http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, Early Online: 1–4
! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.971746
ORIGINAL ARTICLE
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Washington University Library on 01/15/15 For personal use only.
Packed red blood cell transfusion (PRBC) attenuates intestinal blood flow responses to feedings in pre-term neonates with normalization at 24 hours Aaron Pitzele1, Mohammad Rahimi1, Eric Armbrecht2, and Thomas Havranek3 1
Department of Pediatrics, 2Center for Outcomes Research, Saint Louis University School of Medicine, St. Louis, MO, USA, and 3Department of Pediatrics/Neonatology, Albert Einstein University/Montefiore Medical Center, Bronx, NY, USA Abstract
Keywords
Objective: To determine whether packed red blood cell (PRBC) transfusion affects post-prandial superior mesenteric artery blood flow velocities (SMA BFVs) in very-low birth weight (VLBW) neonates and if so, at what time point after transfusion restoration of previous SMA BFV patterns occurs. Design/Methods: VLBW pre-term neonates, older than 14 days and tolerating bolus enteral feedings administered every 3 h were enrolled in this prospective observational study. Pulsed Doppler ultrasound was used to measure pre- and post-prandial (at 45 min) time-averaged mean, peak and end diastolic velocities (TAMV, PSV, EDV) immediately before and after 15 ml/kg of PRBC transfusion was given over 3 h; patent ductus arteriosus (PDA) status was also evaluated. Subsequent pre- and post-prandial SMA BFVs were recorded 24 and 48 h after the transfusion. Results: Pre- and post-prandial measurements were obtained for 21 out of 25 enrolled infants. Post-prandial SMA BFVs were attenuated during the feedings immediately after transfusion; at 24 and 48 h after transfusion, changes in post-prandial SMA BFVs were similar to those measured prior to transfusion; the presence of the PDA did not affect results. Conclusions: PRBC transfusion blunted SMA BFV responses to feedings immediately after the transfusion with normalization observed 24 h post-transfusion.
Doppler ultrasound, intestinal blood flow, packed red blood cell transfusion, preterm neonate
Introduction Necrotizing enterocolitis (NEC) is a complication of prematurity with high morbidity and mortality. About 7 to 13% of all very low birth weight infants admitted to neonatal intensive care units (NICU) develop NEC, with mortality ranging from 10 to 44% [1–3]. The incidence of NEC has remained mostly unchanged over the years [4,5]. NEC is considered as a multifactorial disorder with final common clinical manifestation related to a variety of etiologic mechanisms, including ischemia/reperfusion, infection, increased viscosity, rapid advancement of enteral intake as well as immunological dysfunction [1,6–8]. Not surprisingly, there is no definite agreement on causation [9]. Recently, packed red blood cell (PRBC) transfusion has emerged as a potential causative factor in the development of NEC. Several retrospective studies report increased incidence of NEC 22 h [10] or 48–72 h [11] after PRBC transfusion and increased odds of NEC development within 48 h post-transfusion [12]. Singh et al. [13] measured the strength Address for correspondence: Thomas Havranek, MD, Albert Einstein University/Montefiore Medical Center, 1825 Eastchester Road, Bronx, NY, USA. Tel: +718-904-4105. Fax: +718-904-2659. E-mail:
[email protected]
History Received 20 June 2014 Revised 9 September 2014 Accepted 19 September 2014 Published online 20 October 2014
of association between the NEC and PRBC transfusion, and reported the association as strong, less strong and absent for periods of 524 h, 548 h and at 96 h, respectively. Importantly, the majority of the infants in these studies were stable premature neonates on full enteral feeds, who decompensated and developed NEC after being transfused. Based on limited or no evidence, many units have established policies urging the withholding of feedings in premature infants during the blood transfusion [14]. A small recent prospective trial [11] reported a decreased incidence of NEC (from 5.3 to 1.3%) when infants were not fed during the transfusion. Similarly, limited investigation of the superior mesenteric artery blood flow velocities (SMA BFVs) revealed that the expected postprandial increase in SMA BFV disappeared following PRBC transfusion, placing the fed neonates receiving blood transfusion at higher risk for NEC [15]. In our study, we intended to further our understanding of the hemodynamic consequences of PRBC transfusion in more complex, VLBW neonates at high risk of developing transfusion-associated necrotizing enterocolitis by measuring pre- and post-prandial SMA BFV in neonates before and at several time points after PRBC transfusion. We hypothesized that infants would have attenuated post-prandial SMA BFVs immediately after versus before PRBC transfusion.
