Predictors of Mortality among Patients Hospitalized with COVID-19 during the First Wave in India: A Multisite Case-Control Study.

Severe acute respiratory syndrome coronavirus 2 disease (COVID-19) has caused more than 6 million deaths globally. Understanding predictors of mortality will help in prioritizing patient care and preventive approaches. This was a multicentric, unmatched, hospital-based case-control study conducted in nine teaching hospitals in India. Cases were microbiologically confirmed COVID-19 patients who died in the hospital during the period of study and controls were microbiologically confirmed COVID-19 patients who were discharged from the same hospital after recovery. Cases were recruited sequentially from March 2020 until December-March 2021. All information regarding cases and controls was extracted retrospectively from the medical records of patients by trained physicians. Univariable and multivariable logistic regression was done to assess the association between various predictor variables and deaths due to COVID-19. A total of 2,431 patients (1,137 cases and 1,294 controls) were included in the study. The mean age of patients was 52.8 years (SD: 16.5 years), and 32.1% were females. Breathlessness was the most common symptom at the time of admission (53.2%). Increasing age (adjusted odds ratio [aOR]: 46-59 years, 3.4 [95% CI: 1.5-7.7]; 60-74 years, 4.1 [95% CI: 1.7-9.5]; and ≥ 75 years, 11.0 [95% CI: 4.0-30.6]); preexisting diabetes mellitus (aOR: 1.9 [95% CI: 1.2-2.9]); malignancy (aOR: 3.1 [95% CI: 1.3-7.8]); pulmonary tuberculosis (aOR: 3.3 [95% CI: 1.2-8.8]); breathlessness at the time of admission (aOR: 2.2 [95% CI: 1.4-3.5]); high quick Sequential Organ Failure Assessment score at the time of admission (aOR: 5.6 [95% CI: 2.7-11.4]); and oxygen saturation < 94% at the time of admission (aOR: 2.5 [95% CI: 1.6-3.9]) were associated with mortality due to COVID-19. These results can be used to prioritize patients who are at increased risk of death and to rationalize therapy to reduce mortality due to COVID-19.

High Prevalence of Fungal and NDM-OXA Producing Gram-Negative Bacterial Superinfections in the Second Wave of Coronavirus Disease 2019 in India: Experience from a Dedicated Coronavirus Disease 2019 Hospital in North India.

Introduction: During the second wave of coronavirus disease 2019 (COVID-19), superinfection caused by fungus and multidrug-resistant bacteria worsened the severity of illness in COVID-19 patients. Limited studies from India reported the antimicrobial resistance pattern of secondary infections. In this study, we aim to study the epidemiology of pathogens causing superinfections and genotyping of Gram-negative isolates in COVID-19 patients. Methods: This retrospective study was conducted at a dedicated COVID-19 center, India. The identification of bacteria/fungi was done by Vitek2® and matrix-assisted laser desorption/ionization-time of flight mass spectrometry system. Identification of beta-lactamase genes was done using thermal cycler. The diagnosis of mucormycosis was based on 10% potassium hydroxide direct microscopy. Statistical analyses were performed using STATA version 15.1 (StataCorp., College Station, TX, USA). For continuous variables, mean and standard deviation were computed. For comparing proportions of secondary infections across admission location and outcomes, the Chi-squared test of independence was used. To compare the mean and median between intensive care units and outcomes, an independent t-test and a Mann-Whitney test were used. Results: Of all the clinical samples, 45.4% of samples were cultured positive for secondary infections. Acinetobacter baumannii (35%) was the most common Gram-negative pathogen, while among Gram positive, it was Enterococcus faecium (40%). Among fungus, Candida spp. (61%) predominates followed by molds. Colistin and tigecycline proved effective against these pathogens. blaNDM was the most prevalent gene followed by the blaOX among the carbapenemase genes. Conclusions: The mortality rate among COVID-19 patients with secondary infection was significantly higher compared to the overall mortality rate in COVID-19 patients.

Rate of shed of SARS COV-2 viral RNA from COVID-19 cadavers.

