Predictors of mortality in tocilizumab treated severe Covid-19

doi:10.21203/rs.3.rs-1594765/v1

Download PDF

Research Article

Predictors of mortality in tocilizumab treated severe Covid-19

https://doi.org/10.21203/rs.3.rs-1594765/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 20 Jun, 2022

Read the published version in Vaccines →

Version 1

posted

You are reading this latest preprint version

Purpose

Tocilizumab is associated with positive outcomes in severe Covid-19. We wanted to describe the characteristics of non-responders to treatment.

Methods

Retrospective multi-center study in two respiratory departments, investigating adverse outcomes at 90 days from diagnosis in subjects treated with tocilizumab (8mg/kg intravenously single dose) for severe progressive covid-19.

Results

121 subjects, 62% males, 9% fully vaccinated. 96 (79.4%) survived and 25 died (20.6%). Compared to survivors (S), non-survivors (NS) were older (median 57 versus 75 years of age), had more comorbidities (charlson comorbidity index 2 versus 5) and higher rates of intubation/mechanical ventilation (p < 0,05). On admission, NS had lower PO2/FiO2 ratio, higher blood ferritin and higher troponin and on clinical progression (day of tocilizumab treatment) had lower PO2/FiO2 ratio, decreased lymphocytes, increased neutrophil to lymphocyte ratio, increased ferritin and lactate dehydrogenase (LDH), disease located centrally on computed tomography scan and increased late c-reactive protein. Cox proportional hazards regression analysis identified age and LDH on deterioration as predictors of death; admission PO2/FiO2 ratio and LDH as predictors of intubation; PO2/FiO2 ratios, LDH and central lung disease on radiology as predictors of non-invasive ventilation (NIV) (a < 0,05). Log-rank test of mortality yielded the same results (p < 0,001). ROC analysis of above predictors in a separate validation cohort yielded significant results.

Conclusion

Older age and high serum LDH on deterioration are predictors of mortality in tocilizumab treated severe covid-19. Hypoxia levels, LDH and central pulmonary involvement radiologically are associated with intubation and NIV.

covid-19

tocilizumab

mortality

outcomes

severity

As of 31st January 2022, novel coronavirus designated as severe acute respiratory syndrome coronavirus-2 (SARC-CoV-2) causing Covid-19 disease, is responsible for 5,658,702 deaths worldwide. Infection fatality rates vary by geographic location, age, risk factors and vaccination status, ranging from 0.15-1% in unvaccinated individuals, however exceeding 15% in humans > 85 years of age [1, 2].

In-hospital fatality rates also vary, as a mortality rate of 11.4% is reported in a United States 2021 survey, while a mortality rate around 30% is anticipated in critical Covid-19 cases [3, 4]. Pro-inflammatory cytokines like IL-6 are implicated in severe Covid-19. Tocilizumab, an anti-IL-6 receptor antagonist used for the treatment of rheumatoid arthritis, has shown efficacy against Covid-19 in two large randomized trials. RECOVERY trial showed a reduction in mortality rate from 35–31% in conjunction with corticosteroids in hospitalized patients with moderate, severe and critical Covid-19 with evidence of inflammation [5]. In REMAP-CAP trial, critical Covid-19 cases treated with tocilizumab or sarilumab had a mortality rate of 27% versus 36% in those who received only usual care [6]. Three subsequent meta-analyses supported the use of tocilizumab against Covid-19, the first reporting a pooled mortality prevalence of 19% in tocilizumab arm, the second a reduced 28-day mortality (RR 0.89, 95% confidence interval 0.82 to 0.97) and the latter a 28-day mortality risk of 22% (Odds Ratios 0.83, 95% CI, 0.74–0.92) [7–9].

