Incidence and Procedural Risk Factors for Symptomatic Cholecystitis After Yttrium-90 Transarterial Radioembolization: A Systematic Review and Meta-analysis

Technical Summary: Incidence and Procedural Risk Factors for Symptomatic Cholecystitis After Y-90 Transarterial Radioembolization

Problem Statement
Yttrium-90 transarterial radioembolization (Y-90 TARE) is a standard treatment for primary and metastatic liver malignancies. However, the procedure carries a risk of nontarget delivery to extrahepatic structures, specifically the gallbladder. Because the cystic artery frequently arises from the right hepatic arterial circulation, the gallbladder is vulnerable to injury via two potential mechanisms: direct radiation injury from microspheres reaching the cystic arterial bed, or ischemic injury resulting from coil/particulate occlusion, vasospasm, or flow redistribution.

Existing literature presents conflicting data regarding the incidence of clinically symptomatic cholecystitis and the efficacy of various preventive strategies, such as distal catheter positioning or permanent cystic-artery (CA) coiling. Furthermore, previous reports often conflate imaging-only abnormalities (e.g., wall thickening, uptake) with clinical disease, and many studies exclude zero-event cohorts, potentially skewing incidence estimates. This review aims to resolve these ambiguities by estimating the true incidence of clinically symptomatic cholecystitis and synthesizing evidence on procedural risk factors.

Methodology
This study is a systematic review and meta-analysis adhering to PRISMA 2020 and PRISMA-S principles.

• Data Sources: A comprehensive search was conducted across PubMed/MEDLINE, CINAHL, and Google Scholar (inception through September 22, 2026), supplemented by citation chaining.
• Eligibility Criteria: The review included adult cohorts treated with Y-90 glass or resin microspheres that reported an exact numerator (symptomatic events) and denominator (treated patients). Crucially, the primary outcome was strictly defined as clinically symptomatic cholecystitis; imaging-only abnormalities without compatible clinical disease were excluded. Zero-event cohorts were explicitly retained.
• Statistical Analysis: The primary pooled incidence was calculated using a one-stage random-intercept binomial-normal generalized linear mixed model (GLMM) with a logit link. This approach allowed for the retention of zero-event studies without continuity correction. Heterogeneity was assessed using tau, tau-squared, and I-squared statistics. An approximate 95% prediction interval was calculated to estimate the dispersion of underlying risks across future settings. Sensitivity analyses were performed to test the influence of cohort types (general vs. anatomy-enriched), microsphere types, and publication eras.

Key Contributions

1. Strict Clinical Definition: The review distinguishes between radiographic findings and clinical disease, addressing a common source of overestimation in prior literature.
2. Inclusion of Zero-Event Cohorts: By retaining studies with no reported events, the meta-analysis provides a more accurate, conservative estimate of incidence for rare adverse events.
3. Differentiation of Cohort Types: The analysis separates "general" cohorts from "anatomy- or technique-enriched" cohorts (e.g., those specifically studying patients with high-risk anatomy or specific embolization techniques), clarifying that higher event rates in the latter are likely due to selection bias rather than the technique itself.
4. Procedural Risk Synthesis: The paper provides a narrative synthesis of comparative evidence regarding catheter positioning and cystic-artery embolization, moving beyond simple event counts to evaluate the quality of comparative data.

Results

• Incidence: From 44 eligible reports (43 independent cohorts), 5,380 patients were analyzed, yielding 67 symptomatic events. The pooled incidence of clinically symptomatic cholecystitis was 1.32% (95% profile-likelihood CI, 0.88%–1.89%). The crude aggregate proportion was 1.25%.
• Heterogeneity: Moderate between-study heterogeneity was observed (I² = 50.4%). The 95% prediction interval was wide (0.28%–6.05%), indicating that incidence can vary significantly depending on patient selection, vascular anatomy, and surveillance methods.
• Cohort Comparisons: General and mapping cohorts showed a pooled incidence of 1.21%, whereas anatomy- and technique-enriched cohorts showed a higher rate of 2.65%. The authors note this elevation reflects the higher baseline risk of the selected populations rather than a causal effect of the specific interventions studied.
• Procedural Factors:
  • Catheter Positioning: Evidence from Piasecki et al. suggests that proximal catheter positioning relative to the cystic artery origin is associated with a higher risk of technetium-99m macroaggregated albumin (MAA) leakage (RR ≈ 2.5) compared to distal positioning.
  • Cystic-Artery Coiling: Comparative evidence does not support routine permanent cystic-artery coil embolization. Studies showed no significant reduction in event rates with coiling (e.g., 2.9% vs 2.7% in Powerski et al.) and raised concerns that permanent occlusion may contribute to ischemic injury or collateralization.
  • Imaging vs. Clinical Disease: Gallbladder uptake and wall thickening were found to be poorly specific for clinical cholecystitis. For instance, Sag et al. reported imaging abnormalities in 84 of 85 cases but only one clinical event.
• Timing: Clinical presentation varies by mechanism. Embolic/ischemic events tend to occur early (within 48 hours), while radiogenic injury may present weeks later.

Significance and Claims
The paper concludes that clinically symptomatic cholecystitis after Y-90 TARE is an uncommon event, occurring in approximately 1.3% of patients on average. However, the wide prediction interval underscores that risk is not uniform and depends heavily on local anatomy and procedural execution.

The authors assert that the available evidence supports careful vascular mapping and distal catheter positioning (when tumor coverage permits) as the most effective preventive strategy. Conversely, the review explicitly states that routine permanent cystic-artery coiling is not supported by comparative data and may introduce ischemic risks. Finally, the paper emphasizes that imaging abnormalities alone should not be diagnosed as symptomatic cholecystitis; clinical correlation is required. The findings suggest that prevention should prioritize anatomy-guided technical precision over routine prophylactic embolization.