A key issue we face in reporting Covid 19 results is determining the Indeterminate or Presumptive positive result.
This can take two forms, firstly a rising ct value that doesn't reach the threshold for positivity of the assay. Does this mean it is an early infection and how do we manage this clinically and explain that to the patient/clinician. Something is there and do we need a new sample in a few days time? Also what is the management of these patients?
Secondly is the issue around a positive, non-specific, marker like the E-gene (a lot of beta coronavirus's have this gene,) with a negative specific marker such as N2. The current best practice in my region is to repeat the assay, ideally via another method. However what happens if this result has a repeated nonspecific positive result ? Or it comes back as negative? What is the best guidelines to manage these patients? Community testing is simple enough, self isolation is a low risk control for these people.
But what if they are In-patients do we move them to the Covid only wards or with the negative/indeterminate result or a negative result but with a low level rising ct via PCR? Do we allow them to stay on the regular ward? Which result may be more accurate and what is the risk reward for the patient if they are indeed negative or have just a commensal coronavirus (with a E-gene) and move them into a high risk area?
Would be interested to see if anyone else has any opinions or thoughts. There is no right answer in this. There are big Pro's and Con's to each approach for both the individual and community depending on the plan in place, especially for In-patients. Where does that duty of care lie in this scenario.
Looking forward to seeing what people think and stay safe everyone
The sensitivity and specificity of a diagnostic test depend on several factors. Not just the kit. I do not have any data on the accuracy of the kit.
However the viral load varies with site of sampling. Please see:
Wenling Wang; Yanli Xu; Ruqin Gao; et al Detection of SARS-CoV-2 in Different Types of Clinical Specimens. Jama 2020 (in press).
So false positives and false negatives are possible.
It may be better to combine PCR data with imaging such as lung ultrasound. Lung ultrasound may detect abnormalities even in asymptomatic patients. These abnormalities resolve as the patient improves.
Some doctors argue that the molecules of the epidemic are difficult to break down and that they are developing themselves
Rajkumar Rajendram , False positive and negatives are always going to be an issue for any test but how do we interpret them when two tests contradict, especially if the patient isnt in a hospital or imaging is indeterminate at this point? Do we assume the worst outcome and corroborate clinically where possible? When testing in the millions a small issue like this leads to large numbers and maybe not enough capacity to follow up tests with imaging
Rajkumar Rajendram " the molecules of the epidemic are difficult to break down and that they are developing themselves " could you elaborate. I am not sure what you are trying to get at.
Dear Stephen,
One of the main issues with diagnostics relays on to a general misleading concept: when applying direct (ie PCR) or indirect (serology) methods we are searching for evidence of the agent in tissues, but certainly we are not diagnosing "disease". As with many other infections, there are a significant number of asymptomatic cases: people carrying SARS-Cov-2 but not sick (ie not Covid19).
In the frame of the wide and fast viral transmission a "false positive" or even a doubtful result in a RT-PCR, It should be interpreted as "potential transmitter" and therefore proceed with the isolation or ward measures.
The bigger problem for the RT-PCR are the "false negative", because there is still a chance to be infected and transmit virus. Any negative result should be confirmed by other method or by repeating the sampling on alternate days before considering the patient discharge or the movement permissions. Moreover, if the clinical suspicious persists after a negative result, the patient should be treated as a "positive-like" to avoid or minimize transmission risk.
Gastón Moré, thanks for the reply. I will note that asymptomatic does not mean 'not Covid 19', they still have the virus present even if asymptomatic and can still shed and infect, though at a potentially lower level.
But I do agree, it the realm of doubt with a result/patient history we should assume the worst case scenario until more evidence is brought to light.
What I find fascinating with False Negatives is the fact that even with very sensitive methods you could nearly count them as 'true' negatives at that time point due to the long incubation period though you cannot call someone having a false negative test when they are infected after that test.
Secondly the feasibility of mass testing people when negative is tricky, if we cannot trust the negative result completely, i.e it has a high NPV instead of low, then when do we stop testing a person?
Dear Stephen,
I mean that for most "diseases" there are no a unique method capable to certify you have symptoms due to an agent. In the actual situation "be infected" is the real point to identify (which includes sick individuals and asymptomatic carriers).
Your question is a key one and I guess there in no universal answer. In regards to potential false negative for molecular methods and if you check here: https://www.bmj.com/content/369/bmj.m1808, it is posible to assume the testing will never ends!
Nevertheless, the higher risk for transmission relays on the confirmed positives, if a person is tested negative 2 or 3 times, could be considered at least as low to no spreader. The second big issue is the apparent low immunity generated, therefore, we do not know if a person negative, or even positive today could be spreader and positive in a couple of months....
- Badges
- Science topic
- Similar topics
- Microbiology
- Assays