- Email: [email protected]
Sensitivity of modern computed tomography in detecting acute subarachnoid haemorrhage
Abstract / Clinical Radiology 71 (2016) S1eS10
S7
In patients referred from their general practitioner (GP), knee pain in the elderly is often non-acute and degenerative cause of knee pain should be sought first. We audit the MRI knee diagnostic yield in elderly patients (>60 years) referred from GPs and whether the referral was appropriate. Methods and materials: Data was collected from three centres e a specialist musculoskeletal hospital, a teaching hospital and a district general hospital. Retrospective review of 50 consecutive MRI knee referrals for elderly (>60 years) patients from GP’s for each centre were assessed over a period of time in 2015. Each case was assessed on the picture archiving and communication system (PACS) and computerised radiology information system (CRIS) for appropriateness. This was ascertained by evaluating if they had a radiograph of the knee prior to MRI referral and, if so, by analysing the radiograph. The MRI knee and radiograph findings were also analysed. Results: The age and sex distribution of the cohorts of the three centres were comparable. Only a third of the cohort had prior radiographs and half the MRI referrals were inappropriate. A third had moderate to severe degenerative change on MRI and a further third had degenerative meniscal tears. Conclusion: All MRI knee referrals in elderly patients (over 60 years old) from GPs should have a radiograph prior to referral. If the radiograph demonstrates moderate or severe osteoarthritic changes then MRI does not seem to add any further information and hence is inappropriate.
Symptom onset to imaging: 0e14 days. In negative scans, limitations of CT was mentioned in 121 cases (50%) and successful LP rate was 146/242 (60%). A definitive xanthochromia result was available in 136 cases e only one was positive. Review revealed a tiny equivocal convexity hyperdensity on CTB with corresponding blooming artefact on subsequent susceptibility weighted magnetic resonance imaging (MRI) but no relevant abnormality on vascular imaging including Digital subtraction angiography (DSA). Of those with no LP (96) or equivocal xanthochromia result (ten), 20 had further imaging with computed tomography angiography (CT-A)/ computed tomography venography (CT-V)/MRI with no occult SAH cases. Sensitivity of CT for SAH in our cohort, including positive scans and negative scans with either corroborating definitive xanthochromia result or further vascular imaging/MRI is 92% (confidence interval [CI] 62e100%). Specificity 100% (CI 97e100%). If the case with positive xanthochromia but equivocal imaging is excluded, sensitivity is 100% (CI 70e100%). Conclusion: Modern CT is a highly sensitive tool for detecting SAH. Sensitivity was between 92e100% with one indeterminate case. LP uptake was 60% in negative CTB with a small number having further relevant imaging while limitations of CT was mentioned in 50%.
Radiology registrars in the hot seat: accuracy of trauma CT cervical spine reports issued during on-calls
Authors: Michael Paddock, Ramdas Senasi, Caroline Smith, Xing-Chang Wei, Ashok Raghavan, Seemab Seemab
Authors: Nyla Alam Khan, Sheetal Gagrani, Sangoh Lee, Raj Bhatt
Purpose: Both Sheffield Children’s Hospital (Sheffield, UK) and Alberta Children’s Hospital (Calgary, Canada) use a mixture of the lateral contrast swallow (LCS) and the lateral tube oesophagram (LTO) technique for the diagnosis of H-type tracheoesophageal fistula (H-TEF). We present our combined experiences using LCS and LTO in the radiological diagnosis of H-TEF. Methods and materials: Medical notes and imaging were interrogated in patients with a suspected diagnosis of H-TEF across both centers from 1999 to 2014. Results: 11 patients were diagnosed with H-TEF. n¼2 were diagnosed surgically; no documentation was available for one patient. n¼8 had a diagnostic fluoroscopic investigation, of which three were diagnosed on the first attempt using LCS. Of the remaining five patients that were subsequently proven positive with a diagnostic LTO, three were found negative on initial LCS. Three patients had evidence of pulmonary aspiration with one patient requiring transfer to intensive care. All patients had access to the hospital resuscitation team. Conclusion: Based on our combined experience, a repeat study with LTO would be complimentary if the initial LCS were negative, allowing for a greater confidence in the radiological diagnosis of H-TEF. The decision for further or repeat imaging is guided by discussion with the clinical team and the index of suspicion. Resuscitation support is essential.
