Sir—We read with great interest the findings from the Royal College of Radiologists (RCR) national audit, which demonstrated the low sensitivity for the identification of COVID-19 in a preoperative group of patients making it difficult to justify computed tomography (CT) screening during the pandemic in the UK.1 This corroborates our own experience2 and that of another institution3 in demonstrating a low diagnostic yield of CT for COVID-19 in this cohort of patients; however, in this letter, we would like to also highlight the potential impact that our reports may have in a specific cohort: renal transplant patients.
At our institution, renal transplantation continued throughout the pandemic. In the early stages of the pandemic, we performed preoperative thoracic CT as reverse transcription polymerase chain reaction (RT-PCR) testing was not available for preoperative patients. Subsequently, CT was performed in cases where RT-PCR results would not be available in an acceptable time frame prior to the operation. Here, we correlate imaging findings with PCR and re-evaluate the imaging of those patients who had CT findings classic or indeterminate for COVID-19 pneumonia, as well as determine how the radiology report affected clinical care.
We performed a retrospective study of the potential renal transplant recipients who have undergone CT chest screening for COVID-19 from 27/3/20 to 9/6/20. The study took place as part of a Trust retrospective audit and the requirement for informed consent was waived. Seventy-four CT chest examinations took place in 68 patients, 60% male, median age 53 years (range 24–77 years). CT examinations were reported according to the BSTI guidelines, with 57 reported as normal, 7 as indeterminate, 3 as classic/probable COVID-19, and 7 were reported as non-COVID-19 (four fluid overload, three infection). One patient with indeterminate findings went on to have another CT chest 25 days later, which remained indeterminate, whilst another patient with indeterminate findings had a subsequent normal CT 31 days later. 20 were reported by thoracic radiologists, whilst 54 were reported by other cross-sectional radiologists. 49 patients had nasopharyngeal swab RT-PCR, in 55% the result of RT-PCR was available within a day of the issued CT report. All patients who had a normal CT and went on to have RT-PCR had a negative RT-PCR. The only patient in the overall cohort to have a positive RT-PCR test had an indeterminate CT examination; it later transpired that this patient was symptomatic with a fever.
Eight of nine patients were suspended from the transplant waiting list because of the report (of classic/probable COVID-19 or indeterminate for COVID-19). At the time of writing, five remain on the waiting list, whilst three patients have subsequently received their transplant following a delay of 25, 33, and 61 days.
Two thoracic radiologists (H.P., L.W.) undertook the review of the nine CT examinations reported as classic/probable or indeterminate for COVID-19, blinded to the RT-PCR result, and issued a report without knowledge of the purpose of the study and cohort (see Table 1 and Fig 1 ). In three cases, both thoracic radiologists interpreted CT examinations as non-COVID-19 (a combination of fluid overload, infection and expiratory study) when originally the examinations had been reported as indeterminate by non-thoracic cross-sectional radiologists. There was a 48.15% overall agreement amongst the three raters (original report and re-review). The interobserver agreement was poor (free-marginal kappa 0.31, 95% confidence interval [CI]: 0.05–0.56). Interobserver agreement was better between the two chest radiologists on blinded review: 66.7% overall agreement with a moderate agreement of free-marginal kappa 0.56 (95% CI: 0.12–0.99). The interpretation and representative images are provided below, alongside patient outcomes.
Table 1.
Results from second reporting of the nine CT examinations reported as classic/probable or indeterminate for COVID-19.
| Case | Initial report | Reporter 1 | Reporter 2 | PCR | Outcome |
|---|---|---|---|---|---|
| 1 | Classic/probable (general radiologist) | Indeterminate | Classic/probable | Negative | Suspended: remaining on waiting list |
| 2 | Classic/probable (thoracic radiologist) | Classic/probable | Classic/probable | Negative | Postponed and happened 61 days later |
| 3 | Indeterminate (general radiologist) | Non-COVID-19 (other infection) | Non-COVID-19 (other infection) | Positive | Suspended: remaining on waiting list |
| 4 | Indeterminate (general radiologist) | Indeterminate | Indeterminate | Negative | Postponed and happened 25 days later |
| 5 | Indeterminate (general radiologist) | Non-COVID-19 (fluid overload) | Non-COVID-19 (fluid overload) | Negative | Postponed and happened 33 days later |
| 6 | Classic/probable (general radiologist) | Indeterminate | Indeterminate | Negative | Suspended: remaining on waiting list |
| 7 | Indeterminate (thoracic radiologist) | Indeterminate | Non-COVID-19 (other infection) | Negative | Proceeded as planned |
| 8 | Indeterminate (general radiologist) | Indeterminate | Non-COVID-19 (infection/atelectasis) | Negative | Suspended: remaining on waiting list |
| 9 | Indeterminate (general radiologist) | Non-COVID-19 (expiratory study/viral infection/fluid) | Non-COVID-19 (expiratory study/viral infection/fluid) | Negative | Suspended: remaining on waiting list |
Figure 1.
Example images from the nine cases classified as indeterminate for COVID-19 or classic/probable COVID-19 on initial report.
Our findings highlight the difficulties in classification of CT findings in this cohort and emphasise the challenges in differentiating between British Society of Thoracic Imaging (BSTI) categories, particularly those of indeterminate and non-COVID with poor interobserver agreement; this has also been found on chest radiography.4 In this cohort, ground-glass opacity, the hallmark of COVID-19 infection, may in fact be caused by a degree of fluid overload secondary to end-stage renal failure, and this is supported by the reclassification of two cases as volume overload by reinterpretation by thoracic radiologists. Time since dialysis and patient's admission weight (in comparison to their “target” or “dry” weight) may lend important additional clinical information when interpreting preoperative scans in this cohort. Furthermore, this cohort of patients is also more susceptible to bacterial infection and other comorbidities or medication effect, which may challenge interpretation. Meanwhile, the patient who was symptomatic and RT-PCR positive had indeterminate features on CT: unilateral, relatively central consolidation/ground glass. Overall, therefore, these findings support the assertion made by National Institute for Health and Care Excellence (NICE) guidelines5 that CT should not be performed in patients preoperatively with no symptoms.
Conflict of interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.Callaway M., Harden S., Ramsden W. A national UK audit for diagnostic accuracy of preoperative CT chest in emergency and elective surgery during COVID-19 pandemic. Clin Radiol. 2020;75(9):705–708. doi: 10.1016/j.crad.2020.06.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chetan M.R., Tsakok M.T., Shaw R. Chest CT screening for COVID-19 in elective and emergency surgical patients: experience from a UK tertiary centre. Clin Radiol. 2020;75(8):599–605. doi: 10.1016/j.crad.2020.06.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hamilton N.E., Adam G.H., Ifan D.L. Diagnostic utility of additional whole-chest CT as part of an acute abdominal pain CT imaging pathway during the COVID-19 pandemic. Clin Radiol. 2020;75(8):592–598. doi: 10.1016/j.crad.2020.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hare S.S., Tavare A.N., Dattani V. Validation of the British Society of Thoracic Imaging guidelines for COVID-19 chest radiograph reporting. Clin Radiol. 2020;75(9) doi: 10.1016/j.crad.2020.06.005. 710.e9–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.NICE. 5 Care at the time of the transplant. COVID 19 rapid guideline: renal transplantation. NICE guideline [NG178]. Published date: 19 June 2020https://www.nice.org.uk/guidance/ng178/chapter/5-Care-at-the-time-of-the-transplant (accessed: 21/8/2020).