key: cord-0737207-pqz5r1c3
authors: Zarabanda, David; Vukkadala, Neelaysh; Phillips, Katie M.; Qian, Z. Jason; Mfuh, Kenji O.; Hatter, Matthew J.; Lee, Ivan T.; Rao, Vidya K.; Hwang, Peter H.; Domb, George; Patel, Zara M.; Pinsky, Benjamin A.; Nayak, Jayakar V.
title: The Effect of Povidone‐Iodine Nasal Spray on Nasopharyngeal SARS‐CoV‐2 Viral Load: A Randomized Control Trial
date: 2021-11-09
journal: Laryngoscope
DOI: 10.1002/lary.29935
sha: 0cdd2796fac21f16b8ea9dacf47ebfe0ff577ce1
doc_id: 737207
cord_uid: pqz5r1c3

OBJECTIVES/HYPOTHESIS: To determine the effect of povidone‐iodine (PVP‐I) nasal sprays on nasopharyngeal (NP) viral load as assessed by cycle threshold (Ct) on quantitative polymerase chain reaction (qPCR) of SARS‐CoV‐2 in outpatients. STUDY DESIGN: Three arm, triple blinded, randomized, placebo‐controlled clinical trial. METHODS: Participants were randomized within 5 days of testing positive for COVID‐19 to receive nasal sprays containing placebo (0.9% saline), 0.5% PVP‐I, or 2.0% PVP‐I. NP swabs for qPCR analysis were taken at baseline, 1‐hour post‐PVP‐I spray (two sprays/nostril), and 3 days post‐PVP‐I spray (20 sprays/nostril). Symptom and adverse event questionnaires were completed at baseline, day 3, and day 5. University of Pennsylvania Smell Identification Tests (UPSIT) were completed at baseline and day 30. RESULTS: Mean Ct values increased over time in all groups, indicating declining viral loads, with no statistically significant difference noted in the rate of change between placebo and PVP‐I groups. The 2.0% PVP‐I group showed statistically significant improvement in all symptom categories; however, it also reported a high rate of nasal burning. Olfaction via UPSIT showed improvement by at least one category in all groups. There were no hospitalizations or mortalities within 30 days of study enrollment. CONCLUSIONS: Saline and low concentration PVP‐I nasal sprays are well tolerated. Similar reductions in SARS‐CoV‐2 NP viral load were seen over time in all groups. All treatment groups showed improvement in olfaction over 30 days. These data suggest that dilute versions of PVP‐I nasal spray are safe for topical use in the nasal cavity, but that PVP‐I does not demonstrate virucidal activity in COVID‐19 positive outpatients. LEVEL OF EVIDENCE: II Laryngoscope, 2021

The emergence and spread of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes the inflammatory respiratory disease termed coronavirus disease 2019 (COVID- 19) , has created a worldwide health crisis. Intensive research efforts have led to the formulation of effective vaccines, 1 but human-to-human transmission of SARS-CoV-2 remains widespread. 2 There is an urgent need to develop other affordable and readily available strategies to reduce viral transmission and/or mitigate symptoms of COVID-19.

Transmission of SARS-CoV-2 occurs primarily through the spread of aerosolized droplets into the nose. 3, 4 The virus gains access by binding to angiotensin-converting enzyme 2 receptors present in the cilia of the upper and lower airway respiratory epithelium. 4 Recent studies have demonstrated a correlation between SARS-CoV-2 viral load and COVID-19 disease severity, mortality, and infectivity. 5, 6 Therefore, direct application of topical virucidal treatment to the nasopharyngeal (NP) tissues to arrest SARS-CoV-2 replication would be expected to decrease viral burden in exposed patients.

One possible topical therapy for reducing SARS-CoV-2 viral titers is povidone-iodine (PVP-I) solution, a widely available and inexpensive antiseptic medication. PVP-I is frequently used to treat intranasal methicillinresistant Staphylococcus aureus colonization 7 and to reduce the incidence of surgical wound infections in the operating room. 8 Prior in vitro studies have demonstrated that dilute PVP-I solutions, in concentrations as low at 0.5%, produce a ≥4 log 10 decrease in SARS-CoV-2 viral titers (inactivation of ≥99.99% of virus) within 15 seconds of exposure. 9,10 PVP-I nasal sprays of up to 4.4% concentration have also been used in healthy volunteers without adverse effects. 11 These reports have generated intense interest in intranasal PVP-I treatments worldwide. 12 Despite the in vitro efficacy, it is unclear whether intranasal PVP-I would be effective in vivo given the surface area of the nasal mucosa, rapid mucociliary clearance, volume and viscosity of mucus present. 13 Recently, Guenezan and colleagues published the first human clinical trial comparing PVP-I mouth rinses and nasal sprays to nonintervention in COVID-19 patients and did not find significant differences in SARS-CoV-2 NP viral load between the groups. 14 This study was neither blinded nor placebo-controlled.

