Docstoc

hypothermia protocol - Homepage of the neonatal neuroscience

Document Sample
hypothermia protocol - Homepage of the neonatal neuroscience Powered By Docstoc
					                                                                                     1




 5
     neo.nEuro.network


10
     Study protocol:

     Induced systemic hypothermia in asphyxiated new-born
     infants: a randomized, controlled, multicenter study
15


     Appendix 1:     Procedures and definitions for neurological assessment,
                     aEEG and EEG assessment
20
     Appendix 2:     Scientific basis for inclusion criteria, classification into sub-
                     groups and sample size calculations

     Appendix 3:     Specifications for cooling mattress and aEEG device
25
     Appendix 4:     Monitoring schedule

     Appendix 5:     Case Record Form

30

     Principal Investigator and Organizer:
     Simbruner G, M.D., Professor of Pediatrics
     Division Neonatology, University Childrens´s Clinic,
35   Ludwig-Maximilians University, Munich
     Lindwurmstraße 4, D - 80337 Munich, Germany
     Fax : + 49 89 5160 - 4419 ;
     e-mail > simbruner@kk-i.med.uni-muenchen.de

40
     Status:
     3 May 2000
                                                                                              2

45   Study protocol :
     Induced systemic hypothermia in asphyxiated new-born infants:
     a randomized, controlled, multicenter study.

     ________________________________________________________________
50

     Study Group
     Participants in Germany and other European Countries within the framework of the
     ”neo-nEuro-network” (abbreviated n.n.n), an non-juridal association of neonatologists
55   interested in neurological and neuro-intensive care investigations.


     Prinicipal Investigator:
     Simbruner G, Division Neonatology, Ludwig-Maximilians University, Munich
60   Fax : + 49 89 5160 - 4419 ; e-mail > simbruner@kk-i.med.uni-muenchen.de


     Scientific Committee of the nnn Hypothermia Study:
     Blennow M, Huddige Hospital , Huddige, Sweden,
65   Gaus W, Univ Ulm, Ulm, Germany
     Greisen G, Univ Copenhagen, Copenhagen, Denmark
     Obladen M, Charite, Humboldt Univ, Berlin, Germany
     Pohlandt F, Univ Ulm, Ulm, Germany
     Simbruner G, LMU, Munich,Germany
70   Thoresen M, Univ Bristol, Bristol , UK
     Zupan V, Univ Paris, Paris, France


     Biometry:
75   W. Gaus, Department of Biometry and Medical Documentation
     University of Ulm, Germany
     Fax: +49 731 50-26902; e-mail: wilhelm.gaus@medizin.uni-ulm.de


80   Regional co-ordinators (Status 12 April, 2000):
     Copenhagen, Denmark: Gorm Greisen
     Munchen, Germany: Georg Simbruner
     Paris, France: Veronique Zupan

85
     Acknowledgement :
     We thank the principal investigator Peter D Gluckman, Auckland, New Zealand of the
     Olympic Medical Brain Cooling Trial and Olympic Medical for providing a copy of that
     protocol. Inclusion and exclusion criteria and the primary endpoint is identical in the two
90   protocols. The brain cooling trial uses a cooling cap and a target body temperature of
     34.5C whereas we will use systemic cooling to 33.5C and all patients receive mor-
     phine for pain and sedation. We think that the results from the two trials can add up to
                                                                                                   3

      document a potential benefit of hypothermia as well as suggest a difference between
      the two approaches. I thank Professors Azzopardi D and Edwards DA for also providing
 95   insight into their study protocol.


      Synopsis

100   This randomized, controlled multicenter study aims to determine whether inducing sys-
      temic hypothermia in birth-asphyxiated new-born infants born at term increases the
      chance of survival without severe neurodevelopmental handicap. Secondary objectives
      are to determine whether the treatment benefit is greater in milder rather than more se-
      vere asphyxia, and whether systemic hypothermia is associated with significant side-
105   effects.

      New-born infants with a gestational age of > 36 weeks , asphyxiated at birth, who show
      evidence of encephalopathy and who have abnormal EEG within 6 hrs of postnatal age,
      will be randomized to hypothermia or to normothermia. Hypothermia (target rectal tem-
110   perature 33.5°C) will be induced by a cooling matters, perfused with water. Hypothermia
      will be maintained for 72 hours. Infants in both groups will receive morphine at a seda-
      tive and analgesic dosage.

      Neurological assessment will be done at 7 days, 6 and 18 months. Further, at
115   18 months the Griffith General Quotient will be assessed. The primary outcome is (1)
      death, or (2) inability to sit unsupported, or (3) Griffith GQ < 85, or (43) bilateral cortical
      blindness at 18 months or a combination thereof. The statistical analysis will examine
      whether the hypothermia and control group differ in their proportion of infants surviving
      without disability .
120

      Background

      Ethical considerations.
125   Severe birth asphyxia results in brain damage and death or lifelong handicap. The sci-
      entific community so far failed to provide an effective protective or therapeutic interven-
      tion.
      Controlled mild hypothermia is the best candidate for a post-asphyxia intervention. Hy-
      pothermia has been proven effective in several animal species.
130   Extensive clinical experience with short term, deep hypothermia exists from heart sur-
      gery practice. Some clinical experience exists with long-term moderate hypothermia. Al-
      though hypothermia may have side effects, even if serious, they are likely to be acute,
      whereas the benefits are likely to be long term.
      It is therefore reasonable to proceed to a randomized controlled trial. A trial of selective
135   head cooling is under way in New Zealand, USA, Canada and the UK. The present
      protocol is sufficiently similar and sufficiently different to the Olympic Medical Head
      Cooling Trial to be likely to add significantly to the evidence provided by that study.
      The major ethical concern is the limited possibility to obtain truly informed, free consent
      by the parents. There is no solution to this problem. It will be the duty of the investiga-
                                                                                              4

140   tors to remain sensitive to parental wishes, to obtain the best possible consent, and to
      offer a high level of information to the parents throughout the study.


      Epidemiolgy.
145   The incidence of asphyxiated full-term infants still ranges between 2 and 4 infants per
      1000 deliveries in high income countries (1, 2) and may be significantly higher in low
      income countries. Birth asphyxia results in longterm neurological sequelae in up to 20%
      of infants, if moderate and in up to 100% of infants, if severe. Birth asphyxia causes
      high costs for posthospital care and severe suffering of families (3, 4).
150
      Systemic and selective hypothermia.
      From animal experiments a vast body of knowledge demonstrates that a low intra-
      ischemic brain temperature or lowering brain temperature by 2-4 °C post ischemia may
      protect the asphyxiated brain from neuronal damage and cell death and improve long
155   term neurological outcome (5-7, 47). In factually all animal experiments, hypothermia of
      the brain was induced and associated with hypothermia of the rest of the body. Selec-
      tive brain cooling while maintaining normal rectal temperature was demonstrated to be
      possible in new-born piglets (8), but evidence of long lasting effects or superiority to
      systemic hypothermia is still lacking.
160   In analogy, all human hypothermia studies in adults were based on systemic body
      cooling. Mellegard P. demonstrated that effective selective cooling of the brain was not
      possible in adult neurosurgical patients (9). Simply heat balance calculations or complex
      mathematical model indicate, that selective brain cooling is only possible if one accepts
      extreme temperature gradients across both, body and
165   brain (10).



