Original Article
747
Postpartum Triage, Services Provided, and Length of Stay for Infants Born At 35 Weeks Gestation James I. Hagadorn, MD, MS1,2 Saritha Salikooti, MD1,2 Mariann Pappagallo, MD1,2 Jose Arias-Camison, MD3 Scott Weiner, MD4 Jorge Alba, DO5 Victor Herson, MD1,2 1 Divison of Neonatology, Connecticut Children’s Medical Center,
Hartford, Connecticut 2 Department of Pediatrics, University of Connecticut School of Medicine, Farmington, Connecticut 3 Divison of Neonatology, St. Francis Hospital, Hartford, Connecticut 4 Divison of Neonatology, The Hospital of Central Connecticut, New Britain, Connecticut 5 Divison of Neonatology, Manchester Memorial Hospital, Manchester, Connecticut
Address for correspondence James I. Hagadorn, MD, MS, Connecticut Children’s Medical Center, 282 Washington Street, Hartford, CT 06106 (e-mail: [email protected]).
Abstract
Keywords
► ► ► ► ►
infant premature length of stay health resources late preterm
Objectives The aim of this article is to (1) compare the care setting to which 35-week infants are initially triaged postpartum to the level of services subsequently provided; and (2) identify factors known at delivery or immediately postpartum associated with services received and length of stay during the birth hospitalization. Study Design In this multicenter retrospective study of 35-week infants born between 2007 and 2008, service capabilities of the initial postpartum care setting were categorized as level 1 or neonatal intensive care unit (NICU) using American Academy of Pediatrics definitions. Subsequent services actually provided were categorized as routine care, level 1, or >level 1. Results Over half of 431 studied infants were sent to a level 1 nursery postpartum. Of these, over 90% ultimately received routine care or level 1 services. Of 200 infants triaged to a NICU, the majority received only routine care or level 1 services. The great majority of infants requiring > level 1 services were identified promptly postpartum. Initial triage to the NICU was associated with significantly (p < 0.05) increased length of stay despite provision of similar services. Conclusions This study suggests a need for improved triage of 35-week infants and provides tools for this purpose. Validation of the models presented here is warranted.
Clinical Perspective • Among 35-week gestation infants triaged from delivery room to the NICU, a majority received only routine care or level 1 services. • Compared with initial triage to level 1 nursery, triage to the NICU was associated with increased length of stay despite provision of similar services. • Hospitals may be overutilizing the NICU setting for these infants, increasing length of stay, infant–parent separation, and costs.
received June 5, 2014 accepted after revision October 1, 2014 published online December 23, 2014
• This multicenter study documents a need for improved triage of 35-week infants and provides potential tools for this purpose. Postpartum care needs of late preterm infants (LPI)—those born 340/7 to 366/7 weeks’ gestation—vary considerably. Compared with term, LPI are at increased risk for postpartum hypoglycemia, respiratory distress, hyperbilirubinemia, or ineffective feeding.1–7 The public health impact of these morbidities is large, as LPI represent nearly 10% of U.S. births and three-quarters of this country’s premature deliveries.8–10
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1396689. ISSN 0735-1631.
Downloaded by: University of Pittsburgh. Copyrighted material.
Am J Perinatol 2015;32:747–754.
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
Cumulative incidence of short-term postpartum morbidity in LPI decreases from over 50% at 34 weeks to approximately 10% at 36 weeks.11 Consequently, LPI present a dilemma for providers choosing a postpartum neonatal care setting. Published guidelines encourage each delivery service to determine where observation can best be accomplished while simultaneously anticipating infant care needs and supporting the mother–infant dyad.12,13 However, tools to assist this potentially difficult determination are few. In our community, an informal consensus has developed in which infants at 36 weeks’ gestation are generally observed initially in a level 1 nursery, while 34-week infants are thought better served by transitioning in a level 2, level 3, or level 4 neonatal intensive care unit (NICU) setting. For infants born 350/7 to 356/7 weeks no such consensus or common protocol has emerged. Instead, 35-week infants are triaged on a case-bycase basis. In the absence of a clear indication for intensive care, such triage decisions may present a challenge. The National Institute of Child Health and Human Development panel’s call for research to improve precision of risk assessment in LPI subgroups has recently been reiterated.14,15 Likewise, the American Academy of Pediatrics (AAP) has suggested that more evidence is needed to determine outcomes for LPI by level of care.16 The purpose of this study was to examine postpartum triage of infants born at 35 weeks’ gestation. We wished to (1) compare the capabilities of settings to which 35-week infants are initially triaged postpartum to services subsequently provided and (2) identify factors known at delivery or immediately postpartum associated with services received and length of stay during the birth hospitalization. We hypothesized that discrepancies exist between the care capabilities of the initial care setting and subsequent services received, and that identifiable prenatal, intrapartum, and immediate postpartum factors are associated with subsequent services received and length of neonatal hospital stay (LOS).