On the other hand, tocilizumab is clearly not effective in all Covid-19 cases. It has been postulated that early drug administration may be of benefit [10]. In Greece, health authorities recommend its use in worsening severe or critical covid-19 with evidence of inflammation (with severe Covid-19 defined by World Health Organization as resting oxygen saturation < 90% on room air or < 94% plus respiratory rate > 30/min and/or lung infiltrates > 50% of total lung fields) [11]. Tocilizumab is potent anti-inflammatory drug with this effect found by others to reflect in lowering c-reactive protein (CRP) albeit not always with clinical benefit, while additional clinical parameters such as level of hypoxia may be of importance to determine treatment escalation or to determine prognosis, as suggested by other authors [12, 13]. The purpose of the present study is to elaborate on the characteristics of patients who received the drug, on clinical indices associated with responsiveness or non-responsiveness to tocilizumab and identify predictors of mortality and of other adverse outcomes related to its` use.

Study Design

This is a multi-center retrospective study of covid-19 subjects hospitalized from May 2021 till November 2021 in two dedicated Respiratory Departments at referral hospitals, namely Pulmonary Department of Corfu General Hospital and 5th Pulmonary Department of SOTIRIA Hospital for Chest Diseases. The study is conducted based on declaration of Helsinki on human subjects` studies and approved by local Institutional Review Boards (25 − 01/12/2021, consent form not required). Inclusion criteria were a covid-19 diagnosis by polymerase chain reaction, WHO clinical progression scale ≥ 5 and treatment with tocilizumab according to national guidelines (severe Covid-19 pneumonia as defined above) [11]. All subjects were instructed to return for a follow-up visit at three months post-diagnosis (Day 0) and if that was not feasible, a telephone communication took place, in order to establish 90-day mortality. The only exclusion criterion was the unavailability of data on survival at 3 months post diagnosis. The majority of patients had a computed tomography (CT) scan of the chest, but that was not a prerequisite for inclusion in the study. Clinical predictors that were identified for three clinical outcomes (mortality, intubation, non-invasive ventilation [NIV]) were further explorated for confirmation of findings in a separate validation cohort.

Clinical investigations

Demographics, body mass index (BMI), Covid-19 vaccination status, comorbidities, charlson comorbidity index (CCI) were recorded. Day of tocilizumab treatment in relation to in-hospital days and disease onset and dose as well as concomitant treatments were also obtained. Partial arterial pressure of oxygen/oxygen delivery concentration ratio (PO₂/FiO₂) was recorded on admission and on deterioration/clinical worsening day (designated as the day tocilizumab was administered, D_T). The same applied for c-reactive protein (CRP), serum Lactate dehydrogenase (LDH), d-dimers, ferritin, blood neutrophils, blood lymphocytes and neutrophil to lymphocyte ratio (NLR) and platelets. Aspartate aminotransferase (SGOT), alanine aminotransferase (SGPT) and troponin were recorded on admission solely and procalcitonin (PCT) on D_T. Charts were assessed for implementation of NIV (high flow nasal cannulae or non-invasive positive pressure ventilation), intubation and mechanical ventilation and all-cause mortality at 90 days. Discharge CRP was also recorded.

Radiology investigations

In subjects that a CT scan was available, two dedicated pulmonologists (one from each department) scored the findings per quality (ground glass opacities, consolidation or mixed pattern), per quantity (occupying more or less than 50% of total lung fields) and per distribution (central perihilar, peripheral or mixed).

Statistical analysis

Continuous variables are presented as median with 25%-75% interquartile range and analyzed with unpaired non-parametric two-tailed Mann-Whitney test. Categorical variables are presented in absolute numbers (frequency %) and analyzed with Fischer`s exact test or Chi-square test. CCI is recorded numerically. Statistically significant variables were further analyzed with Cox proportional hazards multi-variable regression analysis to identify predictors of event(s) (event = death, intubation, NIV). Log-likelihood ratio was examined for goodness of fit. Significant variables were used to plot survival curves using Log-rank (Mantel-Cox) test. Subjects were censored at 90 days. P and alpha value below 0,05 were considered significant. GraphPad Prism version 9.3.1 was used (USA).

Validation cohort

Predictors of event(s) recognized with Cox regression analysis were afterwards validated in a separate validation cohort consisting of retrospectively identified subjects hospitalized between 09/2020-12/2020, who were not treated with tocilizumab and were matched for sex and disease severity with study subjects. Predictors identified in the initial tocilizumab cohort were analyzed with ROC analysis to validate their performance per the same events (mortality, intubation, NIV).