BACKGROUND: At what rate does the RNA of SARS CoV-2 shed from cadavers? Although, there have been numerous studies which have demonstrated the persistence of the virus on dead bodies, there is a lack of conclusive evidence regarding the variation of viral RNA content in cadavers. This has led to a knowledge gap regarding the safe handling/management of COVID-19 decedents, posing a barrier in forensic investigations. METHODS: In this study, we report the presence of RNA of SARS CoV-2 by real time RT-PCR, in nasopharyngeal swabs collected after death from two groups of bodies - one who died due to COVID-19 and the other who died due to other diagnoses. A prospective study on 199 corpses, who had tested positive for COVID-19 ante-mortem, was conducted at a tertiary care center. RNA testing was conducted at different time intervals (T1-T5). RESULTS: 112(56.3%) died primarily due to COVID-19 and 87(43.7%) died due to other diagnoses. 144(72.4%) were male and 55(27.6%) were female. A total of 115 (57.8%) tested positive for COVID-19 after death at different time points. The mean age was 50.7 ± 18.9 years and the length of hospitalization ranged from 1 to 50 days with a mean of 9.2 ± 7.6 days. Realtime RT-PCR positivity of SARS CoV-2 RNA decreases with time. CONCLUSION: We observed that real time RT-PCR positivity, indicating viral RNA detection, decreases with time. Therefore, it is advisable to follow appropriate COVID-19 precautions to carry out scientific studies, medico-legal investigations and mortuary services on suspected/confirmed COVID-19 corpses.

Secondary Bacterial Infections in Mucormycosis-COVID-19 Cases: Experience during the Second COVID-19 Wave in India.

In the second wave of COVID-19 in India, there was a new challenge in the form of mucormycosis. Coinfection with mucormycosis was perilous as both conditions required a prolonged hospital stay, thus serving as an ideal platform for secondary infections. Using a retrospective observational study, we studied secondary infections and their impact on the outcome in COVID-19 patients with mucormycosis. The outcome in these patients was evaluated and compared with COVID-19 patients with mucormycosis but without any secondary infection. SPSS V-20 was used for data analysis. Fifty-five patients tested positive for mucormycosis (55/140; 39.28). Twelve out of these 55 (21.8%) developed secondary infections during their hospital stay. Bloodstream infection was the most common (42.86%) secondary infection. The Gram-negative (GN) organisms were more common (11/16; 68.75%) compared with the Gram-positives (GP) (5/16; 31.25%). But the most common isolate was Enterococcus faecium (5/16; 31.25%). A high percentage of microorganisms isolated were multidrug-resistant (15/16; 93.75%). Two out of five (40%) isolates of Enterococcus faecium were vancomycin-resistant (VRE). High resistance to carbapenems was noted in the GN isolates (9/11; 81.81%). The comparison of length of stay in both subgroups was statistically significant (P value <0.001). When compared, the length of stay in people with adverse outcomes was also statistically significant (P value <0.001). Procalcitonin (PCT) had a positive predictive value for the development of secondary bacterial infections (P value <0.001). Antimicrobial stewardship and strict infection control practices are the need of the hour. IMPORTANCE Although our knowledge about COVID-19 and secondary infections in patients is increasing daily, little is known about the secondary infections in COVID-19-mucormycosis patients. Thus, we have intended to share our experience regarding this subgroup. The importance of this study is that it brings to light the type of secondary infections seen in COVID-19-mucormycosis patients. These secondary infections were partially responsible for the mortality and morbidity of the unfortunate ones. We, as health care workers, can learn the lesson and disseminate the knowledge so that in similar situations, health care workers, even in other parts of the world, know what to expect.

Association between Hypocalcemia and Outcome in COVID-19 Patients: A Retrospective Study.

