OBJECTIVE: The American Academy of Pediatrics recommends observation of preterm and at-risk infants in their car seat before hospital discharge to screen for breathing problems. This observation period, which we refer to as the infant car seat challenge (ICSC), is used to determine readiness for travel in a car seat. Infants who fail the ICSC are recommended to travel in a car bed. Limited data exist to identify risk factors for failing the ICSC, and no guidelines are available to facilitate transition from car beds to car seats. The purpose of this study was to determine whether weight and age are predictors for passing the ICSC.
METHODS: This retrospective study evaluated 43 infants referred to the Children's Hospital Boston Center for Healthy Infant Lung Development for a repeat ICSC after initial failure and 37 infants who passed or failed their initial ICSC at the Caritas Saint Elizabeth's Medical Center. Gender, birth weight, gestational age (GA), weight (ICSC weight), corrected GA, and chronological age at time of ICSC were extracted, and logistic regression analysis was performed.
RESULTS: The average GA at birth of infants referred was 35 weeks 2 days (±2 weeks 3 days), and almost equal numbers of boys and girls were referred. A majority of infants passed their initial rechallenge (38 [88%] of 43). Infants who failed the repeat challenge were slightly smaller (3327 ± 927 vs 3913 ± 936 g) and younger at time of retesting (CGA 40 weeks 5 days vs 42 weeks 5 days; chronological age 39.2 vs 52.2 days). Neither weight nor age at initial or repeat ICSC predicted passing the ICSC.
CONCLUSIONS: This study suggests that ICSC screenings remain the safest method for transitioning preterm infants from a car bed to a car seat.
WHAT'S KNOWN ON THIS SUBJECT:
Preterm and infants at risk are screened for breathing problems in car seats. Infants who fail the screen are discharged from the hospital in car beds. There are no guidelines for transitioning infants from a car bed to a car seat.
WHAT THIS STUDY ADDS:
This is the first study to follow infants who failed their initial car seat screen and were rescreened after hospital discharge. We sought to determine an age or weight that predicts passage of the initial and repeat car seat screen.
The American Academy of Pediatrics (AAP) recommends that preterm and at-risk newborns be observed for breathing problems for a period of time in their car seat, before hospital discharge.1,–,3 There is no universally accepted term for this observation period; therefore, for the purpose of clarity, we refer to this observation period as the infant car seat challenge (ICSC). In response to the AAP's recommendation, the ICSC has become a routine predischarge screening procedure, and many infants nationally and internationally have been screened by some form of ICSC.4,5
Of the >4 million births in the United States each year, almost 13% (>500 000 infants) are preterm (<37 weeks' gestation), making them eligible for the recommended screening by ICSC.6 Estimates of compliance with the AAP guidelines for screening suggest that close to 90% of hospitals are performing recommended predischarge ICSC, although details of the exact testing process vary.7 Estimates of failed ICSCs range from 12% to 26%8,–,11; this suggests that several thousand infants in the United States currently fail the ICSC each year.
In our center, infants who have breathing problems in a traditional car seat are subsequently screened for respiratory stability in the only available alternative for automobile transport of infants, a car bed that meets Federal Motor Vehicle Safety Standards 213. This is because studies have indicated that even some healthy term infants can experience oxygen desaturation while positioned in a car bed.12,13 We believe, in our institution, that preterm infants also require an observation period in their car bed because they may be at even greater risk for respiratory instability than term infants as a result of their fragile respiratory status. Furthermore, screening of infants in car beds is consistent with the most current AAP guidelines.3 Infants who experience oxygen desaturations in a car bed are recommended to undergo additional diagnostic evaluation for cardiorespiratory issues, whereas infants who demonstrate respiratory stability in a car bed are recommended to use this type of restraint for travel in automobiles.
Despite the large number of infants who are discharged from the hospital in car beds each year, guidelines for transition from a car bed back to a car seat are unavailable; therefore, we conducted a survey of pediatricians in Massachusetts to determine the practice patterns of pediatricians who transition infants from a car bed to a car seat. From this survey, we found that management strategies for transitioning infants from car beds to car seats varied dramatically among the respondents, and this survey highlighted problems with discharging preterm and newborns who are at risk in car beds. One major problem cited by the pediatricians was that most hospitals do not have in place a system to retest infants after hospital discharge, leaving few options for retesting. The pediatricians also indicated that there was a lack of resources such as equipment and staff available to perform a repeat ICSC. Furthermore, many pediatricians desired that a hospital-based or office-based car seat challenge follow-up program that employs clinicians with expert knowledge about the ICSC and infant motor vehicle restraints perform the follow-up ICSC. The findings from our survey study supported the establishment of the Infant Car Seat Challenge Follow-up Program in our institution.