Subjects and Methods This was a retrospective record review at five hospitals, each with a level 1 normal newborn nursery and a higher level (2, 3, or 4) NICU.16 Institutional Review Board approval was obtained at all sites. Participating hospitals—Hartford Hospital/Connecticut Children’s Medical Center Hartford NICU, University of Connecticut Health Center/Connecticut Children’s Farmington NICU, St. Francis Hospital, Hospital of Central Connecticut, and Manchester Memorial Hospital— included all facilities in the Hartford metropolitan area with both a delivery service and an NICU. At each hospital, initial triage of 35-week gestation infants following delivery was performed on a case-by-case basis by the hospitalist, neonatal nurse practitioner, neonatal fellow, or neonatologist attending the delivery. Triage decisions were made promptly following delivery if necessary, or after 1 to 2 hours of postpartum observation if the baby’s condition allowed. When warranted, such observation could include maternal–infant skin-to-skin care, measurement of temperature and blood glucose, serial respiratory assessment, measureAmerican Journal of Perinatology
Vol. 32
No. 8/2015
ment of oxygen saturation with pulse oximetry, and attempted feeding. Obstetric, neonatal, and nurse practitioner staffs for the hospitals were separate.
Subjects All studied infants were born from January 1, 2007 to December 31, 2008 with estimated gestational age (EGA) between 350/7 and 356/7 weeks. Infants were screened for inclusion using delivery room, nursery, and NICU logs. Infants were excluded if they required intubation, chest compressions or medications in the delivery room, for congenital anomalies requiring NICU admission, or for grossly incomplete records. Data regarding prenatal and intrapartum diagnoses and interventions, delivery room interventions, and the postnatal hospital course were collected from individual medical record reviews. Mothers were classified as diabetic if given that diagnosis in their obstetric record, and status regarding insulin therapy at the time of delivery was recorded. Presence of prenatal labor was as diagnosed by the obstetric care team. Date and time of birth, of initiation of specific treatments, and of discharge were recorded. LOS and age at initiation of treatment in hours were calculated. Highest level of services each infant received was determined using 2004 AAP definitions.17 Services classified as level 1 care included phototherapy, intravenous fluids or antibiotics, supplemental oxygen without positive pressure, and nasogastric feeding. Positive pressure via continuous positive airway pressure (CPAP), mechanical ventilation, or any need for surgery were classified as >level 1 services. Thus, if a 35-week infant was admitted to neonatal intensive care but nasogastric feeding was the highest level service provided, the infant was categorized as triaged to NICU but receiving only level 1 services. Infants requiring no services in the level 1 or >level 1 categories were classified as receiving routine care only. Infants were classified as appropriate, large (>90th percentile) or small (level 1 services during the birth
Downloaded by: University of Pittsburgh. Copyrighted material.
748
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
749
hospitalization (outcome) versus routine care only (model I), and provision of >level 1 services (outcome) versus routine care/level 1 (model II). Model performance was assessed by calculating sensitivity, specificity, positive predictive value, negative predictive value, and C statistic (area beneath the receiver-operator characteristic curve). Model calibration was assessed with the Hosmer–Lemeshow goodness-of-fit test and by contrasting observed and predicted rates of outcome within subgroups. Regression analyses were repeated using hierarchical linear or logistic regression to adjust for clustering of multiples within pregnancies where appropriate.
Screening hospital logs yielded 462 births at EGA 350/7 to 356/ 7 weeks during the study period. Of the 462 births 15 were subsequently excluded, as EGA on record review was not 35 weeks at the time of delivery. Six were excluded for congenital malformations and ten for receiving cardiopulmonary resuscitation in the delivery room. Of the 462 infants, 431 35-week infants from 384 pregnancies were studied. EGA was based on first trimester ultrasound in 58% of infants (n ¼ 248), date of implantation of technology-assisted pregnancy in 11% (n ¼ 46), date of last menstrual period in 8% (n ¼ 33), 2nd trimester ultrasound in 6% (n ¼ 25), 3rd trimester ultrasound in 2% (n ¼ 10), transvaginal ultrasound in 2% (n ¼ 9), and lacked a clear basis in the maternal record in 14% (n ¼ 60). Overall 11% (n ¼ 44) of mothers had diabetes mellitus. Deliveries were 53% vaginal (n ¼ 230), 34% primary cesarean section (n ¼ 147), and 13% repeat cesarean section (n ¼ 54). A total of 75 infants (17%) were delivered by cesarean section without labor. Studied infants had a mean birth weight 2,413 442 g, with 50% (n ¼ 218) white, 20% (n ¼ 84) black, 19% (n ¼ 80) Hispanic, and 50% (n ¼ 229) female. One-fifth (n ¼ 93) of infants received antenatal steroids, whereas 17% (n ¼ 72) were exposed to intrapartum magnesium sulfate. There were no positive blood cultures, surgeries, or deaths among study infants during their birth hospitalization.