Background Calcium has been shown to play a vital role in the pathophysiology of severe acute respiratory syndrome-coronavirus-2 and middle east respiratory syndrome coronavirus diseases, but less is known about hypocalcemia in coronavirus disease 2019 (COVID-19) patients and its association with the disease severity and the final outcome. Therefore, this study was conducted with an aim to assess clinical features in COVID-19 patients having hypocalcemia and to observe its impact on COVID-19 disease severity and the final outcome. Methods In this retrospective study, consecutive COVID-19 patients of all age groups were enrolled. Demographical, clinical, and laboratory details were collected and analyzed. On the basis of albumin-corrected calcium levels, patients were classified into normocalcemic ( n = 51) and hypocalcemic ( n = 110) groups. Death was the primary outcome. Results The mean age of patients in the hypocalcemic group was significantly lower ( p < 0.05). A significantly higher number of hypocalcemic patients had severe COVID-19 infection (92.73%; p < 0.01), had comorbidities (82.73%, p < 0.05), and required ventilator support (39.09%; p < 0.01) compared with normocalcemic patients. The mortality rate was significantly higher in the hypocalcemic patients (33.63%; p < 0.05). Hemoglobin ( p < 0.01), hematocrit ( p < 0.01), and red cell count ( p < 0.01) were significantly lower with higher levels of absolute neutrophil count (ANC; p < 0.05) and neutrophil-to-lymphocyte ratio (NLR; p < 0.01) in the hypocalcemic patients. Albumin-corrected calcium levels had a significant positive correlation with hemoglobin levels, hematocrit, red cell count, total protein, albumin, and albumin-to-globulin ratio and a significant negative correlation with ANC and NLR. Conclusion The disease severity, ventilator requirement, and mortality were considerably higher in hypocalcemic COVID-19 patients.

Variation in Laboratory Utilization and Correlation with Hospital Bed Utilization: Experience of a Trauma-Care Hospital during the COVID-19 Pandemic

Objectives  The present study was planned with the following objectives: (i) to calculate the difference in frequency of laboratory test ordered and use of consumables between the prepandemic and pandemic phases, (ii) to determine and compare the monthly average number of tests ordered per patient between the prepandemic and pandemic phases, and (iii) to correlate the monthly test ordering frequency with the monthly bed occupancy rate in both phases. Materials and Methods  Records of laboratory tests ordered and use of consumables were collected for the prepandemic phase (1.8.2019 to 31.3.2020) and the pandemic phase (1.4.2020 to 31.10.2020). The absolute and relative differences were calculated. Monthly average number of tests ordered per patient and bed occupancy rate between prepandemic and pandemic phases was determined, compared, and correlated. Statistical Analysis  The absolute and the relative differences between the two periods were calculated. The continuous variables were analyzed between groups using Mann–Whitney U test. Spearman correlation was used to correlate the monthly test ordering frequency with the monthly bed occupancy rate in both phases. Results  A total of 946,421 tests were ordered, of which 370,270 (39%) tests were ordered during the pandemic period. There was a decrease in the number of the overall laboratory tests ordered (12%), and in the use of blood collection tubes (34%), and an increase in the consumption of sanitizers (18%), disinfectants (3%), masks (1633%), and gloves (7011%) during the pandemic period. Also, the monthly average number of tests ordered per patients significantly reduced ( p -value < 0.001). Test ordering frequency had strong positive correlation with bed occupancy rate during pandemic (Spearman co-efficient = 0.73, p -value = 0.03). Conclusions  An overall decline in laboratory utilization during pandemic period was observed. Understanding and correlating the trends with hospital bed utilization can maximize the productivity of the laboratory and help in better preparedness for the challenges imposed during similar exigencies.

Clinicoepidemiological Features and Mortality Analysis of Deceased Patients with COVID-19 in a Tertiary Care Center.

