Pretérminos tardíos

Noviembre 2006

 

Factores de riesgo para morbilidad y mortalidad neonatal entre recién nacidos prematuros tardíos “sanos”

 

Carrie K. Shapiro-Mendoza, PhD, MPH,* Kay M. Tomashek, MD, MPH,* Milton Kotelchuck, PhD, MPH,† Wanda Barfield, MD, MPH,‡ Judith Weiss, ScD,† and Stephen Evans, MPH†

 

*Centers for Disease Control and Prevention, Division of Reproductive Health, Maternal and Infant Health Branch, Atlanta, GA.

†Department of Maternal and Child Health, Boston University School of Public Health, Boston, MA.

‡Massachusetts Department of Pubic Health, Bureau of Family and Community Health, Boston, MA.

 

Semin Perinatol 30:54-60. Abril 2006

 

Colaboración : Dr Gerardo Flores Henríquez    Pediatra Neonatólogo   Puerto Montt     Chile

 

Los niños pretérminos tardíos (desde 34 semanas a 36 6/7 semanas de gestación) son más propensos a presentar problemas para alimentarse, inestabilidad de temperatura corporal, sepsis, retrasos en excreción de bilirrubina, hipoglicemia y problemas respiratorios y problemas respiratorios que los niños de término debido a su inmadurez.1,2 A pesar de su riesgo aumentado de morbilidad, los neonatos pretérmino tardíos a menudo son cuidados en puericultura despues del nacimiento y son dados de alta del hospital a los  2 - 3 días de edad.2 Los niños pretérmino tardíos son a menudo tratados como niños de término , maduros desde el punto de vista del desarrollo, debido a que muchos tienen peso de nacimiento normal ( ³ 2500 g) y el mismo tamaño que los niños de término.

La proporción de todos los nacimientos en EEUU que son pretérmino tardíos aumentó 16%, desde 7.3% en 1992 a 8.5% en 2002. En 2002, los nacimientos pretérmino tardíos comprendieron 71% de todos los los nacimientos pretérmino (< 37 semanas) y 8.5% de todos los nacimientos en EEUU. 3 La explicación más probable para el aumento en nacimientos de pretérmino tardíos desde 1992 a 2002 es el aumento en las intervenciones médicas en embarazos a ó más allá de las 34 semanas de gestación.4 Un estudio estudio reciente de Wang y cols. 2 examinó la morbilidad y costo acumulado durante la hospitalización entre los niños pretérmino tardíos y de término. Los autores encontraron que los niños pretérmino tardíos de 35 a 36 semanas de gestación tenían significativamente más problemas médicos durante la hospitalización del parto y aumentaban los costos hospitalarios en comparación con los niños de término. Si el número de niños pretérmino tardíos aumenta, su impacto sobre los costos hospitalarios continuará aumentando. Los esfuerzos para prevenir sus complicaciones postparto son dificultados por la limitada investigación sobre nacimientos pretérmino tardíos. No hay estudios que hayan examinado los factores de riesgo entre niños pretérmino tardiís “sanos,” para reingreso hospitalario ó estadía para observación en el período neonatal. La comprensión de cuales características de los niños pretérmino tardíos implican el mayor riesgo de morbilidad neonatal es esencial para prevenir las complicaciones al alta postparto. El conocimiento de los factores de riesgo para morbilidad neonatal entre los niños pretérmino tardíos “sanos” puede ser usado para identificar los niños que necesitan monitoreo más estricto y seguimiento más precoz después del alta hospitalaria.

 

Métodos

 

Of the 365.53 vaginal singleton deliveries in our study population during the study period, 15.147 occurred at 34 to 36 weeks’ gestation. Of these, we excluded infants with a birth weight lower than 2000 g (n = 523), those who were discharged home 4 nights or later postdelivery (n = 2694), those who had significant health problems or procedures (n = 1416), those who were transferred following birth (n = 972), and those who did not survive to discharge after birth (n = 20). Our final study population was 9522 infants.

 

Definitions

Late preterm infants are infants born at 34 to 36 weeks’ completed gestation. “Healthy,” late preterm infants were defined as those who: (1) were discharged home before a fourth-night hospital stay, (2) weighed at least 2000 g at birth, and (3) did not have any significant health problems or procedures (ie, those classified with uncomplicated DRG (diagnosis-Related Group) codes 620, 629, 657, or 678). We did not include infants who were 34 to 36 weeks’ gestational age and weighed less than 2000 g at birth.

Neonatal morbidity was defined as any medical condition resulting in postdelivery inpatient  hospital readmission, observational stay, or mortality in the first 28 days of life. We included neonatal deaths because we considered death to be the result of severe morbidity. Principal diagnostic codes (ICD-9) used to define jaundice, infection, and other re-hospitalization and observational stay diagnoses are available from the authors on request. To calculate gestational age, we did not round up to the next of gestation. For example, if a baby was born at 36 weeks and 4/7, it was coded as 36 weeks.

