key: cord-0731003-52m27gnx
authors: Claassen, Nadè; van Wyk, Gerhard; van Staden, Sanet; Basson, Michiel M.D.
title: Experiencing COVID-19 at a large district level hospital in Cape Town: A retrospective analysis of the first wave
date: 2022-03-29
journal: S Afr J Infect Dis
DOI: 10.4102/sajid.v37i1.317
sha: b8a6a2ce5664042b3005eed1ebd20d03fb79d228
doc_id: 731003
cord_uid: 52m27gnx

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic in tertiary hospitals from South Africa and world wide have been well described, but limited data are published on the findings. This article aimed to describe patients admitted to a large district hospital in Cape Town, South Africa, during the first wave of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Infections. To compare the clinical features and further investigate survivors and deceased COVID-19 patients. METHODS: A single centre retrospective review of clinical records and laboratory data of patients admitted with a positive SARS-CoV-2 polymerase chain reaction (PCR) from April 2020 to August 2020. RESULTS: A total of 568 patients with a positive SARS-CoV-2 PCR were admitted to the study centre for one night or longer and of these patients 154 (27%) died of COVID-19. The median age of patients who died of COVID-19 was 66 years and 53 years for survivors. Hypertension, diabetes mellitus and obesity were the commonest comorbidities in patients who survived and died of COVID-19. There were no major differences when comparing the severity of infiltrates on chest X-rays (CXR) of COVID-19 survivors with deceased patients. More than half (58%) of deceased patients died within 3 days following admission to hospital. A substantial number of patients who died of COVID-19 had associated acute kidney injury (n = 79, 51%). CONCLUSION: Acute kidney injury had a high prevalence amongst patients who died of COVID-19. Delays in transfer to intensive care unit (ICU), limited ICU capacity and disease severity contributed to a substantial number of patients dying within 3 days of admission.

Since the advent of the outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) worldwide, the lives of more than five million individuals have been lost because of coronavirus disease 2019 (COVID- 19) . 1 Worldwide the case fatality rate is estimated to be 2.0% -3.4%. 1 Models estimate the South African case fatality rate to be comparable with the rate reported in the rest of the world. The under-reporting of cases and deaths, limited availability of testing in rural areas and different testing strategies are challenges impeding the calculation of a more accurate case fatality rate. When looking at the South African Medical Research Council data on excess deaths since the start of the pandemic, the official South African death rate seems to be significantly underestimated. 2, 3 The COVID-19 pandemic has brought the spotlight on sub-Saharan Africa and what the response will be in a population already under the heavy burden of HIV and AIDS, TB, and non-communicable diseases. With primary and district healthcare services having even more limited resources and bed capacity, 8, 9, 10, 11 mechanisms had to be put in place to ensure the appropriate management and referral of patients with COVID-19 who are critically ill. This also raised concern for primary care platforms servicing people living with HIV, TB and non-communicable diseases being side-tracked by an overwhelming number of COVID-19 cases.

Much has been learned in terms of disease process and therapeutics since the first wave and several factors associated with severe COVID-19 have been identified. Amongst these factors are demographic factors (male gender, smoking status and increasing age), biochemical and radiological parameters and the sequential organ failure assessment (SOFA) score. 12, 13, 14, 15, 16, 17, 18 Certain comorbidities are also associated with increased risk of severe COVID-19 and risk of death, 12, 13, 14, 15, 16, 17 with knowledge regarding the interplay between diseases such as HIV and AIDS and associated opportunistic infections and COVID-19 growing by the day. 15, 17, 19, 20, 21, 22, 23 The Western Cape and specifically the Cape Town metropole was the area hit the hardest in the initial stages of the pandemic in South Africa, its peak being towards the end of June 2020. 24 The province's COVID-19 response was met with several challenges and triumphs, but despite a robust contingency plan in place, the lives of thousands of patients were lost to The subsequent second and third waves of SARS-CoV-2 infections were predominated by the new and more transmissible Beta (B.1.351) and Delta (B.1.617.2) variants, which are also known to cause more severe disease in infected individuals. 25, 26, 27 With a fourth wave of COVID-19 infections looming, only a small fraction of the South African population vaccinated against COVID-19 and new SARS-CoV-2 variants arising rapidly, it is uncertain for how long the pandemic will continue to cause death and devastation around the world. 28, 29 We compare a cohort of patients who died of COVID-19 (either because of direct effects or consequences of disease) with COVID-19 survivors in terms of clinical, radiological and biochemical parameters.

