Use of Convalescent Plasma Therapy among Hospitalized Coronavirus Disease 2019 (COVID-19) Patients: A Single-Center Experience

ABSTRACT: COVID-19 Disease has strained our healthcare system. Convalescent plasma has been used to treat emerging infectious diseases - Influenza A/B, SARS-CoV, Ebola virus and now SARS-CoV 2. OBJECTIVE: This study aims to determine the outcome and clinical course of COVID-19 patients who received convalescent plasma transfusion at Cardinal Santos Medical Center. METHODS: This is a retrospective cohort analytical study of 75 patients who received convalescent plasma. RESULTS: Median time from admission to CP transfusion was 3 days. Majority of patients received additional therapies including dexamethasone (100%); Remdesivir (95%); antibiotics (100%), tocilizumab (65%); hemoperfusion (88%) or combination of these. Among the survivors, the median LOS was 15 days while non-survivors have a median LOS of 6 days. One patient (1.33%) had mild transfusion reaction. Four patients (5.33%) developed DVT despite anti-coagulation. There was improvement in the inflammatory markers (LDH pvalue 0.04, CRP pvalue 0.00, Ferritin pvalue 0.0001). There was improvement in the pulmonary parameters - increase in mean PaO2, mean SaO2, and mean PFR; and decrease in mean FiO2 and mean RR post-treatment. Median LOS is 14 days for the CP group vs 11 days for the non-CP group. Mortality rate among the CP group is 25.33% while the non-CP group was 26.67%. LOS and mortality rate did not reach statistical significance. CONCLUSIONS: There was no significant difference in mortality and length of hospital stay in patients given CP vs controls. CP might have a role in the improvement of inflammatory markers and pulmonary status.


CONCLUSIONS:
There was no significant difference in mortality and length of hospital stay in patients given CP vs controls. CP might have a role in the improvement of inflammatory markers and pulmonary status.

BACKGROUND:
The Novel Coronavirus Disease 2019 (COVID-19) has gripped our country with overwhelming strain in our health system, while there is no single medication which offers cure, we would like to look into the possible benefits of convalescent plasma (CP) in limiting complications and in treating COVID-19. Plasma is the liquid portion of blood containing proteins like albumin, coagulation cascade factors, complement, and a variety of antibodies or immunoglobulins and enzymes. 1 The primary function of all immunoglobulins is the recognition and binding of specific antigenic determinants, whether soluble (including toxins), particulate, or cellular (such as pathogens). The consequences of immunoglobulin binding depend on the nature of the antigen, and at its simplest, may be the physical prevention of antigen penetration through the epithelium. 2 This is the neutralizing effect of immunoglobulins. Secondary effects of immunoglobulin are complement activation, opsonization and antibody dependent cell mediated toxicity by different cells.
The premise behind use of convalescent plasma in COVID-19 patients is the passive transfer of antibodies from recovered individuals to patients with the virus manifesting with diseases. There is both historical precedence and biological plausibility for the use of CP as it has been utilized in previous pandemics. In the Spanish flu of 1918, there . CC-BY 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 February 17, 2021. ; https://doi.org/10.1101/2021.02. 16.21251824 doi: medRxiv preprint was a systematic review on the use of convalescent blood products which showed case fatality rate of 19% in patients given CP < 4days vs. 59% for those given > 4 days from onset of symptoms. 3 In 1974, the Argentine hemorrhagic fever outbreak, there was a double blind placebo controlled study which showed 1.1% case fatality rate in those given immune plasma vs 16.5% in those given normal plasma. In 2011 during the Influenza A/B outbreak where patients were given CP + standard of care vs. standard of care alone, it showed a non-significant reduction in time to normalization of patients' respiratory status; significant reduction in mortality was noted. It was also noted to be safe and well tolerated. 4 The EBOLA outbreak in 2015 in west Africa, the study showed there was no difference in the risk of death at 31% in CP group vs 38% in the control group and that there were no adverse events noted. 5 CP was also used in other coronavirus such as the SARS epidemic in 2003. A retrospective study in Hongkong showed good clinical outcome in 33 out of 80 patients and improved outcome was associated with early administration of CP.
The use of convalescent plasma in COVID-19 is still in the experimental stage and is not yet included as a standard of care for COVID-19. With the unequivocal efficacy data and good safety profile from published studies, we believe that more studies are needed. It has been utilized in our institution through the Hematology Section Convalescent Plasma Program as an adjunct treatment and it is our aim to collate, analyze and report data of patients who received CP.
. CC-BY 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)

Research design:
This is a retrospective cohort analytical study.

Population and Sampling:
Inclusion criteria: 1. Patients must be 18 years of age or older . CC-BY 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.

Controls/Non-CP Group
The patients who received CP were matched based on age, gender and severity of COVID-19 pneumonia to patients who were admitted from April 1, 2020 to October 31, 2020 who received other COVID-19 treatment except convalescent plasma. There were several reasons why these patients did not received convalescent plasma: 1) attending physicians discretion; 2) no available plasma with the same blood type; 3) they were admitted during the time where CP program was not yet available in our institution; OR 4) they were referred for CPT but did not give consent for plasma transfusion.

