Preterm birth and its associated complications account for significant neonatal morbidity and mortality,1,2 including an increased incidence of intraventricular hemorrhage, bronchopulmonary dysplasia (BPD), sepsis, necrotizing enterocolitis, and hemodynamic instability.3,4 The appropriate use of progesterone therapy to reduce the occurrence or recurrence of preterm birth and corticosteroids and magnesium sulfate to reduce neonatal morbidity if preterm birth occurs are examples of prenatal interventions. However, delayed cord clamping is an intrapartum procedure, which may also benefit the premature neonate.
Delaying clamping of the umbilical cord in preterm neonates improves hemodynamic stability and immune function by increasing blood volume and oxygenation as a result of the additional iron, immunoglobulins, and stem cells delivered from the placenta.5 Multiple randomized controlled trials have demonstrated preterm neonates who receive delayed cord clamping are less likely to require intubation in the delivery room, receive red blood cell transfusions in the first week of life, or develop intraventricular hemorrhage.6 Delayed cord clamping is also associated with decreased risk of hypothermia, necrotizing enterocolitis, sepsis, and need for blood pressure support.6,7 Additionally, this intervention has not been found to result in adverse consequences.8,9 Given this evidence, in December 2012, the American College of Obstetricians and Gynecologists (the College) recommended 30–60 seconds of delayed umbilical cord clamping in all premature neonates when feasible10 and recently expanded these guidelines to include a recommendation for delayed cord clamping for all neonates.11 These recommendations were further endorsed by the American Academy of Pediatrics in 2013.12
Despite the College’s recommendation, the actual practice of delayed cord clamping has varied widely among obstetricians nationwide, even in settings where protocols have been attempted, and patient factors that influence health care provider adherence to delayed cord clamping remain poorly understood.10,13 Therefore, our aim was to evaluate a delayed cord clamping protocol introduced in an institution with a large, diverse obstetric population. The study objectives were 1) to perform a quality improvement evaluation of adherence to the delayed cord clamping protocol and 2) to investigate neonatal outcomes associated with the implementation of delayed cord clamping.
MATERIALS AND METHODS
This is a retrospective cohort study of women delivering singleton neonates at 23–32 weeks of gestation in the 2 years before and 2 years after institution of a delayed cord clamping protocol at a large-volume academic center. At this medical center, approximately 12,000 births occur per year, and deliveries are performed by both private and university-employed faculty with residents typically involved in preterm deliveries. Health care provider types include maternal–fetal medicine physicians, obstetric physicians, and midwives. Delayed cord clamping was not routinely performed before introduction of this protocol.
This institution's 2013 delayed cord clamping protocol, based on a review of existing literature and College and American Association of Pediatrics recommendations, includes clamping the umbilical cord 30 seconds after delivery for preterm neonates at 23–32 weeks of gestation. For vaginal deliveries, proper technique includes wrapping the neonate in a sterile towel and holding him or her below the level of the maternal introitus, whereas cesarean deliveries involve wrapping the neonate in a sterile towel and placing him or her on the lower maternal abdomen or legs. A designated member of the delivery team tracks the time of the delay and ensures proper documentation and team communication. Neonatal resuscitation is initiated by the neonatal team immediately after clamping and cutting the umbilical cord. Exclusion criteria include multiple gestations, estimated gestational age 32 weeks, general anesthesia, major fetal anomalies, or no intent to resuscitate. The procedure can also be terminated at any time at the discretion of the obstetrician or neonatologist for a newborn thought to have more immediate resuscitative needs. Planned cord blood banking or donation is not considered an indication to abstain from performing delayed cord clamping.14 Neonates eligible for delayed cord clamping by this protocol constituted this study's analysis sample.
