Content uploaded by Cadegiani Flávio
Author content
All content in this area was uploaded by Cadegiani Flávio on Jan 10, 2022
Content may be subject to copyright.
1
Ivermectin Prophylaxis Used for COVID-19 Reduces COVID-19 Infection and
1
Mortality Rates: A City-Wide, Prospective Observational Study of 223,128
2
Subjects Using Propensity Score Matching.
3
4
Lucy Kerr, MD, ARDMS1, Flavio A. Cadegiani, MD, MSc, PhD2,3, Fernando Baldi,
5
PhD4, Raysildo Barbosa Lôbo, PhD5, Washington Luiz Olivato Assagra6, Fernando Carlos
6
Proença7, Pierre Kory, MD, MPA3, Jennifer A. Hibberd, DDS, DPD, MRCDC8, Juan J
7
Chamie-Quintero9
8
9
1Instituto Kerr de Ensino e Pesquisa, São Paulo, Brazil
10
2Corpometria Institute, Brasilia, Brazil
11
3Front-Line Covid-19 Critical Care Alliance (FLCCC), USA
12
4Department of Animal Sciences, State University of São Paulo, São Paulo, Brazil
13
5Department of Genetics, Ribeirão Preto Medical School, University of São Paulo, Ribeirão
14
Preto, Brazil.
15
6Centro Técnico de Avaliação Genômica C.T.A.G, Ribeirão Preto, Brazil
16
7Itajaí City Hall, Itajaí, Brazil
17
8University of Toronto, Toronto, Canada
18
9Data Analysis, Universidad EAFIT, Cambridge, USA
19
20
*Corresponding author:
21
Flávio A. Cadegiani, MD, MSc, PhD
22
Corpometria Institute
23
SGAS 915 Centro Clínico Advance, Rooms 260/262/264, Brasília, DF, Brazil
24
f.cadegiani@gmail.com, flavio.cadegiani@unifesp.br, flavio@flccc.net
25
+55 61 981.395.395
26
27
28
Key-words: COVID-19, SARS-CoV-2, ivermectin, prophylaxis, prevention,
29
coronavirus
30
31
Acromyums: COPD = Chronic Obstructive Pulmonary Disease; CVD = cardiovascular
32
disease; MI = Myocardial infarction; T2D = Type 2 Diabetes
33
34
35
36
37
38
39
40
2
Abstract
41
42
Background: Ivermectin has demonstrated different mechanisms of action that
43
potentially protect from both COVID-19 infection and COVID-19-related comorbidities.
44
Based on the studies suggesting efficacy in prophylaxis combined with the known safety
45
profile of ivermectin, a citywide prevention program using ivermectin for COVID-19 was
46
implemented in Itajai, a Southern city in Brazil in the state of Santa Catarina. The
47
objective of this study was to evaluate the impact of regular ivermectin use on subsequent
48
COVID-19 infection and mortality rates.
49
Materials and methods: We analyzed data from a prospective, observational study of
50
the citywide COVID-19 prevention with ivermectin program which occurred between
51
July 2020 to December of 2020 in Itajaí, Brazil. Study design, institutional review board
52
approval, and analysis of registry data occurred after completion of the program. The
53
program consisted of inviting the entire population of Itajaí to a medical visit in order to
54
enroll in the program and to compile baseline, personal, demographic and medical
55
information. In the absence of contraindications, ivermectin was offered as an optional
56
treatment to be taken 2 consecutive days every 15 days at a dose of 0.2mg/kg/day. In
57
cases where a participating citizen of Itajai became ill with COVID-19, they were
58
recommended to not use ivermectin or any other medication in early outpatient treatment.
59
Clinical outcomes of infection, hospitalization, and death were automatically reported
60
and entered into the registry in real time. Study analysis consisted of comparing
61
ivermectin users with non-users using cohorts of infected patients propensity score
62
matched (PSM) by age, sex, and comorbidities. COVID-19 infection and mortality rates
63
were analyzed with and without use of propensity score matching.
64
Results: Of the 223,128 citizens of Itajaí considered for the study, a total of 159,561
65
subjects were included in the analysis; 113,845 (71.3%) regular ivermectin users and
66
45,716 (23.3%) non-users. Of these, 4,311 ivermectin users were infected, among
67
which 4,197 from the city of Itajaí (3.7% infection rate) and 3,034 non-users (from
68
Itajaí) were infected (6.6% infection rate), a 44% reduction in COVID-19 infection
69
rate (Risk ratio (RR), 0.56; 95% confidence interval (95%CI), 0.53 – 0.58; p <
70
0.0001). Using PSM, two cohorts of 3,034 subjects suffering COVID-19 infection were
71
compared. The regular use of ivermectin led to a 68% reduction in COVID-19 mortality
72
[25 (0.8%) versus 79 (2.6%) among ivermectin non-users; risk ratio (RR), 0.32; 95%
73
confidence interval (CI), 0.20 – 0.49; p < 0.0001]. When adjusted for residual variables,
74
3
reduction in mortality rate was 70% (RR, 0.30; 95%CI 0.19 – 0.46; p < 0.0001). There
75
was a 56% reduction in hospitalization rate (44 versus 99 hospitalizations among
76
ivermectin users and non-users, respectively; RR, 0.44; 95%CI, 0.31 – 0.63; p < 0.0001).
77
After adjustment for residual variables, reduction in hospitalization rate was 67% (RR,
78
0.33; 95%CI 023 – 0.66; p < 0.0001).
79
Conclusion: In this large, propensity score matched study, regular use of ivermectin as a
80
prophylactic agent was associated with significantly reduced COVID-19 infection,
81
hospitalization, and mortality rates.
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
4
Introduction
109
110
Ivermectin has been demonstrated to have not only extensive anti-parasitic actions1,2, but
111
alsoanti-viral, anti-bacterial, and anti-protozoan properties. Ivermectin has been long
112
proposed for use as a repurposed antiviral agent4-6. Indeed, antiviral effects of ivermectin
113
have been reported against both RNA and DNA types of viruses, including HIV-1,
114
Yellow fever (YFV), Japanese encephalitis, tick-borne encephalitis, West Nile, Zika
115
(ZKV), Dengue fever, Chikungunya (CHIKV), Venezuelan equine encephalitis and the
116
Pseudorabies virus3,5,7, as well as functioning in regulation of proteins involved in
117
antiviral responses8.
118
119
Additional actions of ivermectin described include agonism activity to the X-LBD
120
binding receptor (FXR), with multiple potential metabolic benefits9,10; neuronal
121
regeneration11,12, prevention of muscle hypoxia13, anti-inflammatory activity to
122
Interferon (INF)14, nuclear factor-κB (NF-κB), lipopolysaccharide (LPS)15 and JAK-
123
STAT pathway, PAI-116,17; generation of P21 activated Kinase 1 (PAK-1)18,19; reduction
124
of Interleukin-6 (IL-6) levels15; allosteric modulation of P2X4 receptor20; inhibition of
125
high mobility group box 1 (HMGB1)21,22; suppression of mucus hypersecretion,
126
diminished recruitment of immune cells and production of cytokines in the lung23.
127
ivermectin is also described to induce Th1-type immune response against protozoans24,
128
and anti-coagulant action through binding to the S protein of some viruses25.
129
130
The hypothesis that ivermectin could be protective against COVID-19 is
131
substantiated by its multi-pathway, anti-inflammatory effects15,26 and multi-antiviral
132
mechanisms. COVID-19 pathogenesis is largely understood as an inflammation-mediated
133
hemagglutinating infection disrupting pulmonary, vascular and endothelial systems,
134
leading to a multi-systemic disease. In vitro and in-silico, ivermectin has demonstrated
135
anti-SARS-CoV-2 activity through more than 20 direct and indirect mechanisms2,27,28.
136
137
Ivermectin has demonstrated preliminary protective effects against SARS-CoV-2
138
infection in terms of reducing times to clinical recovery, rates of disease progression and
139
mortality2,29,30. However, more robust studies with larger sample sizes are still
140
5
recommended to confirm the possible beneficial effects ivermectin confers in COVID-
141
19.
