Small-bowel capsule endoscopy (SBCE) has undergone substantial technological evolution over the past two decades, accompanied by changes in clinical indications and referral patterns. From 2002–2013, the PillCam™ SB1/SB2 platforms were used at our center, while from 2014–2025 the PillCam™ SB3 system—offering higher image resolution, adaptive frame rate, and prolonged battery life—was introduced. The aim of this study is to compare indications, completion rates, transit times, and diagnostic yield of SBCE across two consecutive 11-year periods (2002–2013 and 2014–2025).

We performed a retrospective analysis of 1.484 consecutive SBCE examinations (1.048 in 2002–2013 and 436 in 2014–2025). Indications were classified according to ESGE/ASGE guidelines as “appropriate” (obscure GI bleeding, iron-deficiency anemia [IDA], Crohn’s disease, inherited polyposis syndromes, imaging abnormalities, and celiac disease) or “low-yield” (abdominal pain, diarrhea, malabsorption). Indications outside ESGE/ASGE criteria were also recorded. Diagnostic yield was defined as the presence of clinically relevant findings at the time of examination. Statistical analysis was performed using t-test and chi-square testing.

Significantly more SBCEs were performed in the first period than in the second (1,048 vs. 436, p<0.001). Patients examined in the later period were significantly older (mean age 60 vs. 66 years, p<0.001). IDA and obscure GI bleeding became more frequent indications in the second period (IDA: 39% vs. 54%; GI bleeding: 14% vs. 25%; both p<0.001), whereas Crohn’s disease was significantly less common (13% vs. 5%, p<0.001). Examinations performed outside ESGE/ASGE guidelines decreased markedly from 2.76% to 0.07% (p<0.01). Completion rate improved substantially from 76% to 94% (p<0.00001), and small-bowel transit time increased from 281 to 321 minutes (p=0.0001). Estimated diagnostic yield decreased significantly in the second period for GI bleeding (86% to 61%, p<0.001), IDA (78% to 53%, p<0.001), and Crohn’s disease (75% to 45%, p<0.01).

Over the last two decades, stricter adherence to guideline-based indications, refined interpretation criteria, and technological enhancements of SBCE have resulted in fewer but more targeted examinations, improved completion rates, and more reliable evaluation. Despite technological advancements, diagnostic yield decreased, likely reflecting increasingly selective patient referral and more stringent diagnostic thresholds rather than diminished test performance.