As the respiratory tract is a common site of infection, it is an important component of the post-mortem examination to be able to accurately detect the causative microorganisms [1]. Key findings from the results of this study using retrospective data from Coronial post-mortem examinations are (1) of all the samples used, nasopharyngeal swabs yielded the greatest proportion of respiratory tract viral infection and (2) lung tissue alone was not an optimal specimen for detection of respiratory tract viral infection by multiplex PCR.
The prevalence of viruses detected was not unexpected. Rhinovirus, which is most commonly associated with the “common cold” was the most common. Whilst it is considered to cause a relatively benign illness, it is implicated in the exacerbation of respiratory diseases including chronic obstructive pulmonary disease, asthma and cystic fibrosis [5]. However, it is difficult to comment on the role of rhinovirus in this study, as in half of the cases when rhinovirus was detected, the cause of death was either unascertained, or considered to be due to a non-infective cause. Respiratory Syncytial Virus (RSV) and influenza were the next most common viruses that were identified [6]. It was noted that influenza A was more commonly seen in those under 60, and vice-versa for RSV, but this was not statistically significant. The observation may be attributed to the high influenza vaccination rates of the elderly in Australia and the fact that RSV increasingly affects the elderly, both in care facilities and the community [7].
For the optimal sampling regime at autopsy, it has been previously suggested that a minimum of four samples should be taken in cases of sudden-unexpected death with respiratory symptoms at any age, namely: lung, nasopharyngeal swab, bronchial swab, and at least one other swab from the affected tissue [2]. In addition, throat swabs are recommended when tonsillitis is apparent, and when empyema is suspected, it is recommended to sample pleural fluid [2]. Whilst this certainly offers the greatest opportunity for maximal detection of an infectious agent, it could be argued that there is a redundancy when it is possible to detect the overwhelming majority of respiratory tract viral infection using only a nasopharyngeal swab. Nasopharyngeal swabs have been shown to have a high yield of detecting viral respiratory tract infection [8]. Due to the mechanics of the mucocillary escalator system, the airways are more likely to contain downstream pathogens contributing to disease, hence making the nasopharyngeal area a prime location for collection [9]. The nasopharyngeal swab has previously been advocated in proposed guidelines relating to sampling in the setting of suspected flu or viral respiratory infection [3]. Certainly, the results of this study serve as confirmation of this strategy. This study was performed before the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic due to corona virus (COVID-19) infection. Nonetheless, the findings support the current United States of America Centers for Disease Control and Prevention recommendation for post-mortem investigation, which is to submit a nasopharyngeal swab for virological testing with a lung swab if an autopsy is performed [10]. Nasopharyngeal swabs have been advocated for post-mortem sampling for COVID-19 [11] with positivity from nasopharyngeal swabs and failure to identify COVID-19 from lung swabs being reported [12]. Incorporating bronchial swabs has also been suggested [11, 13]. Others, including the Royal College of Pathologists [14] suggest using a strategy based on testing in living subjects, including use of sputum and bronchoalveolar lavage samples [15]. A further consideration could be use of rectal swabs [16]. It is not possible to comment on the potential utility of samples others than those used in this reported study.
It was considered that it might be possible to minimize the number of samples taken and consequently reduce costs, by obtaining a sample of lung tissue for the purpose of both multiplex PCR viral and bacteriological culture study. However, the results indicate this would not be ideal, as lung samples yielded the poorest results overall, only detecting 63% of viruses. Nonetheless, even with a lower detection rate, all respiratory tract viral infections regarded as notifiable by the State’s Communicable Disease branch [17] would have been identified by analysis of a sample of lung. One of the main issues with microbiological investigations on lung tissue is that in cases of focal infection, there is a possibility of generating false negative results if the sampling is undertaken outside of the affected region, which could explain the low viral yield [18]. It was also considered the higher yield from lung swabs might reflect that these were bathed in viral transport medium compared with the lung samples that were not. Further investigation would be required, but studies, including for a possible effect of viral transport medium, were outside the scope of this study.
In this study that retrospectively reviewed 94 cases, from which four respiratory tract samples (nasopharyngeal swab, tracheal swab, lung swab, and lung tissue) had been obtained for multiplex PCR viral detection, it was found that a multiple sample strategy detected only one case of viral infection that would not have been reported from a nasopharyngeal swab alone. Although this was a small, retrospective study with only 27 viruses identified from 24 cases, it is concluded a nasopharyngeal swab is an optimal sample for detection of respiratory tract viral infection.