Beginning in July 2013, the delayed cord clamping policy was communicated to obstetric and pediatric providers and nurses and reinforced through intermittent educational sessions on the benefits of delayed cord clamping for preterm neonates. Health care providers were strongly recommended to delay cord clamping in eligible neonates per this quality improvement initiative but were not mandated to do so. At this institution, all very preterm births are attended by resident physicians who are always supervised by attending obstetricians; all neonatal resuscitation teams include an attending neonatologist. Educational sessions were directed by a delayed cord clamping quality improvement team composed of maternal–fetal medicine, nursing, and neonatology personnel. At these sessions, audit and feedback with regard to protocol adherence were provided by one investigator (L.M.Y.) along with neonatal intensive care unit leadership. At periodic intervals (Fig. 1), reeducation to residents, fellows, nurses, and attending physicians was provided through updates at dedicated nursing, resident, and fellow lectures; brief educational lectures at labor and delivery sign-out; departmental grand rounds; signage and flyers; email reminders; and in-person reminders at the time of deliveries. After 12 months of the delayed cord clamping protocol and data review (performed routinely as part of the quality improvement process), nurse-led initiation of the delayed cord clamping procedure was added to the protocol such that neonatal and obstetric nursing remind the accoucheur to perform delayed cord clamping immediately before delivery, count the time aloud during the delay, and document the procedure occurrence in the maternal and neonatal electronic medical record. The electronic labor board is additionally utilized to remind teams about intent for delayed cord clamping.
This analysis includes the 2 years before (preprotocol period: January 1, 2011, to June 30, 2013) and the 2 years after (postprotocol period: July 1, 2013, to June 30, 2015) the introduction of the delayed cord clamping protocol. Data were abstracted from the electronic medical record. First, eligible deliveries within the postprotocol period were assessed for the quality improvement aim of adherence to and completion of the protocol. After introduction of the protocol, the frequency with which eligible neonates received delayed cord clamping was calculated on a monthly basis and plotted on a run chart. Run charts can be used to demonstrate signals of improvement in a process and to examine whether changes in a process lead to improvements over time, such as increase in health care provider adherence to the protocol. Each point graphically represents a percentage adherence to a given quality improvement measure. In this study, each data point represents the proportion of neonates in a given month who received delayed cord clamping out of all who were eligible. Nonrandom patterns of improvement were analyzed using the median and standard run chart evaluation rules. The two run chart rules used for the study include a shift, a period in which six or more consecutive points all fall above or below the median, and a trend, a period in which five or more consecutive points all rise or fall.15 We then compared maternal and perinatal characteristics between eligible women who did and did not receive delayed cord clamping in the postprotocol period to identify factors independently associated with the likelihood of performance of the protocol.
To address the second aim, outcomes associated with implementation of delayed cord clamping, we assessed the association between delayed cord clamping and neonatal hemodynamic and hematologic outcomes in two manners. Maternal and neonatal characteristics in the preprotocol and postprotocol cohorts were compared. To assess the overall effect of delayed cord clamping protocol introduction, we compared all neonates in the preprotocol period (none of whom had received delayed cord clamping) with all neonates in the postprotocol period. To assess the effect of actual performance of delayed cord clamping, we performed a per-protocol analysis, comparing all neonates in the preprotocol period with the subgroup of neonates in the postprotocol period who received delayed cord clamping per guidelines.
Neonatal outcome data included a number of postdelivery hemodynamic, procedural, and laboratory outcomes such as initial temperature, birth hematocrit, red blood cell transfusions, length of mechanical ventilation, peak bilirubin, need for pharmacologic blood pressure support, BPD, necrotizing enterocolitis, late-onset sepsis, grade III or IV intraventricular hemorrhage, and neonatal death. Need for pharmacologic blood pressure support was defined as any need for inotropic support in the neonate. Bronchopulmonary dysplasia was defined as continued need for supplemental oxygen at 36 weeks corrected gestational age. Necrotizing enterocolitis was defined as Bell's stage 2 or greater with confirmation on abdominal radiography.16 Late-onset sepsis was defined as culture-positive sepsis at least 3 days after delivery. Grade III and IV intraventricular hemorrhage was recorded based on the Papile et al criteria.17
This study was approved by the institutional review board of Northwestern University. Maternal and neonatal characteristics and outcomes were compared using the Student t test or Mann-Whitney U test where applicable for continuous variables and χ2 or Fisher exact test where applicable for categorical variables. Multivariable logistic regression was used to identify factors associated with delayed cord clamping performance in the postprotocol period. For the assessment of neonatal outcomes, linear and logistic regression models were used to assess the independent association between 1) cohort and neonatal outcomes and 2) receipt of delayed cord clamping and neonatal outcomes. Multivariable models adjusted for demographic and clinical characteristics that were significant at P<.1 in bivariable analysis. All tests were two-tailed and P<.05 denoted significance. All statistical analyses were performed with SPSS 22. The run chart was created using Microsoft Excel and analyzed using standard run chart analytical techniques15 to apply probability-based rules for determining nonrandom changes in data.
