69as increased risk if they had at least one (or more) prior resection within 10 years,
or resection for a penetrating complication (abscess, fistula), or perianal fistulizing
disease, or active smoking. Patients were randomized by number of risk factors (1
or ≥1). Patients were allowed to continue oral mesalamine or immunosuppressives
at stable doses. Antibiotics and steroids were prohibited. Primary endpoint in this
study was a composite endpoint of both clinical recurrence defined by ≥70-point
CDAI increase and total CDAI ≥ 200 and evidence of endoscopic recurrence
(Rutgeerts ≥i2) or new penetrating complication at week 76 postoperative. If clini-
cal recurrence occurred, patients could have infliximab increased to 10 mg/kg every
8 weeks. A total of 297 patients were randomized. The study was terminated at
week 104 because the primary endpoint was not met. Prophylactic IFX was associ-
ated with a numerical, but not statistically significant, reduction in clinical recur-
rence rates (12.9% IFX vs. 20.0% placebo, p = 0.097). Similarly, composite clinical
recurrence and endoscopic recurrence rates were lower in the IFX compared to
placebo groups (4.1% vs. 9.3%) but failed to reach statistical significance (p = 0.056).
In a secondary endpoint analysis, rates of endoscopic recurrence alone (22.4% IFX
vs. 51.3% placebo) or endoscopic recurrence or new penetrating complication were
significantly reduced in the IFX group (30.6% IFX vs. 60.0% placebo, p < 0.001).
Several reasons were postulated by the authors to explain the primary endpoint
failure in this study. First, the placebo clinical recurrence rate in this study was
smaller than previously reported (20.0% vs. 38.5%, respectively). The majority of
the study population (69.6%) only had one risk factor, and 57.4% were undergoing
their first CD intestinal resection perhaps diluting the effect of a “high-risk” popula-
tion. Furthermore, the additive effect of risk factors hypothesized in the study has
not been formally replicated. These may have led to an overestimation of IFX effect.
Additionally, there was a low median CDAI score in the study population (105.5),
which required many patients to double their CDAI to meet the clinical recurrence
cutoff of CDAI ≥200. This likely limited the rates of composite recurrence. Lastly,
there was a lower rate of immunosuppressive use in the PREVENT trial compared
to the prior 2009 Regueiro et al. study (17.5% vs. 45.8%, respectively).
Immunomodulators increase IFX levels, reduce immunogenicity, and increase effi-
cacy of IFX. Lastly, the composite endpoint utilized in this study had not been pre-
viously investigated or validated. Thus, within the limitations of the study design,
prophylactic IFX did not significantly reduce clinical recurrence but did reduce
endoscopic recurrence.
Localized injection of infliximab has also been investigated in a pilot open-label
study of eight CD patients with localized (<5 cm length) endoscopic recurrence
without clinical recurrence (CDAI < 150) [ 51 ]. This study found no significant
reduction in median endoscopic or histologic score after 14–21 months of
follow-up.
Comparing IFX to thiopurines, in an open-label pilot study of 22 high-risk post-
operative CD patients to compare AZA (2.5 mg/kg/day vs. IFX (standard induction
followed by 5 mg/kg maintenance)) for the prevention of POR, Armuzzi et al. found
a numerical but nonsignificant reduction in endoscopic recurrence rates with IFX
(40% AZA vs. 9% IFX, p = 0.14) [ 52 ]. There was significantly less histologic
5 Use of Biologics in the Postoperative Management of Crohn’s Disease