LOW–DOSE NALTREXONE FOR LONG COVID
by

Micha Kingston
B.Sc.N., University of Victoria, 2014

PROJECT SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE IN NURSING (FAMILY NURSE PRACTITIONER)

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
April 2025
© Micha Kingston, 2025

ii
Abstract
In August 2024, the number of people impacted by Long COVID (LC) was estimated to be
about 400 million globally (Al-Aly et al., 2024). Currently, there are no FDA–approved
treatments for LC. Many of the centres treating LC have noticed the similarity to myalgic
encephalomyelitis/chronic fatigue syndrome (ME/CFS) in some patients. Low–dose
naltrexone (LDN) is a drug that has recently shown potential in treating ME/CFS. In this
integrative review, ten articles were selected for analysis to investigate whether LDN could
be a viable treatment for LC. Although the studies conducted on LDN and LC are small,
non–randomized, and unblinded, a few interesting themes emerged from this analysis that
could guide future studies and treatment decisions. LDN seems to be most effective for
individuals with a LC phenotype that mimics ME/CFS, as, of all the symptoms of LC, it was
found to be most effective for fatigue and pain. In the future, pre–screening tools will likely
be developed to identify patients most likely to respond to LDN. Two double–blind
randomized controlled trials (RCTs) are currently underway that will be published next year,
yielding a higher degree of evidence and certainty around LDN for LC. In the meantime,
initial findings support consideration of LDN for patients with LC whose primary complaints
are fatigue or pain.

iii
Table of Contents
Abstract

i

Table of Contents

iii

Glossary

iv

Introduction

1

Background

2

Methodology

17

Findings

19

Discussion

25

Limitations

32

Conclusion

33

References

34

Appendix A: PRISMA Flow Diagram

52

Appendix B: Data Extraction Matrix

53

iv
Glossary
Term:

Definition

Hormesis

a biphasic dose response to an environmental agent that has
beneficial effects at low doses and inhibitory or harmful effects at
high doses.

Long COVID

The lingering syndrome affecting multi–organ systems that may
follow an acute infection of the SARS–CoV–2 Virus. In some
instances, this disease state persists for years after initial infection.

Phenotype

The observable characteristics of a disease. Used to sub-categorize
different “symptom clusters” in Long COVID.

Seroconversion

The development of specific antibodies in the blood serum due to
infection or immunization, including vaccination.

Titrate

The process of slowly adjusting the medication dosage to
minimize side effects.

Acronyms:
LC

Long COVID

LDN

Low–dose naltrexone – naltrexone is used in amounts much lower
than classical dosing. No exact range is specified, but it generally
averages less than 10 milligrams/day.

ME/CFS

Myalgic encephalomyelitis/chronic fatigue syndrome

NK cell

Natural killer cell, an immune cell that kills diseased and
dysfunctional cells.

PASC

Post–acute sequelae of SARS–CoV–2 infection, another term for
Long COVID

PCC

Post–COVID-19 Condition, one common medical term for Long
COVID

PESE

symptoms such as disabling fatigue or exhaustion, difficulty
thinking, pain, exercise intolerance, and other symptoms that are
made worse by exertion

v
POTS

Postural orthostatic tachycardia syndrome – a disorder of the
autonomic nervous system characterized by inappropriate changes
in heart rate when transitioning from sitting to standing.

RCT

Randomized controlled trial

TLR4

a protein embedded in the cell wall of glial cells, which is a key
activator of the innate immune response and plays a central role in
the fight against bacterial infections.

TRPM3

Transient receptor potential cation channel subfamily M member 3
– In natural Killer cells (a type of immune cell), this allows calcium
into the cell. The impairment of the TRPM3 protein seen in CFS
results in the dysfunction of NK cells.

1
Low Dose Naltrexone for Long COVID
The ongoing COVID-19 global pandemic is unprecedented and continues to impact every
aspect of society. Although in many ways life has returned to the pre–pandemic “normal,” the
long–term impacts of COVID-19 infection are still being quantified. Most people by now are
aware of the effects of an acute COVID-19 infection; less well known, however, is the fact that
some individuals who clear the initial acute coronavirus infection are left with persistent
manifestations, affecting multiple organ systems (Greenhalgh et al., 2024). The long–term
impact of acute COVID-19 infection is variably referred to as long COVID (LC), post–acute
sequelae of SARS–CoV–2 infection (PASC) (Collins, 2021), or post–COVID-19 condition
(PCC) (Centers for Disease Control and Prevention [CDC], 2022; Collins, 2021). A primary
concern with LC is the difficulties faced when quantifying exact case counts, which will be
discussed in greater detail below.
For several reasons, LC lacks a standardized treatment plan (Ozanic et al., 2025). In the
early days of the pandemic, clinicians working at LC clinics would identify similarities between
PCC and other post–viral conditions and trial established treatments for the conditions they
resemble (Hurt et al., 2024). Some drugs have also been proposed as treatments based on the
presumed pathophysiological underpinnings of the condition (Bonilla, Peluso, et al., 2023). The
WHO (2025) and the U.S. Centers for Disease Control (2024) list more than 200 possible
symptoms associated with PCC. Symptomatically speaking, some phenotypes (or subtypes) of
LC share many similarities to myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS),
which has led to the implementation of established ME/CFS treatments being considered for LC
(Davis et al., 2023a). Although to date, no randomized controlled trials (RCTs) have been
conducted on it, one drug that has gained attention recently in the treatment of CFS and other

2
autoimmune conditions is low–dose naltrexone (LDN) (Polo et al., 2019; Cabanas et al., 2021;
Steenhuysen, 2022). This integrative review aimed to collect, analyze, and synthesize the
evidence surrounding the efficacy of LDN in people living with PCC.
Background
Scale of COVID-19
Quantifying precisely how many people have had a COVID-19 infection is challenging.
Although the statistics from the World Health Organization (WHO, 2025) suggest the total
global case count to be approximately 780 million, in fact, this number could be dramatically
higher as these WHO figures rely on confirmed cases and do not account for asymptomatic
infections or instances where individuals did not meet diagnostic criteria for COVID-19.
According to Alvarez et al. (2023), ascertaining a definitive number is difficult for a few reasons.
These include issues with the testing process (i.e., factors that influence a person's decision to
seek or not seek testing, and errors made with the administration of tests), limited access to
testing centres or self–tests, and delays in processing and reporting tests. Variation in data
collection methods is also problematic, leading to inconsistent data. Badker et al. (2021) found
that another factor that contributes to the difficulty in accurately estimating cases of both
COVID-19 and LC is an inconsistent case definition. In addition to this, there is the confounding
factor of asymptomatic infections (Bergeri et al., 2022) and variability in individual
seroconversion (McConnell et al., 2021). Blood testing for COVID-19 antibodies has
demonstrated that some individuals who never had symptoms of a COVID-19 infection have
antibodies (or have seroconverted) (Banu et al., 2023). In some cases, the opposite may happen;
when tested later, a person who tested positive for an acute COVID-19 infection does not have
antibodies for COVID-19 (Toh et al., 2022).

3
To address the issue of asymptomatic infection, some areas have conducted blood tests
for COVID-19 antibodies to attain a more accurate understanding of the incidence of COVID-19.
However, even here, there is a large variability in seroconversion depending on region. Another
issue with assessing COVID-19 incidence through seroconversion is that some people who have
had confirmed COVID-19 infection will not seroconvert at all (Toh et al., 2022). To assess the
true incidence of COVID-19 infections, some jurisdictions have conducted wastewater analysis
for viral RNA. In Peel, Ontario, wastewater showed the incidence of infection to likely be 6–18
times higher than the rates found with laboratory testing (Cheng et al., n.d.). In short, no
consensus exists on the precise incidence of COVID-19.
Long COVID
The persistent illness that can follow a COVID-19 infection goes by many names. The
term Long COVID was coined by patients. These individuals were also the first to identify this
condition and advocate to bring awareness (Callard & Perego, 2021). For these reasons, the term
Long COVID (or LC) will be preferentially used in this review; however, when directly quoting
or paraphrasing other sources, the original terms will be preserved.
Controversy abounds in the world of LC, with multiple theories about what causes it,
how best to identify it, and how to treat it, with widely disparate estimates of prevalence. This
general lack of agreement is due to the relative novelty of LC, and it is hoped that as more is
studied and discovered about LC, a growing consensus will emerge (WHO, 2025). The lack of
agreement can also be attributed to the highly variable way LC presents itself. The existence of
diverse clinical phenotypes of LC is generally agreed upon if the phenotypes themselves are
described in various ways by different researchers. Another area of confusion with LC is how to
diagnose it. No biomarkers have been established for diagnosing or monitoring LC (Erlandson et

4
al., 2024). As a result, LC remains an entirely clinical diagnosis. This presents diagnostic
challenges as there is no agreement on definitive diagnostic criteria (Tan & Koh, 2023). The
establishment of biomarkers is imperative as they assist clinicians with the diagnosis and
monitoring of the treatment of LC (Davis et al., 2023). Much research is being done in this area;
Putrino from Mount Sinai and Iwasaki from the Yale School of Medicine (n.d.) are two leading
researchers in this area.
These challenges in ascertaining an accurate instance of acute COVID-19 also contribute
to the difficulty in estimating the exact number of people affected by LC. There are also
additional factors that make it difficult to assess the frequency of LC. These include the absence
of definitive diagnostic tests or criteria and the wide variability in symptom onset and duration
(Davis et al., 2023). Due to these factors, estimates of the incidence of LC vary widely. For
example, one study in California found that, depending on which definition was used, LC rates
after COVID-19 infection ranged from 4.9% to 30.9% (Pry et al., 2024).
Another aspect of LC that remains uncertain is the pathophysiological underpinnings of
the syndrome. Fumagalli et al. (2022) identified several risk factors for developing LC, including
overall frailty, age, female gender, and pre–existing Chronic Obstructive Pulmonary Disease.
Similar factors were also identified in a retrospective cohort analysis of over 270,000 by Taquet
et al. (2021). Song and Giuriato (2023) also found that female gender, older age, and specific
pre–existing medical comorbidities were risk factors for developing LC. Al–Aly et al. (2024)
found that with each additional infection with acute COVID-19, an individual's risk of
developing LC increased.
Although objective measures for diagnosing and tracking the progression of LC have not
yet been determined, some lab values that have been considered for assessment include

5
creatinine, C–reactive protein (CRP), D–dimer, ferritin, interleukin–6 (IL–6), IL–10, interferon–
gamma (IFN–γ), lactate dehydrogenase (LDH), leukocytes or white blood, lymphocytes,
monocytes, neutrophils, neutrophil–lymphocyte ratio, N–terminal prohormone of brain
natriuretic peptide (NT–Pro–BNP), platelets, troponin, TNF–alpha (TNF–α), and fibrinogen.
(Yong et al., 2023). Determining biomarkers for LC would greatly assist with diagnosis and
tracking treatment efficacy.
Despite the increasing understanding of who gets LC, there remains no standardized
treatment for people living with this condition (WHO, 2025; Bonilla et al., 2023a) The
challenges of conducting trials in this population (Munblit, Nicholson et al., 2022) mean that the
existing recommendations are often based on low or very low–quality evidence. Most current
recommendations aim to treat symptoms rather than address underlying pathological pathways
(Bonilla et al., 2023a). The Canadian Guidelines for Post-COVID-19 Condition (CAN–PCC,
2025) only have 22 clinical recommendations for LC. These include Metformin, Antihistamines,
and Beta–Blockers, all of which have a low or very low quality of evidence and focus on
alleviating the symptoms rather than the underlying pathology of LC. It is postulated that LDN
might reverse the underlying pathology of LC, rather than simply masking the symptoms
(Steenhuysen, 2022), which is one reason it may be helpful in ways that existing treatments, such
as antihistamines, are not. The relevant pharmacology will be expanded upon below. The WHO
Clinical Management of COVID-19: Living Guideline (WHO, 2023) focuses mainly on various
physical or behavioural therapies and does not have any recommendations for pharmaceutical
approaches. In The Journal of Infection and Chemotherapy, Seo et al. (2024) published a set of
recommendations for treating LC that focus on addressing individual symptoms with existing
therapies, particularly cognitive behavioral therapy and physical therapy. However, the