Postpartum Triage and Subsequent Services Received ►Fig. 1 shows the initial triage setting and subsequent services received for the entire study cohort. Over one-third of infants (n ¼ 164) received only routine care without the need for any specific level 1 or > level 1 services. The highest level service received was 1 for 37% (n ¼ 158) and >1 for 25% (n ¼ 109). Of the 200 infants initially triaged to NICU, less than half received > level 1 services, whereas the majority received either routine care or level 1 services only. Over half (n ¼ 231) of study infants were initially sent to a level 1 setting following delivery. Of these, the great majority subsequently received only routine care or level 1 services, whereas less than one-tenth ultimately received > level 1 services. All sites triaged 35-week infants to both level 1 and NICU. The proportion triaged to level 1 varied significantly among sites, ranging from 38.5 to 68.5% (p < 0.001, chi-square test). Provision of > level 1 services did not differ significantly
among sites, ranging from 19.1 to 35.9% (p ¼ 0.106, chisquare test). ►Table 1 presents highest level of services received following initial delivery room triage, together with age at initiation and duration of specific services. Services were initiated comparatively promptly among infants receiving > level 1 care, with the median age 0.6 hours for initiation of CPAP and 4.8 hours for mechanical ventilation. Level 1 services were initiated at more varied ages, ranging from median 0.6 hours for supplemental oxygen administration without positive pressure to 52 hours for phototherapy or nasogastric tube feedings. Diagnoses for studied infants included surfactant deficiency (5.9%), transient tachypnea of the newborn (TTN, 22.6%), pneumonia (1.7%), hypoglycemia (20.6%), hyperbilirubinemia (30%), apnea of prematurity (8.1%), and poor feeding (15.6%). Overall 24 infants (5.5%) received antibiotics for longer than 3 days despite negative blood cultures. Hypoglycemia was diagnosed more frequently following initial triage to a level 1 setting (59/231, 25.5%) compared with NICU (30/200, 15%, p ¼ 0.007). In contrast, apnea of prematurity was diagnosed more frequently in infants triaged to NICU (31/200, 15.5%) compared with level 1 (4/231, 1.7%, p < 0.001). Other diagnoses made significantly more frequently following initial triage to NICU were surfactant deficiency, TTN, pneumonia, hyperbilirubinemia, and poor feeding.
Development of Triage Models ►Table 2 shows antenatal, intrapartum, and immediate postpartum characteristics associated with subsequent services in univariate analysis. Maternal or fetal conditions significantly associated in univariate analysis with subsequent level of services received by the infant were preeclampsia, maternal diabetes treated with insulin, preterm labor, and intrauterine growth restriction. Not significant were gender, maternal diabetes, maternal tobacco, or illicit drug use by history or screen, antenatal steroid exposure, intrapartum antibiotic administration, instrumented delivery, epidural anesthesia, placenta previa, abruption, pitocin induction or augmentation, breech presentation, prolonged rupture of membranes, meconium-stained amnionic fluid, repeat American Journal of Perinatology
Vol. 32
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
Results
Fig. 1 Initial triage of study infants from delivery room, and highest level of services subsequently received.
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
Table 1 Initial triage setting and subsequent services received during birth hospitalization Services received
Age at initiation of service (h)
Duration of service (h)
p Value
Initial care setting Level 1 (N ¼ 231)
NICU (N ¼ 200)
Routine care only—no level 1 or >level 1 services documented
138 (59.7)
26 (13)
level 1 services
73 (31.6)
85 (42.5)
level 1 services
20 (8.7)
89 (44.5)
level 1 services, and the great majority of these were identified in the immediate postpartum period. A majority of infants triaged to the NICU required only routine care or level 1 services. Hospitals may be overutilizing the NICU setting for these infants, with consequent increased LOS and costs. This multicenter study documents a need for
753
improved triage of 35-week infants and provides potential tools for this purpose. Further investigations are warranted to validate and refine these results.
Funding Source None. Conflict of Interest The authors have no conflicts to disclose. Clinical Trial Registry Not applicable.
References 1 Laptook AR. Neurologic and metabolic issues in moderately pre-
2
3 4
5 6
7
8 9
10
11
12
13
14
15 16
term, late preterm, and early term infants. Clin Perinatol 2013; 40(4):723–738 Meier P, Patel AL, Wright K, Engstrom JL. Management of breastfeeding during and after the maternity hospitalization for late preterm infants. Clin Perinatol 2013;40(4):689–705 Ramachandrappa A, Jain L. Health issues of the late preterm infant. Pediatr Clin North Am 2009;56(3):565–577 Young PC, Glasgow TS, Li X, Guest-Warnick G, Stoddard G. Mortality of late-preterm (near-term) newborns in Utah. Pediatrics 2007; 119(3):e659–e665 Engle WA. Morbidity and mortality in late preterm and early term newborns: a continuum. Clin Perinatol 2011;38(3):493–516 Hellmeyer L, Herz K, Liedtke B, Wohlmuth P, Schmidt S, Hackeloeer BJ. The underestimation of immaturity in late preterm infants. Arch Gynecol Obstet 2012;286(3):619–626 Medoff Cooper B, Holditch-Davis D, Verklan MT, et al. Newborn clinical outcomes of the AWHONN late preterm infant researchbased practice project. J Obstet Gynecol Neonatal Nurs 2012;41(6): 774–785 Phillips