Cavernous aneurysm meta analysis 2
Niels Pacheco
2024-05-25
———————————- ASYMPTOMATIC CAVERNOUS ANEURYSM ——————————————–
# Load necessary library
library(tibble)
# Create the dataframe
asymptomatic <- tibble::tibble(
Study_ID = c("Fehrenbach et al., 2022", "Goldenberg-Cohen et al., 2003", "Junior et al., 2014", "Junior et al., 2019", "Kupersmith, 1992", "Kupersmith, 2002", "Lievo, 2023", "Linskey et al., 1990", "Stiebel-Kalish, 2005", "van Rooji et al., 2012", "Vercelli et al., 2019"),
Sample = c(64, 31, 100, 201, 70, 174, 250, 37, 206, 85, 194),
asymptomatic = c(23, 12, 32, 135, 12, 24, 201, 15, 36, 22, 147),
size = c(6.6, NA, NA, NA, NA, 19.6, 16.2, NA, NA, 22.71, 8.6)
)
# Display the dataframe
print(asymptomatic)## # A tibble: 11 × 4
## Study_ID Sample asymptomatic size
## <chr> <dbl> <dbl> <dbl>
## 1 Fehrenbach et al., 2022 64 23 6.6
## 2 Goldenberg-Cohen et al., 2003 31 12 NA
## 3 Junior et al., 2014 100 32 NA
## 4 Junior et al., 2019 201 135 NA
## 5 Kupersmith, 1992 70 12 NA
## 6 Kupersmith, 2002 174 24 19.6
## 7 Lievo, 2023 250 201 16.2
## 8 Linskey et al., 1990 37 15 NA
## 9 Stiebel-Kalish, 2005 206 36 NA
## 10 van Rooji et al., 2012 85 22 22.7
## 11 Vercelli et al., 2019 194 147 8.6——————————- Meta-analysis of proportions of asymptomatic cavernous aneurysm —————————
# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}## Loading required package: meta## Loading required package: metadat## Loading 'meta' package (version 7.0-0).
## Type 'help(meta)' for a brief overview.
## Readers of 'Meta-Analysis with R (Use R!)' should install
## older version of 'meta' package: https://tinyurl.com/dt4y5drs# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis1 <- metaprop(event = asymptomatic, n = Sample, data = asymptomatic,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis1)## proportion 95%-CI %W(random)
## Fehrenbach et al., 2022 0.3594 [0.2432; 0.4890] 9.1
## Goldenberg-Cohen et al., 2003 0.3871 [0.2185; 0.5781] 8.7
## Junior et al., 2014 0.3200 [0.2302; 0.4208] 9.2
## Junior et al., 2019 0.6716 [0.6021; 0.7361] 9.3
## Kupersmith, 1992 0.1714 [0.0918; 0.2803] 8.9
## Kupersmith, 2002 0.1379 [0.0904; 0.1982] 9.2
## Lievo, 2023 0.8040 [0.7493; 0.8513] 9.3
## Linskey et al., 1990 0.4054 [0.2475; 0.5790] 8.8
## Stiebel-Kalish, 2005 0.1748 [0.1255; 0.2336] 9.2
## van Rooji et al., 2012 0.2588 [0.1699; 0.3652] 9.1
## Vercelli et al., 2019 0.7577 [0.6912; 0.8162] 9.3
##
## Number of studies: k = 11
## Number of observations: o = 1412
## Number of events: e = 659
##
## proportion 95%-CI
## Random effects model 0.3921 [0.2336; 0.5772]
##
## Quantifying heterogeneity:
## tau^2 = 1.5480 [0.6402; 4.5209]; tau = 1.2442 [0.8001; 2.1262]
## I^2 = 97.1% [96.1%; 97.9%]; H = 5.91 [5.05; 6.92]
##
## Test of heterogeneity:
## Q d.f. p-value
## 349.73 10 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis1, layout = "RevMan")
Meta-analysis of proportions of asymptomatic cavernous aneurysm
——————————- Metaregression of the proportion of asymptomatic and size —————————
m.one.reg <- metareg(meta_analysis1, ~size)## Warning: 6 studies with NAs omitted from model fitting.m.one.reg##
## Mixed-Effects Model (k = 5; tau^2 estimator: DL)
##
## tau^2 (estimated amount of residual heterogeneity): 2.2064 (SE = 1.8968)
## tau (square root of estimated tau^2 value): 1.4854
## I^2 (residual heterogeneity / unaccounted variability): 98.20%
## H^2 (unaccounted variability / sampling variability): 55.50
## R^2 (amount of heterogeneity accounted for): 0.00%
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 166.4962, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.7100, p-val = 0.3994
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt 1.1687 1.7300 0.6755 0.4993 -2.2220 4.5593
## size -0.0912 0.1082 -0.8426 0.3994 -0.3032 0.1209
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1bubble(m.one.reg, studlab = FALSE)
