Gastroenterology Hepatology

Crohn's disease

How can I be sure that the patient has Crohn's disease?

Crohn’s disease (CD) is chronic, affecting the intestinal mucosal in a focal, asymmetric, transmural, granulomatous inflammatory disease that can involve any portion of the digestive tract from the mouth to the anus. The transmural nature of CD leads to potential complications of intestinal strictures, fistulas, and/or abscesses.

Crohn’s disease is distinguished from ulcerative colitis (UC), another variant of inflammatory bowel disease (IBD) manifest by diffuse, continuous and superficial ulceration beginning at the ano-rectal junction and extending proximal to a variable extent of the colon. Both conditions have genetic, immunological, and environmental underpinnings and both can be associated with extraintestinal manifestations (inflammatory arthropathies, cutaneous manifestations such as erythema nodosum and pyoderma gangrenosum, uveitis, episcleritis, nephrolithiasis, and sclerosing cholangitis).

While IBD was initially described in individuals of northern European descent, these are now recognized worldwide as societies develop “first world” hygiene and dietary habits. Although the epidemiology of CD and UC overlap to a considerable extent, patients with Crohn’s disease are more likely to have a family history of Crohn’s disease, are more likely to be smokers or have been exposed to passive smoking, and are less likely than UC patients to have had an appendectomy at an early age. The peak age for CD occurrence is 20 to 30 years old, with a possible second peak in those aged 60 to70. The gender distribution of CD is roughly equal between males and females. While cigarette smoking is “protective against the development of UC," smoking is associated with a worse course of CD; however, this factor cannot account solely for a large proportion of the variable CD phenotypes or for children who develop CD. It is also possible for patients to evolve from UC to CD, in particular, after colectomy and ileoanal anastomoses performed to “cure” UC.

A tabular or chart listing of features and signs and symptoms

Table I summarizes the signs and symptoms of CD that are dependent on the disease’s anatomic location, disease severity, and intestinal complications. The majority of patients have distal ileal and/or colonic disease. Upper gastrointestinal lesions manifest in only 10% to 15% of patients; 20% to 30% of patients have perianal lesions, while 15% to 20% have or have had a fistula.

Table I.

Signs and symptoms of Crohn's disease

Patients with small bowel and proximal colonic CD more typically present with abdominal pain, diarrhea, steatorrhea, weight loss, fatigue, and weakness. Stricturing disease leads obstructive symptoms (nausea vomiting, abdominal pain). Patients with colonic involvement typically present with diarrhea, rectal urgency, hematochezia, and tenesmus. Patients with fistulizing disease may present with perianal skin tags, perianal abscesses, or, in women, vaginal drainage of blood, stool, or purulence. Systemic symptoms such as fatigue, fevers, and weight loss, or, in children, delays in growth or sexual development are common. Signs include pallor, abdominal tenderness, palpable abdominal mass, and perianal drainage or tenderness.

Patients with CD can present with extraintestinal manifestations (EIMs). These symptoms include jaundice and pruritus (related to primary sclerosing cholangitis); photophobia, blurred vision, and injected sclerae (related to uveitis, sclero-conjunctivitis, episcleritis); joint pain and swelling, lower back stiffness (related to peripheral large-joint inflammatory arthritis and sacroiliitis, respectively); skin ulceration or erythema and swelling (related to erythema nodosum [EN] and pyoderma gangrenosum [PG]); or abdominal pain from chole- or nephrolithiasis.

An important diagnostic distinction between IBD and irritable bowel syndrome (IBS) is the finding of inflammation. The diagnosis of CD requires evidence of inflammation at endoscopic or imaging (X-ray, CT, MRI).

Table II reviews the differential diagnosis of CD. Due to the chronic nature of CD, the initial diagnosis must be distinguished from other acute forms of IBD – primarily infections, whereas the differential diagnosis for patients presenting with more long-standing symptoms is based on other forms of chronic digestive symptoms.

Table II.

Alternative diagnoses and key features allowing differentiation

Acutely, the wide variety of diseases that mimic ileal CD include appendicitis; infectious ileitis (Yersinia spp, Salmonella, Clostridium difficile, Mycobacterium tuberculosis, Mycobacterium avium-intracellulare complex, cytomegalovirus, and histoplasmosis); vasculitides, such as systemic lupus erythematosus, polyarteritis nodosa, and Behcet’s disease; ischemia; small-bowel neoplasms; drug-related ileitis (primarily nonsteroidal anti-inflammatory drugs, potassium chloride) and less common forms of infiltrative ileitis, such as eosinophilic gastroenteritis, sarcoidosis, amyloidosis; and endometriosis.

Bloody or nonbloody diarrhea, possibly secondary to colonic CD, could also be secondary to pancreatic insufficiency, celiac disease, bacterial overgrowth, irritable bowel syndrome, colonic neoplasia, viral infection, infectious colitides, intestinal tuberculosis, vascular and ischemic colitis, and NSAID induced colitis.

How can I confirm the diagnosis?

CD is diagnosed with a combination of laboratory, endoscopic, and radiologic tests. No single test is wholly sensitive or specific for CD. Figure 1 shows a diagnostic algorithm for CD. Table III lists the variety of tests used in diagnosis and their possible results, with subsequent follow-up tests.

Figure 1.

Simplified diagnostic algorithm.

Table III.

Diagnostic tests and follow-up

The signs and symptoms of CD will depend on the anatomic extent and severity of intestinal involvement. Patients with small bowel CD are more likely to present with abdominal pain and diarrhea secondary to malabsorption. Patients with colonic disease are more disposed to rectal bleeding, abdominal pain, and perianal manifestations.