142
143
Since the onset of the COVID-19 pandemic, the use of inexpensive options based
144
on a consistently beneficial signal of efficacy, a well-established safety profile,
145
favourable cost-effectiveness, ivermectin is a highly attractive intervention for the patient
146
centred medicine practiced by frontline clinicians, with use aligning strongly with the
147
bioethical principles for medical practice outlined in Article 36 of the Helsinki
148
declaration31.
149
150
However, despite this favorable risk/benefit profile and absense of therapeutic
151
alternatives, ivermectin has yet to be approved for prophylaxis and treatment of COVID-
152
19 by agencies throughout the world, including FDA (Food & Drug Administration;
153
United States of America), EUA (European Medicines Agency; Europe) and ANVISA
154
(Agência Nacional de Vigilância Sanitária – Brazilian Health Regulatory Agency;
155
Brazil).
156
157
The ability to prescribe ivermectin or any other off-label drug for COVID-19 has
158
long been at the discretion of frontline physicians once all risks, uncertainties, potential
159
benefits, and patients’ rights are exposed, and informed consent has been obtained. Of
160
particular note, in Brazil, this follows the medical autonomy to determine the best
161
therapeutic strategies for individuals, as per the Medical Code of Ethics of the Brazilian
162
Board of Medical Doctors; the Federal Council of Medicine – Conselho Federal de
163
Medicina (CFM), that determines the obligations and rights of medical doctors in
164
Brazil32.
165
166
Since vaccines for COVID-19 were not available in Brazil until 2,021, and the
167
lack of prophylactic alternatives in the absence of vaccines, Itajai, a city in the
168
Southern Brazilian state of Santa Catarina, initiated a population wide government
169
program for COVID-19 prophylaxis. The medical-focused decision parameters
170
established are based on the distribution of ivermectin to whole populations in different
171
countries. To ensure the safety of the population, a well-controlled computer program
172
was developed to compile and maintain all relevant demographic and clinical data. The
173
6
use of ivermectin was optional and based on patients’ preferences given its benefits as a
174
preventative agent was unproven.
175
176
This study’s objective is to assess the impact on important clincal outcomes when
177
ivermectin is used as prophylaxis for COVID-19. The prophylaxis program occurred in
178
addition to the standard non-pharmacological strategies of masking and social distancing,
179
as part of a citywide program conducted in outpatient settings.
180
181
182
Material and Methods
183
184
Study population
185
186
This was a prospective, observational study. Although study design, IRB approval, and
187
data analysis occurred after completion of the voluntary prophylaxis program, all data
188
were collected prospectively in real-time with mandated reporting to the registry of all
189
events as they occurred during the citywide governmental COVID-19 prevention with
190
ivermectin program, from July 2020 to December 2020, developed in the city of Itajaí, in
191
the state of Santa Catarina, Brazil. Demographic and clinical data was reported from
192
medical records of patients followed in a large outpatient setting; a provisional outpatient
193
clinic set in the Convention Center of Itajaí, and several secondary outpatient settings, as
194
part of the Universal Health System (SUS).
195
196
The objective was to determine the number of patients affected by COVID-19
197
(positivity rate of rtPCR-SARS-CoV-2), risk of death due to COVID-19 (whether
198
infected or not), and COVID-19 mortality rate (risk of death from COVID-19) of those
199
who used and did not use ivermectin prophylactically for COVID-19. This data was
200
analyzed stratified by age, sex, presence of comorbidities, and correlated demographic
201
characteristics.
202
203
The present retrospective analysis of the prospectively collected data was
204
approved by the CONEP - National Research Ethics Council (CONEP) under the number
205
4.821.082 with the project number CAAE: 47124221.2.0000.5485. Although study
206
design, IRB approval, and data analysis occurred after completion of the voluntary
207
7
prophylaxis program, all data were collected prospectively in real-time with mandated
208
reporting to the registry of all events as they occurred during the citywide governmental
209
COVID-19 prevention with ivermectin program, from July 7, 2020 to December 2,
210
2020, developed in the city of Itajaí, in the state of Santa Catarina, Brazil.
211
212
213
Study procedures and data collection
214
215
Optional, voluntary prophylactic use of ivermectin was offered to patients during regular
216
medical visits between July 7, 2020 and December 31, 2020 in 35 different sites,
217
including 34 local SUS health centres and a large temporary patient setting. Doctors
218
working in these sites were free to prescribe ivermectin prophylactically. Subjects that
219
did not use ivermectin either refused or their primary care physicians opted not to offer
220
ivermectin.
221
222
The program was conducted in all 35 sites, 24/7, with the initial enrollment in the
223
program occurring during a two-week time frame, due to the large number of subjects to
224
evaluate in the entire population of Itajaí. In order to avoid underreported data, strict
225
procedure sequencing was followed: 1. Registration and recording of patient data,
226
documented by assistants; 2. Weighing subjects (Subject weight was essential to calculate
227
the appropriate dose of ivermectin); 3. Brief medical evaluation of past medical history,
228
comorbidities, use of medications and contraindications to drugs; 4. Medical prescription
229
of prophylactic doses of ivermectin, according to medical judgment and following a
230
subject’s informed consent related to potential benefits, risks, and side effects. Regarding
231
drug interactions with ivermectin, use of warfarin was a contra-indication for
232
prophylaxis with ivermectin due to drug interactions. Subjects under chronic use of
233
glucocorticoids, protease inhibitors and anti-epileptics were recommended to
234
schedule regular medical visits every six to eight weeks. Subjects were recommended
235
to inform medical doctors about the use of ivermectin, in case one or more of the
236
following medications were prescribed: warfarin, azithromycin, dexamethasone,
237
prednisone or prednisolone (Hydrocortisone or cortisone are not commercially
238
available in regular pharmacies in Brazil).
239
240
8
All details of this citywide program and campaign had been previously agreed
241
upon between the city local department of the National Healthcare System (SUS), city
242
mayor, and local public prosecutors.
243
244
The following data were analyzed, adjusted as confounding factors, and used as
245
variables for balancing and matching groups for the employment of propensity score
246
matching (PSM) in the present study: age, sex, past medical history, previous diseases;
247
myocardial infarction (MI), stroke: existing comorbidities; type 2 diabetes (T2D), asthma,
248
chronic obstructive pulmonary disease (COPD), hypertension, dyslipidemia,
249
cardiovascular diseases (CVD), cancer (any type), and other pulmonary diseases: habits
250
(past or current smoking). Additional data analyzed included self-reported comorbidities
251
and medications used.
252
253
Patients who presented signs or the diagnosis of COVID-19 before July 7, 2020,
254
were excluded from the sample. Other exclusion criteria were contraindications to
255
ivermectin and subjects below 18 years of age. The dose and frequency of ivermectin
256
treatment was 0.2mg/kg/day; i.e., giving one 6mg-tablet for every 30kg. for 2 consecutive
257
days every 15 days.
258
259
During the study, subjects who became infected with COVID-19 were diagnosed
260
with a positive rtPCR-SARS-CoV-2 and then underwent a specific medical visit to assess
261
COVID-19 clinical manifestations and severity. All subjects were recommended not to
262
use ivermectin, nitazoxanide, hydroxychloroquine, spironolactone or any other drug
263
claimed to be effective against COVID-19. The city did not provide or support any
264
specific pharmacological outpatient treatment for subjects infected with COVID-19.
265
266
They were questioned for the presence of common COVID-19 symptoms. These
267
included chills, high-grade fever, cough, myalgia, fatigue, anosmia, ageusia, sore throat,
268
headache, nasal congestion, sneeze, runny nose, hemoptysis, nauseas, vomiting,
269
abdominal pain, diarrhea, cutaneous rash, arthralgia, chest pain, eye pain and pinkeye,
270
and presence of alert signs, including shortness of breath, signs of hypoxia, signs of
271
coagulation abnormalities and an altered level of consciousness. Systolic and diastolic
272
blood pressure, heart rate, respiratory rate, oxygen saturation, and axillar temperature
273
were measured. The same signs and symptoms, and vital signs were collected at each
274
9
following medical visit during COVID-19. Individual data was compiled and reviewed
275
by the researchers.