———————————- SYMPTOMATIC CAVERNOUS ANEURYSM ——————————————–
# Load necessary library
library(tibble)
# Create the dataframe
symptomatic <- tibble::tibble(
Study_ID = c("Fehrenbach et al., 2022", "Goldenberg-Cohen et al., 2003", "Junior et al., 2014", "Junior et al., 2019", "Kupersmith, 1992", "Kupersmith, 2002", "Lievo, 2023", "Linskey et al., 1990", "Stiebel-Kalish, 2005", "van Rooji et al., 2012", "Vercelli et al., 2019"),
Sample = c(64, 31, 100, 201, 70, 174, 250, 37, 206, 85, 194),
Symptomatic = c(41, 19, 68, 66, 58, 150, 49, 22, 170, 63, 47),
Size = c(6.6, NA, NA, NA, NA, 19.6, 16.2, NA, NA, 22.71, 8.6)
)
# Display the dataframe
print(symptomatic)## # A tibble: 11 × 4
## Study_ID Sample Symptomatic Size
## <chr> <dbl> <dbl> <dbl>
## 1 Fehrenbach et al., 2022 64 41 6.6
## 2 Goldenberg-Cohen et al., 2003 31 19 NA
## 3 Junior et al., 2014 100 68 NA
## 4 Junior et al., 2019 201 66 NA
## 5 Kupersmith, 1992 70 58 NA
## 6 Kupersmith, 2002 174 150 19.6
## 7 Lievo, 2023 250 49 16.2
## 8 Linskey et al., 1990 37 22 NA
## 9 Stiebel-Kalish, 2005 206 170 NA
## 10 van Rooji et al., 2012 85 63 22.7
## 11 Vercelli et al., 2019 194 47 8.6——————————- Meta-analysis of proportions of symptomatic cavernous aneurysm —————————
# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis2 <- metaprop(event = Symptomatic, n = Sample, data = symptomatic,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis2)## proportion 95%-CI %W(random)
## Fehrenbach et al., 2022 0.6406 [0.5110; 0.7568] 9.1
## Goldenberg-Cohen et al., 2003 0.6129 [0.4219; 0.7815] 8.7
## Junior et al., 2014 0.6800 [0.5792; 0.7698] 9.2
## Junior et al., 2019 0.3284 [0.2639; 0.3979] 9.3
## Kupersmith, 1992 0.8286 [0.7197; 0.9082] 8.9
## Kupersmith, 2002 0.8621 [0.8018; 0.9096] 9.2
## Lievo, 2023 0.1960 [0.1487; 0.2507] 9.3
## Linskey et al., 1990 0.5946 [0.4210; 0.7525] 8.8
## Stiebel-Kalish, 2005 0.8252 [0.7664; 0.8745] 9.2
## van Rooji et al., 2012 0.7412 [0.6348; 0.8301] 9.1
## Vercelli et al., 2019 0.2423 [0.1838; 0.3088] 9.3
##
## Number of studies: k = 11
## Number of observations: o = 1412
## Number of events: e = 753
##
## proportion 95%-CI
## Random effects model 0.6079 [0.4228; 0.7664]
##
## Quantifying heterogeneity:
## tau^2 = 1.5480 [0.6402; 4.5209]; tau = 1.2442 [0.8001; 2.1262]
## I^2 = 97.1% [96.1%; 97.9%]; H = 5.91 [5.05; 6.92]
##
## Test of heterogeneity:
## Q d.f. p-value
## 349.73 10 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis2, layout = "RevMan")
Meta-analysis of proportions of symptomatic cavernous aneurysm
——————————- Metaregression of the proportion of Symptomatic and size —————————
m.two.reg <- metareg(meta_analysis2, ~Size)## Warning: 6 studies with NAs omitted from model fitting.m.two.reg##
## Mixed-Effects Model (k = 5; tau^2 estimator: DL)
##
## tau^2 (estimated amount of residual heterogeneity): 2.2064 (SE = 1.8968)
## tau (square root of estimated tau^2 value): 1.4854
## I^2 (residual heterogeneity / unaccounted variability): 98.20%
## H^2 (unaccounted variability / sampling variability): 55.50
## R^2 (amount of heterogeneity accounted for): 0.00%
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 166.4962, p-val < .0001
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.7100, p-val = 0.3994
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -1.1687 1.7300 -0.6755 0.4993 -4.5593 2.2220
## Size 0.0912 0.1082 0.8426 0.3994 -0.1209 0.3032
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1bubble(m.two.reg, studlab = FALSE)
——————————– Meta-analysis of proportions of Diplopia ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
diplopia <- tibble::tibble(
Study_ID = c("Dacus, 2017", "Date 1998", "Goldenberg-Cohen 2003", "Junior 2014", "Junior 2019", "Linskey 1990", "Stiebel-Kalish 2005", "Vercelli 2019, a"),
Sample = c(40, 29, 31, 100, 201, 37, 206, 194),
n_events = c(30, 6, 19, 27, 50, 6, 134, 25)
