Inflammatory bowel disease (IBD) patients exposed to tumor necrosis factor inhibitor (TNFi) prior to surgery do not face increased risk of postoperative infection, according to the findings of the largest and most comprehensive research effort on the subject to date.
Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services Policy
The results of the study led by New York’s Icahn School of Medicine at Mount Sinai and involving Cleveland Clinic and 15 other institutions confirm that recent preoperative TNFi use should not impact surgical decisions for most IBD patients. Instead, the researchers advise that clinicians and patients focus on addressing other modifiable risk factors for postoperative infection identified by the study, including smoking and preoperative systemic corticosteroid use.
IBD is characterized by overactive proinflammatory molecular pathways, dysregulated inflammatory cytokines and resultant autoimmune cascades that afflict the intestinal mucosa and wall. The proinflammatory cytokine TNF-alpha is a key mediator of this abnormal immune response.
Introduced beginning in the late 1990s, TNFis such as infliximab and adalimumab have become a treatment mainstay for moderate to severe Crohn’s disease and ulcerative colitis. Although these biologic therapies are highly effective, a percentage of IBD patients eventually require surgical intervention. Because TNFis are immunosuppressants, there are concerns regarding infection risks, especially for patients who undergo surgery. Previous retrospective studies assessing whether preoperative TNFI administration increases postoperative infection risks produced contradictory outcomes.
A comprehensive approach
The new study, known as PUCCINI (the Prospective Cohort of Ulcerative Colitis and Crohn’s Disease Patients Undergoing Surgery to Identify Risk Factors for Post-Operative Infection I), sought to determine if preoperative exposure to TNFi was an independent risk factor for postoperative infections within 30 days of IBD surgery.
“For years, there has been a lot of confusion in the literature regarding the safety of TNFis, especially in the perioperative period,” says the PUCCINI study’s lead author, Benjamin Cohen, MD, Co-Section Head and Clinical Director for Inflammatory Bowel Diseases in Cleveland Clinic’s Digestive Disease & Surgery Institute. Dr. Cohen notes that many of the previous studies on this topic were retrospective, single-center analyses.
“At the time we conceived of the study, there were no prospective cohorts built to answer this question,” Dr. Cohen says. “And it’s really important that we study this in a prospective manner because there are many potential confounding factors that can contribute to postoperative infections. Not all of them are going to be available accurately in the medical record when conducting a retrospective analysis.”
Unlike in previous studies, the PUCCINI researchers were able to control for these confounding factors, including the influence of steroids, another IBD treatment that poses significant infection risk. The study also is unique because it assessed risk of infection using TNFi exposure defined by serum drug concentrations, Dr. Cohen says.
Methodology and results
PUCCINI enrolled 947 IBD patients undergoing intra-abdominal surgery across 17 sites from 2014 to 2017. Among those patients, 382 reported current TNFi exposure, which was defined as use within 12 weeks prior to surgery. The researchers categorized infectious complications as surgical site infections (SSI) or non-SSI occurring within 30 days following surgery. Serum was collected to analyze TNFi levels by measuring drug concentrations and anti-drug antibody concentrations for infliximab and adalimumab.
PUCCINI’s primary outcome was occurrence of any infection (SSI or non-SSI) within 30 days of surgery. Secondary outcomes included SSI, hospital readmission within 30 days of surgery, re-operation within 30 days of surgery, 30-day post-operative mortality, duration of postoperative hospitalization, thrombotic complication within 30 days of surgery, and hypomotility complication (ileus for more than 5 days or small bowel obstruction).
The researchers conducted a primary statistical analysis to determine the independent contribution of preoperative TNFi exposure to postoperative infection, and a secondary analysis to assess the independent contribution of TNFi exposure to development of SSI.
The data showed that rates of any type of infection and SSIs were similar in patients with and without preoperative exposure to TNFis (18.1% vs. 20.2% for any infection, p=0.469; 12.0% vs. 12.6% for SSIs, p=0.889). The postoperative infection rates in TNFi-exposed and unexposed patients did not differ regardless of how the patients’ perioperative serum drug concentration was measured (using either a commercial assay’s lower limit of quantitation threshold for detectability or any drug concentration > 0 μg/mL as the threshold for detectability).
In multivariable analysis, current preoperative TNFi use was not correlated with any postoperative infection (odds ratio [OR] 1.050; 95% confidence interval [CI] 0.716-1.535) or SSI (OR 1.249; 95% CI 0.793-1.960).
The researchers also reported that the secondary analysis found no association between detectable serum TNFi drug concentrations and any infection or SSI postoperatively.
Preoperative corticosteroid use, current smoking, prior bowel resection, hypertension and diabetes were identified as risk factors independently associated with SSI. Patients with current preoperative TNFi use were more likely than nonTNFI-exposed patients to develop postoperative venous thromboembolism (3.7% vs. 1.4%, p=0.024). The association remained significant even when the researchers controlled for preoperative corticosteroid use. However, the overall number of thrombotic events was too low to perform a multivariable analysis including other potential confounding factors.
Implications for clinical practice
The PUCCINI study’s findings should reassure clinicians that TNFi use is safe and should not affect surgical decisions in the majority of IBD patients, according to Dr. Cohen and his fellow investigators. Rather, the emphasis should be on other infection risk factors.
“I believe this research is going to have a significant impact on how we practice,” Dr. Cohen says. “Among experts, there has been increasing recognition that it was not TNFi use that is associated with postoperative infections and complications, but many of the unmeasured confounders. However, we need to get this study out there to further educate the gastroenterology and surgical communities at large.
“It is important that they understand that surgical decision-making should be based on other modifiable — and even non-modifiable — risk factors that are clearly associated with infection. rather than on whether a patient has used a TNFi,” he says. “My hope is that we’re past the days where, just because a patient may have been on a biologic drug, that’s going to mean that their surgery is delayed or they’re automatically going to undergo an ileostomy.”
While these findings answer questions regarding the safety of TNFis in this patient population, there is still much to learn. “We’re in an era of drug discovery in IBD,” Dr. Cohen says. “There are many new medications with different mechanisms of action coming to market and there will always be questions about the safety of any new drug in the perioperative period.”
Therefore, the ongoing development of prospective research cohorts is critical, although Dr. Cohen acknowledges this can be challenging and will require significant effort and support from all stakeholders.
“As our approach for IBD patients continues to evolve, so will our need for further study of TNFis,” he says. “We may have to explore combinations of different biologic or immunosuppressive drugs, because the field is moving towards combining agents that may change risk profiles. The questions around TNFis alone have been answered, but new questions will come up that we will need to address.”