2
A. Pitzele et al.
In addition, we hypothesized that SMA BFVs would return to pre-transfusion values 24–48 h after transfusion.
Methods The study was approved by Institutional Review Board at Saint Louis University and all infants were enrolled after informed consent was obtained from the patents.
J Matern Fetal Neonatal Med, Early Online: 1–4
to the ostium of the left pulmonary artery (PDA:LPA ratio). A ratio 1 defined a large PDA, 0.5 but 51 a moderate PDA, and 50.5 a small PDA [16]. All SMA BFV measurements as well as PDA evaluation was performed by one of the two ECHO technicians, not familiar with the clinical care of the infants. Precision studies revealed great correlation between the two technicians (r ¼ 0.93, p50.001), with the intraobserver variability of 4 and 4.8%.
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Washington University Library on 01/15/15 For personal use only.
Study design Statistical analyses
Infants in this prospective observational trial were enrolled from February 2012 to August 2013 at Cardinal Glennon Children’s Medical Center/Saint Louis University. SMA BFVs in VLBW neonates who were receiving bolus feedings over 30 min for every 3 h were recorded pre- and postprandially (at 45 min from conclusion of feeding) immediately before and after 15 ml/kg of PRBC transfusion was given over 3 h; neonates were not fed during the transfusion. Subsequent pre- and post-prandial SMA BFVs were recorded 24 and 48 h after the transfusion; patent ductus arteriosus (PDA) status was evaluated as well. The clinical team taking care of the neonate made the decision about the need for PRBC transfusion. Fortified maternal or donor breast-milk was administered to all subjects during all measurements. Clinical data were collected prospectively and included: feeding volume, daily enteral intake, development of NEC (Bell Stage 2), hemoglobin level before and after transfusion, phototherapy, caffeine administration, birth weight, gestational age and vital signs.
Descriptive statistics were calculated for the cohort’s demographic characteristics (e.g. age, gender, birth weight), clinical conditions (e.g. PDA, hemoglobin) and treatment (e.g. feeding volume, caffeine, phototherapy) variables. Preand post-transfusion variables were reported. The pre- and post-prandial SMA BFVs at four different time points relative to PRBC transfusion were measured: pre-transfusion, immediately post-transfusion (3 h), 24 h post-transfusion and 48 h post-transfusion. Blood flow velocity was recorded with three different pulsed Doppler measurements: time average mean (TAMV), peak (PSV) and end (EDV) diastolic velocities. The paired t-test, with alpha 0.05, was employed to assess change in SMA BFV for each time point and Doppler measurement combination separately. The potential for effect modification by PDA was assessed with a stratified analysis with the nonparametric Mann-Whitney U-test at each time point by comparing percent change in SMA BFV for patients with and without PDA.
Subjects
Results
Pre-term neonates were recruited from the NICU at Cardinal Glennon Children’s Medical Center. Inclusion criteria included birth weight 1500 g, singleton gestation, appropriate birth weight for gestational age, tolerance of 20 ml/kg/ day of feeding volumes and 36 weeks corrected gestational age at the time of transfusion. Exclusion criteria include major congenital or chromosomal anomalies, presence of congenital heart disease, shock, vasopressor administration, history of NEC Bell stage 2 or greater. Infants experiencing rare changes in vital signs or oxygen level drop during the Doppler ultrasound measurements needing intervention (such as oxygen increase or stimulation) had studies discontinued and were excluded from further analysis.