)
# Display the dataframe
print(diplopia)## # A tibble: 8 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Dacus, 2017 40 30
## 2 Date 1998 29 6
## 3 Goldenberg-Cohen 2003 31 19
## 4 Junior 2014 100 27
## 5 Junior 2019 201 50
## 6 Linskey 1990 37 6
## 7 Stiebel-Kalish 2005 206 134
## 8 Vercelli 2019, a 194 25# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis3 <- metaprop(event = n_events, n = Sample, data = diplopia,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis3)## proportion 95%-CI %W(random)
## Dacus, 2017 0.7500 [0.5880; 0.8731] 12.2
## Date 1998 0.2069 [0.0799; 0.3972] 11.5
## Goldenberg-Cohen 2003 0.6129 [0.4219; 0.7815] 12.2
## Junior 2014 0.2700 [0.1861; 0.3680] 13.0
## Junior 2019 0.2488 [0.1906; 0.3145] 13.2
## Linskey 1990 0.1622 [0.0619; 0.3201] 11.6
## Stiebel-Kalish 2005 0.6505 [0.5811; 0.7154] 13.3
## Vercelli 2019, a 0.1289 [0.0852; 0.1843] 13.0
##
## Number of studies: k = 8
## Number of observations: o = 838
## Number of events: e = 297
##
## proportion 95%-CI
## Random effects model 0.3549 [0.1974; 0.5516]
##
## Quantifying heterogeneity:
## tau^2 = 1.2495 [0.4666; 5.5738]; tau = 1.1178 [0.6831; 2.3609]
## I^2 = 95.5% [93.1%; 97.1%]; H = 4.73 [3.81; 5.87]
##
## Test of heterogeneity:
## Q d.f. p-value
## 156.69 7 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis3, layout = "RevMan")
Meta-analysis of proportions of Diplopia
——————————– Meta-analysis of proportions of Visual changes ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
visual_changes <- tibble::tibble(
Study_ID = c("Dacus, 2017", "Date, 1998", "Nurminen 2014", "Vercelli 2019"),
Sample = c(40, 29, 21, 194),
n_events = c(7, 6, 2, 5)
)
# Display the dataframe
print(visual_changes)## # A tibble: 4 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Dacus, 2017 40 7
## 2 Date, 1998 29 6
## 3 Nurminen 2014 21 2
## 4 Vercelli 2019 194 5# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis4 <- metaprop(event = n_events, n = Sample, data = visual_changes,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis4)## proportion 95%-CI %W(random)
## Dacus, 2017 0.1750 [0.0734; 0.3278] 27.0
## Date, 1998 0.2069 [0.0799; 0.3972] 26.2
## Nurminen 2014 0.0952 [0.0117; 0.3038] 20.6
## Vercelli 2019 0.0258 [0.0084; 0.0591] 26.3
##
## Number of studies: k = 4
## Number of observations: o = 284
## Number of events: e = 20
##
## proportion 95%-CI
## Random effects model 0.1009 [0.0353; 0.2561]
##
## Quantifying heterogeneity:
## tau^2 = 1.0401 [0.1166; 16.4914]; tau = 1.0198 [0.3414; 4.0610]
## I^2 = 81.1% [50.7%; 92.8%]; H = 2.30 [1.42; 3.72]
##
## Test of heterogeneity:
## Q d.f. p-value
## 15.89 3 0.0012
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis4, layout = "RevMan")
Meta-analysis of proportions of Visual changes
——————————– Meta-analysis of proportions of Ptosis ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
ptosis <- tibble::tibble(
Study_ID = c("Dacus, 2017", "Goldenberg-Cohen 2003", "Vercelli 2019"),
Sample = c(40, 31, 194),
n_events = c(5, 10, 2)
)
# Display the dataframe
print(ptosis)## # A tibble: 3 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Dacus, 2017 40 5
## 2 Goldenberg-Cohen 2003 31 10
## 3 Vercelli 2019 194 2# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis5 <- metaprop(event = n_events, n = Sample, data = ptosis,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis5)## proportion 95%-CI %W(random)
## Dacus, 2017 0.1250 [0.0419; 0.2680] 34.0
## Goldenberg-Cohen 2003 0.3226 [0.1668; 0.5137] 35.0
## Vercelli 2019 0.0103 [0.0013; 0.0367] 31.0
##
## Number of studies: k = 3
## Number of observations: o = 265
## Number of events: e = 17
##
## proportion 95%-CI
## Random effects model 0.0882 [0.0138; 0.4004]
##
## Quantifying heterogeneity:
## tau^2 = 2.6302 [0.6418; >100.0000]; tau = 1.6218 [0.8011; >10.0000]
## I^2 = 91.2% [77.2%; 96.6%]; H = 3.37 [2.09; 5.42]
##
## Test of heterogeneity:
## Q d.f. p-value
## 22.72 2 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis5, layout = "RevMan")
Meta-analysis of proportions of Ptosis
