Risk Factors and Variability of Cerebral Palsy Related to Prematurity

Brain scans as predictors of mild and severe cerebral palsy in preemies.

By Helen Harrison

Brain scans May Not Predict Cerebral Palsy

About half of all preemies diagnosed with CP have normal NICU ultrasounds. NICU ultrasounds give only limited information about what is going on the baby's brain. A "normal" ultrasound does not necessarily mean normal outcome.(1 Nearly one half of all preemies who develop cerebral palsy have normal NICU ultrasounds. (1)

Scans change over time -- sometimes fairly quickly -- and it is easy to miss significant events in the brain unless scans are performed constantly. Although bleeds may "resolve" or cysts may disappear, it does not mean that they didn't happen or that no damage has taken place. A normal NICU scan often does not remain normal in the months and years following NICU discharge. Half of all preemies who leave the NICU with "normal" scans no longer have normal scans on MRIs in adolescence. (2)

As background, cysts or holes in the brain characteristic of PVL can form but eventually "cave in" with surrounding brain tissue moving in to fill up the space. The cysts then "disappear," however the brain tissue that was destroyed in the area of the cyst is still gone (along with the functions controlled by that bit of brain tissue). Sometimes these cysts are quite small ("microcysts") and cannot be seen on scans. If many small cysts form and later on cave in this way, brain volume is diminished and all that is eventually seen on the scan are enlarged or misshapen ventricles, indicating that the ventricles have expanded outward to fill the space created by the diminished brain volume. This is called hydrocephalus "ex vacuo." It does not require shunting, but is a red flag for future developmental problems.

Sometimes, the only sign of PVL is not a cyst, but a transitory "flare" or "echodensity" that may be visible briefly on ultrasound before it disappears. Unless scans are being constantly performed (unfortunately this is not practical) these "flares" may never even be noticed. According to pediatric neurologist Mark Sher "there is probably much more in the way of mild to moderate non-cystic PVL out there that is causing the more pervasive problems with learning and motor control than we think...[we] just don't know how to screen for it." (3)

Changes in the presence or amount of blood from IVH can also be seen on scans. When blood from hemorrhage in the ventricles (or elsewhere in the brain) lessens or disappears, the bleed is said to have "resolved." Unfortunately, this does not mean that all is well and no damage has been done. *All* bleeds eventually "resolve" meaning that the blood is reabsorbed by the body, but the damage that caused or accompanied the bleed remains, although it is not always visible on scans.

3) Damage can occur not only to the areas around the ventricles, but also in less easily visualized areas such as the cerebellum.

How Common is Brain Damage in Preemies?

Twenty percent or more of preemies are thought to suffer some cerebellar damage, which occurs either as an independent event or as an extension of an IVH. (4) (5) Although the prognosis for large cerebellar bleeds is extremely poor in preemies (most are discovered on autopsy), I have not been able to find any information on the outcomes in cases of small bleeds which may be barely visible on scans.

Some sources mention damage to the cerebellum, along with lesions in the brain stem and basal ganglia, as possibly playing a role in later behavioral and psychiatric problems in preemies. (6)

If Your Preemie has an Abnormal Ultrasound

The following information is from the Neonatal Brain Hemorrhage Study (7), a geographically representative sample of 1,105 infants born weighing between 501-2,000 grams of whom 777 were followed to age two. Of these 777 children, 113 were diagnosed with cerebral palsy (15%) by age two. The children with CP were subdivided as to whether they had "disabling CP" or "nondisabling CP."

Disabling CP was defined as having abnormal neurological signs such as abnormal tone, preservation of primitive reflexes, abnormal range of motion in the hip, ankle, etc., along with one or more of the following: 1) inability to walk five steps unaided by age two years, 2) Bayley motor score greater than one standard deviation below the Bayley performance score, 3) the receipt of physical therapy for motor disability, 4) the receipt of surgical intervention for motor disability, and 5) use of braces or other physical assistive devices.

Nondisabling CP was defined as "clearly abnormal neurological findings, commonly increased leg tone with inability to dorsiflex at the ankle and/or limitation of hip extension and adduction with inability to straight leg raise to 180 degrees. Most of these children when examined by a study child neurologist were classified as having mild spastic diplegia." Although these children exhibited the motor abnormalities characteristic of CP, the examiners felt that their abnormalities would not interfere seriously with daily living.

Of the 113 CP cases, 61 were classified as disabling CP and 52 were classified as nondisabling CP.

NICU brain scan status was available for all 777 children examined. 149 children had category 1 scans meaning they had "isolated germinal matrix/intraventricular hemorrhage." This corresponds to grades 1 or 2 IVH. Sixty-three had category 2 scans defined as "parenchymal echodensities/lucencies (PEC) with or without germinal matrix or intraventricular hemorrhage and/or ventricular enlargement with or without germinal matrix or intraventricular hemorrhage." This would include grades 3 and 4 bleeds (since both include ventricular enlargement) and/or PVL and/or ventricular enlargement without bleeding. The researchers used this categorization because every situation in category 2 represented damage to the cerebral white matter of the brain.

Of the 63 children with category 2 scans 33 (52.4%) had disabling cerebral palsy. Of the remaining 30 children, 8 had nondisabling CP. A total of 41 out of the 63 had CP giving an overall CP rate of 65%. Of the children with category 1 scans, 37 out of 149 infants had disabling CP (24.8%). Of the remaining 112 children, nondisabling CP was found in 10, giving a total CP rate of 47/149 or 32%.

