Relation between proton magnetic resonance spectroscopy within 18 hours of birth asphyxia and neurodevelopment at 1 year of age J Donncha Hanrahan MRCPI, Research Fellow, Department of Paediatrics; I Jane Cox PhD, Lecturer, Robert Steiner Magnetic Resonance Unit; Denis Azzopardi FRCP, Senior Lecturer in Paediatrics; Frances M Cowan MRCP, Senior Lecturer in Neonatal Neurology; Department of Paediatrics; Janet Sargentoni DCR(R)MU, Senior Radiographer; Jimmy D Bell PhD, Senior Lecturer; David J Bryant PhD, Senior Lecturer; Robert Steiner Magnetic Resonance Unit; A David Edwards* FRCP, Weston Professor of Neonatology, Department of Paediatrics; Imperial College School of Medicine, Hammersmith Hospital, London, UK. *Correspondence to final author at Department of Paediatrics, Imperial College School of Medicine, Hammersmith Hospital, DuCane Rd, London W12 ONN. E-mail:
[email protected]
The aim of the study was to test the hypotheses that elevated cerebral lactate, detected by proton spectroscopy performed within 18 hours of suspected birth asphyxia, is associated with adverse outcome, and that increased lactate can be used to predict adverse outcome. Thirty-one term infants suspected of having had birth asphyxia and seven control infants underwent proton magnetic resonance spectroscopy, using three-dimensional chemical shift imaging, within 18 hours of birth. Adverse outcome was defined as death or neurodevelopmental impairment at 1 year of age or more. Nine infants had an adverse outcome. The other 22 and all of the control infants remained normal. Median (range) lactate/creatine plus phosphocreatine (lactate/creatine) ratios in the abnormal, the normal, and the control group were 1.14 (0.17 to 3.81), 0.33 (0 to 1.51), and 0.05 (0 to 0.6) respectively (P=0.003). Lactate/creatine >1.0 predicted neurodevelopmental impairment at 1 year of age with sensitivity of 66% and specificity of 95%, positive and negative predictive values of 86% and 88%, and a likelihood ratio of 13.2. Elevated cerebral lactate/creatine within 18 hours of birth asphyxia predicts adverse outcome.
76 Developmental Medicine & Child Neurology 1999, 41: 76–82
Hypoxic–ischaemic injury in the perinatal period remains a significant cause of paediatric mortality and morbidity. Studies using proton magnetic resonance spectroscopy (1H MRS) have demonstrated a relation between altered cerebral metabolism and neurodevelopmental outcome following insult. A fall in the ratio of N-acetylaspartate, a putative neuronal marker, to choline (Peden et al. 1990; Groenendaal et al. 1994) and to creatine (Peden et al. 1993, Amess et al. 1995) has been shown to indicate adverse prognosis. Increased cerebral lactate concentration measured by 1H MRS within days of injury (Groenendaal et al. 1994) has also been shown to suggest poor outcome. In these studies however, MRS data were not collected until days and weeks after birth. We have previously demonstrated (Hanrahan et al. 1996) a relation between the concentration of lactate in the brain, detected by 1H MRS in the first 18 hours of life, and the ratio between phosphocreatine and inorganic phosphate, which is a strong predictor of neurodevelopmental impairment (Azzopardi et al. 1989, Roth et al. 1992, Martin et al. 1996), detected by 31P MRS at 2 to 4 days of age. These results suggested the hypotheses that elevated cerebral lactate/creatine, detected by proton spectroscopy performed within 18 hours of birth asphyxia, is associated with adverse outcome, and that increased lactate/creatine can be used to predict adverse outcome in infants suspected of birth asphyxia. Method Permission for this study was granted by the Research Ethics Committee of the Royal Postgraduate Medical School of Hammersmith Hospital (No 93/4047). Parental consent was obtained in all cases. PATIENTS
