The number of studies, trials and systematic reviews that were included was unclear.
Diagnosis: Radionuclide scanning, computed tomography angiography, and magnetic resonance and standard angiography were discussed, largely without supporting evidence. The radionuclide scanning accuracy rates were reported to range from 24% to 91% across studies, and standard angiography in patients with overt obscure gastrointestinal bleeding was reported to have lower diagnostic yield (20%) than immediate wire capsule endoscopy (53%; p=0.016).
Endoscopic management: Argon plasma coagulation was reported to be safe, available, easy to use and effective. Electrocoagulation had complications and lower efficacy than argon plasma coagulation. Laser photocoagulation was effective, but there were concerns about the risk of perforations. Endoscopic ligation could be effective in the short-term, but its long-term efficacy was unclear. Injection sclerotherapy was reported in two studies, but with no efficacy data. Endoscopic clips and endoscopic resection were mentioned in single case reports.
Transcatheter angiography and intervention: This was stated to be appropriate for specific groups of patients only; those with active gastrointestinal bleeding or who were unsuitable for endoscopic surgery. Success rates of 80 to 90% were reported, with complications in 5% to 9% of patients.
Surgery: No data on effectiveness were provided.
Pharmacotherapy: Hormonal therapy (five studies) was shown to be effective in small, retrospective or uncontrolled studies, with less positive results in two larger studies. Thalidomide (one controlled open-label trial, one cohort study, and one case series) reduced the number of bleeding episodes, with increased adverse events. This was recommended only as a treatment option where other therapies were not suitable. Lenalidomide was mentioned, but no studies were reported. Octreotide (a meta-analysis of three cohort studies, with 62 patients; plus two cohort studies, with 26 patients) was reported to be beneficial for patients with recurrent chronic bleeding, with no serious adverse events.