With the development of the Infant Car Seat Challenge Follow-up Program, we sought to identify accurate and sensitive predictors for passing the ICSC and thereby eliminate the need for cumbersome and potentially costly ICSC screens. This study was designed to identify patient characteristics that predict passing the ICSC.
The institutional review boards of Children's Hospital Boston and Caritas St Elizabeth's Medical Center (CSEMC) approved this study. A retrospective chart review was performed at both facilities. A total of 43 infants who were referred to the Children's Hospital Boston Center for Healthy Infant Lung Development (CHILD Clinic) between August 2008 and March 2009 for repeat ICSC after initial failure at the discharging facility were identified. In addition, 37 infants who were born at the CSEMC between January 1, 2008, and March 1, 2009, at gestational age (GA) <37 weeks and were screened by the ICSC before hospital discharge were identified. Gender, birth weight, weight at ICSC, mode of delivery, GA, corrected GA (CGA), and chronological age (CA) at the time of the initial and repeat car seat challenge (as applicable) and passage or failure of the ICSC were extracted from the medical records and entered into our statistical database. Descriptive and logistic regression analyses were used to describe our population and to assess the association among infant characteristics (gender, birth weight, weight at ICSC, GA, CGA, and CA) and ICSC results.
At CSEMC, the ICSC is performed by using the Hewlett Packard Model 54, M1175A cardiorespiratory monitor and oximeter in the child's individual car seat and lasts for 90 minutes. The bedside nurse performed the ICSCs, and failure of the ICSCs at CSEMC included any episode of apnea that lasted 20 seconds, bradycardia with heart rate <80 beats per minute or 10% below baseline, or oxygen saturation below 90% for >20 seconds.
At the CHILD Clinic, infants were also screened in their car seats for 90 minutes. A Masimo oximeter recorded oxygen saturation and heart rate. A pulmonary function–testing technician remained within 10 ft of the infant throughout the study and responded immediately to any alarms to assess patient motion or inadequate stability of the probe. Criteria for failure of the ICSC included any episodes of desaturation of <90% that lasted >20 seconds or bradycardia with heart <80 or 10% below baseline.
Statistical analyses were performed with SAS 9.1 (SAS Institute, Cary, NC). Categorical variables were analyzed by Fisher's exact tests. Because of a small sample size, Mann-Whitney nonparametric tests were used to evaluate continuous data. Significance was defined as P < .05.
Logistic regression analysis was used to identify predictors for passing the initial ICSC and repeat test after initial failure. Independent variables that were entered into the regression models included mean CGA and weight at repeat ICSC. These variables were selected because pediatricians typically use CGA and weight as criteria for transitioning infants in place of repeating the ICSC or in conjunction with a repeat ICSC when determining a safe time to transition infants from a car bed to a car seat. For the initial ICSC, CA at testing was statistically significant in univariate analysis (P < .05) so it was also included in the models. Effects are reported as odds ratios and 95% confidence intervals.
Forty-three infants were referred for repeat ICSC testing because they failed their initial ICSC. An additional 37 infants were screened for their initial ICSC at CSEMC, and 19% failed (n = 7). Characteristics of the resulting 50 infants who failed their initial ICSC and the 30 infants who passed their initial ICSC are presented in Table 1. None of the children who were referred for repeat testing had significant comorbidities (eg, intraventricular hemorrhage, gastroesophageal reflux).
Predictors of Passing the ICSC
In univariate analysis, only CA at time of testing was associated with passing the initial ICSC. Other variables including gender, birth weight, weight at ICSC, GA, and CGA at time of ICSC were not statistically significant.
Logistic regression analysis was performed to assess the effect of several potential predictors of passing the ICSC. In the initial analysis, we used the means for CGA, weight at ICSC, and CA. In a second analysis, we used the medians for these variables; medians were used because of the small sample sizes. CGA, weight at ICSC, and CA were not significant predictors for passing the ICSC according to these models (Table 2).
Predictors of Passing the Repeat ICSC
Of the 43 infants who were retested in the CHILD Clinic, 38 (88%) passed on their initial retest. Of the remaining 5, 3 passed on the second retest 2 to 4 weeks later and 2 did not return for repeat testing. A comparison of the infants who passed the initial retest and those who failed is shown in Table 3.
In logistic regression analysis, again we used both the mean and median values for CGA, weight at ICSC, and CA in our 2 models. Again, the 2 models showed that CGA, weight at ICSC, and CA were not significant predictors for passing the ICSC (Table 4).