This was a single centre study that included 573 admissions to a large district level hospital in Cape Town, South Africa, during the so-called first wave of SARS-CoV-2 infections. The study hospital services the Northern as well as a part of the Tygerberg substructures of Cape Town, which has a total adult population of 870 826. However, a part of the Tygerberg Substructure drains directly to the Tygerberg Academic Hospital (TBH) and an unknown portion of the population used services provided by the private healthcare sector. The true population size reflected in this cohort is therefore uncertain.

The study site is a district level hospital with 282 beds and a 4-bed high care unit (HCU). The COVID-19 cases admitted to the hospital were primarily managed by the hospital's department of internal medicine. This department was staffed with 2 full time specialist physicians and 11 medical officers and a registrar in internal medicine at the time of this study.

The study included all patients aged 18 years or older with a positive SARS-CoV-2 polymerase chain reaction (PCR) test who were admitted to hospital from 01 April 2020 to 31 August 2020. Severe acute respiratory syndrome coronavirus 2 antigen tests were not yet available for use during the study period. All patients admitted during the study period were included in this comparison, including the large number of patients who were transferred to intermediate care facilities (for further management once stable and acute complications requiring specialist care were ruled out or treated) as well as patients who were transferred to facilities providing a higher level of care or palliative care. Patients were separated into two groups, namely survivors (those who were discharged alive from either the base hospital or the hospital they were transferred to) and COVID-19 deaths (those patients who died during their admission to either the base or referral hospital).

Two burden of disease surveys were carried out during the year 2020 -one just before the first wave of the COVID-19 pandemic in February 2020 and one during the peak of the first wave in June 2020. Data from these surveys were used to assess and compare the number of admissions and admission diagnoses other than COVID-19.

Patients were identified using the hospital's admissions registry for the duration of the study period. Inclusion criterion was all patients aged 18 years or older with a positive SARS-CoV-2 PCR admitted to the hospital for one night or longer regardless of the reason for admission. Patient outcomes were checked to determine the number of deaths, discharges and transfers to other facilities using the admission lists, the electronic continuity of care record (eCCR) and Clinicom patient administration system. All patients who were discharged alive (whether from the study centre or transfer hospital) were included in the survivor analysis group and all patients who died of COVID-19 during their admission (whether at the study centre or at hospitals elsewhere after being transferred) were included in the COVID-19 death analysis group.

Data were extracted from patient files and laboratory records, using a standardised form, including demographic details (age, gender and race), comorbidities (hypertension, diabetes mellitus, HIV infection, active TB, chronic lung disease, chronic kidney disease, cardiovascular disease, dyslipidaemia, connective tissue disease, chronic neuromuscular disease, active malignancy, smoking, obesity, etc.), timing of onset of symptoms to admission and death, clinical features on admission (pulse oximetry, whilst breathing ambient air where available), blood results (full blood and differential counts, serum creatinine, C-reactive protein), complications (acute kidney injury [AKI], acute arterial occlusion, congestive cardiac failure, diabetic ketoacidosis (DKA), haematological abnormalities, healthcare associated infection, hypernatraemia and dehydration, hypoglycaemia and venous thromboembolism) and whether or not corticosteroids were administered.

Blood results were interpreted according to National Health Laboratory Services reference values.

