AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 9, NUMBER 5/6
Sept/Nov 1992
AN EASY AND ACCURATE METHOD OF ESTIMATING NEWBORN BIRTHWEIGHT FOR RESUSCITATION Gary M. Chan, M.D., Laurie Moyer-Mileur, M.S., and Lisa Rallison, M.S.
ABSTRACT
Each year approximately 350,000 infants will be resuscitated in the United States. Often, emergency medications will be required. Drug dosages and fluid therapy used in neonatal resuscitation vary and are based on the patient's body weight. In a crisis situation, health care personnel may be unable to weigh newborn patients, and the weight is then estimated. The estimated weight is then used to calculate vital drugs. Unfortunately, during a hectic resuscitation, this estimate of the infant's weight might be inaccurate. In pediatrics, a measuring tape using the patient's length has been used for estimating weight in older children,1 but there has not been a tape developed for newborns. The purpose of this study was to develop and evaluate a tape measure for newborn infants to estimate birthweight using either length or head circumference.
Downloaded by: National University of Singapore. Copyrighted material.
Emergency administration of medication based on birthweight is often required in newborn resuscitation. Actual weighing is often delayed because of the emergency situation. Therefore drugs are given according to weight estimates by physicians or nurses. The purposes of this study were to develop a measuring tape using the infant's length and head circumference to determine body weight and to evaluate the performance of the tape measurements to staffs' estimates. Weight, length, and head occipitofrontal circumference (OFC) measurements of 200 newborn infants were collected. By regression analyses, the best log curve for both length and OFC versus weight was determined. From these data, a measuring tape was constructed with the corresponding weights marked for both length and OFC. Forty-five newborn infants with gestational ages of 26 to 40 weeks were studied. Prior to actual weighing, estimates of the birthweights were obtained and recorded from the tape measurements using both the length and OFC and staffs' estimates. For infants less than 2 kg, the average percent error was less by the tape length (1.2%) and OFC (0.7%) compared with staffs' estimates (-7.2%). The tape was also useful in estimating weights of small for gestational age infants. The tape OFC (7.8% error) was more accurate than staffs' estimates (26% error) in these growthretarded infants. The clinical precision of the tape was 3% with an intrameasurer variability of 5%. We conclude that estimating the birthweight in infants using our tape method is a practical and more accurate way than staff estimates, especially for low birthweight and small for gestational age infants.
METHODS Two hundred newborn infants admitted to the University newborn intensive care unit were retrospectively evaluated. All infants were Caucasian, free of any congenital anomalies, and were appropriate for gestational age. All infants were between the 10th and 90th percentiles for weight, length, and head size according to the National Center for Health Statistics. Data about weight, length, and head occipitofrontal circumference (OFC) were recorded. By regression analyses, the best log curve for both length and OFC versus birthweight was determined. From the regression line, a measuring tape of the corresponding weights was constructed by labeling the weights in grams instead of units of length (Fig. 1). Using this tape, 45 infants were measured at birth.
Department of Pediatrics, Division of Neonatology, University of Utah, Salt Lake City, Utah Reprint requests: Dr. Chan, Department of Pediatrics, 2A230, 50 North Medical Drive, Salt Lake City, Utah 84132 Copyright © 1992 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
371
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 9, NUMBER 5/6
Sept/Nov 1992
employed to assess the best relationship. Mann-Whitney test was used to compare group differences.
