Article Figures & Data
- Figure 1
Distribution of survey respondents by country, region, occupation and by hospital context. AR, aspect ratio; AUC, area under the curve; PC, posterior circulation; SAH, subarachnoid haemorrhage.
- Figure 2
Calibration curves of the CMB-MIAs nomogram in the derivation (A) and validation (B) set. The predicted probabilities and the actual observed probability of aneurysm rupture were divided into ten groups to create a bar chart for in the derivation (C) and validation (D) set. CMB, clinical and morphological; MIA, multiple intracranial aneurysms.
- Figure 3
Decision curve analysis of CMB-MIAs nomogram in derivation and validation cohort.
- Table 1
Univariate analysis between unruptured and ruptured groups in the derivation and validation set
Characteristic Derivation cohort
(n=1171)Validation cohort
(n=500)Unruptured
N (%)Ruptured
N (%)P value Unruptured
N (%)Ruptured
N (%)P value No of aneurysm 1067 (91.1) 104 (8.9) 446 (89.2) 54 (10.8) Female sex 734 (68.8) 79 (76.0) 0.130 294 (65.9) 40 (74.1) 0.229 Age (years) <50 214 (20.1) 23 (22.1) 0.178 80 (17.9) 12 (22.2) 0.205 50–70 728 (68.2) 63 (60.6) 307 (68.8) 31 (57.4) ≥70 125 (11.7) 18 (17.3) 59 (13.2) 11 (20.4) History of SAH 56 (5.2) 23 (22.1) <0.001 22 (4.9) 12 (22.2) <0.001 No of aneurysms 2 636 (59.6) 65 (62.5) 0.830 271 (60.8) 37 (68.5) 0.442 3–4 382 (35.8) 35 (33.7) 155 (34.8) 16 (29.6) >4 49 (4.6) 4 (3.8) 20 (4.4) 1 (1.9) Comorbidities Hypertension 562 (52.7) 62 (59.6) 0.175 234 (52.5) 35 (64.8) 0.086 Diabetes 128 (12.0) 8 (7.7) 0.191 42 (9.4) 2 (3.7) 0.207 Hypercholesterolaemia 111 (10.4) 10 (9.6) 0.801 50 (11.2) 7 (13.0) 0.702 Heart diseases 113 (10.6) 5 (4.8) 0.061 36 (8.1) 6 (11.1) 0.436 History of stroke 123 (11.5) 8 (7.7) 0.236 53 (11.9) 7 (13.0) 0.818 Smoking 199 (18.7) 20 (19.2) 0.885 84 (18.8) 15 (27.8) 0.119 Drinking 141 (13.2) 18 (17.3) 0.145 70 (15.7) 14 (25.9) 0.058 Morphological characterises Size, mm <3 291 (27.3) 14 (13.5) 0.012 127 (28.5) 135 (27.0) 0.102 3–7 573 (53.7) 65 (62.5) 235 (52.7) 34 (63.0) 7–15 170 (15.9) 23 (22.1) 84 (18.8) 12 (22.2) >15 33 (3.1) 2 (1.9) Neck >3.9 379 (35.5) 41 (39.4) 0.428 168 (37.7) 21(389) 0.861 AR >1.5 196 (18.4) 41 (39.4) <0.001 95 (21.3) 23 (46.2) 0.001 Branching to parent ratio >1.2 187 (17.5) 33 (31.7) <0.001 86 (19.3) 17 (31.5) 0.036 Neck to parent ratio >1.2 325 (30.5) 41 (39.4) 0.060 138 (30.9) 14 (25.9) 0.449 SR >1.7 204 (19.1) 42 (40.4) <0.001 95 (21.3) 21 (38.9) 0.004 Location of PC 109 (10.2) 23 (22.1) <0.001 42 (9.4) 10 (18.5) 0.039 Irregular shape 357 (33.5) 53 (51.0) <0.001 155 (34.8) 35 (64.8) <0.001 Bifurcation location 236 (22.1) 41 (39.4) <0.001 109 (24.4) 24 (44.4) 0.002 Inflow angle, mean (SD) 96 (35) 103 (33) 0.055 99 (31) 104 (31) 0.351 Outflow angle, mean (SD) 102 (31) 102 (29) 0.866 100 (30) 106 (29) 0.172 Branching angle, mean (SD) 137 (34) 128 (36) 0.021 137 (35) 134 (33) 0.045 AR, aspect ratio; PC, posterior circulation; SAH, subarachnoid haemorrhage; SR, size ratio.
- Table 2
Multivariate analysis of all 1671 multiple aneurysms in 700 patients and 400 multiple aneurysms in 158 patients who presented with SAH
Model 1* OR (95% CI) P value VIF Model 2* OR (95% CI) P value VIF SAH history 5.094 (2.848 to 8.929) <0.001 1.026 AR >1.55 2.449 (1.302 to 4.641) 0.006 1.218 Irregular shape 1.883 (1.225 to 2.897) 0.004 1.025 Irregular shape 1.886 (1.070 to 3.336) 0.028 1.120 Alcohol use 2.022 (1.016 to 3.920) 0.040 1.400 Neck >4 mm 1.994 (1.015 to 3.944) 0.046 1.307 Female 1.852 (1.053 to 3.395 0.038 1.419 PC 2.522 (1.156 to 5.667) 0.022 1.007 AR >1.5 2.371 (1.507 to 3.698) <0.001 1.021 Size Bifurcation location 1.762 (1.118 to 2.745) 0.013 1.026 <3 mm Ref Ref PC 2.722 (1.561 to 4.460) <0.001 1.028 3 to 7 mm 1.494 (0.738 to 3.093) 0.270 1.617 >7 mm 3.240 (1.090 to 10.114) 0.038 1.803 *Model one was developed by all 1671 multiple aneurysms in patients with or without SAH; model 2 was developed by 400 multiple aneurysms in 158 patients with SAH.
AR, aspect ratio; PC, posterior circulation; SAH, subarachnoid haemorrhage; VIF, variation inflation factors.
- Table 3
Evaluation of discrimination and calibration abilities of the models
Model 1* Model 2* Derivation Group Validation Group Total group (1671 IAs) Derivation Group Validation Group Total group (400 IAs) Overall performance Brier score 0.073 0.085 0.083 0.199 0.207 0.202 Discrimination AUC (95%) 0.743 (0.690 to 0.796) 0.772 (0.686 to 0.858) 0.753 (0.712 to 0.794) 0.739 (0.677 to 0.801) 0.717 (0.632 to 0.802) 0.730 (0.683 to 0.782) Calibration, goodness of fit P value of Hosmer-Lemeshow test 0.9146 0.4654 0.983 0.893 0.269 0.3663 *Model 1 was developed by all 1671 multiple aneurysms in patients with or without SAH; model 2 was developed by 400 multiple aneurysms in 158 patients with SAH.
AUC, areas under curve; IA, intracranial aneurysm; SAH, subarachnoid haemorrhage.
Supplementary Materials
Supplementary data
Additional Files
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
