RU2325840C1 - Spastic infantile cerebral paralysis prediction for premature children - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to the medicine, in particular to the children's neuroscience and can be used as a tool for spastic infantile cerebral paralysis prediction for premature children. The method includes determination of the gestational age, duration of anhydrous period, artificial pulmonary ventilation, partial parenteral feeding, and a peri- intraventricular hemorrhage of the III-IV degree, periventricular leukomalacia, brain edema, the horns size of the lateral ventricles rear. Each indicator should be provided with the diagnostic rate as far as information content is decreasing. Then the amount of diagnostic coefficients should be determined starting with the amendments. The decision can be taken on the possibility of cerebral palsy child if the amount of diagnostic coefficients in descending order exceeds the upper threshold limit of 0.68. It can also be taken on the absence of cerebral palsy child, if the amount of diagnostic lower coefficients is lower than threshold which equals to -1.2. This method allows to predict the formation of spastic cerebral palsy on the 21^st day of life. It reduces the probability of premature baby's disability, as a result of changes in observation and treatment.

EFFECT: this method allows to predict the formation of spastic cerebral palsy on the 21^st day of life.

1 tbl, 3 ex

Description

The invention relates to medicine, in particular to pediatric neurology, and can be used as a method for predicting cerebral palsy (cerebral palsy) in premature infants.

An objective separation of patients as far as possible the formation of cerebral palsy is the most important classification task of rehabilitation medicine. It has long attracted the attention of researchers, since it is the basis for an individualized approach to dispensary observation and therapy, promotes a differentiated attitude to perinatal brain lesions in premature babies, and allows you to determine the most important perinatal risk factors that contribute to the formation of cerebral palsy.

The relevance of the proposed method is determined by the high frequency of the formation of cerebral palsy among premature babies - from 20 to 35% [Bennett F.C., Scott D.T. Long-term perspective on premature infant outcome and contemporary intervention issues // Semn. Perinatal. - 1997 Jun. - No. 21 (3). - P.190-201.].

It is noted that the method for predicting spastic forms of cerebral palsy, first of all, should characterize the severity of pathomorphological changes in the central nervous system and be based on the most informative factors of the perinatal outcome.

A known method for predicting cerebral palsy using the hardware-software complex "Micromotorics-2000" [Efimov A.P. Systemic understanding of the comprehensive treatment and rehabilitation of children with the consequences of birth injuries. Second International Congress "Restorative Medicine and Rehabilitation 2005". September 20-21. Abstracts, Moscow, 2005. P.24-25], allowing to register and analyze the micromotion of the body under various pathological conditions of the musculoskeletal and nervous systems.

The disadvantages of the method:

1. Not applicable to premature babies, due to the characteristics of their neuromuscular maturation, motor activity [Amiel-Tison C, Stewart A. Follow up studies during the first five years of life: a pervasive assessment of neurological function. Arch Dis. Child 1989; 64: 496-502] until they reach an adjusted age of 40 weeks.

2. Requires expensive equipment.

Closest to the claimed method for predicting cerebral palsy is a neonatal clinical risk index - Neonatal Medical Index (NMI), which takes into account the baby’s body weight at birth, the duration of mechanical ventilation (IVL), the need for additional oxygenation and its duration, persistence of fetal communications, the presence of apnea and bradycardia, the level of hyperbilirubinemia, requiring replacement blood transfusion, the presence and degree of peri-intraventricular hemorrhage (PIVC), periventricular leukemia malignancy (PVL), as well as the presence of neonatal seizure syndrome and meningitis [Korner A.F., Stevenson D. Kraemer H.C. Prediction of the development of low birth weight preterm infants by a new neonatal medical index. J. Dev. Behav. Pediatr. - 1993 Oct. - V. 14 (2), - p.106-111] - prototype.

The disadvantages of the method:

1. Highlights the most important neonatal risk factors for the development of adverse neurological abnormalities, but does not take into account the influence of factors of the intrapartum period.