Using the regression lines from the 200 newborn measurements, the tape was constructed for both length and head circumference. The relationship of birthweight to length and OFC is shown in Table 1. With the infant supine, the measurements of both length and head took about 20 seconds. Both length and OFC were highly correlated to the infant's birthweight. The errors to the actual birthweight of the tape and staff estimates are shown on Table 2. There was no difference between the measurements of the length or head, but the staffs estimates were different from either the tape length or OFC measurements. For infants less than 2 kg, the staff underestimated the weight by an average of 7.2%. For SGA infants, the staff overestimated (26%) the weight compared with the tape OFC estimates (7.8%). The tape length had a higher correlation (r = 0.986) than the tape OFC correlation (r = 0.958). Hospital staff estimates correlation was 0.956. The clinical precision of the tape was 3%. DISCUSSION Estimating the birthweight of infants was found to be easy and accurate by measuring the infant's length or OFC. The tape appears to be valid for low birthweight and SGA infants. For infants less than 2 kg, the staff underestimated the weight by an average of 7.2%, whereas the tape error was 0.7 to 1.2%. For SGA infants, the staff overestimated the weight by 26%, and the tape OFC estimate had an error of 7.8%. Since the OFC is usually spared in SGA infants, we found that the tape OFC estimates were more accurate than the tape length estimates. Medication doses calculated by these estimates may be clinically important during a resuscitation. Most methods used to estimate body weight in children have been based on age. 23 There have been several methods developed that estimate body weight based on Figure 1. A schematic picture of a preterm infant being measured by the tape to determine her weight. She measures approximately 1400 gm. Table 1. Relation of Birthweight to Length and Occipitofrontal Circumference (OFC) Size
Thirty infants were appropriate for their gestation and had no congenital anomalies. Eight infants were small for gestational age (SGA), denned as birthweight less than the 10th percentile. Seven infants were large for gestational age (LGA) whose birthweight was greater than the 90th percentile. Immediately after delivery, estimates of the birthweights from the hospital staff of nurses and physicians were obtained and recorded. All observers were blinded to each other's estimates. There were two different nurses and two different doctors for each evaluation. Finally, the infant was weighed on a Toledo infant scale. Clinical precision of the tape was determined by making repeated measurements of infants. The values of birthweight were tabulated and a coefficient of variation was determined. Standard regression analyses were used in either a 372 linear or nonlinear model. Correlation coefficients were
Weight (gm) 500 700 800 900 1000 1250 1500 1750 2000 2500 3000 4000 5000
Length (cm)*
26 29 31.5
34 35.2 38.6 40.2
43 44.2 46.5
50 54 56
*Weight = 60.19 X 10 2 g ( n = 8) Average error (%) Range SGA (n = 8) Average error (%) Range LGA (n = 7) Average error (%) Range Precision Clinical precision (%) Intermeasurer variability (%) Intrameasurer variability (%)
OFC
*Actual weight - estimate/actual weight - 100. +p
Sept/Nov 1992
AN EASY AND ACCURATE METHOD OF ESTIMATING NEWBORN BIRTHWEIGHT FOR RESUSCITATION Gary M. Chan, M.D., Laurie Moyer-Mileur, M.S., and Lisa Rallison, M.S.
ABSTRACT
Each year approximately 350,000 infants will be resuscitated in the United States. Often, emergency medications will be required. Drug dosages and fluid therapy used in neonatal resuscitation vary and are based on the patient's body weight. In a crisis situation, health care personnel may be unable to weigh newborn patients, and the weight is then estimated. The estimated weight is then used to calculate vital drugs. Unfortunately, during a hectic resuscitation, this estimate of the infant's weight might be inaccurate. In pediatrics, a measuring tape using the patient's length has been used for estimating weight in older children,1 but there has not been a tape developed for newborns. The purpose of this study was to develop and evaluate a tape measure for newborn infants to estimate birthweight using either length or head circumference.
Downloaded by: National University of Singapore. Copyrighted material.
Emergency administration of medication based on birthweight is often required in newborn resuscitation. Actual weighing is often delayed because of the emergency situation. Therefore drugs are given according to weight estimates by physicians or nurses. The purposes of this study were to develop a measuring tape using the infant's length and head circumference to determine body weight and to evaluate the performance of the tape measurements to staffs' estimates. Weight, length, and head occipitofrontal circumference (OFC) measurements of 200 newborn infants were collected. By regression analyses, the best log curve for both length and OFC versus weight was determined. From these data, a measuring tape was constructed with the corresponding weights marked for both length and OFC. Forty-five newborn infants with gestational ages of 26 to 40 weeks were studied. Prior to actual weighing, estimates of the birthweights were obtained and recorded from the tape measurements using both the length and OFC and staffs' estimates. For infants less than 2 kg, the average percent error was less by the tape length (1.2%) and OFC (0.7%) compared with staffs' estimates (-7.2%). The tape was also useful in estimating weights of small for gestational age infants. The tape OFC (7.8% error) was more accurate than staffs' estimates (26% error) in these growthretarded infants. The clinical precision of the tape was 3% with an intrameasurer variability of 5%. We conclude that estimating the birthweight in infants using our tape method is a practical and more accurate way than staff estimates, especially for low birthweight and small for gestational age infants.