Fecal markers, similarly, are nonspecific but represent intestinal inflammation via exudation of leukocytes into the lumen. Calprotectin, released by neutrophils, can distinguish between IBD and IBS. A cutoff of 160 μg/g stool has 100% sensitivity and 80% specificity in predicting histologic evidence of active IBD. Lactoferrin, a major iron-biding glycoprotein found in neutrophils, shares the same diagnostic characteristics. Both tests suggest bowel inflammation and possible CD but require further testing for a definitive diagnosis of IBD. Fecal markers are often ordered in conjunction with stool examinations for enteric pathogens, C. difficile, and ova and parasites to rule out other infectious diagnoses or complications.

While there is no gold standard for the diagnosis of CD, a combination of imaging, endoscopy, and histology are used to document the focal, asymmetric, and transmural inflammatory features of Crohn’s disease. Colonoscopy is particularly utilized to differentiate CD (focal, asymmetric, transmural with perianal involvement in 25% of patients) from the diffuse continuous and superficial inflammatory changes of UC. The presence of skip areas of macroscopically and microscopically uninvolved mucosa; ulceration that is deep, stellate, linear, or serpiginous; strictures; and internal fistula all suggest CD over UC. Ileoscopy can also help in this differentiation, as can upper endoscopy for suspected upper GI symptoms. “Focally enhanced gastritis” is often manifest by nonspecific mucosal changes in the stomach, with no evidence of H. pylori on biopsy.

There are no pathognomonic markers for the diagnosis of CD. Aside from the signs and symptoms, laboratory tests can add further suggestion of CD. Serum tests consistent with CD include elevated acute-phase reactants, such as C-reactive protein; low hemoglobin (whether from iron-deficiency, B12 deficiency, or another cause), and general markers of malabsorption (low albumin, elevated INR, low calcium). Most of the laboratory abnormalities have low sensitivity, specificity, or both. The two best known serologic markers are the perinuclear antineutrophil cytoplasmic antibodies (p-ANCA), present in 5% to 25% of patients with CD (particularly in left colonic CD), and the anti-Saccharomyces cerevisiae antibodies (ASCA), prevalent in 48% to 69% of patients with CD. Both tests have good specificity (~80-90%) but poor sensitivity (~50%) for the diagnosis of IBD.

An emerging endoscopic diagnostic tool is video capsule endoscopy (CE) in the detection of suspected CD of the small bowel, with a yield of 43% to 71%. Studies show sensitivity that parallels that of radiologic studies (up to 83%) but a disappointingly low specificity (as low as 53%). Perhaps its biggest advantage is its high negative predictive value, avoiding additional, unnecessary testing. However, the capsule may not reach the colon in up to 25% of patients, with a risk of capsule retention in patients with known CD of 7% to 13%. With ileocolonoscopy serving as the initial test of choice for suspected CD, CE may be the confirmatory subsequent test to assess the extent of disease.

At other times, a contrast-enhanced radiologic study may be the first study to suggest the diagnosis. In addition, contrast radiography serves as confirmatory testing once a diagnosis is suspected. Aphthous ulceration, fistulae, narrowed bowel lumen, cobblestoning, mural thickening, and abscesses all suggest the diagnosis of CD.

For the diagnosis of ileal CD, CT enterography and MR enterography appear to differentiate between inflammatory and noninflammatory disease and have the highest sensitivity and specificity; small-bowel follow-through, meanwhile, has lower sensitivity, with similar specificity. For patients that require serial imaging, MR enterography may be the modality of choice to avoid exposure to ionizing radiation and contrast.

While several chromosomal loci − including NOD2/CARD15, SLC22A4, and IL-23 − have been identified to be associated with specific phenotypes of CD, the beneficial use of this testing in clinical practice has yet to be established for the diagnosis or prognostication of CD.

What other diseases, conditions, or complications should I look for in patients with Crohn's disease?

Table IV lists the complications and comorbid conditions associated with CD. As previously noted, 20% to 30% of patients present with perianal lesions, but the cumulative risk for perianal disease is about 50%. Fifty to sixty percent of patients with perianal disease have at least one perianal abscess, and 15% to 20% have fistulae; 60% have recurrence of perianal abscesses within 2 years. Diagnosis and evaluation of perianal disease oftentimes involves a combination of anorectal examination under anesthesia to identify fistulous tracts, MRI of the pelvis to delineate the path of the fistulae, and endoscopic ultrasound to further map the course of the disease.

Table IV.

Complications and associated conditions

Fifteen percent of all IBD-related deaths are secondary to colorectal carcinoma (CRC). CRC has been shown to rarely occur if duration of colitis is less than 7 years. Chronic histologic inflammation appears to be a risk for colorectal cancer. In addition, older age, primary sclerosing cholangitis (PSC), family history of CRC, and extent of disease appear to be risk factors, with patients with rectal disease not developing CRC.

The risk of CRC in CD is 4 to 20 times that in the general population, with mean duration of CD to time of CRC diagnosis of 15 to 20 years. Given that colonic dysplasia is a precursor and the best predictor of CRC risk in IBD, guidelines on surveillance examinations for CRC in CD are similar to those for UC. Guidelines suggest beginning colonoscopies at 8 years from onset of symptoms, with multiple biopsies through the colon. Although no randomized controlled trial exists, surveillance colonoscopy has been shown to result in moderate CRC risk reduction. Guidelines also recommend colectomy for high-grade dysplasia or DALM (dysplasia-associated lesion or mass in an area of colitis) in UC but recommendations are less established for CD.

In addition to increased risk of CRC, patients with CD are also at higher risk for lymphoproliferative disorders (LD), particularly for Hodgkin’s and non-Hodgkin’s lymphoma; this risk could be related to the chronic course of IBD itself or to the use of immunosuppressants (as noted below). Small bowel adenocarcinoma also occurs with a RR of 28 compared to the general population. Other solid tumors, such as basal cell carcinoma, are also noted to be more common in the IBD population.