Car seats are essential for preventing death or injury to newborns and infants during motor vehicle travel14; however, most car safety seats are designed for the “average” term newborn infant, so many preterm infants who are discharged from the hospital may face specific issues when using standard infant car safety seats. Specifically, they may be prone to respiratory compromise. Early studies noted that preterm infants who were otherwise ready for hospital discharge had frequent episodes of hypoxia during a 90-minute observation period, and the incidence of hypoxia was inversely related to birth weight and GA at birth.10,15,16 Since then, other at-risk populations of infants have been identified.17,18 Furthermore, we have found from our experience in the CHILD Clinic and from previous investigation that many infants who fail the ICSC are late preterm infants who were born between 35 and 37 weeks' GA.8 Although no long-term studies have determined the prognostic implications of the intermittent oxygen desaturation described in infants who fail the ICSC, the adverse effects of intermittent hypoxia from other disease processes have been well-described19; therefore, the observation that some preterm infants and infants who are at risk develop breathing problems in standard car safety seats has resulted in the need for informed recommendations regarding the proper screening procedures and equipment that is needed for transporting these infants.
The preterm birth rate has remained constant at nearly 13% for the past several years in the United States, translating to an estimated 500 000 preterm infants who are eligible for ICSCs.6 The exact number of infants who fail their ICSC is not known, although 1 single distributor of infant car beds in the United States reported to have sold >4000 car beds in 2009 (private communication). On the basis of estimates of prevalence of ICSC testing and published reports of ICSC failure rates,4,10 it is likely that >20 000 infants in the United States fail their ICSC each year.
Many neonatal units and special care nurseries that discharge preterm infants in the United States have incorporated ICSCs into their routine discharge assessment for preterm infants; however, some late preterm infants may not require special care or intensive care after birth. This is concerning because only 22% of well-infant nurseries (level I) performed ICSCs according to 1 published report, whereas frequencies reported for specialized newborn care nurseries was much higher (81% for level III and 91% for level II).7
The AAP's recommendation to perform some form of ICSC does not include specific guidelines for screening infants; therefore, as may be expected, there is wide variation between units in the duration of observation and monitoring and the criteria for passing the test. Consistent with the recently published AAP guidelines,3 our CHILD Clinic studies children for 90 minutes and records all oximetry data continuously to eliminate artifact and to obtain a minimum of 60 minutes of interpretable data.
In addition to the variability in the testing procedures, there is considerable variation in the recommendations given to parents of infants who fail the ICSC. Some advocate for the use of inserts to provide additional postural support, although the use of these supports does not eliminate oxygen desaturation events.20,21 Others delay hospital discharge until the infant passes the test, but prolonging hospitalization can add to the financial and emotional costs for the family. The 2009 AAP guidelines recommend use of a car bed that meets Federal Motor Vehicle Safety Standards 213 and that allows an infant to travel supine rather than upright, eliminating slouching that has been hypothesized to cause the observed compromise in respiratory function. In our center, we repeat the ICSC in the car bed to document resolution of respiratory compromise, because rarely a nonpostural cause of respiratory compromise may be determined. We have received anecdotal reports of infants who received a diagnosis of congenital heart disease after a failed ICSC; in addition, we have detected a pneumothorax in an otherwise healthy infant who was brought to the NICU for an ICSC.
Car beds occupy more space than a traditional car seat, and the use of a car bed may create anxiety in families who perceive the assignment of their child to a different mode of transportation as a form of stigmatization. Most parents in our clinic have expressed intense desire to transition from the car bed to the car seat, but guidelines to do so safely are not available. Regardless, infant car seat safety experts and the AAP agree that before discontinuing the use of a car bed, a follow-up study should be arranged to determine whether the infant can travel safely in a semireclined position without apnea, bradycardia, or oxygen desaturation.3
To our knowledge, this is the first study to describe the follow-up evaluation of infants who failed their initial ICSC. In a previous survey study of primary care pediatricians in Massachusetts, we noted that a majority of providers used some form of weight or age to guide their decision about when to transition from car bed to car seat. We hoped that the analysis of these infants would provide a minimum age or weight that predicted successful passage of the ICSC, eliminating the need for repeat testing. This study demonstrates that a majority (88%) of infants do pass the ICSC when performed within 5 to 8 weeks of the initial test. We noted that neither weight nor age seemed to be predictive factors for passing the ICSC. Unfortunately, our analysis suggests that repeat testing is the only way to determine definitively whether an individual infant passes or fails and therefore is ready to transition to the car seat.
Our study has several important limitations. First, this study was retrospective and potential predictors were assessed by medical chart review. Second, our testing method included recorded oximetries and heart rates but did not include impedance apnea monitoring; however, our review of all of the records of infants who failed initial ICSCs with impedance monitoring in place demonstrated that no child failed from isolated apnea without associated desaturation or bradycardia. Consistent with this finding, the literature describes that documented ICSC failures are primarily attributable to desaturations and rarely, if ever, to apnea. Third, 1 of our criteria for failure was an oxygen saturation of <90% for >20 seconds. It has been suggested that an oxygen saturation of <93% may be more appropriate.19,22 We chose to use an oxygen saturation of <90% as our criterion for several reasons:
In a previous study, we found that it is rare for an infant to fail an ICSC when the oxygen saturation range is 90 to 92 and that most infants failed their ICSC with profoundly low oxygen levels.8
Our inpatient unit and many of the surrounding health care facilities use <90% as criterion for failure; we wanted to use a consistent criterion so as not to confuse parents.