Purpose: Computed tomography (CT) cervical spine (C-spine) scans are frequently performed within a trauma setting during on-call periods. Initial provisional reports are issued by radiology registrars. These are subsequently checked and verified by consultant radiologists. The aim of this study was to determine the discrepancy rate between the provisional report and the final report and the impact of discordant reports on patient management. Methods and materials: A retrospective study of consecutive CT C-spines reported by registrars while on call from November 2014eNovember 2015. Preliminary registrar reports were compared with the final reports issued by the consultant. Any discordancy was recorded and independently reviewed by two experienced musculoskeletal radiology consultants. Discrepancies between the two reports were classified as major, significant and minor as per The Royal College of Radiologists’ guidelines. Results: 630 CT scans were included in this study. Of these, 34 discrepancies were found giving an overall error rate of 5.4%. There were three misses (0.5%) classified as major, 14 significant (2.2%) and 17 minor (2.7%). There were three false positive errors. No significant association in discrepancy rate between various training years was identified. Conclusion: The incidence of major discrepancies at our institution is low and comparable to published research. They did not significantly impact patient management. Our study provides an insight into common errors made by registrars. We emphasise the importance of sufficient training for registrars. This includes preparation for on-calls by highlighting frequent errors made and the preceding causes for these. Sensitivity of modern computed tomography in detecting acute subarachnoid haemorrhage Authors: Claire McArthur, Scott Blackwell Purpose: Determine sensitivity of modern computed tomography (CT) in detecting subarachnoid haemorrhage (SAH). Ascertain rate of lumbar puncture (LP), further imaging and reporting of limitations of CT in negative cases. Methods and materials: Retrospective study January 2015eMarch 2016 of adults undergoing acute CT brain (CTB) for non- traumatic SAH. Scan result, time from onset of symptoms, comment on limitation of CTB, LP rate, xanthochromia result and relevant further imaging results were recorded. In discordant cases, all imaging reviewed. Results: Of 266 CTB in 257 patients, 242 (91%) showed no acute finding; 12 positive for SAH (4.5%); 12 other acute pathology.
Retrospective review of techniques used for diagnosis of H-type tracheoesophageal fistula across two paediatric regional centres
The five-year Oxford experience of paediatric ultrasound-guided percutaneous biopsies Authors: Nassim Parvizi, Mark Bamber, Shyamal Saujani, Sarah Dillon, Kaye Platt, Subhasis Chakraborty Purpose: Children with cancer require histological diagnosis to guide appropriate therapy. This can be performed by percutaneous imageguided biopsy. The aim of this study was to assess the diagnostic accuracy and safety of percutaneous ultrasound-guided paediatric biopsies in a tertiary referral centre. Methods and materials: A retrospective analysis of clinical data related to percutaneous ultrasound-guided biopsies performed for histological diagnosis in patients aged 0 to 18 years between January 2010 and December 2015 in a tertiary paediatric hospital was conducted. A total of 116 percutaneous ultrasound-guided biopsies were performed in 114 children. The median age of the children was 5.79 years. A primary diagnosis was made in 114 patients and suspected recurrence in two. Most biopsies were performed using 18-gauge core biopsy needles with a minimum of two cores per examination.
S7
In patients referred from their general practitioner (GP), knee pain in the elderly is often non-acute and degenerative cause of knee pain should be sought first. We audit the MRI knee diagnostic yield in elderly patients (>60 years) referred from GPs and whether the referral was appropriate. Methods and materials: Data was collected from three centres e a specialist musculoskeletal hospital, a teaching hospital and a district general hospital. Retrospective review of 50 consecutive MRI knee referrals for elderly (>60 years) patients from GP’s for each centre were assessed over a period of time in 2015. Each case was assessed on the picture archiving and communication system (PACS) and computerised radiology information system (CRIS) for appropriateness. This was ascertained by evaluating if they had a radiograph of the knee prior to MRI referral and, if so, by analysing the radiograph. The MRI knee and radiograph findings were also analysed. Results: The age and sex distribution of the cohorts of the three centres were comparable. Only a third of the cohort had prior radiographs and half the MRI referrals were inappropriate. A third had moderate to severe degenerative change on MRI and a further third had degenerative meniscal tears. Conclusion: All MRI knee referrals in elderly patients (over 60 years old) from GPs should have a radiograph prior to referral. If the radiograph demonstrates moderate or severe osteoarthritic changes then MRI does not seem to add any further information and hence is inappropriate.
Symptom onset to imaging: 0e14 days. In negative scans, limitations of CT was mentioned in 121 cases (50%) and successful LP rate was 146/242 (60%). A definitive xanthochromia result was available in 136 cases e only one was positive. Review revealed a tiny equivocal convexity hyperdensity on CTB with corresponding blooming artefact on subsequent susceptibility weighted magnetic resonance imaging (MRI) but no relevant abnormality on vascular imaging including Digital subtraction angiography (DSA). Of those with no LP (96) or equivocal xanthochromia result (ten), 20 had further imaging with computed tomography angiography (CT-A)/ computed tomography venography (CT-V)/MRI with no occult SAH cases. Sensitivity of CT for SAH in our cohort, including positive scans and negative scans with either corroborating definitive xanthochromia result or further vascular imaging/MRI is 92% (confidence interval [CI] 62e100%). Specificity 100% (CI 97e100%). If the case with positive xanthochromia but equivocal imaging is excluded, sensitivity is 100% (CI 70e100%). Conclusion: Modern CT is a highly sensitive tool for detecting SAH. Sensitivity was between 92e100% with one indeterminate case. LP uptake was 60% in negative CTB with a small number having further relevant imaging while limitations of CT was mentioned in 50%.
