 |
|
|
Figueras-Aloy J, Rodríguez-Miguélez JM, Iriondo-Sanz
M, Salvia-Roiges MD, Botet-Mussons F, Carbonell-Estrany
X. Intravenous immunoglobulin and necrotizing
enterocolitis in newborns with hemolytic disease.
Pediatrics. 2010; 125(1):139-144.
|
|
|
Navarro M, Negre S, Matoses ML, Golombek SG, Vento
M. Necrotizing enterocolitis following
the use of intravenous immunoglobulin for haemolytic
disease of the newborn. Acta Paediatr.
2009; 98(7):1214-1217. |
|
|
Shah PS, Kaufman DA. Antistaphylococcal
immunoglobulins to prevent staphylococcal infection
in very low birth weight infants.
Cochrane Database Syst Rev. 2009; (2):CD006449.
|
|
|
Lambert DK, Christensen RD, Henry E, et al. Necrotizing
enterocolitis in term neonates: data from a multihospital
health-care system. J Perinatol.
2007; 27(7):437-443. |
|
|
Are
intravenous immunoglobulin transfusions a risk factor for
NEC?
A recently published article by Figueras-Aloy and coworkers
suggests an association between IVIG transfusion and NEC
in near-term, predominantly female infants being treated
primarily for hyperbilirubinemia (7 with ABO incompatibility,
4 with Rh disease). Of the total cohort of 492 patients,
half were male, 11 developed NEC, with 9 of these infants
being female). In addition to gender, other significant
differences indicated a highly skewed cohort, including
younger gestational age, smaller birth weight, need for
resuscitation, and poorer 1-minute Apgar score. There were
also strong trends approaching significance for small-for-gestational-age
(SGA) status and 5-minute Apgar score, which proved to be
significant, along with IVIG, in a multivariate analysis.
The only risk factor for NEC that these girls did not exhibit
was having received an excess of formula. It appears that
in a number of cases, only a couple of hours of feeds were
given prior to the development of symptoms (and at least
1 patient did not receive any feeding), which would indicate
that the infants did not have NEC but more likely had ileus
from thrombosis. Since the authors did not define what they
considered to be NEC (specifically, they did not invoke
the editorial “get-out-of-jail-free pass” known as Bell’s
staging criteria, nor did they mention whether pneumatosis
was a required finding), it is impossible to determine if
the patients actually had NEC (generally defined as the
presence of pneumatosis following successful, sustained
feeding).
Fortunately, there is an excellent case series from Spain
by Navarro and colleagues, in which the authors not only
describe the rapid onset of pneumatosis in 3 infants within
hours of receiving IVIG, but also describe “disseminated
thrombi obstructing multiple minor vessels of the mesenteric
circulation” from a pathologic specimen. In short, they
are not describing NEC; rather they are describing an acute
mesenteric thromboembolic event occurring shortly after
high-dose IVIG administration. This latter explanation appears
more palatable, as it accounts for why these patients do
not die very often (in the 2 studies, although 14 infants
developed acute symptoms, only 1 died in the acute period).
Presumably, this is because they do not yet have enough
substrate in their intestines to drive bacterial overgrowth.
Thus, the initial plume of pneumatosis is self-limited.
As an aside, of the 14 pooled infants with IVIG-associated
disease, 11 were female, suggesting once again that there
might be a true sex predilection. We need larger numbers
of patients to test this hypothesis.
What about IVIG in preterm infants?
There are, in fact, fairly good safety data on IVIG preparations in preterm infants. The largest dataset is to be found with
antistaphylococcal-enriched IVIG preparations, which have now been investigated in 2 large, multicenter, randomized trials
(reviewed in a Cochrane analysis by Shah and Kaufman). No association with NEC was demonstrated in either trial. Associations
were examined separately by each of Bell’s 3 staging criteria. With the polyclonal immune globulin Veronate® (Inhibitex,
Inc.; Alpharetta, GA), the authors also delineated NEC vs. SIP and found no association with either entity. Based on these
findings, there is no evidence to support an association between IVIG and NEC in cohorts of patients who are predominantly
preterm.
What is the mechanism by which IVIG causes pneumatosis in term infants?
Term vs. preterm is an important distinction. Clearly, IVIG-associated NEC differs from PRBC-associated NEC in that it affects
near-term infants (see the figure in Review 1). Several years ago, Lambert and associates from the Intermountain Healthcare
group conducted the definitive study on risk factors for term NEC. The authors demonstrated that SGA, slightly less mature,
term infants with perinatal hypoxic events and rapid formula advances were at highest risk for term NEC. Although the
Figueras-Aloy study contains many patients who were designated NEC and had many of those same risk factors, the study does
not appear compelling for IVIG and NEC. However, in combination with the Navarro study, it appears likely that the viscosity
effects of IVIG, which are thought to be quite profound when IVIG is administered too quickly, could induce sudden thrombosis
and yield NEC-like symptoms. Therefore, it can be argued that this disease entity is not NEC, as we have conceptualized it
for the last 50 years, but is rather a form of diffuse mesenteric thromboembolism. In term infants who are barely fed,
morbidity and mortality should be better with this disease entity than in term infants with NEC (which is actually not as bad
as we used to think). There may be confounders, such as high-intensity phototherapy, biliblankets, cephalosporins, and
central lines, which all need to be investigated as possible adjunct thrombosis risks. One of the things that neither the
Figueras-Aloy study nor the Navarro study examined was the route of IVIG administration (one might imagine that
administration of IVIG via an umbilical artery catheter would be an unfortunate correlate). Future research will require
capturing more patients with better data granularity. This would perhaps be an ideal disease entity for an adverse outcomes
registry Web site, so that we could gather cases worldwide.
|
