Results
While there was a significant difference between the groups’ ages (p=0.018), the MANCOVA analysis showed that the observed shape differences were due to the groups (p<0.001) and not due to age (p=0.711). Among parous women, the median parity was 2 and median time since last delivery was 2 years. Overall, it was observed, both visually and quantitatively, that the sacrum and coccyx were straighter in pregnant women (Table 1). This was reflected by significant differences between the three groups in the coccygeal curvature index (p=0.001), sacrococcygeal curvature index (p=0.002), and sacrococcygeal angle (p=0.016) (Table 1). All other measures failed to achieve significance.
Gravidity resulted in a 13.3% increase in the coccygeal curvature index (89.2 ± 10.0) compared to nulliparous women (78.7 ± 6.6, p=0.003). Parous women (80.0 ± 5.5) demonstrated a near return to nulliparous values with a significant decrease of 10.3% relative to the pregnant group (p=0.004; Figure 2a) and no difference from the nulliparous group. The additional analysis of subgroups of pregnant women demonstrated an interesting trend reflecting increased straightening with increased parity (Figure 2d). However, a similar trend was not observed in the parous subgroups, suggesting that most of the observed changes are limited to pregnancy.
The sacrococcygeal curvature index, which includes the combined curvature of the sacrum and coccyx, also demonstrated a significantly straighter (8%) sacrococcygeal shape for pregnant (79.2 ± 3.7) relative to nulliparous women (73.3 ± 5.8, p=0.016). For this metric, the parous group straddled the pregnant and nulliparous groups but, again, was not significantly different from either (Figure 2b).
Not surprisingly, differences for the sacrococcygeal angle followed those of the sacrococcygeal curvature index with the angle for the nulliparous group at 92.8° ± 10.9° versus 109.3° ± 9.4° for the pregnant group (p=0.002) and the parous group straddling both but not different from the nulliparous group (Figure 2c). Both metrics exhibited a similar trend as that shown for the coccygeal curvature index when evaluating the subgroups within the pregnant and parous groups individually (Figure 2e/f). It should be noted that none of the measures describing the shape of the sacrum alone were found to be significant (Table 1). Thus, it is likely that the significant metrics for the sacrum-coccyx are being driven by the rotation/elongation of the coccyx.