The number of ARDS cases caused by viral pneumonia is increasing, resulting in a high mortality rate . Previous studies have mostly focused on influenza viruses, such as H1N1 and H7N9 [15, 16]. Cases of adenovirus-associated ARDS have gradually been increasing, which may be due to recent developments in molecular diagnostic technology . However, there is limited data about the viral etiology of ARDS in patients who required mechanical ventilation. Zhou et al. reported that the incidence of adenovirus pneumonia ranks third among viral pneumonia in adults in China . They have also found that among virus-related ARDS patients with PO2/FiO2 ratio < 200 mmHg, HAdV infection was the most frequently detected virus . Our study showed that in all ARDS patients, virus-related ARDS accounted for 32.87% of infections. Among these infections, adenovirus-associated ARDS accounted for 9.79% of all ARDS patients and 29.79% of virus-related ARDS patients. The prevalence of respiratory viruses varies in different countries and different populations [4–6, 13]. To the best of our knowledge, this is the first study about the viral etiology in ARDS patients who required mechanical ventilation in China.
There are few studies on adenovirus-associated ARDS in adults who required mechanical ventilation [4–6, 13]. In 2000, two non-immunocompromised soldiers became infected with adenovirus, which resulted in ARDS [4–6, 13]. In 2006, there was an outbreak of adenovirus pneumonia caused by HAdV-B11 in the USA; 140 people were diagnosed with HAdV infection, and 24 patients who were diagnosed with ARDS were admitted to the ICU . Some studies have proposed that severe adenovirus infection is likely to occur in children and immunocompromised adults, such as HIV patients, and patients after transplantation [23, 24]. In this study, most of the adenovirus-associated ARDS patients, with a mean age of 54 years, had no underlying diseases. Among these patients, 78.57% of them were men, and 50% of the patients were smokers, suggesting that severe adenovirus pneumonia in non-immunocompromised adults was likely to occur in middle-aged men. Additionally, 85.71% of the patients had a significant decrease in the number of CD3 + CD4 + T cells during the early stage, which suggests that adenovirus infection may cause immune system dysregulation.
Delayed clearance of respiratory adenovirus infection leads to a worse prognosis in these patients, and monitoring the viral load may help to predict the disease severity and the patients’ prognosis [25, 26]. Rapid identification of adenovirus viral infection is critical to reduce the overall costs of patient management. Multiplex Q-PCR is of great value in the early diagnosis of virus infection because of its high sensitivity [27, 28]. However, viral testing of URT and LRT samples may yield different results [27, 28]. Currently, few studies have been published that compare the diagnostic yields of URT and LRT samples to detect adenovirus. In our study, the detection rates of adenovirus from LRT and URT samples were 100% and 64.29%, respectively. Similarly, a European study in 2016 reported that the overall virus positivity rate of URT was lower than that of the LRT specimens (24.5% vs. 44.2%) . In this study, the percentage of positive specimens was higher in LRT than in URT specimens. On average, URT specimens cleared faster than LRT specimens, suggesting that traditional nasopharyngeal diagnostic techniques can miss cases of severe adenovirus infection. This suggests that LRT specimens are more reliable for diagnosing severe adenovirus infection, especially in patients with pneumonia that occurs several days after the infection onset when the frequency of virus detection in the URT has already decreased.
Several studies have shown that the mortality of severe adenovirus-associated ARDS can be as high as 26.7–80% in adults . In our study, among the 14 patients with adenovirus-associated ARDS, there were only two deaths, and the mortality rate was 14.29%. Compared with previous studies, the mortality rate of adenovirus-associated ARDS in this study was relatively low, and this may have several explanations. First, rapid identification of adenovirus viral infection and early intervention are important to reduce the overall mortality rate. In this study, the time from onset to intubation was relatively short. In addition, the PO2/FiO2 ratio for most adenovirus-associated ARDS patients was > 150 in this study, while most other studies showed that the PO2/FiO2 ratio in patients with severe pneumonia was < 150 [6, 23, 31]. Considering that the condition of severe adenovirus pneumonia patients was more advanced, the above results suggested that early intervention in adenovirus pneumonia-related ARDS helped to improve the patients’ prognosis. Second, establishing organ support, such as application of early renal replacement treatment and ECMO are important. Adenovirus-associated ARDS completely resolved in three patients who were supported by ECMO in this study, suggesting that early application of ECMO improved the prognosis of patients with adenovirus-associated ARDS. Finally, timely initiation of antiviral therapy is very important to improve patient outcome. Currently, antiviral therapies for adenovirus infection remain controversial. No specific and effective antiviral drug is available for adenovirus infection . Some studies have shown that cidofovir antiviral therapy in severe adenovirus pneumonia improves the clinical prognosis . However, clinical application of cidofovir is limited due to its toxic side effects and low-quality evidence. Other case reports have also shown that ribavirin can be used to treat adenovirus infection [34, 35]. Ganciclovir has been shown to be effective for treating adenovirus infection in animal experiments . In our study, 11 patients (78.57%) received antiviral therapy with ganciclovir and immunoglobulin therapy after confirming adenovirus infection, which might be the reason for the lower mortality.
This study prospectively observed the viral etiology of ventilated ARDS patients, especially for patients where adenovirus was associated with ARDS. We compared the adenovirus detection rate and adenovirus load in different respiratory tract specimens, which were also a highlight of the study. This study also has several limitations. It was a single-center study with a relatively small number of patients enrolled. In addition, this study used multiplex Q-PCR to detect multiple respiratory viruses at the same time without genotyping the adenovirus.