How Accurate are Brain Scans as Predictors?

The almost exclusive focus on bleeds as a predictor of poor outcome is misleading. BPD (and other factors) seem to be almost as important in predicting developmental disorders as are ultrasound abnormalities.

Most children who develop CP despite a normal ultrasound (and that's about half of all preemies who develop CP) had BPD. Therefore, BPD (or "some unknown antecedent of BPD" Infection? cytokines?) is presumed to be the cause of nearly half of all cases of CP in preemies. (8)

Both BPD and Grade 3-4 bleeds are strongly associated with not being able to walk at age 22 months and motor scores on the Bayley below 70. (p=0.0001 in both cases) and with mental scores on the Bayley below 70. (for BPD p=0.0004, for severe IVH p=0.0001) (9)

In a study of 133 7-year-old preemies who weighed less than 1500 g at birth, those who had severe ultrasound abnormalities (grades 3-4 bleeds and/or PVL and/or persistent ventricular enlargement) had a mean IQ of 81.8 while those without severe abnormalities had a mean IQ of 91.6. Preemies with BPD (need for oxygen at 36 weeks g.a.) had a mean IQ of 81.75 while those who did not have BPD had a mean IQ of 91.58. (10)

In another comparison of BPD and non-BPD VLBW preemies at age 8, the BPD children had a mean IQ of 86 (range of 52 - 108) while the non-BPD preemies had a mean IQ of 96 (range of 64 - 130). (Children with CP or abnormal brain scans were not included in this study). (11)

Other Risk Factors for Cerebral Palsy

Other risk factors that have been shown to have a relationship to CP and developmental problems include:

a) low thyroid levels following birth (12),

b) low CO2 levels (from overly aggressive mechanical ventilation) (13),

c) sepsis -- (10).

Even without risk factors, "healthy," "feeder and grower" preemies show significant differences in the structure and function of their brains when compared to full term babies of the same gestational age. The "ordinary" stresses of preterm birth seem to be enough to derail the normal migration of brain cells from the periventricular area to the cortex, as well as normal myelination (insulation of brain cells) and normal connections among brain cells. These differences in brain growth and development may partially explain the cognitive, motor, and behavioral problems found in preemies who have no other apparent risk factors. (3)

Good News

Despite all this, follow-up studies suggest that over a third of VLBW preemies proceed fairly normally through the school system and do not have problems of the sort that are tested for in outcome studies.


1. Pinto-Martin et al. "Cranial ultrasound prediction of disabling and nondisabling cerebral palsy at age two in a low birth weight population" Pediatrics 1995;95:249-254.
2. Rifkin et al. "Brain MRI at 14-15 years in very preterm infants and its relationship to cranial ultrasound in the newborn period." Presented at 22nd International Congress of Pediatrics, Amsterdam, August 14, 1998.
3. Sher. "Impact of neonatal brain injury on the development of the preterm infant" presented at "Developmental Interventions in Neonatal Care," San Francisco, Nov 17, 1998-- the tape of this lecture can be ordered by calling 800-679-3646; be sure also to ask for the tape of Dr Sher's Nov 16 presentation "The developing neonatal brain."
4. Martin et al. "Massive intracerebellar hemorrhage in low-birth-weight infants" Journal of Pediatrics. 1976; 89:290-293.
5. Also see Volpe. Neurology of the Newborn, Vol. 3, W.B. Saunders, 1995.
6. Whitaker et al. "Psychiatric outcomes in low-birth-weight children at age 6 years: relation to neonatal cranial ultrasound abnormalities" Archives of General Psychiatry.1997; 54:847-856.
7. Pinto-Martin et al. Cranial ultrasound prediction of disabling and nondisabling cerebral palsy at age two in a low birth weight population. Pediatrics. 1995; 95:249-254.
8. O'Shea et al. "Risk factors for cerebral palsy in very low birthweight infants without major cranial ultrasound abnormalities." Pediatric Research. 1998; 43:224A, #1309.
9. Vohr et al. "Neurodevelopmental and functional outcome of extremely low birth weight infants."Pediatric Research 1998; 43:233A, #1362
10. Taylor et al. "Predictors of early school age outcomes in very low birth weight children." Journal of Developmental and Behavioral Pediatrics 1998; 19:235-243.
11. O'Shea et al. "Intelligence and academic achievement at eight years of age in children recovering from neonatal chronic lung disease." Pediatric Research. 1997; 41:207A #1228.
12. Reuss et al. "The relationship of transient hypothyroxinemia in preterm infants to neurological development at two years of age." The New England Journal of Medicine. 1996; 334:821-7
13. Graziani et al. "Mechanical ventilation in preterm infants: neurosonographic and developmental studies" Pediatrics. 1992; 90:515-522
14. Hunt et al. "Further investigations of intellectual status at age 8" in _The Psychological Development of Low Birthweight Children. (eds. Sigman and Friedman) Norwood, NJ: Ablex Publishing, 1992:315-340.
15. Bennett, FC. "Neurodevelopmental outcome in low birthweight infants" in Neonatal Intensive Care. (ed. R.D. Guthrie). New York: Churchill Livingstone, 1988:221-249.

Helen Harrison was the well known author of The Premature Baby Book, for many years referred to as the "Bible of Prematurity" by older preemie parents. These observations are excerpted with permission from posts to the Preemie-child support group.