We report the results from our randomized, tripleblinded, placebo-control clinical trial comparing low and high concentration PVP-I and saline nasal sprays as topical treatments to reduce the NP viral load of SARS-CoV-2 in outpatients.

This PVP-I COVID-19 trial was a prospective, three arm, triple-blinded, placebo-controlled clinical trial conducted at Stanford University School of Medicine and the Stanford Center for Sinus and Endoscopic Skull Base Surgery. The study was approved by the Stanford Institutional Review Board (protocol #56134), and the trial was registered with clinicaltrials.gov under trial number NCT04347954.

The eligible study population was limited to outpatients who had a documented positive NP swab for COVID-19 within 5 days prior to study enrollment. All participants were at least 18 years of age. Exclusion criteria included: 1) a history of adverse reactions to iodine supplementation and/or food dye; 2) current/active use of intranasal steroid sprays; 3) sinonasal surgery within 30 days of study enrollment; 4) pregnancy; and 5) participation in other COVID-19 trials. Study participants received a modest compensation (US$50) for their participation in this study.

Participants were randomized 1:1:1 into three groups: 0.9% NaCl (saline group), 0.5% PVP-I in water, and 2% PVP-I in water. 15 The use of PVP-I in water rather than saline solution was based on prior publications. 9,10 Each study medication was prepared by an independent, licensed compounding pharmacy (A & O Specialty Pharmacy, Salinas, CA), and dispensed into identical, opaque nasal atomizer spray bottles that deliver standardized 0.1 mL spray volume with each pump. The spray bottles were labeled as "A," "B," or "C" and the corresponding treatment solutions were only known to the pharmacist, who was not directly involved in the study. To further ensure blinding of the study team, the pharmacist added food coloring to the 0.9% NaCl saline solution to give it a similar appearance to PVP-I. The study was thereafter triple blinded: the treatment assignment was concealed to both the participants and investigators during all aspects of data collection and analysis, until the primary outcome measure (reverse transcription polymerase chain reaction cycle threshold, or RT-PCR Ct) had been assessed.

On treatment day 1, after providing written informed consent, participants underwent an initial baseline NP swab for RT-PCR testing for SARS-CoV-2. Next, after an educational review of nasal spray administration technique, the assigned randomized spray bottle was provided, and the first treatment of two sprays per nostril was self-administered by each participant.

A second NP swab specimen was then collected from all participants 1 hour following the first treatment. This time point was included to assess for an immediate virucidal effect to the sprays while allowing for sufficient time to sweep the nasal cavity of nonviable virus particles via mucociliary clearance. Patients were then instructed to self-administer nine additional treatments (two sprays per nostril, four times a day at home), using the assigned, randomized spray bottle. A third NP swab was collected on treatment day 3, after 10 treatments, or a total of 20 sprays per nostril, had been self-administered. All NP swabs were performed by the study principal investigator (J.V.N.) on the same side for each patient.

All patients completed baseline questionnaires, which provided demographic and clinical information, as well as a baseline olfactory assessment using the validated University of Pennsylvania Smell Identification Test (UPSIT). 16 Disease-related symptoms and potential adverse reactions related to use of the assigned nasal spray use were monitored in follow-up. On days 1, 3, and 5, participants rated the severity of their symptoms on a 4-point Likert scale with "1" representing "not experiencing that symptom" and "4" representing "severe symptoms." Additionally, on days 3 and 5, participants provided a global assessment of the trajectory of their symptoms by answering the question of "Compared with yesterday, I feel my infection is:" on a 5-point scale ranging from "much worse" to "much better." A final UPSIT was completed by participants 30 days following study enrollment and returned to Stanford by mail. Adverse events were monitored by asking participants if they experienced symptoms such as nasal pain or burning. Participants could then pick adverse event frequency as "never," "rare," "sometimes," "often," or "always." On day 5 of the study, adherence to nasal sprays was queried and participants were asked to report adherence as either "0-25%," "26-50%," "51-75%," or "76-100%" of all doses.