6
guidelines offer limited discussion of pharmacological options, which may be more appropriate
for patients with severe fatigue.
Due to symptom severity, some people with LC may find physical or cognitive
behavioural therapies impossible due to their debilitating fatigue (DeMars et al., 2023). Those
who experience high levels of fatigue are not good candidates for physical or cognitive
behavioural therapies due to the severity of their symptomatology. For example, those with post–
exertional symptom exacerbation (PESE) may not be candidates for physiotherapy (Long
COVID Physio, 2024). CAN–PCC (2025) recommends limiting exercise–based interventions in
people with PESE to only within the context of research settings. The WHO (2023) also
discusses that the use of physical rehabilitation is not appropriate for some people with LC who
have overwhelming PESE. Appelman et al. (2024) demonstrated that the PESE in LC patients
appears to result from mitochondrial dysfunction, and they stress that exercise should be limited
and paced in individuals experiencing fatigue. This leaves a subset of LC sufferers with minimal
treatment options. In cases such as these, where therapy is not possible, pharmacological options
could be of great benefit. Standardized, evidence–based recommendations for pharmaceutical
treatment are urgently needed as LC continues to impact a growing number of people worldwide.
LC symptom clusters
It has been suggested that classifying all LC presentations into one homogenous category
is problematic, given the disparate ways LC can present (Pfaff et al., 2022; Soriano et al., 2022;
Munblit, O’Hara et al., 2022). Currently, the WHO’s International Classification of Diseases
(ICD), the global standard for naming and classifying diseases, only has one code for LC (WHO,
2019). Pfaff et al. (2022) advocate for the ICD codes to be expanded to include different
phenotypes of LC, arguing that this will enable more precise targeting of clinical studies to well–

7
defined cohorts. This, in turn, should support the creation of more effective, precise treatment
and clinical decision support and assist in accurately estimating the incidence of LC.
The precise number of LC phenotypes and how to classify them remain a topic of debate.
RECOVER (Researching COVID-19 to Enhance Recovery) is a comprehensive research
initiative launched by the U.S. National Institutes of Health (NIH) to understand, diagnose, treat,
and prevent LC. In their analysis of 13,647 participants, the RECOVER initiative identified 52
symptoms of LC and noted that these all occurred in every subtype category. They also identified
five subtypes of LC based on the primary symptom (Geng et al., 2025). These five categories are
subtype 1 – change in taste or smell; subtype 2 – chronic cough; subtype 3 – brain fog; subtype 4
– palpitations; and subtype 5 – post–exertional soreness, dizziness, and gastrointestinal
symptoms. Among their 1796 participants, Gentilotti et al. (2023) identified four LC subtypes:
1– chronic fatigue–like syndrome, subtype 2 – respiratory syndrome, subtype 3 – chronic pain
syndrome, and subtype 4 – neurosensorial syndrome. Epsi et al. (2024) also categorize LC
phenotypes based on symptomatology. They describe these symptom clusters as a
fatigue/difficulty thinking cluster, a difficulty breathing/exercise intolerance cluster, and a
sensory cluster. Conversely, Gerritzen et al. (2023) divide LC phenotypes into three categories
based on overall number rather than type of symptoms (phenotype one comprised participants
with a lower–than–average number of symptoms, phenotype 2 with an average number of
symptoms, and phenotype 3 with a higher–than–average number of symptoms). Another way in
which LC could be categorized is by the time of symptom onset after initial infection. Song and
Giuriato (2023) found that, although most diagnoses of LC were made within the first 1–2
months after the initial COVID-19 diagnosis, there is a second peak of diagnoses that occurs
around one year after the initial COVID-19 infection. It is not known if this later presenting LC

8
is fundamentally different from other presentations of LC. However, it may be important in the
differentiation of discrete phenotypes. There is much work that remains to be done in this area.
Even among those who see the value in subcategorizing LC, there is no clear consensus on what
these phenotypes should be or how they should be qualified.
Some have even argued that LC is not, in fact, a discrete entity; instead, it is simply a
variation of what is sometimes known as Post–ICU syndrome. Hodgson et al. (2022) found that
six months out, there was no significant difference in new disability between COVID-19 and
non–COVID–19 patients who had been mechanically ventilated. However, Ma et al. (2023) note
that LC also presents in patients whose initial COVID-19 infection was mild, which suggests that
it is a discrete condition. This lack of consensus on classifying LC makes it challenging to study
and treat, and as the number of individuals impacted continues to grow, this becomes an
increasingly significant problem.
Issues surrounding LC research
Several issues have been identified that make LC challenging to research. Five years into
the pandemic, the pathophysiology of LC remains an enigma. Theories abound, but no consensus
exists on what causes the over 200 symptoms associated with LC, or whether a single unifying
cause exists. A comprehensive review of the pathophysiology of LC found many theories on this
matter (Castanares–Zapatero et al., 2022). Some theories on the pathophysiological causes of LC
are immune dysregulation (Proal et al., 2023), procoagulatory effects (Pretorius et al., 2021),
endothelial and microcirculatory disturbances (Varga et al., 2020), spike protein persistence
(Swank et al., 2023), coinfections (Peluso et al., 2023), viral reactivations, and microbiome
disturbance (Zuo et al., 2020) as well as metabolic effects (Khunti et al., 2021) and inflammatory
effects (Wijeratne & Crewther, 2020).

9
Another research issue is that, since LC is so new, the scales used to assess related
symptomatology, such as fatigue, have not yet been validated (Thomas et al., 2024). In some
cases, scales used in other conditions, such as ME/CFS, are being repurposed for LC, but these
still need to be clinically evaluated and validated. Wong et al. (2023) argue that Patient Reported
Outcome Measures (PROM) are an important part of determining the phenotypes of LC. As a
result, in the future, the scale or tool used to assess symptoms could help determine a patient's
treatment plan. A 2024 scoping review (Thomas et al.) recommended using either the Functional
Status Scale (FSS) (Krupp et al., 1989), Fatigue Assessment Scale (FAS) (Michielsen et al.,
2004), or the Functional Assessment of Chronic Illness Therapy – Fatigue (FACIT-F) (Webster
et al., 2003) for assessing LC–related fatigue. A 2023 meta–analysis of fatigue in LC (Poole–
Wright et al., 2024) noted the difficulty in comparing data from different studies when different
fatigue tools were used. They also advocate that the researchers studying LC use fatigue scales
validated in other conditions with fatigue as a significant symptom, such as ME/CFS or SLE.
Additionally, Campos et al. (2022) recommend including an objective measure of fatigue in
outcome measurements.
Often, Likert scales are used to rate symptoms. Although considered a standard approach,
this use of Likert scaling is not without issue. The Likert scale was developed nearly a century
ago by Rensis Likert (1932) to measure the degree to which a person agrees or disagrees with a
statement. This method is not without controversy for several reasons; one of them being that it
“cannot be assumed that the difference between responses is equidistant even though the
numbers assigned to those responses are” (Sullivan & Artino, 2013, p. 1). Standardized,
validated tools should be used to assess symptoms such as fatigue. Simple single–dimensional
scales such as the Likert scale should be avoided (Behrangrad & Yoosefinejad, 2021).

10
Long COVID in Primary Care
The exact burden of LC in primary care in Canada has not been well–established, but in
December 2023, 2.1 million people were estimated to be living with long–term symptoms of a
COVID-19 infection (Statistics Canada, 2023). Unfortunately, these numbers are expected to
increase; the Government of Canada (2023) found that LC rates in adults were around 14% after
one COVID-19 infection. However, this rate climbed to nearly 38% after three or more
infections. According to the government of British Columbia (BC), the number of nurse
practitioners working in primary care in BC increased by 11.3% from December 2022 to
December 2023 (BC Gov, 2024). As NPs increasingly provide primary care, and LC is becoming
increasingly prevalent in this setting, it can be expected that the treatment of LC will become
increasingly performed by NPs, making this review particularly relevant to NP practice.
History and mechanism of action of Naltrexone
In recent years, naltrexone, classically used in the treatment of opioid addiction and
alcohol use disorder, has been repurposed as an off–label treatment for a variety of autoimmune,
chronic pain, and inflammatory disorders (Toljan & Vrooman, 2018). When it was first
developed in 1963, naltrexone was originally used for its opioid antagonist properties. It was
approved for opiate use disorder by the FDA in 1984 and for alcohol use disorder in 1994
(Milhorn, 2017). In 1984, physician Dr Bernard Bihari, who was working in the field of
addiction, began exploring naltrexone for its potential immunomodulating properties. Most
information about this discovery of LDN is in the form of an interview, published
posthumously in the non-peer-reviewed journal Alternative Therapies in Health and
Medicine (Bihari, 2013). In this interview, Dr. Bihari discusses how, when he learned that
people with HIV had significantly lower levels of endorphins than those without, he

11
postulated that naltrexone, in low doses, may be able to increase the body's production of
endorphins by transiently blocking endorphin receptors. He then made the connection
between the blockade and the resulting upregulation of endorphin receptors he saw in his
trials with heroin users and the potential to treat the newly discovered HIV infection.
Unfortunately, none of Dr. Bihari’s work in this area has been published in peer–reviewed
journals, and he could not continue his research in this vein due to a lack of funding.
However, through presentations to physicians and word of mouth communication within the
HIV/AIDS community, the use of LDN slowly became more well–known and accepted.
What initially was used to boost endorphin levels in people with HIV, in more recent years,
has been investigated for its anti-inflammatory and immunoregulatory properties in many
diseases involving immune dysfunction.
The grassroots history of the development of LDN is reflected in the current situation,
with word of mouth substantially contributing to its uptake (Raknes & Småbrekke, 2017).
Following the release of a popular documentary on LDN, Norway saw a significant increase
in LDN prescriptions. In 2012, there were only 14 people in Norway using LDN; in 2013,
following the airing of the aforementioned documentary on LDN in multiple sclerosis, LDN
prescriptions in Norway rose to more than 11,000 (Raknes & Småbrekke, 2017).
LDN is also being investigated for use in chronic pain, which is controversial and
remains off–label due to a lack of large, high–quality RCTs demonstrating efficacy. Despite
this, naltrexone use has continued to increase, driven in part by the abundant anecdotal
evidence and word of mouth in online chronic pain and autoimmune community spaces
(Raknes & Småbrekke, 2017). An example of this is the popular Facebook group LDN
Research Trust (n.d.), which has over 61,000 members from around the globe.

12
In its FDA–approved role, naltrexone is used in doses of 50–100 mg/day for alcohol
use disorder. At this dose, it acts primarily by blocking the mu–opioid receptor, responsible
for releasing dopamine in the nucleus accumbens of the user's brain, hence the positive
mental state, or high feeling accompanying this dopamine release (Swift, 1999). It has a suite
of proposed effects in much lower doses, ranging from 1–5 mg/day. Although much is
known about the mechanism of action of naltrexone at classical dosing, the precise
mechanism of action of LDN remains unproven (Toljan & Vrooman, 2018). This is likely due
to Naltrexone lacking FDA approval for dosages other than 50–100 mg and settings other
than addiction medicine (US Food and Drug Administration, 2013), so studies looking at the
pharmacology of naltrexone in lower doses are lacking. It also may be because LDN works
on multiple pathways in the body.
Despite the lack of high–quality pharmacological information on LDN, there are
several theories as to why it might be helpful in chronic pain, cancer, and autoimmune
conditions. One mechanism behind LDN’s efficacy in chronic pain, which has been known
since the 1980s, is the temporary blockade of opioid receptors. This blockade results in the
body's compensatory production of endorphins (Kosten et al., 1986), which helps improve
pain level. It is also proposed that naltrexone in low doses exerts a paradoxical effect that leads
to a reduced affinity for the µ–opioid receptor and an increased affinity for the toll–like receptor
4 (TLR4) (Toljan & Brooman, 2018). TLR4s are found in the cell membranes of microglial
cells, the most prominent immune cells of the CNS. In 2013, Stevens et al. found that both
opioids, such as fentanyl and morphine, and opioid antagonists (in this case, naloxone) interact
with TLR4s and so impact inflammation and the immune system. However, this study was
conducted in vitro, so caution must be used when applying these results to the in vivo setting.