276
277
Registry data of all patient records from the city of Itajaí between July 7, 2020 and
278
December 2, 2020, including those who used ivermectin and did not use ivermectin were
279
reviewed. Subjects who tested positive for COVID-19 during the study were considered
280
for this analysis, whether they used ivermectin or not. Of the infected subjects, two groups
281
were considered: subjects who used ivermectin prophylactically (treated group) and
282
subjects who did not use ivermectin prophylactically (untreated group). Any missing data
283
from patients were actively searched by the investigators, via phone or in person. Since
284
this is a citywide program, all recorded data must have matched the exact number of
285
COVID-19 cases and deaths of the city. This strict interval avoids differences in terms of
286
periods of exposure.
287
288
Due to the uncertainty of reinfection with COVID-19, subjects with a history of
289
previous COVID-19 did not participate in the program although they were still permitted
290
to use ivermectin prophylactically. Limiting parameters of the government system
291
allowed the recording of a first episode of COVID-19 infection only. Subjects below 18
292
years old and subjects with diagnosis of COVID-19 before July 7, 2020 were
293
excluded from all datasets and analysis.
294
295
From the registry of the city population (223,128 inhabitants), subjects below
296
18 years old (61,583 subjects) were removed. Of the 161,545 subjects above 18 y/o
297
from the city of Itajaí, we removed the 1,984 COVID-19 cases that occurred before
298
July 7, 2020, among subjects above 18 y/o, remaining 159,561 subjects. Subjects
299
above 18 y/o were considered those who were born before June 30, 2002.
300
301
A total of 147,223 subjects participated in the program of ivermectin
302
prophylaxis used for COVID-19. Of these, 24,304 subjects were below 18 y/o. Of the
303
122,919 ivermectin users above 18 y/o, 8,346 were from other cities, and 728 had
304
COVID-19 before July 7, 2020, although they used ivermectin afterwards. In total,
305
113,845 subjects that participated in the program remained in the dataset. The
306
45,716 non-participants, remaining subjects among the 159,561 subjects were
307
considered as the ivermectin non-users.
308
10
309
Finally, city-wide COVID-19 hospitalization and mortality rates of Itajaí were
310
compared between the period before the program (before July 7, 2020) and during the
311
program between July 7, 2020 and December 2, 2020) aiming to evaluate whether a
312
program of prophylaxis with ivermectin for COVID-19 would cause a positive impact in
313
the overall numbers of the city, despite only partial adoption. Chances of dying from
314
COVID-19 in the overall population, according to use or non-use of ivermectin
315
(irrespective of COVID-19 infection) were only calculated prior to matching. Conversely,
316
mortality rate, i.e., among those who were infected by the SARS-CoV-2, was calculated
317
for both pre and post-matched cohorts. Analysis of hospitalization and mortality rates
318
before matching, mortality rate in subpopulations among ivermectin users, among
319
ivermectin non-users, and mortality rate ratios between iveremctin users and non-users in
320
subpopulations, before and after propensity score matching, and STROBE checklist are
321
presented in the Supplement Appendix 1.
322
323
324
Statistical analysis
325
326
The full underlying data for the present analysis was analyzed by two independent
327
statisticians, and discrepancies evalauted by a third statistics expert. In this outpatient
328
study of those who tested positive for SARS-CoV-2, mortality rate was evaluated
329
according to each parameter, that adjusted against other variables (for multivariate
330
regression analysis) and used for balancing and matching groups, including age intervals,
331
sex, history of smoking, prophylactic ivermectin use, T2D, asthma, COPD,
332
cardiovascular diseases and other pulmonary diseases, hypertension, current cancer (any
333
type), history of stroke and/or MI. Groups, baseline characteristics, and mortality rates
334
were presented before matching and after matching.
335
336
Before matching, a generalized linear mixed model was employed, assuming the
337
binomial distribution for the residues and including the fixed classificatory effects of each
338
of these parameters. Age intervals were adjusted for the evaluation of ivermectin
339
prophylactic use as an independent predictor of death from COVID-19. Unadjusted and
340
multivariate Poisson- adjusted probabilities to survive from COVID-19 (p-value),
341
according to each parameter were provided.
342
11
343
PSM was performed for mortality risk between ivermectin and non-vermectin
344
users. COVID-19 infection rate and risk of dying were also calculated matching for
345
variables. After PSM, a second adjustment (‘double adjustment’) with multivariate linear
346
regression was performed for residual variables33,34.
347
348
The statistical approach for missing data depended on the percentage of missing
349
data for each parameter. However, due to the registry system design mandating that all
350
data variables be filled to be formally included in the registry, only erroneously entered
351
(illogical) data were found. In such instances, medical record review was performed to
352
obtain the accurate data.
353
354
The program used for the analysis was the Statistical Analysis Software
355
(SAS/STAT) (SAS Institute Inc., Care, North Carolina, USA).
356
357
For transparency reasons, two datasets will be made public upon peer-
358
reviewed publication, of the 7,345 COVID-19 cases and of the 113,845 participating
359
subjects considered for the present analysis.
360
361
Results
362
363
Figure 1. Underlying data for the study on ivermectin prophylaxis used for COVID-19.
364
365
366
12
367
368
Population of Itajaí in 2020
Population above 18 y/o
COVID-19 cases
above 18 y/o until
July 7, 2020, among
ivermectin users that
started ivermectin
after July 7, 2020.
Ivermectin users
Other cities
IVERMECTIN PROPHYLAXIS FOR COVID-19
Ivermectin users from
Itajaí above 18 y/o
Below 18 y/o
Ivermectin users from Itajaí
above 18 y/o without
COVID-19 until July 7, 2020
Population of Itajaí above 18 y/o
without COVID-19 until July 7, 2020
223,128
161,545
COVID-19 cases until July 7, 2020
- Total = 2,611
- Above 18 y/o = 1,984
(161,545 –1,984 =) 159,561
147,223
24,304
Ivermectin users above 18 y/o
122,919
8,346
114,573
113,845
728
Above
18 y/o
Ivermectin users + Ivermectin non-users
Ivermectin users
COVID-19 cases until July 7, 2020
above 18 years old that did not use
ivermectin after July 7, 2020
1,256
Ivermectin users
113,845 (71.3%)
Ivermectin non-users
45,716 (28.7%)
X
Positive cases above 18 y/o
among ivermectin users
4,311
Positive cases among ivermectin
users above 18 y/o from Itajaí
4,197
Positive cases above 18
y/o among ivermectin
users from other cities
114
Below 18 y/o
61,583
Population of Itajaí in
2020 without COVID-
19 until JUly 7, 2020
220,517
Ivermectin users
113,845
Population of Itajaí above 18 y/o
without COVID-19 until July 7, 2020 that did
not use ivermectin after July 7, 2020
(159,561 –113,845 =) 45,716
Positive cases above 18 y/o
among ivermectin non-users
3,034
COVID-19 cases after July 7, 2020
Positive cases among ivermectin
non-users above 18 y/o from Itajaí
3,034
Positive cases above 18
y/o among ivermectin non-
users from other cities
0
Cases in
ivermectin users
4,197 (3.7%)
Cases in
non-ivermectin users
3,034 (6.6%)
44% in
COVID-19
infection
rate
(Risk ratio, 0.56;
95% confidence
interval 0.53 –
0.58; p < 0.0001)
13
369
A detailed description of the data considered for the present analysis is
370
illustrated in Figure 1. Of the 220,517 citizens of Itajaí without COVID-19 until July
371
7, 2020, 159,561 were above 18 years old. Of the 159,561 citizens above 18 y/o without
372
COVID-19 until July 7, 2020, 113,845 (71.3% of the population above 18 years old)
373
received ivermec tin before being infected by COVID-19. A total of 45,716 citizens
374
(28.7%) did not receive or did not want to receive ivermectin during the program,
375
including as a prophylactic or as treatment after having COVID-19.