All NP swab samples were placed in 2 mL of saline in 15 mL conical tubes. These tubes were kept on ice until delivered to the Stanford Clinical Virology Laboratory. Specimens were processed in the following fashion. 17 Briefly, total nucleic acids were extracted from 500 μL saline on the QIAsymphony SP using the QIAsymphony DSP Virus/Pathogen Midi Kit (both from Qiagen, Germantown, MD), and eluted in 60 μL buffer AVE as supplied by the manufacturer. SARS-CoV-2 RNA was detected using previously described primer and probe sequences targeting the envelope (E) gene. 18 These were combined in multiplex with RNase P primers and probe. Real-time RT-PCR was performed using the SuperScript III One-Step RT-PCR System with Platinum Taq DNA Polymerase Kit (Invitrogen, Carlsbad, CA) on the Rotor-Gene Q Instrument (Qiagen).

Cycle threshold (Ct) denotes how many PCR cycles are required before the SARS-CoV-2 viral RNA reached a detectable level. Higher Ct values correspond to lower viral copy numbers. For reference, Ct values of 20 correspond to $2.12 Â 10 6 viral copies per milliliter, while a Ct value of 40 is undetectable and is considered the lower limit of this RT-PCR assay. Viral copy number is derived using a formula which applies a logarithmic transformation to the Ct.

During the course of the study, the SARS-CoV-2 strandspecific real-time RT-PCR test for actively replicating SARS-CoV-2 virus was developed and validated at the Stanford Virology laboratory. The detection of SARS-CoV-2 minus-strand RNA was shown to correlate with the presence of active, replicating virus. 19 A post-hoc exploratory analysis was performed on our samples to assess whether there was a detectable effect on minus strand presence as a proxy for determining inactivation of viral replication following spray use.

Because this trial was developed and conducted early in the pandemic at a time when limited information was available on the natural history of the virus and no prior data existed on PVP-I as a therapy, the sample size for this study is not based on statistical assumptions for formal hypothesis testing. We aimed to recruit 45 subjects in total. This number of participants was considered sufficient to provide a descriptive summary of the safety and primary outcome. Statistical analyses were performed using Stata 16 (StataCorp LP, TX). Ct was used as the primary unit of measurement. For our primary outcome of NP viral load, we present the data as Ct, since this utilizes the original output of the qPCR test.

The Shapiro-Wilk's test was used to determine the normality of the data. Performance of treatment arms over time was compared using a mixed effects maximum likelihood linear regression model with random intercept and random slope. Time (in hours since first swab), treatment arm, and Ct value were entered into the model, which then predicted the effect of time and treatment arm on Ct values. The model also corrected for baseline Ct values.

For symptoms, the Friedman test was used to assess changes in symptom scores for each of the tracked symptoms over time in the study arms. Changes in global symptom rating were compared using the chi-square test. Changes in olfaction, as categorized by normative values of the UPSIT were assessed using the chi-square test. Likewise, differences in the frequencies of adverse events were assessed using the chi-square test or Fischer's exact test.

Overall, 265 patients were screened for eligibility and 35 patients were included in the final analysis for the primary outcome measure (saline: 11 participants; Table I . There was a mean of 3.5 days (SD AE 1.9 days) from a positive COVID test to study enrollment, and a mean of 7.0 days (SD AE 5.0 days) from symptom onset to study enrollment.

Mean baseline Ct values for intranasal SARS-CoV-2 were similar across all groups at 23.7 (SD AE 6.9) in the saline group, 26.0 (SD AE 4.9) in the 0.5% PVP-I group, and 28.4 (SD AE 6.4) in the 2.0% PVP-I group (p = .179).

Mixed effects linear regression showed that Ct values increased with time (mean difference = 0.055 cycles/hour; 95% confidence interval [CI] 0.037 to 0.074) in the overall cohort, indicating a decline in SARS-CoV-2 NP viral loads in all groups. However, there was no difference seen in the longitudinal performance between the three nasal spray cohorts over the duration of this study ([saline: reference], [0.5% PVP-I: mean difference = À0.349 cycles/hour; 95% CI À1.584 to 0.886], [2.0% PVP-I: mean difference À1.059 cycles/hour; 95% CI À2.318 to 0.201]), that is, saline and PVP-I nasal sprays were not significantly different in altering Ct in COVID-19 positive patients (Fig. 2) . 

