Diarrhea may be a secondary path of transmission for the novel coronavirus, scientists said Friday following the publication of the latest study reporting patients with abdominal symptoms and loose stool.
The primary path is believed to be virus-laden droplets from an infected person’s cough, though researchers in early cases have said they focused heavily on patients with respiratory symptoms and may have overlooked those linked to the digestive tract.
A total of 14 out of 138 patients (10 percent) in a Wuhan hospital who were studied in the new paper by Chinese authors in the Journal of the American Medical Association (JAMA) initially presented with diarrhea and nausea one or two days prior to development of fever and labored breathing.
The first US patient diagnosed with 2019-nCoV also experienced loose bowel movements for two days and the virus was subsequently detected in his stool, and there have been other such cases in China documented in the Lancet, albeit infrequently.
“Importantly, 2019-nCoV has been reported elsewhere in the feces of patients with atypical abdominal symptoms, similar to SARS which was also shed in urine, suggesting a fecal transmission route which is highly transmissible,” William Keevil, a professor of environmental healthcare at the University of Southampton said in a comment to the UK’s Science Media Centre.
The possibility is not totally surprising to scientists, given that the new virus belongs to the same family as SARS.
Fecal transmission of SARS was implicated in sickening hundreds in Hong Kong’s Amoy Gardens housing estate in 2003.
A rising plume of warm air originating in bathrooms contaminated several apartments and was transported by wind to adjacent buildings in the complex.
Based on the literature, “The 2019-nCoV virus found in stool may be transmitted through fecal spread,” added Jiayu Liao, a bioengineer at the University of California, Riverside.
But, he added, “We still do not know how long this virus can survive outside the body – HIV can only survive roughly 30 minutes outside the body – and what temperature range the 2019-nCoV is sensitive to.”
Fecal spread could present new challenges to the virus’s containment, but is more likely to be a problem inside hospitals, which can become “amplifiers” of epidemics, said David Fisman, an epidemiologist at the University of Toronto.
Benjamin Neuman, a virology expert at Texas A&M University-Texarkana, cautioned that while fecal transmission was “certainly worth considering,” “droplets and touching contaminated surfaces then rubbing eyes, nose or mouth” were likely the main way the virus was transmitted based on current data.
The Health Commission of Hubei province, China, first announced a cluster of unexplained cases of pneumonia on Dec 31, 2019.1 27 patients were initially reported, which was subsequently revised to 41 on Jan 11, 2020, with seven severe cases and one death.2
Some patients were reported to have radiographic ground-glass lung changes; normal or lower than average white blood cell lymphocyte, and platelet counts; hypoxaemia; and deranged liver and renal function.
Most were said to be geographically linked to the Huanan seafood wholesale market, which was subsequently reported by journalists to be selling freshly slaughtered game animals.3
To date, no evidence of person-to-person transmission or affected health-care workers has been published in the scientific literature.
The Chinese health authority said that the patients initially tested negatively for common respiratory viruses and bacteria, but later tested positive for a novel coronavirus.2
The virus was soon isolated and its genome sequenced by a number of Chinese scientists.4 The virus was tentatively named by WHO as the 2019 novel coronavirus (2019-nCoV).
Here, we report the epidemiological, clinical, radiological, laboratory, and genomic findings of a family cluster of five patients in Shenzhen who had a history of travel to Wuhan, and one other family member who has not travelled to Wuhan.
We report here a familial cluster of unexplained pneumonia due to 2019-nCoV. Six of seven family members had radiological changes of viral pneumonia, among whom five (patients 1, 2, 4, 5, and 7) tested positive for 2019-nCoV by RT-PCR. Five patients (patients 1, 2, 3, 4, and 7) had associated symptoms at the time of presentation.
Complete genome sequences of the two strains from patients 2 and 5 showed almost complete nucleotide identity with each other, and were closest to the bat SARS-related coronaviruses reported in 2018. Several possible scenarios of transmission exist.
