We present the results of the study performed using data obtained from 281 patients (156 male & 125 female) and provide recommendations regarding the safe sizes for the South African population as well techniques to accurately and safely measure the columns for individual outliers.
The measurements of the anterior and posterior columns varies as expected.
Gender differences are marked with male pelvis measurments being larger than female pelvis measurements.
This is however not a surprising finding.
The tables below (Table 1 & Table 2) show the exact means,
ranges and standard deviations for each measurement in males and females separately.
Of note is that the narrowest measurement of the pelvis in both males and females,
for safe screw placement,
is the anterior column height (Fig. 4).
This is in keeping with intra-operative experiences at our institution as well findings of previous studies4-7.
Thus the smallest indicator for a safe screw size is the anterior column height (oblique sagittal plane).
The mean measurement for the anterior column height in males is 3,872mm [3,684; 4,06 - 95% confidence intervals].
In females the mean measurement is 3,018mm [2,839; 3,2].
Further sub-group analysis is necessary to determine exact areas of concern regarding screw placement within the anterior column and to determine percentage occurences of the narrowest points within the anterior column height.
The posterior column is wider than the anterior column in all dimensions in both males and females.
The narrowest indicator in the posterior column is the transverse width (coronal plane in Fig. 6).
In males the mean measurement for the posterior column transverse width is 6,216mm [5,907; 6,524] with the mean in females being 5,114mm [4,846; 5,381].
There is a significant proprtion of patients having pelvic and acetabular measurements that fall outside the expected averages (outliers).
The described CT measurement technique provides a good technique to identify these patients.
However for the average patient screws with diameters of 3,5 mm (males) and 3,0mm screws (females) are safe for the pelvic anterior columns,
while in the posterior columns screws of diameters of 6mm (males) and 5mm (females) would be safe to use.
Table 1: Table indicating the results of the 281 patients on whom measurement of the anterior and posterior columns were obtained with breakdown into male and female. The femoral neck measurements were discontinued after 108 observation as no statistically significant relationship was evident by that stage of the study. (Abbreviations: ACL - Anterior column length; ACW - anterior column width; ACH - anterior column height; PCL - posterior column length; PCAPW - posterior column antero-posterior width; PCTW - posterior column transverse width; FN - femoral neck; FH - femoral head).
References: Anith Chacko, Department of Radiology, University of Pretoria - Pretoria/ ZA
Table 2: Table indicating the results of the 108 patients on whom measurement of the anterior and posterior columns as well as femoral neck and femoral head measurements were obtained with breakdown into male and female. The femoral neck measurements were discontinued after 108 observation as no statistically significant relationship was evident by that stage of the study. Very low correlation (highest of 27%) between femoral neck and various acetbular column measurements were obtained. The correlation between femoral head and acetabular column measurements were higher (highest of 40,1%) but still not of a clinically applicable level. With the addition of three variables into the regression analysis the correlation was slightly higher but again not clinically applicable. Indications are that the more measurements obtained around the acetabulum the higher correlation will be obtained, however this is again not clinically practical.
(Abbreviations: ACL - Anterior column length; ACW - anterior column width; ACH - anterior column height; PCL - posterior column length; PCAPW - posterior column antero-posterior width; PCTW - posterior column transverse width; FN - femoral neck; FH - femoral head).
References: Anith Chacko, Department of Radiology, University of Pretoria - Pretoria/ ZA
The secondary objective of describing a relationship between the femoral neck and femoral head measurements was not successful.
The correlation between femoral neck measurements and the various acetabular column measurements resulted in only very small percentage correlations - see Table 2.
This may be due to an actual pelvic dysplasia (similar to sacral dysplasia).
Further investigation into this will be necessary.