——————————– Meta-analysis of proportions of Facial numbness ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
facial_numb <- tibble::tibble(
Study_ID = c("Dacus 2017", "Vercelli 2019"),
Sample = c(40, 194),
n_events = c(2, 1)
)
# Display the dataframe
print(facial_numb)## # A tibble: 2 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Dacus 2017 40 2
## 2 Vercelli 2019 194 1# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis6 <- metaprop(event = n_events, n = Sample, data = facial_numb,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis6)## proportion 95%-CI %W(random)
## Dacus 2017 0.0500 [0.0061; 0.1692] 54.5
## Vercelli 2019 0.0052 [0.0001; 0.0284] 45.5
##
## Number of studies: k = 2
## Number of observations: o = 234
## Number of events: e = 3
##
## proportion 95%-CI
## Random effects model 0.0180 [0.0019; 0.1496]
##
## Quantifying heterogeneity:
## tau^2 = 1.9214; tau = 1.3861; I^2 = 71.5% [0.0%; 93.6%]; H = 1.87 [1.00; 3.95]
##
## Test of heterogeneity:
## Q d.f. p-value
## 3.51 1 0.0610
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis6, layout = "RevMan")
Meta-analysis of proportions of Facial numbness
——————————– Meta-analysis of proportions of Headache ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
headache <- tibble::tibble(
Study_ID = c("Fehrenbach 2022", "Junior 2019", "Nurminen 2014", "Perez de Vasconcellos 2009"),
Sample = c(64, 201, 21, 35),
n_events = c(16, 15, 1, 35)
)
# Display the dataframe
print(headache)## # A tibble: 4 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Fehrenbach 2022 64 16
## 2 Junior 2019 201 15
## 3 Nurminen 2014 21 1
## 4 Perez de Vasconcellos 2009 35 35# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis7 <- metaprop(event = n_events, n = Sample, data = headache,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis7)## proportion 95%-CI %W(random)
## Fehrenbach 2022 0.2500 [0.1502; 0.3740] 31.3
## Junior 2019 0.0746 [0.0424; 0.1201] 31.4
## Nurminen 2014 0.0476 [0.0012; 0.2382] 21.3
## Perez de Vasconcellos 2009 1.0000 [0.9000; 1.0000] 16.1
##
## Number of studies: k = 4
## Number of observations: o = 321
## Number of events: e = 67
##
## proportion 95%-CI
## Random effects model 0.2523 [0.0655; 0.6189]
##
## Quantifying heterogeneity:
## tau^2 = 1.9738 [1.1508; 85.8943]; tau = 1.4049 [1.0728; 9.2679]
## I^2 = 90.8% [79.5%; 95.9%]; H = 3.29 [2.21; 4.92]
##
## Test of heterogeneity:
## Q d.f. p-value
## 32.53 3 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies
## - Continuity correction of 0.5 in studies with zero cell frequencies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis7, layout = "RevMan")
Meta-analysis of proportions of Headache
——————————– Meta-analysis of proportions of Cranial Nerve palsy ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
nerve_palsy <- tibble::tibble(
Study_ID = c("Fehrenbach 2022", "Nurminen 2014", "van Rooji 2012"),
Sample = c(64, 21, 85),
n_events = c(14, 14, 56)
)
# Display the dataframe
print(nerve_palsy)## # A tibble: 3 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Fehrenbach 2022 64 14
## 2 Nurminen 2014 21 14
## 3 van Rooji 2012 85 56# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis8 <- metaprop(event = n_events, n = Sample, data = nerve_palsy,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis8)## proportion 95%-CI %W(random)
## Fehrenbach 2022 0.2188 [0.1251; 0.3397] 33.9
## Nurminen 2014 0.6667 [0.4303; 0.8541] 31.2
## van Rooji 2012 0.6588 [0.5480; 0.7582] 34.9
##
## Number of studies: k = 3
## Number of observations: o = 170
## Number of events: e = 84
##
## proportion 95%-CI
## Random effects model 0.5035 [0.2088; 0.7958]
##
## Quantifying heterogeneity:
## tau^2 = 1.2991 [0.2459; 52.9024]; tau = 1.1398 [0.4958; 7.2734]
## I^2 = 92.9% [82.5%; 97.1%]; H = 3.75 [2.39; 5.88]
##
## Test of heterogeneity:
## Q d.f. p-value
## 28.14 2 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis8, layout = "RevMan")
Meta-analysis of proportions of Cranial Nerve palsy