Thirty-one term infants with a median gestational age of 40 weeks (range 37 to 42 weeks), a median birthweight of 3280 g (range 2140 to 4800 g), and suspected of having had intrapartum hypoxic–ischaemic injury were studied by 1H MRS at a median age of 13 hours (range 4 to 18 hours). They were selected because of clinical suspicion of birth asphyxia, as indicated by the following: a history of fetal distress (late decelerations on cardiotocography, meconium stained liquor, or acidaemia [i.e. pH12 µmol/L] in umbilical cord blood or arterial blood within 30 minutes of birth); the necessity for resuscitation at birth; abnormal postnatal neurological signs consistent with hypoxic–ischaemic encephalopathy graded by the method of Sarnat and Sarnat (1976). Clinical details of these infants’ perinatal periods are shown in Table I. EEG was performed in 28 of the infants within the first 72 hours of life and all 31 infants underwent cerebral ultrasound. Seven term infants with a median gestational age of 39 weeks (range 38 to 41 weeks) and a median birthweight of 3380 g (range 2740g to 3540 g), and with no history of birth asphyxia were also studied for comparison at a median age of 8 hours (range 5 to 16 hours). Infants in the asphyxia group who did not die in the neonatal period and all control infants were assessed neurodevelopmentally at the age of at least 1 year. NEURODEVELOPMENTAL ASSESSMENT
Neurodevelopmental assessment was conducted by workers not involved in spectral analysis. The neurodevelopmental
examination consisted of three parts: (1) The Griffiths’ Developmental Scale (Griffiths 1954). This gave a developmental quotient, determined by dividing the overall developmental age by the chronological age, with the results displayed as a percentage. The overall developmental age was calculated by the average of the developmental ages of each section of the examination, i.e. locomotor, eye/hand, personal/social, hearing/ speech, and performance. (2) Head circumference growth velocity was measured. (3) An Infant Neurological Examination, based on a previously published proforma (Kuenzle et al. 1994) was performed which requires the infant to be examined in the following categories: (i) Posture. The posture of the head, trunk, arms, legs, hands, feet, and degree of asymmetry is assessed. (ii) Tone. This is examined with respect to head control, being pulled to sit, ventral suspension, ‘scarf sign’, passive shoulder elevation, hip adduction, popliteal angle, ankle dorsiflexion, and degree of asymmetry. (iii) Spontaneous movement. This is concerned with the sym-
metry, quantity and quality of movement, with the ability to kick when both supine and in vertical suspension, rolling, movement when prone or held sitting, standing and walking. (iv) Elicited motility. This relates to the smoothness and accuracy of movement when reaching for a ring, a cube, or a box. (v) Reflexes. This section is concerned both with protective reflexes (arm protection, lateral tilt, and parachute) and deep tendon reflexes. (vi) Eye movements. These are deemed normal or abnormal, depending on their smoothness, accuracy, and the presence or absence of strabismus or nystagmus. (vii) Interaction. This section assesses the infant with respect to smiling, vocalizing, babbling, words, vision, and hearing. Each aspect of a category is given a grading of 1 to 3, 1 being significantly abnormal and 3 being normal. Categories are graded for age, and only the number of sections that pertain to the relevant age group are examined. The mean of the scores of each section examined in a category gives the score for that category. The overall ‘optimality score’ results from
