Journal of Perinatology (2014) 34, 819–822 © 2014 Nature America, Inc. All rights reserved 0743-8346/14 www.nature.com/jp

ORIGINAL ARTICLE

Antenatal magnesium sulfate and spontaneous intestinal perforation in infants less than 25 weeks gestation BN Rattray1, DM Kraus2, LR Drinker1, RN Goldberg1, DT Tanaka1 and CM Cotten1 OBJECTIVE: Evaluate spontaneous intestinal perforation (SIP)/death among extremely low birthweight (ELBW) infants before, during and after initiation of an antenatal magnesium for neuroprotection protocol (MgPro). STUDY DESIGN: We tested associations between SIP/death and magnesium exposure, gestational age (GA) and interactions with GA and magnesium exposure in a cohort of inborn ELBW infants before, during and after MgPro. RESULT: One hundred and fifty-five ELBW infants were included, 81 before, 23 during and 51 after MgPro. ELBW infants (78.3%) were exposed to Mg during MgPro compared with 50.6% and 60.8% before and after, respectively. Incidence of SIP on protocol was 30.4% vs 12.9% off protocol (P = 0.03). GA was strongly associated with SIP (P o0.01). Antenatal Mg dose was also associated with SIP/death regardless of epoch (odds ratio 9.3 (1.04–104.6)), but increased SIP/death was limited to those o25 weeks gestation. CONCLUSION: Higher Mg dose was associated with higher SIP and death risk among infants with the lowest birthweights. Validation of this observation in larger populations is warranted. Journal of Perinatology (2014) 34, 819–822; doi:10.1038/jp.2014.106; published online 5 June 2014

INTRODUCTION Magnesium sulfate is widely used in obstetrics as an anticonvulsant for the treatment of eclampsia, prevention of eclampsia in women with preeclampsia and as a tocolytic despite controversy regarding its efficacy.1 Several observational studies suggest a lower prevalence of cerebral palsy in very low birth weight infants exposed to in utero magnesium sulfate.2,3 However, other observational studies and one small randomized control trial failed to find a protective effect.4–9 Hope for a neurologic outcome benefit led to three randomized clinical trials between 2003 and 2008. The studies used different antenatal magnesium dosing strategies and demonstrated varying neuroprotective efficacy.10–12 The data from the multicenter clinical trials prompted the Duke Maternal–Fetal Medicine (MFM) group to begin a Magnesium Sulfate for Neuroprotection Protocol in 2010 (MgPro). After initiation of this protocol at our center, neonatologists noted an increase in the rate of spontaneous intestinal perforation (SIP) and subsequent mortality in the most extremely preterm infants, prompting suspension of the protocol while we collaboratively examined possible explanations and potential risk factors for SIP and mortality in this population other than MgPro. The hiatus also allowed for a period of observation to see if SIP/mortality would return to the pre-MgPro baseline rate. Our objective for this report was to examine changes in exposure to Mg before, during and after MgPro, and assess potential interactions between Mg and other SIP risk factors in our center and their influence on risk of SIP or death. Additionally, after noting higher rates of SIP during MgPro, we hypothesized that total antenatal Mg doses might be higher during MgPro. As a result, we assessed Mg dose effect on risk for SIP and mortality, regardless of epoch.