The first and most likely scenario is that one virologically documented patient with pneumonia (patient 1) acquired the infection from a Wuhan hospital while visiting their relative (relative 1) and then patients 1–5 transmitted the virus to patient 7 on returning to Shenzhen.
The second scenario is that patients 1–5 have directly acquired the infection from relatives 2–5 and transmitted it to patient 7 on returning to Shenzhen.
But this scenario is less likely because patients 1–5 developed symptoms before relatives 2–5. The third scenario is that patients 1–5 acquired the infection from an unknown common source in Wuhan and transmitted it to patient 7 when back in Shenzhen.
For the patients’ relatives (relatives 2–5), they could have acquired the infection from the hospital or the community, although no virological confirmation was possible and they had no animal contacts, game food, or visits to the Huanan seafood wholesale market.
Notably, patient 1 or patient 3 who had visited Wuhan hospital might have been infectious before symptom onset because patient 5 was shedding virus without symptoms.
These findings suggested that person-to-person transmission and intercity spread of 2019-nCoV by air travel are possible, supporting reports of infected Chinese travellers from Wuhan being detected in other geographical regions.
Many of the epidemiological, clinical, laboratory, and radiological features of this novel coronavirus pneumonia were similar to those of SARS patients in 2003.8, 15, 16 The incubation period of the Wuhan pneumonia appeared similar to that of SARS.
The attack rate is rather high, up to 83% if we included the five patients (patients 1, 2, 3, 4, and 5) with unexplained ground-glass radiological changes of the lungs on CT scan as the case definition in this family outbreak after visiting Wuhan.
A rather unexpected finding from the lung CT scan of patient 5, which was done on the insistence by the nervous parents, also showed ground-glass pneumonic changes. Patient 5 was later confirmed virologically to have an asymptomatic infection.
Although asymptomatic patients with SARS were uncommon, they were documented in our retrospective study in the minor 2004 SARS outbreak after reopening of the wildlife market in Guangzhou.17
Notably, patients 3 and 4 were afebrile at presentation to our hospital. These cryptic cases of walking pneumonia might serve as a possible source to propagate the outbreak.
Further studies on the epidemiological significance of these asymptomatic cases are warranted.The symptoms of this novel pneumonia were also non-specific.
The three oldest patients in this family with comorbidities had more severe systemic symptoms of generalised weakness and dry cough. As expected, they might have decreased total white blood cell, lymphocyte, or platelet counts, with also extended activated thromboplastin time and increased C-reactive protein level.
The multifocal ground-glass changes on lung CT scan were typical of viral pneumonia. Their lung involvement was also more diffuse and extensive than those of the younger patients, whose blood test results were largely normal.
Patient 4, who had a history of chronic sinusitis, might have a bacterial superinfection because he had a productive cough instead of a dry cough. He also had a high white blood cell count, although the bacterial test was negative.
Interestingly, the two younger adults (patients 3 and 4) initially had diarrhoea, which was also reported in 10·6% (15 of 142) of our SARS patients at presentation;18 however, the subsequent faecal samples of patients 3 and 4 that were collected 9–10 days after symptom onset were negative for the virus after the diarrhoea had long subsided.
Up to 30% of patients with Middle East respiratory syndrome coronavirus (MERS-CoV) also have diarrhoea.19
Subgenomic RNA indicating viral replication was seen in faecal samples of patients with MERS.20 Moreover, MERS-CoV was shown to survive in simulated fed gastrointestinal juice and the ability to infect intestinal organoid models.20
Diarrhoea and gastrointestinal involvement are well known in coronavirus infections of animals and humans.21
On microbiological testing, we did not find any evidence of other known respiratory viral or bacterial infections, but specific RT-PCR assays for two widely separated genome targets—the highly conserved RdRp and the highly variable S genes—were positive for this novel 2019-nCoV.
Two complete genome sequences of this novel coronavirus were recovered from the nasopharyngeal swab of patient 2 and the sputum sample of patient 5 with an earlier cycle threshold value indicating a higher viral load.