Obstetric covariates included labor and delivery complications (none/any), pregnancy risk factors (none or any), parity (1, 2, 3, 4_; or then primiparous or multiparous), and prenatal care  utilization (Kotelchuck APNCU index: inadequate, intermediate, adequate, or adequate plus) . Ver : http://www.mchlibrary.info/databases/HSNRCPDFs/Overview_APCUIndex.pdf.   We calculated adjusted risk ratios (aRR) and 95% confidence intervals (95% CI) using modified Poisson regression with the robust option in Stata software to simultaneously control for several confounders.5,6 Additionally, we stratified by breastfeeding status at time of discharge and calculated aRR for neonatal morbidity among breastfed and nonbreastfed late preterm infants because we found that the majority (89%) of the neonatal readmissions for jaundice occurred among breastfed infants. All analyses were conducted using Stata.5

 

Results

 

Of the 9552 late preterm, “healthy” infants, 4.8% had an inpatient readmission and 1.3% had an observational stay. Five infants had both an inpatient hospital re-admission and an observational stay. Four infants died in the neonatal period, and of these, 3 infants also had inpatient hospital readmissions. The characteristics of our study population are described in Table 1. In this study, most of these late preterm infants were born at 36 weeks’ gestation, breastfed at discharge, and stayed in the hospital 2 to 3 nights after delivery. Most mothers of these “healthy” late preterm infants were U.S. born, multiparous, married, privately insured, 20 to 34 years old who had had adequate intensive prenatal care utilization and no labor and delivery complications. Among the 577 hospital readmissions and observational stays, three-quarters of late preterm infants had a principal diagnosis of jaundice (63%) or infection (13%). Other principal diagnosis categories each accounted for less than 5% of the total inpatient hospital readmissions and observational stays. The vast majority (95%) of all hospital readmissions for jaundice occurred on or before 7 days of age, with most occurring on days 3 to 5. In contrast, only 13% of the readmissions for infection occurred by day 7 and 30% by day 14. Jaundice was the most common primary diagnosis on hospital readmission for infants of all races and ethnicities, followed by infection.

Predictors of neonatal morbidity among late preterm infants using the adjusted risk ratio are reported in Table 2. The strongest infant risk factor for neonatal morbidity and the strongest overall risk factor was breastfeeding at time of discharge. Although the majority of breastfed infants (93%) did not require rehospitalization or observational stays in the neonatal period, breastfeeders had a higher risk (aRR 1.65, 95%CI 1.33, 2.04) of requiring subsequent hospital care in the neonatal period than their nonbreastfed counterparts. Among the sociodemographic risk factors, infants with Asian/Pacific Islander mothers were more likely to require subsequent hospital care in the neonatal period than non- Hispanic white infants (aRR 1.32, 95%CI 1.01, 1.72). Those at least risk for neonatal morbidity were non-Hispanic blacks (aRR 0.65, 95%CI 0.46, 0.90). Infants who had a public payer source at delivery also had a higher risk for neonatal

morbidity compared with those with a private source (aRR 1.25, 95%CI 1.00, 1.56).

The obstetric risk factors for neonatal morbidity were low parity and complications of labor and delivery. Late preterm infants whose mothers were primipara had a higher risk of neonatal morbidity than infants with multiparous mothers (aRR 1.44, 95%CI 1.23, 1.70). Infants whose mother had at least one labor and delivery complication were at a higher risk of neonatal morbidity than infants whose mothers did not have a labor and delivery complication (aRR 1.25,  5%CI 1.06, 1.47).

When we stratified the adjusted multivariable analysis by breastfeeding status at time of hospital discharge, we found that the risk factors for neonatal morbidity among late preterm infants differed between breastfed and nonbreastfed infants (Table 3). Risk factors for breastfed late preterm infants were similar to the risk factors identified in the unstratified analysis. However, risk factors for neonatal morbidity among nonbreastfed late preterm infants include male sex (aRR 1.62, 95%CI 1.10, 2.38), having any labor and delivery complications (aRR 1.37, 95%CI 1.03, 2.50), and receiving public insurance at the time of delivery (aRR 1.60, 95%CI 1.036, 2.50).

 

Discussion

 

In this population-based study to identify risk factors for neonatal morbidity among “healthy,” singleton late preterm infants, we found that the independent risk factors for neonatal morbidity were: having an Asian/Pacific Islander mother, being firstborn, being breastfed at discharge, having a mother with any reported labor and delivery complications, and having a public payer source at delivery. However, when we stratified by breastfeeding status, we found that breastfed late preterm infants were at increased risk of neonatal morbidity if they were first born or if their mother was an Asian/ Pacific Islander or had any labor and delivery complications. Nonbreastfed late preterm infants were more likely to have neonatal morbidity if they were male, received  public insurance at the time of delivery, or had a mother with any labor and delivery complications.