METHODS We conducted a cohort study of singleton and twin gestation extremely low birth weight infants (o1000 g birth weight) born at Duke University Medical Center (Durham, NC, USA) and admitted to the Duke Neonatal Intensive Care Unit. The following three groups were defined: those born before magnesium sulfate for neuroprotection protocol (MgPro) (1 January 2009 to 14 July 2010); during MgPro (15 July 2010 to 30 November 2010); and after MgPro (1 January 2011 to 30 October 2011). This study was approved by the Duke University Medical Center Institutional Review Board. MgPro was based on the study intervention used by the National Institute of Child Health and Human Development-MFM Units Network (NICHD-MFMU) randomized, controlled trial of magnesium sulfate for the prevention of cerebral palsy.12 Magnesium sulfate for neuroprotection was considered for all patients with singleton or twin gestations at 24 to 32 weeks at imminent risk for spontaneous delivery because of rupture of membranes at 22 to 32 weeks; advanced preterm labor with dilatation of 4 to 8 cm with intact membranes; or indicated preterm delivery anticipated within 2 to 24 h. A loading dose of 6 g followed by a maintenance dose of 2 g h − 1 was begun. If delivery did not occur after 12 h and risk of delivery was no longer imminent, magnesium was discontinued and resumed when delivery was deemed imminent. If >6 h had elapsed since discontinuation of magnesium, then a repeat loading dose was administered. All inborn extremely low birthweight infants were included, but infants born with major congenital malformations or chromosomal anomalies were excluded from the study. Using analysis of variance to test variation among the three epochs, we compared maternal data, including maternal age, ethnicity, weight, delivery type, presence of chorioamnionitis, preeclampsia, number of fetuses, indication for magnesium and total dose of magnesium for the three time periods. We also compared neonatal data including gestational age (GA), birthweight, sex, Apgar scores, antenatal steroid exposure, rates of intestinal perforation, medical or surgical necrotizing enterocolitis (grade IIA or higher by modified Bell’s criteria),13 postnatal hydrocortisone (HC) and dexamethasone exposure, indomethacin treatment, occurrence

1 Department of Pediatrics, Duke University Medical Center, Durham, NC, USA and 2Department of Maternal–Fetal Medicine/Obstetrics, Duke University Medical Center, Durham, NC, USA. Correspondence: Dr BN Rattray, Women’s Hospital of Greensboro, 801 Green Valley Road, Greensboro, NC 27408, USA. E-mail: [email protected] Received 21 February 2014; revised 19 April 2014; accepted 28 April 2014; published online 5 June 2014

Associations between SIP/death and Mg exposure BN Rattray et al

820 of patent ductus arteriosus, hypotension and death among the three periods. The primary outcome of our analysis was the incidence of death before discharge or SIP as defined either by surgical findings or radiographic evidence of intraperitonial air in the absence of pneumotosis. Multivariate logistic regression analysis was used to assess the influence of the independent variables, such as epoch, birth weight, GA, preeclampsia, exposure to antenatal steroids, postnatal HC and indomethacin exposure on neonatal outcomes. After the primary analyses, we used multivariable models using forward stepwise regression constructed to test for an interaction effect between Mg exposure and GA and SIP. Owing to concern that postnatal drug exposures may have a role in SIP, we built another model examining both neonatal HC and indomethacin exposure independently and as an interaction. Data analysis were conducted using JMP 10.0 SAS Institute (Cary, NC, USA). For statistical significance, a P-value o0.05 was regarded as significant.

RESULTS Of the 155 extremely low birthweight infants included, 81 (52%) were before, 23 (15%) during and 51 (33%) after MgPro. Maternal and neonatal characteristics (maternal age, preeclampsia, antenatal steroids, GA at delivery and BW) were similar among epochs (Table 1). Prevalence of antenatal magnesium exposure varied from 50.6%, 78.3% and 60.8% in the pre, during and after epochs, respectively (P = 0.05). For the neonatal outcomes, the primary outcome of SIP/death was most frequent during the MgPro period, but this was not statistically significant. SIP alone was more frequent during the MgPro epoch (30.4%) than the combined before and after epoch (12.9%) (P = 0.03) as was death associated with perforations (30.4% vs 12.1%) (P = 0.02). The incidence of perforations and death are shown in Table 2. Cumulative Mg dose per infant was higher during MgPro, 50.9 ± 45.7 g per infant vs 23.4 ± 34.9 g, before protocol and 28.1 ± 32.5 after protocol (P p 0.01). Total grams of magnesium as an independent variable for SIP/death was not significant

Table 1.

