Page 43 - JSOM Summer 2023

with pelvic circumference to support alternative guidance for Quantitative variables were recorded as mean ± standard devi-
prehospital providers. ation. Statistical differences in anthropometric measurements
between males and females were calculated using an unpaired
Weight is the most commonly used variable for selecting med- Student t-test. Univariate and multivariate linear regressions
ication doses and equipment in pediatric patients. Height has were run to evaluate age, height, and weight as predictors of
been shown to be an adequate marker for weight-based dosing pelvic circumference. The regression coefficient and 95% CI
in both prehospital and hospital emergency settings and is the pelvic circumference for the listed predictive variables were
most commonly used variable when a patient’s weight cannot determined using age- and sex-adjusted linear regressions. The
be practically obtained. 12,13 Our study aimed to explore the odds ratio and 95% CI of SAM Pelvic Sling fit and Pediatric
correlation of age, weight, and height with pelvic circumfer- PelvicBinder fit were determined using age- and sex-adjusted
ence. By establishing accurate predictors of pelvic circumfer- multinomial logistic regressions. Univariate relations of SAM
ences, this study intended to use these data to create a more Pelvic Sling fit and Pediatric PelvicBinder fit were compared
practical sizing guideline for prehospital providers in the selec- using the maximum Broselow weight (36-kg) and the maxi-
tion of PCCDs. mum Broselow length (143-cm) using chi-square analysis. A
p value of <.05 was accepted as statistically significant. All
statistical analysis was conducted using Stata version 15.1
Methods
(StataCorp).
The study population used was a convenience sample of chil-
dren aged 1 year to 14 years recruited in the Johns Hopkins Results
Hospital pediatric emergency department from June 2017
to May 2018. Prior to recruitment, this study protocol was This study cohort consisted of 65 patients (61.5% male) with
approved by the institutional review board of Johns Hop- a mean age of 6.0 years (±4.0 years). The mean height, weight,
kins Hospital. Enrolled subjects included both patients with and pelvic circumferences are shown in Table 1. No statisti-
low-acuity conditions (Emergency Severity Index level 4/5) cally significant difference was found between males and fe-
and nonpatient visitors. Patients were not eligible for enroll- males in any of the measured variables.
ment if they had known or suspected pelvic fractures or pelvic
fixation devices, abdominal complaints, or any immune sys- Univariate and multivariate linear regressions (Table 2) were
tem compromise. After parental consent, study team members run to identify predictors of pelvic circumference size. Uni-
measured subject height and weight directly or obtained these variate regression analysis of height and weight as predictors
data from parents using a standardized data collection form. of pelvic circumference were statistically significant for each
Pelvic, arm, and thigh circumference were measured directly factor (p < .001). Multivariate linear regressions adjusting for
by study team members. Pelvic circumference was measured at age and sex identified weight as the sole positive predictor of
the level of the greater trochanter, reflecting the landmark for pelvic circumferences.
placement of PCCDs. Study personnel directly measured each
subject’s height and estimated weight using the Broselow Tape Of our cohort, the SAM Pelvic Sling and the Pediatric Pelvic-
and recorded the weight estimate in kilograms. We used the Binder fit for 14 (21.5%) and 59 (90.8%) of study patients,
maximum length of the Broselow Tape (i.e., 143-cm or 56-in) respectively. With the minimum pelvic circumference of the
to create a dichotomous variable for Broselow Tape fit. Each SAM Pelvic Sling, provided by the manufacturer, being 68-cm,
subject underwent fitting with one or both of the two widely our mean pelvic circumference of 63.2-cm fell below this min-
available commercial PCCDs: the Pediatric PelvicBinder and imum requirement, accounting for the decreased proportion
the small SAM Pelvic Sling, based on manufacturer indication of sufficient compression in this device. For both compression
for pelvic circumference (Pediatric PelvicBinder, 35.5–81.3-cm; devices, patient weights and lengths within the range of the
small SAM, 68–114.3-cm). The PCCDs were placed at the Broselow Tape scale displayed a significant association with
level of the greater trochanter. Criteria for a successful device device fit (p < .001). Of the 50 patients with actual or reported
fit included (1) no belt overlap at the level of the abdomen, weights that fell on the Broselow tape (<36-kg), 50 (100%)
(2) adequate device closure per manufacturer specifications had successful fit of the Pediatric PelvicBinder, and 0 (0%) had
(e.g., ability to engage and maintain the tension mechanism a successful fit with the SAM Pelvic Sling. Of the 15 patients
in locked position), and (3) the device’s not going above the with actual or reported weights greater than the maximum
top of the iliac crest or placing any pressure on the abdominal Broselow Tape weight, 13 (87%) had a successful device fit
compartment. The primary outcome of this study was the pro- with the SAM Pelvic Sling, and 9 (60%) had a successful fit
portion of subjects who fit each device. with the Pediatric PelvicBinder.

TABLE 1 Patient Demographics and Measurement with p-Values Measuring Statistical Differences by Sex
Overall Male (n = 40) Female (n = 25)
Variable (mean ± SD) (mean ± SD) (mean ± SD) p-Value
Age (y) 6.0 ± 4.0 6.3 ± 4.0 5.4 ± 4.1 .385
Height (cm) 117.9 ± 27.9 120.1 ± 26.6 114.1 ± 30.2 .404
Weight (kg) 27.5 ± 17.7 29.0 ± 18.6 25.2 ± 16.5 .406
Pelvic 63.2 ± 14.3 63.9 ± 14.5 62.1 ± 14.0 .753
Circumference (cm)
Broselow 19.7 ± 7.3 20.0 ± 7.5 18.8 ± 7.3 .579
Weight (kg)

Selection of Pediatric Pelvic Circumferential Compression Devices | 41