As mentioned previously, 25% to 40% of patients with CD can develop extraintestinal manifestations (EIMs), with 25% having more than one EIM.

The most common EIM is musculoskeletal pain. The spine, sacroiliac joint, peripheral joints, or a combination of these can be affected by arthritis. Arthritis is generally more common in patients with colonic than small-bowel disease. Peripheral arthritis involving the large joints occurs in 5% to 20% of IBD patients; knee arthritis, the most common type, is strongly associated with CD activity. Spondylitis occurs in 1% to 26% of patients with IBD, affects more males, and does not depend on CD activity.

Patients with IBD are at risk for osteoporosis. Factors contributing to this risk include active inflammation, corticosteroid therapy, reduced physical activity, bone resorption, calcium malabsorption, vitamin D deficiency, and low albumin levels. Guidelines on monitoring and management are discussed below in the section on corticosteroids.

Two to thirty-four percent of IBD patients present with a dermatologic EIM. Erythema nodosum (EN) is more common in CD and parallels disease activity; as such, treatment focuses on management of intestinal disease. Pyoderma gangrenosum, seen in 0.5% to 20% of CD patients, exhibits pathergy; treatment also involves management of intestinal disease but oftentimes also requires high-dose steroids or anti-TNF biologics. Aphthous stomatitis is treated with topical anesthetics and corticosteroids.

PSC is more common in nonsmokers and, hence, UC but can be seen in 2% of patients with CD. On the other hand, 5% to 10% of patients with PSC have CD; 80% of patients with PSC have a positive pANCA. The test of choice for diagnosis is MRCP. Diagnosis has implications in screening for colorectal cancer (usually begun at the time of diagnosis and continued every 1-2 years) and cholangiocarcinoma.

Cholelithiasis can be seen in 13% to 34% of patients, particularly among those with ileal disease. Nephrolithiasis is ten times more common in CD than UC.

Of all IBD patients, 0% to 5% – particularly those with colitis or ileocolitis – have ocular involvement. Episcleritis is the most common manifestation and parallels disease activity. Scleritis and uveitis, more serious complication that can impair vision, often require high-dose steroids or immunosuppression.

The risks of venous and arterial thromboembolism in IBD are two- to threefold those of the general population. Acquired risk factors play a larger role than congenital ones (e.g., Factor V Leiden mutations) and include inflammation, immobility, hospitalization, steroid therapy, central venous catheter insertion, smoking, oral contraceptive use, and vitamin deficiency. High degree of vigilance for DVT and pulmonary embolism should be the standard for early diagnosis. In addition, the same vigilance should be used for the diagnosis of portal vein thrombosis. Anticoagulation, typically for a minimum of 3 months and up to 6 months, should be considered on a case-by-case basis, dependent on the risk of bleeding.

What is the right therapy for the patient with Crohn's disease ?



The aminosalicylates, sulfasalazine, and mesalamine formulations have minimal effects for the treatment of mild-moderate CD. Given that their action is likely topical rather than systemic, topical formulations such as suppositories and enemas are included in management of colonic disease. pH-dependent and delayed-release formulations have been developed to target specific locations of disease. Common adverse reactions include headache, abdominal pain, and nausea. Rare but notable adverse effects include pancreatitis, myocarditis, oligospermia, and renal failure (oftentimes secondary to interstitial nephritis).

Some patients exhibit a hypersensitivity to mesalamine, manifesting as diarrhea and abdominal pain. In the elderly, mesalamine can rarely cause blood dyscrasias. Finally, the sulfapyridine moiety of sulfasalazine can result in the typical reactions to sulfa products, particularly with regard to rash and hypersensitivity. Because of the class effects, prior to and during therapy, patients’ renal function and CBC (in the case of elderly) should be periodically monitored.


Glucocorticoids are effective for the treatment of moderate-severe CD. However, despite a substantial risk of steroid-dependence, they are not considered maintenance agents due to the litany of systemic side effects.

Prednisone and prednisolone decrease inflammation through multiple mechanisms, including inhibition of leukocyte migration, reduction in activity of the lymphatic system, and inhibition of phospholipase A2.

Budesonide can be utilized for mild-moderate CD involving the ileum and/or right colon and has a high first-pass metabolism through the liver and less systemic bioavailability (10-15%) than prednisone; it thus results in less systemic side effects.

Prolonged use of glucocorticoids results in a number of adverse effects including infection, skin thinning, Cushingoid appearance and weight gain, cataract development, gastritis and ulcer formation, fluid retention and hypertension, psychiatric disturbance, glucose intolerance, and adrenal insufficiency. Also with prolonged use comes musculoskeletal side effects including muscle weakness, avascular necrosis of bones, and osteopenia/osteoporosis. To prevent the latter, avoidance of prolonged use and , if not possible, supplementation with calcium/vitamin D should be considered. A bone mineral density (BMD) at the start of steroid therapy and every 1 to 2 years can be considered in patients with prolonged use and other risk factors. Bisphosphonates can be used in those with confirmed bone loss on BMD.

6-mercaptopurine and azathioprine

The thiopurines mercaptopurine and its prodrug, azathioprine, are purine antagonists but also induce apoptosis of lymphocytes. These agents have steroid-sparing properties and are utilized to maintain steroid-induced remissions or as combination therapies with biologic agents to reduce immunogenicity and may have synergistic effects.

Mercaptopurine and azathioprine are converted to 6-thioguanine, thought to be the active metabolite through a series of constitutive and inducible enzymes. Thiopurine methyltransferase (TMPT) is the major enzyme involved in both agents’ catabolism. While 89% of the population has wild-type TPMT activity, 11% are heterozygous with lower TPMT activity and 0.3% are homozygous with minimal to no activity.

