PEDIATRICS Vol. 106 No. 3 September 2000, pp. 618-620

Cardiopulmonary Resuscitation in Very Low Birth Weight Infants

To the Editor.

We read with interest the recent manuscript entitled "Cardiopulmonary Resuscitation in the Very Low Birth Weight Infant: The Vermont Oxford Network Experience."¹ The authors should be commended for pursuing a complex clinical issue. However, there are significant deficiencies in study design and data analysis that limit the interpretation and significance of the data.

Briefly, as background information, the birth of an infant is associated with abrupt cessation of the fetomaternal circulation with subsequent rapid and profound physiologic changes involving both the cardiac and respiratory systems. Failure of either system to adapt will result in cardiorespiratory compromise and the need for resuscitation. Two cardinal events appear critical in the genesis of compromised cardiorespiratory adaptation and the subsequent need for intensive resuscitation (CPR). These events include failure to establish a functional residual capacity (FRC) and the presence of impaired placental gas exchange as evidenced by profound fetal acidemia (umbilical cord arterial pH 7.00 [2,31]). Failure to establish a FRC may be secondary to ineffective or improper ventilation or an abnormal underlying pulmonary state, eg, pulmonary hypoplasia, pulmonary immaturity, etc. Indeed, we previously documented the prominent association between improper or ineffective ventilatory support and the need for CPR.⁴ The presence of asphyxia may be suspected when there is an associated sentinal event, eg, abruptio placentae, but often is clinically inapparent in the delivery room.⁵ Recovery after CPR is in part influenced by the cause of the cardiorespiratory compromise, the duration of the precipitating event, the establishment of effective ventilation, and restoration of spontaneous circulation (ROSC) with establishment of coronary perfusion pressure.

To briefly illustrate the importance of the above issues, we present data on infants <1500 g birth weight who received CPR (n = 21), ie, chest compression only (n = 17) and additionally epinephrine (n = 4) for the years 1996-1998. It should be noted that in cases of uncertain prognosis (birth weight < 600 g) CPR is not pursued in our delivery room (DR) in all cases. Moreover, prophylactic surfactant is not administered in the DR. The percentage of infants receiving CPR by birth weight is shown in Table 1.

The percentage of infants requiring chest compressions is comparable to the Vermont Oxford Network experience. However, the percentage of infants requiring epinephrine is markedly less. Current National Resuscitation Program (NRP) guidelines are followed in our institution; however, we stress the importance of ventilation in ROSC.

This may in part explain the less frequent use of epinephrine. The perinatal characteristics and short-term outcome of the 21 infants are illustrated in Table 2.

The data highlight several important points that are crucial when assessing the efficacy of CPR including: 1) the duration of CPR which was significantly longer for infants who died, 270 ± 231 versus 43 ± 32 seconds (P = .0004) for all survivors, and even shorter for those survivors without severe IVH 30 ± 12 seconds; 2) lethal conditions incompatible with life which included pulmonary hypoplasia, complex heart disease (unknown in the DR) and trisomy 18; and 3) ineffective or improper ventilatory support, ie, a misplaced endotracheal tube that complicated the resuscitation in at least 4 cases.

This and additional relevant information lacking from the Oxford-Vermont data included: 1) the outcome of infants with a 1-minute Apgar score of 0 or 1 alone, or as compared to infants with an Apgar score of 2; 2) the number of infants who received prophylactic surfactant in the DR that may have complicated cardiorespiratory adaptation; and 3) the absence of important clinical data, ie, the presence or absence of asphyxia. In addition, it is noteworthy that of the 1595 infants of birth weight 501-1500 g, 670 (42%) had a 1-minute Apgar score 2 and 367 infants (19%) had a score 3. Did these latter infants really require chest compressions? The authors fail to comment on the potential neuroprotective influence of antenatal glucocorticoids, ie, decreased incidence of intraventricular hemorrhage, which was administered to 33% of the mothers of infants who received DR-CPR. Because neither the timing and/or number of cranial ultrasound scans performed is provided, cases of late hemorrhage and/or cystic periventricular leukomalacia (PVL) may have been missed. Despite this limitation, it is very bothersome that 7.8% of infants who received DR-CPR had evidence of PVL that was double the anticipated 3.9% incidence in the group of infants without CPR. The diagnosis of PVL is synonymous with poor neurodevelopmental outcome.