13
Cabanas et al. (2019) recently discovered another mechanism of action of LDN: its impact on
regulating the function of natural killer (NK) cells. They found that naltrexone restores function
in dysfunctional NK cells in ME/CFS patients.
LDN is also being considered for psychiatric reasons beyond addiction medicine. A proof
of concept trial (Mischoulon et al., 2017) showed that LDN might benefit patients with major
depressive disorder who are experiencing breakthrough symptoms while on other
antidepressants. This is far from definitive proof of its efficacy, but it does show that it may hold
promise as an adjunctive treatment for depression. Unfortunately, perhaps due to some of the
issues discussed below, LDN has not been further studied for depression; however, a small
retrospective survey study found that LDN reduced anxiety in people with multiple sclerosis
(MS) during the COVID-19 pandemic (McLaughlin et al., 2022). The results of both studies
suggest that LDN may eventually be found to have beneficial effects on psychiatric conditions
such as anxiety and depression. In some instances, mental health concerns, such as anxiety and
depression, are dominant symptoms of LC in some phenotypes (Bautista–Rodriguez et al., 2023).
In these cases, LDN may be found to be a viable treatment. 2–4.5
Dosing and Duration of LDN
Due to the lack of extensive studies, it is challenging to determine the optimal dose of
LDN. Dr. Bihari (2013) conducted a small study, which was never published, looking at dose
size and timing to raise endorphins most effectively. He found that doses between 1.75 and 5 mg
in one nightly dose were the best for most individuals. However, he acknowledged that this was
a small study and that there was significant variation amongst individuals, so he encouraged
future research into this area as the sample size of his studies were too small to be generalizable
to the population at large. Due to a lack of high–quality evidence and the fact that LDN use is

14
exclusively off–label, the most effective dosing remains unproven. One resource for suggestions
on LDN dosing is the LDN Research Trust, a non–profit charity focused on funding research and
disseminating information on LDN. The LDN Research Trust recommends starting at a low dose
and slowly titrating to avoid or minimize side effects (LDN Research Trust, 2024). The schedule
they recommend (based on the analysis of the existing literature and the combined anecdotal
experience of their pharmacists and prescribers) is 0.5– 1 mg daily for 14 days, increasing by 0.5
to 1 mg every 2 weeks until 4.5mg or the highest tolerated dose. In this manner, side effects can
be minimized. This is like the dosing schedule found most effective by Marcus et al. (2024) in
their observational study on the best dosing of LDN for chronic pain. Marcus et al. (2024) also
observed that LDN seems to operate in a hormetic manner, meaning there is an optimal dosing
range where the drug is most effective and that increasing or decreasing beyond this range does
not translate into a better effect; which was also what Bihari found in his early studies in the
1980’s (Bihari, 2013).
The dosing of LDN for chronic pain is generally 2–4.5 mg/day (Rassi–Mariani et al.,
2024). In fibromyalgia, the doses that have been studied range from 0.1–9 mg/day, with most
sources using 4.5 mg/day (Yang et al., 2023). Dosing for most conditions is not well–established
due to a lack of high–quality RCTs, but the usual dose is around 4.5 mg/day for a diverse range
of conditions, including multiple sclerosis, Crohn’s disease, cancer, Hailey–Hailey disease, and
complex regional pain syndrome (Toljan & Vrooman, 2018).
Another consideration when looking at the efficacy of LDN is the duration of therapy
needed to see optimal effects. There is also a lack of high–quality evidence, and much of the
information is anecdotal. The LDN research trust (n.d.) speaks to this, and their physician, Dr
Yusuf Saleeby, recommends that patients trial LDN for at least three months before drawing any

15
conclusions about efficacy. Another factor to consider with duration is that if LDN is titrated
slowly, it can lead to the optimal dose not being reached for several weeks, so the duration of
therapy should be adjusted to accommodate this. Furthermore, one author (Marcus et al., 2024)
found that some patients in their study had previously trialled LDN and found it ineffective.
However, when they used a gradual titration schedule, rather than starting at a relatively higher
dose of ~4.5 mg/day, many of the patients who in the past had not responded to LDN ended up
finding it effective. This finding led Leiber and Parker (2025) to make the following
recommendation: “previous studies ruling out LDN as an effective treatment for certain disease
states might warrant re-evaluation if titration was not utilized” (Leiber & Parker, 2025, p. 12)
Safety considerations
Naltrexone has a well–established safety profile and has been in use at doses of 50–100
mg since the mid–1980s without any recorded serious adverse effects. For years, however, there
was a black box warning on the label for hepatotoxicity. The U.S. Food and Drug Administration
(FDA, 2013) removed this in 2013 because it was based on liver damage seen at doses of 300 mg
or higher, and there are no known cases of hepatic failure resulting from naltrexone.
Furthermore, although dosing for LDN varies, it is generally in the neighbourhood of one–
sixtieth of the dose shown to potentially induce hepatic injury. Because of this much lower dose,
it is thought that LDN is unlikely to provoke hepatic injury in an uncompromised patient.
Despite the lack of significant adverse events associated with naltrexone use, minor adverse
events are relatively common. However, they do not appear to occur more than in placebo
groups (Bolton et al., 2019). The established safety of LDN makes it a good candidate for
investigating for novel uses.

16
Side effects and interactions
Common side effects of LDN are mild and include vivid dreams, insomnia, nausea, and
headache (Leiber & Parker, 2025). These are relatively rare and usually tolerable. These side
effects are often transient, disappear with time, and can be minimized with appropriate titration.
A significant potential interaction to consider with Naltrexone is that it can precipitate
opiate withdrawal and therefore should be avoided in people who are on opiate medications
(U.S. National Library of Medicine, 2021). Naltrexone may also interact with the anti-psychotic
thioridazine and alcohol misuse treatment disulfiram. There are no known interactions with herbs
or vitamins. The American Society of Addiction Medicine (2013) also notes that the above
interactions are for Naltrexone at FDA–approved dosing; it is not known if naltrexone at lower
doses also carries the same risks.
Potential issues regarding naltrexone research
Despite many small–scale studies on a myriad of autoimmune and pain conditions, to
date, there have been few high–quality randomized controlled trials on LDN. One contributing
factor may be that the patent on naltrexone has long since expired; indeed, in its oral form,
naltrexone is only available as a generic (Seabright, 2023). It is well known that as the patent
expires on a drug, the market price of the drug decreases significantly (Vondeling et al., 2018).
Without financial reward, private pharmaceutical corporations have limited incentives to study
generics such as naltrexone in novel therapeutic indications.
This situation has led to alternative funding sources, such as publicly funded studies. In
2010, a pilot study on LDN in MS was funded entirely by private contributions from MS patients
(Cree et al., 2010). Other studies have been funded by patient advocacy groups such as the LDN
Research Trust (Puttick, 2007). Studies of this type raise several concerns, particularly related to

17
the ethics surrounding the potential exploitation of people who are already facing hardship due to
their illness. Additionally, if the patients funding the study are permitted to direct its design, this
raises potential validity issues. Wenner et al. (2015) summarize the many potential issues raised
by publicly funded trials (PFT), stating that
“The reconfiguration of research relationships seen in PFTs is prone to inefficiencies,
market–unproven, ineffective, and perhaps even dangerous interventions to patients
desperate for a cure” (Wenner et al., 2015, p. 1).
Methodology
Study Design
This is an integrative review of low–dose naltrexone in Long COVID. Whittemore and
Knafl's (2005) methodology guided this integrative review. The research question was formatted
as a PIO question (Population, Intervention, Outcome) based on the model developed by
Richardson et al. (1995). The research question was “How does low–dose naltrexone impact
people living with Long COVID?”
Search Strategy
The databases selected were CINAHL and MEDLINE, as I expected this to capture the
most pertinent literature. In addition, a Google Scholar search was completed to ensure that all
relevant literature was included for analysis. Although this review focuses on low–dose
naltrexone, to cast a wide net and ensure that all significant literature was identified, the search
included “naltrexone” rather than “low–dose naltrexone.” The initial search included “COVID19” as a unique term, but it was evident that this yielded many articles that were not specifically
about LC. The MeSH terms settled upon for the final search included the keywords “naltrexone”
and “Long COVID” or “chronic COVID-19” or “post COVID-19” or “long haul COVID-19” or
“post COVID-19”.

18

Inclusion and Exclusion Criteria
All articles were screened initially with a title review, then a review of the abstract, and
finally, a full–text review. Studies that were not published in English were excluded, as were
those that were not full–text. The inclusion criteria were that studies must focus on LC, not acute
COVID-19. Additionally, the dose of naltrexone discussed must be in the “low–dose” range, not
the higher dosage typically used in addiction medicine.
Search Results
The search was conducted on February 8, 2025, and yielded 17 results, 12 from
MEDLINE and five from CINAHL. After duplicates were removed, there were 15 results. Please
see Appendix A for a PRISMA Flow Diagram of the selection process. Of the 15 articles, one
was excluded for not being in English, two for not being full text, and the remaining 12 were
selected for abstract review. Of these, four were excluded for not focusing on LC. This left eight
selected for review. At this point, Google Scholar was searched with the exact keywords, and the
first 30 results were reviewed for inclusion. Of these, two were selected for inclusion. The total
number of articles selected for analysis was 10.
Analysis
The study design was critically appraised using the Critical Appraisal Skills Program
(CASP) checklists, including the cohort, RCT, and cross–sectional study checklists (Critical
Appraisal Skills Program, 2024). The content of each study was then noted in a matrix to
compare similarities and identify themes. This matrix was based on the model suggested by
Toronto and Remington (2020). See Appendix B.

19
Findings
This section describes the characteristics of the selected studies, followed by an analysis
of four themes which emerged in the studies: a) symptoms/phenotypes assessed and those most
improved by LDN, b) delivery methods (titration schedule, dose, duration), c) side effects and
safety, and d) outcome measures tracked (scales used and objective measures). An additional
theme that will be discussed is the findings of the laboratory study. All ten articles selected for
review found that LDN improves symptoms associated with LC.
Study Characteristics
The date ranges of the studies were from 2022 to 2024. Although, thus far, no
randomized controlled trials have been conducted on LDN for LC, the selected articles represent
a diverse range of methodologies. Two of the articles are retrospective cohort reviews (Tamariz
et al., 2024; Bonilla et al., 2023b), one is an interventional pre–post study (O’Kelly et al., 2022),
one is a cross–sectional study (Hurt et al., 2024), and one is an observational study (Isman et al.,
2024). In addition, there is one lab study (Sasso et al., 2024), a “mini–review” (Dietz &
Brondstater, 2024), and a scoping review (Livieratos et al., 2024), as well as one study protocol
(Naik et al., 2024) and one case study (Petracek et al., 2023).
Unsurprisingly, given the international nature of the COVID-19 pandemic, the selected
articles originate from around the globe. Five are from the USA (Dietz & Brondstater, 2024;
Hurt et al., 2024; Isman et al., 2024; Petracek et al., 2023; Tamariz et al., 2024). Of the
remainder, one originates in the Netherlands (Bonilla et al., 2023b), one from Canada (Naik et
al., 2024), one from Greece (Livieratos et al., 2024), one from Australia (Sasso et al., 2024), and
lastly, one from Ireland (O’Kelly et al., 2022).