RM. Multidisciplinary guidelines for the care of late preterm infants. Introduction. J Perinatol 2013;33(Suppl 2):S3–S4 Ananth CV, Friedman AM, Gyamfi-Bannerman C. Epidemiology of moderate preterm, late preterm and early term delivery. Clin Perinatol 2013;40(4):601–610 Mathews TJ, Miniño AM, Osterman MJ, Strobino DM, Guyer B. Annual summary of vital statistics: 2008. Pediatrics 2011;127(1): 146–157 Shapiro-Mendoza CK, Tomashek KM, Kotelchuck M, Barfield W, Weiss J, Evans S. Risk factors for neonatal morbidity and mortality among “healthy,” late preterm newborns. Semin Perinatol 2006; 30(2):54–60 Phillips RM, Goldstein M, Hougland K, et al; National Perinatal Association. Multidisciplinary guidelines for the care of late preterm infants. J Perinatol 2013;33(Suppl 2):S5–S22 Academy of Breastfeeding Medicine. ABM clinical protocol #10: breastfeeding the late preterm infant (34(0/7) to 36(6/7) weeks gestation) (first revision June 2011). Breastfeed Med 2011;6(3): 151–156 Raju TN, Higgins RD, Stark AR, Leveno KJ. Optimizing care and outcome for late-preterm (near-term) infants: a summary of the workshop sponsored by the National Institute of Child Health and Human Development. Pediatrics 2006;118(3): 1207–1214 Raju TN. Moderately preterm, late preterm and early term infants: research needs. Clin Perinatol 2013;40(4):791–797 American Academy of Pediatrics Committee on Fetus And Newborn. Levels of neonatal care. Pediatrics 2012;130(3):587–597 American Journal of Perinatology
Vol. 32
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
Appropriate observation of LPI has been well described.13,14,22 The proper care setting for LPI should be based on medical problems apparent at birth, assessed potential for developing medical problems after birth, scope of care available, and ability of nursery personnel to recognize problems and intervene in a timely manner.8,23 Ongoing care should promote optimal physiological stability and reduce risk of inhospital morbidity.12 Thus, the appropriate care setting for LPI will vary among hospitals and the specific capabilities of their available care settings. This supports a need for tools such as the models in this study to aid triage based on likely service needs. Triage decisions during this study may have erred on the side of admitting to an NICU rather than to a level 1 setting as a response to literature emerging at the time regarding LPI. The landmark 2006 NICHD workshop report by Raju et al14 appeared mere months before the beginning of our study period, and as a result care provider awareness of the issues facing LPI was likely high during the period we studied. Furthermore, hospitals may have changed practices subsequently to accommodate the care needs of LPI, for example, by expanding the transitional care capabilities of their level 1 setting. As a result, triage of 35-week infants to an NICU setting may be less frequent currently than the results we report. However, it is less likely that clinical factors associated with a need for specific clinical services in 35-week infants have changed since the period studied. Further research is needed to document contemporary triage patterns for 35week infants, and to validate the models presented in this study. This retrospective study could not determine whether infants actually needed the services they received or received all the services they needed. For example, we could not ascertain whether infants received unnecessary nasogastric feedings, or developed unappreciated apnea. Furthermore, the thresholds for initiating services may have varied among sites, among care settings within sites, or among providers within care settings.24 Defining the impact of variation in care and discrepancies between care needs and care provided among LPI will require further investigation. EGA in this study was the best obstetric estimate at delivery. It is possible that some infants who had completed more or less than 35 weeks’ gestation were included, whereas some 35-week infants carrying a different EGA were not included. It is impossible to know the extent of such obstetric misclassification. However, this reflects usual conditions at the time of postpartum triage, thus increasing this study’s applicability and generalizability.
Hagadorn et al.
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
17 Stark AR; American Academy of Pediatrics Committee on Fetus and
21 Bérard A, Le Tiec M, De Vera MA. Study of the costs and morbidities
Newborn. Levels of neonatal care. Pediatrics 2004;114(5):1341–1347 18 Olsen IE, Groveman SA, Lawson ML, Clark RH, Zemel BS. New intrauterine growth curves based on United States data. Pediatrics 2010;125(2):e214–e224 19 Escobar GJ, Greene JD, Hulac P, et al. Rehospitalisation after birth hospitalisation: patterns among infants of all gestations. Arch Dis Child 2005;90(2):125–131 20 Tomashek KM, Shapiro-Mendoza CK, Davidoff MJ, Petrini JR. Differences in mortality between late-preterm and term singleton infants in the United States, 1995-2002. J Pediatr 2007;151(5): 450–456, 456.e1
of late-preterm birth. Arch Dis Child Fetal Neonatal Ed 2012;97(5): F329–F334 22 Adamkin DH. Feeding problems in the late preterm infant. Clin Perinatol 2006;33(4):831–837, abstract ix 23 Laptook A, Jackson GL. Cold stress and hypoglycemia in the late preterm (“near-term”) infant: impact on nursery of admission. Semin Perinatol 2006;30(1):24–27 24 Roblin DW, Richardson DK, Thomas E, et al. Variation in the use of alternative levels of hospital care for newborns in a managed care organization. Health Serv Res 2000;34(7): 1535–1553
Downloaded by: University of Pittsburgh. Copyrighted material.