376
377
Of the 113,845 prophylaxed subjects from the city of Itajaí, 4,197 had a positive
378
rtPCR-SARS-CoV-2 (3.7% infection rate), while 3,034 of the 37,027 untreated subjects
379
had positive rtPCR-SARS-CoV-2 (6.6% infection rate), a 44% reduction in COVID-19
380
infection rate (Risk ratio (RR), 0.56; 95% confidence interval (95%CI), 0.53 - 0.58;
381
p < 0.0001). An addition of 114 subjects that used ivermectin and were infected were
382
originally from other cities but were registered as part of the program, in a total of
383
4,311 positive cases among ivermectin users. For the present analysis, the 4,311
384
positive cases among subjects that used ivermectin and 3,034 cases among subjects
385
that did not use ivermectin were considered. After PSM, two cohorts of 3,034 subjects
386
were created.
387
388
Baseline characteristics of the 7,345 subjects included prior to PSM and the
389
baseline characteristics of the 6,068 subjects in the matched groups are shown in Table
390
1. Prior to PSM, ivermectin users had a higher percentage of subjects over 50 years old
391
(p < 0.0001), higher prevalence of T2D (p = 0.0004), hypertension (p < 0.0001), CVD (p
392
= 0.03), and a higher percentage of caucasians (p = 0.004), than non-users. After PSM,
393
all baseline parameters were similar between groups. Figure 2 summarizes the main
394
findings of this study.
395
396
Table 1. Baseline characteristics of subjects enrolled in study before matching and after
397
propensity score matched.
398
Pre-Matching
Propensity Score Matched
Overall
(n =
7,345)
Ivermectin
users
(n = 4,311)
Non-
ivermectin
users
p-value
Overall
(n = 6,068)
Ivermectin
users
(n = 3,034)
Non-
ivermectin
users
14
(n = 3,034)
(n = 3,034)
Age
Mean ± SD
42.0 ±
14.7
43.5 ± 14.9
39.8 ± 14.2
< 0.0001
39.7 ± 14.0
3967 ± 13.8
39.8 ± 14.2
< 30 y/o
1730
(23.6%)
886
(20.5%)
844
(27.8%)
1,691
(27.9%)
844
(27.9%)
847
(27.8%)
30-50 y/o
3703
(50.4%)
2121
(49.2%)
1582
(52.2%)
3,155 (52.0%)
1,573
(51.9%)
1,582
(52.1%)
> 50 y/o
1912
(26.0%)
1304
(30.3%)
608
(20.0%)
1,222
(20,1%)
614
(20.2%)
608
(20.1%)
Sex
0.31
Female
3983
(54.2%)
2359
(54.7%)
1624
(53.5%)
3,231
(53.2%)
1,607
(53.0%)
1,624
(53.5%)
Male
3362
(45.8%)
1952
(45.3%)
1410
(46.5%)
2,837
(46.8%)
1,427
(47.0%)
1,410
(46.5%)
Race
Caucasians
5437
(74.0%)
3245
(75.3%)
2192
(72.2%)
0.004
4,398
(72.5%)
2,206
(72.7%)
2,192
(72.3%)
Afro-
Brazilians
209
(2.8%)
109
(2.5%)
100
(3.3%)
0.052
193
(3.2%)
93
(3.1%)
100
(3.3%)
Mixed
1583
(22.6%)
901
(20.9%)
682
(22.5%)
0.10
1,364
(22.5%)
93
(3.1%)
100
(3.3%)
Asian-
Brazilians
116
(1.6%)
56
(1.3%)
60
(2.0%)
0.023
113
(1.9%)
53
(1.8%)
60
(2.0%)
Type 2
diabetes
0.0004
Yes
214
(2.9%)
151
(3.5%)
63
(2.1%)
141
(2.3%)
78
(2.6%)
63
(2.1%)
No
7131
(97.1%)
4160
(96.5%)
2971
(97.9%)
5,927
(97.7%)
2,956
(97.4%)
2,971
(97.9%)
Asthma
0.067
Yes
26 (0.3%)
20
(0.5%)
6
(0.2%)
21
(0.3%)
15
(0.5%)
6
(0.2%)
No
7319
(99.7%)
4291
(99.5%)
3028
(99.8%)
6,047
(99.7%)
3,019
(99.5%)
3,028
(99.8%)
COPD
0.72
Yes
13
(0.2%)
7
(0.2%)
6
(0.2%)
12
(0.2%)
6
(0.2%)
6
(0.2%)
No
7332
(99.8%)
4304
(99.8%)
3028
(99.8%)
6,056
(99.8%)
3,028
(99.8%)
3,028
(99.8%)
Hypertension
< 0.0001
Yes
528
(7.2%)
362
(8.4%)
166
(5.5%)
343
(5.6%)
177
(5.8%)
166
(5.5%)
No
6817
(92.8%)
3949
(91.6%)
2868
(94.5%)
5,725
(94.4%)
2,857
(94.2%)
2,868
(94.5%)
CVD
0.03
Yes
56 (0.8%)
41
(1.0%)
15
(0.5%)
32
(0.5%)
17
(0.6%)
15
(0.5%)
No
7289
(99.2%)
4270
(99.0%)
3019
(99.5%)
6,036
(99.5%)
3,017
(99.4%)
3,019
(99.5%)
Other
pulmonary
diseases
0.53
Yes
15
(0.2%)
10
(0.2%)
5
(0.2%)
9
(0.1%)
4
(0.1%)
5
(0.1%)
No
7330
(99.8%)
4301
(99.8%)
3029
(99.8%)
6,059
(99.9%)
3,030
(99.9%)
3,029
(99.9%)
Cancer
(any type)
0.66
15
Yes
32
(0.4%)
20
(0.5%)
12
(0.4%)
22
(0.4%)
10
(0.3%)
12
(0.4%)
No
7313
(99.6%)
4291
(99.5%)
3023
(99.6%)
6,046
(99.6%)
3,024
(99.7%)
3,022
(99.6%)
Current
smoking
0.76
Yes
110
(1.5%)
63
(1.5%)
47
(1.5%)
95
(1.6%)
48
(1.6%)
47
(1.6%)
No
7235
(98.5%)
4248
(98.5%)
2987
(98.5%)
5,973
(98.4%)
2,986
(98.4%)
2,987
(98.4%)
History of MI
0.26
Yes
15
(0.2%)
11
(0.3%)
4
(0.1%)
8
(0.1%)
4
(0.1%)
4
(0.1%)
No
7330
(99.8%)
4300
(99.7%)
3030
(99.9%)
6,060
(99.9%)
3,030
(99.9%)
3,030
(99.9%)
History of
stroke
0.56
Yes
21 (0.3%)
11
(0.3%)
10
(0.3%)
21
(0.4%)
11
(0.4%)
10
(0.3%)
No
7324
(99.7%)
4300
(99.7%)
3024
(99.7%)
6,047
(99.6%)
3,023
(99.6%)
3,024
(99.7%)
COPD = chronic obstructive pulmonary disease; CVD = cardiovascular disease; MI = myocardial infarction; SD = standard deviation
399
400
401
Figure 2. Summary of the findings.