20
Apart from the lab study (Sasso et al., 2024), all but one (Isman et al., 2024) of the
assessed studies occurred at a LC clinic. Isman et al. (2024) do not specify the study setting.
During the analysis of the articles, a few key themes emerged, which will be discussed in further
detail below. The first theme that will be expanded upon involves the tracked and treated
symptoms in these studies.
Theme One: Symptoms and Phenotypes Most Improved by LDN
Across the studies assessed in this review, one of the most explored areas was which
symptoms LDN was most effective for and whether some phenotypes may respond better to
LDN. The articles in this review reflect the general lack of consensus about which symptoms
define LC. Overall, fatigue was the most common symptom that was assessed. This was
monitored across all the studies in the review (Bonilla et al., 2023b; Hurt et al., 2024; Isman et
al., 2024; O’Kelly et al., 2022; Petracek et al., 2023; Tamariz et al., 2024) as well as in the two
reviews (Dietz & Brondstater, 2024; Livieratos et al., 2024). Sleep disturbances were also
commonly assessed. Different studies focused on different aspects of sleep. Bonilla et al. (2023b)
looked at unrefreshing sleep and abnormal sleep patterns, Hurt et al. (2024) assessed insomnia,
and O’Kelly et al. (2022) focused on the nonspecific term “sleep disturbances”. Pain was
investigated by Hurt et al. (2024), Isman et al. (2024), O’Kelly et al. (2022), and Tamariz et al.
(2024). Interestingly, shortness of breath was only explicitly stated as being tracked in two cases
(Hurt et al., 2024; O’Kelly et al., 2022).
Symptoms most improved by LDN
Several studies noted that LDN seemed more effective for some symptoms than others.
Bonilla et al. (2023b), O’Kelly et al. (2022), and Tamariz et al. (2024) reported that LDN was
most effective for pain. Both Bonilla et al. (2023b) and O’Kelly et al. (2022) assessed pain using

21
the same scale, the Visual Analog Scale (VAS). However, Tamariz et al. (2024) do not specify
how pain was assessed. Many studies also found fatigue to be one of the symptoms most
improved by LDN; these included Bonilla et al. (2023b), Tamariz et al. (2024), Isman et al.
(2024), and Petracek et al. (2023). Interestingly, Isman et al. (2024) note that in their study of 36
participants aged 29–69 years, fatigue seemed to be more improved in females than males;
however, they also note this may be due to the much lower number of males enrolled (11 vs 25
females) and so improvement in fatigue did not reach statistical significance amongst male
participants.
Bonilla et al. (2023b) also noted an overall lower number of symptoms after LDN and an
overall improvement in the general functionality of participants, which was echoed by Petracek
et al. (2023). Overall, fatigue was the symptom that LDN seemed to help most among the LC
symptomatology profile. In addition, several studies (Bonilla et al., 2023b; Hurt et al., 2024;
Petracek et al., 2023) noted different phenotypes within their studies, including ME/CFS–like,
POTS–like, and fibromyalgia-like symptom clusters.
Theme Two: Variation in Delivery Methods (Schedule, Dose, and Duration)
Another area of variation was how LDN was administered, in terms of dosing, duration,
and different titration schedules. The range of LDN doses in this review reflects the general lack
of standardization. This will be discussed in further detail below.
Dose Range
Bonilla et al. (2023b) and Tamariz et al. (2024) used an individualized dosing schedule
and did not find a correlation between dose and response. Isman et al. (2024) and Petracek et al.
(2023) used the standard 4.5 mg/day dose. In the cross–sectional survey by Hurt et al. (2024), the
doses are not specified, but they do refer to the dosing of naltrexone as being low.

22
Time of Day
Isman et al. (2024) were the only authors to specify the time of day the LDN was
administered. They instructed participants to take LDN at bedtime, which aligns with Dr.
Bihari's suggestions (as interviewed in Alternative Therapies in Health and Medicine, 2013),
who found nighttime dosing was the best for boosting endorphin production. Unfortunately,
other authors did not address the time of day LDN was taken.
Titration
Three studies (Bonilla et al., 2023b; Isman et al., 2024; Tamariz et al., 2024) used a
gradual titration schedule. Isman et al. (2024) up-titrated rapidly to achieve a therapeutic dose
quickly. In contrast, O’Kelly et al. (2022) did one month of LDN therapy at 1 mg, followed by
one month of LDN at 2 mg. Hurt et al. (2024) do not mention if the LDN was titrated, as their
survey was focused on identifying potential therapeutic candidates for LC for future trials, rather
than determining optimal dosing strategies.
Duration
In terms of duration of therapy, given the study designs (retrospective or surveys), there
is a wide range of duration of LDN therapy. For example, in their retrospective review, Bonilla
et al. (2023b) found the duration of LDN use was 77–255 days. The study by Isman et al. (2024)
lasted for 12 weeks. O’Kelly et al. (2022) administered 2 months of LDN. The patient in the case
study by Petracek et al. (2023) had been on LDN for 16 months at the time of publishing. The
participants in the study by Tamariz et al. (2024) were on LDN for at least 4 weeks, some longer,
but the exact amount of time is not specified.

23
To summarize, the range of LDN doses (0.5–6 mg/day) was generally in keeping with
that used in other conditions. The duration of therapy ranged from 77 days to 16 months.
Appropriate titration was used in some, but not all, studies.
Theme three: Adverse effects
A third key theme that emerged across the studies was the absence of severe adverse
effects. The interventional pre–post study by O’Kelly et al. (2022) was, in part, designed to
assess the safety of LDN in a PCC cohort. They found no side effects in 94.7% of participants,
which is in keeping with the known safety profile of naltrexone. The side effects mentioned in
their study were mild, with one participant reporting diarrhea and another fatigue. Three studies
(Bonilla, Tian, et al., 2023; Hurt et al., 2024; Tamariz et al., 2024) do not mention any adverse
effects. It is unclear whether this reflects an absence of adverse effects or a lack of reporting. In
keeping with previous LDN studies, Isman et al. (2024) found that side effects were “mild” and
more common at the beginning of treatment and could generally be managed by decreasing drug
dosage. Four of 36 participants withdrew from this study due to adverse effects, three of whom
left in the first few weeks. Isman et al. (2024) note that, generally, LDN is up-titrated gradually
to minimize adverse events. In contrast, in this study, they titrated rapidly to achieve a
therapeutic dose sooner, which may have increased the rates or severity of side effects.
Theme four: Outcome Measures Tracked (scales and objective measures)
Various scales were used across these studies; only one incorporated objective
data. Many scales were used in the studies in this review. For example, fatigue alone was
assessed with several different scales. Bonilla et al. (2023b) and O’Kelly et al. (2022) used the
Fatigue Severity Scale (FSS). Tamariz et al. (2024) used the FACIT–fatigue scale, and Petracek
et al. (2023) used the Multidimensional Fatigue Inventory (MFI). Isman et al. (2024) use the SF–

24
36 symptom scale and a modified Chalder scale, which includes Likert scaling rather than the
typical binary scale seen in the Chalder scale. Many studies (Bonilla, Tian, et al., 2023; Hurt et
al., 2024; O’Kelly et al., 2022) used Likert scaling of individual symptoms to rate change in
these symptoms. Bonilla et al. (2023b) also used a modified FSS scale for assessing fatigue.
Objective measures
Only Tamariz et al. (2024) included any biomarkers in their study. They looked at CRP
and morning cortisol and found a non–statistically significant normalization in both these levels
in participants who had taken LDN. Although they screened participants for a few biomarkers on
intake, O’Kelly et al. (2022) did so only to rule out other conditions, rather than track therapy.
Theme Five: Laboratory Evidence
One of the articles (Sasso et al., 2024) took a mechanistic approach, exploring
naltrexone's potential effects on LC in a laboratory setting. The authors aim to investigate two
primary outcomes with their study: 1) whether the TRPM3 ion channels in LC patients were
impaired similarly to ME/CFS patients, and 2) to investigate the effects of naltrexone on TRPM3
ion channel activity in LC patients. They confirmed impaired TRPM3 channel function in both
ME/CFS and LC patients, and that TRPM3 currents were significantly restored in naltrexonetreated natural killer cells.
Study Design and Reporting Limitations
There were several limitations shared among the articles assessed in this review. Firstly,
none of the included studies were high–quality RCTs. All the studies lacked randomization,
blinding, and a control arm. In one study (Isman et al., 2024), participants received two
interventions simultaneously (both NAD+ and LDN),

25
Furthermore, most studies did not provide the LDN for their patients. In all but one case (Isman
et al., 2024), the participants were given scripts and retrieved the LDN from a pharmacy of their
choice. In the Bonilla et al. (2023b) study, many participants did not complete the follow–up
questionnaire. Only 59/207 patients participated in the follow–up questionnaire, which is a very
low number.
Another limitation is the number of studies that did not track or report adverse events.
Three studies did not make any mention of adverse events at all (Bonilla et al., 2023b; Hurt et al.,
2024; Tamariz et al., 2024), and it is impossible to say if this is because there were none or
simply that they were not tracked. Despite the limitations, the studies share some interesting
themes that could help guide future investigations.
Discussion
Interpretation of key findings
In this review on LDN for LC, several studies identified fatigue as the symptom for
which LDN appears to be most effective. Other symptoms that positively impacted included pain
and a variety of sleep disturbances. As a result, in future studies and clinical practice, the use of
LDN for LC should focus on the ME/CFS phenotype. This hypothesis is supported by the lab
study conducted by Sasso et al. (2024), who identified that the CFS phenotype of LC involved
Transient Receptor Potential Melastatin 3 (TRPM3) ion channel dysfunction affecting natural
killer (NK) cells which is known to also impact NK cells in ME/CFS (Löhn & Wirth, 2024).
Sasso et al. confirmed the presence of impaired TRPM3 ion channel function in NK cells from
people with LC. They also demonstrated in vitro that faulty TRPM3 currents were significantly
restored in naltrexone–treated NK cells from LC patients. This may explain why the evidence in
this review indicates it is most effective in improving the ME/CFS phenotype. This finding also

26
fits with the limited evidence on the efficacy of LDN in various conditions involving fatigue,
such as ME/CFS(Cabanas et al., 2021) or fibromyalgia(Yang et al., 2023).
In clinical practice, distinct phenotypes have not yet been defined or added to the ICD,
which complicates clinical categorization. In the meantime, initial evidence suggests that LDN is
most useful for fatigue, pain, and sleep disturbances, so clinicians could discuss the potential use
of LDN with patients who present with these as primary symptoms. Despite the lack of high–
quality RCTs, given the promising initial evidence in conjunction with LDN's well–established
safety profile, some patients may consider it worthwhile to trial LDN. In this case, clinicians
should support this choice. This is in keeping with the recommendations made by CAN–PCC
(2025), who state in their guideline on LDN:
“In light of very limited direct evidence for the potential benefits or harms of low–dose
naltrexone among people with post COVID-19 condition, some people with post
COVID-19 condition may still decide to trial low–dose naltrexone if they place a
relatively higher value on some desirable outcomes and a relatively lower value on the
undesirable outcomes (such as adverse effects), especially if they live in a jurisdiction
where low–dose naltrexone treatment is feasible.”
The CAN–PCC intends to revisit its guidelines on LDN in 2026 or sooner if more evidence
becomes available.
Measurement challenges and recommendations
The diverse range of self–reported outcome measures, which use different scales, makes
it hard to compare results between studies. Additionally, the studies that use an unvalidated scale
risk casting their findings into question. Many studies used Likert scaling to rate symptoms,
which, although considered a standard approach, is not without issue as discussed above. One

27
study used an FSS, which aligns with recommendations on studying fatigue in LC. (Thomas et
al., 2024). The FSS scale was explicitly designed by Klok et al. (2020) to track functional status
in LC over time and support research. Therefore, its use by Bonilla et al. (2023b) is
methodologically justifiable.
Assessing fatigue is difficult, as there are many facets to what constitutes fatigue. Fatigue
can impact all aspects of a person's life, so the best assessment instrument for an accurate picture
of this would be a multidimensional tool. It has also been suggested that studies on LC fatigue
should include an objective measure of fatigability as well as subjective assessments of fatigue
(Campos et al., 2022). Due to the lack of objective serum measurements, an objective fatigability
measure, such as a grip test, walking test, or a variety of cognitive tests, could increase the
credibility of future studies.
The Potential Role of Objective Measures of LC
Another factor that could enhance the credibility of future studies on LC would be the use
of objective lab values. As mentioned earlier, there are currently no standardized tests to
diagnose or track LC. Therefore, these studies' lack of objective measures is unsurprising. In the
meantime, using an objective measure of fatigability could add validity to studies conducted in
this area.
Although standardized testing for lab markers has not been established, Yong et al.
(2023) report that CRP, D–dimer, LDH, and leukocyte levels have been found to be elevated in
people with LC compared to their counterparts who recovered completely from their COVID-19
infection. Much work remains to be done in this area. However, future studies that look at LC
should consider adding these objective lab values to further support the results obtained in their
subjective symptom analyses. Only one study in this review tracked any lab tests, and this

