key: cord-0805732-bpyalbrc
authors: Vizcaychipi, M. P.; Shovlin, C. L.; Hayes, M.; Singh, S.; Christie, L.; Sisson, A.; Davies, R.; Lockie, C.; Howard, A.; Brown, A.; McCarthy, A.; Popescu, M.; Gupta, A.; Armstrong, J.; Said, H.; Peters, T.; Keays, R. T.; Consortium, ChelWest COVID-19
title: Early detection of severe COVID-19 disease patterns define near real-time personalised care, bioseverity in males, and decelerating mortality rates.
date: 2020-05-11
journal: nan
DOI: 10.1101/2020.05.08.20088393
sha: db26c52cd3a2ecf1133a945db24f40943eaba6ad
doc_id: 805732
cord_uid: bpyalbrc

BACKGROUND: COVID-19 is a global health emergency. Recent data indicate a 50% mortality rate across UK intensive care units. METHODS: A single institution, two-centre retrospective analysis following implementation of a Decision Support tool and real-time data dashboard for early detection of patients requiring personalised enhanced care, focussing particularly on respiratory rate, diastolic blood pressure, oxygenation indices, C-reactive protein, D-dimer and ferritin. Protocols differing from conventional practice included high-dose prophylactic anticoagulation for all COVID-19 positive patients and antioxidant prescription. RESULTS: By 22nd April 2020, 923 patients tested COVID-19 positive. 569 patients (61.7%) were male. The majority presented with advanced disease: interquartile ranges were C-reactive protein 44.9-179mg/L, D-dimer 1070-3802ng/L, and ferritin 261-1208g/L. Completed case fatality rates were 25.1% [95% CI 20.0, 30.0] in females, 40.5% [95% CI 35.9, 45.0] in males. 139 patients were admitted to intensive care where current death rates are 16.2% [95% CI 3.8, 28.7] in females, 38.2% [95% CI 28.6, 47.8] in males with no trends for differences based on ethnicity. A real-time traffic lights dashboard enabled rapid assessment of patients using critical parameters to accelerate adjustments to management protocols. In total 513 (55.6%) of patients were flagged as high risk for thromboembolic disease, exceeding the numbers flagged for respiratory deteriorations (N=391, 42.4%), or cytokine storm (N=68, 7.4%). There was minimal evidence that age was associated with disease severity, but males had higher levels of all dashboard indices, particularly C-reactive protein and ferritin (p<0.0001) which displayed no relationship with age. CONCLUSIONS: Survival rates are encouraging. Protocols employed (traffic light-driven personalised care, protocolised early therapeutic anticoagulation based on D-dimer >1,000ng/L and/or CRP>200 mg/L, personalised ventilatory strategies and antioxidants) are recommended to other units. Males are at greater risk of severe disease, most likely as the obligate SARS-CoV-2 receptor is on the X-chromosome, and require especially close, and early attention.

Human infection due to the novel coronavirus SARS-CoV-2 (commonly referred to as COVID- 19) was first reported at the end of 2019 [1] . The broad spectrum of disease severity ranges from asymptomatic and mild cases, to severe multiorgan failure and death. In early reports available from China [2] , particular features were of multilobar viral pneumonia, frequent requirement for supplementary oxygen/ ventilatory support, and very high transmission rates to healthcare workers. Subsequently affected countries have been able to benefit from the experience in Wuhan.

In the first 191 patients from Wuhan who had been discharged or had died by Jan 31, 2020, the crude death rate was 54/137 (39.4%) [2] . For intensive care, the initial experience in China was of 97% case fatalities following mechanical ventilation [2] , with the development of adult respiratory distress syndrome (ARDS) in more than half of critically ill patients [2] . Secondary bacterial infection, and acute kidney, cardiac and liver failure were also commonly reported. Risk factors for in-hospital death were older age, higher Sequential Organ Failure Assessment (SOFA) score, and D-dimer greater than 750ng/mL on admission) [2] .