METHODS Two hundred newborn infants admitted to the University newborn intensive care unit were retrospectively evaluated. All infants were Caucasian, free of any congenital anomalies, and were appropriate for gestational age. All infants were between the 10th and 90th percentiles for weight, length, and head size according to the National Center for Health Statistics. Data about weight, length, and head occipitofrontal circumference (OFC) were recorded. By regression analyses, the best log curve for both length and OFC versus birthweight was determined. From the regression line, a measuring tape of the corresponding weights was constructed by labeling the weights in grams instead of units of length (Fig. 1). Using this tape, 45 infants were measured at birth.
Department of Pediatrics, Division of Neonatology, University of Utah, Salt Lake City, Utah Reprint requests: Dr. Chan, Department of Pediatrics, 2A230, 50 North Medical Drive, Salt Lake City, Utah 84132 Copyright © 1992 by Thieme Medical Publishers, Inc., 381 Park Avenue South, New York, NY 10016. All rights reserved.
371
AMERICAN JOURNAL OF PERINATOLOGY/VOLUME 9, NUMBER 5/6
Sept/Nov 1992
employed to assess the best relationship. Mann-Whitney test was used to compare group differences.
Using the regression lines from the 200 newborn measurements, the tape was constructed for both length and head circumference. The relationship of birthweight to length and OFC is shown in Table 1. With the infant supine, the measurements of both length and head took about 20 seconds. Both length and OFC were highly correlated to the infant's birthweight. The errors to the actual birthweight of the tape and staff estimates are shown on Table 2. There was no difference between the measurements of the length or head, but the staffs estimates were different from either the tape length or OFC measurements. For infants less than 2 kg, the staff underestimated the weight by an average of 7.2%. For SGA infants, the staff overestimated (26%) the weight compared with the tape OFC estimates (7.8%). The tape length had a higher correlation (r = 0.986) than the tape OFC correlation (r = 0.958). Hospital staff estimates correlation was 0.956. The clinical precision of the tape was 3%. DISCUSSION Estimating the birthweight of infants was found to be easy and accurate by measuring the infant's length or OFC. The tape appears to be valid for low birthweight and SGA infants. For infants less than 2 kg, the staff underestimated the weight by an average of 7.2%, whereas the tape error was 0.7 to 1.2%. For SGA infants, the staff overestimated the weight by 26%, and the tape OFC estimate had an error of 7.8%. Since the OFC is usually spared in SGA infants, we found that the tape OFC estimates were more accurate than the tape length estimates. Medication doses calculated by these estimates may be clinically important during a resuscitation. Most methods used to estimate body weight in children have been based on age. 23 There have been several methods developed that estimate body weight based on Figure 1. A schematic picture of a preterm infant being measured by the tape to determine her weight. She measures approximately 1400 gm. Table 1. Relation of Birthweight to Length and Occipitofrontal Circumference (OFC) Size
Thirty infants were appropriate for their gestation and had no congenital anomalies. Eight infants were small for gestational age (SGA), denned as birthweight less than the 10th percentile. Seven infants were large for gestational age (LGA) whose birthweight was greater than the 90th percentile. Immediately after delivery, estimates of the birthweights from the hospital staff of nurses and physicians were obtained and recorded. All observers were blinded to each other's estimates. There were two different nurses and two different doctors for each evaluation. Finally, the infant was weighed on a Toledo infant scale. Clinical precision of the tape was determined by making repeated measurements of infants. The values of birthweight were tabulated and a coefficient of variation was determined. Standard regression analyses were used in either a 372 linear or nonlinear model. Correlation coefficients were
Weight (gm) 500 700 800 900 1000 1250 1500 1750 2000 2500 3000 4000 5000
Length (cm)*
26 29 31.5
34 35.2 38.6 40.2
43 44.2 46.5
50 54 56
*Weight = 60.19 X 10 2 g ( n = 8) Average error (%) Range SGA (n = 8) Average error (%) Range LGA (n = 7) Average error (%) Range Precision Clinical precision (%) Intermeasurer variability (%) Intrameasurer variability (%)
OFC
*Actual weight - estimate/actual weight - 100. +p