      EFFECTS
170
      Hypothermic interventions have five domains which appear to determine the balance
      between protective and adverse effects: (1) temperature range of and profile within
      brain tissue, (2) time relation of occurrence of insult to inducing hypothermia, (3) dura-
      tion of cooling, (4) degree of severity of the neuronal injury, and (5) developmental and
175   biological age of the individuum subjected to hypothermia.

      Temperature range.
      Though a controversial issue, most animal studies suggest, that the optimal brain tem-
      perature for neuroprotection ranges between 32 and 34 °C. In some studies mild hypo-
180   thermia (34°C) was found to mitigate neuronal damage to a larger extent than moderate
      (32 - 28°C) or deep hypothermia (28 - 17°C) (11,12). In other studies, mild hypothermia
      was as effective in reducing biochemical and histological markers of neuronal injury and
      in reducing abnormal behaviour (12 - 15).

185   Time relation between insult and hypothermia.
                                                                                               5

      Hypoxic-ischemic injury resulting in encephalopathy evolves in two phases , the primary
      and secundary energy failure. The primary event triggers a series of biochemical and
      immunological responses which are responsible for the pathophysiology of the second
      phase and for the final manifestation of cerebral damage. These biochemical and im-
190   munologic responses include release of cytotoxic neurotransmitters (13), oxygen radi-
      cals (16), inflammatory mediators and signals inducing apoptosis (17). It appears natu-
      ral that an immediate or earlier intervention might be more successful than delayed
      ones. In animal experiments, hypothermia induced at 1 to 5.5 hours were more effective
      compared to later inductions (14,18 -20). The neuroprotective effects diminishes and
195   disappears if cooling is delayed beyond 6 hours (14, 20)

      Duration of hypothermia.
      Duration of cooling clearly relates to short and long term outcome of asphyxia (7, 14,
      20), longer periods being more effective than shorter periods of hypothermia. The dura-
200   tion of hypothermia ranges up to 72 hours in asphyxiated animals, kept at 35°C (20). In
      human trauma patients a hypothermia of about 33°C was applied for 24 hours (21). As-
      phyxiated human new-borns have been kept at a rectal temperature of 35.5 °C for 72
      hours (22) and at a nasopharyngal temperature of 34.5° for an average of 78 hours (23)

205   Severity of the injury and developmental stage of the injured.
      Scarce data in animal and human studies indicate that the neuroprotective impact of
      hypothermia is stronger in mild than severe neuronal injury (18, 21, 24). Few and con-
      troversial studies are available which investigated the effectiveness of hypothermia on
      neuroprotection in dependence of developmental age (25, 26). However, in rats of vari-
210   ous developmental stages, brains of immature rats were found to be less resistant to
      hypoxia (26). The neonatal brain undergoes drastic neuronal remodelling through
      apoptosis before and after birth. Different remodelling and healing capacity in develop-
      ing individuals might be responsible for the fact, that similar therapeutic procedures like
      ECMO generally produce better outcome result in new-born infants than in pediatric or
215   adult patients (see ECMO registry).


      ADVERSE EFFECTS

220   Hypothermia on one hand can be the manifestation of an underlying disease and on the
      other hand the result of an accidental or actively induced heat loss. Hypothermia causes
      a wide spectrum of effects on all organ systems. We suggest to classify these effects
      into (i) physiological changes, which are reversible. Amongst them, low heart and
      breathing rate immediately reverse to normal, when normothermia is reinstituted , (ii)
225   pathologic alterations, which are reversible, probably represent some health risks, but
      have not shown to result in irreversible damage, either without or with treatment .
      Amongst them hypoglycemia, altered coagulation states or low platelets, and (iii) altera-
      tions which result in irreversible damage despite attempts of or due to lack of effective
      treatments, amongst them occurrence of intracranial tissue bleeding or untreatable car-
230   diac arrest due to arrhythmias (although arrhythmias are not reported to occur at mild
      hypothermia). Difficulties in assessing the hypothermic effects arise from possible com-
      binations of causes, their manifestations and their consequences.
                                                                                                 6

      Accidental hypothermia and hypothermia, caused by disease or therapy, is known to
235   affect factually all organ systems (27, 28). Some of these effects are reversed at nor-
      mothermia without causing disease or impairment. Other effects like arrhythmias, co-
      agulation disorders or infections may have long term sequelae (29 - 34). Accidentally
      cooled prematures with a rectal temperature < 34 °C at admission suffered from co-
      agulation disorders, a high incidence of hemorrhage and mortality, whilst another group
240   of newly born prematures with an average rectal temperature of 31 ± 2.7 °C at admis-
      sion exhibited only transient thromobcytopenia and renal failure (31,32).

      In controlled clinical trials on adult trauma or cardiac arrest patients , induced mild to
      moderate hypothermia for up to 24 hours resulted in some beneficial effects (21,35, 36)
245   without any significant, irreversible side-effects. In a study by Metz et al. on ten trauma-
      tized adults, subjected to moderate hypothermia for 24 hours, adverse effects such as
      temporary decrease of cardiac index, thrombocytes and creatinine clearance, and an
      elevated serum lipase activity were reported (37).

250   Whole body cooling had been applied to asphyxiated mature human new-borns after
      delivery by Westin (38) in the sixties. Several trials treating asphyxiated term infants with
      hypothermia followed without reporting details on immediate effects or complications
      (39). Head cooling in 12 asphyxiated new-borns to two temperature levels (six at 36.5-
      36.0°C and six at 35.5 - 35.9°C) was reported not to result in any adverse effects which
255   could be specifically attributed to hypothermia (22). Incidences of adverse side effects
      in asphyxiated new-borns cooled to 34.5° nasopharyngeal temperature for 3 1/2 days
      were not different from those asphyxiated but kept at normal body temperature (23).

      A randomized, controlled study on the effect of mild hypothermia (two arms: control vs.
260   head cooling method) is in process by Gluckman & Wyatt in cooperation with Olympic
      Medical , Seattle, USA with the following known characteristics:
      1) Hypothermia induced by external head cooling (cooling cap Olympic Medical)
      2) Hypothermia maintained for three days
      3) Inclusion of moderate and severe cases of asphyxia (aEEG < 5 uVolt) with an esti-
265        mated chance of death or disability of 80%
      4) Regulatory endpoint Trectal 34. 5± 0.5 °C
      5) Study endpoint : combined rate of mortality and neurodevelopmental
            impairment at 18 month of age.

270
                                                                                               7

      This study will differ from the one of Gluckman & Wyatt in the following respects:
      1) Systemic cooling without any efforts to achieve a temperature difference between
           head and rest-of-the body by cooling the head and to add radiative heat to the
           rest-of-the body.
275   2) Regulatory end-point is T rectal 33-34°C, a mean of 33.5 °C
      3) Routine analgesia and sedation in both groups with morphine.
      4) Prospectively planned assessment of treatment effect in mild to moderate vs.
             severe asphyxia.
      5) Option of cooling before transport
280   6) Trectum measured at 2-3 cm from anus, thus allowing measurements with
           ubiquitous available rectal thermometers.


