Height loss, back pain and severe pain are highly predictive of pregnancy associated osteoporosis.

UNLABELLED

The diagnosis of pregnancy associated osteoporosis (PAO) is often delayed. We evaluated clinical predictors of PAO in 225 cases and 174 controls and found that the combination of height loss, back pain and severe pain was highly predictive of PAO, with a sensitivity and specificity close to 100%.

OBJECTIVES

To identify clinical predictors of pregnancy associated osteoporosis.

METHODS

We conducted an online survey of 225 women with a self-reported diagnosis of pregnancy associated osteoporosis (cases) and gathered details of their medical history, demographic characteristics and symptoms. Their responses were compared with those from a group of 174 controls who had pregnancies around the same time as the affected individuals. Using this information, we constructed a receiver operating characteristic (ROC) curve to identify features that best discriminated between cases and controls.

RESULTS

The mean ± SD age was 33.1 ± 4.9 for cases vs. 32.1 ± 4.2 years for controls. Thromboembolic disease, anticoagulant use and a family history of osteoporosis were more frequent in cases, whereas hypertension and pre-eclampsia/eclampsia were more frequent in controls. Severe pain was reported in 186 (84.2%) cases vs. 3 (1.8%) controls, height loss in 112 (49.8%) vs. 1 (0.6%) and back pain in 200 (88.9%) vs. 77 (44.3%) (p < 0.001 for all features). Severe pain had high sensitivity, 84.2% [95% CI, 78.5 to 88.6%], and specificity, 98.2% [94.6 to 99.5%], for PAO. Back pain, height loss and severe pain were highly predictive of PAO in ROC analysis, with an area under the curve [95% CI] of 0.988 [0.980 to 0.997].

CONCLUSIONS

Height loss, back pain and severity of pain are highly predictive of PAO. Height should be measured at the beginning and end of pregnancy, and pain location and severity assessed to help prioritise women likely to have PAO for skeletal imaging to confirm the diagnosis and commence treatment.