Common short-term adverse effects include myelosuppression (manifesting as leucopenia, thrombocytopenia, and anemia); the onset, nadir, and recovery appear to be within 7 to 10 days, 14 to 16 days, and 21 to 28 days, respectively. High levels of 6-thioguanine nucleotides (6-TG) mediate the purine antagonists’ myelosuppressive effects. Hepatotoxicity, manifesting as intrahepatic cholestasis and centrilobular necrosis, can occur at higher doses and usually presents at 2 months; the range of presentation, however, is from 1 week to 8 years. The metabolite felt to be responsible for the hepatotoxicity is 6-methylmercaptopurine (6-MMP).

Less common side effects include infection, drug fever, rash, nausea/vomiting, pancreatitis, and renal toxicity. Equally serious long-term effects of which patients should be aware include the slight increase in the risk of non-Hodgkin’s and Hodgkin’s lymphoma, and the association with a rare malignancy, hepatosplenic T-cell lymphoma (HSTCL); the latter has been described in young adult males.

Given the above findings, TPMT phenotyping, CBC, liver function tests, and renal function should be checked prior to the initiation of these medications. During treatment, an initial 1 month of weekly CBCs, then 1 month of biweekly CBCs, and then monthly or bimonthly checks throughout the course of therapy should be considered. Liver enzymes can be assessed every 3 months. There may be some role in monitoring metabolites (6-TG and 6-MMP) to maximize the therapeutic dose while minimizing adverse effects.


Methotrexate is an alternative steroid-sparing, maintenance agent for CD. Methotrexate (MTX) is a folic acid antimetabolite that inhibits DNA synthesis, also during the S phase of the cell cycle. In CD, it modulates immune function and inflammatory activity similarly to purine antagonists. It can be administered orally, intramuscularly, and subcutaneously.

Common adverse reactions include headache, myelosuppression (with nadir in 7-10 days), and renal failure. Less common but notable adverse reactions include rash, portal fibrosis, and interstitial pneumonitis. Given its immunomodulator function, lymphomas and infections have also been reported.

Pretreatment evaluation of CBC, liver enzymes, and BUN/creatinine should be a consideration. For the first few months of therapy, CBCs every 2 to 4 weeks, monthly LFTs, and monthly BUN/creatinine should be considered. Subsequent monitoring of CBC, renal function, and liver tests can be on a 2 to 3 month basis.

Controversy exists regarding the utility of a baseline liver biopsy with regard to assessment of baseline fibrosis. Although rheumatologic guidelines suggest a liver biopsy once 1.5 g cumulative dose has been obtained, recent literature suggests that a liver biopsy may not be necessary until 3 to 4 g cumulative dose of MTX is reached. The interval for subsequent liver biopsies is currently considered to be every 1 to 1.5 gm of cumulative dose.


Antibiotics such as metronidazole or ciprofloxacin have been used most effectively to reduce perianal fistula drainage or to treat bacterial overgrowth in the setting of stricturing CD.

Metronidazole interacts with bacterial DNA, resulting in inhibition of protein synthesis. However, it also appears to have some anti-inflammatory properties. Common adverse reactions are disulfiram-like reaction, nausea, anorexia, peripheral neuropathy (particularly with long-term use). Counseling regarding alcohol consumption should be considered prior to initiation.

Ciprofloxacin also inhibits DNA synthesis. It can be associated with headache, nausea, and diarrhea. Rare but notable side effects include QT prolongation, tendon inflammation or rupture, and phototoxicity. A baseline ECG can be considered before starting the medication.

Biological agents

Two classes of biologic agents have been demonstrated to be efficacious for the treatment of CD. Monoclonal antibodies, or antibody fragments, targeting tumor necrosis factor alpha (TNFα) as well as monoclonal antibodies that target α4 integrins, have been demonstrated to have both inductive and maintenance benefits.

Infliximab is a chimeric (75% human, 25% murine) IgG1 monoclonal antibody to tumor necrosis factor alpha (TNFα), a proinflammatory cytokine. It is administered intravenously. Common adverse effects include an increase in liver enzymes, infections, and hypersensitivity reactions within 2 hours of infusion; associated with the latter are antibodies to infliximab and lack of concomitant use of immunosuppressants. Three percent of patients develop antibodies to infliximab. Less common adverse events include flushing, anaphylaxis, central demyelinating disorders, worsening heart failure, hepatitis B reactivation, and latent tuberculosis reactivation.

As for 6MP and azathioprine, reports of increased risk of Hodgkin’s and non-Hodgkin’s lymphoma (including the rare HTSCL more common with the purine analogs) have surfaced.

Prior to starting infliximab, a PPD (or QuantiFERON Gold in the appropriate population) should be placed and read. Tests of hepatitis B status, particularly hepatitis B surface antigen (HBsAg) should be performed. During therapy, monitoring should be considered for infection. If response is lost to infliximab or if patients develop an infusion reaction, antibodies to infliximab (along with infliximab blood levels) can help guide further therapy.

Adalimumab is a fully humanized recombinant IgG1monoclonal antibody that binds also to TNFα; it is administered by via a subcutaneous injection. It suffers from the same adverse effects as infliximab and should be monitored in a similar manner. In addition, injection site reactions and antibodies to adalimumab are also established adverse effects.

Certolizumab is a pegylated humanized antibody Fab’ fragment of TNFα; it is also administered via a subcutaneous injection. Its adverse effects and monitoring parameters are similar to those for infliximab and adalimumab.

Natalizumab is a monoclonal antibody against the alpha-4 subunit of the integrin molecules, involved in adhesion and migration of leukocytes into inflamed tissue. It is administered intravenously. Infections are the most concerning risk although uncommon adverse reactions include headache, fatigue, and nausea. Infusion reactions and infection are real possibilities with natalizumab. Antibody formation, hepatotoxicity are uncommon adverse reactions.