754
American Journal of Perinatology
Vol. 32
No. 8/2015
747
Postpartum Triage, Services Provided, and Length of Stay for Infants Born At 35 Weeks Gestation James I. Hagadorn, MD, MS1,2 Saritha Salikooti, MD1,2 Mariann Pappagallo, MD1,2 Jose Arias-Camison, MD3 Scott Weiner, MD4 Jorge Alba, DO5 Victor Herson, MD1,2 1 Divison of Neonatology, Connecticut Children’s Medical Center,
Hartford, Connecticut 2 Department of Pediatrics, University of Connecticut School of Medicine, Farmington, Connecticut 3 Divison of Neonatology, St. Francis Hospital, Hartford, Connecticut 4 Divison of Neonatology, The Hospital of Central Connecticut, New Britain, Connecticut 5 Divison of Neonatology, Manchester Memorial Hospital, Manchester, Connecticut
Address for correspondence James I. Hagadorn, MD, MS, Connecticut Children’s Medical Center, 282 Washington Street, Hartford, CT 06106 (e-mail: [email protected]).
Abstract
Keywords
► ► ► ► ►
infant premature length of stay health resources late preterm
Objectives The aim of this article is to (1) compare the care setting to which 35-week infants are initially triaged postpartum to the level of services subsequently provided; and (2) identify factors known at delivery or immediately postpartum associated with services received and length of stay during the birth hospitalization. Study Design In this multicenter retrospective study of 35-week infants born between 2007 and 2008, service capabilities of the initial postpartum care setting were categorized as level 1 or neonatal intensive care unit (NICU) using American Academy of Pediatrics definitions. Subsequent services actually provided were categorized as routine care, level 1, or >level 1. Results Over half of 431 studied infants were sent to a level 1 nursery postpartum. Of these, over 90% ultimately received routine care or level 1 services. Of 200 infants triaged to a NICU, the majority received only routine care or level 1 services. The great majority of infants requiring > level 1 services were identified promptly postpartum. Initial triage to the NICU was associated with significantly (p < 0.05) increased length of stay despite provision of similar services. Conclusions This study suggests a need for improved triage of 35-week infants and provides tools for this purpose. Validation of the models presented here is warranted.
Clinical Perspective • Among 35-week gestation infants triaged from delivery room to the NICU, a majority received only routine care or level 1 services. • Compared with initial triage to level 1 nursery, triage to the NICU was associated with increased length of stay despite provision of similar services. • Hospitals may be overutilizing the NICU setting for these infants, increasing length of stay, infant–parent separation, and costs.
received June 5, 2014 accepted after revision October 1, 2014 published online December 23, 2014
• This multicenter study documents a need for improved triage of 35-week infants and provides potential tools for this purpose. Postpartum care needs of late preterm infants (LPI)—those born 340/7 to 366/7 weeks’ gestation—vary considerably. Compared with term, LPI are at increased risk for postpartum hypoglycemia, respiratory distress, hyperbilirubinemia, or ineffective feeding.1–7 The public health impact of these morbidities is large, as LPI represent nearly 10% of U.S. births and three-quarters of this country’s premature deliveries.8–10
Copyright © 2015 by Thieme Medical Publishers, Inc., 333 Seventh Avenue, New York, NY 10001, USA. Tel: +1(212) 584-4662.
DOI http://dx.doi.org/ 10.1055/s-0034-1396689. ISSN 0735-1631.
Downloaded by: University of Pittsburgh. Copyrighted material.
Am J Perinatol 2015;32:747–754.
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
Cumulative incidence of short-term postpartum morbidity in LPI decreases from over 50% at 34 weeks to approximately 10% at 36 weeks.11 Consequently, LPI present a dilemma for providers choosing a postpartum neonatal care setting. Published guidelines encourage each delivery service to determine where observation can best be accomplished while simultaneously anticipating infant care needs and supporting the mother–infant dyad.12,13 However, tools to assist this potentially difficult determination are few. In our community, an informal consensus has developed in which infants at 36 weeks’ gestation are generally observed initially in a level 1 nursery, while 34-week infants are thought better served by transitioning in a level 2, level 3, or level 4 neonatal intensive care unit (NICU) setting. For infants born 350/7 to 356/7 weeks no such consensus or common protocol has emerged. Instead, 35-week infants are triaged on a case-bycase basis. In the absence of a clear indication for intensive care, such triage decisions may present a challenge. The National Institute of Child Health and Human Development panel’s call for research to improve precision of risk assessment in LPI subgroups has recently been reiterated.14,15 Likewise, the American Academy of Pediatrics (AAP) has suggested that more evidence is needed to determine outcomes for LPI by level of care.16 The purpose of this study was to examine postpartum triage of infants born at 35 weeks’ gestation. We wished to (1) compare the capabilities of settings to which 35-week infants are initially triaged postpartum to services subsequently provided and (2) identify factors known at delivery or immediately postpartum associated with services received and length of stay during the birth hospitalization. We hypothesized that discrepancies exist between the care capabilities of the initial care setting and subsequent services received, and that identifiable prenatal, intrapartum, and immediate postpartum factors are associated with subsequent services received and length of neonatal hospital stay (LOS).
