Parenchymatous Brain Injury in Premature Infants: Intraventricular Hemorrhage and Periventricular Leukomalacia

Prematurity is a condition associated with high mortality and overall survived rates are near 77.5% (Stoll et al., 2010).Those who survived are at high risk of severe impairment (Bassler et al., 2009). Two percent of all live births are premature with less than 32 weeks of gestational age and 1.5% of them are very low birth weight (Mathews et al., 2011). The most common injury affecting brain of these children is periventricular leukomalacia (PVL) and intraventricular hemorrhage (IVH). PVL is the main cause of cognitive behavioural, motor and sensory impairments found in children born before 32 weeks of gestational age (Volpe, 2003). IVH has a negative impact on the neurodevelopmental outcome and is due not only to the direct consequences of IVH but also associated lesions, such as posthemorrhagic hydrocephalus (PHH) and PVL. The knowledge, prevention, diagnosis and early treatment of these clinical conditions improve the prognosis and neurological outcomes.


Introduction
Prematurity is a condition associated with high mortality and overall survived rates are near 77.5% (Stoll et al., 2010).Those who survived are at high risk of severe impairment (Bassler et al., 2009).Two percent of all live births are premature with less than 32 weeks of gestational age and 1.5% of them are very low birth weight (Mathews et al., 2011).The most common injury affecting brain of these children is periventricular leukomalacia (PVL) and intraventricular hemorrhage (IVH).PVL is the main cause of cognitive behavioural, motor and sensory impairments found in children born before 32 weeks of gestational age (Volpe, 2003).IVH has a negative impact on the neurodevelopmental outcome and is due not only to the direct consequences of IVH but also associated lesions, such as posthemorrhagic hydrocephalus (PHH) and PVL.The knowledge, prevention, diagnosis and early treatment of these clinical conditions improve the prognosis and neurological outcomes.

Brain injury in the premature infant 2.1 Intraventricular hemorrhage
In the late 1970s the incidence of IVH was near 50%.At the present time it is near 20 to 30%.The absolute number of infants with IVH remains significant due to increase survival rate of premature infants especially in very low birth weight, who are at high risk of IVH.Virtually all IVH in premature infants occurs within the first five days of life, with 50, 25 and 15 percent on the first, second and third day respectively, and 10 percent on the fourth day or after.IVH progresses over three to five days in approximately 20 to 40 percent of cases (Volpe, 2008;Groenendaal et al., 2010).PHH, periventricular hemorrhagic infarction (PVHI), and PVL are the most important sequelae of IVH.The first occurs in approximately 25% of infants with IVH and usually it begins within one to three weeks after the brain bleeding (Murphy et al., 2002).PVHI pathogenesis is thought to result from infarction caused by venous obstruction after a germinal matrix IVH (Bassan, 2009;Volpe, 1998).The parietal and frontal cerebral areas are the most often involved (Bassan et al., 2006a).PVL is the mayor form of brain white matter in neonates especially in premature infants.There is a strong association between PVL and IVH and data suggest the IVH may exacerbate PDL (Bassan, 2009).The short term outcome is closely related to the severity of IVH (Kusters et al., 2009).The long term outcome of infants, who survive of IVH, worsens with increasing severity of IVH and decreasing gestational age (Sherlock et al., 2005;Luu et al., 2009).

Risk factors and pathogenesis
IVH originates from the fragile involuting vessels of subependymal germinal matrix, located in the caudothalamic groove.The pathogenesis of this in preterm infants has been demonstrated to be related to numerous risk factors which can be divided into intravascular, probably the most important and amenable to preventive efforts, vascular and extravascular factors (Table 1).Intravascular factors are ischemia and reperfusion, like in volume infusion after hypotension, fluctuating cerebral blood flow (CBF), like in mechanical ventilation, increase in CBF, like in hypertension, anemia and hipercarbia, increase in cerebral venous pressure and platelet dysfunction and coagulation disturbances.Vascular factors consider tenuous and involuting capillaries with large diameter lumen.Extravascular factors are deficient vascular support and excessive fibrinolytic activity (Perlman et al., 1983;Lou, 1988;Pryds et al., 1989).For developing PVHI the risk factors are low birth gestational age, low Apgar scores, early life acidosis, patent ductus arteriosus, pneumothorax, pulmonary hemorrhage and needed for significant respiratory or blood pressure support (Bassan et al., 2006b).PHH likely relates at least in part to impaired cerebral spinal fluid resorption or obstruction of the acueduct or the foramina of Luschka or Magendie by particulate clot (Larroche, 1972).