Comorbidities were further assessed to get an indication of disease control. Patients with diabetes mellitus were assessed by means of glycated haemoglobin (HbA1c) carried out during admission or within four months prior to admission. Target organ damage (if present) was specifically mentioned. Pharmacy records were audited to assess how many antihypertensive agents patients were on. Patients with HIVinfection were assessed in terms of their most recent CD4+ lymphocyte count, whether they were on antiretroviral medication and their viral loads. Patients were classified as obese or not according to the admitting physician's clinical impression. Acute kidney injury was defined according to the Kidney Disease Improving Global Outcomes (KDIGO) Clinical practice guidelines' definition of AKI. 30 Data from two of the Department of Internal Medicine's burden of disease surveys during the months of February 2020 and June 2020 were included to assess and compare admission diagnoses and patient numbers. This survey is an audit of all inpatients admitted to the internal medicine service. Patients were identified using the hospital's admissions registry that provided information on primary and secondary diagnoses, patient outcomes and length of stay.

Due to the nature of the study, consent was not obtained. Patient confidentiality was ensured by means of anonymising data.

Data were collected and analysed using Microsoft Excel. Normally distributed descriptive numerical data were described using means and standard deviations (s.d.) and non-normally distributed data were described using medians and interquartile ranges (IQRs). Continuous data were compared using the t-test for data with a normal distribution and the Mann-Whitney U test for data with a non-normal distribution. Odds ratios (ORs) (unadjusted) were calculated using bivariate analysis to assess association between exposure and outcome. Statistical significance was set at p < 0.05 and confidence interval of 95% was used.

The study was submitted and approved by the South African Medical Association Research and Ethical Committee (SAMAREC). The SAMAREC is associated with the South African Medical Association.

The department's two burden of disease surveys during February 2020 and June 2020 were compared to assess the different reasons for admission to the study centre's internal medicine service ( Figure 1 ). From this comparison it is COVID-19, coronavirus disease 2019. Reason for admission Of the 154 patients who died the majority were mixed race (45%; n = 69) and black people (38%; n = 59), followed by white people (16%; n = 24) and one Asian. The race of one patient was not clear.

Amongst patients admitted to the study centre with COVID-19, the most common comorbidities included hypertension, diabetes mellitus, obesity, cardiovascular disease, smoking, chronic lung disease, chronic kidney disease and HIV-infection (Table 1) . It is our impression that certain comorbidities such as obesity was under-reported as weight and height was not consistently documented on admission and due to the variability in the clinical impression and note keeping of treating doctors. 

Patients were assessed in terms of clinical and radiological features on admission with specific mention of saturation whilst breathing ambient air and findings of plain chest X-rays (CXR). Patients who died of COVID-19 (n = 110) had a median saturation of 85% (IQR: 75-92) whether whilst breathing ambient air or receiving supplemental oxygen. The admission saturation levels of survivors were not reviewed in this study.

The plain CXR of patients admitted at the study centre were classified as having minimal/no infiltrates, having basal and/or peripheral infiltrates (< 50% of lung fields), diffuse bilateral infiltrates (> 50% of lung fields) or having findings not typical of COVID-19 (Table 2) . Furthermore, CXRs were classified as having signs of pre-existing structural lung disease and cardiomegaly.

Commonly used special investigations were recorded and compared with that of patients admitted to the study centre who survived COVID-19 during the study period.

The baseline laboratory investigations of admitted patients who survived and died of COVID-19 were compared (Table 2) .

Timelines from symptom onset to testing, admission and date of death were recorded for the patients who died of COVID-19. The median (IQR) time from onset of symptoms to being tested was four days (IQR: 2-7) (n = 127), median (IQR) from onset of symptoms to admission five days (IQR: 2-7) (n = 127) and median (IQR) time from onset of symptoms until death was seven days (IQR: 5.0-12.5; n = 127) ( Figure 3 ). More than 58% of the 154 patients (n = 90) died within 3 days of being admitted to hospital.