      Study objectives
285
      The objectives of this randomized, controlled, multicenter study are 1) to determine
      whether inducing systemic hypothermia in birth asphyxiated new-born infants at term
      decreases the proportion of infants who die or survive with severe neurodevelopmental
      handicap when compared to new-born infants whose body temperature is maintained
290   within the normal range, 2) to determine whether the protective effect of hypothermia is
      related to the severity of asphyxia and 3) to evaluate the safety of hypothermia.

      Hypotheses

295   Primary: Hypothermia at 33-34°C, induced by a cooled mattress for three days in se-
      verely asphyxiated new-born infants, reduces the risk of death or severe neurodevelop-
      mental handicap at 18 - 21 months.

      Secondary (a): Hypothermia reduces neurodevelopmental retardation (measured by
300   Griffith GQ) at 18 - 21 months to a significant larger extent in the group with moderately
      abnormal EEG compared to the group with severely abnormal EEG.

      Secondary (b): Systemic hypothermia in the range 33-34°C for 72 hours is safe.

305

      Study design

      1.   Randomized, controlled, multicenter           trial   with   blinded   assessment   of
310        neurodevelopmental outcome.
      2.   The protocol and data acquisition is designed to provide a common minimal
           dataset to test the primary and two secondary hypotheses by participation of
           regional study groups. The common statistical analysis will be carried out by Prof.
           W. Gaus, Department of Biometry and Medical Documentation, University of Ulm,
315        Germany.
      3.   Statistical analysis will be based on intention-to-treat.
                                                                                               8

      4.    Safety and interim analyses at 1/4, 1/2 and 3/4 of patient recruitment in a fixed
            sample design. A safety monitoring committee will be headed by Prof. F Pohlandt,
            Dept of Paediatrics, University of Ulm, Germany. Severe adverse events, likely to
320         be related to hypothermia must be reported to the Department of Biometry and
            Medical Documentation, University of Ulm, Germany, by fax on the same day.
      5.    Regional co-ordinators are responsible for data quality. As a minimum, all entries
            in patient record forms must be dated and signed, and all participating centers
            must be visited by a regional monitor once per year when a sample of case record
325         forms must be checked against patient charts. This quality control must be
            documented.
      6.    Inclusion of new patients must be reported to the regional co-ordinator by fax on
            the same day.
      7.    Individual centres are encouraged to add on studies provided this does not
330         interfere with the objectives of this protocol.
      8.    Centres involved in testing of new, unproven and unpublished neuroprotective
            strategies for asphyxiated infants cannot participate in the study.
      9.    Publication will be in the name of ‘neo-nEuro-network’ with a detailed
            acknowledgement of all contributions.
335   10. Publication of add-on studies can be done as relevant with the mention of the par-
          ticipation in the neo-nEuro-network. Statistics which can be used to test the
          hypotheses of this protocol must be avoided.


340   PATIENT SELECTION
      6.1 Inclusion Criteria

      The infant will be assessed sequentially by criteria A, B and C listed below:

      A. Evidence of severe birth asphyxia in infants >36 weeks gestation admitted to the
         NICU: ONE of the following
345           Apgar score of < 5 at 10 minutes after birth
              Continued need for resuscitation, including endotracheal or mask ventilation, at
                10 minutes after birth
              Acidosis defined as either umbilical cord pH or any arterial pH within 60 minutes
                of birth <7.00
350           Base Deficit  16 mmol/L in umbilical cord blood sample or any blood sample
                within 60 minutes of birth (arterial or venous blood)
      B. Evidence of encephalopathy (assessed by certified study personnel): Moderate to
         severe encephalopathy consisting of altered state of consciousness: lethargy,
         stupor or coma, and at least one or more hypotonia, abnormal reflexes including
355      oculomotor or pupillary abnormalities, an absent or weak suck or clinical seizures.
      C. Evidence of moderate-to-severe neurophysiological dysfunction: At least 30 minutes
         duration of amplitude integrated EEG or standard EEG recording that shows abnor-
                                                                                               9

           mal background EEG activity or seizures (see definitions : Appendix 1). The aEEG
           or EEG may be performed from one hour of age. aEEG or EEG should be read by
360        certified study personnel. Classification of the aEEG is according to al Naqeeb et al.
           (42). The classification of the EEG is according to Lamblin et al (45). The aEEG or
           EEG 30 min following IV anticonvulsant therapy, e.g. phenobarbitone should not be
           used for classification.

      Exclusion Criteria

365      Infants expected to be >5.5 hours of age at the time of randomization.
         Prophylactic administration of high dose anticonvulsants (e.g. >20mg/kg phenobar-
           bitone). After trial entry phenobarbitone or other anticonvulsant therapy may be
           given as clinically indicated to treat seizures (see co-treatment below).
         Major congenital abnormalities, such as diaphragmatic hernia requiring ventilation,
370        or congenital abnormalities suggestive of chromosomal anomaly or other syn-
           dromes that include brain dysgenesis.
         Imperforate anus (since this would prevent rectal temperature recordings).
         Evidence of head trauma or skull fracture causing major intracranial hemorrhage.
         Infants <1,800 g birth weight.
375      Head circumference <(mean2SD) for gestation if birth weight and length are
           >(mean2SD).
         Infants ”in extremis” (those infants for whom no other additional intensive manage-
           ment will be offered in the judgement of the attending neonatologist). Record in de-
           tail reason for exclusion.
380      Overt bleeding
      If all inclusion and no exclusion criteria are met, the infants are eligible for randomiza-
      tion.



385
      STUDY PROCEDURES

      Recording and Classification of aEEG or standard EEG (Appendix 1)

390   The aEEG will be recorded with a Lectromed Cerebral Function Monitor and the use of
      needle electrodes inserted at an exact distance of 5 cm apart from each other at the
      parietal regions. The recording speed is 1mm/min. The classification of the aEEG is
      based on the publication of Niran al Naqeeb et al. (ref 42 : Pediatrics 1999; 103; 1263-
      1271) which yields two subgroups: Moderately abnormal aEEG designates mildly to
395   moderately asphyxiated new-borns and suppressed aEEG severely asphyxiated new-
      borns. This classification is used for subgroup analysis.
                                                                                                10

      The standard EEG is recorded according to the International 10-20 Classification. The
      classification of the EEG is based on the publication of Lamblin MD et al. (ref 45: Neu-
400   rophysiol Clin 1999; 29: 123 - 219)


      Consent

405   Randomisation is done after informed consent from at least one parent. Consent may
      be written or verbal and documented by a wittness.

      Entry criteria and other parts of the minimal data set is entered in the Case Record
      Form.
410

      Temperature control

      All infants are nursed in an open care unit (cot, radiant heater unit, bed etc). If relevant,
415   randomisation may be done in a local hospital and treatment started and continued
      during transfer to a study center, provided temperature control and documentation is
      adequate.