BACKGROUND AND OBJECTIVE: A large number of studies describing the clinicoepidemiological features of coronavirus disease-2019 (COVID-19) patients are available but very few studies have documented similar features of the deceased. This study was aimed to describe the clinicoepidemiological features and the causes of mortality of COVID-19 deceased patients admitted in a dedicated COVID center in India. METHODOLOGY: This was a retrospective study done in adult deceased patients admitted in COVID ICU from April 4 to July 24, 2020. The clinical features, comorbidities, complications, and causes of mortality in these patients were analyzed. Pediatric deceased were analyzed separately. RESULTS: A total of 654 adult patients were admitted in the ICU during the study period and ICU mortality was 37.7% (247/654). Among the adult deceased, 65.9% were males with a median age of 56 years [interquartile range (IQR), 41.5-65] and 94.74% had one or more comorbidities, most common being hypertension (43.3%), diabetes mellitus (34.8%), and chronic kidney disease (20.6%). The most common presenting features in these deceased were fever (75.7%), cough (68.8%), and shortness of breath (67.6%). The mean initial sequential organ failure assessment score was 9.3 ± 4.7 and 24.2% were already intubated at the time of admission. The median duration of hospital stay was 6 days (IQR, 3-11). The most common cause of death was sepsis with multi-organ failure (55.1%) followed by severe acute respiratory distress syndrome (ARDS) (25.5%). All pediatric deceased had comorbid conditions and the most common cause of death in this group was severe ARDS. CONCLUSION: In this cohort of adult deceased, most were young males with age less than 65 years with one or more comorbidities, hypertension being the most common. Only 5% of the deceased had no comorbidities. Sepsis with multi-organ dysfunction syndrome was the most common cause of death. HOW TO CITE THIS ARTICLE: Aggarwal R, Bhatia R, Kulshrestha K, Soni KD, Viswanath R, Singh AK, et al. Clinicoepidemiological Features and Mortality Analysis of Deceased Patients with COVID-19 in a Tertiary Care Center. Indian J Crit Care Med 2021; 25(6):622-628.

Clinical Validation of Standard Q COVID-19 Antigen and IgM/IgG Combo Kit Assay at a Tertiary Care Center in Northern India.

Background Expansion of the testing capacities for severe acute respiratory syndrome-coronavirus-2 is an important issue in the face of ever-increasing case load. So, there is need of point-of-care diagnostic tests in the existing laboratory capacities for early treatment, isolation, and clinical decision making, especially in resource limited settings. Materials and Methods This prospective cohort study was conducted at Jai Prakash Narayan Apex Trauma Center, All India Institute of Medical Sciences, New Delhi. Nasopharyngeal samples and blood samples were collected for antigen and antibody testing. Rapid antigen test was performed as per the kit's instructions. The performance of the kit was compared with the gold standard reverse transcription polymerase chain reaction (RT-PCR) testing. Results Eighty-eight out of 110 patients tested positive by RT-PCR for coronavirus disease 2019 in last 48 to 72 hours were included in the study. Overall, the sensitivity of combined antibody test was 52%, antigen test 26%, and combined sensitivity of both antigen and antibody was 72.7%, respectively. Conclusion The combo kit needs to be used with caution in low prevalence settings, where cases may be missed.

Lockdown imposition due to COVID-19 and its effect on orthopedic emergency department in level 1 trauma center in South Asia.

Background: The coronavirus pandemic brought the entire world to a standstill. One of the most stringent lockdowns in the world was implemented in India. With the entire healthcare system being stretched, emergency orthopaedic services also take a hit. We studied the trends in patient presentation, testing, management, and restructuring of doctors at a tertiary care orthopaedic centre and compared them with the data from the same time period the previous year (2019). Method: Data was collected separately for all the 5 different phases of lockdown and unlock, as well as for the same duration of months in 2019, and was analysed for epidemiological trends. Results: A rapid fall in the total number of cases was seen during the lockdown, followed by a skewed rise during the unlock. Forearm, wrist, and hip fractures were the most common fractures. Once nucleic acid testing of all patients intended to be admitted was started, a steep rise in coronavirus positivity was seen. There was a reduction in the total number of cases compared to 2019, but it was not as significant as would have been expected due to the complete standstill of activity during the lockdown. Conclusion: During a pandemic, with the healthcare system under a crisis of workforce and infrastructure, there needs to be a separate task force for catering to orthopaedic emergencies since all fractures cannot be managed conservatively and the numbers of trauma-related patients did not show a stark fall as compared to normal months of last year. Level of evidence: Level 3 Retrospective Case Series.

Comparison of Abbott ID NOW, a novel isothermal amplification based COVID-19 diagnostic method with RTPCR.