Our criterion includes a provision that permits the clinician to assign a failed status if the infant demonstrates increased work of breathing or exhibits frequent brief oxygen desaturation events even when they do not meet the explicit criteria for failure. We believe that this provision adds a level of protection for the infant by enabling the clinician to use observation skills to help detect infants who are having trouble in their car seat even when they meet the explicit criteria for a pass.
The minimal levels for target mean baseline saturation at all times and the minimal level to use for a single desaturation episode during a 90-minute screening test may not need to be identical. Additional study into the significance of intermittent brief episodes of desaturation versus longer or more frequent episodes of desaturation is certainly warranted.
Fourth, another possible predictor of ICSC failure, head control, was not assessed in this study. In our survey of pediatricians in Massachusetts that we used to guide this study, head control was not frequently used to transition infants from a car bed to a car seat; however, we believe that there are a number of potential risk factors for car seat–related oxygen desaturation that warrant additional study, such as head control and sleep state.23,24 Investigation of head control and sleep state requires sensitive and highly accurate measurement techniques; previous studies lacked these sensitive measures, and therefore associations between sleep, head control, and car seat–related oxygen desaturation have not been confirmed.23,24
Fifth, this study used a convenience sample of children who failed an initial ICSC and were referred to the clinic; however, review of the infants who failed initial testing but were not referred to our clinic had similar clinical characteristics to those who were referred. Finally, although this is the largest description of infants who have failed the ICSC, the sample size may be too small to detect predictors successfully. Standardization of the ICSC would enable tracking of characteristics of infants who pass and fail the ICSC on a larger scale and would potentially assist in identifying risk factors for ICSC failure.
We identified a cohort of children who failed their initial ICSC during an 8-month period. We were unable to identify specific minimum age or weight that allowed prediction for passing an initial or subsequent ICSC. A larger cohort of infants who fail the ICSC may allow successful identification of predictors for passing the ICSC. Until predictors are identified, follow-up ICSC testing is an important service, because it can provide additional information to assist the clinician in determining appropriate timing for safe transition of infants from car beds to car seats.
- Accepted September 11, 2009.
- Address correspondence to Michele DeGrazia, PhD, NNP-BC, Children's Hospital Boston, Neonatal Intensive Care Unit, 7-North NICU, 300 Longwood Ave, Boston, MA 02115. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- AAP =
- American Academy of Pediatrics •
- ICSC =
- infant car seat challenge •
- CSEMC =
- Caritas St Elizabeth's Medical Center •
- CHILD Clinic =
- Center for Healthy Infant Lung Development •
- GA =
- gestational age •
- CGA =
- corrected gestational age •
- CA =
- chronological age
- 1.↵American Academy of Pediatrics. Committee on Injury and Poison Prevention and Committee on Fetus and Newborn. Safe transportation of premature and low birth weight infants. Pediatrics. 1996;97(5):758–769
- Bull M,
- Agran P,
- Laraque D,
- et al
- Bull MJ,
- Engle WA,
- Committee on Injury, Violence, and Poison Prevention and Committee on Fetus and Newborn; American Academy of Pediatrics
- Mullen D,
- Courts J
- 5.↵Fetus and Newborn Committee Canadian Paediatric Society. Assessment of babies for car seat safety before hospital discharge. Paediatric Child Health. 2000;5(1):53–56
- 6.↵Centers for Disease Control. National Vital Statistics System, Birth Data;2009
- Merchant JR,
- Worwa C,
- Porter S,
- Coleman JM,
- deRegnier RO
- Young B,
- Shapira S,
- Finer NN
- Cerar LK,
- Scirica CV,
- Gantar IS,
- Osredkar D,
- Neubauer D,
- Kinane TB
- Kinane TB,
- Murphy J,
- Bass JL,
- Corwin MJ
- 14.↵National Highway Traffic Safety Administration. Child Passenger Safety Fact Sheet. Washington, DC: National Highway Traffic Safety Administration; 2008
- Bass JL,
- Mehta KA,
- Camara J
- Bass JL,
- Mehta KA
- Bass JL,
- Corwin M,
- Gozal D,
- et al
- Tonkin SL,
- McIntosh CG,
- Hadden W,
- Dakin C,
- Rowley S,
- Gunn A
- DeGrazia M
- Copyright © 2010 by the American Academy of Pediatrics