The most significant concern in the use of natalizumab is the occurrence of progressive multifocal leukoencephalopathy (PML) caused by the JC virus, requiring prescribers to be registered with the TOUCH prescribing program. The risk of PML increases with treatment longer than 1 to 2 years and in those with JC viremia. A newly approved serologic assay to look for JC virus exposure can help stratify patient candidacy for natalizumab. Otherwise, as with the other biological agents, periodic monitoring for infection should be considered.

See Table V for treatment options.

Table V.

Options for treatment: magnitude 0-4+ (grade of recommendation)

What is the most effective initial therapy?

Principles of therapy

The transmural nature of CD leads to progression, over time, from manifestations of intestinal ulcerations toward the development of strictures and fistulae. The associated phenotypic evolution also appears to depend on the location of CD within the gut, with colonic disease less likely to develop stricturing than small bowel disease although perianal disease and perianal fistula are more common in the setting of colonic involvement than small bowel disease.

The treatment of CD depends on the location, severity, and complications of the disease. The goals are to induce clinical (and mucosal) remissions and to prevent relapse (maintenance of remission). Factors that exacerbate or mimic CD (smoking and the use of NSAIDs) should be avoided.

Therapies used to induce clinical remissions include aminosalicylates, glucocorticoids, and /or biologic agents such as infliximab, adalimumab, certolizumab, and natalizumab. The second goal is maintenance of remission, and the approaches will depend on the therapy used to induce remission. In CD, the options in this category include sulfasalazine and 5-aminosalicylic acid derivatives; immunomodulators such as 6-mercaptopurine, azathioprine, or methotrexate; the anti-TNF biologic agents, including infliximab, adalimumab, and certolizumab; and antileukocyte adhesion molecules such as natalizumab.

Antibiotics and probiotics have important uses in specific subtypes of patients, as discussed in the following section. Nutritional therapy appears to have the most utility in pediatric CD; elemental diets can induce remission, mucosal healing, and recovery of bone metabolism.

Surgery has some role when medical management fails or when specific disease complications occur. Table V and Table VI list the best initial and subsequent therapy. Table VII lists the main side effects of therapy and monitoring. Finally, Figure 2 presents a therapeutic algorithm.

Table VI.

First- and second-line therapy and failure

Table VII.

Adverse events and monitoring

Listing of usual initial therapeutic options, including guidelines for use, along with expected result of therapy.



Early studies showed that controlled-release mesalamine resulted in induction of clinical remission in 42% by 14 months. However, in double-blind, multicenter trials comparing budesonide to low-dose mesalamine, 8-week and 16-week remission rates were lower in the mesalamine group.


Early studies with oral prednisolone (1 mg/kg/day) showed induction of clinical remission in 92% within 7 weeks; however, only 29% also achieved endoscopic remission. Subsequent studies highlighted prednisone’s superiority (in combination with mesalamine) over enteral nutrition for induction of remission. Budesonide was investigated in a double-blind, multicenter trial using a controlled-release preparation at different daily doses; in an 8-week trial, the 9 mg daily dosing was superior to the 15 mg and 3 mg doses and resulted in more than a two-fold remission rate.

Head-to-head studies of budesonide with prednisolone subsequently showed relatively equal efficacy in inducing remission; however, budesonide was noted to have fewer side effects. The most recent systematic review showed that, whereas budesonide was somewhat less efficacious than conventional corticosteroids in induction of remission in CD of the ileum or ascending colon, it was the preferable therapy given its fewer adverse effects.


Pilot studies with ciprofloxacin showed modest difference in induction of remission compared to placebo. Subsequent studies looking at the addition of ciprofloxacin and metronidazole to budesonide found no significant benefit gain in induction, although patients with colonic CD seemed to benefit the most from the antibiotics. As single agents, probiotics have not shown significant benefit in induction of remission for CD.

Biological Agents

After FDA approval for the treatment of CD, infliximab was shown to have higher remission rates than placebo with a single infusion of up to 4 weeks out. Remission was noted to be even higher on retreatment with infliximab versus placebo at 8-week intervals, out to 44 weeks. In the subsequent large ACCENT I trial, 58% responded by week 2 and 27% achieved remission; higher response at week 10 was achieved by a loading dose of three infusions rather than a single infusion.

The CHARM trial evaluated adalimumab for remission in patients with CD. Fifty-eight percent (58%) of the patients responded to adalimumab and, at week 26, more patients in the adalimumab-treated group (weekly or every other weekly) were in remission compared to placebo.

The CLASSIC I trial added further evidence to this, with differences in outcomes noted at 4 weeks. Patients naive to anti-TNF agents had better response than those who had failed infliximab.

The PRECISE trial was a randomized, double-blind, placebo-controlled trial evaluating the efficacy of certolizumab in treatment of CD; results showed moderate improvement in response rates but not in remission rates. Like adalimumab, patients naive to anti-TNF agents had better response than those who had failed infliximab.

Natalizumab, in a double-blind, placebo-controlled trial, was shown to increase the rate of clinical remission and response and improve quality of life at multiple time points. However, the results of two larger randomized controlled trials showed a small, nonsignificant improvement in response and remission compared to placebo in patients with CD.



Early studies with different oral mesalamine preparations showed significant benefit over placebo (45% over 29%), particularly in groups deemed to be at high risk of relapse. Subsequent multicenter studies looking at starting doses lower than 4 gm/day of mesalamine in maintenance found nearly 80% of patients in remission by the end of average 14 months, with 72% in remission continuously after 12 months of therapy.

Mesalamine has also been show to facilitate steroid withdrawal and reduce relapse rate in certain patient subgroups, including the postsurgical setting (see below), those with ileitis, and those with prolonged disease duration.