——————————- Meta-analysis of proportions of Dizziness ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
dizziness <- tibble::tibble(
Study_ID = c("Fehrenbach 2022", "Nurminen 2014"),
Sample = c(64, 21),
n_events = c(8, 3)
)
# Display the dataframe
print(dizziness)## # A tibble: 2 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Fehrenbach 2022 64 8
## 2 Nurminen 2014 21 3# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis9 <- metaprop(event = n_events, n = Sample, data = dizziness,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis9)## proportion 95%-CI %W(random)
## Fehrenbach 2022 0.1250 [0.0555; 0.2315] 73.1
## Nurminen 2014 0.1429 [0.0305; 0.3634] 26.9
##
## Number of studies: k = 2
## Number of observations: o = 85
## Number of events: e = 11
##
## proportion 95%-CI
## Random effects model 0.1296 [0.0732; 0.2191]
##
## Quantifying heterogeneity:
## tau^2 = 0; tau = 0; I^2 = 0.0%; H = 1.00
##
## Test of heterogeneity:
## Q d.f. p-value
## 0.04 1 0.8326
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis9, layout = "RevMan")
Meta-analysis of proportions of Dizziness
——————————- Meta-analysis of proportions of Trigeminal pain ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
trigeminal_pain <- tibble::tibble(
Study_ID = c("Goldenberg-Cohen et al., 2003", "Junior et al., 2014", "Junior et al., 2019", "Linskey et al., 1990", "Nurminen et al., 2014", "Vercelli et al., 2019"),
Sample = c(31, 100, 201, 37, 21, 194),
n_events = c(3, 8, 9, 1, 3, 21)
)
# Display the dataframe
print(trigeminal_pain)## # A tibble: 6 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Goldenberg-Cohen et al., 2003 31 3
## 2 Junior et al., 2014 100 8
## 3 Junior et al., 2019 201 9
## 4 Linskey et al., 1990 37 1
## 5 Nurminen et al., 2014 21 3
## 6 Vercelli et al., 2019 194 21# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis10 <- metaprop(event = n_events, n = Sample, data = trigeminal_pain,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis10)## proportion 95%-CI %W(random)
## Goldenberg-Cohen et al., 2003 0.0968 [0.0204; 0.2575] 10.3
## Junior et al., 2014 0.0800 [0.0352; 0.1516] 20.7
## Junior et al., 2019 0.0448 [0.0207; 0.0833] 22.6
## Linskey et al., 1990 0.0270 [0.0007; 0.1416] 4.3
## Nurminen et al., 2014 0.1429 [0.0305; 0.3634] 9.9
## Vercelli et al., 2019 0.1082 [0.0683; 0.1607] 32.2
##
## Number of studies: k = 6
## Number of observations: o = 584
## Number of events: e = 45
##
## proportion 95%-CI
## Random effects model 0.0804 [0.0539; 0.1182]
##
## Quantifying heterogeneity:
## tau^2 = 0.0947 [0.0000; 1.7727]; tau = 0.3078 [0.0000; 1.3314]
## I^2 = 35.4% [0.0%; 74.2%]; H = 1.24 [1.00; 1.97]
##
## Test of heterogeneity:
## Q d.f. p-value
## 7.74 5 0.1714
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis10, layout = "RevMan")
Meta-analysis of proportions of Trigeminal pain
——————————- Meta-analysis of proportions of Retro-orbital pain, eye pain ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
eye_pain <- tibble::tibble(
Study_ID = c("Kupersmith 1992", "Linskey 1990", "Nurminen 2014", "Perez de Vasconcellos 2009", "Stiebel-Kalish 2005"),
Sample = c(70, 37, 21, 35, 206),
n_events = c(3, 6, 3, 35, 122)
)
# Display the dataframe
print(eye_pain)## # A tibble: 5 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Kupersmith 1992 70 3
