Prevalence and awareness of sacroiliac joint alterations on lumbar spine CT in low back pain patients younger than 40 years

Background Computed tomography (CT) examinations of the lumbar spine are commonly performed in patients aged ≤40 years due to low back pain (LBP). Purpose To investigate the prevalence and awareness of radiologists for the presence of structural post-inflammatory/other sacroiliac joint (SIJ) alterations on lumbar spine CTs of young patients with LBP. Material and Methods A total of 484 lumbar spine CT examinations (272 men, 212 women; average age, 31 years; age range, 18–40 years) of patients with LBP in which the entire SIJs were visualized were retrospectively reviewed. SIJs were scored (consensus) by two senior radiologists (study reading) for the presence of post-inflammatory structural SIJ findings or other SIJs alterations. The original reports were compared to the study reading. Fifty CT examinations were re-evaluated for reliability assessment (intra-class correlation coefficient [ICC]). Results A total of 150 (31%) abnormal SIJ examinations were registered (ICC: r = 0.7–0.8; P < 0.0001): suspected sacroiliitis = 50 (10.2%); definite sacroiliitis = 16 (3.3%); osteitis-condensans-ilii = 38 (7.8%); diffuse idiopathic skeletal hyperostosis = 24 (5%); degenerative changes = 22 (4.5%); accessory SIJ = 22 (4.5%); and tumor = 1. The SIJs were referenced 39 times (8.0%) in the original readings: pathological findings (n = 15); and normal SIJ (n = 24). Total diagnostic accuracy for these reports only and for the entire readings were 49% and 69%, respectively, and 13% and 1.3%, respectively, for the pathological findings. Conclusion Sacroiliitis and other SIJ alterations are prevalent in young individuals with LBP, albeit, the majority of these alterations are not recognized nor reported by senior radiologists thus may delay efficacious treatment.

Download Full-text

Automated detection of sacroiliitis on plain radiograph using EfficientDet algorithm in young patients with back pain: a pilot study

10.21203/rs.3.rs-272234/v1 ◽

2021 ◽

Author(s):

Sung Hyun Noh ◽

Chansik An ◽

Dain Kim ◽

Seung Hyun Lee ◽

Min-Yung Chang ◽

...

Keyword(s):

Low Back Pain ◽

Deep Learning ◽

Back Pain ◽

Object Detection ◽

Sacroiliac Joint ◽

Plain Radiograph ◽

Young Patients ◽

Training Dataset ◽

Detection Accuracy ◽

Low Back

Abstract Background A computer algorithm that automatically detects sacroiliac joint abnormalities on plain radiograph would help radiologists avoid missing sacroiliitis. This study aimed to develop and validate a deep learning model to detect and diagnose sacroiliitis on plain radiograph in young patients with low back pain. Methods This Institutional Review Board-approved retrospective study included 478 and 468 plain radiographs from 241 and 433 young (< 40 years) patients who complained of low back pain with and without ankylosing spondylitis, respectively. They were randomly split into training and test datasets with a ratio of 8:2. Radiologists reviewed the images and labeled the coordinates of a bounding box and determined the presence or absence of sacroiliitis for each sacroiliac joint. We fine-tined and optimized the EfficientDet-D4 object detection model pre-trained on the COCO 2107 dataset on the training dataset and validated the final model on the test dataset. Results The mean average precision, an evaluation metric for object detection accuracy, was 0.918 at 0.5 intersection over union. In the diagnosis of sacroiliitis, the area under the curve, sensitivity, specificity, accuracy, and F1-score were 0.932 (95% confidence interval, 0.903–0.961), 96.9% (92.9–99.0), 86.8% (81.5–90.9), 91.1% (87.7–93.7), and 90.2% (85.0–93.9), respectively. Conclusions The EfficientDet, a deep learning-based object detection algorithm, could be used to automatically diagnose sacroiliitis on plain radiograph.

Download Full-text

Are general practitioners referring patients with low back pain for CTs appropriately according to the guidelines: a retrospective review of 3609 medical records in Newfoundland using routinely collected data

BMC Family Practice ◽

10.1186/s12875-020-01308-5 ◽

2020 ◽

Vol 21 (1) ◽

Author(s):

Gabrielle S. Logan ◽

Russell Eric Dawe ◽

Kris Aubrey-Bassler ◽

Danielle Coombs ◽

Patrick Parfrey ◽

...

Keyword(s):

Low Back Pain ◽

Back Pain ◽

Lumbar Spine ◽

General Practitioners ◽

Ct Imaging ◽

Low Back ◽

Spine Ct ◽

Red Flag ◽

Radicular Syndrome ◽

Record Review

Abstract Background CT Imaging is often requested for patients with low back pain (LBP) by their general practitioners. It is currently unknown what reasons are common for these referrals and if CT images are ordered according to guidelines in one province in Canada, which has high rates of CT imaging. The objective of this study is to categorise lumbar spine CT referrals into serious spinal pathology, radicular syndrome, and non-specific LBP and evaluate the appropriateness of CT imaging referrals from general practitioners for patients with LBP. Methods A retrospective medical record review of electronic health records was performed in one health region in Newfoundland and Labrador, Canada. Inclusion criteria were lumbar spine CT referrals ordered by general practitioners for adults ≥18 years, and performed between January 1st-December 31st, 2016. Each CT referral was identified from linked databases (Meditech and PACS). To the study authors’ knowledge, guidelines regarding when to refer patients with low back pain for CT imaging had not been actively disseminated to general practitioners or implemented at clinics/hospitals during this time period. Data were manually extracted and categorised into three groups: red flag conditions (judged to be an appropriate referral), radicular syndrome (judged be unclear appropriateness), or nonspecific LBP (determined to be inappropriate). Results Three thousand six hundred nine lumbar spine CTs were included from 2016. The mean age of participants was 54.7 (SD 14 years), with females comprising 54.6% of referrals. 1.9% of lumbar CT referrals were missing/unclear, 6.5% of CTs were ordered on a red-flag suspicion, 75.6% for radicular syndromes, and 16.0% for non-specific LBP; only 6.5% of referrals were clearly appropriate. Key information including patient history and clinical exams performed at appointment were often missing from referrals. Conclusion This audit found high proportions of inappropriate or questionable referrals for lumbar spine CT and many were missing information needed to categorise. Further research to understand the drivers of inappropriate imaging and cost to the healthcare system would be beneficial.

Download Full-text