One hundred and twenty-six subjects fulfilling inclusion criteria were initially recognized. Data for five subjects regarding 90-day mortality were not available and were excluded from the study. From the remaining one hundred and twenty-one subjects analyzed, ninety-six were alive after 3 months (Survivors) and 25 were deceased (Non-Survivors). Mortality rate was 20.6% (25/121). Demographics and baseline characteristics are shown in Table 1. Non-survivors were older, had more comorbidities and were more possibly intubated and on invasive ventilation. Timing and dose of tocilizumab did not vary between survivors and non-survivors. Most subjects received one dose 8mg/kg of tocilizumab intravenously. Prophylactic dose of low-molecular weight heparin showed a non-significant association with survival.

Clinical and radiological features are shown in Table 2. At the evaluation during hospital admission, survivors exhibited higher P/F ratio, lower blood ferritin and troponin levels. At the evaluation upon deterioration/clinical worsening (day of tocilizumab treatment, D_T) survivors exhibited higher P/F ratio, lower blood ferritin, higher blood lymphocytes and lower N/L ratio, lower LDH and less central disease distribution. Survivors also exhibited lower late post tocilizumab CRP values.

All variables shown significant in univariate analysis with no more than 5 missing values were included in Cox proportional hazards multi-variable regression analysis. Outcome was death from all causes and cases were censored at 90 days. A model of 9 variables with no co-variates was preferrable (Log-likelihood ratio (G squared): 63,05) with significant predictors of mortality being age and serum LDH on deterioration (D_T), with CCI and admission ferritin marginally not significant (alpha<0,05) (Table 3). Kaplan-Meier survival analysis using cut-offs for age and LDH combined showed worse survival in these subjects (Log-rank test for trend p<0,0001, Figure 1).

Same analysis with intubation as outcome, identified significant variables being P/F ratio on admission and LDH on D_T, with central radiology distribution marginally non-significant (Table 4). Lastly, as to non-invasive ventilation as outcome, significant variables were found P/F ratio on admission and on D_T,LDH on D_T, and central disease distribution (Table 5).

Validation cohort

This was a precedent without tocilizumab treatment retrospective cohort from our department of fifty Covid-19 cases matched for sex and severity (WHO CPS≥5) to tocilizumab treated study subjects. This cohort consisted of 32 males (32/50, 64%), who were significantly older than study subjects (median age 74 years versus 63 years of age, p<0.05) and had worse outcomes than total tocilizumab group [mortality rate 19/50, 38% versus 25/121, 20.6%; intubation rate 16/50, 32% versus 26/121, 21%; NIV rate 27/50, 54% versus 27/121, 22%].

Predictors of mortality recognized in tocilizumab group (Tables 3-5), significantly predicted adverse outcomes in the validation cohort by ROC analysis (Figure 2). Age above 77 years predicted death with 74% sensitivity and 81% specificity (AUC: 0.79). LDH upon clinical deterioration above 468 IU/l predicted death with 79% sensitivity and 77% specificity (AUC: 0.87). Further, LDH on deterioration and P/F at admission both predicted intubation (AUC: 0.83, AUC: 0.80 respectively); P/F ratios at admission, at deterioration and LDH at D_T all predicted NIV (AUC: 0.74, AUC: 0.88, AUC: 0.82 respectively) (all, p<0.05).

The main findings of this study are that among tocilizumab treated patients with severe progressive Covid-19 pneumonia, older age and particularly abnormal serum LDH on disease progression are associated with worse survival. These findings were confirmed in a separate cohort of critically ill patients irrespective of treatment with tocilizumab. Factors associated with intubation were P/F on admission and LDH upon disease progression, while need for non-invasive ventilation was associated with P/F ratios on admission and on progression, serum LDH on progression and central disease distribution on computed tomography scan.

At first, mortality rate observed in our study is similar to the rates reported in randomized controlled trials of tocilizumab but also in meta-analysis reporting a pooled mortality prevalence around 20–30% [5, 6, 9, 12, 14]. Previous studies have well shown that age is a significant risk factor for covid-19 mortality [15, 16]. Other studies have identified inflammatory markers as neutrophil to lymphocyte ratio, d-dimers, ferritin and c-reactive protein as markers of adverse outcomes [17]. However, data are lacking as to which patient responds to tocilizumab or not.