402
403
404
405
Citywide program of COVID-19
prophylaxis with ivermectin
(Itajaí, Brazil)
113,845 participants
from Itajaí used ivermectin
(71.3%)
Pre-matched
COVID-19 infection
rate risk ratio
0.56 (0.53 –0.58)
[p < 0.0001]
45,716 participants
did NOT use ivermectin
(28.7%)
158,934 participants
3,034 participants
had COVID-19
8.2% infection rate
44 hospitalized patients
due to COVID-19
1.6% hospitalization rate
99 hospitalized patients
due to COVID-19
3.3% hospitalization rate
Post-matching
COVID-19 hospitalization
rate risk ratio
0.44 (0.31 –0.63)
[p < 0.0001}
Post-matching
COVID-19 mortality
rate risk ratio
0.32 (0.20 –0.49)
[p < 0.0001}
IVERMECTIN USE NON IVERMECTIN USE
4,197 participants
from Itajaí had COVID-19
3.7% infection rate
44%
56%
68%
COVID-19
infection rate
COVID-19
hospitalization
rate
COVID-19
mortality
rate
X
3,034 subjects
with COVID-19
3,034 subjects
with COVID-19
Balancing and
matching groups
25 deaths from
COVID-19
0.8% mortality rate
79 deaths from
COVID-19
2.6% mortality rate
114 participants
from other cities
had COVID-19
(Total: 4,311 cases)
16
406
407
408
409
410
411
Hospitalization and mortality rates in ivermectin users and ivermectin non-users in
412
propensity score matched analysis
413
414
As described in Table 2, after employing PSM, of the 6,068 subjects (3,034 in each
415
group), there were 44 hospitalizations among ivermectin users (1.6% hospitalization rate)
416
and 99 hospitalizations (3.3% hospitalization rate) among ivermectin non-users, a 56%
417
reduction in hospitalization rate (RR, 0.44; 95%CI, 0.31 – 0.63). When adjustment for
418
variables was employed, reduction in hospitalization rate was 67% (RR, 0.33; 95%CI 023
419
– 0.66; p < 0.0001).
420
421
There were 25 deaths among ivermectin users (0.8% mortality rate) and 79 deaths
422
among non-ivermectin users (2.6% mortality rate), a 68% reduction in mortality rate (RR,
423
0.32; 95%CI 0.20 – 0.49). When PSM was adjusted, reduction in mortality rate was 70%
424
(RR, 0.30; 95%CI 0.19 – 0.46; p < 0.0001).
425
426
Table 2. Propensity socre matched hospitalization and mortality rate among ivermectin users and
427
non-users.
428
Overall
IVM users
Non-
IVM
users
PSM
mortality risk ratio
(95%CI) and
p-value [p]
Adjusted PSM
mortality risk ratio
(95%CI) and
p-value [p]
COVID-19
infection
Infected population (n)
6,068
3,034
3,034
-
-
COVID-19
hospitalization
Hospitalization due to COVID-
19
143
44
99
-
-
Hospitalization rate*
(in case of COVID-19) (%)
2.3%
1.6%
3.3%
0.44 (0.31 – 0.63)
[< 0.0001]
0.33 (0.23 – 0.46)
[<0.0001]
COVID-19
death
COVID-19 deaths (n)**
104
25
79
-
-
Mortality rate
(among infected subjects) (%)
1.7%
0.8%
2.6%
0.32 (0.20 – 0.49)
[< 0.0001]
0.30 (0.19 – 0.46)
[< 0.0001]
IVM = ivermectin; PSM = propensity score matching; CI = confidence interval; *Only subjects hospitalized in public hospitals; **All deaths, including
429
from public and private hospitals, and in-home.
430
431
17
432
433
434
435
436
437
438
439
Determinants of COIVD-19 mortality through propensity score matched analysis
440
441
Table 3 describes resulting risk factors for COVID-19 death amongst the overall
442
population through PSM analysis. Risk factors for mortality in COVID-19 included aging
443
(p < 0.0001), male sex (p = 0.015), T2D (p < 0.0001), hypertension (p < 0.0001), asthma
444
(p = 0.011), COPD (p < 0.0001), other pulmonary diseases (p = 0.048), history of MI (p
445
= 0.034) and history of stroke (p < 0.0001). To detect independent risk factors, post-PSM
446
adjustment for variables showed that ivermectin (p < 0.0001; 70% reduction in mortality
447
risk) and female sex (p = 0.022; 38% reduction in mortality risk) were protectors, whereas
448
T2D (p = 0.041; 79% increase in mortality risk), hypertension (p = 0.008; 98% increase
449
in mortality risk), and, marginally, other pulmonary diseases (p = 0.061; 468% increase
450
in mortality risk) and history of stroke (p = 0.054; 97% increase in mortality risk) were
451
identified as independent risk factors.
452
Table 3. Propensity score matched COVID-19 mortality rate according to each characteristic, in
453
overall population, ivermectin users, and non-users.
454
Propensity Score Matched Groups
Variable
Overall
(n = 6,068)
Death (%)
Unadjusted
COVID-19 mortality risk
ratio and p-value [p]
Multivariate adjusted
COVID-19 mortality risk
ratio and p-value [p]
Ivermectin use
- n (%)
0.32
(0.20 – 0.49)
[< 0.0001]
0.30
(0.19 – 0.46)
[< 0.0001]
Yes
3,034
25 (0.8%)
No
3,034
79 (2.6%)
Age - n (%)
[< 0.0001]
[< 0.0001]
< 30 y/o
1,691
1
(0.1%)
30-50 y/o
3,155
12 (0.4%)
> 50 y/o
1,222
91 (7.4%)
Sex- n (%)
0.62
0.64
18
(0.42 – 0.91)
[0.015]
(0.44 – 0.93)
[0.022]
Female
3,231
43
(1.3%)
Male
2,837
61
(2.2%)
Race - n (%)
[0.24]
[0.44]
Caucasians
4,398
79
(1.8%)
Afro-Brazilians
193
6
(3.1%)
Mixed
1.364
17
(1.3%)
Asian-Brazilians
113
2
(1.9%)
Type 2 diabetes
- n (%)
10.0
(6.32-15.8)
[< 0.0001]
1.79
(1.03 – 3.12)
[0.041]
Yes
141
20
(14.2%)
No
5,927
84
(1.4%)
Hypertension - n
(%)
8.83
(5.99 – 13.0)
[< 0.0001]
1.98
(1.19 – 3.30)
[0.008]
Yes
343
36
(10.5%)
No
5,725
68
(1.2%)
Asthma
- n (%)
5.64
(1.49 – 21.4)
[0.011]
1.74
(0.52 – 5.81)
[0.36]
Yes
21
2
(9.5%)
No
6,047
102
(1.7%)
COPD
- n (%)
15.0
(5.52 – 40.7)
[< 0.0001]
1.71
(0.68 –4.31)
[0.25]
Yes
12
3
(25.0%)
No
6,056
101
(1.7%)
Cardiovascular
diseases - n (%)
7.54
(2.96 – 19.3)
[< 0.0001]
1.22
(0.44 – 3.37)
[0.70]
Yes
32
4
(12.5%)
No
6,036
100
(1.7%)
Other pulmonary
diseases - n (%)
6.54
(1.02 – 41.9)
[0.048]
5.68
(0.92 – 35.0)
[0.061]
Yes
9
1 (11.1%)
No
6,059
103 (1.7%)
Cancer
(any type)
- n (%)
2.67
(0.39 – 18.3)
[0.32]
1.97
(0.30 – 12.9)
[0.48]
Yes
22
1
(4.6%)
19
No
6,046
103 (1.7%)
Current smoking
- n (%)
1.23
(0.31 – 4.92)
[0.77]
0.36
(0.08 – 1.70)
[0.20]
Yes
95
2 (2.1%)
No
5,973
102 (1.7%)
History of MI
- n (%)
7.35
(1.16 – 46.5)
[0.034]
1.91
(0.17 – 21.6)
[0.60]
Yes
8
1 (12.5%)
No
6,060
103 (1.7%)
History of stroke
- n (%)
17.6
(8.72 – 35.7)
[< 0.0001]
1.97
(0.99 – 3.92)
[0.054]
Yes
21
6 (28.6%)
No
6,047
98 (1.6%)
COPD = Chronic obstructive pulmonary disease; CVD = cardiovascular disease; MI = myocardiac infarction
455
456
In a comparison of city-wide COVID-19 hospitalization rates prior to and during
457
the program, COVID-19 mortality decreased from 6.8% before the program with
458
prophylactic use of ivermectin, to 1.8% after its beginning (RR, 0.27; 95%CO, 0.21 –
459
0.33; p < 0.0001), and in COVID-19 mortality rate, from 3.4% to 1.4% (RR, 0.41; 95%CI
460
0.31 – 0.55; p < 0.0001). (Table 4).