Twenty-five pre-term neonates were enrolled; two infants were made NPO during the study protocol and two had incomplete SMA BFV recordings. The mean birth weight and gestational age of the remaining 21 infants were 819 ± 240 g and 26.1 ± 1.6 weeks, respectively (Table 1). None of the 21
Doppler ultrasound studies and PDA evaluation SMA BFV was measured by pulsed Doppler ultrasound (GE Healthcare, Waukesha, WI) using a real time two-dimensional image from a longitudinal abdominal approach, using a 10.0MHz transducer. The sampling volume of the pulsed Doppler was placed 3 mm distal to the origin of the SMA, insonation angles were kept 525 . When five stable consecutive wave forms (five cardiac cycles) were obtained, the peak systolic blood flow velocity (PSV), the time-averaged mean blood flow velocity (TAMV), the end-diastolic blood flow velocity (EDV) were reported as the mean value from five consecutive wave forms. SMA BFV was recorded pre- and 45 min post-prandially at the above-mentioned time points. The size of the PDA, if present, was defined as small, moderate or large on the basis of the ratio of the smallest ductal diameter
Table 1. Infant demographic characteristics (n ¼ 21). Characteristics Birth weight (g) Mean Range Gestational age (weeks) Mean Range Age at time of study (days) Mean Range Feeding volume 1 (per feeding) (ml) Mean Range Feeding volume (per day) (ml/kg/day) Mean Range Pre-transfusion hemoglobin (g/dl) Mean Range Post-transfusion hemoglobin (g/dl) Mean Range Patent ductus arteriosus present, n (%) Phototherapy, n (%) Caffeine, n (%)
Value 819 ± 240 505–1450 26 ± 1.6 24–29 1/7 34.8 ± 14 16–65 21.5 ± 7.8 9–38 140 ± 31.3 50–171 8.9 ± 0.9 7–10.2 12.8 ± 1.2 10.8–15.5 9 (43) 0 (0) 20 (95)
Effects of PRBC on post-prandial intestinal blood flow
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Washington University Library on 01/15/15 For personal use only.
DOI: 10.3109/14767058.2014.971746
infants died or developed NEC. Infants were 34.8 ± 14 days old when the first SMA BFV measurements were made. The mean feeding volume was 21.5 ± 7.8 ml. Pre- and posttransfusion hemoglobin values were 8.9 ± 0.9 and 12.8 ± 1.2 g/dl; change in Hgb levels (delta Hgb) was 3.5 ± 0.7 g/dl. The time interval between pre-transfusion mesenteric feed study and the onset of the transfusion was 105 ± 12 min for the pre-prandial measurements and 15 ± 8 min for the post-prandial measurements. Feeds were initiated again 85 ± 11 min after the end of the transfusion (time intervals were 66 ± 9 min for the pre-prandial studies and 164 ± 10 min for the post-prandial studies). Clinical variables that may affect SMA BFV are listed in Table 1. There were no significant differences in SMA BFV between infants who did or did not have a PDA. Pre-transfusion TAMVs and PSVs were significantly increased from pre-prandial to post-prandial state, then became non-significant in immediate post-transfusion state. The differences between pre- and post-prandial TAMVs
Table 2. Pre- and post-prandial superior mesenteric artery blood flow velocities at different time points.
Pre-transfusion Pre-prandial Post-prandial p value Immediately Post-transfusion Pre-prandial Post-prandial p value 24 h Post-transfusion Pre-prandial Post-prandial p value 48 h Post-transfusion Pre-prandial Post-prandial p value
PSV (n ¼ 21) Mean Std Dev
EDV (n ¼ 21) Mean Std Dev
TAMV (n ¼ 21) Mean Std Dev
76.7 ± 22.2 89.8 ± 22.9 0.001**
10.7 ± 4.0 11.8 ± 5.1 0.133
34.2 ± 9.6 41.6 ± 10.8 0.001**
71.9 ± 16.3 82.9 ± 22.3 0.079
11.5 ± 5.1 13.1 ± 6.1 0.257
34.2 ± 9.0 37.5 ± 12.1 0.220
70.4 ± 17.3 85.2 ± 22.3 0.008**
11.2 ± 4.3 13.1 ± 5.9 0.118
31.7 ± 8.0 38.1 ± 8.8 0.004**
75.1 ± 19.4 87.8 ± 31.1 0.076
13.0 ± 5.4 13.5 ± 4.3 0.664
34.6 ± 8.6 40.2 ± 12.1 0.014**
**p value was significant for 50.05. Figure 1. Pre- and post-prandial time-averaged mean blood flow velocity at different time points relative to PRBC transfusion.