A total of 427 neonates met inclusion criteria during the study period. Of these, 240 were delivered before and 187 were delivered after the introduction of the protocol. Of the 187 postprotocol patients, 53.5% (n=100) received delayed cord clamping, whereas 0% received delayed cord clamping before protocol introduction.
The run chart in Figure 1 illustrates monthly adherence rates to the delayed cord clamping protocol. The institutional goal was set at 80% after discussion among neonatal and obstetric providers. This chart demonstrates the significant improvement in delayed cord clamping compliance that occurred after initiation of the protocol in July 2013 as determined by standard run chart probability-based rules. First, a shift occurred from February 2014 to September 2014, demonstrating a nonrandom period of sustained improvement in protocol performance. Decreases in compliance were always followed by a rebound in adherence with reeducation. Second, two trends were observed: between July 2013 and November 2013 (protocol initiation) and again between January 2015 and June 2015 (after protocol reeducation). Both periods were evidence of significantly improving performance in distinct time periods. Such findings indicate that the improvement in delayed cord clamping adherence was not the result of chance.
Next, we identified factors associated with adherence to the delayed cord clamping protocol in the postprotocol period. Maternal demographic and clinical characteristics were similar between women who did and did not receive delayed cord clamping (Table 1). The mean gestational age for neonates who received delayed cord clamping was not significantly different than the mean gestational age for those who did not in the postprotocol group (29.5±2.0 weeks compared with 29.2+2.8 weeks of gestation, P=.94). However, on bivariable analysis, women who received delayed cord clamping were less likely to have undergone cesarean delivery (n=28 [28.0%] compared with n=40 [46.0%], P=.01) or to have had meconium-stained amniotic fluid (n=1 [1.0%] compared with n=6 [6.9%], P=.03). After adjusting for potential confounders, only cesarean delivery remained independently associated with a decreased likelihood of receiving delayed cord clamping (adjusted odds ratio 0.49, 95% confidence interval 0.25–0.96). There were no differences in delayed cord clamping adherence by maternal age, parity, body mass index, race or ethnicity, insurance, or gestational age at delivery. In addition, there were no differences in delayed cord clamping adherence by clinical characteristics including use of neuraxial analgesia, receipt of antibiotics, magnesium or betamethasone, birth weight, 1-minute or 5-minute Apgar score, clinical diagnosis of abruption, or chorioamnionitis.
We then assessed whether neonates born after protocol introduction had outcomes different from those born before the protocol, regardless of protocol adherence. There were no significant differences in most maternal demographics, obstetric characteristics, or gestational age between the preprotocol and postprotocol cohorts (Table 2). Women in the postprotocol cohort, however, were more likely to have received neuraxial analgesia and magnesium neuroprophylaxis (88.2% compared with 77.9%, P=.005; 94.7% compared with 85.4%, P=.002, respectively). On bivariable analyses, neonates in the postprotocol cohort were significantly less likely to require pharmacologic blood pressure support (Table 3). However, after adjustment for potential confounders, the presence of a new protocol, regardless of protocol adherence, did not have a statistically significant effect on neonatal outcomes between the preprotocol and postprotocol cohorts.
