We thank Drs Yamamoto and Shiraki for their letter[1] in which they draw attention to elements regarding the time span between anti-tumour necrosis factor (TNF)-α therapy and time of surgery. As stated in the letter, the first meta-analysis on patients with ulcerative colitis (UC) reported an increased risk of short-term postoperative complications, but some patients received the last infusion more than 12 weeks before surgery.[1, 2]

In two studies we found no increased risk of reoperation following anti-TNF-α therapy within the 12 weeks before surgery.[3, 4] Although a sustained biological effect of infliximab beyond 12 weeks might be unlikely,[1, 5] we considered a possible adverse effect of frequent and prolonged anti-TNF-α treatment before surgery on the risk of reoperation due to prolonged suppression of the immune response.

We therefore performed further analyzes on all Danish patients with UC and Crohn's disease (CD) having a first-time operation from 1 January 2003 to 31 December 2010. Patients were classified according to number of anti-TNF-α treatments within the 12 months before surgery (exposed groups: >3 or ≤3 treatments) or no treatment within 12 months (unexposed). All types of anti-TNF-α agents were included.

A total of 263 UC patients received anti-TNF-α agents within the 12 months before surgery, and likewise 222 CD patients. In UC patients receiving >3 treatments (n = 73), 20.5% had a reoperation compared to 23.9% among the unexposed (= 933). In CD patients receiving >3 treatments (= 108), 7.4% had a reoperation compared to 9.7% among the unexposed (= 1501). Details and adjusted relative risks are given in Table 1.

Table 1. Reoperations in cohorts of patients with ulcerative colitis (UC) and Crohn's disease (CD) with a first-time surgery from 1 January 2003 through 2010
UC patients
Reoperation within 30 days after UC surgeryExposed cohort >3, n = 73Exposed cohort ≤3, = 190Unexposed cohort, n = 933
Yes, number (%)15 (20.5%)35 (18.4%)223 (23.9%)
No, number (%)58 (79.5%)155 (81.6%)710 (76.1%)
Crude OR (95% CI)0.82 (0.46–1.48)0.72 (0.48–1.07)Reference
ORa (95% CI)0.96 (0.51–1.78)0.81 (0.53–1.25) 
CD patients
Reoperation within 30 days after CD surgeryExposed cohort >3, n = 108Exposed cohort ≤ 3, n = 114Unexposed cohort, n = 1501
  1. a

    Adjusted for age (years 15–25, 26–55, >55), gender, co-morbidity (no co-morbidity, co-morbidity level 1, co-morbidity level 2), calendar period (2003–2004, 2005–2006, 2007–2008, 2009–2010), duration of UC/CD (<5 years, ≥5 years), steroid prescriptions (no/yes) within 4 weeks before surgery, and number of inpatient days at hospital within 4 weeks before surgery (<2 weeks, ≥2 weeks) in a logistic regression model.

Yes, number (%)8 (7.4%)9 (7.9%)145 (9.7%)
No, number (%)100 (92.6%)105 (92.1%)1356 (90.3%)
Crude OR (95% CI)0.75 (0.36–1.57)0.80 (0.40–1.62)Reference
ORa (95% CI)0.75 (0.34–1.62)0.87 (0.42–1.78) 

Our further analyses thus indicate that repeated treatments with anti-TNF-α agents given within the 12 months before surgery did not increase the relative risk of reoperation within 30 days; and neither did anti-TNF-α therapy within 12 weeks before surgery.[3, 4] The latest meta-analysis (only UC patients) also showed no increased risk of early postoperative complications.[6]


  1. Top of page
  2. Acknowledgement
  3. References

Declaration of personal and funding interests: None.


  1. Top of page
  2. Acknowledgement
  3. References