Incorporation of these experiences into care pathways has led to reports of better, but still poor survival rates following hospitalisation. In the UK, the Intensive Care National Audit and Research Centre (ICNARC) recently reported 48.6% mortality in the first 5,139 patients with COVID-19 completing an Adult Intensive Care Units (AICU) admission [3] . This proportion was unchanged from earlier in the epidemic [4] . Similarly recent Italian and Washington State AICU reports indicate that for completed encounters, mortality was 61.2% and 67% mortality [5, 6] respectively.

These data prompted us to report the outcomes from a single institution which had implemented novel procedural and management protocols, aiming to secure better survival rates for those infected by COVID-19 and requiring hospital admission. 

The Chelsea and Westminster NHS Foundation Trust comprises two hospitals and 12 community-based clinics in North-West London, and delivers acute care to a population of over 1.5 million people with socioeconomic descriptors detailed in [7] . The Accident and Emergency services provided at Chelsea and Westminster Hospital and West Middlesex University Hospital treat over 300,000 patients a year.

A new cloud database was created in Microsoft Azure with a near real-time pipeline from electronic patient records (EPR) and dedicated to COVID-19 reporting from an operational, nationally mandated and clinical perspective. The database was kept in a permanent development cycle to facilitate rapid additions and changes. As the volume of data in both patient numbers and extracted clinical data grew, a limited scale data mining exercise was undertaken to flag any presenting readings with unusually high correlation with certain outcomes, and shared with the first author (MPV). For management of COVID-19 cases, all healthcare staff were supported with personal protective equipment (PPE) appropriate to the level of aerosolisation risk, according to institutional guidance. Captured results of blood tests spanned full blood count including red and white cell counts (with differential for e.g. lymphocytes, neutrophils, monocytes, eosinophils); haemoglobin, haematocrit, and platelet count; coagulation screen including fibrinogen, prothromin time and activated partial thromboplasin time (APTT); D-dimer, and anti-factor Xa where appropriate; electrolytes (sodium, potassium), renal function (urea, creatinine), liver function (bilirubin, alkaline phosphatase, albumin, aspartate transaminase, alanine transaminase, gamma glutamyl transferase (GGT), lactate dehydrogenase (LDH)), iron studies (serum iron, transferrin saturation index, ferritin); and additional markers of the acute phase response (C-reactive protein, fibrinogen).

Elements of therapeutics were also captured spanning ventilatory support including FiO2 (as above), . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 11, 2020. These patients underwent a computerised tomographic pulmonary angiogram (CTPA) to screen for an overt pulmonary embolus (PE) and a non-contrast computerised tomogram of the brain (CTB) using established institutional protocols, to exclude intracranial haemorrhage or an acute cerebral infarct, prior

to initiating systemic anticoagulation. In the absence of visible thrombus on CTPA, a microvascular thrombosis was assumed as previously reported [2] , and therapeutic anticoagulation usually commenced. However, if CTB identified an acute cerebral infarct which could be at risk of haemorrhagic transformation, anticoagulation was maintained at prophylactic dose.

Ventilation and oxygenation assessments: Dashboard markers of a high respiratory rate (> 35/min -1 );

SpO2 <92%, inspired fractional partial pressure of oxygen (FiO2) >60% or deterioration of haemodynamic function were integrated by the AICU Consultant in the emergency department.

Patients meeting the criteria received a 30 minute trial of CPAP in the emergency department. If the respiratory rate fell by 30% or more, CPAP was continued, otherwise patients were intubated and ventilated inmediately by a dedicated intubating team and transferred to AICU for invasive ventilation.