Lastly, to assess the effect of delayed cord clamping adherence on neonatal outcomes, we performed a per-protocol analysis of neonatal outcomes comparing all neonates born preprotocol with the subgroup of neonates born postprotocol who actually received the intervention (Table 4). Neonates born after the protocol who received delayed cord clamping had higher birth hematocrit, fewer total number of red blood cell transfusions, lower frequency of red blood cell transfusions in the first week of life, and lower frequency of pressor support. In multivariable analysis, mean birth hematocrit, total number of blood transfusions, and proportion of neonates requiring blood transfusion in the first week of life remained statistically significantly improved for neonates receiving delayed cord clamping. No other neonatal outcomes were different between groups. Measures of potential harm in delayed cord clamping recipients such as neonatal peak bilirubin were not statistically significant.
Since the release of recommendations for delayed cord clamping by the College,10 many hospitals have adopted institution-wide protocols. However, this intervention has yet to be adopted uniformly by all health care providers for preterm neonates as a result of ongoing questions regarding the balance of need for immediate neonatal resuscitative efforts with benefits of delayed cord clamping. Our study accomplishes two goals with regard to this question. First, it addresses the quality improvement issue of health care provider adherence to the delayed cord clamping protocol within a single institution. We identified a pattern of improvement with delayed cord clamping performance after health care provider education and determined that cesarean delivery was an important factor that negatively influenced health care provider adherence with delayed cord clamping. Second, it provides further evidence in a real-life, large-volume setting that introduction of a standardized protocol for performing delayed cord clamping in preterm, singleton neonates at less than 32 weeks of gestation is feasible and associated with neonatal clinical benefits without evidence of adverse outcomes.
Previous randomized trials and meta-analyses have demonstrated the benefits of performing delayed cord clamping in the preterm neonate.18,19 Similar to prior work, we found delayed cord clamping is associated with a higher birth hematocrit and reduced need for red blood cell transfusions in the first week of life compared with neonates born before the initiation of the protocol. Furthermore, we identified that delayed cord clamping is associated with a decrease in the total number of red blood cell transfusions required over the course of a neonate's hospitalization.20,21 Multivariable analyses could not be performed for the outcomes of neonatal death, need for pharmacologic blood pressure support, or cord pH because there were no neonates in the cohort who received delayed cord clamping. Increases in magnesium prophylaxis administration and neuraxial analgesia across these two time periods may reflect effective obstetric change over time despite the absence of any new or updated institutional practices regarding magnesium sulfate or neuraxial analgesia use during the study period and were controlled for in analyzing neonatal outcomes. No other major institutional protocols changed during this period, suggesting the findings are not the result of other policy changes. Notably, these differences were not evident when comparing all neonates in the postprotocol period (ie, including those who did not receive delayed cord clamping despite eligibility) with those in the preprotocol period, which suggests the differences identified in the per-protocol subgroup analysis can be related to performance of delayed cord clamping. Such findings suggest there may be even greater improvement in overall neonatal outcomes if adherence to the protocol improved.
Moreover, although delayed cord clamping has been demonstrated to benefit premature neonates, data in nontrial settings are lacking. Importantly, our data suggest benefits and lack of adverse outcomes associated with delayed cord clamping in a nontrial setting similar to those in a trial setting.7,8,18,19 Furthermore, our study suggests delayed cord clamping was not associated with inability to collect cord blood gases, because cord blood gases were still recorded for more than 70% of patients undergoing delayed cord clamping.
A central goal of this study was to identify areas for quality improvement and health care provider education. As is often seen with quality improvement initiatives, intermittent decrements in adherence occurred over time that were then reversed with further education. Continuous reeducation and support for delayed cord clamping are necessary to sustain these quality improvement changes. Importantly, cesarean delivery was the only factor significantly associated with failing to receive delayed cord clamping. We identified no difference in adherence to the protocol based on maternal demographic factors, clinical characteristics, or antepartum or intrapartum care, suggesting ongoing interventions to improve protocol performance should be directed at staff education during cesarean deliveries. We hypothesize that cesarean deliveries are associated with decreased likelihood of performing delayed cord clamping as a result of health care provider ambiguity toward optimal positioning of the neonate, technique, and duration and timing of the procedure associated with this mode of delivery. Furthermore, health care providers may have concerns about maternal bleeding from an open hysterotomy during the delay. Physician and nursing education on the neonatal benefits of delayed cord clamping regardless of route of delivery, including absence of data on maternal harm with delayed cord clamping during cesarean delivery, likely will be essential to increasing health care provider comfort with and adherence to the protocol.
