PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 22, 2026The Cardiothoracic Surgeon5 citationsOpen Access

Regional disparities in surgical learning curves and maturity for congenital heart surgery in Brazil’s public health system: when volume is not enough

GMG. MartinsVCVerônica Cristina Carvalho Chaves

Key Result

Structural constraints limit mortality reduction in congenital heart surgery, with high-capacity regions achieving <2% mortality after 1800-3200 cases, while constrained regions plateaued above 2%.

Key Points

  • The study evaluates the impact of cumulative surgical volume on mortality rates for congenital heart defects in Brazil's public health system.
  • Analyzed a nationwide cohort of 48,721 ASD and VSD repairs from 2008 to 2024.
  • Modeled regional learning curves using non-linear smoothing of mortality data.
  • Developed a Surgical Maturity Index combining experience and outcome stability.
  • Procedure volume varied significantly across regions, with Southeast performing 44% of operations.
  • Mortality rates ranged from 1.4% to 4.0% for ASD and 5.8% to 14.0% for VSD.
  • Regions with high case volumes improved mortality to below 2% after a certain number of procedures, while constrained regions plateaued.
  • A strong correlation was found between the Surgical Maturity Index and lower mortality rates.

Structured PICO

Does cumulative surgical volume alone reduce in-hospital mortality for ASD and VSD repairs across Brazil's macro-regions?

P
Population
48,721 atrial septal defect (ASD) and ventricular septal defect (VSD) repairs in Brazil's public health system (SUS) between 2008 and 2024 (29,104 ASD repairs and 19,617 VSD closures).
I
Intervention
Cumulative surgical volume of atrial septal defect (ASD) and ventricular septal defect (VSD) repairs
O
Outcome
Annual in-hospital mortality ratehard clinical

Systemic constraints impose a ceiling on volume-driven improvement in congenital heart surgery, indicating that targeted capability transfer is more effective than volume centralization alone in resource-constrained public health systems.

Limitations

  • Aggregated nature of the data
  • Absence of individual-level outcomes preventing patient- or hospital-level mixed-effects models
  • Data sparsity and collinearity regarding pandemic effects
  • Potential underreporting

Abstract

Abstract Background The volume-outcome relationship in congenital heart surgery is established, but its validity in systems with extreme regional inequalities is unclear. This study assessed whether cumulative volume alone reduces mortality for atrial septal defect (ASD) and ventricular septal defect (VSD) repairs across Brazil’s macro-regions or if structural barriers limit improvement. Methods Nationwide cohort of 48,721 ASD and VSD repairs in Brazil’s public health system (SUS), 2008–2024. Regional learning curves were modeled using non-linear smoothing of in-hospital mortality against cumulative volume. A composite Surgical Maturity Index (IMS) integrated experience and outcome stability. Results Procedure volume was highly unequal (Southeast 44%). Mortality ranged 1.4–4.0% (ASD) and 5.8–14.0% (VSD). High-capacity regions achieved 3000 cases (interaction p = 0.004). IMS strongly correlated with lower mortality ( ρ = − 0.89). The Northeast showed rapid VSD maturity gains (+ 15.2% annual change, p = 0.001) via quality-improvement partnerships. Conclusions Systemic constraints impose a ceiling on volume-driven improvement. Targeted capability transfer accelerates maturity more effectively than volume centralization alone, informing equitable policy in public health systems.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Martins et al. (2026) studied this question. Structural constraints limit mortality reduction in congenital heart surgery, with high-capacity regions achieving <2% mortality after 1800-3200 cases, while constrained regions plateaued above 2%.

synapsesocial.com/papers/6971bd90642b1836717e2372https://doi.org/10.1186/s43057-026-00186-4
Ask AI
Helpful
Bookmark
Share
View Full Paper