BACKGROUND: The emergent crisis of the COVID-19 pandemic has posed enormous challenges for clinical laboratories to speed up diagnostics. The current reference standard for the diagnosis of COVID-19 is real time reverse transcriptase PCR on various platforms. However, even with automation, the turnaround time is huge enough to keep up with ever increasing numbers of patients. With increasing surge of COVID cases we need rapid diagnostic tests with good sensitivity and specificity. OBJECTIVES: Comparison between Abbott ID NOW COVID-19 and real time reverse transcriptase PCR as a reference method. MATERIALS AND METHODS: Specimens from seventy-two individuals were obtained over a period of two months which were processed for ID NOW and RTPCR at a dedicated COVID-19 centre of AIIMS. Dry nasal swabs were used for ID NOW while nasopharyngeal swabs along with throat swab were used for RTPCR. Among the participants, 15 were healthcare workers. Mild COVID was seen in 36 participants, moderate in 19 and severe in 9. Eight participants had non COVID illness. RESULTS: From the given samples, we observed that ID NOW has a sensitivity of 93.22% (55/59) specificity 100% (13/13), PPV 100% (55/55) and NPV 76.47% (13/17). CONCLUSION: ID NOW is a convenient, rapid molecular test which makes it suitable for both in laboratory use and as a point of care test. It can be a rapid rule-in test for COVID-19. Negative results, however, have to be interpreted as per the context.

Clinical Presentation and Course of SARS-CoV-2 Infection in Health-Care Personnel Working in Dedicated COVID-19 Hospital During 2 Pandemic Waves in India.

INTRODUCTION: Health-care personnel (HCPs) are predisposed to infection during direct or indirect patient care as well as due to the community spread of the disease. METHODS: We observed the clinical presentation and course of severe acute respiratory syndrome coronavirus disease 2 (SARS-CoV-2) infection in HCPs working in a dedicated coronavirus disease 2019 (COVID-19) care hospital during the first and the second wave. RESULTS: A total of 100 and 223 HCPs were enrolled for the first wave and the second wave, respectively. Cough, shortness of breath, sore throat, runny nose, and headache was seen in 40 (40%) and 152 (68%) (P < 0.01), 15 (15%) and 64 (29%) (P = 0.006), 40 (40%) and 119 (53.3%) (P = 0.03), 9 (9%) and 66 (30%) (P < 0.01), 20 (20%) and 125 (56%) (P < 0.01), respectively. Persistent symptoms at the time of joining back to work were seen in 31 (31%) HCPs and 152 (68%) HCPs, respectively (P ≤ 0.01). Reinfection was reported in 10 HCPs. CONCLUSIONS: Most of the HCPs had mild to moderate infections. Symptoms persist after joining back to work. Upgradation of home-based care and teleconsultation facilities for active disease and redressal of residual symptoms will be helpful.

Variation in Laboratory Utilization and Correlation with Hospital Bed Utilization: Experience of a Trauma-Care Hospital during the COVID-19 Pandemic.

Objectives The present study was planned with the following objectives: (i) to calculate the difference in frequency of laboratory test ordered and use of consumables between the prepandemic and pandemic phases, (ii) to determine and compare the monthly average number of tests ordered per patient between the prepandemic and pandemic phases, and (iii) to correlate the monthly test ordering frequency with the monthly bed occupancy rate in both phases. Materials and Methods Records of laboratory tests ordered and use of consumables were collected for the prepandemic phase (1.8.2019 to 31.3.2020) and the pandemic phase (1.4.2020 to 31.10.2020). The absolute and relative differences were calculated. Monthly average number of tests ordered per patient and bed occupancy rate between prepandemic and pandemic phases was determined, compared, and correlated. Statistical Analysis The absolute and the relative differences between the two periods were calculated. The continuous variables were analyzed between groups using Mann-Whitney U test. Spearman correlation was used to correlate the monthly test ordering frequency with the monthly bed occupancy rate in both phases. Results A total of 946,421 tests were ordered, of which 370,270 (39%) tests were ordered during the pandemic period. There was a decrease in the number of the overall laboratory tests ordered (12%), and in the use of blood collection tubes (34%), and an increase in the consumption of sanitizers (18%), disinfectants (3%), masks (1633%), and gloves (7011%) during the pandemic period. Also, the monthly average number of tests ordered per patients significantly reduced ( p -value < 0.001). Test ordering frequency had strong positive correlation with bed occupancy rate during pandemic (Spearman co-efficient = 0.73, p -value = 0.03). Conclusions An overall decline in laboratory utilization during pandemic period was observed. Understanding and correlating the trends with hospital bed utilization can maximize the productivity of the laboratory and help in better preparedness for the challenges imposed during similar exigencies.

