Impact of the COVID‐19 outbreak on blood supply in two large university hospitals

This study aimed to examine the relationship between the decrease in elective procedures and the need for blood donation during the novel coronavirus disease (COVID‐19) pandemic at university hospitals.


| INTRODUCTION
The coronavirus disease (COVID- 19), first reported in December 2019 in Wuhan, Hubei province, China, is an virus caused disease leading to an acute respiratory pathology. 1,2On January 7, 2020, the causative agent of this respiratory illness was identified as the novel severe acute respiratory syndrome COVID-19. 3The first case in Germany was reported in Bavaria on January 27, 2020. 4At that time, 4593 people had already been infected worldwide, and more than 100 had died from the viral infection. 5 Germany, the first two deaths were reported on March 8, 2020, and March 11, 2020, when the World Health Organisation (WHO) declared the disease a pandemic. 5,6Twelve days later, the German government reacted to the increasing case numbers with drastic contact bans and severe restrictions on social gatherings.
Infection numbers increased exponentially, and by this time, 18 610 people in Germany had been infected and 55 had died owing to COVID-19. 5e pathophysiology of COVID-19 manifests in a broad clinical spectrum, from an asymptomatic course to acute respiratory failure. 2 In 85%-90% of cases, contracting COVID-19 results in mildto-moderate symptoms.In affected individuals, hospital treatment is usually not required; accordingly, anaemia or thrombocytopenia requiring transfusion is not expected.However, a severe clinical course from an acute respiratory distress syndrome (ARDS) to multi organ failure occurs in 10% of COVID-19 infected cases, and anaemia or thrombocytopenia requiring transfusion is more common. 7,8As is common with other ribonucleic acid (RNA) viruses, the COVID 19 virus has adapted and evolved mutations.Over time, different variants of concern have been identified by the World Health Organisation. 9The different variants have different characteristics in terms of transmissibility, clinical presentation and the effectiveness of detection methods and therapeutics.The first variant is Alpha variant.This variant was 43%-82% more transmissible than the previously existing variant.In addition, infection with this variant showed an increased case fatality rate. 10,11In late 2020, a new variant, the beta variant, was reported.This also showed an increased risk of transmission and reduced neutralisation by monoclonal antibody therapy and convalescent plasma. 12 a result, some countries, such as Italy, quickly experienced high utilisation of intensive care beds and a large influx of patients requiring inpatient care. 13This in turn led to a rapid overload of the healthcare system in several countries in early 2020.To prevent hospital overload, on March 13, 2020, the German Federal Ministry of Health urged hospitals to stop all elective procedures that were not medically urgent. 14 taking this step, the Ministry of Health hoped to preserve resources for treating patients with COVID-19.It was feared that the pandemic could have a significant impact on blood donations and that there would be an increased need for donated blood for severely ill patients (e.g., cancer patients) and patients with COVID-19 infection. 15 drastically reducing elective surgeries at university hospitals, donated blood should have been needed less, thereby increasing its availability for patients with COVID-19.However, no studies have indicated that the measure at university hospitals actually led to the saving of blood products, and thus a potential drop in blood donations could be intercepted.The need for personnel in blood donation facilities and donated blood supply can be determined only with a better understanding of this relationship.Therefore, our study examined the relationship between the decrease in elective procedures and the need for blood donation during the COVID-19 pandemic at two exemplarily university hospitals.

| Study setting
This study was conducted at university hospitals in Essen and Hamburg, Germany.These are two maximum care hospitals that focus on the treatment of patients with haematological and cardiovascular pathologies. 16,17At these hospitals, approximately 80% of platelet (PLT) transfusions occur in these two focus areas.The University Medical Centre Hamburg-Eppendorf treats approximately 497 000 patients annually and is the largest and most important healthcare provider in the northern region, with a population of approx. 2 million. 18,19The University Hospital in Essen treats approximately 340 000 patients annually and plays an important role in healthcare provision in the Ruhr metropolitan area, which has a population of approx.5 million. 20,21In both hospitals, the Institute of Transfusion Medicine is responsible for blood products' planning, regulation, production, and delivery.