Classification -Severity and grading of IVH
There are two main systems for grading the severity of IVH based on the amount of blood in the germinal matrix and lateral ventricles demonstrated by ultrasound or computed tomography (CT) scan (Table 2).By US, Volpe grades the IVH in: I, for germinal matrix hemorrhage (GMH) with no or minimal IVH, with less than 10% of ventricular volume occupying, II for IVH occupying 10%-50% of ventricular area on parasagittal view, III for IVH occupying more than 50% of ventricular area on parasagittal view, usually distends lateral ventricle.Separate notation: persistent periventricular echodensity (Volpe, 2008).Using CT scan Papile grading system is: I for isolated germinal matrix hemorrhage II for IVH without ventricular dilatation, III for IVH with ventricular dilatation and IV for IVH with parenchymal hemorrhage (Papile et al., 1978)

Clinical presentation and diagnosis
Clinical presentation varies from a silent syndrome, recognized only when a routine cranial US is performed, a mild form with decreased levels of consciousness , hypotonia, abnormal eye movements or skew deviation.A catastrophic way with rapid and severe neurological deterioration, with seizures, tonic posturing and coma is more rarely (Soul, 2008).
Whereas any cerebral imaging test is useful, it is almost invariably made by portable cranial US.The diagnosis of HIV and PHH is easy and performing fast, quick to make.Premature infants are fragile patients carry them to make CT scan or magnetic resonance adds more costs, time and risk, and there are hospitals where this exams aren´t available.

Prevention and treatment
The most effective strategy to prevent IVH is the prevention of preterm birth.When it cannot be avoided, the following prenatal and delivery interventions are associated with a reduced risk of IVH.Antenatal corticosteroids, reducing de risk of IVH detected by cranial US examination (OR 0.29, CI 95% 0.14 to 0.61) (Crowley, 2000).Delayed clamping of the umbilical cord more than 30 seconds demonstrated a lower relative risk of IVH versus early clamping (RR 0.56, CI 95% 0.36 to 0.93) (Rabe et al., 2004).Mothers who are at risk of preterm delivery must be transferred to a perinatal center with experience in high risk deliveries and care of prematures.Infants who are transferred after delivery are at higher risk for developing IVH.Inborn patients compared to those who are transported had a lower incidence of IVH (13.2 versus 27.4) and a lower relative distribution of severe IVH (39.2 versus 44.1 percent) (Mohamed & Aly, 2010).Delivery mode does not appear to affect the risk of severe IVH (Riskin et al., 2008) whereas the presence or absence of labor in cesarean delivery the data are conflicting.In the neonatal care units the efforts must be put on: prompt and appropriate resuscitation of the neonate, avoiding hemodynamic instability, hypoxia, hyperoxia, hypercarbia and hypocarbia.All this factors affects the cerebrovascular autoregulation; avoid hypotension and hypertension and hemodynamic instability must be care avoiding large bolus infusions.Metabolic abnormalities such hyperosmolarity, hyperglicemia and hypoglicemia should be prevented.Abnormalities in coagulation should be corrected (Bada et al., 1990;Dani et al., 2009;Perry et al., 1990).
The incidence of IVH is higher in preterm infants with patent ductus arteriosus, they should be treating (Jim et al., 2005).Ineffective interventions for prevention IVH includes: antenatal administration of phenobarbital (Shankaran et al., 1997), antenatal administration of magnesium sulfate and vitamin K to the mother has no benefit in prevent IVH (Crowther et al., 2010;Volpe, 2008).Postnatal indomethacin and vitamin E are associated a lower risk of IVH, but the first increases other risks like gastrointestinal and renal negative effects and the other with an increased risk of sepsis, so it's use is controversial (Fowlie et al., 2010;Brion et al., 2003).
The treatment doesn´t includes a specific therapy, it is supportive and the main goal is to preserve more perfectly cerebral perfusion.Treatment and early detection of complications like seizures and PHH, with serial cranial US, will improve the outcome, minimizing further brain injury.