Has the time to reinstate elective orthopedic procedures come?

BACKGROUND: The coronavirus pandemic changed how we manage and operate patients in orthopaedic practice. Although elective orthopaedic procedures were halted to prevent spread of the disease as well as sustain supplies of essential protective equipment and healthcare workers, trauma services were continued. We studied the orthopaedic trauma cases operated over 6 months of the pandemic, and discuss the protocols used to minimize disease spread. METHODS: Data was collected for all orthopaedic emergency cases operated at our centre from 1 st March - 10 th August 2020. During this time specific protocols were used for first aid, pre-operative care, inside the operation theatre, post-operative stay as well as for follow ups. RESULTS: A total of 851 patients were operated. A sharp decline in surgeries was seen during the lockdown. Average stay in the hospital was 4 days. Only 44% of the patients came for follow-up visits. None of the contacted patients or their relatives developed symptoms or tested positive for COVID after discharge. CONCLUSION: Multiple waves and various mutant strains of COVID-19 have made this pandemic longer than expected. Elective orthopaedic cases cannot be ignored for forever, as it leads to poor quality of life and an increasing burden of such patients. We suggest, that using the protocols used at our centre, we have successfully operated on cases without risking spread of the virus. Thus, we believe it's time to reinstate elective orthopaedic procedures, in a phased manner.

The rise in preanalytical errors during COVID-19 pandemic.

INTRODUCTION: The COVID-19 pandemic has posed several challenges to clinical laboratories across the globe. Amidst the outbreak, errors occurring in the preanalytical phase of sample collection, transport and processing, can further lead to undesirable clinical consequences. Thus, this study was designed with the following objectives: (i) to determine and compare the blood specimen rejection rate of a clinical laboratory and (ii) to characterise and compare the types of preanalytical errors between the pre-pandemic and the pandemic phases. MATERIALS AND METHODS: This retrospective study was carried out in a trauma-care hospital, presently converted to COVID-19 care centre. Data was collected from (i) pre-pandemic phase: 1st October 2019 to 23rd March 2020 and (ii) pandemic phase: 24th March to 31st October 2020. Blood specimen rejection rate was calculated as the proportion of blood collection tubes with preanalytical errors out of the total number received, expressed as percentage. RESULTS: Total of 107,716 blood specimens were screened of which 43,396 (40.3%) were received during the pandemic. The blood specimen rejection rate during the pandemic was significantly higher than the pre-pandemic phase (3.0% versus 1.1%; P < 0.001). Clotted samples were the commonest source of preanalytical errors in both phases. There was a significant increase in the improperly labelled samples (P < 0.001) and samples with insufficient volume (P < 0.001), whereas, a significant decline in samples with inadequate sample-anticoagulant ratio and haemolysed samples (P < 0.001). CONCLUSION: In the ongoing pandemic, preanalytical errors and resultant blood specimen rejection rate in the clinical laboratory have significantly increased due to changed logistics. The study highlights the need for corrective steps at various levels to reduce preanalytical errors in order to optimise patient care and resource utilisation.

Evaluation of COVID-19 Antigen Fluorescence Immunoassay Test for Rapid Detection of SARS-CoV-2.

INTRODUCTION: Tests detecting SARS-CoV-2-specific antigen have recently been developed, and many of them are now commercially available. However, the real-world performance of these assays is uncertain; therefore, their validation is important. In this study, we have evaluated the performance of STANDARD F COVID-19 antigen fluorescence immunoassay (FIA) kit. METHODS: Nasopharyngeal samples collected from patients were subjected to the test as per manufacturer's instructions. The performance of the kit was compared with the gold standard real-time polymerase chain reaction. RESULTS: A total of 354 patients were tested with STANDARD F COVID-19 antigen FIA test kit. The overall sensitivity, specificity, positive predictive value, and negative predictive value of this test were found to be 38%, 99%, 96.2%, and 72%, respectively, with a diagnostic accuracy of 75.7%. CONCLUSION: STANDARD F COVID-19 antigen FIA showed high specificity and positive predictive value but low sensitivity and negative predictive value.