461
462
Table 4. Hospitalization and mortality rates registered in the city of Itajaí, Brazil, before
463
verus after the beginning of the citywide program with ivermectin use as prophylaxis for COVID-
464
19, independent of the ivermectin use status.
465
Overall
Until July
30th
After July
30th
Relative risk ratio
(95%CI)
p-value
Infected COVID-19 population (n)
9956
2663
7293
-
-
Infected non-hospitalized COVID-19 population (n)
9641
2481
7160
-
-
Hospitalized COVID-19 population (n)
315
182
133
-
-
COVID-19 hospitalization rate COVID-19 (%)
3.2%
6.8%
1.8%
0.27 (0.21 – 0.33)
<0.0001
Overall number of COVID-19 deaths
192
90
102
-
-
Overall mortality rate (%)
1.9%
3.4%
1.4%
0.41 (0.31 – 0.55)
<0.0001
466
467
468
469
470
471
20
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
Discussion
495
496
This prospective, citywide COVID-19 ivermectin prophylaxis program resulted
497
in significant reductions of COVID-19 infections, hospitalizations, and deaths. The
498
ivermectin non-users were two times more likely to die from COVID-19 than ivermectin
499
users in the overall population analysis. Since groups were compared for the exposure
500
during the same period, in a parallel manner, changes in transmission rates would
501
affect ivermectin users and non-users equally.
502
503
The city of Itajai, in the state of Santa Catarina, Brazil, started a citywide program of
504
prophylaxis with ivermectin in July 2020 as part of several initiatives to reduce the burden
505
21
of COVID-19. ivermectin was used, based on the existing literature at that time and on
506
the virtual absence of risks. The National Health System (Sistema Único de Saúde – SUS)
507
that functions as a full healthcare support to the entire population allowed the city to
508
establish a non-restricted population program. This program included a support structure
509
consisting of a large outpatient clinic located at the Convention Center of Itajaí. This
510
outpatient clinic became the main locale of assistance for COVID-19 patients, supported
511
by multiple public facilities where general practitioners regularly saw patients.
512
513
The use of ivermectin was optional unless contraindicated, and given upon
514
medical discretion. A structured medical-based program with a medical visit and
515
evaluation of basic demographic characteristics and comorbidities offered ivermectin as
516
an optional prophylaxis to those who agreed to participate in this preventive treatment
517
program. Health status was assessed and data was enterered prospectively throughout the
518
period of the program, in a fully digitzed system provided by the national health system
519
(SUS). Since the system existed prior to the pandemic, a significant number of the
520
population were already registered with their health information, including past and
521
current diseases, use of medications and other characteristics. The adaptations made to
522
the SUS for the pandemic preparedness, prior to the initiation of this ivermectin outpatient
523
program, allowed a structured, well-organized collection of the data that monitored any
524
missing values, reinforcing the reliability of the results.
525
526
An important conservative bias was present. Major risk factors for severe COVID-
527
19 and mortality due to COVID-19, including aging, diabetes, and hypertension, were
528
more present among ivermectin users, which may have underestimated the benefits
529
measured Ivermectin was demonstrated to be particularly effective in subjects above 49
530
years old in terms of reduction of absolute risk, which corresponds to the group at the
531
highest risk for COVID-19. This allows the understanding that prophylactic use of
532
ivermectin can be particularly impactful in older subjects. In addition, ivermectin seemed
533
to reduce the exceeding risk of hypertension, T2D, and other diseases.
534
535
In accordance with the literature, subjects with higher age, diabetes and males
536
were less likely to survive (p < 0.05 for all), only aging remained as an independent risk
537
factor after PSM (p < 0.0001). However, prophylactic ivermectin use appears to mitigate
538
22
the additional risk of COVID-19 death due to T2D, hypertension, and cardiovascular
539
diseases.
540
541
The narrative that using preventive & early treatment therapies will have people
542
relax their caution of remaining socially & physically distanced to allow more COVID-
543
19 related infections is not supported here. This study data demonstrates that the use of
544
preventive ivermectin significantly lowers the infection rate, ands benefits outweigh the
545
supposed increased risk of changes in social behaviours. Hence, we can speculate that the
546
prophylactic use of ivermectin could play an important role in the reduction of the
547
pandemic burden.
548
549
Even after adjustments to measure the most relevant variables that could influence
550
COVID-19 related outcomes, including age, sex, comorbidities, and habits, aiming to
551
avoid overestimation of the effects of ivermectin and to resemble a randomized clinical
552
trial, prophylactic ivermectin proved to be protective for the overall population, with a
553
reduction of 46% in death rate and p < 0.0001 after employment of PSM.
554
555
The protection provided by ivermectin when used prophylactically for COVID-
556
19 may have reflected in the reduction in COVID-19 hospitalization and mortality rates
557
observed in a populational level. Compared to before the beginning of the program,
558
COVID-19 hospitalization and mortality rates were reduced by 73% and 59%,
559
respectively (p < 0.0001 for both). These reductions were obtained when overall
560
population of the city of Itajaí, as well as overall number of COVID-19 cases,
561
hospitalizations, and deaths, were considered, irrespective of the percentage of patients
562
using ivermectin prophylactically. There were no changes in SARS-CoV-2 variants,
563
infectivity and pathogenicity between before and during the program.
564
565
When compared to all other major cities in the State of Santa Catarina, where
566
Itajaí is located, differences in COVID-19 mortality rate between before July 7, 2020 and
567
between July 7, 2020 and December 21, 2020, Itajaí is ranked number one, and far from
568
the second place35. These results indicate that medical-based optional prescription,
569
citywide covered ivermectin can have a positive impact in the healthcare system.
570
However, the present results do not provide sufficient support for the hypothesis
571
that ivermectin could be an alternative to COVID-19 vaccines.
572
23
573
Due to the large number of participants, this citywide program was unable to
574
supervise whether ivermectin users were using ivermectin regularly, although the
575
accumulated number of ivermectin tablets was strictly controlled. This occurred to
576
be a potential another conservative bias, since the effects of ivermectin on prophylaxis
577
could be underestimated due to adherence to the recommended frequency of ivermectin
578
use.
579
580
While ivermectin is a multi-target drug36, its maximum benefits occur when it’s
581
present at minimum concentration in a wide range of sites to inhibit multiple metabolic
582
and inflammatory pathways. However, although the dose of ivermectin employed in the
583
program was smaller than the minimum to reach the concentration required to act in these
584
multiple sites, the reduction in infection, mortality, and death rates in the infected group
585
that used ivermectin prophylactically was surprisingly lower. Long-term or accumulated
586
ivermectin could also play a critical role for its long-term protection against COVID-19.
587
588
Limitations
589
590
Being a prospective observational study which allowed subjects to self select
591
between treatment vs. non-treatment instead of relying on randomization, important
592
confounders may have been differentially present which could otherwise explain the
593
differnces observed. Given that the benefits measured occurred despite negative risk
594
factors being more present in the treatment group, this suggests the benefits are likely
595
accurate and unbiased. Further, studies relying on PSM techniques have been to shown
596
to consistently agree with those employing randomization37,38, again supporting the
597
likelihood the benefits measured are accurate, The prevailing type of SARS-CoV-2 in the
598
city was unknown due to the lack of genotyping surveillance during the period of the
599
program. Whether the prophylaxis proposed in this program would be as effective in other
600
SARS-CoV-2 variants is unclear. Also, there was not a strict control of whether infected
601
subjects used any specific drug in case of COVID-19 infection, this allows the possibility
602
that the differences may be explained by differences in the use of ivermectin or other
603
medications as treatment.