Evolution and prognosis
IVH is still an important cause of injury in premature infants.The negative impact of IVH on neurodevelopmental outcome is due not only to direct consequences of IVH because it is also associated to other lesions like PHH or PVL.Long term prognosis for infants with IVH varies considerably depending on the severity of IVH, complications or other brain lesions such PVL, the most lower birth weight and gestational age add to others significant illness will determinate the outcome.Studies have suggested that preterm infants with grade I-II IVH have an increased risk of cerebral palsy and cognitive impairment compared who those without (Sherlock et al., 2005;Ancel et al., 2006;Patra et al., 2006).Infants with the mayor complications, like PVHI and PHH are at much higher risk of permanent neurologic impairments like cerebral palsy than dose with IVH alone (de Vries et al., 1999).More than 50% of children born before 32 weeks gestational age have school difficulties whether or not they had IVH, although the risk is clearly higher among children and adolescents with a story of IVH and lower birth gestational age or weight (Bowen et al., 2002;van de Bor & den Ouden, 2004).These cognitive or behavioural handicaps are related in part to white matter brain injury.The most effective strategy to prevent IVH is the prevention of preterm birth.

Periventricular leukomalacia
PVL refers to damage of cerebral white matter brain injury.The name is based on the characteristic distribution and consists of periventricular focal necrosis with subsequent cystic formation and more diffuse cerebral white matter injury (Volpe, 2008).
PVL is the mayor form of brain white matter injury that affects premature infants and it is associated with subsequent development of cerebral palsy, intellectual impairment and visual disturbances.The great risk for developing PVL is under 32 weeks of gestational age.The incidence of PVL varies among centers and in relation with imaging testing realized.Based on US, frequency of PVL ranges from 5 to 15% in VLBW infants (Stevenson et al., 1998).Using MRI, white matter abnormalities are found in 21%, and are associated with adverse neurodevelopment outcomes at a corrected age of 2 years.Gray matter abnormalities are present in half of infants, and are also significantly associated, but less strongly, with cognitive delay motor and cerebral palsy (Woodward et al., 2006).PVL is still the principal cause of this neurodevelopmental impairment (Volpe, 2003).Intraventricular Hemorrhage and Periventricular Leukomalacia 91

Risk factors and pathogenesis
PVL, lesion found predominantly in preterm infants, can be caused by ischemia or infection.
The distinctive lesion of PVL found in the immature white matter newborns likely results from the interaction of multiple pathogenic factors (Table 3).To date, the several mayor factors identified are: vascular anatomic factors, pressure-passive cerebral circulation, intrinsic vulnerability of cerebral white matter of the premature neonate and infection/inflammation.The anatomy of the developing cerebral vasculature renders makes the premature infant especially vulnerable to periventricular white matter injury.The area near ventricles result in a boundary zone, with incomplete development of vessels that penetrates the deep and subcortical white matter, therefore this area is vulnerable to reduced flow (Takashima & Tanaka, 1978;Rorke, 1992).The circulatory factors and metabolic disturbances can impair autoregulation, resulting in higher risk for developing PVL, like HIV.In cystic PVL necrotic changes usually affects all cell components.Diffuse non cystic PVL affects predominantly a specific cell lineage, that of oligodendrocytes (Oka et al., 1993).Another risk factor is oxidative stress attributed to cerebral ischemia and reperfusion and or maternal infection (Khwaja & Volpe, 2008).Expression of inducible nitric oxide synthase is increased in brains with PVL (Haynes et al., 2009).Axonal maturation studies suggest that axons may be particularly susceptible to damage at time in development that coincides with the highest risk of PVL (Haynes et al., 2005).Antenatal risk factors like infection increase the risk of premature birth.In a meta-analysis, chorioamnionitis was associated with cystic PVL (RR 3.0) and cerebral palsy (RR 1.9) (Pidcock et al., 1990).Cytokines produced as a consequence of maternal or fetal infection, even the infection is asymptomatic may be associated with PVL, because the white matter is especially susceptible to damage mediated by this inflammation factor (Dammann & Leviton, 1997).