Adult Respiratory Distress Syndrome (ARDS) /Secondary Infection: This pattern was defined in ventilated patients with dashboard markers of high temperature (>38°C), FiO2 >60% with SpO2 <94%, positive end expiratory pressure (PEEP) of more than 10mmHg, CRP >250mg/L, requirements for noradrenaline to support the circulation, and a rising serum fibrinogen >5g/L. In this pattern, platelets were also reduced. Patients were noted to have a poor response to high FiO2, relatively good response to ventilation with prone positioning, but patients were noted to not tolerate the previously recommended conservative fluid balance for (COVID-19) ARDS [8] , with renal function deterioration. <35mmHg, CRP >250mg/L and ferritin >1,000μg/L were used to define patients at medium and high risk of a hyperinflammatory response. Patients meeting the criteria were monitored continuously and fluid management adjusted to replace insensible losses (1ml.kg -1 .h -1 for every degree above 38°C). If there was no response to initial fluid management and diastolic blood pressure remained <35mmHg, or oxygen requirement increased, then the patients were referred to the AICU for further management.

Patients who did not improve following a trial of CPAP, or meeting other criteria for AICU referral, were transferred to intensive care units at the two hospitals. There, they were intubated, ventilated using volume/pressure support, and managed according to conventional protocols, but with four additional modifications: 1) Ventilatory settings aimed for a target SpO2 92 % and generally low initial PEEP of 8cm water (H2O), as lung compliance was observed to be normal/high during the early phase of the disease process. 

Selected dashboard data including the real-time flags for VTE, ARDS, and cytokine storm were downloaded. Notably, for calculation of the population age deciles, all cases with the boundary age were assigned to the lower or upper decile which contained most of the values (none were equally distributed between two deciles).

Separately, to explore genotypic predisposition, DNA variants in the ACE2 gene were downloaded from the Genome Aggregation Consortium (gnomAD) [9] . This had mapped largely non-overlapping . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. . 

By 22 nd April 2020, 923 patients admitted to the two hospitals had tested positive for COVID-19 ( Figure   1 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. 

Whether evaluating all cases, or cases restricted to known outcomes only, males displayed worse outcomes than females (χ 2 p<0.0001, Figure 1B) 

From initial admission in the Emergency units, patient data were autopopulated into the hospital dashboard, and updated every 10 minutes. Real time-reports were sent to an experienced critical care clinician (MPV) for review. Results were interpreted and fed back to the clinical teams. Patterns identified allowed rapid modification of treatment pathways that were individualised to each patient, but particularly regarded anticoagulation, imaging, trial of continuous positive pressure ventilation, and early transfer to AICU for additional interventions (Figure 2) .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. . . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. 

Noting that males were more likely to die, we examined differences in biomarkers between males and for illustration: as shown in Figure 4 , the increase in CRP in males was sustained across all age groups.

Similar findings were observed for ferritin, illustrated in Supplementary Figure 1 .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. As indicated in Figure 4 , there was minimal association between age and CRP in either sex.

Furthermore, while there were physiological parameters that did differ according to the age of the patient, in the COVID-19 cohort there was no age-relationship with the core critical markers of CRP or ferritin, and only a marginal increase in male D-dimer values with age (Supplementary Table 1 , Supplementary Figure 1) .

Notably, all relationships with gender persisted after adjustment for age (Supplementary Table 1 ).

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. . have no "normal" ACE2 sequence ( Figure 5 ). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. 