Radiology registrars in the hot seat: accuracy of trauma CT cervical spine reports issued during on-calls
Authors: Michael Paddock, Ramdas Senasi, Caroline Smith, Xing-Chang Wei, Ashok Raghavan, Seemab Seemab
Authors: Nyla Alam Khan, Sheetal Gagrani, Sangoh Lee, Raj Bhatt
Purpose: Both Sheffield Children’s Hospital (Sheffield, UK) and Alberta Children’s Hospital (Calgary, Canada) use a mixture of the lateral contrast swallow (LCS) and the lateral tube oesophagram (LTO) technique for the diagnosis of H-type tracheoesophageal fistula (H-TEF). We present our combined experiences using LCS and LTO in the radiological diagnosis of H-TEF. Methods and materials: Medical notes and imaging were interrogated in patients with a suspected diagnosis of H-TEF across both centers from 1999 to 2014. Results: 11 patients were diagnosed with H-TEF. n¼2 were diagnosed surgically; no documentation was available for one patient. n¼8 had a diagnostic fluoroscopic investigation, of which three were diagnosed on the first attempt using LCS. Of the remaining five patients that were subsequently proven positive with a diagnostic LTO, three were found negative on initial LCS. Three patients had evidence of pulmonary aspiration with one patient requiring transfer to intensive care. All patients had access to the hospital resuscitation team. Conclusion: Based on our combined experience, a repeat study with LTO would be complimentary if the initial LCS were negative, allowing for a greater confidence in the radiological diagnosis of H-TEF. The decision for further or repeat imaging is guided by discussion with the clinical team and the index of suspicion. Resuscitation support is essential.
Purpose: Computed tomography (CT) cervical spine (C-spine) scans are frequently performed within a trauma setting during on-call periods. Initial provisional reports are issued by radiology registrars. These are subsequently checked and verified by consultant radiologists. The aim of this study was to determine the discrepancy rate between the provisional report and the final report and the impact of discordant reports on patient management. Methods and materials: A retrospective study of consecutive CT C-spines reported by registrars while on call from November 2014eNovember 2015. Preliminary registrar reports were compared with the final reports issued by the consultant. Any discordancy was recorded and independently reviewed by two experienced musculoskeletal radiology consultants. Discrepancies between the two reports were classified as major, significant and minor as per The Royal College of Radiologists’ guidelines. Results: 630 CT scans were included in this study. Of these, 34 discrepancies were found giving an overall error rate of 5.4%. There were three misses (0.5%) classified as major, 14 significant (2.2%) and 17 minor (2.7%). There were three false positive errors. No significant association in discrepancy rate between various training years was identified. Conclusion: The incidence of major discrepancies at our institution is low and comparable to published research. They did not significantly impact patient management. Our study provides an insight into common errors made by registrars. We emphasise the importance of sufficient training for registrars. This includes preparation for on-calls by highlighting frequent errors made and the preceding causes for these. Sensitivity of modern computed tomography in detecting acute subarachnoid haemorrhage Authors: Claire McArthur, Scott Blackwell Purpose: Determine sensitivity of modern computed tomography (CT) in detecting subarachnoid haemorrhage (SAH). Ascertain rate of lumbar puncture (LP), further imaging and reporting of limitations of CT in negative cases. Methods and materials: Retrospective study January 2015eMarch 2016 of adults undergoing acute CT brain (CTB) for non- traumatic SAH. Scan result, time from onset of symptoms, comment on limitation of CTB, LP rate, xanthochromia result and relevant further imaging results were recorded. In discordant cases, all imaging reviewed. Results: Of 266 CTB in 257 patients, 242 (91%) showed no acute finding; 12 positive for SAH (4.5%); 12 other acute pathology.
Retrospective review of techniques used for diagnosis of H-type tracheoesophageal fistula across two paediatric regional centres
The five-year Oxford experience of paediatric ultrasound-guided percutaneous biopsies Authors: Nassim Parvizi, Mark Bamber, Shyamal Saujani, Sarah Dillon, Kaye Platt, Subhasis Chakraborty Purpose: Children with cancer require histological diagnosis to guide appropriate therapy. This can be performed by percutaneous imageguided biopsy. The aim of this study was to assess the diagnostic accuracy and safety of percutaneous ultrasound-guided paediatric biopsies in a tertiary referral centre. Methods and materials: A retrospective analysis of clinical data related to percutaneous ultrasound-guided biopsies performed for histological diagnosis in patients aged 0 to 18 years between January 2010 and December 2015 in a tertiary paediatric hospital was conducted. A total of 116 percutaneous ultrasound-guided biopsies were performed in 114 children. The median age of the children was 5.79 years. A primary diagnosis was made in 114 patients and suspected recurrence in two. Most biopsies were performed using 18-gauge core biopsy needles with a minimum of two cores per examination.