Table I: Perinatal clinical details of asphyxiated infants studied within 18 hours of birth Nr
Length of gestation (wk)
1 2a 3 4 5a 6a 7 8 9 10a 11a 12a 13 14 15a 16 17a 18 19a 20a 21a 22a 23 24 25 26 27a 28a 29a 30a 31a
41 38 42 41 42 40 38 39 38 41 41 40 38 41 39 37 40 37 41 39 41 42 40 41 42 39 42 39 41 39 39
Birth- Occipito-frontal weight head (g) circumference at birth (cm) 3545 3060 3520 2750 3715 3640 3020 3435 2370 3305 2860 2900 3440 2750 3280 2270 2535 2580 3340 2550 4110 2860 4040 4800 4000 2520 3300 3190 3990 2140 3550
34.6 35 37.5 32.3 35.5 35.3 35 35 33 35 33.5 34 36 33 36.2 33.4 34 32 34 33 36 34 36.3 36 36.2 33.9 34 34.8 35.8 35 34.5
Fetal distress
Delivery
Mode of resuscitation
pH of cord blood
Base excess of cord blood (µmol/L)
Ctg Ctg Ctg Ctg/Mec Ctg/Mec Ctg/Mec Ctg Ctg Ctg Ctg/Mec None None Ctg Ctg/Mec None Ctg Ctg None Ctg Ctg Mec None None Ctg/Mec Ctg/Mec Ctg/Mec Ctg Ctg/Mec Ctg/Mec Ctg/Mec Ctg
Normal Normal Em C/S Normal Normal Em C/S Breech Forceps Em C/S Forceps Normal Normal Forceps Em C/S Normal Normal Em C/S Normal Ventouse Forceps Ventouse Home Dystocia Em C/S Em C/S Normal Em C/S Em C/S Ventouse Em C/S Ventouse
Bag/mask ET ET ET, CM, Ad Bag/mask ET ET Bag/mask ET Bag/mask ET Bag/mask ET ET ET ET, Ad ET Bag/mask ET, CM, Ad Bag/mask Bag/mask Bag/mask ET, CM, Ad ET, CM, Ad ET, CM, Ad ET ET, CM, Ad ET, CM, Ad ET ET, CM, Ad ET, CM, Ad
6.92c 7.18c 6.80c 6.90b 7.25b 7.06c 7.14b 6.96c 7.10b 6.87c 6.97c 7.24c 6.80c 6.89c 7.13c 7.09c 6.70c 6.98c 7.00c 7.10c 7.15b NRc NRc 7.00c 6.98b 6.80c 6.90c NRc NRc 7.10c 7.08c
–12c NRc –20c –17b –14b –12c –20b –14c –18b NRc –13c –9c –17c –22c –21c –14c –29c –18c –16c –14c –15b NRc NRc –12c –24b –25b –17c NRc NRc –25b NR c
Apgar score Grade after of hypoxic– 1 min (5 min) ischaemic encephalopathyc 5 (7) 5 (10) 1 (4) 1 (2) 2 (6) 6 (9) 3 (6) 3 (6) 0 (4) 3 (7) 2 (7) 2 (4) 4 (7) 3 (6) 2 (3) 1 (4) 0 (5) 5 (6) 1 (7) 4 (7) 2 (6) 3 (4) 0 (0) 0 (1) 2 (5) 2 (4) 1 (3) 1 (3) 0 (4) 1 (3) 3 (2)
1 1 2 1 1 1 1 1 1 2 1 1 2 1 1 1 1 1 1 1 1 1 3 3 3 2 3 2 – 3 3
a infants whose MRS data has already been published. b measurements of arterial blood within 1 hour after birth (cord blood not recorded). C graded by Sarnat and Sarnat (1976). Mec, meconium stained liquor; Ctg, fetal distress seen on late cardiotocography; Em C/S, emergency caesarean section; ET, endotracheal intubation; Ad, adrenaline; CM, cardiac massage; Home, home delivery; Dystocia, shoulder dystocia; NR, not recorded; –, no data.
Cerebral Lactate and Outcome After Asphyxia J D Hanrahan et al. 77
the sum of the mean scores for each of the seven categories. As each category gives a score of 1 to 3, this can range from 7 (extremely abnormal) to 21 (completely normal). OUTCOME GROUPS
Infants with adverse outcome were those who either died or had abnormal neurodevelopment. Infants were deemed to have a normal neurodevelopment if they achieved a Griffiths’ quotient of ≥ 85% and an optimality score of ≥ 20. They were considered to have an abnormal development if they had a Griffiths’ quotient of