We defined three critical disease groups by using a real-time, traffic lights dashboard that enabled rapid adjustment of management protocols. There was minimal evidence that age was associated with disease severity, but males were significantly more likely to have severe disease biomarkers, and to die, Given the current cohort were already at an advanced disease state on presentation, one reason for their generally good survival rates is likely to be the real time Clinical Decision Support tool that enabled rapid escalation of treatments, particularly relevant since the commonest critical scenario was thromboembolic disease for which anticoagulation is available: It was recognised early that the thrombotic process was most amenable to preventative and early treatment strategies, and more recent data have emphasised this: Overall thrombotic complication rates in intensive care patients are now reported as exceeding 30% [13] [14] [15] . Recent data from severely affected patients suggest the coagulopathy is not consistent with acute disseminated intravascular coagulation (DIC) and instead supports hypercoagulability together with a severe inflammatory state [16] . Additionally, autopsy series published in recent weeks highlight particularly the thrombosis of small vessels [17, 18] , and multiple groupings are now advising on how best to manage anticoagulation in COVID-positive patients [19] [20] [21] [22] [23] . Recognising evidence of microvascular injury [17, 18] , it is plausible that early, stratified intermediate prophylactic or full therapeutic anticoagulation, where appropriate, with heparin may have modified the disease process and led to improved outomes in our protocolised case series compared to all series [4-7; 12] , and low overall admission mortality rates for a cohort already at an advanced stage of disease. Furthermore, independent of heparin's anti-coagulant properties, it may exhibit important anti-viral properties. Lang and colleagues [24] have shown that heparin inhibits SARS-CoV infection in vitro and it has been shown to interact with the spike protein of SARS-CoV-2 causing conformational change, which may inhibit interaction with the ACE2 receptor and thereby block host cell entry [25] .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. The addition of an antioxidant in the form of ascorbic acid, for all patients admitted to hospital, could also have contributed to lower mortality in our case series. Although not shown to improve outcome in small sepsis [26] and ARDS [27] studies, there is growing interest in the use of ascorbic acid in sepsis [28] . Ascorbic acid has been shown to shorten the duration of the common cold and reduce hospital stay in influenza-A associated pneumonia. [29] In vitro studies have demonstrated that ascorbic acid enhances the proliferation, differentiation and function of lymphocytes, and stimulates interferon release, [30] which could enhance viral clearance.

There is substantial media discussion that older individuals, and people from ethnic minorities, are more at risk from COVID-19. While we cannot discount such signals, the evidence was not strong in the current cohort. Instead the most striking feature was the male predominance-in ward admissions, in admission to AICU, for biomarkers of severe disease (especially CRP and ferritin that displayed no relationships with age), and for deaths. We are not the first to report male susceptibility to severe COVID-19 [1-7], but the evidence for susceptibility to all aspects, across all age-groups, is noteworthy.

The genomic evidence from ACE2 adds to the compelling clinical data that emphasise the importance of increasing targeted and personalised care to males who are biologically more at risk of severe complications if infected by COVID-19. The presented biomarker data and excess mortality seen on AICU imply however, that even with the accelerated in-hospital care provided by the traffic light system, as a group, males are not benefitting as much as females. The nature of putative X-chromosome variants that modify the expression of the viral receptor are yet to be determined, thus for now we suggest that all males should be currently considered to be at higher risk.

In conclusion we define the three major patterns of COVID-19 mortality, present a traffic light system that enables early anticoagulation and other accelerated care pathways, and report encouraging survival rates in a hospitalised cohort admitted with advanced disease. We found no evidence that age was associated with biomarkers of disease severity-in contrast these emphasised the males at risk. We suggest the presented protocols are adapted more widely. Earlier presentation of males to hospital is also encouraged, in efforts to prevent the pathophysiological deterioration to which they are biologically susceptible.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 11, 2020. Bold face indicates the more abnormal variable between the sexes, and for p-values, those meeting p<0.0001. Females were older, there was no difference in the percentage of males and females by ethnicity, but all dashboard variables tended to be more abnormal in males, particularly CRP (illustrated in main manuscript Figure 4) , ferritin, creatinine (p-values <0.0001); respiratory rate and diastolic blood pressure (p<0.0001). BP, blood pressure, SpO2, oxygen saturation, IQR, interquartile range. P values for sex calculated by Mann Whitney rank sum. * Age dependent, SpO2 normal low 92% (for older patients), normal high 98.5%. The final columns indicates the p-values following exploratory multivariate regression analyses (regression coefficients not indicated for clarity). Superimposed for completed encounters, are whether the patient was discharged alive (green pipe), or died (red square). A) Admission CRP values including 95% confidence intervals [CI] for males and females, and illustrating higher values in males that becomes less pronounced at older age as female ferritin tends to increase. C) Natural log-transformed Ddimer. Note there was again no relationship with age for females (95% CI not shown for clarity), but that D-dimer increased with age in males. However, the regression coefficient for ln (D-dimer) predicted an averaged increase in the order of 10ng/mL per decade which we considered unlikely to have a bearing on the values under discussion.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 11, 2020. . https://doi.org/10.1101/2020.05.08.20088393 doi: medRxiv preprint