| Study design
The study was approved by the Institutional Ethics Committee of the Medical Faculty of the University Duisburg-Essen.
The data were collected at two department of transfusion medicine at Essen University Hospital and University Medical Centre Hamburg-Eppendorf which produce approximately 1% of all red blood cells (RBCs) and 4% of all PLT in Germany. 22is retrospective study investigated the number of packed RBC and PLT transfusions and requests for blood group typing and crossmatches during the first phase of the COVID-19 pandemic in Germany compared with the same period in 2019.Data acquisition started on February 3, 2020 (week 1), 1 week after the first confirmed case was reported in Germany.The study period lasted 26 weeks and ended on August 2, 2020 (week 26).At that time, the nationwide infection rate averaged fewer 1000 cases per day for 12 weeks.The following week, there was a significant increase in the number of cases, indicating the beginning of a new phase of the pandemic. 23e number of cross-match requests and blood group typing were determined using our laboratory information system.The total number of RBC and PLT transfusions was compiled from our laboratory information system.Additionally, a specific analysis was performed at the Clinic for Haematology and Stem Cell Transplantation, Trauma Surgery, and Thoracic Surgery in Essen, as most blood products are transfused in these departments.In relation to the Essen site, 54.5% of all blood transfusions take place in the aforementioned departments of the university hospital.
The case mix index (CMI) for all clinical departments and institutes at the university hospitals were compiled from our hospital database.The CMI represents a comparative value and valuation value for the patient mix of a hospital and aims to represent the average severity of treated cases. 24,25ta of patients treated in hospitals were obtained from the central hospital database.The University Hospital Essen is a specialised centre for extracorporeal membrane oxygenation (ECMO) therapy in Metropole Ruhr.More than 42% of patients who were COVID-19-positive with ARDS were admitted from other hospitals for ECMO therapy. 26Therefore, we analysed and compared ECMO cases in Essen between 2020 and 2019.

| Statistical analysis
The data were provided and analysed in Excel spreadsheets (Microsoft Office 2016, Microsoft Corp., Redmond).Statistical analyses and graphical representations were performed using the Graph-Pad Prism 8.4 (GraphPad Software, San Diego).Data were reported as the number of laboratory tests performed and blood products dispensed per week during the same study period in 2019 and 2020.For statistical comparisons between the periods an unpaired sample t-test was used.We calculated the CMI from the sum of all relative weights from the diagnosis-related groups (DRGs) and divided this result by the total number of treatment cases.The graphical representation was in the form of bar charts for 2019 and 2020.Data on the number of blood products administered during ECMO therapy were reported as the median with the associated confidence interval.

| RESULTS
The direction to stop elective procedures by the German government was adhered to by all institutes and departments of the investigated hospitals.Thus, from March 16, 2020 (the start of the seventh week of investigation), all non-urgent surgical procedures were postponed.This affected the overall operational caseload at both sites, particularly in March and April 2020.Table 1 shows the surgical case numbers of different specialties at University Hospital Essen.Only urgent medical procedures were performed at both sites.The reduction in elective cases led not only to a decrease in surgical cases at both hospitals but also to a significant decrease in the total number of outpatient and inpatient cases and bed utilisation (Table 2).
Due to the drop in caseload, we observed a dramatic year-on-year decline in crossmatch applications in our study in the following weeks, beginning in the seventh study week.However, the number of blood transfusions, especially of packed RBC and PLT, was reduced to a far lesser extent (Figure 1).Between weeks 7-10 of the study, we observed a significant decrease in crossmatch determination in both Essen (19%, p < 0.0005) and Hamburg (10%, p < 0.05) compared to the values in 2019.We observed an even greater decrease in both Essen (34%, p < 0.0005) and Hamburg (31%, p < 0.0005) in blood group determinations compared with the previous year.Despite the lower need for crossmatching and blood grouping, erythrocyte consumption at both hospitals did not decrease significantly (Essen, p = 0.24 and Hamburg, p = 0.73), and PLT consumption increased compared to the previous year during the study period (Figure 1).After 4 weeks, from April 13, 2020, scheduled procedures at the two clinics resumed after the number of COVID-19 infections in Germany showed a declining trend, and an overload on the healthcare system could be avoided.As the local incidence in Germany declined, elective procedures resumed from April 13, 2020   (start of study week 11), and there was a significant increase in case  numbers (Table 2).Weekly crossmatch determination was similar to that in the previous year.Moreover, there was an increase in the number of blood group determinations; however, this number was lower than that of the previous year.The number of transfused blood products was similar to that of before (study week 10), and RBC and PLT transfusions significantly increased compared to 2019.
To determine why blood product consumption hardly changed despite the cancellation of elective interventions, but the number of requests for crossmatch and blood grouping increased, a more detailed analysis was conducted in Essen for Haematology and Stem Cell Transplantation, thoracic surgery, and trauma surgery (Figure 2).
Blood product consumption remained almost unchanged among the departments examined.The thoracic surgery department consumed more erythrocytes and PLT in the period under review than in the previous year.
To determine whether fewer severe and complex cases were treated at the two hospitals after the cessation of elective procedures compared to 2019, we examined the CMI of the two hospitals.In both Essen and Hamburg, the CMI increased in 2020 compared with 2019, even after elective procedures were stopped, despite lower numbers of inpatients, outpatients, and operative cases.(Table 3) During the study period, with the increasing number of COVID-19 cases in Germany, there was also an increased admission of patients with COVID-19 at both hospitals (data not shown).As Essen specialised as a centre for ECMO therapy for patients with ARDS and COVID-19 in the region, there was a significant increase in ECMO cases in 2020 compared to that in 2019 (Figure 3).In 2019, 75 patients required ECMO therapy.Of these, 71% were men and 29% were women and their average age was 54 years.In 2020, 295 patients underwent ECMO therapy.Of these, 67% were men and 33% were women.Their average age was 55 years.Blood products are highly consumed during ECMO. 27,28In 2019, the median value of required units of RBCs for patients registered for ECMO treatment was 11 (0-170), for 2020 the median value was 12 (0-116).In 2019 the median value of required units of PLT was 1 (0-36) and for 2020 the median value was 0 (0-54; Figure 3).On average, there was no major difference in the number of blood products administered compared with that in the previous year; however, the consumption of blood products increased significantly owing to the increased number of ECMO cases.In summary, this study showed that the number of crossmatches and blood group determinations at both sites decreased significantly over a period of 4 weeks after the cessation of all elective procedures, but there was no significant difference in the consumption of blood products during the same period.