Classification -Severity and grading PVL
Using cranial ultrasound PVL can be classified according to Volpe or de Vries (Volpe, 1990;de Vries et al., 1992).Volpe categorizes PVL in mild, with micro cysts smaller than 0.2 mm in specially in parasagittal view, moderate with cysts between 0.2 to 0.5 mm, and severe when exists multiple cyst bilaterally bigger than 0.5 mm.According to de Vries, PVL can be classified from I to IV grades (Table 4).The existence of echogenic lesions without cysts formation for more than 14 days is considered as persistent periventricular echogenicity (Larroque et al., 2003).

Reference
For RMI the grading of PVL is more descriptive.Several studies have shown that MRI is more sensitive than cranial US for detection of PVL especially for non cystic form of PVL (Maalouf et al., 2001;Roelants-van Rijn et al., 2001).

Diagnosis
PVL is detected in newborns by brain imaging using US or CT or RMI.Ultrasound is the initial standard method because is portable and less expensive.The criteria for US diagnosis are not well defined.A standard examination includes coronal and sagital view (Veyrac et al., 2006).The US findings evolve on repeated examinations.The cysts appear after one or three weeks and disappear after one or three m o n t h s i f t h e y a r e m o d e r a t e o r s e v e r e ventriculomegaly may results (Blankenberg et al., 1997).US has limited sensitivity and specificity to detect PVL specially if the lesion are less than 0.5 cm.Sonograms may detect only one third of lesion identified at autopsy (Papile, 1997) (Keeney et al., 1991).Whereas MRI is the most sensible examination the routing use of MRI scans for all premature infants has not be recommended although it may be useful in some high risk premature infants or older infant or child born prematurely who presents with cognitive, motor or sensory impairment (Ment et al., 2002).

Prevention and treatment
The strategies to prevent PVL emphasize the maintenance of cerebral perfusion.All conditions that impair cerebrovascular autoregulation should avoid by correcting abnormalities in blood pressure and blood gases.Antenatal exposure to betamethasone may be associated with decrease risk of cystic PVL, more than in infants whose mothers received dexamethasone or no received glucocorticoid treatment (Baud et al., 1999).Management of PDL after discharge from the hospital is directed at identification of any cognitive, sensory or motor impairments, and appropriate therapies for any such impairment.Promising studies of neuroprotective strategies to prevent or minimize PVL are being conducted in animal models, but human trials of such agents are probably still years away (Oka et al., 1993;Follett et al., 2004).

Evolution and prognosis
Cranial US and brain MRI may yield prognostic information in neonates and children with PVL.Infants with more extensive white matter injuries and persistent ventricular enlargements are more likely to have severe motor and cognitive deficits.When PVL and cystic formation are found in the neonatal period using US, exist a subsequent risk for developing cerebral palsy.In a twelve studies review 50% of infants with periventricular echolucency developed cerebral palsy compared with 2.6% with infants with normal cranial US scans (Holling & Leviton, 1999).Considering the increase on survival rates of extremely preterm infants (birth weight less than 1000g it would be reasonable to consider MRI to add more prognostic information for such high risk infants.Abnormal white matter a gray matter findings on brain MRI at term are predictors of adverse neurodevelopment outcome. In VLBW there is an approximately 10% incidence of cerebral palsy and up to 50%incidence of school difficulties largely due to PVL with PVHI being the other cerebral lesion that contributes significantly to neurologic disabilities.The incidence of neurologic impairments increases with lower gestational age at birth.Cerebral palsy in children born before 36 weeks gestational age is 20%.Spastic diparesis is the most common form of cerebral palsy in children born prematurely because PVL typically affects the periventricular white matter closest to the ventricles (Ancel et al., 2006).Despite of these negative numbers thanks to therapy and family support some of these children get a social adaptation that increases the quality of their lives.