Complications whilst in hospital were identified by auditing clinical notes and laboratory data. It is our impression that complications might be under-reported as the data reflect note keeping. The most common documented complication was AKI. Laboratory data were audited to demonstrate an acute deterioration in renal function from baseline using the KDIGO definition of AKI defined by at least 1.5-fold increase in serum creatinine (SCr), increase in SCr of 26.5 µmol/L or more or urine output less than 0.5 mL/kg per hour for 6 h. 30 

All patients admitted to the study centre were treated according to standard of care with supplemental oxygen as required to maintain adequate oxygen saturation, antipyretics, analgesia and low molecular weight heparin in prophylactic or therapeutic doses, which was individualised on a patient-to-patient basis, unless specifically contraindicated. At the time of this study, although there was still uncertainty regarding the benefit of using thiamine and (Table 3 ).

CD4+ count - - 134 30-248 - - N/A - - - - - N/A - N/A WHO clinical stage - - - - 3.2 0.8 N/A - - - - - N/A - N/A On ART (% of HIV-infected) 17 94.0 - - - - N/A - - - - - N/A - N/A

This retrospective study revealed valuable information about the history, clinical features and disease course amongst patients who died of COVID-19 in a large district level hospital in Cape Town, South Africa during the first wave of SARS-CoV-2 infections.

What is striking in the comparison of the two burden of disease surveys between February 2020 and June 2020 is the similar amounts of admissions for routine internal medicine diagnoses but with an extra 205 COVID-19 cases during the month of June 2020. It seems the burden of internal medicine diagnoses (other than COVID-19) did not decrease or plateau during the peak of the first COVID-19 wave despite the so-called level five lockdown and theoretical hesitancy for patients to seek medical help in fear of being exposed to SARS-CoV-2 at healthcare facilities. Another finding is the notable increase in cases admitted for hyperglycaemic emergencies during June 2021 independent of COVID-19 (Table 3) . Amongst patients who died of COVID-19 the prevalence of non-communicable diseases such as hypertension, diabetes mellitus and obesity were higher than those who survived COVID-19. Having any two or more of the above-mentioned comorbidities carried an OR ranging between 2.7 and 3.6 ( Table 3 ). The variability in history taking, note keeping and clinical impression of doctors treating patients with COVID-19 proved to be a challenge when recording variables such as comorbidities and complications from clinical records. We suspect that the prevalence of certain important comorbidities such as obesity, hypertension and smoking history may have been under-reported. Even though there was suspected under-reporting of acute complications in hospitalised patients, AKI seems to have been very prevalent amongst patients who died of COVID-19 and amongst the study population the presence of acute renal failure carried a high OR of dying from COVID-19 (Table 3) .

When comparing the CXRs of those patients who died with those who survived COVID-19, there did not seem to be a significant difference in the extent of infiltrates. A remarkable number of patients died very shortly following admission. This can be because of various theories including severity of disease and timing of (late) presentation to hospital. Another factor possibly contributing to early death was the limited capacity for escalation of care offered at district level. Several Upon the publication of the RECOVERY trial on 16 June 2020 the use of systemic glucocorticoid therapy were included in national clinical guidelines. 31 The addition of glucocorticoid therapy was followed by a reduction in the case fatality ratio amongst patients admitted with COVID-19 from before (29.7%; n = 70) to after (25.3%; n = 84) 16 June 2020. Although small study numbers may limit definitive conclusions in terms of therapeutic benefit, plausible explanations for a decrease in mortality after the introduction of glucocorticoid therapy include direct benefit from treatment with systemic glucocorticoids and advances in therapeutic and intervention strategies as the pandemic progressed.

This was a retrospective study and data sets were not complete for all variables as they were dependent on the history taking, clinical impressions, diagnostic skills and note keeping of the clinicians who managed these patients. An example is patients being classified as overweight or not, as all patients did not have their weight and height measured on admission.

We suspect that the cause for the disparity in the data were due to the under-reporting of acute complications and comorbid diseases. Even though all possible care has been taken to identify all cases admitted to the study centre with COVID-19, the possibility exists that the total number of COVID-19 cases is under-reported because of lack of a robust system where discharge diagnoses are recorded.