| DISCUSSION
Cessation of elective procedures at the hospitals in Germany led to a significant decrease in outpatients, inpatients, and surgical cases. 29,30is was confirmed from our study (Table 2).
The COVID-19 pandemic had a major impact not only on the economic and social systems but also on the healthcare system, including Weekly crossmatch and blood group determinations over the period of 26 weeks in the Clinic for Haematology and stem cell transplantation at the Essen University Hospital, (B) in vascular surgery, and (C) in orthopaedic surgery.In the area highlighted in grey, the elective procedures were suspended and postponed.
on the supply of donated blood.Our study revealed a decrease in crossmatches and blood group requests during the first phase of the pandemic (Figure 1).However, at the same time, the consumption of packed RBCs and PLT did not change significantly.Compared with the previous year, there was only a slight decrease in the transfusion of RBCs and a modest increase in that of PLTs.
We observed this trend for 4 weeks after the elective procedures were stopped in hospitals.Table 1 shows that in the 4 weeks under review, the number of surgical cases decreased significantly.In January 2020, there was already a decrease in surgical cases compared to 2019.This was due to the fact that the first staff members fell ill with COVID-19 already in this month and thus some surgeries had to be cancelled.But also on the patient side, operations had to be cancelled because of the first COVID illnesses.However, the decrease in surgical cases was never more than 25% compared to the previous year.The effect of stopping elective surgeries is particularly evident in the months of March and April, where the decline was sometimes more than 50%.In the following months of May to June, the number of surgical cases still fell slightly, since, as is already known from the literature, many patients postponed operations or visited medical facilities less frequently for fear of contracting the virus.From July 2020 onwards, the number of cases increased slowly, as the first COVID-19 wave showed a decline and many patients resumed surgery appointments in the summer months.With the outbreak of the second COVID-19 wave at the end of 2020, the operative case numbers also decreased again from October 2020.Hardly any fluctuations can be seen in trauma surgery.
2][33][34] A study from Italy has reported that fewer blood products were needed during March to May 2020. 35ere was a restructuring of blood donation to ensure fewer preparations, and consequently, there was less wastage.
We assume that the continued increase in the proportion of transfused units during the study period correlated with the size of the hospital and its patient spectrum.Our study indicates that suspension of elective procedures at larger institutions had a limited impact on the use of donated blood (Figure 1).In our opinion, the continuous or even increased consumption of RBCs and PLT (Figure 2) reflects transfusion for patients in haematology, transplantation, and  cardiothoracic surgery; and for those who were severely ill with COVID-19 and in need of ECMO.In these scenarios, the ratio of transfusion-to-crossmatch was higher than that in elective surgery.
We observe an increase in crossmatch requirements in Essen at the beginning of 2020 (Figure 1).There is a simple explanation for this.At the beginning of 2020, a new partner hospital with a large cardiothoracic surgery unit was added.This partner hospital was then also supplied by the blood donation centre of the university hospital from January 1, 2020.Consequently, the number of crossmatches has increased significantly with the new supply contract.
7][38] Thus, we interpret the decrease in crossmatches and blood group determination as a result of the suspension of elective surgeries.
Certainly, the consumption of blood products is impacted not only the patients who are chronically and acutely ill and treated at major hospitals during the pandemic, but also those who have COVID-19 infection and require ECMO therapy.ECMO treatment is sometimes necessary for patients who are critically ill with the COVID-19 infection and is associated with a higher bleeding tendency and an increased need for packed RBCs and PLT. 27,28In our study, we were able to demonstrate that ECMO therapy is associated with a significantly increased need for erythrocyte and PLT preparations, independent of the COVID-19 infection (Figure 3).Consequently, the increase in the number of ECMO cases has increased the demand for blood products.
The CMI confirmed the assumption that large hospitals continued to treat critically ill and medically complex cases during the pandemic.
This suggests that significantly more severe cases were treated in 2020 than in 2019, as the CMI indicates case severity and the high technical effort of treatments. 39Especially in the 4 weeks after the stop of elective procedures, a significant increase of the CMI was shown in Essen as well as in Hamburg (Table 3).Therefore, a decrease in the number of laboratory tests cannot be explained by case severity.
Other countries had also ceased elective surgeries and medical procedures, with an associated decrease in blood transfusions. 15,33,40 were unable to determine whether the decrease in elective procedures was associated with a significant decrease in the use of donated blood.However, we observed a significant decrease in the requests for blood group determination and cross-matches.This indicates that, even during the pandemic, seriously ill and medically complex cases were treated at large institutions.Thus, we conclude that during the COVID-19 pandemic, transfusion medicine, especially at large institutions, faces an unchanged demand for donated blood, which should be considered during personnel and blood donation allocations.
The cessation of elective procedures led to a reduction in laboratory requirements, and thus relief in the daily work of the staff, which was important in this period considering the increased sick leave among the staff due to contracting the COVID-19 infection.Considering the demographic changes and the relative shortage of blood donors leading to the shortage of stored blood products, in the future, even without a pandemic, the requirements of cross-matching should be handled in a more restrained manner, especially for scheduled surgeries. 41,42This is because the lower rate of crossed RBC preparation implies that fewer products need to be stocked, and therefore potential discards can be reduced.In contrast to other studies, there was a continued high consumption of blood products at both sites.As a consequence, we developed an automated monitor, that allows real-time monitoring of stored blood products with a latency of less than 1 min.
A simple graphical representation provides an overview of the available and patient-specific products.Quick and easy viewing of in-stock and expiring blood products can help optimise the management of this valuable resource.Further studies are required to determine the extent to which this new automated technology can reduce blood product deterioration.This would be interesting from economic, medical, and ethical points of view, considering the possible shortage of blood products in the near future. 41,42is study had a few limitations.It was conducted in only two large maximum care hospitals.As previous reports from small hospitals tended to show a decrease in consumption, additional research should be conducted in larger hospitals to confirm our assumptions.
Consumption and number of laboratory requests were similar at both sites.However, significantly more erythrocyte concentrates were produced at the Hamburg site than at Essen, which implies that the impact of new automated techniques, such as the one we developed, could differ between the sites.This influence was not investigated in our study and should be evaluated in future studies.

