Probiotics Prevent GI Infection in Premature Infants

PERTH, Australia, May 11 -- Preterm babies given probiotics to pump up beneficial bacteria may be less likely to suffer necrotizing enterocolitis, according to a meta-analysis.

But, probiotics cannot yet be recommended for routine use in underweight infants born before 33 weeks gestation, said Sanjay Patole, M.D., of the King Edward Memorial Hospital for Women here, and colleagues.

Questions of probiotic type, dose, duration, and safety need to be answered first, they cautioned in the May 12 issue of The Lancet.

"If a large, well-designed trial confirms our results, it could make a very strong case for the routine use of probiotics in preterm neonates," they wrote.

"However, since prematurity is the single most important and unpreventable risk factor," they continued, "probiotic supplementation alone might not turn out to be the panacea for prevention of this illness (necrotizing enterocolitis) in preterm neonates."

Necrotizing enterocolitis affects 5% to 10% of very-low-birthweight neonates, those who weigh less than 1,500 g (about 3.3 lbs). The infection, which kills the intestinal lining, is associated with 20% to 40% mortality and often causes long-term neurodevelopmental damage.

Colonization of the gut with pathogenic microbes is thought to be a contributing factor.

Preterm infants are typically slower to develop a stable and functionally active intestinal flora compared with infants carried to term, particularly if they are delivered by Caesarean section.

"Neonates with very low birthweight usually acquire microbial flora mainly from the intensive care environment rather than from their mother," the researchers wrote.

So, a rapidly growing body of evidence for probiotics has developed, noted Carlo Caffarelli, M.D., and Sergio Bernasconi, M.D., both of the Universit di Parma, Italy, in an accompanying editorial.

The researchers found seven such controlled trials including a total of 1,393 preterm neonates of very low birthweight. They were randomized to control or at least one week of enteral probiotic administration beginning within the first 10 days of life.

Nearly all of the studies used once-daily dosing, but the agents used varied widely.

Five trials used a single type of probiotic (Bifidobacterium breve, Lactobacillus GG, Saccharomyces boulardii, L. casei, and B. lactis, respectively). Two used combinations (B. infantis, Streptococcus thermophilus, and B. bifidus in one trial and L. acidophilus and B. infantis in the other).

For the primary efficacy endpoint, fewer infants in the probiotic groups compared with the control groups developed stage 2 or greater necrotizing enterocolitis (2% versus 6%, relative risk 0.36, 95% confidence interval 0.20 to 0.65).

The number needed to treat to prevent one necrotizing enterocolitis case was 25 (95% CI 17 to 50).

The probiotic group was also better able to tolerate feeding. The weighted mean time to first full feed was 2.74 days sooner (95% CI -4.98 to -0.51).

In the five trials with mortality data, all-cause mortality was likewise improved with probiotic administration (RR 0.47, 95% CI 0.30 to 0.73). The number needed to treat to prevent one death of any cause was 20 (95% CI 12 to 50).

However, mortality from necrotizing enterocolitis was not significantly different between groups in the four trials with available data (RR 0.14, 95% CI 0.02 to 1.15).

For the primary safety endpoint, both groups had similar risk of blood-culture-positive sepsis in available data from six trials (RR 0.94, 95% CI 0.74 to 1.20).

While it was reassuring that there were no reported cases of sepsis caused by the probiotic organisms in the studies included in the meta-analysis, long- and short-term safety is a concern, editorialists Drs. Caffarelli and Bernasconi said.

Sepsis from Lactobacillus GG has been reported in premature infants with gastrointestinal diseases as have several cases of fungemia with S. boulardii, they said.

The range of patient demographics, ages at commencement, and type, dose, and duration of probiotics hampered the findings, though, the researchers said.

Some organisms could be expected to be more effective than others because of differing rates of colonization, they noted, but maturity of the recipient's intestinal tract and other factors are also important to success.

"Inappropriate colonization of the gut might not be the only or the most important factor in the pathogenesis of necrotizing enterocolitis," they wrote, "failure to achieve striking benefits with probiotic supplements may, therefore, not be surprising."

The range of feeding and antibiotic strategies in neonatal intensive care units could also be expected to affect the efficacy of probiotic supplementation, Dr. Patole said.

Nevertheless, "The remarkably consistent results, despite the distinct differences in dose, timing, and type of organisms used, suggest that substantial latitude might be available in the choice of an effective probiotic regimen in the design of further trials," he concluded.