Subjects and Methods This was a retrospective record review at five hospitals, each with a level 1 normal newborn nursery and a higher level (2, 3, or 4) NICU.16 Institutional Review Board approval was obtained at all sites. Participating hospitals—Hartford Hospital/Connecticut Children’s Medical Center Hartford NICU, University of Connecticut Health Center/Connecticut Children’s Farmington NICU, St. Francis Hospital, Hospital of Central Connecticut, and Manchester Memorial Hospital— included all facilities in the Hartford metropolitan area with both a delivery service and an NICU. At each hospital, initial triage of 35-week gestation infants following delivery was performed on a case-by-case basis by the hospitalist, neonatal nurse practitioner, neonatal fellow, or neonatologist attending the delivery. Triage decisions were made promptly following delivery if necessary, or after 1 to 2 hours of postpartum observation if the baby’s condition allowed. When warranted, such observation could include maternal–infant skin-to-skin care, measurement of temperature and blood glucose, serial respiratory assessment, measureAmerican Journal of Perinatology
Vol. 32
No. 8/2015
ment of oxygen saturation with pulse oximetry, and attempted feeding. Obstetric, neonatal, and nurse practitioner staffs for the hospitals were separate.
Subjects All studied infants were born from January 1, 2007 to December 31, 2008 with estimated gestational age (EGA) between 350/7 and 356/7 weeks. Infants were screened for inclusion using delivery room, nursery, and NICU logs. Infants were excluded if they required intubation, chest compressions or medications in the delivery room, for congenital anomalies requiring NICU admission, or for grossly incomplete records. Data regarding prenatal and intrapartum diagnoses and interventions, delivery room interventions, and the postnatal hospital course were collected from individual medical record reviews. Mothers were classified as diabetic if given that diagnosis in their obstetric record, and status regarding insulin therapy at the time of delivery was recorded. Presence of prenatal labor was as diagnosed by the obstetric care team. Date and time of birth, of initiation of specific treatments, and of discharge were recorded. LOS and age at initiation of treatment in hours were calculated. Highest level of services each infant received was determined using 2004 AAP definitions.17 Services classified as level 1 care included phototherapy, intravenous fluids or antibiotics, supplemental oxygen without positive pressure, and nasogastric feeding. Positive pressure via continuous positive airway pressure (CPAP), mechanical ventilation, or any need for surgery were classified as >level 1 services. Thus, if a 35-week infant was admitted to neonatal intensive care but nasogastric feeding was the highest level service provided, the infant was categorized as triaged to NICU but receiving only level 1 services. Infants requiring no services in the level 1 or >level 1 categories were classified as receiving routine care only. Infants were classified as appropriate, large (>90th percentile) or small (level 1 services during the birth
Downloaded by: University of Pittsburgh. Copyrighted material.
748
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
749
hospitalization (outcome) versus routine care only (model I), and provision of >level 1 services (outcome) versus routine care/level 1 (model II). Model performance was assessed by calculating sensitivity, specificity, positive predictive value, negative predictive value, and C statistic (area beneath the receiver-operator characteristic curve). Model calibration was assessed with the Hosmer–Lemeshow goodness-of-fit test and by contrasting observed and predicted rates of outcome within subgroups. Regression analyses were repeated using hierarchical linear or logistic regression to adjust for clustering of multiples within pregnancies where appropriate.
Screening hospital logs yielded 462 births at EGA 350/7 to 356/ 7 weeks during the study period. Of the 462 births 15 were subsequently excluded, as EGA on record review was not 35 weeks at the time of delivery. Six were excluded for congenital malformations and ten for receiving cardiopulmonary resuscitation in the delivery room. Of the 462 infants, 431 35-week infants from 384 pregnancies were studied. EGA was based on first trimester ultrasound in 58% of infants (n ¼ 248), date of implantation of technology-assisted pregnancy in 11% (n ¼ 46), date of last menstrual period in 8% (n ¼ 33), 2nd trimester ultrasound in 6% (n ¼ 25), 3rd trimester ultrasound in 2% (n ¼ 10), transvaginal ultrasound in 2% (n ¼ 9), and lacked a clear basis in the maternal record in 14% (n ¼ 60). Overall 11% (n ¼ 44) of mothers had diabetes mellitus. Deliveries were 53% vaginal (n ¼ 230), 34% primary cesarean section (n ¼ 147), and 13% repeat cesarean section (n ¼ 54). A total of 75 infants (17%) were delivered by cesarean section without labor. Studied infants had a mean birth weight 2,413 442 g, with 50% (n ¼ 218) white, 20% (n ¼ 84) black, 19% (n ¼ 80) Hispanic, and 50% (n ¼ 229) female. One-fifth (n ¼ 93) of infants received antenatal steroids, whereas 17% (n ¼ 72) were exposed to intrapartum magnesium sulfate. There were no positive blood cultures, surgeries, or deaths among study infants during their birth hospitalization.