      Body temperature is measured rectally, at least 2 cm from the anus, preferentially with a
420   permanent probe .

      Control Group

      Infants randomized to the control group are treated according to present standards of
425   postnatal care with the aim of achieving a normal body temperature i.e. T rectum of 37°C
      (range: 36.5 to 37.5°C) . All handling and treatments otherwise should be identical to
      the one of the hypothermia group.

      Hypothermia Group
430
      Infants randomised to the hypothermia group will be nursed naked on a cooling mat-
      tress (cooling device and matrass see: Appendix 3) which is perfused by circulating
      water at a variable temperature and covered by normal linen. The infant will have a dia-
      per around the lower half of the abdomen. The rectal temperature is targeted at 33.5 °C
435   (range of 33 to 34 °C). The duration of hypothermia is 72 hours.

      Induction of hypothermia

      Target temperature should be achieved within 60 min. The water temperature should
440   initially be set to 10°C. The body temperature should be checked every 10 min. If the
      temperature drop is insufficient the cooling mattress should be wrapped around the in-
      fant or cool packs be applied.

      Maintaining hypothermia
445
                                                                                             11

      Normally a mattress temperature of 32-33°C will maintain the temperature within range.
      The temperature must be recorded every hour. Be prepared to reduce matters tem-
      perature in case of seizures, to increase it at initiation of mechanical ventilation etc.

450   Rewarming

      Rewarming is started by stopping the cooling of the mattress. Check the temperature
      every hour. The infant should reach a normal rectal temperature in 6 -12 hours after
      stopping cooling. If the infant does not warm up spontaneously, heat may be added
455   from a radiant heater. Check the temperature more frequently. If the temperature rises
      more than 0.5°C per hour, the cooling mattress should be used to control the rise.

      After rewarming

460   Temperature is managed according to routine care. Beware of rebound hypothermia.


      Co-treatments

465   1.   All infants in the hypothermia and control group will receive morphine at 0.1 mg/kg
           every 4 hours during the first three days as brain edema might and as cold does
           induce discomfort, restlessness, anxiety and irritation (References: in Physiological
           and Behavioral Temperature Regulation. Chapter 57. Eds. Hardy JD, Gagge AP,
           Stolwijk JAJ, 1970 Charles C Thomas Publisher) and pain (Reference: Maria Fitz-
470        gerald, London, at 11 Anual ESIC Congress, Stockholm, 1998). Hypothermia in-
           creases stress hormones (48).
      2.   In ventilated infants PCO2 should be maintained between 40 and 60 mmHg. In
           spontaneous breathing infants hyperventilation will not be treated, but hypoventi-
           lation with PCO2 > 60 mmHg will be treated.
475   3.   All other treatments will be given according to clinical routines

      Discontinuing hypothermia before 72 hours

      1.   Parents withdraw consent.
480   2.   The attending neonatologist decides, if possible after consultation with the regional
           co-ordinator (Record reason for withdrawal). Potential reasons might include for
           example bleeding, thrombosis, pulmonary hypertension or arrhythmia which is
           difficult to control, or continued inability to maintain rectal temperature in the
           desired range.
485   3.   Need for ECMO.
      If hypothermia is discontinued prior to 72 hours, rewarming will be done according to
      clinical routine. The follow-up procedures will be done unless the parents withdraw their
      permission for this.
      Improvement on continuous aEEG or standard EEG recording, after trial entry, is not an
490   indication for discontinuing treatment.
                                                                                                12



      Follow-up

      7 ± 1 days postnatal : Neurological assessment by the Thompson score
495   (ref. 41) and brain ultrasound.

      6 months: Bodyweight, length, and head circumference and gross neurological as-
      sessment (blinded assessment by certified study personnel).

500   12 months: Parental questionnaire.

      18 month: Neurological examination and determination of Griffith General Quotient
      (GQ) (blinded assessment by certified study personnel). Persistent neurological signs
      consistent with a central motor deficit are classified according to five level classification
505   of Palisano et al (40).



      Primary Outcome Measure
510
      Survival at 18–21 months of postnatal age free of severe neurodevelopmental handi-
      cap.

      Severe neurodevelopmental handicap is defined as one or more of
515   Neurological deficit with a functional score of 3-5 as defined by Palisano et al. (40),
        i.e. sitting only with low back support or worse.
      Griffith General Quotient of less than 85 (i.e. < 2 SD)
      Severe bilateral cortical visual deficit (no reaction to a threat to the eye, which has a
        clear anterior chamber and normal fundi)
520
      Secondary outcome measures

      1. Thompson Score (ref. 41) at day 7
      2. Head Circumference at 6 months
525
                                                                                            13

      Outcome measure for Subgroup Analysis

      Subgroup analysis of infants with (1) moderately abnormal and (2) suppressed aEEG in
530   order to determine which of these subgroups is more responsive to hypothermia treat-
      ment. Griffith General Quotient will be used as outcome measure .




535   ADVERSE EFFECTS

      Definitions

      Potential adverse effects to hypothermia may also be consequences of asphyxia, and
      thus statistical testing is essential to determine whether cooling may be a contributing
      factor. Severe adverse events that in particular may be due to hypothermia will be im-
540   mediately reported to the Data Safety Monitoring Committee (DSMC) . They are:

       Cardiac arrhythmia
       Severe hypotension despite full inotrope support and volume correction
       Major venous thrombosis not related to an infusion line
545    Severe bleeding, e.g. from the lungs or in the brain


      Other complications that may occur during the first 7 days of life may include:

         Death
         Electrocardiographic evidence of cardiac arrhythmias or myocardial ischemia or hy-
550        potension <40mm Hg
         Coagulopathy (clinical bleeding with abnormal clotting studies, consistent with dis-
           seminated intravascular coagulation or hepatic coagulopathy)
         Abnormal renal function (urine output <0.5 ml /kg/hour for >24 hours after birth,
           maximum serum creatinine (>0.09 mM).
555      Hyponatremia (<130 mM), hypokalemia (<3.5 mM).
         Bone marrow depression (platelet count <100,000 per mm 3)
         Elevated liver enzyme levels (AST >200 IU, ALT >100 IU)
         Metabolic acidosis (BE <-5) following after entry into study
         Need for mechanical ventilation
560      Need for nitric oxide
         Need for ECMO
         Systemic infection (blood, CSF or urine cultures)
                                                                                              14

         Hemoconcentration (increase of hematocrit by 20% or more) not associated with
           transfusions
565      Hypoglycemia (<2.6 mmol/L)
         Hypocalcemia (<2 mmol/L) adjusted for albumin levels, or <1.0 mmol/L on ionised
           calcium measurement.

      Adverse events related directly to cooling device:
          Difficulties in temperature control
570


      STATISTICAL ANALYSIS

575   Randomization

      Randomization will be stratified for hospitals as well as for the severity of asphyxia
      (moderate/severe). Before the start of the trial, randomization plans will be established
      in the Department of Biometry and Medical Documentation of the University of Ulm,
580   Germany, for all hospital. The resulting group assignment for each consecutive patient
      will be concealed in sealed envelopes. The envelopes are identified by

      Title of study : Induced systemic hypothermia...
      Name of the hospital
585   moderate / severe asphyxia
      A patient number.