Despite these inherent weaknesses, the data are important vis-à-vis the antenatal counseling of parents as to the potential for CPR in the DR, ie, 10% for the tiniest infants (<750 g) versus 2% for the larger infants (>1250 g). Given the rarity of asphyxia in this patient population, ineffective ventilation and failure to establish a FRC should be the primary concern for those infants with persistent bradycardia. For the tiniest infants, pulmonary immaturity is the most likely explanation in most cases. The duration of CPR would appear to be the most useful clinical determinant of short and long adverse outcome, an observation consistent with data from other patient population groups, ie, children and adults.

Myra Wyckoff
Jeffrey Perlman
Department of Pediatrics
University of Texas Southwestern Medical Center
Dallas, TX 75235

REFERENCES

1. Finer NN, Horbar JD, Carpenter JH Cardiopulmonary resuscitation in the very low birth weight infant: the Vermont Oxford Network experience. Pediatrics. 1999; 104:428-434 [Abstract/Free Full Text]
2. Vyas H, Field D, Milner AD, Hopkin IE Determinants of the first inspiratory volume and functional residual capacity at birth. Pediatr Pulmonol. 1986; 2:189-193 [Medline]
3. American College of Obstetricians and Gynecologists. Assessment of Fetal and Newborn Acid-Base Studies. Washington, DC: American College of Obstetricians and Gynecologists; 1989:1-4. ACOG Technical Bulletin No. 127
4. Perlman JM, Risser RR Cardiopulmonary resuscitation in the delivery room: associated clinical events. Arch Pediatr Adolesc Med. 1995; 149:20-25 [Abstract/Free Full Text]
5. King TA, Jackson GL, Josey AS, The effect of profound umbilical artery acidemia in term neonates admitted to a newborn nursery. J Pediatr. 1998; 132:571-572 [CrossRef][Medline]

In Reply.

We thank Drs Wyckoff and Perlman for their correspondence regarding our manuscript. In their letter they emphasize the need to establish a functional residual capacity (FRC) shortly after delivery, discuss some of the shortcomings of the data presented in our manuscript, and review their own experience with DR-CPR. We are pleased that our paper has stimulated others to review their own experience and provide additional information, which may be of clinical importance.

Wyckoff and Perlman discuss events related to poor cardiopulmonary adaptation at birth. We agree that impaired placental gas exchange predisposes infants, especially extremely very low birth weight (VLBW) infants, to poor adaptation, but, as has been previously pointed out, cord pH per se is often inadequate to explain a poor transition at birth or subsequent neonatal events.¹ Other factors including the arterial base deficit and Apgars may be more predictive of neonatal outcome. The failure to establish spontaneous breathing or to make effective respiratory efforts may delay the development of an adequate FRC and the use of larger, longer breaths during resuscitation may be effective in promoting the development of FRC.² In addition, previously compromised fetuses are often intolerant of labor and delivery. We agree that prompt and effective resuscitation, including the early and effective establishment of FRC, is essential to maximize intact recovery. In addition, the reflex response by the premature infant to positive pressure ventilation appears to be important in improving tidal exchange and establishing and FRC.³ Such reflex responses may be inhibited by degrees of asphyxia during labor.

Wyckoff and Perlman indicate that certain information was lacking from our review. To facilitate the use of the Vermont Oxford Network database at a large number of neonatal units, the database has been limited a set of core data items. Asphyxia is not a data item in the database. The timing of surfactant therapy was not an item in the database during the time period of the study, but has been subsequently added and is now routinely collected. We indicated the percentage of all the infants whose mothers received antenatal steroids, but limited our discussion to the aspects of resuscitation and neonatal outcome. Wyckoff and Perlman mistakenly indicate that steroids were "administered to 33% of mothers of infants who received DR-CPR." As shown in Table 1 of our report, 33% of the 497 infants weighing 401 to 500 g and 64% of the 27 210 infants weighing 501 to 1500 g were exposed to antenatal steroid therapy.⁴ We did not report steroid exposure separately for the infants who received DR-CPR.

We specifically limited our discussion of PVL because were not able to review the neuroanatomic evaluations and their timing as stated on page 432.⁴ We were also concerned about the overall incidence of PVL, but because we could not evaluate the extent of this abnormality, we refrained from further comment. Until our review, there was no suggestion that infants <750 g at birth would survive if they required DR-CPR. Our efforts were directed at determining the proportion of infants who survived with and without evidence of neuromorbidity. However, our study did not evaluate long-term neurodevelopment.

