 |
|
Brief Reports Indian Pediatrics 2001; 38: 762-766 |
|||||||||||||||||
|
Pulse Oximetry in Asphyxiated Newborns in the Delivery Room |
|||||||||||||||||
|
From
the Neonatal Division, Department of Pediatrics, Maulana Azad Medical
College, New Delhi 110 002, India.
Effective management of asphyxiated newborns in the initial few minutes of life may significantly alter their outcome. Apgar scores have been used traditionally to assess the well being of the newborn infant in the immediate neonatal period. The falacies of this assessment to determine adequacy of resuscitation are well known. Pulse oximetry offers a reliable non-invasive, real-time objective method for monitoring oxygen saturation. Use of the pulse oximeter in the labor room has demonstrated the efficacy as well as the sensitivity of this tool in assessing the cardiopulmonary adaptation of the newborn both in normal and asphyxiated newborn infants(1-2). Its use to monitor oxygenation during neonatal resuscitation has been insufficiently documented. The present study was designed to evaluate the ability of pulse oximetry in recording oxygen saturation in asphyxiated neonates and serial changes in oxygenation during their resuscitation. Subjects and Methods Asphyxiated neonates who had been enrolled as part of a large multi-centric clinical trial on resuscitation(3) were the subjects for evaluating the utility of pulse oximetry during resuscitation in the delivery room. The neonates enrolled had birth weights more than 999 g, and at birth were apneic or gasping and had heart rates <80/minute (the asphyxia group). Those with lethal anomalies, hydrops fetalis and congenital cyanotic heart disease had been excluded from the study. Neonates born during the study period with birth weights >2499 g and without asphyxia were used as controls. Primary outcome variable for the study was the time taken to reach an oxygen saturation of 90% recorded by pulse oximeter. The secondary outcome variables were the time taken for first record and time taken to reach 75% saturation. Oximetry recordings were done using a Novametrix 515A® pulse oximeter. The Y-sensor was placed on the ulnar aspect of the right palm of the neonate immediately after the baby was placed under the radiant warmer and dried (usually 30-40 seconds after birth). A clock was started as soon as the umbilical cord was cut. Recordings were made continuously till the oxygen saturation reached 90%. The time to first record and time to reach 75% were also noted. In babies needing resuscitation, standard protocols as per the guidelines of the AHA-AAP were used(4). In all neonates baseline information was recorded with regard to maternal age, parity, gestational age, obstetri-cal complications, birth weight and mode of delivery and Apgar scores. Statistical Analysis All continuous normally distributed variables such as age, parity, gestation and birth weight were analyzed by comparing the means using the Student ‘t’ test. Non para-metric continuous variables such as time to different saturation values were analyzed using Mann-Whitney U test. Proportions were compared using Chi-square/Fisher exact test. A probability of 5% was taken as significant. Results Ninety-five asphyxiated infants and thirty controls were enrolled into the study. The two groups were comparable with regard to gestation, birth weights and the mode of delivery. The mean heart rate at one minute of age was significantly lower in the asphyxiated group (95.6 ± 37.9 per minute) as compared to the controls (161.1 ± 19.8 per minute). There were no significant differences in heart rates between the groups thereafter. In 10 (10.5%) infants in the asphyxiated group and 2 (6.7%) infants in the control group, no recordings were obtained despite the appropriate application of the sensor onto the neonate’s palm (p >0.05). Pulse oximetry recordings were obtained in 6.3, 36.5, 63.1, and 80% of the asphyxiated infants and 10, 80, 90, and 90% of the control infants at 1,3,5, and 10 minutes of age respectively. The percentage of oximetry records obtained at 3 and 5 minutes were significantly lower (p <0.01) in the asphyxiated group compared to the control, but were not significant at other ages. The median time (inter-quartile range) for the first recording to detect oxygen saturation was significantly longer in asphyxiated neonates (195, 170-300 sec) compared to controls (162.5, 90-180 sec). Table I provides the oxygen saturation values in the two groups during the first 10 minutes of life. There were no significant differences (p >0.05) between the asphyxiated and control subjects. The median (inter-quartile range) time to reach 75% saturation was 300 (180-435) sec in the asphyxiated group compared to 170 (100-266) sec in the control group (p <0.001). This difference in saturation trends was also reflected in achieving 90% saturation. The median (inter-quartile range) time taken for the saturation to reach 90% was 435 (270-660) sec in the asphyxia group compared to 260 (175-350) sec in the control group. Discussion Earlier studies in normal newborn infants in the delivery room have observed a very wide range in the proportion of neonates on whom pulse oximetry recordings could be obtained at 1 min (38-75%), but by 5 minutes the recordings could be obtained in more than 90% of the neonates(1,2,5,6). In the present study the percentage of normal neonates on whom a record was obtained in the first minute was lower than that reported in the earlier studies; however, the 5-minute recordings were similar to that reported by earlier studies. The delay in obtaining recordings initially could be due to several reasons - decrease in peripheral pulse volume during post-natal adaptation, move-ment artifacts, probe mal-position, excess pressure and venous occlusion while securing the probe, and the time taken for the instrument in detecting the pulsatile arterial wave-form(7,8). In asphyxiated infants, weak waveform due to decreased perfusion may contribute to decreased signal detection, resulting in lower recordings in them during the first few minutes of life. The asphyxiated infants in whom no saturation recordings could be obtained in this study eventually had a poor neonatal outcome. This suggests that inability to obtain recordings during the first 5-10 minutes of life in asphyxiated neonates could predict a poor outcome.
The upper limb was chosen for sensor placement in the present study based on earlier reports of faster signal detection in the upper limb. Saturation was recorded in 50% of infants at 1.3 min and in 90% by 4 min in the upper limb compared to 50% of infants at 3.1 min and in 90% by 9 min in the lower limb(7). The mean SaO2 recorded by pulse oximetry at one minute of life may vary from 43% to 77% in normal newborn infants. This rises to mean values in the eighties by 5 minutes, 88% at ten minutes, 90% by 15 minutes, 92% by 60 minutes and 96% by 5 hours of age (1,2,5,9). The observations in the present study support the earlier results. Infants in some of these studies however also received oxygen supplementation after birth. House et al.(1) however, failed to demonstrate any differences in saturation in infants who received oxygen after birth from those who did not. In the present study about 60% asphyxiated infants received supplemental oxygen during resus-citation; the supplemented group had signifi-cantly higher SaO2-recordings at 5 and 10 min. There are few studies evaluating the efficacy of pulse oximeter as a monitoring tool during neonatal resuscitation. In a recent experimental study in lambs, the mean (SD) SaO2 recorded in normally delivered lambs and those with asphyxia were 67 (15)%, 84 (9)%, 83 (9)% and 61 (15)%, 69 (16)%, 69 (19)% at 1, 5, and 10 minutes of age, respectively (10). Maxwell et at.(11) evaluated oxygenation during resuscitation of four asphyxiated preterm infants. They were able to demonstrate the adequacy of bag and mask ventilation by rising SaO2 whereas attempts at endotracheal intubation resulted in falling SaO2. They were also able to detect displaced endotracheal tube before it was detected clinically. Sendak et al.(12) during resuscitation of four term newborn infants were successfully able to document the effectiveness of bag and mask ventilation by documenting rising SaO2. In the present study too, rising oxygen saturation with resuscitation was demonstrated in asphyxiated newborns. It was further observed that by three minutes of age the SaO2 recordings were comparable between asphyxiated and normal neonates; thus indicating the efficacy of resuscitation procedures that were used. We have found the pulse oximeter to be a useful monitoring tool during resuscitation of asphyxiated newborn infants. The adequacy of oxygenation during resuscitation being reflected by the rising oxygen saturation. While clinical parameters still remain important guidelines in neonatal resuscitation, pulse oximetry is an additional noninvasive monitoring tool that can help identify those with arterial desaturation and hypoxemia. Contributors: RR collected the data and drafted the manuscript. SR coordinated the study, performed the statistical analysis and was responsible for overall drafting of the manuscript. He will act as the guarantor for the paper. Competing
interests: None stated.
| |||||||||||||||||
|
References | |||||||||||||||||
|