      Each participating hospital will receive two piles of envelops, one pile for moderate as-
      phyxia and one pile for severe asphyxia.
590
      Envelopes have to be used by increasing patient number, i.e. the envelope with the
      lowest patient number first.

      The envelopes will contain a self adhesive label with title of study, name of the hospital,
595   the degree of asphyxia (moderate or severe), the patient number and finally the treat-
      ment (hypothermia or control). This label has to be placed on the case report form on
      admission of a patient to the study. Then, a filled out form with the label has to be
      transmitted by fax to the regional co-ordinator within 24 hours.

600
                                                                                               15

      Statistical Evaluation

      The primary null-hypothesis is: ”The proportion of surviving infants without severe neu-
605   rodevelopmental handicap at the age of 18-21 months is not reduced by hypothermia”.
      The test is one-sided since it is not relevant to confirm statistically whether hypothermia
      increases the proportions of death or handicap.
      The statistical evaluation will be performed on all admitted patients (full sample analysis,
      intent to treat analysis). The result will be summarized by a 2 x 2 contingency table (out-
610   come by group); Fisher’s exact test (level of significance: 5%) will be used to compare
      the treatment groups. This test is confirmative.

      The statistical evaluation of the data will include all necessary descriptive measures.
      The Griffith General Quotient will be compared using a Wilcoxon test for two parallel
615   groups. All further secondary outcome measures will be evaluated using the appropriate
      statistical procedures. These tests are one-sided, too, with a level of significance of 5%.
      The levels of significance will not be adjusted for multiple testing. Hence, these tests are
      only intended for exploration. At regular intervals, interim analyses will be carried out
      (see below), their maximum number is 3.
620

      Sample size (see APPENDIX 2)

      The sample size of 2 x 75 patients has been determined under the following assump-
625   tions: 60% of the patients suffer from moderate asphyxia, 40% from severe asphyxia.
      Further, 75% of the patients with moderate asphyxia and 95$ of the patients with severe
      asphyxia will be dead or handicapped after 18-21 months in the control group. If 40% of
      these cases can be prevented by hypothermia under moderate asphyxia and 10% un-
      der severe asphyxia, the overall proportion of dead or handicapped will be 83% in the
630   control group and 61% in the hypothermia group. These assumptions lead to a sample
      size of 61 patients per group when testing a one-sided hypothesis at a level of 5% with
      a power of 80% under the given treatment effects. It is realistic to assume that 20% of
      infants will be lost to follow up. Thus 75 patients will have to be recruited in each group.

635

      Data Monitoring and Interim Analysis

      DATA COLLECTION & MONITORING
640
      Data collection and monitoring of study centres is done by the regional co-ordinators.
      Data will be passed on to the Department of Biometry and Medical Documentation, Uni-
      versity of Ulm, for analysis. Safety and interim analyses will be done when 1/4, 1/2 and
      3/4 of the patients are recruited. Severe adverse effects will be reported on the same
645   day to Ulm. Safety data and Interim analysis will be reviewed by the Safety and Data
      Monitoring Committee (Pohlandt F, Ulm and Versmold H, Berlin and N.N).

      Regional co-ordinators will meet every 6 months to maintain consistency of procedures
      when documentation of site visits and sample chart reviews by regional co-ordinators
650   will be reviewed by the principal investigator.
                                                                                                16




      INTERIM ANALYSIS
655
      The estimated mortality in the control group is 20% to 30% (assuming a mortality in
      moderate asphyxia of about 5% and in severe asphyxia of 60%, further a case mix of
      60% moderate and 40% severe asphyxia).The majority of deaths after severe birth as-
      phyxia are a consequence of brain damage, and occur within the first week of life. This
660   means that the study hypothesis, assuming a reduction in adverse outcome (death and
      handicap) of about 30%, includes an expectation of reducing death rate from 20 to
      15%; this is a difference of 5% death before 7 days. The interim analyses will be aimed
      at detecting strong evidence against this. Furthermore, the adverse effects of cooling
      are recorded within 7 days of life. Safety analysis will compare adverse effects in the
665   hypothermia and control groups.


      Interim analysis will be done when 40, 75 and 115 of the patients in each group have
      been recruited and 7 day data is available.
670
      If the interim analyses show that it is highly unlikely (p < 0.0025 in the first analysis,
      p < 0.005 in the following two interim analyses) that the mortality will be reduced by an
      absolute difference 5% or if safety analysis shows that it is highly likely (p-values as
      above) that hypothermia is associated with increased number of severe adverse effects,
675   premature stopping of the trial will be considered by the Data Safety Monitoring Com-
      mittee.

      If the trial is stopped prematurely the patients already included in the study will be fol-
      lowed up as scheduled by this study protocol.
680

      Final Analysis

      Final analysis will be based on intent-to-treat and carried out after follow-up is closed for
685   all patients on study.

      The analysis will include
       Source data listing
       An elementary descriptive analysis of all variables observed
690    Assessment of data quality achieved
       Description of study patients
       Assessment of drop-outs and comparability of groups
       Evaluation of efficacy including scheduled subgroups analysis
       Evaluation on adverse events, tolerability and risks of hypothermia
695    Explorative analysis if further interesting results are supposed
       Biometrical assessment on validity of study results
                                                                                           17

      In case there will be a serious proportion of drop-outs the most important parts of the
      analysis have to be done for all admitted patients (full sample analysis) as well as for
700   those patients valid for efficacy.


      Financial Support

705   G. Simbruner will file an application at the German Federation for Research (Deutsche
      Forschungsgemeinschaft) to finance insurance, 1/2 post for a statistician, 1 post for
      data monitoring doctor, and up to 10 Lectromed Ceerebral Function Monitors to in-
      crease the number of participants in Germany. Other regional co-ordinators will have to
      obtain their own funds to fulfil their roles.
710


      LITERATURE
      1    Finer NN, Robertson CM, Richards RT, Pinnell LE, Peters KL. Hypoxic-ischemic
715        encephalopahty in term neonates: Perinatal factors and outcome. J Pediatr 1981;
           98: 112-117

      2     Thornberg E. Birth asphyxia: incidence, clinical course and outcome in a Swe-
            dish population. Acta Paediatr 1995; 84:927-932
720
      3     Robertson CM, Finer NN,Grace MG. School performance of survivors of neonatal
            encephalopathy associated with birth asphyxia at term. J Pediatr 1989; 114: 753-
            60

725   4     Shankaran S, Woldt E, Koepke T, Bedard MP, Nadyal R. Acute neonatal morbid-
            ity and long-term central nervous system sequelae of perinatal asphyxia in term
            infants. Early Hum Dev 1991; 25: 136-148

      5     Busto R, Dietrich WD, Globus MYT, Valdes I, Scheinberg P, Ginsberg MD
730         Small differences in intra-ischemic brain temperature critically determine the ex-
            tent of ischemic neuronal injury. J Cerebr Blood Flow Metab 1987; 7:
            729 - 38