Exclusion criteria:
Patients with incomplete medical records/data were excluded.

Sampling:
The study used convenience sampling. All patients who received convalescent plasma from April 1, 2020 to October 31, 2020 were included.
. CC-BY 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)

Data Collection
Investigators collected data using google forms and excel sheets. The following data were captured: demographic data (age, sex); medical data (diagnosis, date of admission, date of discharge, date of CP transfusion and day of illness, comorbidities, adverse events); treatment data (medications received; outcome). The sources of data were patient chart from the medical records section and/or census. Data custodian ensured completeness of data and consistency with source data.

Data Analysis
Descriptive statistics using frequency, mean, median and mode were used. For the comparison of pulmonary and inflammatory parameters pre and post-CP transfusion, Wilcoxon signed rank test was used. For the comparison between CP and non-CP group with mortality and length of hospital stay, Mann Whitney U test was used.

Ethical Considerations
The study proposal was reviewed and approved by the CSMC Research Ethics Review Committee.

RESULTS:
There were seventy five (75) COVID-19 patients diagnosed using QRT-PCR for SARS COV-2 on NPS/OPS given convalescent plasma from April 1, 2020 to Oct 31, 2020 who were included in this retrospective analysis. One patient was excluded because of incomplete data and death in <48 hours. Demographics and baseline data of the participants are shown in Table 1 The majority of patients received additional therapies including dexamethasone (100%); Remdesivir (95%); all were given antibiotics either piperacillin-tazobactam or meropenem or ceftriaxone and/or azithromycin, tocilizumab (65%); hemoperfusion . CC-BY 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.

PRE and POST TREATMENT DATA
Laboratory parameters were collected pre-treatment which was either laboratory results during admission or the nearest prior to CP transfusion and post-treatment which was within 7 days after CP transfusion shown in Table 3. There was a decrease in the mean hemoglobin at 13.24 g/dL pre-treatment to 12.90 g/dL post-treatment. There was an increase in mean WBC from 10.00 to 11.60 pre-and post-treatment. These were statistically significant. There was no significant change in the mean platelet level from 235.60 to 251.20, very few had mild thrombocytopenia.
There was a noted trend towards improvement in the inflammatory markers, which were statistically significant. The mean LDH from a baseline was 485.82, which decreased to 453.24 ng/mL, mean pre-treatment CRP was 206.02 which decreased to mean posttreatment of 53.98 and mean serum ferritin also decreased from 3480.41 to 2713.54.
These were all statistically significant. . CC-BY 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.

Comparison of CP group vs matched-controls
Patients given CP were matched by age, gender and severity of illness with control patients who were given other forms of treatment except convalescent plasma. Table 4 summarizes the comparison between the two groups with regards to the length of hospital stay and mortality. Among the survivors from the CP group, the median LOS is 15 days with most of them staying in the range of 10 to 20 days while the median LOS among the non-CP group is 11.5 days with an IQR of 9-17 days. Among the nonsurvivors from the CP group, the median LOS is 6 days with an IQR of 6 to 15.5 days, while the median LOS among the Non-CP group non-survivors is 11 days with an IQR of 6.5 to 26 days. The median LOS of all patients is 14 days with an IQR of 9 to 20 days for the CP group vs 11 days for the non-CP group with an IQR of 8 to 17 days. All values when compared did not reach statistical significance.
The mortality rate among the CP group is 25.33% (19/75) while the non-CP group was 26.67% (20/75) and was not statistically different.