The 4.8% risk for rehospitalization in this population is consistent with the 4.4% re-hospitalization risk reported by Escobar and colleagues among similarly aged, preterm infants who had no prior history of receiving neonatal intensive care.1 Close to two-thirds of the neonatal morbidity was attributable to a principal diagnosis of jaundice. Preterm infants have developmentally immature livers compared with full-term infants, and this immaturity limits their ability to remove bilirubin fromthe circulatory system.7,8 The finding that late preterm, non-Hispanic black infants had lower neonatal morbidity than non-Hispanic white and Asian/Pacific Islander infants may initially be surprising; however, it is consistent with findings from previous published studies of risk factors for neonatal morbidity and mortality among term infants.1,9-11 A chief reason for this is that jaundice, the primary diagnosis reported for the majority of neonatal hospital readmissions, is more common among Asian/Pacific Islander and non-Hispanic white infants than non-Hispanic black infants.12,13 Also, non-Hispanic black mothers are less likely to breastfeed and therefore are less at risk for hospital readmission due to feeding-related problems, including jaundice.10,14

Late preterm infants who are breastfed are more likely to have feeding-related problems and are at greater risk for hospital readmission or observation stay related to inadequate weight gain, jaundice, or feeding difficulties.7,15 We found that 63% (n = 365) of the 577 late preterm infants who were readmitted to the hospital or for an observational stay had jaundice as a primary diagnosis, and 89% of those infants diagnosed with jaundice were breastfed. Although breastfeeding is the preferred nutrition for neonates and has many advantages to infants and mothers, not establishing successful breastfeeding can result in failure-to-thrive and dehydration. 16-18 Improving clinical assessments for jaundice and adequate intake in the breastfed infant, together with providing more comprehensive parental breastfeeding education and support before hospital discharge, may allow for earlier identification and management of breastfeeding-related problems. Improved assessment of the breastfed newborn would help detect early physiologic jaundice, which is common, from more exaggerated jaundice that might signal bilirubin encephalopathy or kernicterus.7,15,16

Infants who are discharged early may also be healthier than infants with longer lengths of stay. Infants with longer than 3-night hospital stays were excluded from the analysis as “nonhealthy” infants, so we could not assess later discharge as a protective factor. We examined the effect of prenatal care utilization because studies have suggested that adequate prenatal care results in the more frequent use of child preventive care.25 Because we are uncertain about the quality of this variable as a proxy for the use of child preventive services, we did not draw any inferences about its effect on neonatal morbidity risk. Removing prenatal care utilization from the final models had little effect on the adjusted risk ratios for the other covariates and would not change our study findings or conclusions. The current findings, if replicated, may assist in the prevention of neonatal hospital readmissions or observational stays postdischarge by allowing health care professionals to better assess “healthy,” late preterm infant readiness for hospital discharge. These findings could also help identify mothers who may require additional breastfeeding instruction before discharge and earlier follow-up after discharged home. Breastfed, late preterm infants should be monitored more closely and may require earlier follow-up than nonbreastfed, late preterm infants to prevent postdelivery discharge rehospitalizations or observational stays related to jaundice. This may be especially true for Asian/Pacific Islanders, firstborn infants, and those with mothers who had a labor and delivery complication.

This study adds to the small but increasing literature for late preterm births, an increasingly significant segment of the U.S. birth cohort.

 

Summary Points

Table 1 Select Characteristics of “Healthy” Late Preterm Newborns and Neonatal Morbidity Incidence, Massachusetts, 1998–2002  (Late preterm is defined 34–36 completed weeks’ gestation. “Healthy,” late preterm infants were those who were vaginally delivered, discharged home after less than a 4-night stay, had birth weights of at least 2000 grams, and had hospital DRG codes not pertaining to significant health problems or procedures. Morbidity refers to subsequent hospital care (hospital inpatient or observational stay care) or mortality in the first 28 days of life.)

 

Table 2 Adjusted Risk Ratios and 95% Confidence Intervals for Neonatal Morbidity Among “Healthy” Late Preterm Infants, by Selected Infant, Obstetric, and Sociodemographic Characteristics, Massachusetts, 1998–2002

 

Table 3 Adjusted Risk Ratios and 95% Confidence Intervals for Neonatal Morbidity Among “Healthy” Late Preterm Infants, by Selected Infant, Obstetric, and Sociodemographic Characteristics, Massachusetts, 1998–2002, Stratified by Breastfeeding Status at Discharge, Massachusetts, 1998–2002