      6     Dietrich WD. The Importance of Brain Temperature in Cerebral Injury.
735         J Neurotrauma 1992; 9: Suppl 2: S-475- 485

      7     Thoreson M, Wyatt J. Keeping a cool head, post-hypoxic hypothermia - an old
            idea revisited. Acta Paediatr 1997; 86: 1029-33

740   8     Gelman B, Schleien CL, Lohe A, Kuluz JW. Selective brain cooling in infant pig-
            lets after cardiac arrest and resuscitation. Crit Care Med 1996; 24: 1009-1017

      9     Mellegard P. Changes in human intracerebral temperature in response to differ-
            ent methods of brain cooling. Neurosurgery 1992; 31: 671-677
745
                                                                                          18

      10   Simbruner G.Thermodynamic models for diagnosic purposes in the new-born and
           fetus; Facultas Verlag, Wien; 1983, ISBN 3-85076-133-9

      11   Weinrauch V, Safar P,Tisherman S, Kuboyama K, Radovsky A. Beneficial effects
750        of mild hypothermia and detrimental effect of deep hypothermia after cardiac ar-
           rest in dogs. Stroke 1992; 23: 1454-1462

      12   Chopp M, Knight R, Tidwell CD, Helpern JA, Brown E Welch KMA. The metabolic
           effects of mild hypothermia on global cerebral ischemia and recirculation in the
755        cat: comparison to normothermia and hyperthermia. J Cereb Blood Flow Metabol
           1989; 9: 141148

      13   Busto R, Globus MYT, Dietrich D, Martinez E, Valdes I,Ginsberg MD. Effect of
           mild hypothermia on ischemia-induced release of neurotransmitters and free fatty
760        acids in rat brain. Stroke 1989; 20: 904-910

      14   Coimbra C, Wieloch T. Moderate hypothermia mitigates neuronal damage in the
           rat brain when initiated several hours following transient cerebral ischemia. Acta
           Neuropathol 1994; 87: 325-331
765
      15   Clifton GL, Jiang JY, Lyeth BG, Jenkins LW, Hamm RJ, Hayes RL. Marked pro-
           tection by moderate hypothermia after experimental traumatic brain injury. J
           Cereb Blood Flow Metab 1991; 1:114-121

770   16   Armstead WM, Mirro R, Busija DW, Leffler CW. Post-ischemic generation of su-
           peroxide anion by newborn pig brain. Am J Physiol 1988; 255: H401-403

      17   Edwards AD, Yue X, Cox P, Hope PL, Azzopardi D, Squier MV, Mehmet H.
           Apoptosis in the brains of infants suffering intrauterine cerebral injury. Pediatr
775        Res 1997; 42(5) 684-689

      18   Colbourne F, Corbett D. Delayed and prolonged post-ischemic hypothermia in
           neuroprotective in the gerbil. Brain Research 1994; 654: 265-272

780   19   Gunn AJ, Gunn TR, Gunning MI, Williams CE, Gluckman PD. Neuroprotection
           with prolonged head cooling started before post-ischemic seizures in fetal
           sheep.Pediatrics 1998; 102 (5): 1098-1106
                                                                                           19

785   20   Sirimane ES, Blumberg RM, Bossano E, Gunning M, Edwards AD, Gluckman
           PD,Williams CE. The effect of prolonged modification of cerebral temperature on
           outcome after hypoxic-ischemic brain injury in the infant rat. Pediatr Res 1996; 9:
           591-597

790   21   Marion-DW; Penrod-LE; Kelsey-SF; Obrist-WD; Kochanek-PM; Palmer-AM; Wis-
           niewski-SR; DeKosky-ST. Treatment of traumatic brain injury with moderate hy-
           pothermia. N Engl J Med 1997; 336: 540-546

      22   Gunn AJ, Gluckman PD, Gunn TR. Selective head cooling in newborn infants
795        following perinatal asphyxia; a safety study. Pediatrics 1998; 102 : 885 - 892

      23   Simbruner G, Haberl C, Harrison V, Linley L. Induced brain hypothermia in as-
           phyxiated human newborn infants: a retrospective chart analysis of physiological
           and adverse effects . Int Care Med 1999; 25:1111-1117
800
      24   Chopp M, Chen H, Dereski MO, Garcia JH. Mild hypothermic intervention after
           graded ischemic stress in rats. Stroke 1991; 22: 37-43

      25   Berger R, Jensen A, Hossmann KA, Paschen W. Effect of mild hypothermia dur-
805        ing and after transient in vitro ischemia on metabolic disturbances in hippocampal
           slices at different stages of development. Dev Brain Res 1998; 105: 67-77

      26   Yager JY, Shuaib A, Thornhill J.The effect of age on susceptibility to brain dam-
           age in a model of global hemispheric hypoxia-ischmia. Dev Brain Res 1996; 93:
810        143-154

      27   Cornell HM. Accidental hypothermia. J Pediatr 1992; 120: 671-79. 28

      28   Ilievich UM, Spiss CK. Hypothermic therapy for the injured brain. Current Opin-
815        ions in Anesthesiology 1994; 7:394-400

      29   Okada M. The cardiac rhythm in accidental hypothermia. J Electrocardiol 1984;
           17: 123-128

820   30   Mann TP, Elliot RIK. Neonatal cold injury due to accidental exposure to cold.
           Lancet 1957; 1, 229

      31   Chadd Ma, Gray OP. Hypothermia and coagulation defects in the newborn. Arch
           Dis Child 1972: 41: 819-821
825
      32   Kaplan M, Eidelman AI. Improved prognosis in severely hypothermic newborn
           infants treated by rapid rewarming. J Pediatr 1984; 105: 470-74

      33   Staab BD, Sorensen VJ, Fath JJ, Raman SBK, Horst HM, Obeid FN. Coagulation
830        defects resulting from ambient temperature-induced hypothermia. J Trauma
           1994; 36: 634-8
                                                                                          20

      34    El-Radhi AS, Jawad MH, Ibrahim M, Jamil II. Infection in neonatal hypothermia.
            Arch Dis Child 1983; 58 (2): 143-145
835
      35    Shiozaki T, Sugimoto H, Tenada M, Yoshida H, Iwai A, Yoshioka T, Sugimoto T.
            Effect of mild hypothermia on uncontrollable intracranial hypertension after se-
            vere head injury. J Neurosurger 1993; 79: 363-368

840   36    Bernard SA, Jones BM,Horne MK. Clincal trial of induced hypothermia in coma-
            tose survivers of out-of-hospital cardiac arrest. Ann Emerg Med 1997; 30: 146-53

      37    Metz C, Holzschuh M, Bein T, Woertgen C, Frey A, Frey I, Taeger K, Brawanski
            A. Moderate hypothermia in patients with severe head injury: cerebral and extra-
845         cerebral effects. J Neurosurgery 1996; 85: 533-541

      38    Westin B, Nyberg R, Miller JA, Wedenberg E. Hypothermia and transfusion with
            oxygenated blood in the treatment of asphyxia neonatorum. Acta paediatr Scand
            1962; 51; Suppl 139: 1-80
850
      39    Miller JA, Miller FS, Westin B. Hypothermia in the treatment of asphyxia neona-
            torum. Biol Neonate 1964; 6: 148-163