## 2 Linskey 1990 37 6
## 3 Nurminen 2014 21 3
## 4 Perez de Vasconcellos 2009 35 35
## 5 Stiebel-Kalish 2005 206 122# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis11 <- metaprop(event = n_events, n = Sample, data = eye_pain,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis11)## proportion 95%-CI %W(random)
## Kupersmith 1992 0.0429 [0.0089; 0.1202] 20.8
## Linskey 1990 0.1622 [0.0619; 0.3201] 21.7
## Nurminen 2014 0.1429 [0.0305; 0.3634] 20.5
## Perez de Vasconcellos 2009 1.0000 [0.9000; 1.0000] 14.1
## Stiebel-Kalish 2005 0.5922 [0.5218; 0.6600] 22.9
##
## Number of studies: k = 5
## Number of observations: o = 369
## Number of events: e = 169
##
## proportion 95%-CI
## Random effects model 0.3354 [0.0860; 0.7301]
##
## Quantifying heterogeneity:
## tau^2 = 3.1836 [1.1642; 41.9939]; tau = 1.7843 [1.0790; 6.4803]
## I^2 = 93.9% [88.6%; 96.7%]; H = 4.04 [2.96; 5.52]
##
## Test of heterogeneity:
## Q d.f. p-value
## 65.38 4 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies
## - Continuity correction of 0.5 in studies with zero cell frequencies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis11, layout = "RevMan")
Meta-analysis of proportions of Retro-orbital pain, eye pain
——————————- Meta-analysis of proportions of Fistulas ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
CCF <- tibble::tibble(
Study_ID = c("Kupersmith, 1992", "Kupersmith, 2002", "van Rooji et al., 2012"),
Sample = c(70, 174, 85),
n_events = c(5, 13, 8)
)
# Display the dataframe
print(CCF)## # A tibble: 3 × 3
## Study_ID Sample n_events
## <chr> <dbl> <dbl>
## 1 Kupersmith, 1992 70 5
## 2 Kupersmith, 2002 174 13
## 3 van Rooji et al., 2012 85 8# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis12 <- metaprop(event = n_events, n = Sample, data = CCF,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis12)## proportion 95%-CI %W(random)
## Kupersmith, 1992 0.0714 [0.0236; 0.1589] 19.4
## Kupersmith, 2002 0.0747 [0.0404; 0.1244] 50.3
## van Rooji et al., 2012 0.0941 [0.0415; 0.1771] 30.3
##
## Number of studies: k = 3
## Number of observations: o = 329
## Number of events: e = 26
##
## proportion 95%-CI
## Random effects model 0.0795 [0.0547; 0.1142]
##
## Quantifying heterogeneity:
## tau^2 = 0 [0.0000; 0.8558]; tau = 0 [0.0000; 0.9251]
## I^2 = 0.0% [0.0%; 89.6%]; H = 1.00 [1.00; 3.10]
##
## Test of heterogeneity:
## Q d.f. p-value
## 0.36 2 0.8335
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis12, layout = "RevMan")
Meta-analysis of proportions of Fistulas
——————————- Meta-analysis of proportions of morbidity ———————————–
# Load necessary library
library(tibble)
# Create the dataframe
morbidity <- tibble::tibble(
Study_ID = c("Fehrenbach, 2022", "Junior, 2014", "Kupersmith, 1992", "Kupersmith, 2002", "Lievo, 2023", "Linskey, 1990", "Stiebel-Kalish, 2005"),
Sample = c(64, 100, 70, 174, 250, 37, 206),
morbidity = c(25, 36, 31, 3, 26, 14, 74)
)
# Display the dataframe
print(morbidity)## # A tibble: 7 × 3
## Study_ID Sample morbidity
## <chr> <dbl> <dbl>
## 1 Fehrenbach, 2022 64 25
## 2 Junior, 2014 100 36
## 3 Kupersmith, 1992 70 31
## 4 Kupersmith, 2002 174 3
## 5 Lievo, 2023 250 26
## 6 Linskey, 1990 37 14
## 7 Stiebel-Kalish, 2005 206 74# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis13 <- metaprop(event = morbidity, n = Sample, data = morbidity,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis13)## proportion 95%-CI %W(random)
## Fehrenbach, 2022 0.3906 [0.2710; 0.5207] 14.7
## Junior, 2014 0.3600 [0.2664; 0.4621] 15.1
## Kupersmith, 1992 0.4429 [0.3241; 0.5666] 14.8
## Kupersmith, 2002 0.0172 [0.0036; 0.0496] 10.9
## Lievo, 2023 0.1040 [0.0691; 0.1487] 15.1
## Linskey, 1990 0.3784 [0.2246; 0.5524] 13.8
## Stiebel-Kalish, 2005 0.3592 [0.2937; 0.4288] 15.6
##
## Number of studies: k = 7
## Number of observations: o = 901
## Number of events: e = 209
##
## proportion 95%-CI
## Random effects model 0.2477 [0.1445; 0.3908]
##
## Quantifying heterogeneity:
## tau^2 = 0.7233 [0.4070; 5.9167]; tau = 0.8505 [0.6379; 2.4324]
## I^2 = 93.0% [88.0%; 95.9%]; H = 3.77 [2.88; 4.92]
##
## Test of heterogeneity:
## Q d.f. p-value
## 85.11 6 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis13, layout = "RevMan")
Meta-analysis of proportions of morbidity