28
showed that the participants who took LDN had a non–significant reduction in CRP and an
increase (normalization) in cortisol (Tamariz et al., 2024).
Also, in future studies, serum testing could be performed to identify who will likely be a
responder or non–responder to LDN. A study in fibromyalgia found that the people who
responded best to LDN were those whose baseline erythrocyte sedimentation rate was higher.
(Younger et al., 2009). Ideally, in the future, objective measures such as this will be identified in
people living with LC so treatments can be tailored to their specific phenotypes.
As these objective measurements are discovered, they need to be included in future
research into LC. Though measuring subjective symptoms such as fatigue is important, looking
at biomarkers can give a clear sense of whether a treatment is improving the underlying
pathology and not simply the symptoms. If the end goal is to cure, objective measures must be
identified and included in future studies.
Dosing and Duration Considerations for Future LDN Trials
Another factor that should be considered in future studies using LDN is the duration of
therapy. Many patients report that the full effect of LDN may not be felt for up to several months
(LDN Research Trust, 2023). This suggests that any studies evaluating the efficacy of LDN
should ideally run for at least several months to assess its impact adequately.
The lack of adequate titration, coupled with the short duration of therapy in many of the
reviewed studies, limits the ability to assess the efficacy of LDN. Based on preliminary evidence
from its use in other conditions and anecdotal evidence from the LDN research trust (2024),
LDN may require a gradual titration and a longer duration of treatment to ensure full therapeutic
effect. Therefore, it is possible that some participants who may have responded to LDN under

29
ideal conditions did not experience its benefits within the timeframes and the titration schedules
used. Future trials should consider these factors and incorporate them into the study design.
The dosages of LDN seen in this review (1–6 mg daily) were in line with what is
generally used in other autoimmune or pain conditions. (Rupp et al., 2023; Vatvani et al., 2024).
Most commonly, patients were prescribed the same dose for the duration of the study. As
previously outlined, the LDN Research Trust (2024) recommends a slow titration schedule to
minimize adverse events, starting at 0.5–1 mg daily and increasing every two weeks until
reaching a target dose.
Although many of the reviewed studies used gradual titration, all of them did so in a
faster manner than is recommended by the LDN trust. Future studies should take this into
account when deciding how to dose. Like many drugs, if patients are started on an adequate dose
from the beginning, side effects are more likely (Caffrey & Borrelli, 2020). To minimize these
risks, LDN should be up–titrated gradually, in line with the recommendations of the LDN
Research Trust.
Another consideration with LDN is the duration of treatment needed to see results.
Although there is a dearth of high–quality evidence discussing the time frame in which the
effects of LDN are optimally felt, there is an abundance of anecdotal information surrounding
this topic. As a slow-titration process delays reaching the therapeutic dose and experiencing full
effects, ideally, any future studies on LDN would consider this, and to get an accurate idea of
efficacy, trial LDN for an extended period before reassessing. Aitcheson et al. (2023)
recommend trialling LDN for at least 3 months before making any assumptions about its
efficacy.

30
Despite the funding challenges associated with the study of generic drugs, a few ongoing
trials are looking at LDN for LC. The NALCOVID-19 study in Australia will include 56 LC
patients and be a double–blinded RCT (Marshall–Gradisnik, 2024–2025), currently in the
recruitment phase. The University of British Columbia also has a study underway of 160
participants for LDN in LC (Nacul, 2024–2025). The duration of treatment will be 16 weeks, and
the LDN dose will be 4.5 mg a day, which aligns with the recommendations regarding dosing
and duration of LDN. The study also incorporates a gradual titration schedule, which includes an
increase of 1 mg per week to a ceiling dose of 4.5 mg/day or the highest tolerated dose. In
addition to self–reported ratings of fatigue and other symptoms, the researchers also plan to track
several blood measures and objective measures of fatigability. This study will be a double–blind
RCT, and its high–quality study design should provide more robust data regarding LDN and LC.
This trial is expected to conclude in August 2025.
Clinical and Research Implications
One important consideration regarding the clinical use of the LC phenotype is its
potential to facilitate applying for long–term disability (LTD). According to Littlejohn Barristers
(n.d.), having a medical diagnosis can improve the likelihood of LTD approval. As discussed
above, for some individuals with LC, particularly those whose symptoms persist for an extended
period, returning to work might not be possible. In cases where a phenotype mimics a more
established condition, such as ME/CFS or POTS, using this as the diagnosis may facilitate an
LTD application more effectively than the more ambiguous diagnosis of LC. This clustering of
patients into phenotypical categories was justified by Hurt et al. (2024), who state:
For example, if LC patients meet the criteria for these complex chronic illnesses, we
would diagnose them as LC–induced ME/CFS if we determine COVID-19 to be the
inducing factor. … In addition to guiding treatment choices with some modicum of
success, this approach has helped when LC patients file for medical disability (p.8)

31
Limitations of the Reviewed Studies
There were many limitations found in the articles looked at in this review. The most
obvious is the study design, which introduced a strong possibility of a placebo effect. Many
studies were single–arm, lacking a control group. (Bonilla et al., 2023b; Hurt et al., 2024; Isman
et al., 2024; O’Kelly et al., 2022; Petracek et al., 2023; Tamariz et al., 2024). None of the studies
used proper double–blinding, so in most cases, participants were aware they were taking LDN,
which might have influenced self–reported outcome measures. Bonilla et al. (2023b) also noted
that many participants did not complete the follow–up questionnaire, which may have skewed
the results towards participants with positive experiences with LDN. Another factor contributing
to the potential for a placebo effect was that, in many cases, they independently sourced their
own LDN, raising concerns of variation in drug formulation (i.e., different fillers) and blinding.
This also led to offloading the cost onto the trial participant. In one case (Isman et al., 2024)
LDN was given along with NAD+, which made it impossible to say which effects, either positive
or negative, were due to which intervention. The authors originally intended this to be an RCT
but changed the study design to a single–arm study due to low enrolment levels.
Another possible sign of a placebo effect is that some authors (e.g., Bonilla et al., 2023b)
observed no dose–response effect, which may suggest that improvements could have been partly
due to a placebo effect. Despite these flaws in study design, the fact that all the studies report
improvement in symptoms does suggest LDN may have therapeutic potential in the treatment of
LC. However, these findings must be interpreted cautiously until more rigorous RCTs are
completed.
Another potential issue is the possibility that participants may not have had LC, for
example, in the study by Isman et al., (2024) the range of time from a positive COVID-19 test to

32
the initiation of therapy was 27–624 days, meaning that at least one study participant (the one
with 27 days since the positive COVID-19 test) may not have met the diagnostic threshold for
Long COVID. It may have instead been experiencing a longer course of acute COVID-19
infection.
In one case, the study was funded by a private anti–aging company called AgelessRx. This
could introduce bias, as the company markets and sells LDN on their website (AgelessRx, n.d.)
Limitations in this review
There are a few potential limitations in this review. Firstly, due to the many names for
LC, it is possible that not all relevant literature was identified and analysed due to inadequate
search terms. This possibility was minimized by several MeSH terms (LC, PCC, PASC) and
supplementing database searches using a Google Scholar search, which uses a search algorithm
different from that of medical databases, to ensure all relevant literature was captured.
A second potential limitation is that the author herself is an advocate of LDN and uses it
personally to manage health issues, so, despite all attempts to maintain neutrality, there is still a
possibility of interpretive bias. Thirdly, by excluding non–English articles, it is possible that
some relevant information was missed. Although only one non–English study was found in the
search, it is still possible that this linguistic limitation could have led to a less comprehensive
review.
Future Directions for Research
The information presented in this review suggests three key areas that should be
prioritized for future research. Firstly, there is a critical need for standardized subcategories of
LC, based on phenotypes, to be identified and codified. This would facilitate future research and
assist with establishing standardized treatments for these phenotypes. Secondly, these

33
phenotypes should be added to the International Classification of Diseases (ICD) for the same
reasons and for more accurate epidemiological tracking. Thirdly, increased funding is needed to
support trials for generic drugs such as naltrexone. With limited profitability, pharmaceutical
corporations are unlikely to conduct these trials, making public and non–profit funding sources
essential to conducting these studies.
Conclusion
This integrative review examined the impact of LDN on LC. CINAHL,
MEDLINE, and Google Scholar were searched, and a total of ten articles met the inclusion
criteria and were analyzed to answer the research question, “How does low–dose naltrexone
impact people living with Long COVID?”
During the analysis, several themes emerged that could be used to guide future research
into this area. Key themes included the potential that LDN was effective for specific symptoms
or phenotypes of LC (specifically fatigue, pain, and sleep disturbances), a general lack of
significant adverse events, and inconsistency between dosage/duration recommendations and
those seen in the studies. Until the results of ongoing RCTs are published, it is difficult to make
definitive recommendations regarding the use of LDN for LC. However, given the encouraging
preliminary findings, the lack of FDA–approved treatments for LC, and the well–known safety
profile of naltrexone, it may be reasonable for the primary care provider to consider initiating a
trial of LDN in LC patients whose primary complaint is fatigue or pain. Once the ongoing trials
by Nacul et al. and Marshall–Gradisnik et al. are completed and published, a clearer picture of
the impact of LDN on LC should emerge.

34
References
AgelessRx. (n.d.). LDN. Retrieved 1 April, 2025 from https://agelessrx.com/ldn/
Aitcheson, N., Lin, Z., & Tynan, K. (2023). Low–dose naltrexone in the treatment of
fibromyalgia. Australian Journal of General Practice, 52(4), 231–235. Retrieved April
18, 2025, from https://www1.racgp.org.au/ajgp/2023/april/low–dose–naltrexone–in–the–
treatment–of–fibromyalg
Al–Aly, Z., Davis, H., McCorkell, L., Soares, L., Wulf–Hanson, S., Iwasaki, A., & Topol, E. J.
(2024). Long COVID science, research, and policy. Nature Medicine, 30(8), 2148–2164.
https://doi.org/10.1038/s41591–024–03173–6
Alvarez, E., Bielska, I. A., Hopkins, S., Belal, A. A., Goldstein, D. M., Slick, J.,
Pavalagantharajah, S., Wynfield, A., Dakey, S., Gedeon, M.–C., Alam, E., & Bouzanis,
K. (2023). Limitations of COVID-19 testing and case data for evidence–informed health
policy and practice. Health Research Policy and Systems, 21(1), 11.
https://doi.org/10.1186/s12961–023–00963–1
American Society of Addiction Medicine. (2013). Naltrexone: Clinical use and pharmacology.
Principles of Addiction Medicine (5th ed.).
https://www.ncbi.nlm.nih.gov/books/NBK143177/
Badker, R., Miller, K., Pardee, C., Oppenheim, B., Stephenson, N., Ash, B., Philippsen, T.,
Ngoon, C., Savage, P., Lam, C., & Madhav, N. (2021). Challenges in reported COVID19 data: Best practices and recommendations for future epidemics. BMJ Global Health,
6(5). https://doi.org/10.1136/bmjgh–2021–005542