An exploratory chi-square analysis was used to compare the number of participants in each group who initially had a detectable minus-strand SARS-CoV-2 RNA (indicating active viral replication) on study day 1, who subsequently converted to an undetectable minus-strand by day 3 (suggesting viral inactivation). Given the limited sample size, the two PVP-I spray groups were combined for this analysis with seven subjects in the PVP-I and seven subjects in the saline group. In each group, five subjects converted to an undetectable minus strand by day 3, while two subjects did not. The mean change in Ct value for the minus strand between baseline and 1 hour post administration was also assessed, and again there was no significant difference between the groups (saline mean change in Ct: 0.31, PVP-I combined group mean change in Ct: 0.29, p = .89).

A total of 33 of 35 participants completed and returned day 30 UPSIT studies (saline: 11; 0.5% PVP-I: 11; 2.0% PVP-I: 11). When comparing the changes in the degree of anosmia from baseline to post-treatment, there was no statistically significant difference in the percentage of patients who improved in each group (saline: 70%; 0.5% PVP-I: 70%; 2.0% PVP-I: 89%; p = .50). No participant in this clinical trial demonstrated worsening olfaction (Table II) .

Mean adherence was 100% in the 2.0% PVP-I group, 100% in the 0.5% PVP-I group, and 91% in the saline group. There were no differences in the reported frequencies of baseline symptoms between groups. The 2.0% PVP-I group demonstrated statistically significant improvement in all symptoms-fevers, chills, fatigue, smell, taste, congestion, and sore throat over time. The 0.5% PVP-I group only reported significant improvements in taste and sore throat. Meanwhile the saline spray group reported significant improvements in fever, chills, fatigue, and congestion (Table III) . In terms of overall subjective health status, on days 3 and 5, all participants reported feeling either the same or better than they had in the prior 24 hours. The number of participants (saline: 82%; 0.5% PVP-I: 73%; 2.0% 

Subjects randomized to 2.0% PVP-I reported statistically higher rates of burning sensation and pain in the nose (day 3: 92.9%; day 5: 92.9%) as compared to saline (day 3: 16.7%; day 5: 16.7%) and 0.5% PVP-I (day 3: 28.5%; day 5: 14.3%) groups (Table IV) . Of note, no participants in our study required visits to the emergency department or hospitalization for COVID-19, and there there were no mortalities at 30 days after study enrollment.

In this randomized, triple-blinded, placebo-controlled trial of PVP-I nasal sprays, there were similar improvements in Ct values (i.e., similar reduction in SARS-CoV-2 viral loads) between subjects who received saline nasal sprays and PVP-I nasal sprays for treatment of COVID-19. Notably, no participants in our study reported emergency department visits or hospital admissions for COVID-19-related symptoms. Participants in all three groups reported subjective improvement in their condition on days 3 and 5. Finally, no participants reported any worsening in clinical status over the course of the study.

Recently, a randomized, nonblinded, and nonplacebo controlled trial compared a 12 person intervention group of 1% PVP-I oral rinse, spray, and anterior nasal application of 10% PVP-I ointment to a 12 person nonintervention group. 14 This study did not find differences in SARS-CoV-2 viral load between the treatment and placebo groups based on RT-PCR analysis. Similarly, in our study, we found that the saline group had comparable rates of decline in SARS-CoV-2 viral load when compared to both PVP-I groups as assessed by qPCR over the 3-day study period. Although there were only 35 participants in our pilot study, there was no trend in the data toward a difference in viral load reduction between groups so it is less likely that increasing the sample size would lead to identification a clinically significant difference.

Anecdotally, many providers apply PVP-I nasal sprays or oral rinses prior to examination of patients in the outpatient setting. 20 However, our 1-hour NP swab time point after initial dosing of sprays showed no significant reduction in Ct and viral load for SARS-CoV-2 across our three groups. qPCR simply measures a specific segment of viral RNA but cannot demonstrate whether the virus is viable and capable of infecting cells. As part of our exploratory analysis, we used a minus strand SARS-CoV-2 RT-PCR test which correlates with culturable live virus to determine whether any of the treatment arms had an effect on viral replication (i.e., as a proxy for live virus). However, both the saline and PVP-I groups had minimal changes in Ct at the 1-hour time point. This indicates that neither saline nor PVP-I was effective in reducing viral replication after a single administration of therapy. The inclusion of the minus strand test lends a unique dimension to our study as it more directly assesses for changes in viral replication when compared to traditional PCR assessments of viral load. Further research will be needed to demonstrate whether these prophylactic measures indeed provide any benefit in reducing live virus and the risk of transmission.