The discrepancies observed in positive trials (Recovery, REMAP-CAP) and negative trials (CORIMUNO-19, RCT-TCZ-COVID-19 Study Group, COV-AID study group) may be at least partially explained by our study [12, 18, 19]. CRP in our study was elevated, reflecting disease severity and the appropriate national administration guidance, but it was not associated with response to tocilizumab treatment, even when measured serially during disease course. Similarly, in most negative studies, increased serum pro-inflammatory markers IL-6 and CRP were not guaranteeing response to immunomodulatory treatment [20]. These findings taken together point to the necessary discovery of additional biomarkers to guide treatment.

On the contrary, serum LDH has been reported by other authors at the beginning of the course of Covid-19 or to estimate disease resolution and not as a predictor of treatment responsiveness as in our study [21]. Increased serum LDH ≥ 2x to 3x upper limit of normal at time of clinical worsening efficiently predicted all adverse outcomes in our study and was further validated in a severely compromised cohort. Tocilizumab dose and timing of administration (2nd in-hospital day, 10th day of symptoms) was identical between responders and non-responders in our study and similar to previous literature. It is intriguing to suggest, based on these results, that older patients with rising LDH values, irrespective of CRP levels, might benefit from earlier treatment intensification with tocilizumab.

P/F ratio is a crucial indicator for Acute Lung Injury (≤ 300), Acute Respiratory Distress Syndrome (≤ 200) and severe ARDS (≤ 100). Responders in our study had a median P/F ratio 150 when they received tocilizumab, while non responders significantly lower, 100. P/F ratios of > 200 on admission protected from intubation while P/F > 150 upon clinical deterioration protected from the use of NIV. Negative studies employed patients with a ratio > 200 while other investigators focused on patients with P/F ratio < 200 [18, 19, 13]. Studies that would evaluate “earlier” tocilizumab treatment based on P/F ratio in the range of > 100 to < 200 would be useful, in our opinion.

To our knowledge, central disease distribution association with adverse outcomes, is a novel finding not previously reported. Subpleural radiology has been linked with covid-19 resolution [21]. This finding is in line with everyday clinical practice as well. This was significant for NIV, but also, albeit not statistically, for the other two outcomes mortality and intubation. These patients might also need prioritization to treatment and could be the target of future clinical trials.

We acknowledge our study has certain limitations, one being its retrospective design, lack of data posing restrictions in multi-variable analysis and absence of a control group. Further, adverse outcomes reported in this study (all-cause mortality) might be related not solely to severe covid-19 but to other etiologies (i.e pulmonary embolism, sepsis or concomitant other diseases). Lastly, radiology findings were classified by pulmonologists, since this was not designed as a radiology study.

On the other hand, our findings on LDH, P/F ratios and radiology, are relatively novel and of clinical significance. Advantages of our study are the relatively large size of tocilizumab treated subjects, the multi-center design, the 90-days data, the serially recorded clinical parameters and the confirmation of our findings in a separate validation cohort. The high predictive value of our findings, as shown by the significant AUC values in the tocilizumab untreated validation cohort patients, imply that our findings relate to the disease (Covid-19) itself and may be tocilizumab independent.

In our retrospective multi-center study, death occurred in 20.6% of tocilizumab treated severe Covid-19 patients. Predictors of mortality were increased age and particularly increasing serum levels of LDH during the disease course. Additional predictors of adverse outcomes were P/F ratios and central disease distribution. These findings should be used to define patients` sub-populations in prospective clinical trials with tocilizumab or other treatments.

The authors declare that no funds, grants or other support were received during the preparation of this manuscript.

The authors have no relevant financial or non-financial interests to disclose.

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Antonakis, Pandi, Chaini, Digalaki, Kosti, Koukidou and Afthinos. The first draft of the manuscript was written by Pagkratis, Chrysikos, Hillas and Markatis. All authors commented on previous versions of the manuscript. Dimakou and Papanikolaou reviewed the analysis, data collection and final versions of the manuscript. All authors read and approved the final manuscript.

