Top down therapy is particularly important for patients who are likely to develop a complicated, disabling disease course. While varying definitions exist for “complicated” and “disabling,” studies more often use evidence of bowel damage including strictures, fistulas and abscesses, need for hospitalization, immunomodulators , or surgery within the first 5 years of diagnosis [40, 41]. Several studies have identified risk factors that predict progression to disabling disease (Fig. 48.3). Beaugerie and colleagues examined the natural history of over 1100 patients with CD over at least 5 years and identified young age at diagnosis (≤40 years) (OR 2.1), presence of perianal disease at diagnosis (OR 1.6), stricturing disease, and an initial need for corticosteroids (OR 3.1) as predictive factors for progression to disabling disease. When two of these factors were present at diagnosis, 84 % of patients developed disabling disease within 5 years; with three factors, percentage of patients with disabling disease increased to 91 % [40]. Other studies have corroborated perianal disease, stricturing behavior, and need for steroids [41, 42] and have also suggested ileal or ileocolonic location as independent risk factors for surgery (P < .001) [31, 42, 43]. Active smoking has been associated with increased risk surgeries and disease complications [44–46].

While predominantly retrospective, data on serologic biomarkers has expanded, particularly regarding biomarkers as potential predictors of CD natural history. Positive anti-Saccharomyces cerevisiae antibody (ASCA ) status has been associated with early surgery [47], ileal disease [48], and more severe clinical outcome [49] and might therefore be a useful marker in the selection of patients. Both anti-I2 (P < 0.003) and anti-outer membrane protein C (anti-OmpC) (P < 0.0006) were associated with internal penetrating and/or stricturing disease (IP/S) [50]. Additionally, the presence of all four immune responses to ASCA , anti-OmpC, anti-I2, and anti-CBir1 flagellin (anti-CBir1) yielded the highest odds of developing penetrating or stricturing complications and need for intestinal surgery in pediatric patients (OR (95 % CI): 11 (1.5–80.4); P = 0.03) [50]. In another study, a well-defined correlation was found between triple positive status for ASCA , anti-OmpC as well as anti-I2 and small bowel surgery [51]. Newer markers have been evaluated including antibodies against carbohydrate epitopes [52]. In Reider et al.’s study, positivities for ASCA , anti-mannobioside IgG (AMCA), anti-chitobioside IgG (ACCA), and anti-laminarin IgA (anti-L) were each independently and cumulatively associated with shorter time to disease complications and need for surgery [53].

Data, while lacking, also suggests that genetic predictors exist. Most data have identified mutations to NOD2 locus . The presence of at least 1 NOD2 variant conferred a pooled relative risk for the presence of strictures or fistulas of 1.17 (95 % CI: 1.10–1.24) [54] and was associated with a higher risk of surgery [54–58]. Further studies evaluating the influence of genetics on disease phenotype are needed.

An important issue related to early intensive treatment is the safety of long-term use of biologics and immunomodulators . Patients are exposed to potentially toxic agents, and therefore appropriate safety measures should be undertaken before immunomodulators and biologics are initiated (Fig. 48.4). Safety measures are aimed at prevention of infections and awareness of several rare complications.

Patients treated with immunomodulators have an increased risk for influenza infections, pneumococcal infections, and Salmonella supp. infections. Therefore, influenza vaccination, pneumococcal vaccination, and appropriate food hygiene (avoiding raw eggs, unpasteurized milk, raw meat) is recommended [59].

Screening for cytomegalovirus (CMV) and herpes simplex virus (HSV) before starting immunomodulator therapy is not recommended. Latent subclinical CMV infection is not a contraindication for starting immunomodulators . However, CMV colitis should be excluded in refractory IBD cases and in case of systemic CMV, immunomodulator therapy should be discontinued. In contrast, given its relationship to lymphoproliferative disease during immunomodulator therapy, Epstein–Barr virus (EBV) screening and antiviral therapy should be considered. In cases of severe EBV infection, immunomodulator therapy should be discontinued. Regular cervical cancer screening and human papilloma virus vaccination are highly recommended for women with IBD, especially if treated with immunomodulators [59].

Also, reactivation of latent hepatitis B (HBV) is considered a serious risk, and therefore all patients with inflammatory bowel disease (IBD) should be tested to exclude HBV. Patients with active chronic HBV infection should be treated according to standard antiviral therapy. Nucleoside/nucleotide analogs are preferred since interferon therapy might exacerbate the colitis. Seronegative patients should receive HBV vaccination. Patients might need a higher dose of immunizing antigen since vaccination efficacy is affected by number of immunomodulators [59].

Anti-TNF use is associated with an approximately 21-fold increased risk of tuberculosis (TB) without appropriate safety measures [60] (Fig. 48.5). TB incidence has decreased with suitable safety measures. Approximately 78% of TB cases present during the first 3 months of treatment and have an atypical presentation, which makes the diagnosis more complicated [61]. For that reason, guidelines advise to assess the risk of TB at the time of diagnosis and before starting treatment with an anti-TNF agent, including patient history, chest X-ray, tuberculin skin test, and interferon-gamma release assays (IGRA) . Latent TB may be suspected in case of a positive initial tuberculin skin test and when the patient has recently been exposed to the disease. Physicians should be aware of the possibility of false-negative skin tests, especially when patients are immunocompromised. When the patient is diagnosed with latent TB, treatment with the full therapeutic antituberculous regimen should be initiated and it is advised to delay anti-TNF treatment for at least 3 weeks after initiation of anti-TB regimen. When active TB is diagnosed, anti-TNF treatment is ideally delayed until anti-TB treatment has been completed. However, solid data on the ideal timing during anti-TB treatment are lacking and infectious disease consultation is recommended. When TB is diagnosed during anti-TNF treatment, the anti-TNF agent should be discontinued and TB therapy should be started. Anti-TNF therapy can be resumed if needed after 2 months. All patients should be monitored carefully for signs of cough, fever and weight loss and treating physicians should be aware of uncommon extrapulmonary TB as well as the more common lung disease [59].