      40    Palisano R, Rosenbaum P, Walter S, Russell D, Wood E, Galuppi B.
855         Development and reliability of a system to classify gross motor function in chil-
            dren with cerebral palsy. Dev Med Child Neurology 1997; 39: 214-223

      41    Thompson CM, Puterman AS, Linley LL, Hann FM, van der Elst CW, Molteno
            CD, Malan AF. The value of a scoring system for hypoxic ischaemic encepha-
860         lopathy in predicting neurodevelopmental outcome. Acta Paediatr 1997; 86: 757-
            61


      42    al Naqeeb N, Edwards AD, Cowan FM, Azzopardi D. Assessment of Neonatal
865         Encephalopathy by Amplitude-intergrated Electroencephalography. Pediatrics
            1999; 103; 1263-1271

      43     Hellström-Westas L, Rosén I, Svenningsen NW (1995) Predictive value of
             early continuous amplitude integrated EEG recordings on outcome after severe
870   birth asphyxia in full term infants. Arch Dis Child 72: F34-F38

      44    Toet MC, Eken P, Groenendaal F, de Vries LS Comparison of amplitude
            integrated EEG in birth asphyxiated term neonates between 3 and 6 hours
            after birth (Abstr 1902, Neurology; Ped Res 1998; 43, Part 2 of 2)
875
                                                                                       21

      45   Lamblin MD, Andre M, Challamel MJ, Curzi-Dascalova L, dÁllest AM, De Gio-
           vanni E , Moussalli-Salefranque F, Navelet Y, Plouin P, Radvanyi-Bouvet MF,
           Samson-Dollfus D, Vecchierini-Blineau MF. Electroencephalographie du nou-
           veau-ne premature at a term. Aspects maturativs et glossaire. Neurophysiol Clin
880        1999; 29: 123 - 219

      46   Thoresen M and Wyatt J. Review article . Acta Paediatr 1997; 86: 1029-33
           Keeping a cool head, post-hypoxic hypothrmia-an old idea revistied
           (References 29- 34)
885
      47   Thoresen M, Cooling the newborn aftrer asphyxia - physiological and
           experimental background and its clinical use. Semin Neonatol 2000; 5:61-73

      48   Thoresen M, Satas S, Loberg EM, Lindgren C, Acolet D, Steen PA, Haug E,
890        Whitelaw A. Pediatr Res 1999; 45: 46 A




895
                                                                                            22


      APPENDIX 1

      Procedures and definitions for neurological assessment, aEEG and EEG
900   assessment


      Assessment of Neurological Signs

905   Neurological signs are observed and scored according to Thompson et al (41).The
      score consists of a clinical assessment of nine signs. Each sign is scored from 0 to 3
      and the score for each day is totalled. The higher the score the more severely affected
      the infant. The maximum possible score on any day is 22. The score is equally applica-
      ble to ventilated infants . It cannot be applied to paralyzed infants.
910
                                            Score
      Sign           0              1                  2                     3

      Tone         normal       hyper               hypo                   flaccid
915   LOC          normal       hyperalert, stare   lethargic              comatouse
      Fits         none         < 3 per day         > 2 per day
      Posture      normal       fisting, cylcing    strong distal flexion decerebrate
      Moro         normal       partial             absent
      Grasp        normal       poor                absent
920   Suck         normal       poor                absent ± bites
      Respir       normal       hyperventilation    brief apnea           IPPV (apnea)
      Fontanell    normal       full, not tense     tense


925   Recording of aEEG and Standard EEG

      The aEEG will be recorded with a Lectromed Cerebral Function Monitor and the use of
      needle electrodes inserted at an exact distance of 5 cm apart from each other at the
      temples. Different distances result in different voltage (Personal communication: Hell-
930   ström-Westas L, Blennow M). The recording speed is 1mm/min. At least one half hour
      recording is required, latest between 5.0 and 5.5 hours of age. A 30 min decrease of
      lower voltage margine below 5uV must not be associated with sleep cylces, manipula-
      tions or drug administration.

935   The Standard EEG is recorded according to the International 10-20 classification.


      Classification of aEEG for Inclusion Criteria

940   The classification of the aEEG is based on the publication of Niran al Naqeeb et al.
      (ref 42 : Pediatrics 1999; 103; 1263-1271). It yields three subgroups: normal aEEG,
      moderately abnormal aEEG and suppressed aEEG. Moderately abnormal aEEG desig-
      nates mildly to moderately asphyxiated new-born and suppressed aEEE severely as-
      phyxiated new-borns. This classification is used for subgroup analysis. Patients with the
                                                                                          23

945   unlikely aEEG combination of upper margin < 10 uV and lower margin >5uV will remain
      unclassified and excluded from the study.


      Classification of standard EEG for Inclusion criteria
950
      The classification of the standard EEG is based on the publication of Lamblin MD et al.
      (ref 45: Neurophysiol Clin 1999; 29: 123 - 219)

      Definition of normal EEG:
955   Continuous , no intervals > 3 sec and no amplitudes < 10 uV

      Definition of abnormal EEG:
      Continous low voltage < 25 uV or discontinuous (burst supression)
      see exact definition 4.1 , 4.2 and 4.4 , 4.5 in article of Lamblin et al.
960
      Sampling rate: Every 5 min a time sample of 20 sec long tracing



965
                                                                                              24


       APPENDIX 2
 970
       Scientific basis for inclusion criteria, classification into subgroups and
       sample size calculations.

       This study includes all asphyxiated new-borns whose aEEG or standard EEG is not
 975   normal at any time within the first six hours of age. Infants with a normal aEEG or Stan-
       dard EEG, not included into the study, may later turn out to have an neurodevelopmen-
       tal impairment. The false negative rate (normal EEG, but abnormal outcome) might
       range from 4% to 15% (42 - 44). The study population will contain mild to moderate en-
       cephalopathy and severe encephalopathy. These two subgroups will be classified be-
 980   fore the intervention and within the first 6 hours of age according to their aEEG or stan-
       dard EEG (see APPENDIX 1) .


       Event rate
 985
       The event rate is defined as death or neurodevelopmental impairment of survivors. De-
       pending on the severity, asphyxia causes a certain rate of neurodevelopmental impair-
       ments. The table shows the event rates of ”death and neurodevelopmental impairment
       ”in dependence of the initial severity classified according to aEEG into the two sub-
 990   groups. Estimated from data from Hellström-Westas et al, Toet et al and AlNaqeeb et
       al. (42 - 44).


