal were affected (CP1,died5). A combination with Sc/Wm-injury was seen in 26: watershed injury (WI) in 5 (CP0,died2), extensive cortical lesions (ECL) in 20 (CP1,died18) or punctate white matter haemorrhage (PWMH) in 1. In 17 only the Sc/Wm was affected: WI in 5 (CP1,died0), ECL in 3 (CP0,died1) or PWMH in 4 (CP1,died1). In 5 isolated diffuse leukomalacia was seen (CP1,died1). There was a significant difference in involvement of caudate in severe (17/28) and mild (8/32) Sc/Wm-injury (p=0,005, Fisher's-exact-test).

Conclusion: The adBarko includes new patterns. Among others isolated diffuse leukomalacia is seen with a pathophysiological mechanism of axonopathy rather than selective-neuronal-necrosis. There is a significant difference in involment of caudate in severe and mild Sc/Wm-injury. At this moment relation to outcome is difficult (many died or age under 18 months yet).