However, more recently, a systematic review including seven placebo-controlled trials showed 5-ASA not to be effective for maintenance of medically induced remission over 12 to 24 months. These results appear to apply to the class as a whole, although some formulations (pH 7-dependent mesalazine formulations) appear to have some potential benefit.

6-Mercaptopurine and azathioprine

Initial trials confirming the role of azathioprine in maintenance showed significant relapse upon withdrawal after 6 months of maintenance. Subsequent double-blind, placebo-controlled trials confirmed a significant advantage (42% vs. 7%) over placebo in maintenance of remission in CD.

An initial systematic review including 5 randomized, double-blind, placebo-controlled trials of azathioprine showed two-fold odds of remission and a steroid sparing effect. An updated systematic review that included several trials of azathioprine and one trial of 6-MP showed an even higher odds ratio for remission and a significant steroid sparing effect. Furthermore, maintenance benefit appears to remain significant for those who continue therapy for up to 5 years.


Subcutaneous methotrexate, 25 mg weekly, has been effective with prednisone to induce remissions in refractory CD patients and to allow steroid-tapering. After entering remission with methotrexate, 65% of those assigned to maintenance with methotrexate remained in remission compared to 39% with placebo. Methotrexate may particularly be a good option for those intolerant of purine analogs.


Systemic steroids like prednisone are not ideal maintenance medications, given the side-effect profile over the long term. Early systematic reviews of corticosteroids for maintaining remission of CD, including three randomized double-blind, placebo-controlled trials of patients in clinical remission, showed no significant reduction in relapse over placebo over a 2-year period of follow-up. A subsequent systematic review looking only at budesonide in comparison to placebo included eleven studies and reached a similar conclusion, although some benefit (in the form of reduced CDAI scores) was noted in the budesonide group over 3 months.


While antibiotics may have some role in the management of postoperative and perianal CD, these do not have a proven role in maintenance of remission.

Biological agents

In patients who previously responded to infliximab, maintenance every 8 weeks showed significant superiority over placebo for up to 2 years (62% vs. 37%). In the ACCENT-1 trial, retreatment with infliximab was effective in maintaining remission at 30 and 54 weeks in responders and allowed withdrawal of corticosteroids. Subsequent trials confirmed adequacy of 5 mg/kg dosing every 8 weeks. However, up to 30% of patients failed to response: predictors of response include nonsmoker status, concurrent immunomodulator therapy, isolated colitis, and low-levels of TNFα.

The development of antibodies against infliximab has been shown to reduce duration of response to treatment; scheduled maintenance therapy and concomitant immunosuppressive therapy may reduce the magnitude of this loss of response.

Adalimumab has been evaluated in several RCTs, including CLASSIC I and CLASSIC II. Benefit for prolonged remission seems to be highest for those who achieved clinical response and underwent induction with adalimumab. Weekly or every-other-weekly dosing of the subcutaneous injections are equally effective in maintaining remission.

Certolizumab has considerably less data on remission compared to the other biological agents. In the PRECISE trials, compared to placebo, certolizumab appears to have had a clear advantage for maintenance of remission (provided it was the agent used for induction); however, its results were less impressive than the other biologics.

In ENACT-2, a large randomized placebo-controlled natalizumab maintenance study, patients who initially responded to natalizumab in the ENACT-1 induction trial were re-randomized and showed marked improvement in remission rates (61% vs. 28%) through 5 years. Subsequent enrollment of those who achieved remission provided even longer-term data of the efficacy of natalizumab for maintenance in CD.

Early therapy and combination therapy

The prognosis in CD depends on the disease extent, severity and duration. Patients who present at an early age and those with extensive disease have a worse prognosis as do patients with deep ulcerations at endoscopy, fistula, and individuals who smoke.

There is now evidence that early combined immunosuppression (and steroid-sparing strategies) may be the better way to manage newly diagnosed CD, as the rates of remission and reduction of corticosteroids appear improved. The recently published SONIC trial demonstrated that a combination of infliximab and azathioprine had improved outcomes (and less side effects) than either infliximab or azathioprine monotherapy for immunosuppressive-naïve patients with relatively early disease.


After resection of disease in patients with CD, those without treatment will almost universally have disease recurrence. Expressed actuarially, one study found recurrence at 1 year to be 27.5%, at 2 years to be 60.8%, and at 3 years to be 77.3%. Highest risk factors for recurrence include smoking, undergoing surgery at a young age or shortly after diagnosis, a perforating phenotype, extraintestinal manifestations, and perianal disease. Thus, early thought should be given to providing treatment to prevent or delay recurrence.


In double-blind, placebo-controlled trials evaluating mesalamine in the prevention of endoscopic recurrence, 3 gm/day mesalamine, beginning within 1 month after surgery resulted in less frequent and less severe endoscopic lesions at 12 months, with an odds ratio of recurrence with placebo versus mesalamine of 4. Subsequent larger studies, however, failed to confirm this significant benefit. A recent systematic review showed only modest potential benefit in this setting, plagued by publication bias and significant heterogeneity. Perhaps the best benefit with mesalamine can be found in patients with isolated small bowel disease. Its use can also be considered in those in whom immunosuppressive therapy is either not warranted or contraindicated.

6-Mercaptopurine and azathioprine

In one multicenter, double-blind trial comparing low-dose 6-MP to mesalamine and placebo, 6MP was more effective than placebo at preventing postoperative recurrence of CD over the 2 years of the study. In a recent meta-analysis of four trials using azathioprine and 6MP, the purine analogs were better at preventing 1-year and 2-year clinical and endoscopic postoperative recurrence compared to placebo or mesalamine. The purine analogs may have added benefit when combined in the short-term with antibiotics (see below).