A recent study has evaluated VLBW infants who survived following DR-CPR and has shown that intact survival occurred in 70% of the 15 survivors who were assessed. In that study at least 1 CPR survivor with white-matter injury was neurodevelopmentally normal.⁵ We welcome further data which demonstrate that intact neonatal survival can occur after DR-CPR, and note that 2 of the 6 infants <750 g shown by Wyckoff and Perlman survived apparently intact.

Wyckoff and Perlman have provided information from a 4-year period from their own institution. They report that during this period 21 (4%) of 525 infants weighing 501 to 1500 g received CPR. This is similar to the overall rate of cardiac compressions of 4.8% that we reported. Furthermore, 11 (52%) of their 21 infants survived and only 3 of the 11 survivors had evidence of severe IVH. This is consistent with our findings. The difference in the proportion of infants receiving epinephrine during resuscitation is not surprising because our report is based on data for >1600 resuscitations at 196 centers whereas theirs is a single center report. Clearly, there is marked variation among units in DR practice including the use of epinephrine. For instance, in 1998 there were 4654 infants weighing 501 to 750 g treated at the 295 neonatal intensive care units (NICUs) participating in the Vermont Oxford Network that year. Overall, 9% of these infants received epinephrine in the DR. However, there was marked variation among centers; with 25% of NICUs having a 0% rate for this treatment and 25% of NICUs having rates >17%.⁶

Wyckoff and Perlman indicate that they follow NRP guidelines at their institution and stress the importance of ventilation and the ROSC. They report 3 infants who received 180 seconds of DR-CPR, at least 1 of whom had this intervention initiated at 5 minutes, and none of these infants received epinephrine. These authors have commented that infants with Apgar scores of 2 at 1 minute were unlikely to require compressions. Although we agree and so stated in the manuscript on page 431,⁴ it is of interest to us that 6 of their 21 infants who received chest compressions had an Apgar at 1 minute of 2 or greater. We believe that their information supports our belief that current practices of neonatal resuscitation are variable and do not always follow the teachings of the AAP/AHA NRP.⁷ As we stated in our "Conclusion," "Further information is urgently required regarding the indications, application, and longer-term neurodevelopmental outcome of DR-CPR so that we may fully assess its role in the initial treatment of these fragile infants." We hope that our report will stimulate additional research on the issue of DR-CPR for VLBW infants.

Neil N. Finer
Department of Pediatrics
University of California, San Diego
School of Medicine
San Diego, CA 92103-8774

Jeffrey D. Horbar
University of Vermont
College of Medicine
Vermont Oxford Network
Burlington, VT 05401

REFERENCES

1. Sehdev HM, Stamilio DM, Macones GA, Graham E, Morgan MA Predictive factors for neonatal morbidity in neonates with an umbilical arterial cord pH less than 7.00. Am J Obstet Gynecol. 1997; 177:1030-1034 [CrossRef][Medline]
2. Vyas H, Milner AD, Hopkin IE, Physiologic responses to prolonged and slow rise inflation in the resuscitation of the asphyxiated newborn infant. J Pediatr. 1981; 99:635 [CrossRef][Medline]
3. Hoskyns EW, Milner AD, Boon AW, Vyas H, Hopkin IE Endotracheal resuscitation of preterm infants at birth. Arch Dis Child. 1987; 62:663-666 [Abstract/Free Full Text]
4. Finer NN, Horbar JD, Carpenter JH Cardiopulmonary resuscitation in the very low birth weight infant: the Vermont Oxford Network experience. Pediatrics. 1999; 104:428-434
5. Finer NN, Tarin T, Vaucher YE, Barrington K, Bejar R. Intact survival in extremely low birth weight infants after delivery room resuscitation. Pediatrics. 1999:104(4). URL: http://www.pediatrics.org/cgi/content/fall/104/4/e40
6. Vermont Oxford Network, Vermont Oxford Network 1998 Database Summary. Burlington, VT: Vermont Oxford Network; 1999
7. Bloom RS, Cropley C and the AHA/AAP Neonatal Resuscitation Program Steering Committee. Textbook of Neonatal Resuscitation. Dallas, TX: American Heart Association; 1994