(P = 0.107); however, analysis of the entire cohort, before, during and after MgPro, revealed that the interaction between lower GA and increasing grams of antenatal magnesium was associated with higher risk of SIP/death (P o0.01). Figure 1 demonstrates this interaction, with doses of Mg having their greatest effect on infants of the lowest GA (Table 3). To examine the influence of postnatal medications known to add to risk of SIP, we examined the use of HC, prophylactic indomethacin and indomethacin treatment.14 We performed a multivariate analysis examining the independent effect of HC, indomethacin and Mg on risk of SIP or death. In this model, HC (P = 0.021) and magnesium dose as a function of GA (P = 0.006) both were associated with increased risk for SIP. DISCUSSION Analysis of our single center data demonstrated a higher rate of SIP and death during MgPro vs the before/after time epochs, with

Table 2.

Outcomes Before Mg (n = 81) (%)

During Mg (n = 23) (%)

After Mg (n = 51) (%)

P-values

18.5 12.4 11.1

30.4 30.4 30.4

23.5 13.7 13.7

0.45 0.09 0.07

Before+after Mg (n = 132)

During Mg (n = 23)

20.5 12.9 12.1

30.4 30.4 30.4

Death/SIP SIP Death

Death/SIP SIP Death

0.28 0.03 0.02

Abbreviations: MgPro, magnesium sulfate for neuroprotection protocol; SIP, spontaneous intestinal perforation.

Maternal and neonatal characteristics Before MgPro (n = 81)

During MgPro (n = 23)

After MgPro (n = 51)

Maternal characteristics Maternal age

29.1

28.8

29.7

Maternal race (%) Black White Hispanic Asian Other

50.6 43.2 0 2.5 3.7

56.5 26.1 8.7 4.4 4.4

41.2 47.1 5.9 3.9 2.0

% Mg recipients Magnesium Dose Preeclampsia (%) Chorioamnionitis Cesarean (%)

50.6 23.7 ± 36.4 27.4 38.8 75.3

78.3 50.9 ± 45.7 28.6 30.4 65.2

60.8 28.1 ± 32.5 20.0 29.4 70.6

26.3 ± 1.8 760.9 ± 126.1 54.3 19.8 95.1 61.7 21.0

26.2 ± 2.1 754.1 ± 158.1 60.9 8.7 100 65.2 39.1

26.5 ± 2.0 798.2 ± 133.3 43.1 19.6 94.1 72.6 25.5

Neonatal characteristics Gestational age Birthweight Gender (% male) % Multiples Antenatal steroids Postnatal hydrocortisone Postnatal NSAIDS

P-values 0.83

0.05 o0.01 0.59 0.50 0.60 0.74 0.23 0.29 0.45 0.51 0.44 0.21

Abbreviations: MgPro, magnesium sulfate for neuroprotection protocol; NSAID, nonsteroidal anti-inflammatory drug.

Journal of Perinatology (2014), 819 – 822

© 2014 Nature America, Inc.

Associations between SIP/death and Mg exposure BN Rattray et al

821

Figure 1. Interaction profile of death/perforation based on GA and magnesium exposure. Top: interaction profile of probability of spontaneous intestinal perforation or death by gestational age ⩽ 25 weeks (blue line) vs gestation age > 25 weeks (red line) based on magnesium dose in grams. Bottom: interaction profile of probability of spontaneous intestinal perforation or death by decreasing gestational age alone among infants without magnesium exposure (red line) vs probability of spontaneous intestinal perforation or death by decreasing gestational among infants with magnesium exposure up to 198 g (blue line). A color version of the figure is available online.

Table 3.

Multivariable analysis for primary outcome, SIP/death

GA Cumulative Mg

Mg dose × GA

OR

95% CI

6.0 9.3

2.4–18.2 1.04–104.6

LR

Prob>χ2

9.7

o0.01

Abbreviations: CI, confidence interval; GA, gestational age; OR, odds ratio; SIP, spontaneous intestinal perforation.