Very few studies exist that address the role of methotrexate in the postoperative setting. The only substantial study was a prospective pilot study combining infliximab with low-dose methotrexate (10 mg/week PO) compared with mesalamine. The combination group had better outcomes, but the study must be interpreted as a pilot study as the number of subjects was very small and no randomization was performed.


Conventional corticosteroids are not effective in postoperative prevention of recurrence. Budesonide has been studied at a dose of 6 mg daily in a randomized, double-blind trial and offered no benefit in prevention of endoscopic recurrence. The benefit, if any, from budesonide in prevention of disease recurrence is short-term and does not appear to hold for a year.


Initial studies with metronidazole 20mg/kg/day for 3 months post-op showed statistically reduced clinical recurrence out to 1 year (4% vs. 25%). When combined with azathioprine, the effect is additive with 78% versus 55% recurrence in the metronidazole group versus combination group, respectively.

Ornidazole, another nitroimidazole antibiotic, at a dose of 1 gram/day for 12 months reduced recurrence at 3 years with an RR of 0.20. When combined in a meta-analysis, data shows an RR of 0.23 for clinical and 0.44 for endoscopic recurrence at 3 years for the nitroimidazole antibiotics. However, both antibiotics are limited by their side effect profiles.

Biological agents

A recent, randomized, double-blind, placebo-controlled study enrolling 20 patients who received either infliximab or placebo and followed for a year demonstrated that infliximab significantly decreased endoscopic and histologic recurrence for the first year after ileal resective surgery. Subsequent smaller studies suggested that lower dose (3 mg/kg) may be sufficient to prevent recurrence at 1 year. Finally, a 2-year postoperative update of these data showed that patients who started infliximab in response to recurrence had an attenuated response versus those who received prophylactic postoperative infliximab. There have yet to be controlled trials that evaluate either adalimumab or certolizumab pegol.

Penetrating disease

6-mercaptopurine and azathioprine

No randomized study has been done to assess the role of purine analogs on fistula healing. One study that included patients from a previous randomized study treated patients for 6 months with 6MP; 39% achieved complete fistulae closure, with mean time to response being 3 months. In combination with infliximab, a prospective pilot study suggested that concomitant 6-MP/azathioprine could prolong the effects of infliximab on fistula closure.


An initial study with 8 patients placed on metronidazole at 1 to1.5 g/day showed a 20-fold reduction in number of draining fistulae and a 50% reduction in the number of detectable fistulous openings; however, side effects limited the experience of the patients and fistula did re-open after discontinuation.

Biological agents

In a randomized, multicenter, double-blind, placebo-controlled trial of patients with draining abdominal or perianal fistulas for longer than 3 months, infliximab (at 5 or 10 mg/kg) performed twice as well as placebo in a 50% reduction in the number of draining fistulas on subsequent study visits; there was also a significant difference in complete closure of fistulas. To evaluate whether infliximab is an effective maintenance therapy for patients with fistulas, a different such trial showed that patients who respond to induction therapy with infliximab have an increased likelihood of a sustained response over a 54-week period with continued maintenance therapy.

Case series using adalimumab have shown some effectiveness in patients with enterocutaneous fistulas, but larger trials are needed. This observation seems to also hold in patients who have failed infliximab. Certolizumab has also shown some benefit although even less high-quality data exists regarding its utility.

A listing of a subset of second-line therapies, including guidelines for choosing and using these salvage therapies


Surgery is rarely curative in Crohn’s disease, and the disease predictably recurs proximal to an ileo-colonic anastomosis. Despite use of medical therapy, progression of anatomic damage can lead to complications that require surgical management. At 20 years of disease, up to 70% to 80% of patients with CD will have required intestinal surgery. Up to 30% may require a second operation by 10 years later. The yearly incidence of surgery for CD is 3% to 5%, within a mean time frame of 15 to 20 years. While the immune modulators and biologics appear to improve these figures somewhat, these need to be used early on in the disease to alter the course. Therefore, the indications for surgery include disease that is refractory to medical therapy or the presence of complications such as obstructing strictures, symptomatic fistulae, abscesses, intractable hemorrhage, or neoplasia (dysplasia or cancer).

Early in the study of surgical treatment of CD, the patterns of surgical treatment began to change, with increased utility of nonresectional options (e.g., strictureplasty for duodenal and jejunoileal disease) to reduce future short gut. Even in the presence of extensive colonic involvement, segmental colonic resection was shown to best maintain colonic length and intestinal continuity. Although segmental colonic resections may be sufficient for stricturing or focally fistulizing disease, permanent fecal diversion may be needed in more than 10% of CD patients; the latter may be performed for chronic active perianal disease or colorectal CD with anal stenosis.

Surgical intervention in perianal disease relies on whether the disease is symptomatic; aggressive intervention in the asymptomatic patient is often complicated by disease recurrence. Hemorrhoidectomy (removal of perianal skin tags) in selected patients can be undertaken safely but is frequently complicated by poor wound healing. Simple abscesses can be treated with incision and drainage as close to the anal verge as possible. If a concomitant fistula exists, seton placement may be warranted. Fistulotomy is best reserved for low-lying fistula with minimum sphincter involvement. More complicated fistula should be mapped out with diagnostic studies noted previously with surgical options including setons, or endoanal advancement flaps. Anorectal stenosis initially show response to dilation but may eventually require diversion; dilation should be avoided in patients with active proctitis.

Management of more complicated fistula (rectovaginal and anovaginal) first involves confirmation of the site of origin of the fistula, treatment of the inflammation associated with the disease, and identification of abscesses. Diversion may oftentimes be necessary before permanent repair. Endorectal flaps have also been evaluated in this setting with some success.

Endoscopic dilation of short-segment (< 4cm) strictures is gaining some popularity, but segmental resection or strictureplasty is more commonly used for fibrostenotic stricturing disease.

Listing of these, including any guidelines for monitoring side effects.