2. There is no differentiated approach to delayed neurological abnormalities in the state of health of premature babies (motor, mental, cognitive and other unfavorable abnormalities), which is crucial when choosing the tactics of dispensary observation and treatment of this contingent of children.

3. The accuracy of the method is not known.

4. The method is used in infancy, starting from one month of life, does not allow for early diagnosis in the first weeks of life.

5. Given the impossibility of continuous monitoring of the respiratory rate and heart rate in all premature infants for a long time, there is the possibility of underestimating apnea and / or bradycardia and, therefore, incorrect determination of the prognosis.

The problem to which the invention is directed is to early predict the formation of cerebral palsy in order to reduce the number of children with disabilities from childhood.

The technical result that is achieved by using the invention consists in an earlier, objective, comprehensive, reliable, in comparison with the existing methods of forecasting, determination of the likelihood of cerebral palsy, and, therefore, to reduce the likelihood of disability of premature babies, as a result of an early change in observation tactics and treatment.

This is achieved by the fact that the gestational age at birth, the duration of the anhydrous period, the duration of partial parenteral nutrition, as well as the presence of cerebral edema, the size of the posterior horns of the lateral ventricles (according to neurosonography (NSG), are estimated at 0-14 days of life), while each considered indicator is set diagnostic coefficient (DK) in descending order of information content.

Then, the obtained diagnostic coefficients are summarized, starting from the correction, and upon reaching the upper threshold (VP) equal to 0.68, the child belongs to the group of children with cerebral palsy, and when the lower threshold (NP) is equal to -1.2, the child belongs to the group of children without cerebral palsy .

With this method, due to the expansion of the considered indicators and their subsequent processing, early and more reliable prediction of cerebral palsy is provided.

Patented forecasting method is illustrated in table 1 and was defined as follows.

Table 1
The most informative indicators and diagnostic coefficients Indicators Value DK Gestational age ≤30 weeks 0.22 > 30 weeks -0.84 Dimensions of the posterior horns of the lateral ventricles ≥13 mm 0.51 <13 mm -0.35 Anhydrous Duration > 60 hours 0.72 <60 hours -0.22 Duration artificial ≥10 days 0.63 lung ventilation <10 days -0.25 Peri-intraventricular hemorrhages of the III-IV degree there is 0.71 No -0.21 Duration of Partial Parenteral Nutrition ≥18 days 0.34 <18 days -0.34 Periventricular leukomalacia No -0.20 there is 0.58 Cerebral edema there is 0.58 No -0.20 Amendment - -0.34

In order to determine the most informative factors of the perinatal outcome, we conducted a dynamic prospective study of 97 preterm infants. All children were treated in the neonatal departments of the Regional Children's Clinical Hospital No. 1 (CSTO) of Yekaterinburg. In the hospital, NSG was carried out through a large fontanel with 5 and 7.5 MHz sector sensors using Siemens, Aloca-500, and SIM 5000 plus devices by sequentially obtaining 10 standard sections in the coronary and sagittal planes according to the generally accepted method (total - 198 studies). The presence of neurosonographic signs of grade III-IV PIVC, PVL, cerebral edema was determined, and the sizes of the posterior horns of the lateral ventricles were measured. Depending on the formation of spastic forms of cerebral palsy in 12-18 months of life, all patients were divided into 2 clinical groups:

Group I - children who had spastic forms of cerebral palsy (n = 16);

Group II - children with psychomotor development corresponding to the adjusted age at 12-18 months of life (n = 36).

The selection criteria for children in group II were:

- the absence of a delay in psychomotor development (pronounced or tempo). For this reason, 19 children were removed from the study;

- the absence of significant eye pathology (blindness or low vision), because children with sensory impairments have features of static-motor and / or psychomotor development. For this reason, 12 children were removed from the study.

- surviving children. For this reason, 4 dead children were removed from the study;

- regularly observed children.