3
and PSVs then became again highly significant at 24 h post-transfusion time point; 48 h post-transfusion TAMVs displayed statistical significance, PSV changes were nonsignificant (Table 2). The differences between pre- and postprandial EDVs were not significant at any study time point. TAMV changes are illustrated in Figure 1. PDA evaluation was performed on all 21 subjects. Twelve had no PDA detected, six had a small PDA, for total of 18 newborns with PDA:LPA ratio50.5 (no/small PDA category). One newborn had a moderate PDA (ratio 0.5 but 51.0) and two had large PDAs (ratio 1.0). In our limited sample size, the presence of a PDA had no effect on postprandial SMA BFVs.
Discussion The main findings of our study are normal post-prandial responses to feedings in the anemic, pre-transfusion state with attenuated, blunted response during the feeding immediately after the PRBC transfusion. Intestinal BFV responses to feedings were normalized at 24 and 48 h after transfusion. Our study results are best compared to the trial by Krimmel et al. [15]. While we used a different study design and enrolled exclusively VLBW neonates who had lower mean birth weights and gestational ages, we demonstrated normal post-prandial responses prior to PRBC transfusion. This is contrary to the Krimmel’s data which reported abnormal feeding responses in the anemic, pre-transfusion state in the subset of infants whose weight was 51250 g. While the degree of anemia and the sample size of both studies were similar, we did not perform the subgroup analysis to confirm or refute the findings in the anemic state. In addition, all of our infants were fed exclusively with fortified maternal or donor breast milk, which have different effects on SMA BFV compared to formula diet [16]. Importantly, we report a highly significant post-prandial increase in PSVs and TAMVs 24 h and 48 h after transfusion with no change in EDVs. Similarly, Krimmel et al. demonstrated significant post-prandial increases in PSVs without changes in EDVs and mean blood flow velocities at 48 and 96 h. These combined data suggest that normalization of postprandial responses after immediate post-transfusion
J Matern Fetal Neonatal Med Downloaded from informahealthcare.com by Washington University Library on 01/15/15 For personal use only.
4
A. Pitzele et al.
‘‘blunting’’ may occur within 3 and 24 h after transfusion, which may have clinical implications for feedings administration. Interestingly, while the presence of a moderate PDA affected post-prandial responses to feedings in an animal model [17], we were unable to demonstrate an effect in our cohort of VLBW neonates. In addition, recently published data from our group [18] suggest that the presence of a small or moderate PDA did not affect post-prandial changes in mesenteric perfusion, contrary to the presence of large PDA (PDA:LPA ratio41.0). In our study, only 2 out of 9 PDA infants had a large PDA and 1 had a moderate PDA, while the majority (six patients) had a small PDA (PDA:LPA ratio50.5), possibly explaining the lack of a combined PDA effect on post-prandial SMA BFVs in our group. While the effect of feeding volume on SMA BFVs has been described previously [19], we did not analyze this variable, since the majority of the patients had feeding volumes unchanged over the 48-hour measurement period and the entire group was receiving large volumes of feeds (140 ± 31 ml/kg/day) during the SMA BFVs measurements. As mentioned previously, all infants were fed with fortified maternal or donor breast milk and this variable remained constant throughout the study period. In conclusion, this study confirms that PRBC transfusion adversely affects post-prandial responses to feedings 3 h posttransfusion with normalization of mesenteric perfusion at 24 h. Future studies, using Doppler ultrasound or recently validated near-infrared spectroscopy [20] should more precisely define the time period needed for restoration of pretransfusion feeding responses. This information will help clinicians manage feeding initiation after PRBC transfusion administration.