———————————- RUPTURE ——————————————–
# Load necessary library
library(tibble)
# Create the dataframe
rupture <- tibble::tibble(
Study_ID = c("Dacus, 2017", "Fehrenbach, 2022", "Goldenberg-Cohen, 2003", "Junior, 2014", "Kupersmith, 1992", "Kupersmith, 2002", "Lievo, 2023", "Linskey, 1990", "Nurminen, 2014", "Penchet, 2015", "Stiebel-Kalish, 2005", "van Rooji, 2012", "Vercelli, 2019", "Weir, 2002"),
Patients = c(40, 64, 31, 100, 70, 174, 250, 37, 21, 27, 206, 85, 194, 18),
Rupture = c(2, 1, 0, 0, 6, 14, 12, 0, 9, 1, 12, 9, 1, 0),
Size = c(NA, 6.6, NA, NA, NA, 19.6, 16.2, NA, 32, 28, NA, 22.71, 8.6, 11.2)
)
# Display the dataframe
print(rupture)## # A tibble: 14 × 4
## Study_ID Patients Rupture Size
## <chr> <dbl> <dbl> <dbl>
## 1 Dacus, 2017 40 2 NA
## 2 Fehrenbach, 2022 64 1 6.6
## 3 Goldenberg-Cohen, 2003 31 0 NA
## 4 Junior, 2014 100 0 NA
## 5 Kupersmith, 1992 70 6 NA
## 6 Kupersmith, 2002 174 14 19.6
## 7 Lievo, 2023 250 12 16.2
## 8 Linskey, 1990 37 0 NA
## 9 Nurminen, 2014 21 9 32
## 10 Penchet, 2015 27 1 28
## 11 Stiebel-Kalish, 2005 206 12 NA
## 12 van Rooji, 2012 85 9 22.7
## 13 Vercelli, 2019 194 1 8.6
## 14 Weir, 2002 18 0 11.2——————————- Meta-analysis of proportions of rupture among cavernous aneurysm —————————
# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis14 <- metaprop(event = Rupture, n = Patients, data = rupture,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis14)## proportion 95%-CI %W(random)
## Dacus, 2017 0.0500 [0.0061; 0.1692] 7.2
## Fehrenbach, 2022 0.0156 [0.0004; 0.0840] 5.1
## Goldenberg-Cohen, 2003 0.0000 [0.0000; 0.1122] 3.2
## Junior, 2014 0.0000 [0.0000; 0.0362] 3.2
## Kupersmith, 1992 0.0857 [0.0321; 0.1773] 10.0
## Kupersmith, 2002 0.0805 [0.0447; 0.1313] 11.4
## Lievo, 2023 0.0480 [0.0250; 0.0823] 11.3
## Linskey, 1990 0.0000 [0.0000; 0.0949] 3.2
## Nurminen, 2014 0.4286 [0.2182; 0.6598] 9.9
## Penchet, 2015 0.0370 [0.0009; 0.1897] 5.0
## Stiebel-Kalish, 2005 0.0583 [0.0305; 0.0995] 11.3
## van Rooji, 2012 0.1059 [0.0496; 0.1915] 10.8
## Vercelli, 2019 0.0052 [0.0001; 0.0284] 5.1
## Weir, 2002 0.0000 [0.0000; 0.1853] 3.2
##
## Number of studies: k = 14
## Number of observations: o = 1317
## Number of events: e = 67
##
## proportion 95%-CI
## Random effects model 0.0552 [0.0317; 0.0942]
##
## Quantifying heterogeneity:
## tau^2 = 0.6835 [0.2631; 3.5286]; tau = 0.8267 [0.5129; 1.8785]
## I^2 = 73.2% [54.4%; 84.3%]; H = 1.93 [1.48; 2.52]
##
## Test of heterogeneity:
## Q d.f. p-value
## 48.57 13 < 0.0001
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies
## - Continuity correction of 0.5 in studies with zero cell frequencies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis14, layout = "RevMan")
Meta-analysis of proportions of rupture among cavernous aneurysm
——————————- Metaregression of the proportion of rupture and size —————————
m.tres.reg <- metareg(meta_analysis14, ~Size)## Warning: 6 studies with NAs omitted from model fitting.m.tres.reg##
## Mixed-Effects Model (k = 8; tau^2 estimator: DL)
##
## tau^2 (estimated amount of residual heterogeneity): 0 (SE = 0.1308)
## tau (square root of estimated tau^2 value): 0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability): 1.00
## R^2 (amount of heterogeneity accounted for): 100.00%
##
## Test for Residual Heterogeneity:
## QE(df = 6) = 5.9400, p-val = 0.4299
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 34.1025, p-val < .0001
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -5.6155 0.5718 -9.8208 <.0001 -6.7362 -4.4948 ***
## Size 0.1579 0.0270 5.8397 <.0001 0.1049 0.2109 ***
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1bubble(m.tres.reg, studlab = FALSE)