To build a diagnostic model, a probabilistic diagnostic model was used, based on the Bayes theorem.

Since the type of distribution of the diagnostic parameters is unknown, the values of the indicators were optimally divided into two ranges and the conditional probabilities in the ranges were calculated for each of the two possible states of the patient.

To obtain the simplest type of decision rule and small sample size, the assumption of independence of attributes is used. For each symptom, diagnostic coefficients (DCs) are calculated and the forecast is constructed by the Wald sequential analysis method (Gubler E.V., Genkin A.A. Application of nonparametric statistics criteria in biomedical research. - L .: Medicine, 1973. - 141 p. )

All the studied characteristics are ranked in descending order of informativeness, which is calculated by the Kullback information measure. To predict the formation of cerebral palsy, 39 signs were initially used, of which 8 most informative signs were identified.

The forecast is based on the amount of DC and the decision is made when the sum reaches the upper (VP) or lower thresholds (NP). The values of VP and NP are determined by the level of acceptable probability of diagnostic errors.

The error in skipping cerebral palsy was considered the most dangerous and its probability was assumed to be 5% (the sensitivity of the forecast method is 95%). The error in the overdiagnosis of cerebral palsy was considered less dangerous and its probability was taken equal to 20% (specificity of the forecast method is 80%). The method for determining VP and NP is described in (Gubler E.V., Genkin A.A. Application of nonparametric statistical criteria in biomedical research. - L .: Medicine, 1973. - 141 p.). The obtained threshold values: VP = 0.68 and NP = -1.20.

The a priori probability of the occurrence of cerebral palsy is 31.4%, and for non-cerebral palsy - 68.6%. The relations of a priori probabilities for cerebral palsy and non-cerebral palsy are taken into account when building the forecast.

The forecast for the possible development of cerebral palsy is based on the following rule:

1. To the correction taking into account the ratio of a priori probabilities equal to -0.34, the values of DC corresponding to the gradation of the symptom found in the patient are added sequentially (in descending order of informativeness).

2. If the amount of DC in the descending order of informativeness at a certain step of the analysis is equal to or higher than VP, then a decision is made to assign the patient to the group of cerebral palsy.

3. If the amount of DC in the descending order of informativeness at a certain step of the analysis is equal to or less than NP, then a decision is made to classify the patient as a group without cerebral palsy.

4. If the amount of DC remained in the interval between NP and VP, then the summation of the DC continues until either one of the thresholds is reached or the list of informative signs is not exhausted. In the latter case, the forecast is not defined, since the available information is not enough to make a decision with a given level of reliability.

We give clinical examples of the analysis of the information content of signs and determining the prognosis of the formation of cerebral palsy.

Example 1

Z., girl, medical history number 9181.

A child from III pregnancy, I childbirth. Delivery at 26 weeks of pregnancy, natural, quick (6 hours). A premature girl with a body weight of 1200 was born, an Apgar score of 2/5 points. From birth, she was on mechanical ventilation; the duration of mechanical ventilation was 11 days.

Using the identified most informative features, the following equalities were compiled:

Amendment = -0.34 + 0.22 (gestational age = 26 weeks <30 weeks.) = - 0.12 + 0.51 (ZRBZH = 17 mm> 13 mm) = 0.39-0.22 (anhydrous period = 6 h <60 h) = 0.17 + 0.63 (duration of mechanical ventilation = 11 s> 10 days) = 0.80> VP = 0.68, therefore, when determining the sum of the DC of four informative signs, a value exceeding the VP was obtained. Conclusion: Conclusion: there is data for cerebral palsy.

At the age of 11 months of life, a diagnosis was made: cerebral palsy.

Example 2

B., girl, medical history number 3503. A child from I pregnancy, I childbirth. Childbirth in the period of 28 weeks of pregnancy, operational. A premature girl with a body weight of 1090 was born, an Apgar score of 6/7 points. From birth, she was on mechanical ventilation; the duration of mechanical ventilation was 9 days.

