Pericardiectomy for Constrictive Tuberculous Pericarditis: A Systematic Review and Meta-analysis on the Etiology, Patients’ Characteristics, and the Outcomes.

Figures

• 

  Figure 1. The Newcastle-Ottawa scale for evaluation of methodological quality and bias of studies.

• 

  Figure 2. PRISMA flow diagram-2020 for systematic reviews which include searches of databases and registers only. PRISMA - Preferred Reporting Item for Systematic Reviews and Meta-analysis; n - number of articles

• 

  Figure 3. All-cause mortality after pericardiectomy in patients with constrictive tuberculous pericarditis and other etiologies. TB Events: Mortality after pericardiectomy in patients with constrictive pericarditis with tuberculosis as etiology. Other Events: Mortality after pericardiectomy in patients with constrictive pericarditis with etiologies other than tuberculosis. The Cochran-Mantel-Haenszel method and the random-effects model were used to calculate the pooled risk ratio. M-H - Mantel-Haenszel Test; Tau2 - Tau-squared test for random effects model; Chi2 - Chi-squared test; df - degree of freedom; I2 - I2 test for heterogeneity; Z - Standard score References [2,8,10]

• 

  Figure 4. In-hospital mortality versus one-year mortality post-pericardiectomy in patients with constrictive pericarditis. In-hospital mortality events: Mortality in patients within 30 days of pericardiectomy for constrictive pericarditis. One-year mortality Events: Mortality in patients within one year of pericardiectomy for constrictive pericarditis. The Cochran-Mantel-Haenszel method and the random-effects model were used to calculate the pooled risk ratio. M-H - Mantel-Haenszel Test; Tau2 - Tau-squared test for random effects model; Chi2 - Chi-squared test; df - degree of freedom; I2 - I2 test for heterogeneity; Z - Standard score References [2,9,13,15]

• 

  Figure 5. Pre-operative NYHA grade III and IV versus post-operative NYHA grade III and IV after one year in patients with constrictive tuberculous pericarditis. Pre-op NYHA grade 3/4: Pre-operative New York Heart Association (NYHA) grades III and IV in patients with constrictive tuberculous pericarditis. Post-op NYHA grade 3/4: Post-operative NYHA grades III and IV after one year in patients with constrictive tuberculous pericarditis. The Cochran- Mantel- Haenszel method and the random-effects model were used to calculate the risk ratio. M-H - Mantel-Haenszel Test; Tau2 - Tau-squared test for random effects model; Chi2 - Chi-squared test; df - degree of freedom; I2 - I2 test for heterogeneity; Z - Standard score References [7,12,16]

• 

  Figure 6. Mean of postoperative hospital stay for studies with more than 80% of TB cases. Random-effects DerSimonian-Laird model was used to calculate the pooled mean. τ2 - Hotelling's t-squared statistic; H2 - H2 test; θ - Vector of parameters of a probability distribution; τ: Between study variance References [3,7,12,14,16]

• 

  Figure 7. Mean of the standard deviation for the postoperative hospital stay for studies with more than 80% of TB cases. Random-effects DerSimonian-Laird model was used to calculate the pooled mean standard deviation. τ2 - Hotelling's t-squared statistic; H2 - H2 test; θ - Vector of parameters of a probability distribution; τ: Between study variance; I2 - I2 test for heterogeneity References [3,7,12,14,16]

• 

  Figure 8. Mean of postoperative ICU stay in studies with more than 80% of TB cases. Random-effects DerSimonian-Laird model was used to calculate the pooled mean. τ2 - Hotelling's t-squared statistic; H2 - H2 test; θ - Vector of parameters of a probability distribution; τ: Between study variance; I2 - I2 test for heterogeneity References [3,7,10,12,14,16]

• 

  Figure 9. The mean standard deviation for postoperative ICU stay for studies with more than 80% of TB cases. Random-effects DerSimonian-Laird model was used to calculate the pooled mean. τ2 - Hotelling's t-squared statistic; H2 - H2 test; θ - Vector of parameters of a probability distribution; τ: Between study variance; I2 - I2 test for heterogeneity References [3,7,10,12,14,16]

• 

  Figure 10. Mean in-hospital mortality in studies with more than 80% TB cases. Random-effects DerSimonian-Laird model was used to calculate the pooled mean. τ2 - Hotelling's t-squared statistic; H2 - H2 test; θ - Vector of parameters of a probability distribution; τ: Between study variance; I2 - I2 test for heterogeneity References [3,7,10,12,14,16]

• 

  Figure 11. Funnel plot for studies with more than 80% of TB cases. References [3,7,10,12,14,16]

• 

  Figure 12. Mean of postoperative hospital stay in studies with mixed etiologies. Random-effects DerSimonian-Laird model was used to calculate the pooled mean. τ2 - Hotelling's t-squared statistic; H2 - H2 test; θ - Vector of parameters of a probability distribution; τ: Between study variance; I2 - I2 test for heterogeneity References [9,11,13,15]