This study is unique in that it was performed at a high-volume urban academic center with a diverse patient population. Additionally, although prior data have demonstrated the efficacy of delayed cord clamping in the setting of clinical trials, this study demonstrates the effectiveness of delayed cord clamping and the barriers to its use in a nontrial setting. Moreover, this institution is a high-volume delivery unit, with more than 120 delivering attending obstetricians, 51 residents and fellows, and hundreds of nursing staff, thus requiring substantial resources to introduce and disseminate a new practice. The demonstration of increasingly successful protocol introduction in this setting suggests such techniques can be applied in other similarly large and diverse settings. At the same time, our data show the sustained quality improvement efforts required to ensure reasonable, if not ideal, adherence to a new protocol. Our institution's experience may offer lessons for other institutions who are implementing or expanding a delayed cord clamping protocol. For example, the close collaboration of obstetrics, neonatology, and nursing was key to improving adherence; leadership from each of these clinical arenas frequently reinforced the importance of delayed cord clamping, performed specialty-specific education, and worked together to solve problems with protocol uptake. The inclusion of both formal (eg, grand rounds) and informal (eg, didactic lectures and discussions at team labor board sign-out) was also critical to improvement. Lastly, we did not perform individual health care provider feedback on performance of delayed cord clamping; one potential area for improvement would be to offer individual health care providers personalized feedback on their adherence to the protocol.
However, there are a number of limitations to consider. Although we attempted to account for significant differences between the preprotocol and postprotocol cohorts in the analyses, there may have been other unmeasured changes in practices over the 4-year study period for which we could not account, but there were no explicit policy changes in this time period. Unlike some previous trials, our analysis did not show a significant association between delayed cord clamping and the incidence of BPD, necrotizing enterocolitis, late-onset sepsis, or grade III or IV intraventricular hemorrhage22,23; however, our study was not powered to detect differences in these less common outcomes and thus the negative findings may not be generalizable. In addition to the set of exclusion criteria for the study, neonates at less than 32 weeks of gestation with major fetal anomalies, who were ineligible for intervention, or who required urgent intensive care were immediately transferred to the Ann & Robert H. Lurie Children's Hospital after birth and were not included in the study as a result of lack of available outcomes data after their transfer. Furthermore, our data cannot identify reasons why delayed cord clamping was undertaken less frequently during cesarean deliveries and whether there may have been additional reasons beyond health care provider knowledge and comfort. Additionally, these data do not address the optimal timing and duration of delayed cord clamping, because this may affect the benefit and outcomes of the procedure.23,24 Although previous trials have cited between 30 and 90 seconds of delayed cord clamping, this is largely based on institution or health care provider preference. Comparison of duration of cord clamping on neonatal outcomes in future studies may address whether timing of the procedure results in increased benefit for the neonate. Finally, incorporating specific health care provider demographic data would be a valuable addition to the analysis and should be considered in future studies. A limitation of these data is the lack of specific health care provider demographic details; however, all health care providers in this study are attending obstetricians with resident involvement in all preterm deliveries. Examining individual health care provider compliance rates to delayed cord clamping and directing education efforts toward these health care providers may further improve adherence.
In summary, our study provides further evidence that a delayed cord clamping protocol can be implemented in a high-volume academic setting and that, when used, is associated with benefits in the very preterm neonate. Education on delayed cord clamping, especially regarding performance during cesarean deliveries, may be required to improve health care provider adherence to the procedure.
1. WHO, March of Dimes, Partnership for Maternal, Newborn & Child Health, Save the Children. Born too soon: the global action report on preterm birth. Available at: www.who.int/maternal_child_adolescent/documents/born_too_soon/en
. Retrieved July 5, 2016.
2. Mathews TJ, MacDorman MF. Infant mortality statistics from the 2005 period linked birth/infant death data set. Natl Vital Stat Rep 2008;57:1–32.
3. Friedrich MJ. Premature birth complications top cause of death in children younger than 5 years. JAMA 2015;313:235.
4. Stoll BJ, Hansen NI, Bell EF, Shankaran S, Laptook AR, Walsh MC, et al. Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics 2010;126:443–56.