Meyerowitz-Katz G, Merone L. A systematic review and meta-analysis of published research data on COVID-19 infection fatality rates. Int J Infect Dis. 2020 Dec;101:138–148. doi: 10.1016/j.ijid.2020.09.1464. Epub 2020 Sep 29. PMID: 33007452; PMCID: PMC7524446.
Levin AT, Hanage WP, Owusu-Boaitey N, Cochran KB, Walsh SP, Meyerowitz-Katz G. Assessing the age specificity of infection fatality rates for COVID-19: systematic review, meta-analysis, and public policy implications. Eur J Epidemiol. 2020 Dec;35(12):1123–1138. doi: 10.1007/s10654-020-00698-1. Epub 2020 Dec 8. PMID: 33289900; PMCID: PMC7721859.
Garg S, Patel K, Pham H, et al. Clinical Trends Among U.S. Adults Hospitalized With COVID-19, March to December 2020: A Cross-Sectional Study. Ann Intern Med. 2021 Oct;174(10):1409–1419. doi: 10.7326/M21-1991. Epub 2021 Aug 10. PMID: 34370517; PMCID: PMC8381761.
Anesi GL, Jablonski J, Harhay MO, et al. Characteristics, Outcomes, and Trends of Patients With COVID-19-Related Critical Illness at a Learning Health System in the United States. Ann Intern Med. 2021 May;174(5):613–621. doi: 10.7326/M20-5327. Epub 2021 Jan 19. PMID: 33460330; PMCID: PMC7901669.
RECOVERY Collaborative Group. Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial. Lancet. 2021 May 1;397(10285):1637–1645. doi: 10.1016/S0140-6736(21)00676-0. PMID: 33933206; PMCID: PMC8084355.
REMAP-CAP Investigators, Gordon AC, Mouncey PR,et al. Interleukin-6 Receptor Antagonists in Critically Ill Patients with Covid-19. N Engl J Med. 2021 Apr 22;384(16):1491–1502. doi: 10.1056/NEJMoa2100433. Epub 2021 Feb 25. PMID: 33631065; PMCID: PMC7953461.
Berardicurti O, Ruscitti P, Ursini F, et al. Mortality in tocilizumab-treated patients with COVID-19: a systematic review and meta-analysis. Clin Exp Rheumatol. 2020 Nov-Dec;38(6):1247–1254. Epub 2020 Dec 3. PMID: 33275094.
Ghosn L, Chaimani A, Evrenoglou T,et al. Interleukin-6 blocking agents for treating COVID-19: a living systematic review. Cochrane Database Syst Rev. 2021 Mar 18;3(3):CD013881. doi: 10.1002/14651858.CD013881. PMID: 33734435; PMCID: PMC8406988.
WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group, Shankar-Hari M, Vale CL, Godolphin PJ, et al. Association Between Administration of IL-6 Antagonists and Mortality Among Patients Hospitalized for COVID-19: A Meta-analysis. JAMA. 2021 Aug 10;326(6):499–518. doi: 10.1001/jama.2021.11330. PMID: 34228774; PMCID: PMC8261689.
San-Juan R, Fernández-Ruiz M, López-Medrano F, Carretero O, et al; H12O Immunomodulation Therapy for COVID-19 Group, the Spanish Network for Research in Infectious Diseases (REIPI). Analysis of the factors predicting clinical response to tocilizumab therapy in patients with severe COVID-19. Int J Infect Dis. 2022 Jan 23:S1201-9712(22)00050 – 9. doi: 10.1016/j.ijid.2022.01.040. Epub ahead of print. PMID: 35081417; PMCID: PMC8783837.
https://eody.gov.gr/covid_19_algorithmos-nosileuomenon_20211227/.