Declaration of interest The authors report no declaration of interest.
References 1. Stoll BJ. Epidemiology of necrotizing enterocolitis. Clin Perinatol 1994;21:205. 2. Luig M, Lui K. Epidemiology of necrotizing enterocolitis-part II: risk and susceptibility of premature infants during the surfactant era: regional study. J Paediatr Child Health 2005;41:174–9.
J Matern Fetal Neonatal Med, Early Online: 1–4
3. Holman RC, Stoll BJ, Clarke MJ, Glass RI. The epidemiology of necrotizing enterocolitis infant mortality in the United States. Am J Public Health 1997;87:2026–31. 4. Fanaroff AA, Hack M, Walsh MC. The NICHD neonatal research network: changes in practice and outcomes during the first 15 years. Semin Perinatol 2003;27:281–7. 5. Srinivasan PS, Brandler MD, D’Souza A. Necrotizing enterocolitis. Clin Perinatol 2008;35:251–72. 6. Simmonds A, LaGamma E. Addressing the ‘‘new’’ NEC, part I: rediscovering the basics. Indian J Pediatr 2006;73:1011–18. 7. La Gamma EF, Browne LE. Feeding practives for infants weighing less than 1500 g at birth and the pathogenesis of necrotizing enterocolitis. Clin Perinatol 1994;21:271–306. 8. Uauy R, Fanaroff A, Korones S, et al. Necrotizing enterocolitis in very low birth weight: biodemographic and clinical correlates: National Institute of Child Health and Human Development Neonatal Research Network. J Pediatr 1991;119:630–8. 9. Obladen M. Necrotizing enterocolitis: 150 years of fruitless search for the cause. Neonatology 2009;96:203–10. 10. Mally P, Golombek SG, Mishra R, et al. Association of necrotizing enterocolitis with elective packed red blood cell transfusions in stable, growing, premature neonates. Am J Perinatol 2006;23: 451–8. 11. El-Dib M, Narang S, Lee E, et al. Red blood cell transfusion, feeding and necrotizing enterocolitis in preterm infants. J Perinatol 2011;31:183–7. 12. Paul DA, Mackley A, Novitsky A, et al. Increased odds of necrotizing enterocolitis after transfusion of red blood cells in premature infants. Pediatrics 2011;127:635–41. 13. Singh R, Visintainer PF, Frantz ID, et al. Association of necrotizing enterocolitis with anemia and packed red blood cell transfusions in preterm infants. J Perinatol 2011;31:176–82. 14. Agwu JC, Narchi H. In a preterm infant, does blood transfusion increase the risk of necrotizing enterocolitis? Arch Dis Child 2005; 90:102–3. 15. Krimmel GA, Baker R, Yanowitz TD. Blood transfusion alters the superior mesenteric artery blood flow velocity response to feeding in premature infants. Am J Perinatol 2009;26:99–105. 16. Ozkan H, Oren H, Erdag N, Cevik N. Breast milk versus infant formulas: effects on intestinal blood flow in neonates. Indian J Pediatr 1994;61:703–9. 17. McCurnin D, Clyman RI. Effects of a patent ductus arteriosus on postprandial mesenteric perfusion in premature baboons. Pediatrics 2008;122:e1262–7. 18. Havranek T, Rahimi M, Hall H, Armbrecht E. Feeding preterm neonates with patent ductus arteriosus (PDA): intestinal blood flow characteristics and clinical outcomes. J Matern Fetal Neonatal Med 2014;29:1–5 [Epub ahead of print]. 19. Leidig E. Doppler analysis of superior mesenteric artery blood flow in preterm infants. Arch Dis Child 1989;64:476–80. 20. Gillam-Krakauer M, Cochran CM, Slaughter JC, et al. Correlation of abdominal rSO2 with superior mesenteric artery velocities in preterm infants. J Perinatol 2013;33:609–12.