       Correlation of ampliutde-integrated EEG to neurological outcome
 995
       Severity            Author       Nr     Event rate (%)             Event rate (%)
                                               Disabilty and Death        No Disability

       Normal              al Naqueeb 9         0     (0%)                 9     (100%)
1000                       Hellström  26        1     (4%)                25     (96%)
                           Toet       26        4     (15%)               22     (85%)

       Moderate            al Naqeeb    17     13     (75%)                 4    (25%)
       abnormal            Hellström    14     11     (77%)                 3    (23%)
1005   aEEG                Toet         13     10     (77%)                 3    (23%)



       Suppressed          al Naqeeb    12     12     (100%)                0    (0 %)
1010   aEEG                Hellström     5      5     (100%)                0    (0%)
                           Toet          5      5     (100%)                0    (0%)
                                                                                                25

       Summarized Results of aEEG Studies
1015
                                  No disability          Disability         Nr.
                                                         and death
       Normal aEEG                      56                    5              61
       Abnormal aEEG                    10                  56               66
1020                                    66                  61              239

       Sensitivity 56/61 = 92 % ; Specificity 56/66 = 85%


1025
       Correlation of standard EEG to neurological outcome
       (analyzed according to criteria of Lamblin MD et al., to be published d ´Allest AM et al.)

                           No disability           Disability               Nr.
1030                                              and Death

       Normal EEG          128    (96%)            5     (4%)         133 (55.6 %)

       Mod abn EEG           17 (54%)             14     (46%)         31 (13.0 %)
1035
       Severe abn EEG          2 (3%)             73     (97%)         75 (31.4 %)

                            147                   30                  239

1040
       Summarized Results of Standard EEG

                                  No disability          Disability          Nr.
                                                         and death
1045
       Normal EEG                 128                     5                 133
       Abnormal EEG                19                    87                 106
                                  147                    92                 239

1050   Sensitivity 87/92 = 95 % ; Specificity 128/147 = 87 %


       These data on aEEG and standard EEG on neurodevelopmental outcome demonstrate:
       (i) that aEEG and standard EEG have a similarly high predicitve power (92% vs 95%
1055   sensitivity and 85% vs 87% specificity). Consequently both methods are valid for the
       inclusion of asphyxiated infants.
       (ii) that the ratio of moderate to severe aEEG findings is 66 % and 33% respectively in
       the compiled aEEG studies (ref 42-44) and the ratio of moderate to severe in the stan-
       dard EEG Data (see above) is 30% vs 70%.
                                                                                           26

1060   Consequently an overall ratio of 60% moderate and 40% severe asphyxia was assumed
       for sample size calculation and
       iii) because of the lack of congruence of subgroups defined by aEEG and standard
       EEG, subgroup analysis will be done only on patients with aEEG inclusion criteria.

1065
       Treatment effect

       In six new-born animal studies the neuroprotection (expressed as reduction in % cell
       damage or abnormal neurobehavior) of mild to moderate hypothermia starting within
1070   30 min after the ischemic-hypoxic insult varied between 25 to 80% (the neuroprotective
       effect was > 50%, when those studies where hypothermia was induced only for <3 hrs,
       were excluded (46)). No data for the treatment effect of asphyxiated human new-borns
       are available. The study population will contain infants with mild to moderate encepha-
       lopathy. In animal experiments hypothermia has been shown to be more effective in
1075   mild to moderate asphyxia than in severe to fatal asphyxia (Chopp M et al. Mild hypo-
       thermic intervention after graded ischemic stress in rats. Stroke 1991; 22: 37-43).
       Gluckman et al in the Olympic Medical Protocol assumed a 30 % reduction of neurode-
       velopmental impairment and death (an event rate of 70% in the control and 49% in the
       intervention group) in a group of mixed (moderately and severely) asphyxiated new-
1080   borns. We thus assumed a 40% reduction (less than the 50% in animal experiments,
       but more than the 30% for the mixed group) for the mildly to moderately and 10% (less
       than the mixed group) for the severely asphyxiated group.


1085
                                                                                          27


       APPENDIX 3

       Specifications for cooling mattress and aEEG device

1090
       Cerebral Function Monitor (CFM)

       The CFM to be used for the study is the one from Lectromed or proven equivalent.

1095   Specification:
       LECTROMED Cerebral Function Monitor
       Lectromed MT2-5330 CFM System

       Representative in Germany:
1100   FBI Fred Berninger Importe OHG
       Bergstraße 12,
       D-82024 Taufkirchen, Germany
       Tel.: + 49 89 61 453 453 or +49 161 2832755
       Fax.: + 49 89 61 453 453
1105   e-mail > BERNIMPORT@AOL.com


       Cooling Device

1110   The cooling device to be used for the study is the one of TEC.COM GMBH. It consists
       of an apparatus which provides a perfusion flow of a temperature regulated fluid and a
       mattress. A simpler version Tecovit 200 for NICU use and transport, with battery 12 V
       powering, is available.

1115   Specification:
       Cooling device Tecotherm TS med 200 (about 3200 Euro) , size 420 x 190 x 350 mm,
       ca 10 kg; Temperatue range : 5 - 42°C ; hand controlled
       OR
       Tecovit 200 (about 1750 Euro)
1120   plus CSZ Cooling matrass, rectangular, cleanable surface, ca.80 x 60 cm (about
       90 Euro)

       TEC.COM GMBH Prof Dr K Berndt                  e-mail >berndt-halle@t-online.de<
       Product Manager                       OR >SCHOEPKE:teccom@t-online.de
1125   Böllberger Weg 170
       D - 06128 Halle/Saale, Germany
       Tel.: + 49 (0) 345 – 120 52 04
       Fax.: + 49 (0) 345 – 120 52 11

1130
                                                                                                     28


       APPENDIX 4

       Monitoring Schedule

1135   Time                     adm before 3             12       24      48       72       80 hrs
       Data Sets

       CerFunction aEEG         x----- continous------x           x       x        x cont x

1140   Rect Temp ( Thermom )    x        x -----------------hourly-----------------------   x
       Cooling Fluid Temp x     x -----------------hourly----------------------- x
       Environm Temp            x        x        x        x     x        x        x        x

       Blood Pressure           x     x -----------------hourly-----------------------      x
1145   Heart rate               x     x -----------------hourly-----------------------      x
       Breathing rate           x     x -----------------hourly-----------------------      x
       SpO2                     x     x -----------------hourly-----------------------      x
       PtcCO2                   x     x -----------------hourly-----------------------      x
       Respir Variables         x     x        x        x     x        x        x           x
1150   Urine output (3hrs)          x x        x        x     x        x        x           x
       Blood Samples
              blood gases       x       x        x       x        x       x        x        x
              glucose           x       x        x       x        x       x        x        x

1155
       Clinical Examination     x       x                x        x       x        x        x
       Neurol Examination x     x                x       x        x       x        x

       Ultasound Imaging                x
1160
       Organ Failure                    x                         x       x        x
       Blood samples
             electrolytes               x                         x       x        x
             hemoglobin                 x                         x       x        x
1165         coagulation                x                         x       x        x
             infection screen           x                         x       x        x
             kidney function            x                         x       x        x

              amylse                    x                                          x
1170          endocrine                 x                                          x
              lactate                   x                                          x

       Liquor
       lumbal                                            x - between - x
1175   Blood S-100, NSE                                  x - between - x

       Time                     adm before 3             12       24      48       72       80 hrs
                           29


       Appendix 5
1180

       Case Report Forms

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:295
posted:6/3/2010
language:English
pages:29