Salvarani C, Dolci G, Massari M, et al; RCT-TCZ-COVID-19 Study Group. Effect of Tocilizumab vs Standard Care on Clinical Worsening in Patients Hospitalized With COVID-19 Pneumonia: A Randomized Clinical Trial. JAMA Intern Med. 2021 Jan 1;181(1):24–31. doi: 10.1001/jamainternmed.2020.6615. PMID: 33080005; PMCID: PMC7577199.
Karampitsakos T, Malakounidou E, Papaioannou O, et al. Tocilizumab improves 28-day survival in hospitalized patients with severe COVID-19: an open label, prospective study. Respir Res. 2021 Dec 22;22(1):317. doi: 10.1186/s12931-021-01914-6. PMID: 34937570; PMCID: PMC8692825.
Berardicurti O, Ruscitti P, Ursini F, et al. Mortality in tocilizumab-treated patients with COVID-19: a systematic review and meta-analysis. Clin Exp Rheumatol. 2020 Nov-Dec;38(6):1247–1254. Epub 2020 Dec 3. PMID: 33275094.
Duarte-Millán MA, Mesa-Plaza N, Guerrero-Santillán M, et al. Prognostic factors and combined use of tocilizumab and corticosteroids in a Spanish cohort of elderly COVID-19 patients. J Med Virol. 2021 Nov 29. doi: 10.1002/jmv.27488. Epub ahead of print. PMID: 34845754.
AlQahtani H, AlBilal S, Mahmoud E, et al. Outcomes associated with tocilizumab with or without corticosteroid versus dexamethasone for treatment of patients with severe to critical COVID-19 pneumonia. J Infect Public Health. 2022 Jan;15(1):36–41. doi: 10.1016/j.jiph.2021.11.017. Epub 2021 Nov 24. PMID: 34883296; PMCID: PMC8611824.
Brook R, Lim HY, Ho P, Choy KW. Risk factors and early prediction of clinical deterioration and mortality in adult COVID-19 inpatients: an Australian tertiary hospital experience. Intern Med J. 2021 Nov 22. doi: 10.1111/imj.15631. Epub ahead of print. PMID: 34806276.
Hermine O, Mariette X, Tharaux PL, Resche-Rigon M, Porcher R, Ravaud P; CORIMUNO-19 Collaborative Group. Effect of Tocilizumab vs Usual Care in Adults Hospitalized With COVID-19 and Moderate or Severe Pneumonia: A Randomized Clinical Trial. JAMA Intern Med. 2021 Jan 1;181(1):32–40. doi:
10.1001/jamainternmed.2020.6820. Erratum in: JAMA Intern Med. 2021 Jan 1;181(1):144. Erratum in: JAMA Intern Med. 2021 Jul 1;181(7):1021. PMID: 33080017; PMCID: PMC7577198.
Declercq J, Van Damme KFA, De Leeuw E, et al. Effect of anti-interleukin drugs in patients with COVID-19 and signs of cytokine release syndrome (COV-AID): a factorial, randomised, controlled trial. Lancet Respir Med. 2021 Dec;9(12):1427–1438. doi: 10.1016/S2213-2600(21)00377-5. Epub 2021 Oct 29. PMID: 34756178; PMCID: PMC8555973.
Cassone G, Dolci G, Besutti G, et al. Predictive factors of clinical outcomes in patients with COVID-19 treated with tocilizumab: A monocentric retrospective analysis. PLoS One. 2022 Jan 26;17(1):e0262908. doi: 10.1371/journal.pone.0262908. PMID: 35081151; PMCID: PMC8791493.
Zhao Y, Wang D, Mei N, et al. Longitudinal Radiological Findings in Patients With COVID-19 With Different Severities: From Onset to Long-Term Follow-Up After Discharge. Front Med (Lausanne). 2021 Sep 21;8:711435. doi: 10.3389/fmed.2021.711435. PMID: 34621760; PMCID: PMC8490620.