Profile of co-infections & secondary infections in COVID-19 patients at a dedicated COVID-19 facility of a tertiary care Indian hospital: Implication on antimicrobial resistance.

BACKGROUND: The COVID-19 pandemic has raised concerns over secondary infections because it has limited treatment options and empiric antimicrobial treatment poses serious risks of aggravating antimicrobial resistance (AMR). Studies have shown that COVID-19 patients are predisposed to develop secondary infections. This study was conducted to ascertain the prevalence and profiles of co- & secondary infections in patients at the COVID-19 facility in North India. METHODS: We studied the profile of pathogens isolated from 290 clinical samples. Bacterial and fungal pathogens were identified, and antimicrobial susceptibility was determined by the Vitek2® system. Additionally, respiratory samples were tested for any viral/atypical bacterial co-infections and the presence of AMR genes by FilmArray test. The clinical and outcome data of these patients were also recorded for demographic and outcome measures analyses. RESULTS: A total of 151 (13%) patients had secondary infections, and most got infected within the first 14 days of hospital admission. Patients aged >50 years developed severe symptoms (p = 0.0004) and/or had a fatal outcome (p = 0.0005). In-hospital mortality was 33%.K.pneumoniae (33.3%) was the predominant pathogen, followed by A. baumannii (27.1%). The overall resistance was up to 84%.Majority of the organisms were multidrug-resistant (MDR) harbouring MDR genes. CONCLUSION: A high rate of secondary infections with resistant pathogens in COVID-19 patients highlights the importance of antimicrobial stewardship programs focussing on supporting the optimal selection of empiric treatment and rapid-de-escalation, based on culture reports.

Artificial Intelligence-assisted chest X-ray assessment scheme for COVID-19.

OBJECTIVES: To study whether a trained convolutional neural network (CNN) can be of assistance to radiologists in differentiating Coronavirus disease (COVID)-positive from COVID-negative patients using chest X-ray (CXR) through an ambispective clinical study. To identify subgroups of patients where artificial intelligence (AI) can be of particular value and analyse what imaging features may have contributed to the performance of AI by means of visualisation techniques. METHODS: CXR of 487 patients were classified into [4] categories-normal, classical COVID, indeterminate, and non-COVID by consensus opinion of 2 radiologists. CXR which were classified as "normal" and "indeterminate" were then subjected to analysis by AI, and final categorisation provided as guided by prediction of the network. Precision and recall of the radiologist alone and radiologist assisted by AI were calculated in comparison to reverse transcriptase-polymerase chain reaction (RT-PCR) as the gold standard. Attention maps of the CNN were analysed to understand regions in the CXR important to the AI algorithm in making a prediction. RESULTS: The precision of radiologists improved from 65.9 to 81.9% and recall improved from 17.5 to 71.75 when assistance with AI was provided. AI showed 92% accuracy in classifying "normal" CXR into COVID or non-COVID. Analysis of attention maps revealed attention on the cardiac shadow in these "normal" radiographs. CONCLUSION: This study shows how deployment of an AI algorithm can complement a human expert in the determination of COVID status. Analysis of the detected features suggests possible subtle cardiac changes, laying ground for further investigative studies into possible cardiac changes. KEY POINTS: • Through an ambispective clinical study, we show how assistance with an AI algorithm can improve recall (sensitivity) and precision (positive predictive value) of radiologists in assessing CXR for possible COVID in comparison to RT-PCR. • We show that AI achieves the best results in images classified as "normal" by radiologists. We conjecture that possible subtle cardiac in the CXR, imperceptible to the human eye, may have contributed to this prediction. • The reported results may pave the way for a human computer collaboration whereby the expert with some help from the AI algorithm achieves higher accuracy in predicting COVID status on CXR than previously thought possible when considering either alone.