———————————- SAH ——————————————–
# Load necessary library
library(tibble)
# Create the dataframe
sah <- tibble::tibble(
Study_ID = c("Dacus, 2017", "Fehrenbach, 2022", "Junior, 2014", "Kupersmith, 1992", "Kupersmith, 2002", "Lievo, 2023", "Linskey, 1990", "Stiebel-Kalish, 2005", "van Rooji, 2012", "Vercelli, 2019"),
Patients = c(40, 64, 100, 70, 174, 250, 37, 206, 85, 194),
SAH = c(2, 1, 0, 1, 1, 0, 0, 0, 1, 0),
Size = c(NA, 6.6, NA, NA, 19.6, 16.2, NA, NA, 22.71, 8.6)
)
# Display the dataframe
print(sah)## # A tibble: 10 × 4
## Study_ID Patients SAH Size
## <chr> <dbl> <dbl> <dbl>
## 1 Dacus, 2017 40 2 NA
## 2 Fehrenbach, 2022 64 1 6.6
## 3 Junior, 2014 100 0 NA
## 4 Kupersmith, 1992 70 1 NA
## 5 Kupersmith, 2002 174 1 19.6
## 6 Lievo, 2023 250 0 16.2
## 7 Linskey, 1990 37 0 NA
## 8 Stiebel-Kalish, 2005 206 0 NA
## 9 van Rooji, 2012 85 1 22.7
## 10 Vercelli, 2019 194 0 8.6——————————- Meta-analysis of proportions of SAH among cavernous aneurysm —————————
# Install the meta package if not already installed
if (!require(meta)) {
install.packages("meta")
library(meta)
}
# Assuming 'df' is your DataFrame loaded as previously shown
# Meta-analysis of proportions with study labels
# Calculate the pooled proportion using metaprop function, labeling each study by the First Author
meta_analysis15 <- metaprop(event = SAH, n = Patients, data = sah,
sm = "PLO", method.tau = "DL",
prediction = FALSE, comb.fixed = FALSE,
comb.random = TRUE, studlab = Study_ID
)
# Summary of the meta-analysis
summary(meta_analysis15)## proportion 95%-CI %W(random)
## Dacus, 2017 0.0500 [0.0061; 0.1692] 22.2
## Fehrenbach, 2022 0.0156 [0.0004; 0.0840] 11.8
## Junior, 2014 0.0000 [0.0000; 0.0362] 6.1
## Kupersmith, 1992 0.0143 [0.0004; 0.0770] 11.8
## Kupersmith, 2002 0.0057 [0.0001; 0.0316] 11.9
## Lievo, 2023 0.0000 [0.0000; 0.0146] 6.1
## Linskey, 1990 0.0000 [0.0000; 0.0949] 6.0
## Stiebel-Kalish, 2005 0.0000 [0.0000; 0.0177] 6.1
## van Rooji, 2012 0.0118 [0.0003; 0.0638] 11.9
## Vercelli, 2019 0.0000 [0.0000; 0.0188] 6.1
##
## Number of studies: k = 10
## Number of observations: o = 1220
## Number of events: e = 6
##
## proportion 95%-CI
## Random effects model 0.0113 [0.0057; 0.0222]
##
## Quantifying heterogeneity:
## tau^2 = 0.0332 [0.0000; 2.6024]; tau = 0.1823 [0.0000; 1.6132]
## I^2 = 2.6% [0.0%; 63.4%]; H = 1.01 [1.00; 1.65]
##
## Test of heterogeneity:
## Q d.f. p-value
## 9.24 9 0.4152
##
## Details on meta-analytical method:
## - Inverse variance method
## - DerSimonian-Laird estimator for tau^2
## - Jackson method for confidence interval of tau^2 and tau
## - Logit transformation
## - Clopper-Pearson confidence interval for individual studies
## - Continuity correction of 0.5 in studies with zero cell frequencies# To visualize the results, you can plot a forest plot
meta::forest(meta_analysis15, layout = "RevMan")
Meta-analysis of proportions of SAH among cavernous aneurysm
——————————- Metaregression of the proportion of SAH and size —————————
m.four.reg <- metareg(meta_analysis15, ~Size)## Warning: 5 studies with NAs omitted from model fitting.m.four.reg##
## Mixed-Effects Model (k = 5; tau^2 estimator: DL)
##
## tau^2 (estimated amount of residual heterogeneity): 0 (SE = 1.1246)
## tau (square root of estimated tau^2 value): 0
## I^2 (residual heterogeneity / unaccounted variability): 0.00%
## H^2 (unaccounted variability / sampling variability): 1.00
## R^2 (amount of heterogeneity accounted for): 0.00%
##
## Test for Residual Heterogeneity:
## QE(df = 3) = 2.2531, p-val = 0.5216
##
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0079, p-val = 0.9292
##
## Model Results:
##
## estimate se zval pval ci.lb ci.ub
## intrcpt -4.8536 1.2791 -3.7945 0.0001 -7.3607 -2.3466 ***
## Size -0.0068 0.0767 -0.0888 0.9292 -0.1572 0.1435
##
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1bubble(m.four.reg, studlab = FALSE)