Patient 2 had more underlying comorbidities and clinical features and radiological findings of more severe disease than the other patients included here. Moreover, the serum sample of patient 2 was also positive for 2019-nCoV, which might indicate some virus spillover from the more severely infected lung into the systemic circulation, as previously reported in patients with SARS.22
Sputum samples were available for testing from patients 5 and 7. The cycle threshold values of the sputum samples were 8–13 cycles earlier than those of throat swabs, indicating higher viral loads detected in the lower respiratory tract.
This finding is consistent with the observations in patients with MERS who had higher viral loads in lower respiratory tract samples than in upper respiratory tract samples.23 Thus, repeat testing of upper respiratory tract samples or testing of lower respiratory tract samples are warranted in clinically suspected cases with an initially negative result in nasopharyngeal or throat swab.
Unlike our patients in the 2003 SARS outbreak,22 we found no evidence of viral shedding in urine and faeces in these six patients. However, improved systematic serial collection and testing of an increased number of such samples is warranted.
Coronaviruses are enveloped, positive-sense, single-stranded RNA viruses, capable of rapid mutation and recombination.
They are classified into alphacoronaviruses and betacoronaviruses, which both have their gene source from bats and are mainly found in mammals such as bats, rodents, civets, and humans; and gammacoronaviruses and deltacoronaviruses, which both have their gene source from birds and are mainly found in birds.24, 25, 26
Phylogenetic analysis of the PCR amplicon fragments from five of our six patients and the complete virus genome of 29·8 kilobases from patients 2 and 5 showed that the virus is a novel betacoronavirus belonging to the lineage B or subgenus sarbecovirus, which also includes the human SARS coronavirus.
The genome of our virus strains are phylogenetically closest to the bat SARS-related coronaviruses first found in the Chinese horseshoe bats, Rhinolophus sinicus, captured in Zhoushan, Zhejiang province, China, between 2015 and 2017.27
Notably, the first SARS-related coronavirus was also discovered in the R sinicus found in Hong Kong, and central and south China in 2005.28, 29 The full virus genome had about an 89% nucleotide identity with bat-SL-CoVZC45, which makes it a new species.
Moreover, the Spike protein of our virus has an 84% nucleotide identity with the bat-SL-CoVZC45 coronavirus and an 78% nucleotide identity with the human SARS coronavirus. Although substantial genetic differences exist between this and other betacoronaviruses, cross reactions in RT-PCR or antibody assays for SARS or other betacoronaviruses are possible if the primers and antigenic epitopes are not carefully chosen, as previously reported.30
Further studies on the optimal diagnostic tests are warranted.In summary, an outbreak of novel coronavirus is ongoing at Wuhan in the winter of 2019–20.
Similar to the 2003 SARS outbreak in Guangzhou, Wuhan is also a rapidly flourishing capital city of the Hubei province and the traffic hub of central China.
Moreover, both outbreaks were initially connected to wet markets where game animals and meat were sold. In the case of SARS, person-to-person transmission was efficient and super-spreading events had led to major outbreaks in hotels and hospitals.
Learning from the SARS outbreak, which started as animal-to-human transmission during the first phase of the epidemic, all game meat trades should be optimally regulated to terminate this portal of transmission.
But as shown in this study, it is still crucial to isolate patients and trace and quarantine contacts as early as possible because asymptomatic infection appears possible (as shown in one of our patients), educate the public on both food and personal hygiene, and alert health-care workers on compliance to infection control to prevent super-spreading events.
Unlike the 2003 SARS outbreak, the improved surveillance network and laboratory capability of China was able to recognise this outbreak within a few weeks and announced the virus genome sequences that would allow the development of rapid diagnostic tests and efficient epidemiological control.
Our study showed that person-to-person transmission in family homes or hospital, and intercity spread of this novel coronavirus are possible, and therefore vigilant control measures are warranted at this early stage of the epidemic.