Table 1: Study subjects` demographics, along with treatment variables. Continuous variables are presented as median (25^th-75^thIQR), categorical variables as n (frequency %). CCI: charlson comorbidity index.

	Survivors (n=96)	Non-Survivors (n=25)	p
Age, years	57 (48-67)	75 (66-83)	<0,001
Male sex	60 (62.5)	15 (60)	0,9
Fully vaccinated	7 (7)	4 (16)	0,3
CCI	2 (1-3.7)	5 (3.5-7)	<0,001
Diseases: Chronic pulmonary disease Cardiovascular disease Diabetes Chronic kidney disease Malignancies Connective tissue disease	10 (10) 21 (21) 14 (14) 3 (3) 6 (6) 3 (3)	10 (40) 14 (56) 10 (40) 2 (8) 2 (8) 2 (8)	0,9
Body mass index	27 (25-30)	28 (26-29)	0,4
Hospitalization days	16 (12-21)	18 (11-26)	0,6
Hospital days until Tocilizumab treatment	2 (1-4)	2 (2-3.5)	0,9
Symptom days until Tocilizumab treatment	10 (8-12)	10 (6-12)	0,3
Tocilizumab total dose mg	600 (505-720)	640 (560-960)	0,17
Other treatments Remdesivir Corticosteroids Antibiotics Anti-coagulation prophylactic dose intermediate dose full therapeutic dose	54 (56) 96 (100) 35 (36) 28 (29) 53 (55) 15 (16)	17 (68) 25 (100) 13 (52) 3 (12) 14 (56) 8 (32)	0,3 0,9 0,17 0,078
Non-invasive ventilation	20 (20)	7 (28)	0,4
Intubation-invasive ventilation	8 (8)	18 (72)	<0,001

Τable 2: Clinical and radiological features of study subjects. Continuous variables presented as median (25^th-75^th IQR), categorical variables as n (frequency %). P/F ratio: partial oxygen pressure/oxygen concentration, CRP: c-reactive protein, CPK: creatine kinase, LDH: lactate dehydrogenase, N/L ratio: blood neutrophils/blood lymphocytes, SGOT: aspartate aminotransferase, SGPT: alanine aminotransferase, ALP: alkaline phosphatase, γGT: gloutamyl-tranferase, PCT: procalcitonin, TLF: total lung fields, GGO: ground glass opacities. In case of missing values n is depicted.

	Survivors (n=96)	Non-Survivors (n=25)	p
P/F admission	281 (238-314)	214 (188-271)	<0,001
Neutrophils/mm³ admission	4,680 (3,388-7,410)	5,740 (2,990-10,725)	0,3
Lymphocytes/mm³ admission	895 (680-1,175)	710 (495-1,165)	0,12
N/L ratio admission	5.5 (3.8-8.2)	6.6 (3.4-16.5)	0,2
D-dimers ng/l admission	0.72 (0.51-1.1)	0.9 (0.47-1.54)	0,3
Ferritin admission (ng/ml)	587 (256-986) (n=90)	1,271 (480-2,431)	0,004
Platelets x10³/μladmission	180 (133-228)	175 (115-275)	0,8
CRP admission mg/l	8.7 (5-14)	8.3 (5.8-12.5)	0,8
CPK admission U/l	126 (79-219) (n=50)	105 (69-246) (n=20)	0,5
LDH admission U/l	342 (274-436)	413 (312-540)	0,09
SGOT admission U/l	36 (29-54)	38 (28-59)	0,8
SGPT admission U/l	35 (25-57)	31 (20-36)	0,076
ALP admission U/l	57 (50-78) (n=50)	66 (57-78) (n=20)	0,2
γ-GT admission U/l	42 (24-59) (n=50)	28 (19-43) (n=20)	0,078
Troponin ng/l admission	9.4 (5-15) (n=50)	24 (11-30) (n=20)	<0,001
P/F D_T	150 (111-189)	100 (73-129)	<0,001
Neutrophils/mm³ D_T	6,880 (4,100-9,100)	8,100 (5,530-12,265)	0,10
Lymphocytes/mm³ D_T	860 (605-1,180)	560 (460-915)	0,0068
N/L ratio D_T	6.8 (4.5-11.3)	16.6 (5-23)	0,0032
D-dimers ng/l D_T	0.9 (0.6-1.6) (n=60)	1.3 (0.7-2.7)	0,079
Ferritin D_T (ng/ml)	674 (457-1,246) (n=70)	1,565 (529-2,594)	0,022
Platelets x 10³ /μl D_T	223 (171-298)	222 (119-263)	0,2
CRP D_T mg/l	8.6 (5-13)	10.8 (4-14)	0,4
CPK D_T U/l	91 (48-147) (n=50)	116 (77-213) (n=20)	0,19
LDH D_TU/l	354 (297-444)	530 (369-665)	<0.001
PCT ng/ml D_T	0.1 (0.02-1.1) (n=10)	0.5 (0.07-1.2) (n=5)	0,6
CRP late (time of discharge or death) mg/l	0.09 (0.0-0.4) (n=94)	7.6 (2.6-27) (n=22)	<0,001
Radiology Extent >50% TLF Central/peripheral/mixed Ggo/consolidation/mixed	n=92 65 (70) 5/27/60 33/8/51	n=24 21 (87) 5/2/17 8/2/14	0,15 0,013 0,9