| CONCLUSIONS
By stopping elective admissions and interventions, the German government aimed to ease the burden on its healthcare system.This study shows that this has a limited impact on large hospitals that treat complex and critical cases regarding transfusion medicine.
Although it was possible to reduce the number of laboratory requirements and thus, create some relief, the consumption of blood products in large institutions remained largely unchanged and presented the institutes of transfusion medicine with a major challenge regarding both the personnel and organisation during this period.These challenges will become even greater in the future owing to demographic changes and shortage of blood products, and the use of modern and automated techniques can be of great help in solving this problem.

F
I G U R E 1 (A) Weekly crossmatch and blood group determinations over the period of 26 weeks at the Essen University Hospital and (B) at the University Medical Center Hamburg-Eppendorf.(C) Weekly transfused units of red blood cells (RBC) and platelets (PLT) at the Essen University Hospital and (D) at the University Medical Center Hamburg-Eppendorf.In the area highlighted in grey, the elective procedures were suspended and postponed.

T A B L E 3
Case Mix Index (CMI) and percentage change for the study period of 26 weeks at the University Hospitals of Essen and Hamburg for the years 2019 and 2020.

F
I G U R E 3 (A) Number of extracorporeal membrane oxygenation (ECMO) cases at Essen University Hospital in 2019 and 2020.(B) Plot of median value of red blood cell (RBC) and platelet (PLT) concentrate units required per ECMO case.In addition to the representation of the median value, a 95% confidence interval is also displayed.
Operative case numbers at the Essen University Hospital in 2019 and 2020.Inpatient and outpatient case numbers and bed utilisation in 2019 and 2020 in Essen.
aIn trauma surgery, there was no change in the number of operative procedures in March and April 2020 compared with that in the same period in 2019.BRIESKE ET AL.T A B L E 2 aThe bed utilisation rate is shown as a percentage of all beds are occupied by patients.