a suggestion that antenatal magnesium exerts a quantitative exposure effect that is strongly influenced by GA, rather than a general effect on infants of all GAs and birthweights. Three of the largest trials, the Australasian collaborative (ACTOMgSO4),10 European Trial (PREMAG)11 and the US NICHDMFMU,12 tested antenatal magnesium sulfate for neuroprophylaxis and suggest efficacy in reduction of cerebral palsy for the overall cohorts.15 The average birth weight and gestation of infants were higher than our identified highest risk group. None of the trials included intestinal perforation as an examined outcome, nor was this high-risk group analyzed separately. The ACTOMgSO4 study used a lower loading dose (4 g) and a lower follow-up dose (1 mg h − 1) than that of the NICHD-MFMU trial, and included infants born with lowest average birthweight (approximately 1000 g) and GA (27 weeks). Mortality was not statistically different between groups (13.8% vs 17.1%, P = 0.19).16 The PREMAG study single dose (4 g) trial included subjects with an average birthweight >1300 g. Mortality was not significantly different between the Mg group when compared with controls (9.4% vs 10.4%, P = 0.59).17 Birthweight and GA in the MFM-NICHD trial was >1400 g and 29 weeks and SIP was not reported as a selected © 2014 Nature America, Inc.

outcome. Mortality rates were similar between groups (9.5% vs 8.5%, P = 0.41).12 A Cochrane meta-analysis found similar efficacy in decreasing motor disorders without an increase in overall mortality.15 However, the interpretation of safety data is limited by the studies pooled in the meta-analysis, which did not include spontaneous perforations as an included outcome and were comprised of infants who were at a larger birth weight and GA than those at highest risk in our study. In addition, the metaanalysis may hide important center and group differences. Magnesium and HC exert biologic effects that may contribute to the pathogenesis of SIP. Magnesium ions cross the placenta rapidly, and fetal as well as newborn levels increase in proportion with maternal levels.18 Magnesium has a role in the regulation of vascular tone by acting as a competitive antagonist with calcium, thereby decreasing the intracellular free calcium necessary for the actin–myosin interaction of smooth muscle contractility.16 Magnesium is known to produce intestinal atony and fecal impaction, as seen in mothers exposed to high levels of magnesium.18 In animals, magnesium administration causes dilation of cerebral and mesenteric resistance arteries and reductions in renal and mesenteric blood flow.17 Similar reductions in cerebral blood flow and perfusion has been demonstrated in preterm infants.19 In contrast to a single retrospective study measuring Doppler velocities in infants exposed to magnesium compared with intestinal blood flow among non-exposed infants, investigators did not find a difference in intestinal blood flow velocity.17 However, the gestation ages of infants in this study was higher (>27 weeks) than those infants in whom we observed the greatest incidence of SIP. Magnesium also depresses cholinergic activity at the myoneural junction, likely producing intestinal hypomotility.20 Hypomotility may lead to increased water absorption, the formation of stool plugs and an overgrowth of bacteria. An intestinal plug could lead to increased proximal intraluminal pressure, which would be particularly detrimental in intestinal smooth muscle tissue concomitantly exposed to and thinned by HC.21,22 In clinical practice for those infants exposed to in utero magnesium sulfate, we have been cautious not to add immediately magnesium to hyperalimentation, but rather to follow serially magnesium levels and supplement once the level has decreased into the normal range. Clinically, we have observed a pattern of either delayed stooling or lack of stooling before perforation. The reason for this is likely multifactorial with a primary effect resulting from magnesium and secondary exacerbation from other commonly used medications, which are known to slow gut motility such as morphine. Steroids have been demonstrated to exert multiple effects on the ileal crypt cells by increasing insulin-like growth factor (IGF) availability and increasing epidermal growth factor receptormediated signaling. IGF-I is abundant in mesenchymal tissues within the ileum. It is diminished with steroid exposure and instead redistributed to the villus lamina propria and intraepithelia basolateral spaces. This process occurs via a steroid-induced upregulation of cathepsin L, a protease that cleaves IGF-binding proteins from IGF-I, increasing available IGF-1, which increases cell migration. HC has been shown to potentiate epidermal growth factor-dependent signaling in intestinal epithelial cells. The IGFs and epidermal growth factor share common downstream intermediaries, resulting in increased cell proliferation and cell migration. The net result is an aberrant trophism with submucosal thinning and mucosal hyperplasia.23 Therefore, a harmful synergism may exist between magnesium-induced impaired intestinal motility and submucosal thinning owing to HC. There are no reports of increased risk of SIP in HC-exposed infants in conjunction with antenatal magnesium; however, the combination of GA and commonly used medications such as morphine and indocin might exacerbate the risk of SIP. Journal of Perinatology (2014), 819 – 822