After all, when taking appropriate safety measures, anti-TNF and immunomodulator therapy appears to be relatively safe. Safety data from referral centers and randomized controlled trials do not show an increased risk of malignancies or infections in anti-TNF treated patients. In a large meta-analysis of 21 placebo-controlled trials including 5356 patients, no increased risk of death, serious infection or malignancy compared to controls was reported [62]. In line with this observation, no increased risk was found in infections, mortality or malignancy in 734 anti-TNF treated patients compared to controls, with a median follow-up of 58 months [63]. In addition, no increased risk of malignancy was observed in patients treated with anti-TNF in a large cohort of CD patients [64]. However, long-term safety data are not available yet and therefore awareness of (serious) side effects is warranted.

Monitoring of disease activity is a key component of IBD care. Traditionally, patients are monitored clinically using validated indices including Crohn’s Disease Activity Index (CDAI) or the Harvey–Bradshaw Index (HBI). Clinicians are also able to monitor using inflammatory serologic markers such as C-reactive protein, or fecal markers including fecal calprotectin or lactoferrin. Increased levels of calprotectin and lactoferrin have been shown to correlate with the validated Crohn’s Disease Endoscopic Index of Severity scores [65, 66], and therefore, these markers might be useful tools to assess disease activity.

Newer evidence in the literature promotes mucosal healing as a method to assess efficacy of ongoing therapy. Mucosal healing predicts long-term outcome and is associated with reduced subsequent disease activity and risk of relapse, increased steroid-free remission rates and less surgery and hospitalization [5–7, 10, 11, 67]. While Bouguen and colleagues noted that fewer than 26 weeks between endoscopic procedures (hazard ratio, 2.35; 95 % confidence interval, 1.15–4.97; P = .035) and adjustment to medical therapy when mucosal healing was not observed (hazard ratio, 4.28; 95 % confidence interval, 1.9–11.5; P = .0003) are factors associated with mucosal healing [68], further data is needed to determine optimal time intervals between endoscopic assessments for mucosal healing.

To evaluate the risk benefit of early intensive treatment, careful monitoring of potential and sometimes avoidable adverse events is mandatory (Fig. 48.6). Patients on immunosuppressive therapy and with malnutrition are at risk for opportunistic infections [59, 69]. For that reason, patients with fever, cough and systemic illness should be carefully examined and Mycobacterium tuberculosis (TB) should be excluded. Patients diagnosed with pneumonia should be treated with an antibiotic covering S. pneumoniae [59]. In case of an active infection, it is advised to withdraw immunomodulators until the resolution of infection. Furthermore, about 70 % of the population has latent JC virus, which can cause progressive multifocal leukoencephalopathy upon reactivation. Since reactivation of the JC virus is associated with systemic immunosuppression, treating physicians ought to be aware of this rare situation. PML has been associated with natalizumab with a risk estimate of approximately 1:1000 (0.2–2.8 per thousand) with a mean exposure of 18 months of therapy [70–72]. Patient presenting with new-onset neurological symptoms should receive an MRI scan and lumbar puncture.

Studies show divergent data on the occurrence of malignant lymphoma in IBD patients receiving immunomodulators . Whereas some studies do not show an increased risk, other studies do find a moderately elevated risk, especially in patients on thiopurine therapy [73–77]. Most recently, a meta-analysis by Williams et al. examined 22 randomized controlled trials with over 7000 patients comparing anti-TNF treatment with placebo and found no increased risk of malignancy; however, no trials provided data beyond 1 year of treatment. Hence, long-term risk needs to be further assessed [77]. Nonetheless, the absolute risk appears to be low and should be weighed against the benefits of immunomodulator therapy.

Once remission is achieved, it is important to know if, when and how to de-escalate therapy. Although severe adverse events are rare, especially when appropriate safety measures are undertaken and patients are correctly monitored, they can occur and therefore establishing an individualized risk-benefit ratio is encouraged. The disadvantages of discontinuation of therapy should be taken into account, including relapse, possibly lower response to re-induction therapy, infusion reactions and surgery. Particularly important when considering de-escalation is the earlier pattern of the disease and response to therapies. In addition, several factors may predict relapse, including smoking, previous steroid use and elevated fecal calprotectin and CRP.

Early aggressive therapy in CD is beneficial in selected patients who are likely to develop a severe and disabling disease course. Predictors include young age at diagnosis, the presence of perianal disease, stricturing disease, and the initial need for corticosteroids. This approach appears to be relatively safe, but the risks of opportunistic infections and neoplasm should always be weighed against the benefits of therapy. Patients should be monitored carefully on a regular basis and de-escalation of therapy should be considered for those at high risk of adverse events and those in deep remission. The value of serological, immunologic, and genetic markers in monitoring and predicting disease is currently under investigation, and could be helpful to further optimize therapy.