Postpartum Triage and Subsequent Services Received ►Fig. 1 shows the initial triage setting and subsequent services received for the entire study cohort. Over one-third of infants (n ¼ 164) received only routine care without the need for any specific level 1 or > level 1 services. The highest level service received was 1 for 37% (n ¼ 158) and >1 for 25% (n ¼ 109). Of the 200 infants initially triaged to NICU, less than half received > level 1 services, whereas the majority received either routine care or level 1 services only. Over half (n ¼ 231) of study infants were initially sent to a level 1 setting following delivery. Of these, the great majority subsequently received only routine care or level 1 services, whereas less than one-tenth ultimately received > level 1 services. All sites triaged 35-week infants to both level 1 and NICU. The proportion triaged to level 1 varied significantly among sites, ranging from 38.5 to 68.5% (p < 0.001, chi-square test). Provision of > level 1 services did not differ significantly
among sites, ranging from 19.1 to 35.9% (p ¼ 0.106, chisquare test). ►Table 1 presents highest level of services received following initial delivery room triage, together with age at initiation and duration of specific services. Services were initiated comparatively promptly among infants receiving > level 1 care, with the median age 0.6 hours for initiation of CPAP and 4.8 hours for mechanical ventilation. Level 1 services were initiated at more varied ages, ranging from median 0.6 hours for supplemental oxygen administration without positive pressure to 52 hours for phototherapy or nasogastric tube feedings. Diagnoses for studied infants included surfactant deficiency (5.9%), transient tachypnea of the newborn (TTN, 22.6%), pneumonia (1.7%), hypoglycemia (20.6%), hyperbilirubinemia (30%), apnea of prematurity (8.1%), and poor feeding (15.6%). Overall 24 infants (5.5%) received antibiotics for longer than 3 days despite negative blood cultures. Hypoglycemia was diagnosed more frequently following initial triage to a level 1 setting (59/231, 25.5%) compared with NICU (30/200, 15%, p ¼ 0.007). In contrast, apnea of prematurity was diagnosed more frequently in infants triaged to NICU (31/200, 15.5%) compared with level 1 (4/231, 1.7%, p < 0.001). Other diagnoses made significantly more frequently following initial triage to NICU were surfactant deficiency, TTN, pneumonia, hyperbilirubinemia, and poor feeding.
Development of Triage Models ►Table 2 shows antenatal, intrapartum, and immediate postpartum characteristics associated with subsequent services in univariate analysis. Maternal or fetal conditions significantly associated in univariate analysis with subsequent level of services received by the infant were preeclampsia, maternal diabetes treated with insulin, preterm labor, and intrauterine growth restriction. Not significant were gender, maternal diabetes, maternal tobacco, or illicit drug use by history or screen, antenatal steroid exposure, intrapartum antibiotic administration, instrumented delivery, epidural anesthesia, placenta previa, abruption, pitocin induction or augmentation, breech presentation, prolonged rupture of membranes, meconium-stained amnionic fluid, repeat American Journal of Perinatology
Vol. 32
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
Results
Fig. 1 Initial triage of study infants from delivery room, and highest level of services subsequently received.
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
Table 1 Initial triage setting and subsequent services received during birth hospitalization Services received
Age at initiation of service (h)
Duration of service (h)
p Value
Initial care setting Level 1 (N ¼ 231)
NICU (N ¼ 200)
Routine care only—no level 1 or >level 1 services documented
138 (59.7)
26 (13)
level 1 services
73 (31.6)
85 (42.5)
level 1 services
20 (8.7)
89 (44.5)
level 1 services, and the great majority of these were identified in the immediate postpartum period. A majority of infants triaged to the NICU required only routine care or level 1 services. Hospitals may be overutilizing the NICU setting for these infants, with consequent increased LOS and costs. This multicenter study documents a need for
753
improved triage of 35-week infants and provides potential tools for this purpose. Further investigations are warranted to validate and refine these results.
Funding Source None. Conflict of Interest The authors have no conflicts to disclose. Clinical Trial Registry Not applicable.
References 1 Laptook AR. Neurologic and metabolic issues in moderately pre-
2
3 4
5 6
7
8 9
10
11
12
13
14
15 16
term, late preterm, and early term infants. Clin Perinatol 2013; 40(4):723–738 Meier P, Patel AL, Wright K, Engstrom JL. Management of breastfeeding during and after the maternity hospitalization for late preterm infants. Clin Perinatol 2013;40(4):689–705 Ramachandrappa A, Jain L. Health issues of the late preterm infant. Pediatr Clin North Am 2009;56(3):565–577 Young PC, Glasgow TS, Li X, Guest-Warnick G, Stoddard G. Mortality of late-preterm (near-term) newborns in Utah. Pediatrics 2007; 119(3):e659–e665 Engle WA. Morbidity and mortality in late preterm and early term newborns: a continuum. Clin Perinatol 2011;38(3):493–516 Hellmeyer L, Herz K, Liedtke B, Wohlmuth P, Schmidt S, Hackeloeer BJ. The underestimation of immaturity in late preterm infants. Arch Gynecol Obstet 2012;286(3):619–626 Medoff Cooper B, Holditch-Davis D, Verklan MT, et al. Newborn clinical outcomes of the AWHONN late preterm infant researchbased practice project. J Obstet Gynecol Neonatal Nurs 2012;41(6): 774–785 Phillips RM. Multidisciplinary guidelines for the care of late preterm infants. Introduction. J Perinatol 2013;33(Suppl 2):S3–S4 Ananth CV, Friedman AM, Gyamfi-Bannerman C. Epidemiology of moderate preterm, late preterm and early term delivery. Clin Perinatol 2013;40(4):601–610 Mathews TJ, Miniño AM, Osterman MJ, Strobino DM, Guyer B. Annual summary of vital statistics: 2008. Pediatrics 2011;127(1): 146–157 Shapiro-Mendoza CK, Tomashek KM, Kotelchuck M, Barfield W, Weiss J, Evans S. Risk factors for neonatal morbidity and mortality among “healthy,” late preterm newborns. Semin Perinatol 2006; 30(2):54–60 Phillips RM, Goldstein M, Hougland K, et al; National Perinatal Association. Multidisciplinary guidelines for the care of late preterm infants. J Perinatol 2013;33(Suppl 2):S5–S22 Academy of Breastfeeding Medicine. ABM clinical protocol #10: breastfeeding the late preterm infant (34(0/7) to 36(6/7) weeks gestation) (first revision June 2011). Breastfeed Med 2011;6(3): 151–156 Raju TN, Higgins RD, Stark AR, Leveno KJ. Optimizing care and outcome for late-preterm (near-term) infants: a summary of the workshop sponsored by the National Institute of Child Health and Human Development. Pediatrics 2006;118(3): 1207–1214 Raju TN. Moderately preterm, late preterm and early term infants: research needs. Clin Perinatol 2013;40(4):791–797 American Academy of Pediatrics Committee on Fetus And Newborn. Levels of neonatal care. Pediatrics 2012;130(3):587–597 American Journal of Perinatology
Vol. 32
No. 8/2015
Downloaded by: University of Pittsburgh. Copyrighted material.