A Chilean experience
From the years 2001 to 2006, we carried out a prospective cohort study at a tertiary care hospital at Valdivia, Chile that includes all inborn neonates with gestational age of 32 weeks or less and birth weight of 1500 g or less (Barria & Flandez, 2008).Neonates dead within first seven days of age, with major malformation at birth or transferred to another center without ultrasound evaluation were excluded.For this study an experienced pediatric neurologist performed serial cranial ultrasound to 164 neonates according the protocol of the Chilean Ministry of Health.The images were obtained by using a 7.5-MHz transducer with Medison Mysono 201 ® equipment.The US began within the first week of life, and then at 15 and 30 days.
The following epidemiological and clinical variables were obtained from all premature infants: gestational age, birth weight, adequacy of weight to gestational age, gender, Apgar score, antenatal use of corticosteroids, type of delivery, rupture of membranes, maternal pre-eclampsia and chorioamnionitis.
Based on the main outcome (IVH: yes/no; PVL: yes/no), variables among groups were compared using Fisher exact test for categorical data and the Student t-test or Mann-Whitney U test for continuous variables.Relative risk (RR) and its 95% confidence interval (CI 95% ) were calculated as univariate estimation of the risk of IVH and PVL.Adjusted odds ratios (OR) were estimated using multiple logistic regression with backward stepwise incorporation into the initial model of every variable showing a P value ≤ .25 after the univariate analysis.The established level of statistical significance was p < .05.Data processing and analysis were carried out using Stata 8.1 (Stata Corporation, College Station, Texas).

Incidence of IVH and PVL, and factors related
The accumulated incidence of IVH was 18.3% (30/164), distributed in 30% grade I (9/30), 36.7% grade II, (11/30), 20% grade III (6/30) and 13.3% grade IV (4/30).The percentual distribution of IVH by birth weight and gestational age is show in figure 1 and 2, respectively.In the univariate analysis IVH was significantly associated with birth weight, estimating an OR of 5.6 (CI 95% 1.4 to 21.8) in infants under 750g.In other weight categories there was no significant association.According with gestational age, comparing with the group over 30 weeks, a significantly association was found in infants among 24-26 weeks (OR 5.4, CI 95% 1.72 to 16.7) and among 27-29 weeks (OR 5.4, CI 95% 1.72 to 16.7).On the other hand, the Apgar at first minute showed a significant reduction in the risk of IVH for each additional point in the score (OR 0.82, CI 95% 0.70 to 0.95).Other factors analyzed (chorioamnionitis, rupture of membranes, pre-eclampsia, etc.) were not significantly associated with IVH.In multivariate analysis, only Apgar at one minute was associated independently of the outcome of interest.
Ultrasound assessment allowed the detection of 61 neonates (37.2%) with abnormal white matter, 22 with PPVE and 39 with cystic lesion.Thus, the overall incidence for each event was 13.4 and 23.8% respectively.PVL was classified as mild in 64.1% (25/39), moderate in 30.8% (12/39) and severe in 5.1% (2/39).In addition, 23.3% of children with IVH also developed PVL (7/30).While PPVE was found in 50% (2/4) at 24 weeks of gestational age, a lower incidence remained relatively stable between 25 and 30 weeks, fluctuating between 14.3 and 16.7%, and declined clearly from week 31.For its part, the c-PVL was found most often below 28 weeks, accumulating 66.7% of cases (26/39).Between 24 and 27 weeks, the PVL incidence reached at least 50%, and significantly decreased from week 28 (Figure 3).A similar pattern occurred in the distribution of c-PVL and PPVE by weight strata, which the highest incidence occurred under 1000 g, reaching to below 750 g up to 21.4 and 64.3%, respectively.Additionally, the severity of periventricular lesions was higher in most premature infants, finding medians of gestational age for PPVE, mild c-PVL and moderate to severe c-PVL of 29, 27 and 26.5 weeks, respectively.Birth weight, gestational age, Apgar score and maternal hypertension were significantly associated with c-PVL (p<0.05).There were no significant differences in other characteristics assessed such as chorioamnionitis, antenatal corticosteroids nor other perinatal factors.In the univariate risk estimation, there was an increased risk of c-PVL in extremely low-birth weight newborns (<1000g) (RR 5.18, CI 95% 2.8 to 9.61) and in infant under 28 weeks of gestational age (RR 4.83, CI 95% 2.72 to 8.58).A risk reduction effect at border of statistic significance was detected for the presence of maternal hypertension (RR 0.48, CI 95% 0.23 to 1.02).There was no effect of Apgar score and other perinatal conditions on the development of c-PVL.To PPVE, we found an increased risk with Apgar score ≤ 3 at one minute and presence of IVH.The male showed a risk reduction in the limit of statistical significance (p=0.058).The extremely low-birth weight and prematurity below 28 weeks showed no association with PPVE.
Our results verify the effect of gestational age (prematurity) and birth weight commented previously in this chapter.Highlight of our findings the association found between PVL and pre-eclampsia.Consistent with this finding, a study previously has reported that children of whose mothers developed preeclampsia with intrauterine growth retardation had a low incidence of PVL (0.9%) and showed a significantly lower risk of cystic lesions (OR 0.08, CI95% 0,02 to 0.41) (Baud et al., 2000).In this sense, other study showed that none of the infants with a history of preeclampsia had c-PVL (Murata et al., 2005).However, other researchers found that this effect has been limited until 32 weeks, observing, on the contrary, an increased risk in children between 33 and 35 weeks of gestational age (Resch et al., 2000).Consequently, although the evidence is unclear, there is pathophysiologic support for considering hypertension in pregnancy as a protective factor on the incidence of c-PVL, based on self-regulatory mechanisms of the fetoplacental circulation developed in response to fluctuations of vascular tone.This would allow deal adequately potential hypoxicischemic episodes in the fetal brain.In this function, activation of the renin-angiotensin system in the fetoplacental unit caused by pre-eclampsia, would have a principal action (Ito et al., 2002).Likewise, hypertension and growth restriction can accelerate neurological maturity (Hadi, 1984), suggesting that early maturation may reduce brain disorders.It is also likely, however, that there are intermediate factors in this causal chain that must be clarified.