35
Banu ST, Vinotha S, Katragadda R, Vanaja R. (2023). Seroconversion in COVID-19 Infection
and Comparison of Antibody Responses in Symptomatic Versus Asymptomatic
Individuals. J Pure Appl Microbiol 17(1). 590–596. doi: 10.22207/JPAM.17.1.56
Bautista–Rodriguez, E., Cortés–Álvarez, N., Vuelvas–Olmos, C., Reyes–Meza, V., González–
López, T., Flores–delosÁngeles, C., Pérez–Silva, N., Aguirre–Alarcón, H., Cortez–
Sanchez, J., Rocha–Rocha, V., Escobedo–Straffon,J., Contreras–Mioni, L., & and Reyes–
Vergara, M.–L. (2023). Stress, anxiety, depression, and long COVID symptoms. Fatigue:
Biomedicine, Health & Behavior, 11(1), 35–54.
https://doi.org/10.1080/21641846.2022.2154500
Bihari B. (2013). Low–dose naltrexone for normalizing immune system function. Alternative
Therapies in Health and Medicine, 19(2), 56–65.
Bolton, M., Hodkinson, A., Boda, S., Mould, A., Panagioti, M., Rhodes, S., Riste, L., & van
Marwijk, H. (2019). Serious adverse events reported in placebo randomised controlled
trials of oral naltrexone: A systematic review and meta–analysis. BMC Medicine, 17(1),
10. https://doi.org/10.1186/s12916–018–1242–0
Bonilla, H., Peluso, M. J., Rodgers, K., Aberg, J. A., Patterson, T. F., Tamburro, R., Baizer, L.,
Goldman, J. D., Rouphael, N., Deitchman, A., Fine, J., Fontelo, P., Kim, A. Y., Shaw, G.,
Stratford, J., Ceger, P., Costantine, M. M., Fisher, L., O’Brien, L., … McComsey, G. A.
(2023a). Therapeutic trials for long COVID: A call to action from the interventions
taskforce of the RECOVER initiative. Frontiers in Immunology, 14, 1129459.
https://doi.org/10.3389/fimmu.2023.1129459
Bonilla, H., Tian, L., Marconi, V. C., Shafer, R., McComsey, G. A., Miglis, M., Yang, P.,
Bonilla, A., Eggert, L., & Geng, L. N. (2023b). Low–dose naltrexone use for the

36
management of post–acute sequelae of COVID-19. International Immunopharmacology,
124(Pt B), 110966. https://doi.org/10.1016/j.intimp.2023.110966
Cabanas, H., Muraki, K., Staines, D., & Marshall–Gradisnik, S. (2019). Naltrexone Restores
Impaired Transient Receptor Potential Melastatin 3 Ion Channel Function in Natural
Killer Cells From Myalgic Encephalomyelitis/Chronic Fatigue Syndrome Patients.
Frontiers in Immunology, 10, 2545. https://doi.org/10.3389/fimmu.2019.02545
Cabanas, H., Muraki, K., Eaton–Fitch, N., Staines, D. R., & Marshall–Gradisnik, S. (2021).
Potential Therapeutic Benefit of Low Dose Naltrexone in Myalgic
Encephalomyelitis/Chronic Fatigue Syndrome: Role of Transient Receptor Potential
Melastatin 3 Ion Channels in Pathophysiology and Treatment. Frontiers in Immunology,
12. https://doi.org/10.3389/fimmu.2021.687806
Caffrey, A. R., & Borrelli, E. P. (2020). The art and science of drug titration. Therapeutic
Advances in Drug Safety, 11, 2042098620958910.
https://doi.org/10.1177/2042098620958910
Campos, M. C., Nery, T., Starke, A. C., de Bem Alves, A. C., Speck, A. E., & S Aguiar, A.
(2022). Post–viral fatigue in COVID-19: A review of symptom assessment methods,
mental, cognitive, and physical impairment. Neuroscience & Biobehavioral Reviews,
142, 104902. https://doi.org/10.1016/j.neubiorev.2022.104902
Canadian Guidelines for post COVID-19 condition (CAN–PCC). Recommendation topic:
Clinical interventions for PCC. Retrieved 25 March, 2025 from https://can–
pcc.recmap.org/recommendations?recommendationIntent=clinical_interventions

37
Canadian Guidelines for Post COVID-19 Condition (CAN–PCC). (2025). Recommendation:
[Recommendation on Low–Dose Naltrexone]. Retrieved April 18, 2025, from
https://can–pcc.recmap.org/recommendation/24c96691–691b–4dca–aeef–49a45791d319
Castanares–Zapatero, D., Chalon ,P., Kohn ,L., Dauvrin ,M., Detollenaere ,J., Maertens de
Noordhout ,C., Primus–de Jong ,C., Cleemput ,I., & and Van den Heede, K. (2022).
Pathophysiology and mechanism of long COVID: A comprehensive review. Annals of
Medicine, 54(1), 1473–1487. https://doi.org/10.1080/07853890.2022.2076901
Centres for Disease Control and Prevention, (2022) Post–COVID-19 conditions: Information for
healthcare providers. Retrieved 15 April from
https://archive.cdc.gov/www_cdc_gov/coronavirus/2019–ncov/hcp/clinical–care/post–
COVID-19 –conditions.html?
Centers for Disease Control and Prevention (2024). Signs and symptoms of long COVID.
https://www.cdc.gov/COVID-19 /long–term–effects/long–COVID-19 –signs–
symptoms.html
Cheng, L., Dhiyebi, H. A., Varia, M., Atanas, K., Srikanthan, N., Hayat, S., Ikert, H., Fuzzen,
M., Sing–Judge, C., Badlani, Y., Zeeb, E., Bragg, L. M., Delatolla, R., Giesy, J. P.,
Gilliland, E., & Servos, M. R. (2023). Omicron COVID-19 Case Estimates Based on
Previous SARS–CoV–2 Wastewater Load, Regional Municipality of Peel, Ontario,
Canada—Volume 29, Number 8—August 2023—Emerging Infectious Diseases journal—
CDC. https://doi.org/10.3201/eid2908.221580
Collins, F. (2021). NIH launches new initiative to study “Long COVID”. National Institutes of
Health (NIH). https://www.nih.gov/about–nih/who–we–are/nih–director/statements/nih–
launches–new–initiative–study–long–COVID

38
Critical Appraisal Skills Programme (2024). CASP Randomised Controlled Trial) Checklist.
[online] Retrieved 10 December, 2024 from https://casp–uk.net/casp–tools–
checklists/randomised–controlled–trial–rct–checklist/
Critical Appraisal Skills Programme (2024). CASP Cohort Study Checklist. [online] Retrieved
10 December, 2024 from https://casp–uk.net/casp–tools–checklists/cohort–study–
checklist/
Critical Appraisal Skills Programme (2024). CASP Cross–sectional studies Checklist. [online] ]
Retrieved 10 December, 2024 from https://casp–uk.net/casp–tools–checklists/cross–
sectional–studies–checklist/
Cree, B. A. C., Kornyeyeva, E., & Goodin, D. S. (2010). Pilot trial of low–dose naltrexone and
quality of life in multiple sclerosis. Annals of Neurology, 68(2), 145–150.
https://doi.org/10.1002/ana.22006
Davis, H. E., McCorkell, L., Vogel, J. M., & Topol, E. J. (2023). Long COVID: Major findings,
mechanisms, and recommendations. Nature Reviews Microbiology, 21(3), 133–146.
https://doi.org/10.1038/s41579–022–00846–2
DeMars, J., Brown, D. A., Angelidis, I., Jones, F., McGuire, F., O’Brien, K. K., Oller, D.,
Pemberton, S., Tarrant, R., Verduzco–Gutierrez, M., & Gross, D. P. (2023). What is Safe
Long COVID Rehabilitation? Journal of Occupational Rehabilitation, 33(2), 227–230.
https://doi.org/10.1007/s10926–022–10075–2
Dietz, T. K., & Brondstater, K. N. (2024). Long COVID management: A mini review of current
recommendations and underutilized modalities. Frontiers in Medicine, 11, 1430444.
https://doi.org/10.3389/fmed.2024.1430444

39
Erlandson, K. M., Geng, L. N., Selvaggi, C. A., Thaweethai, T., Chen, P., Erdmann, N. B.,
Goldman, J. D., Henrich, T. J., Hornig, M., Karlson, E. W., Katz, S. D., Kim, C., Cribbs,
S. K., Laiyemo, A. O., Letts, R., Lin, J. Y., Marathe, J., Parthasarathy, S., Patterson, T.
F., … RECOVER–Adult Cohort. (2024). Differentiation of Prior SARS–CoV–2 Infection
and Postacute Sequelae by Standard Clinical Laboratory Measurements in the
RECOVER Cohort. Annals of Internal Medicine, 177(9), 1209–1221.
https://doi.org/10.7326/M24–0737
Food and Drug Administration (2024). Human Drug Compounding. Retrieved 18 April, 2025
from https://www.fda.gov/drugs/guidance–compliance–regulatory–information/human–
drug–compounding
Fumagalli, C., Zocchi, C., Tassetti, L., Silverii, M. V., Amato, C., Livi, L., Giovannoni, L.,
Verrillo, F., Bartoloni, A., Marcucci, R., Lavorini, F., Fumagalli, S., Ungar, A., Olivotto,
I., Rasero, L., Fattirolli, F., & Marchionni, N. (2022). Factors associated with persistence
of symptoms 1 year after COVID-19: A longitudinal, prospective phone–based interview
follow–up cohort study. European Journal of Internal Medicine, 97, 36–41.
https://doi.org/10.1016/j.ejim.2021.11.018
Geng, L. N., Erlandson, K. M., Hornig, M., Letts, R., Selvaggi, C., Ashktorab, H., Atieh, O.,
Bartram, L., Brim, H., Brosnahan, S. B., Brown, J., Castro, M., Charney, A., Chen, P.,
Deeks, S. G., Erdmann, N., Flaherman, V. J., Ghamloush, M. A., Goepfert, P., …
RECOVER Consortium. (2025). 2024 Update of the RECOVER–Adult Long COVID
Research Index. JAMA, 333(8), 694–700. https://doi.org/10.1001/jama.2024.24184
Gentilotti, E., Górska, A., Tami, A., Gusinow, R., Mirandola, M., Baño, J. R., Baena, Z. R. P.,
Rossi, E., Hasenauer, J., Lopes–Rafegas, I., Righi, E., Caroccia, N., Cataudella, S.,

40
Pasquini, Z., Osmo, T., Piccolo, L. D., Savoldi, A., Kumar–Singh, S., Mazzaferri, F., …
Tacconelli, E. (2023). Clinical phenotypes and quality of life to define post–COVID-19
syndrome: A cluster analysis of the multinational, prospective ORCHESTRA cohort.
eClinicalMedicine, 62. https://doi.org/10.1016/j.eclinm.2023.102107
Gerritzen, I., Brus, I. M., Spronk, I., Biere–Rafi, S., Polinder, S., & Haagsma, J. A. (2023).
Identification of post–COVID-19 condition phenotypes, and differences in health–related
quality of life and healthcare use: A cluster analysis. Epidemiology and Infection, 151,
e123. https://doi.org/10.1017/S0950268823001139
Greenhalgh, T., Sivan, M., Perlowski, A., & Nikolich, J. Ž. (2024). Long COVID: A clinical
update. The Lancet, 404(10453), 707–724. https://doi.org/10.1016/S0140–
6736(24)01136–X
Hodgson, C. L., Higgins, A. M., Bailey, M. J., Mather, A. M., Beach, L., Bellomo, R., Bissett,
B., Boden, I. J., Bradley, S., Burrell, A., Cooper, D. J., Fulcher, B. J., Haines, K. J.,
Hodgson, I. T., Hopkins, J., Jones, A. Y. M., Lane, S., Lawrence, D., van der Lee, L., …
Scott, B. (2022). Comparison of 6–Month Outcomes of Survivors of COVID-19 versus
Non–COVID-19 Critical Illness. American Journal of Respiratory and Critical Care
Medicine, 205(10), 1159–1168. https://doi.org/10.1164/rccm.202110–2335OC
Hou, Y., Gu, T., Ni, Z., Shi, X., Ranney, M. L., & Mukherjee, B. (2025). Global Prevalence of
Long COVID, its Subtypes and Risk factors: An Updated Systematic Review and Meta–
Analysis (p. 2025.01.01.24319384). medRxiv.
https://doi.org/10.1101/2025.01.01.24319384
Hurt, R. T., Yadav, S., Schroeder, D. R., Croghan, I. T., Mueller, M. R., Grach, S. L., Aakre, C.
A., Gilman, E. A., Stephenson, C. R., Overgaard, J., Collins, N. M., Lawson, D. K.,

41
Thompson, A. M., Natividad, L. T., Mohamed Elfadil, O., & Ganesh, R. (2024).
Longitudinal Progression of Patients with Long COVID Treated in a Post–COVID-19
Clinic: A Cross–Sectional Survey. Journal of Primary Care & Community Health, 1–11.
https://doi.org/10.1177/21501319241258671
Isman, A., Nyquist, A., Strecker, B., Harinath, G., Lee, V., Zhang, X., & Zalzala, S. (2024).
Low–dose naltrexone and NAD+ for the treatment of patients with persistent fatigue
symptoms after COVID-19 -19. Brain, Behavior, & Immunity – Health, 36, 100733.
https://doi.org/10.1016/j.bbih.2024.100733
Klok, F. A., Boon, G. J. A. M., Barco, S., Endres, M., Geelhoed, J. J. M., Knauss, S., Rezek, S.
A., Spruit, M. A., Vehreschild, J., & Siegerink, B. (2020). The Post–COVID-19
Functional Status scale: A tool to measure functional status over time after COVID-19 19. The European Respiratory Journal, 56(1), 2001494.
https://doi.org/10.1183/13993003.01494–2020
LDN Research Trust (n.d.). Home [Facebook Page]. Facebook. Retrieved 30 March, 2025, from
https://www.facebook.com/groups/LDNRT
LDN Research Trust. (n.d.). How long does it take to notice the effects of low dose naltrexone
(LDN)? LDN Research Trust. Retrieved 5 March, 2025 from
https://ldnresearchtrust.org/how–long–does–it–take–notice–effects–low–dose–
naltrexone–ldn%C2%A0
LDN Research Trust (2020). LDN 2020 dosing information for prescribers. Retrieved March 1,
2025, from
https://ldnresearchtrust.org/sites/default/files/2020–04/Dosing–Info–a4_0.pdf
LDN Research Trust (2023). How long does it take LDN to work?