See Table VII for adverse events and monitoring.

Pregnancy and Crohn’s disease

Women with CD have similar fertility to the general population (5-14%), but surgical intervention may result in decreased fecundity and fertility. In addition, sexual dysfunction may be impaired. Pregnancy outcomes are generally the same for CD patients although disease activity at conception can result in an increased risk of fetal loss and preterm birth. No identifiable congenital malformations, however, have been noted.

The rate of flare in pregnancy is similar to that without pregnancy (26-34%). While most births can be planned as vaginal deliveries, women with significant perianal disease are usually recommended to undergo cesarean section.

Mesalamine agents and anti-TNF biologics are considered by the FDA to be category B. Category C medications include olsalazine, budesonide, ciprofloxacin, corticosteroids, cyclosporine, and tacrolimus. Category D medications include azathioprine/6-mercaptopurine. Category X medications include methotrexate.

How should I monitor the patient with Crohn's disease?

Monitoring disease complications and progress

As outlined below, patients on medical therapy should be monitored continuously for infection and be screened for disease complications, such as CRC.

In clinical trials, the CD Activity Index (CDAI) is used to define response and remission (CDAI <150); in clinical, practice, however, traditional monitoring of disease progress has relied more on global clinical symptoms included in the CDAI, such as the number of daily liquid stools, abdominal pain, general well-being, symptoms of EIM, use of antidiarrheal medications, the presence of an abdominal mass, and hematocrit.

More recently, the concepts of “mucosal healing” and deep remissions have come to light. There is a poor correlation between clinical symptoms and CRP or the severity of endoscopic lesions. Most therapies prior to the introduction of anti-TNF monoclonal antibodies have focused on reduction of clinical symptoms and signs. While different treatments result in different rates of mucosal healing, improved outcomes with accomplishing mucosal healing include sustained clinical remission, reduced need for surgery or hospitalization, and better quality of life.

With further evidence, endoscopic healing could become the standard goal of treatment. For now, however, disease progress can be monitored with the same diagnostic tests noted previously (namely, ileocolonoscopy, video capsule endoscopy (VCE), stool calprotectin or lactoferrin, serum tests such as CRP, and contrast radiography). For example, 6 months after surgery, guidelines recommend an ileocolonoscopy 6 months after surgery to assess for endoscopic recurrence.

See Table VI, Figure 2a, and Figure 2b.

Figure 2a.

2a Simplified CD treatment algorithm.

Figure 2b.

Simplified perianal CD treatment algorithm.

What's the evidence?

Nikolaus, S, Schreiber, S. "Diagnostics of inflammatory bowel disease". Gastroenterology. 2007. pp. 1670-89.

(Up-to-date review of diagnosis and diagnostic algorithms to guide therapy.)

Lichtenstein, GR, Hanauer, SB, Sandborn, WJ. "Management of Crohn's disease in adults". Am J Gastroenterol. vol. 104. 2009. pp. 465-834.

(Up-to-date practice guidelines from the American College of Gastroenterology.)

Farraye, FA. "AGA technical review on the diagnosis and management of colorectal neoplasia in inflammatory bowel disease". Gastroenterology. vol. 138. 2010. pp. 746-74.

(AGA review on management of colorectal neoplasia in inflammatory bowel disease.)

Gisbert, JP, Gonzalez-Lama, Y, Mate, J. "5-aminosalicylates and renal function in inflammatory bowel disease: a systematic review". Inflamm Bowel Dis. vol. 13. 2007. pp. 629-38.

(Review of an important class of treatment and adverse effects.)

Hanauer, SB. "Advances in the management of Crohn's disease: economic and clinical potential of infliximab". Clin Ther. vol. 20. 1998. pp. 1009-28.

(Important implications on the year of approval of infliximab for Crohn's disease.)

Camma, C. "Mesalamine in the maintenance treatment of Crohn's disease: a meta-analysis adjusted for confounding variables". Gastroenterology. vol. 113. 1997. pp. 1465-73.

(Meta-analysis of an important class of therapeutics in Crohn's disease.)

Prefontaine, E. "Azathioprine or 6-mercaptopurine for maintenance of remission in Crohn's disease". Cochrane Database Syst Rev. 2009. pp. CD000067.

(Meta-analysis of an important class of therapeutics in Crohn's disease.)

Peyrin-Biroulet, L. "Azathioprine and 6-mercaptopurine for the prevention of postoperative recurrence in Crohn's disease: a meta-analysis". Am J Gastroenterol. vol. 104. 2009. pp. 2089-96.

(Meta-analysis of key post-operative management.)

Ford, AC, Khan, KJ, Talley, NJ. "Moayyedi. 5-aminosalicylates prevent relapse of Crohn's disease after surgically induced remission: systematic review and meta-analysis". Am J Gastroenterol. vol. 106. 2011. pp. 413-20.

(Meta-analysis of key post-operative management.)

Gordon, M, Naidoo, K, Thomas, AG. "Oral 5-aminosalicylic acid for maintenance of surgically-induced remission in Crohn's disease". Cochrane Database Syst Rev. 2011. pp. CD008414.

(Meta-analysis of post-operative management.)
You must be a registered member of ONA to post a comment.

Sign Up for Free e-newsletters

Regimen and Drug Listings


Bone Cancer Regimens Drugs
Brain Cancer Regimens Drugs
Breast Cancer Regimens Drugs
Endocrine Cancer Regimens Drugs
Gastrointestinal Cancer Regimens Drugs
Genitourinary Cancer Regimens Drugs
Gynecologic Cancer Regimens Drugs
Head and Neck Cancer Regimens Drugs
Hematologic Cancer Regimens Drugs
Lung Cancer Regimens Drugs
Other Cancers Regimens
Rare Cancers Regimens
Skin Cancer Regimens Drugs