It will be a parallel group, quadruple blind-randomized controlled pilot trial annexed
with a laboratory based study. Clinical trial will be carried out at the Aga Khan
University Hospital (AKUH), Karachi, Pakistan. Patients will be inducted from the pool of
known patients (laboratory confirmed COVID-19 participants) within 7 days of the onset of
viral infection symptoms, already admitted at AKUH. Molecular and immunological testing
will be done at the Juma laboratory of AKUH. The intervention drugs (Povidone-Iodine,
Hydrogen Peroxide and Hypertonic Saline will be obtained from the AKUH distribution
department and/ or AKUH pharmacy. The Neem extract will be compounded at the Chemistry
department, University of Karachi/ HEJ institute of Organic Chemistry, University of
Karachi.

We will need 50 patients. There will be five study groups. Group A (n=10) patients on 10
ml gargle and nasal lavage with of 0.2% Povidone-Iodine for 20-30 seconds, twice daily
for 5 days. Group B (n=10) patients will be subjected to 10 ml gargle and nasal lavage
with 1% Hydrogen peroxide for 20-30 seconds, twice daily for 5 days. Group C will
comprise of (n=10) subjects using 10ml gargle and nasal lavage with Neem extract
(Azardirachta indica) for 20-30 seconds, twice daily for 5 days. Group D (n=10) patients
will use 2% hypertonic saline gargle and nasal lavage for a similar time period. Group E
(n=10) will serve as positive controls. These will be given simple distilled water
gargles and nasal lavage for 20-30 seconds, thrice daily for six days. For nasal lavage,
a special douche syringe will be provided to each participant. Its use will be thoroughly
explained by the data collection officer.

Data collection: The baseline oral swab will be taken from the posterior pharyngeal wall/
tonsillar area on day one before initiating the gargles by trained dentist. The end-point
oral swab will be taken on day 7, just after using the prescribed gargle. Patient will be
provided with a special hood so that they themselves don't generate aerosol in the
immediate vicinity while carrying out the gargling and nasal lavage.

Data will be analyzed using SPSS for Windows (version 23.0 SPSS) and Graph Pad Prism 7.0
software.

Mean and standard deviation of the continuous variables (age, COVID-19 viral counts and
cytokine profiles at baseline and endpoint etc.) will be computed. Frequency distribution
of the categorical variables will be determined (gender, co-morbids, presence of
hypertension, diabetes, periodontal status etc.) Repeated measures ANOVA will be used to
compare the reduction in intra-oral viral load and the changes in the inflammatory
biomarkers in the study groups. In case of substantially low count of participants in the
study groups, non-parametric tests such as Kruskal-Wallis or Friedman test will be
employed. Multiple linear regression (MLR) will done to predict changes in the Covid-19
viral load and cytokine profiles. A subset analysis using Zero inflated negative binomial
(ZINB) model will be employed, provided a significant reduction in viral load is observed
in any of the study groups. A p-value of <0.05 will be taken as statistically
significant. {20a}

The biases and confounders can be taken care at following four levels:

At recruitment stage, we will use restriction i.e. only those COVID positive subjects
will be recruited who have no other major co-morbidities. At intervention assignment, the
random group allocation will ensure even distribution of the confounders in the study
groups thus their differential effect on the outcome will be addressed. At analysis, we
will employ regression analysis (as described above) to account for biases and
confounders and lastly, the subset analysis will stratify the data and nullify the effect
of biases.

Interim analysis {20b} is not planned on account of a short duration of the study.
Missing data/ uncompliant patients {20c} will be omitted from the final analysis.