Overall, sprays were well tolerated with excellent compliance in each group. The 2.0% PVP-I group had significantly higher reports of nasal burning and sneezing as compared to the other groups. One theoretical risk of PVP-I is chemosensory dysfunction due to in vitro cytotoxicity noted with concentrations above 2.5%. 10 Reassuringly, we did not observe any worsening of olfaction in any individual patient in any treatment arm as assessed by the UPSIT smell test. There was a trend toward greater recovery of smell function in the PVP-I groups with a higher percentage of patients improving by at least one olfaction category, and overall higher mean improvement in UPSIT scores; however, these differences did not meet the threshold for statistical significance.

There was no negative control group (i.e., received no spray treatment), which limits the ability to determine how much of the observed declines in Ct over time was due to natural decreases in viral load over 3 days' time versus decline in intranasal virus induced by the administered treatments. Participants used nasal sprays as these were felt to be easier to provide in a blinded fashion and to ensure consistent use; however, the sprays may not have reached as many surfaces as sinus irrigations. A saline nasal spray group was selected instead of a negative control group given concerns that the inconvenience of study participation would have prevented adequate enrollment if participants did not have the benefit of potentially receiving a treatment for SARS-CoV-2.

Although patients presented, on average, within 3 days following a positive test, the time from self-reported symptom onset was an average of approximately 7 days in the PVP-I groups, which may have been outside the window of peak viral load. 21 We are unable to determine whether earlier use of PVP-I might have been more efficacious. Additionally, SARS-CoV-2 RT-PCR is a proxy for and not a direct measure of the amount of viable virus present.

There is a significant need for safe and effective treatments against SARS-CoV-2. In this randomized, tripleblinded, placebo-controlled clinical trial, we compared topical nasal saline to low and high concentration PVP-I as potential therapies in COVID-19 patients. We found that saline and low dose (0.5%) PVP-I nasal sprays are overall well-tolerated, though more nasal burning symptoms were reported in the 2.0% PVP-I group. All three nasal sprays yielded similar reductions in SARS-CoV-2 NP viral load over 3 days. Olfaction improved in all groups over 30 days, and notably did not decline for any participant. Taken together, these data suggest that dilute PVP-I sprays may be safe for topical exposure on nasal mucosal tissues, but do not appear to confer increased virucidal activity in COVID-19 positive patients. Future work is needed to fully quantify the benefit of nasal sprays for the treatment of COVID-19.

Safety and efficacy of the BNT162b2 mRNA Covid-19 vaccine

Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine

SARS-CoV-2 entry factors are highly expressed in nasal epithelial cells together with innate immune genes

ACE2 localizes to the respiratory cilia and is not increased by ACE inhibitors or ARBs

SARS-CoV-2 viral load is associated with increased disease severity and mortality

Association of SARS-CoV-2 genomic load with COVID-19 patient outcomes

Efficacy of skin and nasal povidone-iodine preparation against mupirocin-resistant methicillinresistant Staphylococcus aureus and S. aureus within the anterior nares

Guideline for prevention of surgical site infection

Povidone-iodine demonstrates rapid in vitro virucidal activity against SARS-CoV-2, the virus causing COVID-19 disease

In vitro efficacy of a povidone-iodine nasal antiseptic for rapid inactivation of SARS-CoV-2

A clinical study on the tolerability of a liposomal povidone-iodine nasal spray: implications for further development

The use of povidone iodine nasal spray and mouthwash during the current COVID-19 pandemic may reduce cross infection and protect healthcare workers

Association of chemosensory dysfunction and Covid-19 in patients presenting with influenza-like symptoms

Povidone iodine mouthwash, gargle, and nasal spray to reduce nasopharyngeal viral load in patients with COVID-19

Development of the University of Pennsylvania Smell Identification Test: a standardized microencapsulated test of olfactory function

Comparison of the Panther Fusion and a laboratory-developed test targeting the envelope gene for detection of SARS-CoV-2

Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

Strand-specific reverse transcription PCR for detection of replicating SARS-CoV-2

More on betadine from JAMA. 2020. Available at

SARS-CoV-2, SARS-CoV, and MERS-CoV viral load dynamics, duration of viral shedding, and infectiousness: a systematic review and meta-analysis

The authors wish to thank the Stanford Clinical Virology Laboratory team for their assistance with testing of samples, the A&O Specialty Pharmacy for nasal spray preparation and packaging for this clinical trial, and finally all study subjects for their willingness to undergo serial NP swabs and self-administer nasal sprays for this clinical trial during the COVID-19 pandemic in the US.