5. Fanaroff AA, Stoll BJ, Wright LL, Carlo WA, Ehrenkranz RA, Stark AR, et al. Trends in neonatal morbidity and mortality for very low birthweight infants. Am J Obstet Gynecol 2007;196:147.e1–8.
6. McDonald SJ, Middleton P. Effect of timing of umbilical cord clamping of term infants on maternal and neonatal outcomes. The Cochrane Database of Systematic Reviews 2008, Issue 2. Art. No.: CD004074. DOI: 10.1002/14651858.CD004074.pub2.
7. Chiruvolu A, Tolia VN, Qin H, Stone GL, Rich D, Conant RJ, et al. Effect of delayed cord clamping on very preterm infants. Am J Obstet Gynecol 2015;213:676.e1–7.
8. Aziz K, Chinnery H, Lacaze-Masmonteil T. A single center experience of implementing delayed cord clamping in babies born at less than 33 weeks' gestational age. Adv Neonatal Care 2012;12:371–6.
9. Rabe H, Diaz-Rossello JL, Duley L, Dowswell T. Effect of timing of umbilical cord clamping and other strategies to influence placental transfusion at preterm birth on maternal and infant outcomes. The Cochrane Database of Systematic Reviews 2012, Issue 8. Art. No.: CD003248. DOI: 10.1002/14651858.CD003248.pub3.
10. Timing of umbilical cord clamping after birth. Committee Opinion No. 543. American College of Obstetricians and Gynecologists. Obstet Gynecol 2012;120:1522–6.
11. Delayed umbilical cord clamping after birth. Committee Opinion No. 684. American College of Obstetricians and Gynecologists. Obstet Gynecol 2017;129:e5–10.
12. American Academy of Pediatrics. Statement of endorsement: timing of umbilical cord clamping after birth. Pediatrics 2013;131:e1323.
13. Jelin AC, Zlatnik MG, Kuppermann M, Gregorich SE, Nakagawa S, Clyman R. Clamp late and maintain perfusion (CLAMP) policy: delayed cord clamping in preterm infants. J Matern Fetal Neonatal Med 2016;29:1705–9.
14. Umbilical cord blood banking. ACOG Committee Opinion No. 648. American College of Obstetricians and Gynecologists. Obstet Gynecol 2015;126:e127–29.
15. Perla RJ, Provost LP, Murray SK. The run chart: a simple analytical tool for learning from variation in healthcare processes. BMJ Qual Saf 2011;20:46–51.
16. Bell MH, Temberg JL, Felgin RD, Keating JP, Marshall R, Barton L, et al. Neonatal necrotizing enterocolitis: therapeutic decisions based upon clinical staging. Ann Surg 1978;187:1–7.
17. Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34.
18. Jahazi A, Kordi M, Mirbehbahani NB, Mazloom SR. The effect of early and late umbilical cord clamping on neonatal hematocrit. J Perinatol 2008;28:523–5.
19. Kugelman A, Borenstein-Levin L, Riskin A, Chistyakov I, Ohel G, Gonen R, et al. Immediate versus delayed umbilical cord clamping in premature neonates born <35 weeks: a prospective, randomized, controlled study. Am J Perinatol 2007;24:307–15.
20. Mercer J, Erickson-Owens D. Delayed cord clamping increases infants' iron stores. Lancet 2006;367:1956–8.
21. Sanberg PR, Park DH, Borlongan CV. Stem cell transplants at childbirth. Stem Cell Rev 2010;6:27–30.
22. Mercer JS, Vohr BR, McGrath MM, Padbury JF, Wallach M, Oh W. Delayed cord clamping in very preterm infants reduces the incidence of intraventricular hemorrhage and late-onset sepsis: a randomized, controlled trial. Pediatrics 2006;117:1235–42.
23. Philip AG. Delayed cord clamping in preterm infants. Pediatrics 2006;117:1434–5.
24. Levy T, Blickstein I. Timing of cord clamping revisited. J Perinat Med 2006;34:293–7.