Associations between SIP/death and Mg exposure BN Rattray et al

822 Multiple countries have released clinical practice guidelines recommending antenatal magnesium sulfate before very preterm birth.15 This has led us to initiate MgPro again, with ongoing observation of total exposure, morbidity and mortality. We acknowledge the weight of evidence from randomized clinical trials in the appropriate target population, and the hazards involved with the subgroup analysis. That said, we believe it prudent to monitor infants who have higher amounts of antenatal magnesium exposure, and consider implementing MgPro at the lowest GAs with the lower doses used in the ACTOMg trial.10 We acknowledge that we are limited by the small size of our cohort, and that the effect we noted may have been from unmeasured local characteristics of management, in particular the use of HC, morphine, early indomethacin or from obstetric practices leading to prolonged exposure to magnesium. In our data set, infants with a GA of o25 weeks were at highest risk. Our data indicate that in our population, magnesium sulfate when given peripartum in the doses reported in the NICHD-MFM Network Trial for fetal neuroprophylaxis in infants o 25 weeks may result in SIP. This was especially true as the amount of antenatal magnesium provided increased, with a marked increase noted once the total antenatal magnesium exposure reached 100 g. In summary, we remain concerned about the association between SIP and death and Mg exposure in extremely preterm infants, particularly those born before 25 weeks gestation. Accumulation and analysis of large numbers of extremely preterm infants and cumulative dose exposure data are needed to verify this safety observation. CONFLICT OF INTEREST The authors declare no conflict of interest.

REFERENCES 1 Mittendorf R. Magnesium sulfate tocolysis: time to quit. Obstet Gynecol 2007; 109 (5): 1204–1205. 2 Nelson KB, Grether JK. Can magnesium sulfate reduce the risk of cerebral palsy in very low birthweight infants? Pediatrics 1995; 95(2): 263–269. 3 Schendel DE, Berg CJ, Yeargin-Allsopp M, Boyle CA, Decoufle P. Prenatal magnesium sulfate exposure and the risk for cerebral palsy or mental retardation among very low-birth-weight children aged 3 to 5 years. JAMA 1996; 276(22): 1805–1810. 4 Boyle CA, Yeargin-Allsopp M, Schendel DE, Holmgreen P, Oakley GP. Tocolytic magnesium sulfate exposure and risk of cerebral palsy among children with birth weights less than 1,750 grams. Am J Epidemiol 2000; 152(2): 120–124. 5 Grether JK, Hoogstrate J, Walsh-Greene E, Nelson KB. Magnesium sulfate for tocolysis and risk of spastic cerebral palsy in premature children born to women without preeclampsia. Am J Obstetr Gynecol 2000; 183(3): 717–725. 6 Mittendorf R, Covert R, Boman J, Khoshnood B, Lee KS, Siegler M et al. Is tocolytic magnesium sulphate associated with increased total paediatric mortality? Lancet 1997; 350(9090): 1517–1518.

