South African hospital outbreak: at least 119 infected from one emergency room patient; 15 of 39 hospital patients died: Science magazine

Durban, South Africa by Liesl Muhl courtesy of pixabay.com

Krisp.org published a story on May 22 (which is excerpted in Science magazine on May 25) about an outbreak in a Durban, South Africa hospital that started on March 9 with a patient who entered the emergency room with symptoms of COVID-19.  The infection jumped to another patient who was in the ED (emergency department) at the same time and was admitted to C-ICU (cardiac intensive care unit) with symptoms of a stroke.  From there, the virus spread, apparently mainly through “indirect contact and fomite transmission” to patients on at least five wards.  119 cases at the hospital (80 staff and 39 patients) were affected, and “smaller outbreaks” occurred at a nursing home (four cases) and an outpatient dialysis unit (nine patients and eight staff).  A total of up to 135 cases may have been seeded from one patient attending the emergency room.

Eighteen genomes were sequenced, and all had the same clade: type A2, associated with infections from Europe.  There was “limited genomic diversity” in these genomes, suggesting that all of them came originally from one infection.

Fifteen of the 39 patients infected died, all with multiple co-morbidities and all “elderly.”  “There was no evidence” that “any specific intervention” would have prevented their deaths once they were infected (none of the patients received remdesivir.)  The first opportunity for prevention of additional infections was when an 81 y/o woman (patient 3) (who turned out to have a “transient ischemic attack” (TIA), a precursor to stroke) developed an unexplained fever on March 13.  This woman was discharged from the medical ward to the nursing home on March 16 and readmitted with severe pneumonia on March 22.  During her time on the medical ward, she appears to have infected five other patients on the same ward, although she had no cough or other apparent symptoms except for the transient fever.

Around the same time, a 46 y/o woman (patient 4) (one of the five mentioned above) was admitted with an “acute respiratory illness” on March 21.  Two other patients and a nurse were infected by this patient.  On March 23, a nurse from the C-ICU (who had treated the TIA patient) was found to have confirmed COVID-19.

Patients 3 and 4 appear to have infected nine other patients when they were in the medical intensive care unit between March 22-27.  Patient 7, one of those nine, then infected eight other people on medical ward one.

Patient 8, from medical ward one, was moved to surgical ward one, where he infected two other patients and then died on April 1.  Patient 12 was moved from medical intensive care on March 31 (when it was closed because of the infection cluster) to the surgical intensive care unit, where he infected eight other patients.

A total of 1892 patients were recalled because they had been in the hospital in March; 191 were brought in for testing, and seven of these tested positive.  Some of these patients may have been infected outside the hospital, but it is also possible that some of the 1700+ patients who were not called in for testing had asymptomatic COVID-19 and were missed (that detail is not mentioned in the complete report.)

The case fatality rate among the confirmed COVID-19 cases who were patients (rather than staff) was 38.5%, with a median age at death of 79.  All of those who died had some co-morbidity, eleven with hypertension (high blood pressure), seven with diabetes mellitus (DM), and two with cancer.  None was known to be HIV positive.  In all but three cases, patients who died were not intubated because of their known poor prognosis at the time the decision was made.

A total of 1171 staff members were tested at least once, of whom 80 (roughly 5%) had positive tests.  The positive staff members were mostly from the medical intensive care unit and medical ward one; none of them came from the specialized COVID-19 intensive care unit, possibly due to the extra care taken there against infections.  All staff members eventually recovered; only fourteen were admitted to the specialized COVID-19 ward.

The nursing home involved took care of patient 3, and four other patients there developed COVID-19.  These were isolated in a separate building and no staff members at the home tested positive.  Nineteen COVID-19 cases (11 patients and eight staff) were found at the dialysis unit.  All patients and staff at the dialysis unit (133 and 36, respectively) were tested.  Five of the eleven cases in dialysis (45.5%) died.

A failure of personal protective equipment and infection control procedures (IPC) led to the spread of the virus to a total of 135 people, with fifteen deaths.  It should be noted that there were only two known patients in South Africa with COVID-19 at the time the first patient was seen.

The article recommends separation of patients at low and high risk for COVID-19, with separate entrances.  They recommend environmental cleaning to reduce fomite transmission, “aligned with national COVID-19 IPC guidelines and the national IPC framework manual”, plus monitoring of cleaning with fluorescent markers and visual re-inspection.

In addition,  they recommend physical distancing within the hospital with floor markings and signs.  Finally, they recommend weekly PCR (polymerase chain reaction) (from sputum or nasopharyngeal swabs?) testing of all frontline staff.

The only difficult ask in this country would be weekly testing of frontline hospital staff, but the recent increased availability of test kits should be beneficial here.  Additionally, the throat/sputum test recently developed should make testing much easier and quicker to implement– with no need for staff involvement in specimen collection and a much less invasive procedure than the nasopharyngeal swab that has been the exclusive method so far.

This tragic series of events should be studied by infection control professionals in all hospitals and the lessons learned should be implemented universally if a recurrence with multiple fatalities is to be prevented.

A .pdf file of the comprehensive report is available here and makes for very interesting reading.  It has details of how each infected person interacted with the previous person, thus how the virus was passed on from person to person.

The article states, “KRISP has been created by the coordinated effort of the University of KwaZulu-Natal (UKZN), the Technology Innovation Agency (TIA) and the South African Medical Research Countil [sic] (SAMRC).”  One of the authors is the director of the Nelson R Mandela School of Medicine at UKZN in Durban, South Africa.