42
Retrieved February 20, 2025 from https://ldnresearchtrust.org/how–long–does–it–take–
low–dose–naltrexone–work–news
LDN Research Trust (2024). Dosing Guide for Prescribers
Retrieved 17 February, 2025 from https://ldnresearchtrust.org/sites/default/files/2024–
02/Dosing–Guide–2024_0.pdf
Leiber, K. K., & Parker R. W. (2025). Therapeutic Uses and Efficacy of Low–Dose Naltrexone:
A Scoping Review. Cureus, 17. https://doi.org/10.7759/cureus.81086
Likert, R. (1932). A technique for the measurement of attitudes. Archives of Psychology, 22
(140), 55
Littlejohn Barristers. (n.d.). The role of medical evidence in Canadian long–term disability
claims. Retrieved April 18, 2025, from https://www.littlejohnbarristers.com/post/the–
role–of–medical–evidence–in–canadian–long–term–disability–claims
Livieratos, A., Gogos, C., & Akinosoglou, K. (2024). Beyond Antivirals: Alternative Therapies
for Long COVID. Viruses, 16(11). https://doi.org/10.3390/v16111795
Long COVID-19 Physio. (n.d.). Post–exertional symptom exacerbation (PESE). Retrieved
March 15, 2025, from https://longCOVID-19 .physio/post–exertional–symptom–
exacerbation,
Ma, Y., Deng, J., Liu, Q., Du, M., Liu, M., & Liu, J. (2023). Long–Term Consequences of
Asymptomatic SARS–CoV–2 Infection: A Systematic Review and Meta–Analysis.
International Journal of Environmental Research and Public Health, 20(2), 1613.
https://doi.org/10.3390/ijerph20021613

43
Marcus, N. J., Robbins, L., Araki, A., Gracely, E. J., & Theoharides, T. C. (2024). Effective
Doses of Low–Dose Naltrexone for Chronic Pain – An Observational Study. Journal of
Pain Research, 17, 1273–1284. https://doi.org/10.2147/JPR.S451183
Marshall–Gradisnik, S. (Principal Investigator) (2024–2025)
Efficacy of Low Dose Naltrexone for the treatment of symptoms of Post COVID-19
Condition (NALCOVID-19 study) [Clinical Trial ACTRN12623001042639]
https://www.anzctr.org.au/Trial/Registration/TrialReview.aspx?id=386522
McConnell, D., Hickey, C., Bargary, N., Trela–Larsen, L., Walsh, C., & Barry, M. (2021).
Understanding the challenges and uncertainties of seroprevalence studies for SARS–
CoV–2. International Journal of Environmental Research and Public Health, 18(9),
4640. https://doi.org/10.3390/ijerph18094640
McLaughlin, P. J., Odom, L. B., Arnett, P. A., Orehek, S., Thomas, G. A., & Zagon, I. S. (2022).
Low–dose naltrexone reduced anxiety in persons with multiple sclerosis during the
COVID-19 pandemic. International Immunopharmacology, 113, 109438.
https://doi.org/10.1016/j.intimp.2022.109438
MedWatch Safety Information. (2013). U.S.: Department of Health and Human Services; Safety
Labeling Changes Approved By FDA Center for Drug Evaluation and Research
(CDER). https://wayback.archive–
it.org/7993/20161023084600/http://www.fda.gov/Safety/MedWatch/SafetyInformation/u
cm374896.htm. Accessed 15 Jan 2025
Milhorn, H. T. (2017). Substance Use Disorders: A Guide for the Primary Care Provider.
Springer.

44
Mischoulon, D., Hylek, L., Yeung, A. S., Clain, A. J., Baer, L., Cusin, C., Ionescu, D. F., Alpert,
J. E., Soskin, D. P., & Fava, M. (2017). Randomized, proof–of–concept trial of low dose
naltrexone for patients with breakthrough symptoms of major depressive disorder on
antidepressants. Journal of Affective Disorders, 208, 6–14.
https://doi.org/10.1016/j.jad.2016.08.029
Munblit, D., Nicholson, T. R., Needham, D. M., Seylanova, N., Parr, C., Chen, J., Kokorina, A.,
Sigfrid, L., Buonsenso, D., Bhatnagar, S., Thiruvengadam, R., Parker, A. M., Preller, J.,
Avdeev, S., Klok, F. A., Tong, A., Diaz, J. V., Groote, W. D., Schiess, N., …
Williamson, P. R. (2022). Studying the post–COVID–19 condition: Research challenges,
strategies, and importance of Core Outcome Set development. BMC Medicine, 20(1), 50.
https://doi.org/10.1186/s12916–021–02222–y
Munblit, D., O’Hara, M. E., Akrami, A., Perego, E., Olliaro, P., & Needham, D. M. (2022). Long
COVID: Aiming for a consensus. The Lancet Respiratory Medicine, 10(7), 632–634.
https://doi.org/10.1016/S2213–2600(22)00135–7
Nacul, L. (Principal investigator) (2024–2025) Low–dose Naltrexone for Post–COVID-19
Fatigue Syndrome. [Clinical Trial NCT:05430152]
https://clinicaltrials.gov/study/NCT05430152?term=low%20dose%20naltrexone%20and
%20COVID-19 &rank=1
Naik, H., Cooke, E., Boulter, T., Dyer, R., Bone, J. N., Tsai, M., Cristobal, J., McKay, R. J.,
Song, X., & Nacul, L. (2024). Low–dose naltrexone for post–COVID-19 fatigue
syndrome: A study protocol for a double–blind, randomised trial in British Columbia.
BMJ Open, 14(5), e085272. https://doi.org/10.1136/bmjopen–2024–085272

45
NIH launches new initiative to study “Long COVID”. (2021, February 23). National Institutes of
Health (NIH). https://www.nih.gov/about–nih/who–we–are/nih–director/statements/nih–
launches–new–initiative–study–long–COVID-19
O’Kelly, B., Vidal, L., McHugh, T., Woo, J., Avramovic, G., & Lambert, J. S. (2022). Safety
and efficacy of low dose naltrexone in a long COVID cohort; an interventional pre–post
study. Brain, Behavior, & Immunity – Health, 24, 100485.
https://doi.org/10.1016/j.bbih.2022.100485
Ozanic, K., Watanabe, A. S. A., Machado, A. B. F., da Silva, V. L., Dias, V. C., & Diniz, C. G.
(2025). Long COVID: General Perceptions and Challenges in Diagnosis and
Management. COVID-19, 5(3), Article 3. https://doi.org/10.3390/COVID-19 5030041
Petracek, L. S., Broussard, C. A., Swope, R. L., & Rowe, P. C. (2023). A Case Study of
Successful Application of the Principles of ME/CFS Care to an Individual with Long
COVID. Healthcare (2227–9032), 11(6), 865.
https://doi.org/10.3390/healthcare11060865
Pfaff, E. R., Madlock–Brown, C., Baratta, J. M., Bhatia, A., Davis, H., Girvin, A., Hill, E., Kelly,
L., Kostka, K., Loomba, J., McMurry, J. A., Wong, R., Bennett, T. D., Moffitt, R., Chute,
C. G., & Haendel, M. (2022). Coding Long COVID: Characterizing a new disease
through an ICD–10 lens. medRxiv, 2022.04.18.22273968.
https://doi.org/10.1101/2022.04.18.22273968
Polo, O., Pesonen, P., & Tuominen, E. (2019). Low–dose naltrexone in the treatment of myalgic
encephalomyelitis/chronic fatigue syndrome (ME/CFS). Fatigue: Biomedicine, Health &
Behavior, 7(4), 207–217. https://doi.org/10.1080/21641846.2019.1692770

46
Poole–Wright, K., Guennouni, I., Sterry, O., Evans, R. A., Gaughran, F., & Chalder, T. (2023).
Fatigue outcomes following COVID: A systematic review and meta–analysis. BMJ Open,
13(4), e063969. https://doi.org/10.1136/bmjopen–2022–063969
Pry, J. M., McCullough, K., Lai, K. W.–J., Lim, E., Mehrotra, M. L., Lamba, K., & Jain, S.
(2024). Defining long COVID using a population–based SARS–CoV–2 survey in
California. Vaccine, 42(26), 126358. https://doi.org/10.1016/j.vaccine.2024.126358
Public Health Agency of Canada. (2023). Post COVID-19 condition (PCC). Health Infobase.
https://health–infobase.canada.ca/COVID-19 -19/post–COVID-19 –condition/
Puttick, H. (2007). MS hope from heroin addiction drug. The Herald. Retrieved 14 January 2025
from https://www.heraldscotland.com/news/1494742.0/
Raknes, G., & Småbrekke, L. (2017). Low dose naltrexone in multiple sclerosis: Effects on
medication use. A quasi–experimental study. PLOS ONE, 12(11), e0187423.
https://doi.org/10.1371/journal.pone.0187423
Rassi–Mariani, V., Barreto, E. S. R., Antunes, C. R., Alencar, V. B., Falcão Lins–Kusterer, L. E.,
Azi, L. M. T. de A., & Kraychete, D. C. (2024). The use of naltrexone in the treatment of
chronic pain: A systematic review. Pain Management, 14(8), 453–463.
https://doi.org/10.1080/17581869.2024.2401769
Richardson, W. S., Wilson, M. C., Nishikawa, J., & Hayward, R. S. A. (1995). The well–built
clinical question: A key to evidence–based decisions. ACP Journal Club, 123(3), A12.
https://doi.org/10.7326/ACPJC–1995–123–3–A12
Rupp, A., Young, E., & Chadwick, A. L. (2023). Low–dose naltrexone’s utility for non–cancer
centralized pain conditions: A scoping review. Pain Medicine, 24(11), 1270–1281.
https://doi.org/10.1093/pm/pnad074