China Novel Coronavirus Investigating and Research Team. A Novel Coronavirus from Patients with Pneumonia in China

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Intensive Care National Audit and Research Centre (ICNARC)

Implications for COVID-19 triage from the ICNARC report of 2,204 COVID-19 cases managed in UK adult intensive care units

Baseline Characteristics and Outcomes of 1591 Patients Infected With SARS-CoV-2 Admitted to ICUs of the Lombardy Region

Characteristics and Outcomes of 21 Critically Ill Patients With COVID-19 in Washington State

PERSONALISED EARLY INTERVENTION COVID19 CARE PATHWAYS INFORMED BY PATTERNS OF CRITICAL DISEASE IN ACUTE HOSPITAL SETTINGS IDENTIFY BIOSEVERITY IN MALES, AND DECELERATING MORTALITY RATES. UPLOADED 03.05.2020

Geographical analysis of socioeconomic factors in risk of domestic burn injury in London

Surviving Sepsis Campaign: guidelines on the management of critically ill adults with Coronavirus Disease 2019 (COVID-19)

The Genome Aggregation Database Consortium

National Center for Biotechnology Information Genome Reference Consortium Human Build 38/hg38

Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the

Incidence of thrombotic complications in critically ill ICU patients with COVID-19

COVID-19-Related Severe Hypercoagulability in Patients Admitted to Intensive Care Unit for Acute Respiratory Failure

High incidence of venous thromboembolic events in anticoagulated severe COVID-19 patients

Hypercoagulability of COVID-19 patients in Intensive Care Unit. A Report of Thromboelastography Findings and other Parameters of Hemostasis

Vander Heide RS. Pulmonary and Cardiac Pathology in Covid-19: The First Autopsy Series

Complement associated microvascular injury and thrombosis in the pathogenesis of severe COVID-19 infection: A report of five cases

ISTH interim guidance on recognition and management of coagulopathy in COVID-19

Thromboprophylaxis and laboratory monitoring for in-hospital patients with COVID-19 -a Swiss consensus statement by the Working Party Hemostasis

Consensus Statement before Guidelines. Thromb Haemost

BTS Guidance on Venous Thromboembolic Disease in patients with COVID-19

Inhibition of SARS pseudovirus cell entry by lactoferrin binding to heparan sulfate proteoglycans

The 2019 coronavirus (SARS-CoV-2) surface protein

Receptor Binding Domain undergoes conformational change upon heparin binding

Bellomo R; VITAMINS Trial Investigators. Effect of Vitamin C, Hydrocortisone, and Thiamine vs Hydrocortisone Alone on Time Alive and Free of Vasopressor Support Among Patients With Septic Shock: The VITAMINS Randomized Clinical Trial

PERSONALISED EARLY INTERVENTION COVID19 CARE PATHWAYS INFORMED BY PATTERNS OF CRITICAL DISEASE IN ACUTE HOSPITAL SETTINGS IDENTIFY BIOSEVERITY IN MALES, AND DECELERATING MORTALITY RATES. UPLOADED 03.05.2020

Effect of Vitamin C Infusion on Organ Failure and Biomarkers of Inflammation and Vascular Injury in Patients With Sepsis and Severe Acute Respiratory Failure: The CITRIS-ALI Randomized Clinical Trial

The Emerging Role of Vitamin C as a Treatment for Sepsis

Colored precipitation reaction of the urine according to Kimbarowski as an index of the effect of ascorbic acid during treatment of viral influenza

Vitamin C and Immune Function