Appropriate observation of LPI has been well described.13,14,22 The proper care setting for LPI should be based on medical problems apparent at birth, assessed potential for developing medical problems after birth, scope of care available, and ability of nursery personnel to recognize problems and intervene in a timely manner.8,23 Ongoing care should promote optimal physiological stability and reduce risk of inhospital morbidity.12 Thus, the appropriate care setting for LPI will vary among hospitals and the specific capabilities of their available care settings. This supports a need for tools such as the models in this study to aid triage based on likely service needs. Triage decisions during this study may have erred on the side of admitting to an NICU rather than to a level 1 setting as a response to literature emerging at the time regarding LPI. The landmark 2006 NICHD workshop report by Raju et al14 appeared mere months before the beginning of our study period, and as a result care provider awareness of the issues facing LPI was likely high during the period we studied. Furthermore, hospitals may have changed practices subsequently to accommodate the care needs of LPI, for example, by expanding the transitional care capabilities of their level 1 setting. As a result, triage of 35-week infants to an NICU setting may be less frequent currently than the results we report. However, it is less likely that clinical factors associated with a need for specific clinical services in 35-week infants have changed since the period studied. Further research is needed to document contemporary triage patterns for 35week infants, and to validate the models presented in this study. This retrospective study could not determine whether infants actually needed the services they received or received all the services they needed. For example, we could not ascertain whether infants received unnecessary nasogastric feedings, or developed unappreciated apnea. Furthermore, the thresholds for initiating services may have varied among sites, among care settings within sites, or among providers within care settings.24 Defining the impact of variation in care and discrepancies between care needs and care provided among LPI will require further investigation. EGA in this study was the best obstetric estimate at delivery. It is possible that some infants who had completed more or less than 35 weeks’ gestation were included, whereas some 35-week infants carrying a different EGA were not included. It is impossible to know the extent of such obstetric misclassification. However, this reflects usual conditions at the time of postpartum triage, thus increasing this study’s applicability and generalizability.
Hagadorn et al.
Postpartum Triage of the 35-Week Infant
Hagadorn et al.
17 Stark AR; American Academy of Pediatrics Committee on Fetus and
21 Bérard A, Le Tiec M, De Vera MA. Study of the costs and morbidities
Newborn. Levels of neonatal care. Pediatrics 2004;114(5):1341–1347 18 Olsen IE, Groveman SA, Lawson ML, Clark RH, Zemel BS. New intrauterine growth curves based on United States data. Pediatrics 2010;125(2):e214–e224 19 Escobar GJ, Greene JD, Hulac P, et al. Rehospitalisation after birth hospitalisation: patterns among infants of all gestations. Arch Dis Child 2005;90(2):125–131 20 Tomashek KM, Shapiro-Mendoza CK, Davidoff MJ, Petrini JR. Differences in mortality between late-preterm and term singleton infants in the United States, 1995-2002. J Pediatr 2007;151(5): 450–456, 456.e1
of late-preterm birth. Arch Dis Child Fetal Neonatal Ed 2012;97(5): F329–F334 22 Adamkin DH. Feeding problems in the late preterm infant. Clin Perinatol 2006;33(4):831–837, abstract ix 23 Laptook A, Jackson GL. Cold stress and hypoglycemia in the late preterm (“near-term”) infant: impact on nursery of admission. Semin Perinatol 2006;30(1):24–27 24 Roblin DW, Richardson DK, Thomas E, et al. Variation in the use of alternative levels of hospital care for newborns in a managed care organization. Health Serv Res 2000;34(7): 1535–1553
Downloaded by: University of Pittsburgh. Copyrighted material.
754
American Journal of Perinatology
Vol. 32
No. 8/2015