Table 3: Cox regression multi-variable analysis (outcome:death). P/F ratio: partial arterial oxygen pressure/oxygen concentration, LDH: Lactate dehydrogenase. D_T: Day of tocilizumab treatment.

Variable	Hazard ratio	95% Confidence Interval	p
Age	1,082	1,01 to 1,15	0,010
Charlson comorbidity index	1,30	0,97 to 1,71	0,063
P/F admission	1,000	0,99 to 1,009	0,95
Ferritin admission	1,000	1,000 to 1,000	0,0504
Lymphocytes D_T	0,999	0,999 to 1,000	0,34
LDH D_T	1,004	1,002 to 1,007	0,0013
P/F D_T	0,991	0,97 to 1,002	0,29
Mixed distribution	1,744	0,382 to 18,02	0,54
Central distribution	3,769	0,538 to 43,22	0.211

Table 4: Cox regression multi-variables analysis (outcome: intubation) (Log-likelihood ratio: 37). P/F ratio: partial arterial oxygen pressure/oxygen concentration, N/L: blood neutrophils/blood lymphocytes, LDH: Lactate dehydrogenase, D_T: Day of tocilizumab treatment.

Variable	Hazard ratio	95% Confidence Interval	p
Age	1,005	0,95 to 1,06	0,85
Charlson comorbidity index	1,018	0,76 to 1,31	0,89
P/F admission	0,990	0,985 to 0,996	0,001
Ferritin admission	1,000	0,999 to 1,000	0,12
N/L D_T	1,033	0,99 to 1,07	0,11
LDH D_T	1,003	1,000 to 1,005	0,019
P/F D_T	1,00	0,995 to 1,005	0,99
Mixed distribution	1,66	0,47 to 8,99	0,48
Central distribution	5,16	0,95 to 33,05	0,058

Table 5: Cox regression multi-variables analysis (outcome: non-invasive ventilation) (Log-likelihood ratio: 44). P/F ratio: partial arterial oxygen pressure/oxygen concentration, N/L: blood neutrophils/blood lymphocytes, LDH: Lactate dehydrogenase, D_T: Day of tocilizumab treatment.

Variable	Hazard Ratio	95% Confidence Interval	p
Age	1,005	0,96 to 1,04	0,78
Charlson comorbidity index	0,913	0,726 to 1,126	0,41
P/F admission	0,993	0,988 to 0,998	0,005
Ferritin admission	1,000	0,999 to 1,000	0,86
N/L D_T	1,029	0,99 to 1,06	0,11
LDH D_T	1,002	1,00 to 1,004	0,017
P/F D_T	0,992	0,985 to 0,998	0,036
Mixed distribution	2,066	0,84 to 6,31	0,15
Central distribution	4,49	1,27 to 16,96	0,020

No competing interests reported.

Download PDF

Journal Publication

published 20 Jun, 2022

Read the published version in Vaccines →

Version 1

posted

You are reading this latest preprint version

Predictors of mortality in tocilizumab treated severe Covid-19

Status:

Journal Publication

Version 1

Abstract

Purpose

Methods

Results

Conclusion

Figures

Introduction

Materials And Methods

Study Design

Clinical investigations

Radiology investigations

Statistical analysis

Validation cohort

Results

Discussion

Conclusions

Declarations

References

Tables

Additional Declarations

Status:

Journal Publication

Version 1