Understanding respiratory disease 'due to' or 'with' COVID-19 to assess appropriate treatmentGA Marraro, C Spada
Healthcare Accountability Lab, University of Milan, Italy
The outbreak of coronavirus disease 2019 (COVID-19), a complex multisystem disorder caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2),, has had a dramatic impact on the social, economic, and in particular, health sphere, causing significant morbidity and mortality globally, determining the necessary interventions for prevention and treatment. Interventions hampered by the changing nature of the COVID-19 pandemic have also brought unprecedented levels of uncertainty. Paucity of evidence of scientific COVID-19 knowledge is still related to the identification of diseases and death of patients 'due to' or 'with' COVID-19., Indeed, clear criteria for the definition of COVID-19–related deaths are not available. Accessible statistics are based on defining COVID-19–related deaths as those occurring in patients who test positive for SARS-CoV-2 independently from preexisting diseases (comorbidity or multimorbidity) that may have caused death.
To detect SARS-CoV-2 RNA, in clinical practice, the most widely utilized nucleic acid amplification test (NAAT) is the reverse transcriptase-polymerase chain reaction (RT-PCR). Consequences of a false-positive RT-PCR or other NAAT test results have also been highlighted from various perspectives.,,
To avoid/reduce false-positive test results, it is necessary to formulate the correct diagnosis by a molecular test accompanied by a radiographic investigation in a symptomatic patient. In the presence of lung disease, a differential diagnosis should be made between common lung disease and lung disease caused by SARS-CoV-2 infection. This distinction is essential to avoid diagnostic errors that can compromise the treatments to be applied and the knowledge of the morbidity and mortality of the COVID-19 pandemic.
Currently, the accuracy of SARS-CoV-2 test is still a larger long-term problem, posing challenges in identifying the causes that may lead to false-positive RT-PCR or other NAAT test results and the role that the presence of SARS-CoV-2 infection may have had in the postoperative course and therefore in the outcome.,, In patients with severe disease or undergoing emergency surgery, it is necessary to distinguish between lung disease caused directly by SARS-CoV-2 infection and respiratory compromise caused by the underlying disease that the patient is suffering from.
The multisystem disorder caused by SARS-CoV-2 causes a clear respiratory disease even if, as with many other viruses, other organs can be variously affected. It is therefore important to distinguish the condition of the patient with respiratory pathology directly caused by COVID-19 from that of the patient with preexisting comorbidity or multimorbidity diseases who can however be positive on the RT-PCR test. A positive RT-PCR test alone can create confusing conditions for the diagnosis and treatment of lung disease. Even the previous radiographic picture is difficult to differentiate from the pathology of COVID-19.
The diagnosis of pulmonary pathology attributable to the direct cause (“due to COVID-19”) or as a contributing cause of other conditions (“involving COVID-19”) is a problem that has also been studied some time for acute respiratory distress syndrome (ARDS). The difficulty consists in defining the criteria for the inclusion of critically ill patients with respiratory failure for the diagnosis of ARDS. For a long time, negative outcomes of respiratory therapies have been attributed to “ARDS” that were indeed related to extra respiratory problems or inappropriate treatments burdened by high morbidity and mortality and not to the direct cause of respiratory pathology. The “Berlin Consensus Conference” clearly defined the diagnostic criteria for ARDS, thus differentiating the direct cause of respiratory diseases from contributing causes of other origins and helping to define the real incidence of ARDS in the light of current evidence.,,,, Just as was done to reach the correct definition of diagnostic criteria for ARDS, the same should be attempted for COVID-19 in order to avoid negative outcomes that can be incorrectly attributed to COVID-19. This pathology can easily be indicted for covering up diagnostic and clinical errors.
The correct diagnosis of COVID-19 is performed with positivity of the SARS-CoV-2 RT-PCR test in symptomatic patients but it is also necessary to consider that patients with a positive test may be pauci-symptomatic or asymptomatic. Positive SARS-CoV-2 RT-PCR test confirms that there is an ongoing COVID-19 infection but does not ensure that the other disease the patient is suffering from is due to or triggered by COVID-19. Also, the test has a greater diagnostic value in defining whether the disease is only COVID-19–related if performed immediately upon the onset of the respiratory disease. Unfortunately, the patient often comes to appropriate health observation only when his clinical picture is complex, and the symptoms of COVID-19 can be associated or superimposed to other symptoms of pathology of which the patient is a carrier.
Thus, in our opinion, it is necessary to differentiate what is “due to COVID-19” from other conditions “involving COVID-19” to apply a specific treatment and evaluate the outcome. This distinction is not easy but it is necessary for obtaining a correct diagnosis and attributing the clinical conditions to COVID-19 or to other causes. Specificically, it can be useful to:
In this issue of the Journal of Postgraduate Medicine, the retrospective cohort study carried out by Karna et al. is of interest because it investigates, albeit in a limited number of patients, the performance of prognostic risk scores and assesses the postoperative morbidity and mortality in COVID-19 patients after emergency surgery. The authors are to be commended for their research that, thanks to the type of inquiry they chose, offers us an opportunity for a thorough reappraisal of our understanding of the therapeutical approach to a specific disease in COVID-19 patients and its actual application in today's clinical practice.
But, being a retrospective study and without the possibility of control cases, the preexisting morbidity and mortality for the specific pathologies considered are unknown. It would have been necessary to analyze the determining conditions in relation to the patient's basic conditions and the specific emergency surgery rather than the condition of positivity to SARS-CoV-2 infection. Having included in the study many pathologies, and therefore different urgent surgical interventions—each with specific operating conditions—makes the aggregate results too general and less relevant than the possibility of stratifying according to pathologies or if the number of cases had been larger.
A further limitation is highlighted by the univocal classification of respiratory pathology as ARDS. Given the widely existing knowledge in the literature on postoperative pulmonary pathology, the diagnosis of ARDS seems to be forced as the pulmonary compromise can easily be attributed to surgical and anesthetic complications that can significantly affect the postoperative course and outcomes.
The study with a control group would have allowed comparing the same complications but without any correlation with COVID-19. The cases, as they are described, suggest that most patients suffered from “mild or moderate” respiratory failure. Reasonably, it was a disease dependent not directly on COVID-19 but resulting from the normal evolution of the underlying disease, from surgical treatment and postoperative treatment. Although the COVID-19 pandemic has disrupted the prevention and treatment of a host of diseases, attention must be placed on the risk of generalizing without proper evidence inferring that diseases are directly related and caused by COVID-19.
It is often difficult to make inferences about causality due to the challenge to deduce, legitimately, a cause-and-effect relationship between events or variables, solely on the basis of an observed association or correlation between them. Actually, robust statistical methods can contribute objectively to readily quantify associations between observed variables and quantities derived from these measures., Much of the published COVID-19 observational research has been fuelled by genuine scientific curiosity, and we should take into account the fact that outcomes of observational studies are only hypothesis forming, offering closer examination of the situation analyzed to identify whether a true causation can be established between the variables.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.