604
605
Final discussion
606
24
607
In this city-wide ivermectin prophylaxis program, a large, statistically significant
608
decrease in mortality rate was observed after the program began among the entire
609
population of city residents. When comparing subjects that used ivermectin regularly,
610
non-users were two times more likely to die from COVID-19 while ivermectin users were
611
7% less likely to be infected with SARS-CoV-2 (p = 0.003).
612
613
Although this study is not a randomized, double-blind, placebo-controlled clinical
614
trial, the data was prospectively collected and resulted in a massive study sample that
615
allowed adjustment for numerous confounding factors, thus strengthening the findings of
616
the present study.
617
618
Due to the well-established, long-term safety profile of ivermectin, with rare
619
adverse effects, the absence of proven therapeutic options to prevent death caused by
620
COVID-19, and while effectiveness of vaccines in real-life all-cause mortality analyses
621
lacks, we recommend that ivermectin could be considered as a preventive strategy, in
622
particular for those at higher risk of complications from COVID-19 or at higher risk of
623
contracting the illness, not as a substitute for COVID-19 vaccines, but as an additional
624
tool, particularly during periods of high transmition rates.
625
626
627
Conclusion
628
629
In a city-wide ivermectin program with prophylactic, optional ivermectin use for COVID-
630
19, ivermectin was associated with significantly reduced COVID-19 infection,
631
hospitalization, and death rates from COVID-19.
632
633
634
635
636
637
Statements
638
639
Conflict of Interest
640
25
641
The authors declare no conflict of interest regarding the drug, ivermectin, and potential
642
commercial benefits of the expansion of its use for COVID-19, or any other related gains.
643
Dr Lucy Kerr received funding from Vitamedic, that manufactures ivermectin, unrelated
644
to this study. Dr. Flavio A. Cadegiani was contracted by Vitamedic for consulting services
645
unrelated to this study, and donated the full budget for COVID-19 patient care and
646
research. Other authors have no conflicts of interest.
647
648
Data availability statement
649
650
Dataset is available under reasonable request by institutions and organizations.
651
652
Author contributions
653
654
Lucy Kerr designed the study. Washington Luiz Olivato Assagra and Fernando Carlos
655
Proença developed the computer program, compiled and ran the data. Raysildo Barbosa
656
Lôbo, Fernando Baldi, Flavio A. Cadegiani and Juan J. Chamie designed and performed
657
the statistical analyses. Lucy Kerr, Flavio A. Cadegiani, Fernando Baldi and Pierre Kory
658
performed the analyses and interpretation of clinical and demographic data generated by
659
the statistical analysis. Fernando Carlos Proença was responsible for the medical
660
surveillance, subjects follow-up and other aspects related to the program administration
661
of the present analysis. Raysildo Barbosa Lôbo and Lucy Kerr were responsible for
662
resources, supervision and project administration related to the analyses. Pierre Kory,
663
Juan J Chamie and Jennifer Hibberd reviewed the data and the manuscript. All authors
664
contributed to the writing of the original draft and final reviewed manuscript. All authors
665
have read and approved the manuscript.
666
667
Funding
668
669
The city of Itajaí acquired the ivermectin, provided the medical and assistant staff and the
670
sites where the citywide programs were conducted. No other funding sources were
671
obtained.
672
673
Acknowledgements
674
675
We acknowledge Dr. Volnei José Morastoni, the city mayor of Itajaí, state of Santa
676
Catarina, Brazil, for developing and enabling the citywide program of ivermectin for
677
COVID-19 prophylaxis. We also acknowledge all the staff that worked at the citywide
678
program for COVID-19 prevention with ivermectin in Itajaí, state of Santa Catarina,
679
Brazil. Also, those who direct- or indirectly offered pro bono support for the subjects that
680
participated in the program, compilation of data, or were involved in any other step that
681
led to the present analysis.
682
683
684
685
686
References
687
688
689
26
1. Chen IS, Kubo Y. Ivermectin and its target molecules: shared and unique
690
modulation mechanisms of ion channels and receptors by Ivermectin. J Physiol.
691
2018 May 15;596(10):1833-1845. doi: 10.1113/JP275236.
692
2. Kaur H, Shekhar N, Sharma S, Sarma P, Prakash A, Medhi B. Ivermectin as a
693
potential drug for treatment of COVID-19: an in-sync review with clinical and
694
computational attributes. Pharmacol Rep. 2021 Jan 3:1-14. doi: 10.1007/s43440-
695
020-00195-y.
696
3. Martin RJ, Robertson AP, Choudhary S. Ivermectin: An Anthelmintic, an
697
Insecticide, and Much More. Trends Parasitol. 2021 Jan;37(1):48-64. doi:
698
10.1016/j.pt.2020.10.005.
699
4. Mastrangelo E, Pezzullo M, De Burghgraeve T, Kaptein S, Pastorino B, Dallmeier
700
K, de Lamballerie X, Neyts J, Hanson AM, Frick DN, Bolognesi M, Milani M.
701
Ivermectin is a potent inhibitor of flavivirus replication specifically targeting NS3
702
helicase activity: new prospects for an old drug. J Antimicrob Chemother. 2012
703
Aug;67(8):1884-94. doi: 10.1093/jac/dks147. Epub 2012 Apr 25. PMID:
704
22535622; PMCID: PMC3888155.
705
5. Wagstaff KM, Sivakumaran H, Heaton SM, Harrich D, Jans DA. Ivermectin is a
706
specific inhibitor of importin α/β-mediated nuclear import able to inhibit
707
replication of HIV-1 and dengue virus. Biochem J. 2012 May 1;443(3):851-6. doi:
708
10.1042/BJ20120150. PMID: 22417684; PMCID: PMC3327999.
709
6. Crump A. Ivermectin: enigmatic multifaceted 'wonder' drug continues to surprise
710
and exceed expectations. J Antibiot (Tokyo). 2017 May;70(5):495-505. doi:
711
10.1038/ja.2017.11.
712
7. Heidary F, Gharebaghi R. Ivermectin: a systematic review from antiviral effects
713
to COVID-19 complementary regimen. J Antibiot (Tokyo). 2020 Sep;73(9):593-
714
602. doi: 10.1038/s41429-020-0336-z.
715
8. Li N, Zhao L, Zhan X. Quantitative proteomics reveals a broad-spectrum antiviral
716
property of ivermectin, benefiting for COVID-19 treatment. J Cell Physiol. 2021
717
Apr;236(4):2959-2975. doi: 10.1002/jcp.30055. Epub 2020 Sep 22. PMID:
718
32959892; PMCID: PMC7536980.
719
9. Jin L, Feng X, Rong H. et al. The antiparasitic drug Ivermectin is a novel FXR
720
ligand that regulates metabolism. Nat Commun 4, 1937 (2013).
721
https://doi.org/10.1038/ncomms2924
722
10. Yang JS, Qi W, Farias-Pereira R, Choi S, Clark JM, Kim D, Park Y. Permethrin
723
and ivermectin modulate lipid metabolism in steatosis-induced HepG2
724
hepatocyte. Food Chem Toxicol. 2019 Mar;125:595-604. doi:
725
10.1016/j.fct.2019.02.005. Epub 2019 Feb 6. PMID: 30738135; PMCID:
726
PMC6527113.
727
11. Cairns DM, Giordano JE, Conte S, Levin M, and Kaplan DL.Ivermectin
728
Promotes Peripheral Nerve Regeneration during Wound Healing. ACS Omega
729
2018 3 (10), 12392-12402. DOI: 10.1021/acsomega.8b01451
730
12. Zheng YY, Ma YT, Zhang JY, et al. COVID-19 and the cardiovascular system.
731
Nat Rev Cardiol. 2020;17:259–60. https://doi.org/10.1038/s41569-020-0360-5
732
13. Nagai H, Satomi T, Abiru A, Miyamoto K, Nagasawa K, Maruyama M, et al.
733
Antihypertrophic effects of small molecules that maintain mitochondrial ATP
734
levels under hypoxia. EBioMedicine 2017;24:147–58.