47
Sasso, E. M., Muraki, K., Eaton–Fitch, N., Smith, P., Jeremijenko, A., Griffin, P., & Marshall–
Gradisnik, S. (2024). Investigation into the restoration of TRPM3 ion channel activity in
post–COVID–19 condition: A potential pharmacotherapeutic target. Frontiers in
Immunology, 15, 1264702. https://doi.org/10.3389/fimmu.2024.1264702
Seabright, J. (2023, October 26). Naltrexone (Vivitrol) and cost. MedicalNewsToday. Retrieved
5 January 2025 from https://www.medicalnewstoday.com/articles/drugs–naltrexone–
cost#faq
Seo, J.–W., Kim, S. E., Kim, Y., Kim, E. J., Kim, T., Kim, T., Lee, S. H., Lee, E., Lee, J., Seo,
Y. B., Jeong, Y.–H., Jung, Y. H., Choi, Y. J., & Song, J. Y. (2024). Updated Clinical
Practice Guidelines for the Diagnosis and Management of Long COVID. Infection &
Chemotherapy, 56(1), 122–157. https://doi.org/10.3947/ic.2024.0024
Song, Z., & Giuriato, M. (2023). Demographic And Clinical Factors Associated With Long
COVID. Health Affairs, 42(3), 433–442. https://doi.org/10.1377/hlthaff.2022.00991
Soriano, J. B., Murthy, S., Marshall, J. C., Relan, P., & Diaz, J. V. (2022). A clinical case
definition of post–COVID–19 condition by a Delphi consensus. The Lancet Infectious
Diseases, 22(4), e102–e107. https://doi.org/10.1016/S1473–3099(21)00703–9
Statistics Canada. (2023). Between April and August 2022, 98% of Canadians had antibodies
against COVID-19 and 54% had antibodies from a previous infection. The Daily.
Retrieved April 13, 2025, from https://www150.statcan.gc.ca/n1/daily–
quotidien/230327/dq230327b–eng.htm
Statistics Canada. (2023). Experiences of Canadians with long–term symptoms following
COVID-19. The Daily. https://www150.statcan.gc.ca/n1/daily–
quotidien/231208/dq231208a–eng.htm

48
Steenhuysen, J. (2022, October 18). Focus: Addiction drug shows promise lifting long COVID
brain fog, fatigue. Reuters. https://www.reuters.com/business/healthcare–
pharmaceuticals/addiction–drug–shows–promise–lifting–long–COVID-19 –brain–fog–
fatigue–2022–10–18/
Stevens, C., Aravind, S., Das, S., & Davis, R. (2013). Pharmacological characterization of LPS
and opioid interactions at the toll–like receptor 4. British Journal of Pharmacology,
168(6), 1421–1429. https://doi.org/10.1111/bph.12028
Sullivan, G. M., & Artino, A. R. (2013). Analyzing and Interpreting Data From Likert–Type
Scales. Journal of Graduate Medical Education, 5(4), 541–542.
https://doi.org/10.4300/JGME–5–4–18
Tamariz, L., Bast, E., Klimas, N., & Palacio, A. (2024). Low–dose Naltrexone Improves post–
COVID-19 condition Symptoms. Clinical Therapeutics, 46(3), e101–e106.
Taquet, M., Dercon, Q., Luciano, S., Geddes, J. R., Husain, M., & Harrison, P. J. (2021).
Incidence, co-occurrence, and evolution of long COVID features: A 6 6-month
retrospective cohort study of 273,618 survivors of COVID-19 -19. PLOS Medicine,
18(9), e1003773. https://doi.org/10.1371/journal.pmed.1003773
Toh, Z. Q., Anderson, J., Mazarakis, N., Neeland, M., Higgins, R. A., Rautenbacher, K., Dohle,
K., Nguyen, J., Overmars, I., Donato, C., Sarkar, S., Clifford, V., Daley, A., Nicholson,
S., Mordant, F. L., Subbarao, K., Burgner, D. P., Curtis, N., Bines, J. E., … Licciardi, P.
V. (2022). Comparison of Seroconversion in Children and Adults With Mild COVID-19.
JAMA Network Open, 5(3), e221313.
https://doi.org/10.1001/jamanetworkopen.2022.1313

49
Toljan, K., & Vrooman, B. (2018). Low–dose naltrexone (LDN)–review of therapeutic
utilization. Medical Sciences (Basel, Switzerland), 6(4), 82.
https://doi.org/10.3390/medsci6040082
Toronto, C. E., & Remington, R. (Eds.). (2020). A Step–by–Step Guide to Conducting an
Integrative Review. Springer International Publishing. https://doi.org/10.1007/978–3–
030–37504–1
Vatvani, A. D., Patel, P., Hariyanto, T. I., & Yanto, T. A. (2024). Efficacy and safety of low–
dose naltrexone for the management of fibromyalgia: A systematic review and meta–
analysis of randomized controlled trials with trial sequential analysis. The Korean
Journal of Pain, 37(4), 367–378. https://doi.org/10.3344/kjp.24202
U.S. Food and Drug Administration. (2013). Revia (naltrexone hydrochloride) tablets, for oral
use: Prescribing information.
https://www.accessdata.fda.gov/drugsatfda_docs/label/2013/018932s017lbl.pdf
U.S. National Library of Medicine. (2021). Naltrexone [Drug Information]. In StatPearls.
StatPearls Publishing. https://www.ncbi.nlm.nih.gov/books/NBK534811/
Vondeling, G. T., Cao, Q., Postma, M. J., & Rozenbaum, M. H. (2018). The Impact of Patent
Expiry on Drug Prices: A Systematic Literature Review. Applied Health Economics and
Health Policy, 16(5), 653–660. https://doi.org/10.1007/s40258–018–0406–6
Wenner, D. M., Kimmelman, J., & London, A. J. (2015). Patient–Funded Trials: Opportunity or
Liability? Cell Stem Cell, 17(2), 135–137. https://doi.org/10.1016/j.stem.2015.07.016
Whittemore, R., & Knafl, K. (2005). The integrative review: Updated methodology. Journal of
Advanced Nursing, 52(5), 546–553. https://doi.org/10.1111/j.1365–2648.2005.03621.x

50
World Health Organization (2019). International statistical classification of diseases and related
health problems (11th ed. Retrieved 1 March 2025 from https://icd.who.int/en/
World Health Organization. (2022, February 3). True extent of SARS–CoV–2 infection through
seroprevalence studies. https://www.who.int/news/item/03–02–2022–true–extent–of–
sars–cov–2–infection–through–seroprevalence–studies
World Health Organization (2023). Clinical management of COVID-19: Living guideline, 18
August 2023. Retrieved 25 March 2025, from
https://www.who.int/publications/i/item/WHO–2019–nCoV–clinical–2023.2
World Health Organization (2025). COVID-19 dashboard. Retrieved 25 March 2025, from
https://data.who.int/dashboards/COVID-19 19/cases
World Health Organization (2025). Post COVID-19 condition (Long COVID). Retrieved 31
March 2025, from https://www.who.int/europe/news–room/fact–sheets/item/post–
COVID-19 -19–condition
Yale School of Medicine. (n.d.). Immunology of Long COVID. Yale University. Retrieved April
15, 2025, from https://medicine.yale.edu/lab/iwasaki/projects/immunology–long–
COVID-19 /
Yong, S. J., Halim, A., Halim, M., Liu, S., Aljeldah, M., Al Shammari, B. R., Alwarthan, S.,
Alhajri, M., Alawfi, A., Alshengeti, A., Khamis, F., Alsalman, J., Alshukairi, A. N.,
Abukhamis, N. A., Almaghrabi, F. S., Almuthree, S. A., Alsulaiman, A. M., Alshehail, B.
M., Alfaraj, A. H., … Rabaan, A. A. (2023). Inflammatory and vascular biomarkers in
post–COVID–19 syndrome: A systematic review and meta–analysis of over 20
biomarkers. Reviews in Medical Virology, 33(2), e2424. https://doi.org/10.1002/rmv.2424

51
Younger J & Mackey S. (2009). Fibromyalgia symptoms are reduced by low–dose naltrexone: A
pilot study. Pain Medicine (Malden, Mass.), 10(4), 663–672.
https://doi.org/10.1111/j.1526–4637.2009.00613.x

52
Appendix A: Prisma Flow Diagram

Screening

Identification

Identification of studies via databases and other methods

Records identified through:
Database searching (n=17)
CINAHL (EBSCO): 5
MEDLINE (EBSCO): 12
Google: 2

Records removed before
screening:
Duplicate records removed
(n = 2)
Records removed for other
reasons (n =2)
Reason 1: Not full text

Abstracts/Records screened with
inclusion criteria.
(n =1) – Not English

Articles/records excluded**
(n =3)

Full-text articles sought for
retrieval
(n =12)

Articles not retrieved
(n =0)

Included

Full-text articles assessed for
eligibility
(n =12)

Records excluded:
Reason 1: Not related to
PCC (n =4)

Records included in review
(n =10)
Systematic Reviews: 2
Laboratory studies: 1
Quantitative: 7
Case Study:1

PRISMA 2020 flow diagram adapted from Page MJ, McKenzie JE, Bossuyt PM, Boutron I,
Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for
reporting systematic reviews. BMJ 2021;372:n71. Doi: 10.1136/bmj.n71
From Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The
PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ
2021;372:n71. Doi: 10.1136/bmj.n71
For more information, visit: http://www.prisma–statement.org/

53

Appendix B: Data Extraction Matrix

Author
etc.

Titration
schedule

Bonilla
et al.
(2023)

Dosage
and
time of
day

Dosedepend
ent
effects

Duratio Side
n of
effects
therapy

Sympto
ms
tracked
and most
improve
d

Scales
and
biomar
kers

Setting

Individual 0.5–6
ized
mg
dose–
daily
titration
ranging
from 0.5
mg daily
to 6.0 mg
daily

Effects
are not
dosedepend
ent

77–255 Not
discuss
days
ed

29
symptom
s
tracked.
Fatigue,
postexertiona
l
malaise,
sleep
issues
most
improve
d

Likert
scale,
Post
COVID
-19
Functio
nal
Status
Scale

Stanford
PASC
clinic

Dietz &
Brondst
ater
(2024)

Not
mentione
d

Not
mentio
ned

Not
mentio
ned

Not
mentio
ned

Not
mentio
ned

Fatigue, Not
chest
mentio
pain,
ned
palpitati
ons,
brain
fog,
sleep
disturban
ces

Based on
Isman,
Bonilla,
and
O’Kelly

Hurt et
al.
(2024)

Not
mentione
d

Not
mentio
ned

Not
specifi
ed

None
mentio
ned

None

Fatigue,
brain
fog,
shortness
of
breath,
muscle
pain,

Long
COVID
clinic

Likert
scaling
of
sympto
ms

54
headache
, etc.
Isman
et al.
(2024)

Slow
titration
over 9
days

4.5 mg
at
bedtim
e
(some
up to 6
mg/day
)

Nausea Persisten
,
t fatigue
fatigue,
dizzine
ss, low
mood

SF-36
and
modifie
d
Chalder
scale

Ageless
Rx clinic

Livierat
os et al.
(2024)

Reported
from
other
studies

0.5–6
Not
mg/day origina
l data

Not
origina
l data

Not
origina
l data

Alternati
ve
therapies
for LC
discusse
d

Not
original
data

Literature
review

Naik et
al.
(2024)

Week 1–4
titration
up to 4.5
mg/day

4.5
mg/day
or max
tolerate
d dose

Planne
d
gradual
titratio
n

16
weeks
(12
weeks
at full
dose)

Monito
red

Primaril
y fatigue
and
biomark
ers

Fatigue
Severit
y Scale
(FSS)

BC-wide,
centered
on BC
Women
and
Children'
s Hospital

O’Kelly
et al.
(2022)

1 mg for 1 1–2
Not
month, 2
mg/day discuss
mg for 1
ed
month

2
months

2/38
withdre
w due
to side
effects

Wide
symptom
tracking
includin
g
fatigue,
brain
fog,
sleep
issues

Likert
scale
for
sympto
ms

Mater
Misericor
diae
Universit
y
Hospital,
Dublin

Petrace
k et al.
(2023)

Not
specified

16+
months

None
reporte
d

Helped
Patient
ME/CFS opinion
and
fibromya
lgia
symptom
s

4.5 mg

Not
12
directly weeks
assesse
d

Not
discuss
ed

Single
case
study

55
Sasso et
al.
(2024).

Not
applicable
(lab
study)

Tamariz Not
et al.
specified
(2024)

Not
applica
ble

Not
applica
ble

Not
applica
ble

Not
applica
ble

TRPM3
ion
channel
function
studied

Lab
outcom
es

Lab study

1.5–4.5 Not
mg/day discuss
ed

Minim
um 1
month

Not
tracked

Fatigue,
pain,
brain
fog,
dyspnea

CRP
and
cortisol
biomar
kers

VA
Medical
Center,
Miami