DISCUSSION:
This study reports our experience on the use of convalescent plasma in COVID-19 not just in severe and critical patients but also COVID-19 with moderate pneumonia. Our patient profile consisted of the majority elderly wherein 82.67% belonged to the >50 years old age group and male gender making up 60% of the subjects. This is consistent with the epidemiologic study of Haw et al. that COVID-19 in the Philippines is more common in males and older age groups. 6 Genetics, behavior/lifestyle and immune system were factors cited for this biological differences. 7 Hematologic profile showed anemia which may be due to inflammation and/or bleeding during the course of hospitalization. Leukopenia common in viral infection, was less frequently seen in our study. Normal to mild thrombocytopenia was frequent. The NLR which is a marker of inflammation was high in 88% of the subjects. There were many studies showing a vast range of hematological abnormalities in COVID-19 patients, the most common findings include lymphocytopenia, neutrophilia, eosinopenia, and mild thrombocytopenia. [8][9][10][11] According to a meta-analysis, increased white blood cell count, and decreased lymphocyte and decrease platelet counts are associated with severe disease and higher mortality. 12 A study by Ellinghaus et al., showed that "O" blood type was associated with lower risk of infection compared to non "O" blood type, while "A" blood type had greater risk than non A persons. 13 In our study, though we did not look into the association of blood groups to severity of disease, it was noted that most of our COVID patients are O+ at 44% followed by B+ at 32%. This may be due to the fact that most Filipinos have O+ blood type at 44 to 46%. 14 . CC-BY 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. given CP which showed improvement in fever, SOFA score, P/F ratio, and decrease viral load. Four patients recovered from acute respiratory disease syndrome (ARDS), and three were weaned from mechanical ventilation within 2 weeks of treatment. 15 Duan et al. reported 10 critically ill COVID patients who were given CP and compared them with control group. This study showed clinical, laboratory and radiologic improvement in those given CP, mortality rate was 0 in CP vs 3/10 in the historical controls. 16 An RCT from Spain involving 81 patients showed improved survival in the convalescent plasma group (mortality at 15 and 29 days, 0% with convalescent plasma and 9% [4 of 43] in controls. The likelihood of progression was also lower with convalescent plasma (0% with convalescent plasma versus 14% of controls). 17 Li and Zhang et al study was an RCT from china involving 103 patients who received CP vs standard of care, mortality rate was 16% vs 24%; time to improvement was 2 days vs 5.3 days. 18 Rasheed et al. a multicenter study in Iraq with a population of 49 comparing those who were given CP (21) vs best available treatment (BAT) (28). The results . CC-BY 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 February 17, 2021. ; https://doi.org/10.1101/2021.02.16.21251824 doi: medRxiv preprint showed a mortality of 1/21 (5%) vs 8/28 (29%) and a shorter length of hospital stay. 19 The biggest study so far was from the Mayo clinic collaboration in USA with a total of 35,322 subjects which showed transfusions of CP within 3 days of disease yielded greater reduction in mortality that is 8.7% vs 11.9% in late CP; a higher antibody titer was also associated with lower mortality. 20 There were also negative studies, Zeng et al. Results showed a non-significant reduction in mortality that is 14.8% vs 24.3% (p=0.09) and shorter length of hospital stay 9 days vs 13 days (p=0.002). 22 The PLACID trial is the first RCT with a large population involving 464 patients from India which showed CP was not associated with a reduction in the progression from moderate to severe covid-19 and no reduction in all-cause mortality. 23 And finally, a recent living systematic review by Piechotta et al. showed uncertain effect on mortality with a risk ratio of 0.89 (CI 0.61 to 1.31). 24 With regard to safety of CP, published data are consistent with the low incidence of adverse events. Our study showed a mild infusion reaction of 1/75 (1.33%) and acute DVT of 4/75 (5.33%). Joyner et al. in a multicenter collaborative study of 20,000 hospitalized patients who received CP reported <1% transfusion reaction, <1% thromboembolic events and 3% cardiac events. 25 Most of the studies mentioned above also noted none to low incidence (<1%) of adverse events after CP transfusion.
. CC-BY 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 February 17, 2021. ; https://doi.org/10.1101/2021.02. 16.21251824 doi: medRxiv preprint Although it seems that overall outcome in terms of mortality and length of hospital stay did not significantly impact those who received CP compared to controls, our study showed that there was a trend in the improvement of inflammatory markers, pulmonary parameters and oxygenation status pre and post treatment in the majority of patients and these were statistically significant. These findings were also reported in many case series and observational studies. An early study in china reported 4 patients recovered from ARDS and 3 weaned from mechanical ventilator within 2 weeks of CP administration. 15 Duan's pilot study showed laboratory, clinical and radiologic improvement in 10 critically ill covid-19 patients. 16 A case series from South Korea, reported improvement in 3 COVID-19 patients with ARDS on mechanical ventilator. 26 Eight out of 10 critically ill COVID-19 patients from Mexico and 18 out of 20 patients from a US study reported improvement in inflammatory markers, clinical and radiologic improvement. [27][28] Whether it is due to convalescent plasma and/or other therapies given to patient, our study is not designed to determine this.
One limitation of this study, is that antibody titer in the donors' plasma was not quantitatively determined because of the unavailability of the test. It was seen in a large multi-center study by Joyner et al. that there's a significant survival advantage in COVID-19 patients who were given high antibody titer plasma (>18.45 S/co) compared to low titer (<4.62 s/co). 20 It might have contributed to the non-significant difference in the mortality rate and hospital stay seen in our study.
. CC-BY 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 February 17, 2021. ; https://doi.org/10.1101/2021.02.16.21251824 doi: medRxiv preprint

CONCLUSION:
In conclusion, although there was no difference in mortality and length of hospital stay in patients given CP vs controls. CP might have a role in the improvement of inflammatory markers and pulmonary status of COVID-19 patients. And finally, CP transfusion is relatively safe with low incidence of adverse events.

CONFLICT OF INTEREST:
All authors declare no conflict of interest

ACKNOWLEDGEMENTS:
We are sincerely thankful to our colleagues who helped us in completing this endeavor: Dr. Arvin Jay Bautista, Dr. Erika Chua, Dr. Matt Fontarum, Dr. Elijah Adamos, Dr. Marge Mesina and Dr. Sharlene Dizon.
. CC-BY 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 February 17, 2021. ; https://doi.org/10.1101/2021.02.16.21251824 doi: medRxiv preprint