Follow up at four years
From 153 children potentially eligible born between September 2001 and June 2005 were included 81 preterm infants <32 weeks and/or VLBW with neurological evaluation at the fourth year of life (52.9%).Of the remainder, in 12 cases there was no record and in 60 there was no neurological control at this age.We calculated the incidence of neurological disorders and assessed association for different variables using t-test and Fisher exact test.Risks were calculated for overall and specific neurological areas for different variables, by estimating crude and adjusted odds ratio using multiple logistic regression.
In multivariate analysis, the gestational age was significantly associated with motor disorder estimating an adjusted OR 0.57 for each additional week (CI 95% 0.40 to 0.81).The history of IVH also showed a significant association estimating an OR 4.3(CI 95% 1.1 to 17.7).Similarly, for cognitive impairment, was estimated a lower risk with higher gestational age (OR 0.64, (CI 95% 0.42 to 0.95).

Conclusion
Gestational age is an important predictor of neurologic outcomes and therefore the systematic monitoring of premature infants allows the diagnosis need to target intervention and/or rehabilitation for improving the quality of life of the child and family.Since prematures improve surviving, decreasing the adverse outcomes taking the best strategies to prevent brain injury is must necessary.Preventing the premature birth is the main goal.When it isn´t possible, knowing the pathofisiology of the brain lesions and it´s risk factors can minimizes the final results in terms of neurodevelopment.Strategies in early intervention for lower motor handicaps, specially in whose have cranial US altered is needed.Their periodic neurological evaluation for finding behavioural and cognitive impairments is needed for giving them the best expectations in their quality of live.
. The routine ultrasound recommended by the Quality Standard Subcommittee of the America Academy of Neurology and the Practice Committee of the Child Neurology Society made the following recommendations: Routine ultrasound screening should be performed on all infants with gestational age less than 30 weeks.Screening should be performed at 7 to 14 days of age and repeated at 36 to 40 weeks postmenstrual age.This strategy is designed to detect unsuspected PVHI, development of PHH or ventriculomegaly.CT scanning is less useful for Intraventricular Hemorrhage and Periventricular Leukomalacia 93 the diagnosis of PVL in the very preterm infant because it detects fewer lesions than does MRI or US