The incidence of INSTIA among consecutive patients undergoing ileo-colonoscopic examination at our center was 8.6%. Most common symptoms seen in patients with INSTIA (n=60) were pain abdomen (61.6%), diarrhoea (41.6%), and constipation (40%). Ulcers were the most common findings [60% (62.1% in group A vs 58.1% in group B) on ileo-colonoscopy followed by nodularity [28.3% (27.65 in group A vs 32.3% in group B)], and erosions [11.6% (10.3% in group A vs 9.6% in group B)]. The baseline histopathologic findings were comparable in both groups. Most patients had non-specific inflammation (69.0% vs 58.1% in groups A and B), followed by no specific pathology/normal histology (27.6% vs 29.0% in groups A and B). Majority of the study subjects had improvement in clinical symptoms, endoscopic and histopathologic findings after a mean follow-up of 3.36 ± 0.27 months without any significant difference between the groups. The study results suggest that natural history of INSTIA follow a favorable course with spontaneous mucosal healing within 3 months in most patients. Further, there was no impact of empirical antimicrobials on the rate of mucosal healing.
The natural history of asymptomatic cases of terminal ileitis has been described in a few studies but data on patients with symptomatic terminal ileitis of non-specific etiology is scarce. [10, 11] The outcome of any protocolized and symptomatic treatment on these patients has not been explored yet. Jeong et al (2008)  prospectively analyzed ileo-colonoscopic findings in 3921 patients and reported macroscopic abnormalities in the terminal ileum in 125 patients (3.7%). Clinically significant histopathology was observed in 11 cases, giving a diagnostic yield of 0.3% in all ileoscopies. Seven out of 11 patients with clinically significant histopathology were diagnosed with Crohn's disease. Analysis of clinical data from all 3921 patients showed that the most common symptoms reported by patients were diarrhoea (27%), abdominal pain (31.9%), constipation (7%), hematochezia (10.2%).  These findings are different from our study due to dissimilar inclusion criteria. On the other hand, an Indian study by Kedia et al (2016)  reported a higher rate of abdominal pain (82.2%), diarrhoea (35.6%), gastrointestinal bleed (15.6%). These findings are different from our results, which may also be attributed to difference in the inclusion criteria. In the study by Kedia et al, 68.8 % of patients with isolated terminal ileitis were associated with specific diagnoses such as tuberculosis or Crohn’s disease, whereas in our study, the subgroup of patients with specific pathology or etiology (like Crohn’s disease and tuberculosis) were excluded. Furthermore, most of the study population in the study by Kedia et al comprised of patients referred for IBD and tuberculosis to quaternary referral hospitals (AIIMS, New Delhi). Conversely in our study, most patients had undiagnosed symptoms and many of our cases may have been suffering only from irritable bowel syndrome (IBS).
In our study, the majority of the patients had non-specific inflammation (69.0 % vs 58.1% in groups A and B), followed by no-specific pathology/normal histology (27.6% vs 29.0% in groups A and B). Chronic inflammation was found in 3.4% of the patients in the protocolized treatment group and 9.7% in the symptomatic treatment group. Kedia et al (2016) , found that out of a total of 45 patients with terminal ileitis, 38 (84.4%) patients had non-specific chronic inflammation even when they have included patients with specific pathology such as Crohn's disease or intestinal tuberculosis.
There is very little data available on the therapeutic management of patients with INSTIA. Curville et al (2009)  conducted a clinico-pathological study of 29 patients with isolated ileitis without any history of IBD. Only patients with at least 2 years of follow-up after colonoscopy (range: 2.2 to 12.6 years) were included. Out of all these patients, nearly half were symptomatic. Of these symptomatic patients, 80% developed Crohn’s disease on follow-up. In contrast, 40% of asymptomatic patients developed Crohn’s disease on follow-up. The author concluded that the presence of symptoms may be a significant predictor of Crohn’s disease. These findings were not like our results. In the present study, after a mean follow-up of 3.36 ± 0.27 months, repeat ileo-colonoscopy showed resolution of INSTIA in most patients, though a longer follow-up is not available.
Wang et al (2011)  studied seven patients with small bowel ulcers of non-specific aetiology. On follow-up examination, it was found that five out of seven patients improved clinically on empirical treatment while the ulcers persisted in four patients and one patient had recurrent symptoms. None of the patients experienced the aggravation of disease condition or complication during the follow-up period of 7 years. A few findings of Wang et al (2011) study such as clinical improvement are similar to our findings. A recent Indian study conducted by Mehta et al (2017),  followed up 60 patients with the terminal ileal disease. The majority (55%) of the cases had a diagnosis of Crohn's disease, intestinal tuberculosis, NSAIDs induced enteropathy, and eosinophilic enteritis. However, almost half of the patients had nonspecific terminal ileal ulcers on ileo-colonoscopy. In this study, more than 88% of the patients with non-specific ulcers improved clinically after symptomatic treatment which suggests that non-specific terminal ileal ulcers usually have a mild benign clinical course. Three patients, who had a diagnosis of non-specific ulcer developed cryptitis or crypt abnormality and villous atrophy/abnormality and were treated as Crohn’s disease. These patients neither exhibited any clinical aggravation nor experienced any serious complications which warranted any aggressive therapeutic management.  These study findings are concurrent with our results.
To the best of our knowledge, ours is the first study that has explored the efficacy of a protocolized treatment. Although there are few studies that showed the efficacy of symptomatic treatment on the non-specific ileal ulcer but none of them compared the effect with the addition of an arm of treatment with antimicrobial agents. Our study showed that both groups with or without protocolized treatment showed comparable response with symptomatic treatment in decreasing the symptoms, improving the ileo-colonoscopy findings like ulcers and nodularity, improving the histopathologic findings. In other words, randomization to the protocolized treatment arm did not have any significant impact on the course of INSTIA.
In usual clinical practice, Crohn's disease, NSAIDs induced enteropathy, and intestinal tuberculosis in a country like India are the common differential diagnosis for patients with symptomatic terminal ileitis. But in many instances such background diagnoses are absent and histopathologic features are inconclusive which makes the clinical diagnosis much more difficult. Many such patients may be erroneously given a trial of treatment for Crohn’s disease or tuberculosis. The findings of the present study seem to resolve this clinical dilemma and suggest that patients with INSTIA may be treated symptomatically with a reasonably good outcome in most. However, if the possibility of Crohn’s disease or tuberculosis cannot be ruled out, then a follow-up colonoscopy with histopathological evaluation may be considered for diagnosis, and patients with persistent abnormalities need to be followed up much longer.
Our study is a pilot study and has several limitations. Firstly, our study is a single center study with small sample size, which limits the generalizability of our results in the general population. Another limitation may be the absence of a complete small bowel workup, which leaves a possibility of undetected ulcers proximal to the terminal ileum. Although we had performed the CT enterography in selected cases (patients with clinical suspicion of IBD/intestinal TB or histological feature of chronic inflammation), it was not done in all the patients. Patients who underwent CT enterography did not have any small bowel abnormalities. Finally, it is possible that such abnormalities in the terminal ileum may appear and disappear in a recurrent manner. A relatively longer follow-up period could have provided more insight into long-term outcomes. Hence, long-term follow-up studies with large sample sizes and the prospective validation of our results are required to substantiate these findings.