Journal of Perinatology (2014), 819 – 822

7 Mittendorf R, Dambrosia J, Pryde PG, Lee KS, Gianopoulos JG, Besinger RE et al. Association between the use of antenatal magnesium sulfate in preterm labor and adverse health outcomes in infants. Am J Obstet Gynecol 2002; 186(6): 1111–1118. 8 O'Shea TM, Klinepeter KL, Dillard RG. Prenatal events and the risk of cerebral palsy in very low birth weight infants. Am J Epidemiol 1998; 147(4): 362–369. 9 Paneth N, Jetton J, Pinto-Martin J, Susser M. Magnesium sulfate in labor and risk of neonatal brain lesions and cerebral palsy in low birth weight infants. Pediatrics 1997; 99(5): p. art. no.–e1. 10 Crowther CA, Jetton J, Pinto-Martin J, Susser M et al. Effect of magnesium sulfate given for neuroprotection before preterm birth: a randomized controlled trial. JAMA 2003; 290(20): 2669–2676. 11 Marret S, Marpeau L, Follet-Bouhamed C, Cambonie G, Austruc D, Delaporte B et al. Effect of magnesium sulphate on mortality and neurologic morbidity of the very-preterm newborn (of less than 33 weeks) with two-year neurological outcome: results of the prospective PREMAG trial. Gynecol Obstet Fertil 2008; 36 (3): 278–288. 12 Rouse DJ, Hirtz DG, Thom E, Varner MW, Spong CY, Mercer BM et al. A randomized, controlled trial of magnesium sulfate for the prevention of cerebral palsy. N Engl J Med 2008; 359(9): 895–905. 13 Walsh MC, Kliegman RM. Necrotizing enterocolitis: treatment based on staging criteria. Pediatr Clin N Am 1986; 33(1): 179–201. 14 Watterberg KL, Gerdes JS, Cole CH, Aucott SW, Thilo EH, Mammel MC et al. Prophylaxis of early adrenal insufficiency to prevent bronchopulmonary dysplasia: a multicenter trial. Pediatrics 2004; 114(6): 1649–1657. 15 Doyle LW, Crowther CA, Middleton P, Marret S, Rouse D. Magnesium sulphate for women at risk of preterm birth for neuroprotection of the fetus. Cochrane Database Syst Rev 2009; (1): CD004661. 16 Satake K, Lee JD, Shimizu H, Uzui H, Mitsuke Y, Yue H et al. Effects of magnesium on prostacyclin synthesis and intracellular free calcium concentration in vascular cells. Magnes Res 2004; 17(1): 20–27. 17 Havranek T, Ashmeade TL, Afanador M, Carver JD. Effects of maternal magnesium sulfate administration on intestinal blood flow velocity in preterm neonates. Neonatology 2011; 100(1): 44–49. 18 Lu JF, Nightingale CH. Magnesium sulfate in eclampsia and pre-eclampsia: pharmacokinetic principles. Clin Pharmacokinet 2000; 38(4): 305–314. 19 Shokry M, Elsedfy GO, Bassiouny MM, Anmin M, Abozid H. Effects of antenatal magnesium sulfate therapy on cerebral and systemic hemodynamics in preterm newborns. Acta Obstet Gynecol Scand 2010; 89(6): 801–806. 20 Santafe MM, Garcia N, Lanuza MA, Tomàs M, Besalduch N, Tomàs J. Presynaptic muscarinic receptors, calcium channels, and protein kinase C modulate the functional disconnection of weak inputs at polyinnervated neonatal neuromuscular synapses. J Neurosci Res 2009; 87(5): 1195–1206. 21 Gordon PV, Price WA, Stiles AD, Rutledge JC. Early postnatal dexamethasone diminishes transforming growth factor alpha localization within the ileal muscularis propria of newborn mice and extremely low-birth-weight infants. Pediatr Dev Pathol 2001; 4(6): 532–537. 22 Gordon PV, Paxton JB, Herman AC, Carlisle EM, Fox NS. Igf-I accelerates ileal epithelial cell migration in culture and newborn mice and may be a mediator of steroid-induced maturation. Pediatr Res 2004; 55(1): 34–41. 23 Gordon PV. Understanding intestinal vulnerability to perforation in the extremely low birth weight infant. Pediatr Res 2009; 65(2): 138–144.

© 2014 Nature America, Inc.

Antenatal magnesium sulfate and spontaneous intestinal perforation in infants less than 25 weeks gestation.

Evaluate spontaneous intestinal perforation (SIP)/death among extremely low birthweight (ELBW) infants before, during and after initiation of an anten...
200KB Sizes 2 Downloads 3 Views