This study describes the experiences of a district level hospital during the first wave of the COVID-19 pandemic.

Given the small study population, incomplete data sets and numerous variables, limited definitive conclusions can be drawn from this study.

Patients who died of COVID-19 seemed to have similar comorbidities that are associated with increased risk of death. Amongst the study population, AKI seemed to have been very prevalent in patients who died of COVID-19. As a result of challenges with accurate and consistent note keeping and lack of standardisation leading to inter-clinician variability of clinical impressions, some comorbidities and complications seemed under-reported.

Apart from late presentation to hospital, limited ICU beds and logistical reasons causing delays in escalation of care contributed to a significant number of patients dying of COVID-19 very shortly after admission to the study hospital.

Worldometer coronavirus

Report on weekly deaths in South Africa

Report on weekly deaths in South Africa

From brain drain to brain gain: Understanding and managing the movement of medical doctors in the South African health system

African Institute for Health and Leadership Development

Africa Check puts together the numbers on doctor-patient ratios

Medical Brief

Poverty, inequality and COVID-19: The forgotten vulnerable. Public Health

Non-communicable diseases in South Africa: A challenge to economic development

Primary health care in South Africa

Building on shaky ground' -Challenges to and solutions for primary care guideline implementation in four provinces in South Africa: A qualitative study

Primary health services at district level in South Africa: A critique of the primary health care approach

Challenges of quality improvement in the healthcare of South Africa post-apartheid: A critical review

Prognostic factors associated with mortality risk and disease progression in 639 critically ill patients with COVID-19 in Europe: Initial report of the international RISC-19-ICU prospective observational cohort. EClinicalMedicine [serial online

Predictors of adverse prognosis in COVID-19: A systematic review and meta-analysis

Age and multimorbidity predict death among COVID-19 patients. Hypertension [serial online

Risk factors for COVID-19-related in-hospital mortality in a high HIV and tuberculosis prevalence setting in South Africa: A cohort study

Difference in mortality among individuals admitted to hospital with COVID-19 during the first and second waves in South Africa: A cohort study

Risk factors for coronavirus disease 2019 (COVID-19) death in a population cohort study from the Western Cape Province, South Africa

Risk factors and prognosis for COVID-19-induced acute kidney injury: A meta-analysis

High HIV prevalence in an early cohort of hospital admissions with COVID-19 in Cape Town

HIV and SARS-CoV-2 co-infection: The diagnostic challenges of dual pandemics

Coronavirus disease 2019 and Pneumocystis jirovecii pneumonia: A diagnostic dilemma in HIV

Epidemiology and outcomes of COVID-19 in HIV-infected individuals: A systematic review and meta-analysis. Sci Rep

Clinical outcomes of patients with COVID-19 and HIV coinfection

Western Cape Department of Health. Western Cape COVID-19 Dashboard

Hospital admission and emergency care attendance risk for SARS-CoV-2 delta (B.1.617.2) compared with alpha (B.1.1.7) variants of concern: A cohort study

COVID-19 update: Delta variant in South Africa

Letter from South Africa -COVID-19: The good, the bad and the ugly

South African Government: Department of Health

World Health Organization. Classification of Omicron (B.1.1.529): SARS-CoV-2 variant of concern

KDIGO clinical practice guideline for acute kidney injury. Kidney Int Suppl

Dexamethasone in hospitalized patients with Covid-19 -Preliminary report

The authors would like to thank Michiel Heyns Claassen who assisted with data collection with his Excel skills and Dr. Zirkia Joubert for assisting with conceptualisation and the use of her office and computer.

The research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

The data that support the findings of this study are available on reasonable request from the corresponding author, N.C.

The views and opinions expressed in this article are those of the authors and do not reflect the official policy or position of Internal Medicine Department at Karl Bremer Hospital.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.