Using the identified most informative features, the following equality was compiled:

Amendment = -0.34 + 0.22 (gestational age = 26 weeks <30 weeks.) = - 0.12-0.35 (ZRBZH = 7 mm <13 mm) = - 0.47-0.22 (duration of the anhydrous period = 0 h <60 h) = - 0.69 -0.25 (duration of mechanical ventilation = 9 s <10 s) = 0.94-0.21 (IVH 3-4 st = no) = - 1.15-0.34 (duration of partial parenteral nutrition = 16 days <18 days) = - 1.49 <-1.2 .

Conclusion: data for cerebral palsy - no.

At the age of 15 months of life, at the next appointment with a neurologist, the child was diagnosed with RCSC. Myatonic syndrome.

Example 3

P., boy, case history number No. 10234.

A child from II pregnancy, I childbirth. Childbirth in the period of 31 weeks of pregnancy, natural. A premature boy with a body weight of 1460 was born, Apgar score - 1/3 points. From birth, he was on mechanical ventilation; the duration of mechanical ventilation was 9 days.

Using the most informative signs identified, the following equalities were compiled: Correction = -0.34-0.84 (gestational age = 31 weeks> 30 weeks) = - 1.18 + 0.51 (ZRBZH = 15 mm> 13 mm) = - 0.67-0.22 (duration of the anhydrous period = 10 hours <60 hours) = - 0.99-0.25 (duration of mechanical ventilation = 9 days <10 days) = - 1.24 <-1.2. Conclusion: data for cerebral palsy - no.

Claims (1)

A method for predicting spastic forms of cerebral palsy in premature infants by determining the duration of mechanical ventilation, the degree of peri-intraventricular hemorrhage, the presence of periventricular leukomalacia, characterized in that the gestational age, the duration of the anhydrous period, the duration of partial parenteral nutrition, and the presence of cerebral edema are additionally evaluated. and the sizes of the posterior horns of the lateral ventricles in 1-14 days of life, while each indicator taken vayut diagnostic coefficients (DC) in descending order of informativeness, namely at ≤30 weeks gestational age, Diagnostic adjusted coefficient of 0.22, and at gestation period> 30 weeks, Diagnostic ratio (-) 0.84; with the sizes of the posterior horns of the lateral ventricles ≥13 mm, a diagnostic coefficient of 0.51 is established, and with the sizes of the posterior horns of the lateral ventricles <13 mm, the diagnostic coefficient (-) 0.35; for a duration of an anhydrous period> 60 hours, a diagnostic coefficient of 0.72 is established; for a duration of <60 hours, a diagnostic coefficient of (-) 0.22; with a duration of mechanical ventilation ≥10 days, a diagnostic coefficient of 0.63 is established; with a duration of <10 days, a diagnostic coefficient (-) of 0.25; in the presence of peri-intraventricular hemorrhages of III-IV degree, a diagnostic coefficient of 0.71 is established, in the absence of a diagnostic coefficient of (-) 0.21; with a duration of partial parenteral nutrition ≥18 days, a diagnostic coefficient of 0.34 is established; with a duration of <18 days, a diagnostic coefficient (-) of 0.34; in the presence of periventricular leukomalacia, a diagnostic coefficient of 0.58 is established; in the absence of a diagnostic coefficient (-) of 0.20; in the presence of cerebral edema, a diagnostic coefficient of 0.58 is established; in the absence of cerebral edema, a diagnostic coefficient of (-) 0.20; wherein the amendment is (-) 0.34; after which the sum of the diagnostic coefficients is determined, starting with the amendment, and a decision is made to assign the patient to the group of cerebral palsy if the sum of the diagnostic coefficients in descending order of information at a certain analysis step is not less than 0.68, and to the group of non-cerebral palsy if the sum of the diagnostic coefficients is equal to or less than (-) 1.2.