735
https://doi.org/10.1016/j.ebiom.2017.09.022
736
14. Park A, Iwasaki A, Type I. and type III interferons—induction, signaling, evasion,
737
and application to combat COVID-19. Cell Host Microbe. 2020;27:870–8. doi:
738
27
10.1016/j.chom.2020.05.008. Epub 2020 May 27. PMID: 32464097; PMCID:
739
PMC7255347.
740
15. Zhang X, Song Y, Ci X, et al. Ivermectin inhibits LPS-induced production of
741
inflammatory cytokines and improves LPS-induced survival in mice. Inflamm
742
Res. 2008;57:524–9. https://doi.org/10.1007/s00011-008-8007-8.
743
16. Zaidi, A.K., Dehgani-Mobaraki, P. The mechanisms of action of Ivermectin
744
against SARS-CoV-2: An evidence-based clinical review article. J Antibiot
745
(2021). https://doi.org/10.1038/s41429-021-00430-5
746
17. Matsuyama T, Kubli SP, Yoshinaga SK, et al. An aberrant STAT pathway is
747
central to COVID-19. Cell Death Differ. 2020;27:3209–25.
748
https://doi.org/10.1038/s41418-020-00633-7
749
18. Wang K, Gao W, Dou Q, et al. Ivermectin induces PAK1-mediated cytostatic
750
autophagy in breast cancer. Autophagy. 2016 Dec;12(12):2498-2499. doi:
751
10.1080/15548627.2016.1231494. Dou Q, Chen HN, Wang K, et al.
752
Ivermectin Induces Cytostatic Autophagy by Blocking the PAK1/Akt Axis in
753
Breast Cancer. Cancer Res. 2016 Aug 1;76(15):4457-69. doi: 10.1158/0008-
754
5472.CAN-15-2887. Epub 2016 Jun 14. PMID: 27302166.
755
19. Layhadi JA, Turner J, Crossman D, Fountain SJ. ATP evokes Ca2+ responses and
756
CXCL5 secretion via P2X4 receptor activation in human monocyte-derived
757
macrophages. J Immunol Balt Md 1950 2018;200:1159.
758
https://doi.org/10.4049/jimmunol.1700965
759
20. Juarez M, Schcolnik-Cabrera A, Dueñas-Gonzalez A. The multitargeted drug
760
Ivermectin: from an antiparasitic agent to a repositioned cancer drug. Am J Cancer
761
Res. 2018;8:317–31. Published 2018 Feb 1
762
21. Andersson U, Ottestad W, Tracey KJ. Extracellular HMGB1: a therapeutic target
763
in severe pulmonary inflammation including COVID-19? Mol Med. 2020 May
764
7;26(1):42. doi: 10.1186/s10020-020-00172-4. PMID: 32380958; PMCID:
765
PMC7203545.
766
22. Yan S, Ci X, Chen N, et al. Anti-inflammatory effects of ivermectin in mouse
767
model of allergic asthma. Inflamm Res. 2011 Jun;60(6):589-96. doi:
768
10.1007/s00011-011-0307-8. Epub 2011 Jan 29. PMID: 21279416.
769
23. Reis TAR, Oliveira-da-Silva JA, Tavares GSV, Mendonça DVC, Freitas CS et al.
770
Ivermectin presents effective and selective antileishmanial activity in vitro and in
771
vivo against Leishmania infantum and is therapeutic against visceral
772
leishmaniasis.Exp Parasitol. 2021 Feb;221:108059. doi:
773
10.1016/j.exppara.2020.108059.
774
24. Scheim DE.Ivermectin for COVID 19 treatment Clinical response at quasi-
775
threshold doses via hypothesized alleviation of CD147 mediated vascular
776
occlusive (June, 2020) SS RN:https://SSRN.com/abstract=3636557.
777
25. Ci X, Li H, Yu Q, Zhang X, Yu L, Chen N, Song Y, Deng X. Avermectin exerts
778
anti-inflammatory effect by downregulating the nuclear transcription factor
779
kappa-B and mitogen-activated protein kinase activation pathway. Fundam Clin
780
Pharmacol. 2009 Aug;23(4):449-55. doi: 10.1111/j.1472-8206.2009.00684.x.
781
26. Zaidi AK, Dehgani-Mobaraki P. The mechanisms of action of Ivermectin against
782
SARS-CoV-2: An evidence-based clinical review article. J Antibiot (Tokyo).
783
2021 Jun 15:1-13. doi: 10.1038/s41429-021-00430-5.
784
27. Kalfas S, Visvanathan K, Chan K Drago J. The therapeutic potential of
785
Ivermectin for COVID-19: a systematic review of mechanisms and evidence. doi:
786
https://doi.org/10.1101/2020.11.30.20236570 - PREPRINT
787
28
28. Behera P, Patro BK, Singh AK, Chandanshive PD, S R R, Pradhan SK, Pentapati
788
SSK, Batmanabane G, Mohapatra PR, Padhy BM, Bal SK, Singh SR, Mohanty
789
RR. Role of Ivermectin in the prevention of SARS-CoV-2 infection among
790
healthcare workers in India: A matched case-control study. PLoS One. 2021 Feb
791
16;16(2):e0247163. doi: 10.1371/journal.pone.0247163.
792
29. Hellwig MD, Maia A. A COVID-19 prophylaxis? Lower incidence associated
793
with prophylactic administration of Ivermectin. Int J Antimicrob Agents. 2021
794
Jan;57(1):106248. doi: 10.1016/j.ijantimicag.2020.106248.
795
30. https://www.wma.net/what-we-do/medical-ethics/declaration-of-helsinki/doh-
796
oct2000/ (Last acess: December 24, 2021)
797
31. https://portal.cfm.org.br/images/stories/biblioteca/codigo%20de%20etica%20me
798
dica.pdf (Last acess: December 24, 2021)
799
32. Nguyen TL, Collins GS, Spence J et al. Double-adjustment in propensity score
800
matching analysis: choosing a threshold for considering residual imbalance. BMC
801
Med Res Methodol 17, 78 (2017). https://doi.org/10.1186/s12874-017-0338-0
802
33. Zhang Z, Kim HJ, Lonjon G, Zhu Y; written on behalf of AME Big-Data Clinical
803
Trial Collaborative Group. Balance diagnostics after propensity score
804
matching. Ann Transl Med. 2019;7(1):16. doi:10.21037/atm.2018.12.10.
805
34. http://www.dive.sc.gov.br (Last access: December 24, 2021)
806
35. Choudhury A, Das NC, Patra R, Bhattacharya M, Ghosh P, Patra BC, et al.
807
Exploring the binding efficacy of ivermectin against the key proteins of SARS-
808
CoV-2 pathogenesis: an in silico approach. Future Virol. 2021;10.2217/fvl-2020-
809
0342. https://doi.org/10.2217/fvl-2020-0342.
810
36. Austin PC. An Introduction to Propensity Score Methods for Reducing the Effects
811
of Confounding in Observational Studies. Multivariate Behav Res.
812
2011;46(3):399-424. doi:10.1080/00273171.2011.568786.
813
37. Gray L. Propensity Score Matching in the Absence of Randomized Controlled
814
Trials: A Case Study on the Effects of Breastfeeding on Childhood Obesity. SAGE
815
Research Methods Cases. London: 2020. doi:10.4135/9781529719475. Accessed
816
December 24, 2021.
817
818
819
820
821
Table list
822
823
Table 1. Baseline Characteristics of Subjects Enrolled in Study.
824
Table 2. Infection, horpitalization, death, and mortality rate among ivermectin users and
825
non-users.
826
Table 3. COVID-19 mortality rate according to each characteristic, in overall population,
827
ivermectin users, and non-users.
828
Table 4. Hospitalization and mortality rates registered in the city of Itajaí, Brazil, before
829
verus after the beginning of the citywide program with ivermectin use as prophylaxis for
830
COVID-19, independent of the ivermectin use status.
831
832
833
834
Figure list
835
836
29
Figure 1. Underlying data for the study on ivermectin prophylaxis used for COVID-19.
837
Figure 2. Summary of the findings.
838
839
840
841
842
